Whole Conference (3 days): 375 EUR
Daily Rate: 150 EUR
Half-Day Workshop: 100 EUR
Full-Day Workshop: 200 EUR
|Before Aug 14|| After Aug 14|
| SAN ||25%||15% |
| AAPB, ISNR, Eastern Europeans||15%||5% |
*Discounts may not be combined.
Convenient on campus overnight accommodation in new, single, 4-star ensuite
rooms is available to book at 63 Euro/night, including breakfast.
Swansea Bay is many destinations rolled into one! The maritime City of Swansea, birthplace of Catherine Zeta Jones and Dylan Thomas, the Victorian seaside resort of Mumbles, the Gower Peninsula and the Waterfalls Country combine into a breathtaking mix of spectacular contrasts, all you need for a fabulous holiday!
Absract Submissions Now Closed
To submit a poster abstract, please use the online Abstract Submission Form .
In collaboration with
Symposia, Papers, Posters, Round tables 17th-19th September 2006
Workshops, Training Courses 14-16th September
COST 15-16th September
Call Deadline, to include student scholarships, April 30th 2006
This conference will bring together both scientists and practitioners
from a wide range of specialist fields.
|Audio-Visual Stimulation||Neurofeedback - EEG, fMRI, MEG|
|Applied Clinical Neuroscience||Brain-Computer Interface|
|Peak Performance||EEG and Cognition|
|Clinical Psychopathology||QEEG Diagnosis|
|Complimentary Evidence Base||Trans-Cranial Magnetic Stimulation|
|Clinical Outcomes||Peripheral Biofeedback|
|Behavioural Measurement||Near Infrared Spectroscopy|
|LORETA and Methodology||Virtual Reality|
The Inaugural Conference of the Society of Applied Neuroscience, attended by close to 150 participants, really could hardly have gone better, and has set standards for the future that will be hard to live up to.
Thank you everyone for making the inaugural meeting of our Society such a rewarding and memorable experience.
Affiliation: (1) Siberian Branch of the Russian Academy of Medical Sciences, Novosibirsk, (2) Macedonian Academy of Sciences and Arts, Skopje (3) Macedonian Philharmony, Skopje, (4) Faculty of Music Art, Skopje (5) Faculty of Medicine, Skopje (6) Novosibirsk State Musical College, Novosibirsk
Background: EEG and EMG studies demonstrated an increased alpha activity and decreased amplitude EMG activity in high skilled performing musicians. The objective of this study was to determine the effects of individual alpha neurofeedback and EMG biofeedback protocols for improvement musical performance in musicians.
Methods: The sample consisted of 27 musical students (10 pianists, 10 violinists, 5 winder instruments players and 2 viola players) 12 from Faculty of Music Art, Skopje and 15 from Novosinirsk Music College mean age of 19.01 and SD = 2.98. Twelve of them had low individual alpha peak frequency (IAPF) (< 10 Hz), and fifteen high IAPF (≥ 10 Hz). The sample was divided in two groups. The students from the experimental group participated in 20 sessions of biofeedback (alpha/EMG), combined with musical practice, while the students from the control group did only musical practice. Average absolute power, interhemispheric coherence in alpha band, IAPF, individual alpha band width (IABW), surface forearm integrated EMG power (IEMG), score on musical performance and inventories measuring anxiety were assessed.
Results: Alpha/EMG-B was associated with significant increase in average alpha power, IAPF and IABW in all participants and with decreases in IEMG only in high- IAPF musicians. The biofeedback training success was positively correlated with the alpha power, IcoH, IAPF, IABW and baseline level of IAPF and IABW.
Conclusions: Alpha/EMG Biofeedback is capable to increase voluntary self regulation and the quality of musical performance. The efficiency of biofeedback training depends on the baseline EEG alpha activity status, in particular the IAPF.
Affiliation: (1) Institute of Medical Psychology and Behavioral Neurobiology, University of Tübingen. (2) Department of Psychology, Goldsmiths College, University of London
Professional musicians have been the focus of scientific interest in the past, in neuroscience for instance, instrumental musicians have commonly been studied to investigate cerebral correlates of highly specialized motor skills. However, studies involving professional classical singers have remained absent. In the current study, we aimed at improving performance quality of elite classical singers by means of EEG-biofeedback, based on a series of studies by Egner & Gruzelier (2003), including assessments of cerebral correlates of singing before and after biofeedback. Elite singers from the Stuttgart State Opera and the Music Conservatory Stuttgart were randomly assigned to 3 groups: two training groups and one non-training control group. EEG biofeedback training in group 1 involved 10 sessions that aimed at increasing Sensory Motor Rhythm (SMR, 12-15Hz) activity at electrode position C4 in an eyes open resting state. EEG biofeedback in group two involved 10 sessions of alpha/theta training at electrode position Pz that aimed at increasing theta wave activity (5-8 Hz) over alpha wave activity (8-11 Hz) in an eyes closed resting state. Both trainings were conceptualized as relaxation training. The former one based on the association of SMR activity with reduced sensorimotor excitability and the latter one based on the association of increased theta wave activity during meditation and flow experiences (e.g. in marathon running). Pre-post training evaluation involved functional magnetic resonance imaging (fMRI) during executed and imagined singing of part of an aria, musical performance in front of an audience, analysis of spectral difference in the singing voice, skin conductance level for pre and post performance, as well as pychometric assessments, including NEO-FFI and the State-Trait Anxiety Inventory. We will present our first data of this ongoing study at the SAN conference with a focus on fMRI measures.
Affiliation: (1) Department of Psychology, Goldsmiths College, University of London.
(2) Western Eye Hospital, Imperial College Medical Faculty.
The present study examines the application of peak performance neurofeedback to the novel area of microsurgery assessing two common neurofeedback protocols. 20 NHS trainee eye surgeons were randomly assigned to either SMR and Alpha-Theta (AT) protocols or a wait-list 'no-treatment' control group (N = 8) who were subsequently randomized to the training protocols. Both groups received eight up to 30-minute sessions of training. Pre and post- assessment included a skills lab surgical procedure with objective timed measures and subjective ratings from video-recordings by consultant surgeons who were blind to group and order, together with state/trait and mood (POMS) questionnaires, and the Attention Network Test (ANT). Evidence of voluntary EEG self-regulation was shown by: 1) within session increases in SMR (p<0.002) and concomitant decreases in theta (p<0.001): and a between session increase in the SMR-theta ratio across five sessions (p<0.001), followed by a decline to baseline. 2) Alpha/theta training within session increases in the theta-alpha ratio which were significant up until the 24th minute (p<0.001); and between sessions a significant increase in ratio across all sessions (p<0.001).With SMR training there were significant advantages not present in the control group. There was improvement in surgical technique according to 1) the expert ratings: technical skill overall (p<0.038) and knot task (p<0.018) (judges' reliability r = 0.85); 2) task speed (total task time, p<0.021), while everyday anxiety (trait) decreased by approximately 10% (p<0.017), and importantly this decrease in anxiety correlated with improved skill. Within session SMR learning correlated positively with skill pause time (r = 0.584). Apha/theta training produced only a marginal improvement, evinced by overall performance time (p<0.10), which was accompanied by a large standard error indicative of uncontrolled individual differences. Notwithstanding, successful within session elevation of the theta-alpha ratio correlated positively with overall technical skill (r = 0.638), while across session learning correlated with overall skill time (r = -0.523).The SMR benefits were despite the fact that learning peaked at the fifth session and fell off for the final three sessions. Importantly the falloff was attributable to the poorer performers, as disclosed by splitting the group into equal halves of 5 subjects on the basis of improvements in task time, the SMR-theta ration continued to increase in the better performers. This further attests to the validity of the relation between SMR training and performance enhancement (Egner & Gruzelier, 2001, 2004; Vernon et al, 2003). Similarly despite the significant incremental learning effect across all 8 A/T sessions, a result not achieved in our previous studies, (Gruzelier & Egner, 2005), the alpha/theta training contingencies were suboptimal within sessions for there was a tail off at the end of the sessions not found in previous studies that produced enhancement in performance and mood (Egner and Gruzelier, 2003; Raymond et al, 2005). This implies that the A/T training sessions may have been too long for the doctors who sandwiched sessions into a busy schedule. This might also account here for the lack of mood benefits. No reliable advantages were found on the ANT task.
Conclusion: SMR neurofeedback training provides significant enhancements in surgical technique while considerably reducing time of instrument contact with the eye. There is also evidence that Alpha-Theta training marginally reduces total surgery time, in spite of suboptimal training efficacies. All in all the data set provides encouraging relationships validating the optimizing of performance on a complex professional skill through neurofeedback training.
Affiliation: State-Research Institute of Molecular Biology and Biophysics (1) and State-Research Institute of Physiology (2) Siberian Branch, Russian Academy of Medical Sciences,
Background: The very important issues of the Neurofeedback training consists in the fact that its efficiency runs up to no more than 75 % [Monastra et al., 2005]. It has been suggested that such inefficiency is due to individual variability of EEG in the frequency domain. The large variance in peak and width begs the question about necessity of use a large or narrower band to assess dominant frequency activity. The aim was to test of whether Neurofeedback training applied in order to increase or decrease power of individual EEG frequency ranges is more efficient than Neurofeedback training of standard EEG frequency ranges.
Methods: The sessions of theta/beta decreasing and alpha stimulating trainings were carried out on two outpatients with attention deficit disorder (the schoolboy) and functional pain contraction (professional musician). In both cases the procedure consisted of few sessions:
Neurofeedback sessions with the use standard frequency domain power of the ranges (SNFB): theta (4 - 8 Hz), alpha (8 - 12 Hz) and beta-1 (13-18 Hz) and Neurofeedback sessions with the use individual frequency domain power (INFB), before and after Neurofeedback sessions assessment of individual alpha indices were provided during resting sessions (with closed and open eyes) with the help of WinEEG program
Results: The Neurofeedback with standard frequency ranges was inefficient and even resulted in aggravation of symptoms of disorders in both cases. The Neurofeedback training with individual frequency ranges resulted in substantial clinical and psychometric traits improvement, increased individual alpha peak frequency, expansion alpha band width and the increase of EEG reactivity.
Conclusions: In the present research, due to the individual frequency EEG analysis, the necessity of an individualization of the approaches for Neurofeedback technology is convincingly shown.
Affiliation: Korea Institute of Brain Science
We have previously shown that two children trained in a Heightened Sensory Perception programme (HSP) have the ability to perceive colour whilst wearing a blindfold . Performance was dependent upon the visual information, as gradually reducing the stimulus intensity greatly affected performance until with near zero stimulus intensity, the performance was no better than chance. These promising results suggested that we test the repeatability of HSP colour recognition in a greater number of subjects and with a more robust stimulus. We therefore designed a stimulus delivery programme (HSPia) that presented the colours via an LCD display. Additionally, we tested whether correct HSP performance was due to peculiarities with our blindfold, so we created a blindfold goggle for comparison. Each subject (n = 7; 3 males, 4 females) performed 60 trials per day for 3 consecutive days. Subsequently the 3 male subjects performed 60 trials with the goggle blindfold, and 60 trials per day for a further 3 days with the computer monitor covered with black sheet plastic. The results suggest that HSP colour perception is a repeatable phenomenon, which is not dependent upon the blindfold used. Our current results also support our previous findings, suggesting that removing or reducing the visible information significantly reduces performance to chance levels. In addition fMRI and EEG studies are being carried out.
This study's main objectives were to recruit children with a primary diagnosis of ADHD to participate in an attention training (EEG biofeedback) paradigm to determine the effect of this treatment on the cardinal symptoms of ADHD. This was the first study to utilize a placebo condition and double-blind interface with this type of treatment.
Methods: The design of the study consisted of a diagnostic workup, 40 sessions (with a crossover after 20 sessions), and pre-, mid-, and post-assessments. Testing measurements were the Connor's parent and teacher rating scales, and the Integrated Visual and Auditory continuous performance test (IVA). During the sessions each child played Sony PlayStation games with an active sensor placed at FZ. The children were randomized into two groups. Group 1 received 20 sessions playing video games while brainwave activity was monitored and then received 20 sessions of brainwave-modulated video games. Group 2 received treatment in the opposite order.
Results: Results are based on 53 seven- to 11-year-old children completing mid-point and 42 children completing post-assessments. All data were normalized. Statistical analyses was the ANOVA of crossover trials with baseline run-in observations. This function evaluated effectiveness of intervention vs placebo, treatment effects, and period effects. For the children on the IVA, Response Control Quotients showed Effectiveness of Intervention (p < .0001) and Treatment Effect (p < .0001). On the IVA Attention Quotient there was significant Effectiveness of Intervention (p = .0004) and Treatment Effect (p = .0004). On the Connor's parent DSM-IV Inattentive scale, there was a significant Effectiveness of Intervention (p = .0002), Treatment Effect (p < .0001), and Period Effect (p < .0001). Connor's teacher DSM-IV Inattentive scale results indicated significant Effectiveness of Intervention (p = .05) and Treatment Effect (p = .046). The DSM-IV Hyperactive-Impulsive scale showed Effectiveness of Intervention (p < .0001) and Treatment Effect (p < .0001).
Conclusions: In summary, the analyses indicated that the attention training via EEG biofeedback was effective not only as a treatment, but effective when compared to an active placebo. More specifically, children improved on responding and attention on a computerized attention task (IVA); parents indicated improved DSM-IV inattentive symptoms; and teachers reported improved DSM-IV inattentive and hyperactive-impulsive symptoms. Additional findings and implications of this research will be discussed.
Affiliation: ADD Centres Ltd., Biofeedback Institute of Toronto
Background: This paper reviews our EEG, QEEG and LORETA findings which correspond to known functional neuroanatomical sites associated with the symptoms of Asperger's. We show how we use that information to guide successful neurofeedback interventions. Theoretical formulations arise from work with more than 1000 autistic spectrum clients and EEG neurofeedback analysis and results with more than 150 of these clients.
Methods: Before NFB we used medications, behavior therapies, psychotherapies, education and speech therapies (Thompson and Havelkova 1982). When neurofeedback became available clients were trained to decrease slow wave activity and beta spindling and raise 13-15 Hz often at FCz. Pre and post testing (after 40 sessions) used the WISC, WRAT, TOVA, IVA, questionnaires and QEEG changes. A selection of these clients have had 19 channel EEG assessments and analysis using LORETA.
Results: 126 Asperger's clients reviewed showed significant improvement (P < .001) on all measures. Brain maps were done on a sample of the clients. These revealed low activation P4 T6 in Asperger's cases and decreased activation Cz and Pz in the Autistic group. LORETA showed anterior cingulate Brodmann area 24 dysfunction in Asperger's and dysfunction in the posterior cingulate Brodmann area 31 and in the cuneus and precuneus in the autistic cases.
Conclusions: Findings to date suggest an axis of disturbed functioning (anterior and posterior cingulate Brodmann areas 24, 31) often with high amplitude 3-10 Hz and spindling beta (> 20 Hz) and low amplitude 13-18 Hz usually at FCz (LORETA abnormalities seen in the cingulate, medial and orbital frontal and/or prefrontal cortex). Similar findings are seen with anxiety disorders. Asperger's is unique in also having right temporal-parietal cortex inactivity corresponding to sensory aprosodia. Beta spindling suggests that the cortex is irritable, unstable and/or easily kindled. Increasing sensorimotor rhythm (SMR) using neurofeedback has a stabilizing effect (Sterman 2000).
Affiliation: Institute of the Human Brain of Russian Academy of Sciences, St. Petersburg, Russia. Norwegian University for Science and Technology, Trondheim, Norway.
Introduction: This paper presents a methodological approach developed in the Institute of the Human Brain for assessment electrophysiological indexes of executive functions of ADHD population.
Methods: 19-channel EEG was recorded in eyes open, eyes closed and different task conditions in 250 normal children (age 7-17) and 300 ADHD children of the same age (the data were collected and analyzed together with Andreas Muller and Valery Ponomarev). The tasks included 1) GO/NOGO task for assessment motor and executive functions, 2) Math task for assessment abstract reasoning, 3) Reading task for assessment reading and speech understanding, 4) Acoustic tasks (including auditory oddball) for assessment auditory information processing. Artifacts were corrected by spatial filtration based on artifact topographies extracted by means of independent component analysis (ICA). Absolute and relative power and amplitude spectra, spectral ratios, spectral asymmetries, coherence, phase are normalized for various age groups and were used for calculation z-scores for each of above mentioned parameters in each recording condition for a tested subject. ICA method is applied for separation independent ERP components associated with separate psychological operations, such as early stages of visual and auditory processing, engagement operations, comparison and monitoring operations and some others. For mapping EEG parameters into Talairach space LORETA and s-LORETA are used.
Results: The paper discusses the results of application of the data base for separating ADHD sub-types out of an ADHD group. Application of EEG spectrograms, ERPs and ERDs for constructing individual protocols of neurotherapy (neurofeedback n transcranial direct current stimulation) will be presented.
Conclusion: This study is the first to present the normative QEEG data base that include not only conventional spectral parameters of EEG but ERPs independent components as indexes of sensory-motor and cognitive functions.
Affiliation: (1) Imperial College Medical Faculty. (2) Goldsmiths College. (3) Northwick Park Hospital. (4) University of Westminster.
Growing controlled evidence suggests that EEG-neurofeedback (NF) training may be a valuable adjunct or even alternative to medication for attention deficit hyperactivity disorder (ADHD). In setting out to contribute to this evidence, we randomly assigned children to neurofeedback or a computerised attention training task (CT), a necessary control condition that up until now has been lacking in the field. Children between 8 and 12 years were diagnosed by a consultant psychiatrist and received a DSM-IV subtype, which informed the training protocol for those receiving neurofeedback. Exclusion criteria included psychiatric, behavioural or neurological co-morbidities, and medication other than methylphenidate. There were 23 participants (2 girls) divided into 4 groups: NF/unmedicated N = 4, NF/medicated N = 6, CT/unmedicated N = 6, CT/medicated N = 7. Pre and post-training assessment consisted of the arithmetic scale from the Wide Ranging Aptitude Test (WRAT), the Suffolk Reading Scale, the Strengths and Difficulties Questionnaire (SDQ) for parents, the Attention Network Test (ANT), the Test of Variables of Attention (TOVA) and a QEEG (to be analysed - also conducted midway along with the TOVA). The NF protocol depended on the subtype: Combined ADHD 15 sessions of enhance SMR, inhibit theta and 15 sessions enhance beta 1, inhibit theta; Hyperactive 30 sessions SMR/theta; Inattentive 30 sessions beta 1 /theta. Whereas both groups improved behaviourally on the stress and hyperactivity subscales of the SDQ, the two cognitive attention tests showed some advantages to NF. On the TOVA, omission errors initially decreased with NF but increased with CT (p<0.02), while on the ANT, the NF group improved in accuracy whereas the CT group did not change (p<0.017). The implications of the results will be discussed.
Affiliation: Department of Psychology, University of Southampton.
A prospective controlled trial was conducted to compare the efficacy of an analgesic cream (eutectic mixture of local anesthetics, or EMLA) with a combination of EMLA with hypnosis in the relief of lumbar puncture-induced pain and anxiety in 45 pediatric cancer patients (age 6-16 years). The study also explored whether young patients can be taught and can use hypnosis independently as well as whether the therapeutic benefit depends on hypnotizability. Patients were randomized to 1 of 3 groups: local anesthetic, local anesthetic plus hypnosis, and local anesthetic plus attention. Results confirmed that patients in the local anesthetic plus hypnosis group reported less anticipatory anxiety and less procedure related pain and anxiety and that they were rated as demonstrating less behavioral distress during the procedure. The level of hypnotizability was significantly associated with the magnitude of treatment benefit, and this benefit was maintained when patients used hypnosis independently.
Affiliation: (1) Living Health Center for Research and Education, Istanbul Turkey.
Background: According to DSM-IV, the essential feature of Mental Retardation is significantly subaverage general intellectual functioning that is accompanied by significant limitations in adaptive functioning in at least two of the following skill areas: 1-Communication 2- Self-care 3- Home living 4- Social/Interpersonal Skills 5- Use of Community resources 6- Self â"direction 7- Functional academic skills 8- Work 9-Leisure 10-Health and Safety. Neurofeedback is a learning strategy that enables persons to alter their brain waves.There are clinical evidences concerning the use of neurofeedback in brain regulation (Monastra,2000).At the Child and Adolescent Psychiatric Clinics of North America (2005) Hirshberg and colleques summarized the relevant research findings ,and examined the implications of EBF research for practice standards based on the guidelines for recommending evidence based treatment EBF as developed by the American Academy of Child and Adolescent Psychiatry for attention deficit hyperactivity disorder(ADHD).They concluded that EBF meets the 'Clinical Guidelines' standard for ADHD, seizure disorders, anxiety, depression, and traumatic brain injury. Many clinicians and researchers use this technique in ADHD (Monastra,2000) and GLDO(Othmer,1998) so while studying with the mentally retarded in order to increase attention, improve learning capacity we discovered something special that there is significant statistical improvement in speech and language problems. When treatment schedule and data of the 25 patients who are previously diagnosed as mentally retarded came to our center were examined we ascertain the effectiveness of neurofeedback in speech - language deficits. The problem levels of the elements (articulation, phonetic, comprehension, lexicon, semantic, syntax, pragmatic) which implies prelanguage and language development and also if stuttering was present in this group were established with the nonstandardized scales prepared by our center. Considering the etiology there were 8 patients with Down Syndrome (2 of them mosaic type) 1 patient with phenylketonuria the etiology of the 16 patients were not known. 5 of the patients had epilepsy.
Methods: We studied with 25 subjects ranging from 6-24 years old, previously diagnosed mentally retarded. Sessions were completed between 40-120 depending on the case.Five out of twentyfive were under medication related to epilepsy,the others were medication free. Evaluation measures include F.D.A approved QEEG-NxLink Normative databank (John, 1993),clinical observations and two different nonstandardized scales prepared by the center. Questions of the nonstandardized scales include the problem levels of the elements (articulation, phonetic, comprehension, lexicon, semantic, syntax, pragmatic) which implies prelanguage and language development and also if stuttering was present in this group were established. Neurofeedback trainings were performed by Lexicor Biolex software.QEEG signals were sampled at 128 Hz. Samples were normative neurometric analyzed in NXLink databank pre-post and every 40 sessions in order to establish the abnormal electrical activity of the brain.Electrodes were sited according to QEEG analysis at F7-T5,T3-T5,P3-T5 bipolar montaged with a 0-4 Hz delta down reward and inhibits at 4-8 Hz.CZ-C4 with a 12-15 Hz SMR reward and inhibits at 4-8 Hz.
Results: The evaluation results of 14 boys and 11 girls will be presented for this study. The mean age for this study group is 11.64 and standart deviation is 4.33. Pre and Post treatment comparison results were provided by student's t test in this group. (n= 25) (Compared to pre treatment significant improvement observed) For stuttering and pointing finger comparison results Wilcoxon was used. Number of subject is less than 5. 2 cases out of 25 showed no progress during and after treatment. Subcortex damage might be thought with these two cases. Many people underestimate the abilities of these patients and assume that only certain achievements can be accomplished. Neurofeedback can change this view of teachers,special educators, child psychiatrists and neurologists in future.Further study with controls and additional outcome measures is warranted.
Affiliation: France Telecom, R&D Laboratory TECH/IDEA/TIPS
Background: We couple the standardized low-resolution electromagnetic tomography (sLORETA), an inverse solution for electroencephalography (EEG) and the common spatial pattern, which is here conceived as a data-driven beamformer, to classify the benchmark BCI (Brain Computer Interface) competition 2003, data set IV. The data-set is from an experiment where a subject performed a self-paced left and right finger tapping task.
Methods: Available for analysis are 314 training trials whereas 100 unlabeled test trials have to be classified. The EEG data from 28 electrodes comprise the recording of the 500 ms before the actual finger movements, hence represents uniquely the left and right finger movement intention.
Results: Despite our use of an untrained classifier, and we extract only one attribute per class, our method yields accuracy similar to the winners of the competition for this data-set.
Conclusions: The distinct advantages of the approach presented here are the use of an untrained classifier and the processing speed, which make the method suitable for actual BCI applications. The proposed method is favourable over existing classification methods based on EEG inverse solution, which either rely on iterative algorithms for single-trial independent component analysis or on trained classifiers.
Affiliation: (1) Faculty of Medicine, University of Skopje, Macedonia, (2) Macedonian Academy of Sciences and Arts
Background: In a previous paper (Pop-Jordanova and Pop-Jordanov, 2005) the concept of spectrum-weighted EEG frequency (brain-rate) correlated to mental arousal has been introduced. The objective of this study is to investigate the viability of brain-rate for preliminary checking the general mental activation level, in addition to heart-rate, blood pressure and temperature as standard indicators of general bodily activation.
Methods: The first sample is consisted of 25 children, mean age 8.08 years, diagnosed as ADHD and treated by neurofeedback (SMR followed by theta/beta). The second group comprised 27 children, mean age 12.8 years, dwelling nearby ironworks and showing increased blood lead levels. Cognitive psychological tests and multimodal biofeedback assessments are compared with the brain-rate values calculated from four-band data measured in Cz. The corresponding mental arousal indices are also evaluated.
Results: For the first sample, the results of neurofeedback training showed an increase of brain-rate, simultaneously with the improvement in psychometric parameters. In the second group, a positive correlation between blood lead level and ADD is registered, both being inversely correlated with the brain-rate values.
Conclusions: The obtained preliminary results for brain-rate, confirmed to be consistent with clinical and psychometric assessments, support its use as a preliminary diagnostic indicator of general mental activation level. It is also reasonable to use the brain-rate, related to spectrum shift, as a general multi-band biofeedback parameter, in supplement to the more specific few-band biofeedback parameters. Further studies of diagnostic advantages and limitations of brain-rate concept are needed.
Affiliation: BrainMaster Technologies, Inc.
Background: A method is introduced that provides specific training for synchronous EEG events, based upon a simple, yet powerful signal processing model. This method reveals activity of the type reported by Freeman et. al. (2003), Schack et. al. (2002), Von Stein and Sarnthein (2000), and Lutz et. al. (2004).
Methods: Channel recombination (addition and subtraction) is used to differentially emphasize both synchronous, and asynchronous, EEG activity, in different derived signals. This method combines the benefits of differential training with those of synchrony training (Fehmi and Sundor, 1989). Joint Time-Frequency Analysis provides a further method for distinguishing detected events in both time and frequency.
Results: It is possible to visualize a single brief transient in any band of interest, including gamma, using this method. Data will be shown illustrating the phase and frequency selectivity of this method, and quantifying its ability to discern synchronous vs. asynchronous activity.
Conclusions: This approach is useful for assessment and visualization, as well as for training EEG rhythms. It is immediately applicable for real-time processing, and can be implemented on a wide range of systems. Results will be presented illustrating the ability to graphically depict rare and brief EEG transients, as well as the ability to train such rhythms for neurofeedback applications
Affiliation: University of Salzburg
Background: The circular mean resultant length (also called MRL, PLI, PLV etc.) has been used extensively in the past years to analyze oscillatory properties within an EEG signal. The calculation of such methods is based on circular statistics of time/frequency transformed EEG epochs. The variability of the phase locking measure decreases with increasing number of trials, and for many experimental setups, it is crucial to know the critical number of trials required to get a reliable phase locking measure. Thus, we investigate the statistical properties of the MRL, its variance and mean squared error in order to find unbiased estimates.
Methods: The MRL is a measure for the phase coherency of a circular random variable. Commonly, the estimation of the MRL is based on a random sample. However, we concentrate on its properties as an estimator, rather than just the descriptive measure. The MRL is a biased but consistent estimator. In order to calculate the value of the non-squared MRL, we apply a simulation for some special distributions to see how the expected values of the estimators behave for varying sample sizes.
Results: The conventional MRL overestimates the correct value while the modified version underestimates it. Especially for small MRLs the modified estimator is better. But also for greater MRL it can beat the conventional one. However, the conventional MRL seems to be more robust for discrete or "jumpy" distributions. Both measures are estimates of the real MRL and it is the variance and the mean squared error, which are the indicators to the critical number of trials required to get a reliable estimator.
Conclusions: For many experimental setups the number of trials available for analysis is rather small. In such cases it is important to analyze the properties of measures such as the MRL before statistical evaluation between groups.
Affiliation: (1) Biomedical Engineering Centre, Tallinn University of Technology
(2) Institute of Cybernetics at TUT.
Background: The EEG signal has stochastic character and high fluctuations of the signal makes difficult to detect small changes in the signal - these are masked by natural variability. The aim of this study is to develop methods sensitive to small hidden changes in time variability and energy of the EEG signal.
Methods: Different methods for EEG analysis were evaluated on the same database. Database for evaluation of methods consists of resting EEG recordings of 15 subjects without and with low-level microwave exposure (450 MHz, 0.16 mw/cm2, modulation 40 Hz). Different traditional methods as power spectral density (PSD), correlation, bispectral, fractal dimension, were applied for EEG analysis. Two new methods were developed. Nonlinear method of scaling analysis of the EEG signal based on the length distribution of low variability periods (LDLVP) has been developed and successfully applied for heart rate variability analysis. LDLVP method provides a simple route to detecting the multifractal characteristics of a time-series and yields somewhat better temporal resolution than the traditional multifractal analysis. The method was adapted for EEG analysis (Bachmann et al. Med. Biol. Eng. Comp., 43, 142-149, 2005). The method of modulation with further integration of energy from signal segments is widely used in radio engineering. This method was adapted for EEG analysis with integration of relative changes in energy of the EEG segments with and without microwave exposure.
Results: Analysis of the recorded signals by traditional methods (PSD, coherence, bispectral, fractal dimension) did not reveal significant changes between signals with and without exposure for any subject. The significant differences between recordings with and without exposure were detected by the LDLVP method for 3 subjects and energy integration method for 2 subjects within the group.
Conclusions: The results show that small changes in the EEG signals hidden for visual inspections can be detected by the LDLVP and energy integration methods.
Affiliation: Institute of Medical Psychology and Behavioral Neurobiology, University of Tübingen.
A series of experiments using FMRI feedback from motor cortex, amygdala, anterior insula and frontal areas are reported: self control of local BOLD is possible within 2-10 sessions and strong and area specific behavioural consequences are measured.
Affiliation: (1) Swansea University, UK, (2) Vrije Universiteit Brussels, Belgium.
Background: Pioneering research in cats has demonstrated that SMR feedback training results in a higher sleep spindle density and less sleep stage changes. This beneficial effect has been confirmed in insomnia patients to some extent. The present study investigates whether EEG neurofeedback training has a significant influence on sleep quality in healthy sleepers.
Methods: A blinded randomized placebo-controlled study was performed. After diagnostic workup, twenty healthy subjects (11M/9F) participated (mean age = 24.1 SD 3.1), randomized over three conditions: SMR (12-15 Hz), alpha/theta feedback, and sham control with non-contingent feedback. Both the EEG recorded during wake time and the sleep EEG was recorded from C3. This within-subject design was counterbalanced using a 1 week interval. Two consecutive EEG-feedback sessions of 30 min. each were carried out the evening prior to sleep recording. Sleep was blindly scored by neural network software, measuring sleep latency, sleep efficiency, wake after sleep onset, slow wave sleep, number of arousals, number of micro-arousals, and micro-arousal index.
Results: SMR training significantly reduced sleep latency (19.9 vs. 29.8 min.) and the wake time after sleep onset (56.3 vs. 80.4 min.). As a result, sleep efficiency also improved (87.2 vs. 81.3 %). No other differences were found between SMR or alpha/theta feedback versus sham control.
Conclusions: Two sessions of SMR training versus placebo-control in normal sleepers significantly improved sleep latency, wake after sleep onset, and sleep efficiency.
Affiliation: ADD Centres Ltd., Biofeedback Institute of Toronto
Background: Stress induced anxiety and emotional intensity may correlate with ruminating and can correspond to bursts of high frequency beta activity or beta spindling (23-35 Hz) with corresponding low amplitude SMR and/or low high frequency alpha activity at FCz. LORETA often showed the origin of this activity to be anterior cingulate Brodmann area 24. These
EEG changes corresponded to psychophysiological findings including: poor RSA, HRV, shallow, frequent, irregular respiration, increased heart rate, drop in skin temperature, rise in EDR and an increase in muscle tension. This review quantified EEG changes that could stand as a clinical marker.
Methods: Clients are asked to experiment with feedback to find how their mental states correlate with specific EEG frequency band amplitudes and to find how these correlate with psychophysiological variable changes. Clients are trained to self-regulate those variables that correspond to feelings of anxiety, tension and stress. They are taught how to generalize control of those variables that correlate with stress to their every day living. A chart review was carried out on165 consecutive clients who presented to one room of the Biofeedback Institute to ascertain which EEG variables correlated with stress and anxiety.
Results: All of the clients who complained of tension and anxiety with even minor stress demonstrated high frequency beta 23-35 Hz / SMR 13-15 Hz and/ or 19-22 Hz / 11-12 Hz ratios that were higher than controls. All reported much improved control with training.
Conclusions: Symptoms and signs, combined with the EEG and stress assessment findings, and knowledge of neuroanatomy and neurophysiology, can lead to an appropriate and specific NFB & BFB intervention.
Affiliation: (1) Goldsmiths College, University of London. (2) John Moores University Liverpool. (3) Imperial College, London. (4) University of Nottingham.
Hypnosis has been shown to be efficacious in a range of clinical conditions, including the management of chronic pain. However, not all individuals are able to enter a hypnotic state, thereby limiting the clinical utility of this technique. We sought to determine whether hypnotic susceptibility could be increased using three methods thought to facilitate relaxation, with particular interest in an EEG neurofeedback protocol which elevated the theta to alpha ratio. This was compared with progressive muscle relaxation and self-hypnosis. Ten subjects with moderate levels of susceptibility (2-7/12) were randomly assigned to each condition and assessed for hypnotic susceptibility prior to and upon completion of ten sessions of training. Hypnotic susceptibility increased post training in all groups, providing further evidence that operant control over the theta/alpha ratio is possible, but contrary to our predictions, elevation of the theta/alpha ratio proved no more successful than the other interventions. Nonetheless, all three techniques successfully enhanced hypnotic susceptibility in over half of the participants (17/30), a similar incidence to that reported using other methods. As previously reported, the majority who were not susceptible to modification were at the lower levels of susceptibility, and the greater increases tended to occur in the more susceptible subjects. However, here enhancement was disclosed in some at low levels, and capability was found of reaching high levels, both features not typically reported. Further research is warranted.
Affiliation: (1) Human Neuropsychology Laboratory, School of Psychology, Department of Experimental Psychology, University of Seville, Spain., (2) Center for Brain Injury Rehabilitation (CRECER), Seville, Spain., (3) School of Biomedical Engineering, Science and Health Systems, Drexel. University, Philadelphia, USA., (4) Applied Mathematics Department, School of Optics, Universidad Complutense de Madrid, Spain
Background: Most neuroimaging studies on emotion tend to focus on signal changes that occur during exposure to stimuli. Few, if any, focus on exposure following stimulus cessation. Prefrontal cortex (PFC) was chosen as our region of interest primarily because research links it to emotions and feelings. We hypothesized that changes in cerebral blood oxygenation (CBO) would not return to baseline after emotion-eliciting stimulus cessation, but would show either overshoot (continued activation) or undershoot (decreased activation)
Methods: We introduce a new paradigm in the study of emotional processes through functional neuroimaging. We studied the evoked-CBO changes in PFC during direct exposure to different emotion-eliciting stimuli ('on' period) and during the period directly following stimulus cessation ('off' period) using functional near-infrared spectroscopy (fNIRS). Thirty healthy participants were studied. We selected 9 stimuli consisting of video clips with different emotional content, rated by subjects into valence and arousal categories. The total sample consisted of 270 trials
Results: Results show more robust activation within PFC during the 'off' period than during the 'on' period. Data shows that clips rated as arousing had a higher degree of activation during the "off" period. The unpleasant valence category was associated with higher oxy-Hb concentration in right inferior PFC, while arousal categories were not related to activation in any specific area. Gender differences in oxy-Hb concentration related to emotional processing are also presented.
Conclusions: Given that PFC activation is related to affective dimensions, we suggest that subjective valence is related to cognitive aspects of emotion and subjective arousal to bioregulatory processes. This fNIRS study provides the first evidence directly linking emotional dimensions to dorsolateral prefrontal activity. For clinicians, our data suggests that alterations in the hemodynamic course of emotional neural substrate may lead to neuropsychiatric disorders.
Affiliation: (1) Human Neuropsychology Laboratory, School of Psychology, Department of Experimental Psychology, University of Seville, Spain
(2) Center for Brain Injury Rehabilitation (CRECER), Seville, Spain
(3) School of Biomedical Engineering, Science and Health Systems, Drexel University, Philadelphia, USA
(4) Applied Mathematics Department, School of Optics, Universidad Complutense de Madrid, Spain
Background: Religious behaviour has become an important issue in the social and scientific forum. Nevertheless, the question of whether religious experience is a cultural, a psychological or a neural phenomenon remains open to debate. There are a number of functional neuroimaging studies that have implicated prefrontal cortex in the generation of religious feelings.
Methods: Functional near-infrared spectroscopy (fNIRS) was used to measure activation (as measured by oxygenated hemoglobin concentration levels) in the prefrontal cortex (PFC) of subjects during direct viewing of a subject-pertinent religious image, and immediately following the viewing.
Results: The hemodynamic course of religious condition differed from non-religious condition, showing a higher oxy-Hb concentration even after cessation of the stimulation. Gender-specific activation patterns were apparent, with men showing higher activation levels than women.
Conclusions: Results assign an important role to PFC in the temporal integration of conscious religious experience and suggest that the DLPFC is a critical element in the cortical-limbic network involved in proactive religious feelings.
Affiliation: (1) Human Neuropsychology Laboratory, School of Psychology, Department of Experimental Psychology, University of Seville, Spain. (2) Center for Brain Injury Rehabilitation (CRECER), Seville, Spain. (3) School of Biomedical Engineering, Science and Health Systems, Drexel University, Philadelphia, USA. (4) Applied Mathematics Department, School of Optics, Universidad Complutense de Madrid, Spain
Background: A fundamental question in human sexuality is regarding the cerebral hemodynamic course of sexual arousal self-regulation. Lesion and neuroimaging studies suggest that dorsolateral prefrontal cortex (DLPFC) plays an important role in regulating the processing of visual sexual stimulation.
Methods: fNIRS was used to image the evoked cerebral blood oxygenation (CBO) response in 15 male and 15 female subjects. Using a new paradigm, we measured the relative oxygenated hemoglobin concentrations in DLPFC while subjects were directly exposed to selected stimuli ('on' periods) and immediately following stimuli cessation ('off' periods). The stimuli were two film clips, the first being an erotic scene showing a Roman orgy, and the second, deemed as neutral, showing a man walking in a crowd.
Results: Neutral stimulus viewing produced no PFC activation in subjects and post-stimulus viewing produced a fall in oxygenated hemoglobin. During exposure to the erotic stimulus, we observed rapidly ascendant overshoot in the DLPFC, which became even more pronounced following stimulus cessation. Furthermore, we found evidence indicating that men experience greater and more rapid sexual arousal than do women when exposed to erotic stimuli.
Conclusions: Our results show that self-regulation of DLPFC activation is a process modulated by subjective arousal and that cognitive appraisal of the sexual stimulus (valence) plays a secondary role in this regulation. Although further study is required to identify factors and interactions underlying the differences found in degree of arousal, this method can be used to study the hemodynamic course of sexual behaviour in patients with neuropsychiatric disorders.
Affiliation: (1) Department of Psychology, Applied Neuropsychology, University of Graz, (2) BCI-Lab, Institute for Knowledge Discovery, Graz University of Technology
Brain Computer Interfaces (BCI) provide users with an alternative output channel other than the normal output path of the brain, i.e. the efferent nervous system and muscles. The purpose of a BCI is to detect physiological signals from the brain, typically electrical signals resulting from neural firing and by using an algorithm to translate this signal in order to control an output device. Most of the current BCIs, which rely on the brain's electrical activity, use surface EEG signals or implanted electrodes. The Graz BCI for example, is based on event-related changes - ERD (Event-Related-Desynchronization) and ERS (Event-Related-Synchronization) - of mu and beta rhythms. It aims at distinguishing between the EEG associated with imagination of different simple motor actions, such as right or left hand or foot movement, and thereby enabling the user to control several devices (Pfurtscheller & Neuper, 2001; Neuper et al., 2003).
The main aim of the current research focuses on the potential of Near-Infrared-Spectroscopy (NIRS) as a novel signal acquisition tool for BCI development. The optical response denoting functional brain activation can be used as an alternative to electrical signals, with the intention of a more practical, user-friendly BCI. The presented optical BCI was implemented with the same logic as EEG-based BCI systems, but focusing on the concentration changes of deoxygenated and oxygenated blood flow during motor imagery. First results of near-infrared-spectroscopy on motor imagery will be presented and discussed in the context of its suitability for next generation brain-computer-interfaces.
Affiliation: Assisted Learning Centre Starnberg
Methods: Four Clients, with widely varying backgrounds, will be presented who were referred because of severe academic difficulties. Three were in danger of being asked to leave the educational establishment. Traditional educational methods had failed and we were asked to help with academic rehabilitation. Clients were pre-tested with various tools including
1) Traditional standardised
2) QEEG / IVA
3) Parent interview
Customised training protocols, based on the results of the specialised testing were provided for each client, and were implemented over an average of 12 months. Each client used computer assisted technology. All Protocols included use of QEEG based Neurofeedback and / or Fast
ForWord. (Fast ForWord is a computer based cognitive / language training program. fMRI studies show that is enhances performance in the left temporal lobes.) Clients were re-tested after 12 months
Results: All four clients showed excellent progress as measured by
a) Parent report
b) Standardised testing
c) Visible changes in the QEEG
d) Enhanced IVA score
e) School success
All four students were able to continue with their academic careers the next school year.
Conclusions: Academic rehabilitation is greatly enhanced by the judicious use of technology. Timely referral to outside specialists is crucial when a school is unable to provide the necessary services.
There are a plethora of claims regarding the use of neurofeedback for use in performance enhancement training. Invariably the aim of such training is for an individual to complete a specific function or task with fewer errors and/or greater efficiency. This research reviews the evidence from neurofeedback training studies to enhance performance in a particular area. Previous research has documented associations between specific cortical states and optimum levels of performance in a range of tasks. This information provides a plausible rationale for the use of neurofeedback to train individuals to enhance their performance. An examination of the literature revealed that neurofeedback training has been utilised to enhance performance from three main areas; sport, cognitive and artistic performance. The review examines evidence from neurofeedback training studies within each of these three areas. Some suggestive findings have been reported with regard to the use of neurofeedback training to enhance performance. However, due to a range of methodological limitations and a general failure to elicit unambiguous changes in baseline EEG activity, a clear association between neurofeedback training and enhanced performance has yet to be established.
Background: Neurofeedback represents a technique whereby the individual can work to gain some control over the electrical activity of their brain. Research that shows clear associations between particular cognitive states and specific patterns of cortical activity provides a rationale for the use of such a technique to influence cognition. Based on research showing that those who score higher on imagery tasks exhibit higher alpha amplitude (Williams et al, 1995) we examined whether training individuals to enhance the amplitude of their alpha rhythm, via neurofeedback, would benefit performance on a mental rotation task.
Methods: Ten participants were recruited and randomly allocated to either a neurofeedback training group or a non-contingent control group. Both groups completed a standard mental rotation task (after Cooper & Shepard, 1970) prior to and post the neurofeedback training. Those in the neurofeedback training group completed ten 10-minute sessions aimed at enhancing the amplitude of alpha (8-12Hz) while simultaneously inhibiting theta (4-8Hz) and beta (13-20Hz).
Results: The behavioural data showed that only those undertaking the neurofeedback training exhibited improved response times. However, there was only very limited evidence of any change in alpha amplitude.
Conclusions: As such this provides only partial support for the notion that neurofeedback training of alpha can influence cognition.
Affiliation: Ben-Gurion University of the Negev, Israel.
Background: In this work we propose the new methodology for biofeedback system which is based on biochemical reactions discrete chaotic dynamics theory (DCD). DCD based mathematical modeling of the networks of interconnected neurons (brain activity) resulted to the variety of internal neuron\'s constituents\' concentrations oscillatory regimes. Integrated within the neuron\'s networks oscillations could be related to the experimentally observed EEG signals. In the same time DCD model allows to present instant network\'s distributed neuron\'s states (amplitudes of the neuron\'s oscillatory regime at time t) in a form of 2D colored patterns. It was shown that the simulation of neuron\'s networks activity could result to the variety of oscillatory regimes including chaotic ones and to the variety of corresponded to these oscillations 2D patterns including symmetrical ones (mandalas).
Methods: In this work we intend to correlate oscillatory neuron\'s networks regimes with the experimentally received EEG while the corresponded 2D symmetrical patterns in a form of colored mandalas will be used as visual stimuli for biofeedback system. The idea to use mandalas as visual stimuli based on Jung\'s observation that this kind of images, being created spontaneously by an individual, has a therapeutic effect on their Authors. We propose to use the patient\'s EEG data as an input of DCD mathematical model for transformation of received signals into mandala images which are somehow should reflect personal ongoing brain activity.
Results: Different oscillatory regimes and corresponded mandala images generated by the DCD mathematical models will be demonstrated. Mathematically modeled signals and mandalas will be compared with the experimentally observed EEG signals and mandala images created by Jung\'s patients. The prototype of future biofeedback system exploring proposed method will be presented and discussed.
Conclusions: Based on the presented results we can expect that biofeedback systems based on the proposed methodology will enhance therapeutic impact on the patient.
Affiliation: (1) Vrije Universiteit Brussel, Belgium, (2) Universitair Centrum Sint-Jozef, Catholic University Louvain, Belgium.
Background: Insomnia is a frequent symptom in psychiatric disorders. Moreover, psychiatric disorders are often characterized by specific wake EEG characteristics, and insomnia appears to be characterized by specific polysomnographic characteristics. In this pilot study we explored the effects of neurofeedback on comorbid insomnia in 4 chronic schizophrenic patients.
Methods: Four chronic schizophrenia/schizoaffective inpatients received a combination of sensorimotor rhythm (SMR: 12-15 Hz) and alpha (8-12 Hz) training and five inpatients served as a control group. No major changes were made in medication. They received five sessions a week of approximately 30 minutes for a total of eight weeks. To achieve optimum effect this standard protocol was adjusted to the individual EEG characteristics of every participant after two weeks. Sleep was measured using Biosomnia Plus (Oxford Biosignals Ltd. Oxford) pre and post treatment. Latency to Persistent Sleep (LPS), Wake After Sleep Onset (WASO), Total Sleep Time (TST), Sleep efficiency (SE) were our primary outcome variables.
Results: Concentration problems resulted in a lack of attention for the presented feedback halfway every session. One patient of the experimental group ended the study after 26 sessions. In the experimental group, two patients improved on TST and WASO, two improved on LPS and one showed an enhancement of SE. In the control group, two patients deteriorated on TST and WASO, two patients improved on WASO. One patient deteriorated on SE and one patient showed the same impairment in TST and SE at both measurements.
Conclusions: All four schizophrenia inpatients who participated in this pilot study showed sufficient motivation to participate in a neurofeedback protocol, but attentional problems indicate that session duration need to be shortened. The improvements in the sleep variables evaluated suggest a potential positive effect of neurofeedback on sleep quality, and indicate that future research in this field might be of interest.
Affiliation: (1) University of Porsmouth, UK. (2) University of Wales Swansea, UK. (3) University of Wales Swansea, UK
Background: According to Gray and McNaughton (2000), the Behavioural Inhibition System (BIS) functions as a general risk assessment system, charged with evaluating potential danger, and is specifically activated during goal-conflict (e.g., approach-avoidance conflict) in human beings, this is anxious rumination. However, in human beings, there have been few attempts to characterise the neural correlates of BIS activation. During rest, increased theta coherence in BIS sensitive individuals have been suggested to reflect limbic-cortical systems involved in the emotional regulation of cognitive functions, e.g., retrieval of affectively-laden memories. However, the use of a resting condition makes it difficult to validate if participants engage in goal-conflict processing. Therefore, characterising neural activity during goal-conflict processing (e.g., anxious rumination) is necessary and forms the principal aim of this experiment.
Methods: 60 participants (30 males) had EEG (128 channels), heart rate, skin conductance and pre and post-subjective ratings of mood recorded with eyes-closed during (a) personal rumination, (b) nominal rumination and (c) goal-conflict free cognitive processing periods of 2 minutes. The two rumination periods were counterbalanced and silent counting periods were used as baseline.
Results: Results will indicate (a) whether trait anxiety (i.e. BIS sensitivity) increases the type, severity and emotional impact of rumination and (b) if the severity of the experienced goal-conflict will induce widespread theta coherence and (c) whether theta coherence will be more prominent in high BIS participants.
Conclusions: Understanding the neural processes underlying rumination may lay the foundation for the development of neurofeedback protocols aimed at modifying the specific motivational neural processes underlying anxiety.
Affiliation: (1) Norwegian University of Science and Technology, Trondheim, Norway.
(2) Russian Academy of Sciences, St. Petersburg, Russia.
Background: A paradigm for assessment of episodic memory has been developed and adapted for testing with qEEG. Neurofeedback has previously been successfully applied to enhance attention and reduce impulsivity in children with ADHD. Even if promising it is still unclear whether neurofeedback can reliably enhance cognition in normal subjects. This study presents results that show differential neural activations during episodic memory load before and after neurofeedback training. Using qEEG to study neural activation during cognitive load to provide parameters for individually adapted neurofeedback protocols to enhance performance is our main objective.
Methods: Neural activity was recorded using qEEG (19 electrodes, 10-20 montage) in 22 healthy volunteers (aged 19-30, recruited from a student population) during an episodic memory task. The test protocol involved noun encoding and retention of list members or synonym distractors. The project was presented as an EEG pretest-posttest paradigm with intervening neurofeedback training. Subjects were randomly assigned to a waiting list condition, an attention trainer or an individually adapted temporal lobe beta trainer.
Results: Using qEEG analysis, we extracted ERP and ERD/ERS and constructed an averaged database. Main ERP components provided positional estimates using LORETA. Based on error percentage and response time, we extracted databases with better-than-average and worse-than-average performers. Differences in ERP power correlated with performance. Loci were brodmann area 22, bilaterally, and an area slightly anterior. Using wavelet ERD and ERS analysis, we found alpha desynchronization and beta synchronization in abovementioned areas. The group of good performers had notably stronger beta synchronization than the group of bad performers. Furthermore; individual ERD/ERS frequencies were used as a basis for individually adapted neurofeedback.
Conclusions: Based upon qEEG pretesting results, neurofeedback training is now in progress. ERP analysis has shown a correlation between localized ERP power and test performance. Results from qEEG posttests following individually adapted neurofeedback will be presented.
Affiliation: (1) Norwegian University of Science and Technology, Trondheim, Norway.
(2) Russian Academy of Sciences, St. Petersburg, Russia.
Background: Mental rotation is a transformational operation involved in several cognitive processes. Studies have shown that response time increases in a linear manner with angular differences. Neurofeedback has previously been successfully applied to enhance attention and reduce impulsivity in children with ADHD. Even if promising it is still unclear whether neurofeedback can reliably enhance cognition in normal subjects. This study presents differential neural activations during mental rotation before and after neurofeedback training. Using qEEG to study neural activation during cognitive load to provide parameters for individually adapted neurofeedback protocols to enhance performance is our main objective.
Methods: Neural activity activity was recorded using qEEG (19 electrodes, 10-20 montage) in 32 healthy volunteers (aged 19-37, recruited from a student population) during a mental rotation task. Test protocol involved comparing two 3D objects presented incrementally. The project was presented as an EEG pretest-posttest paradigm with intervening neurofeedback training. Subjects were randomly assigned to a waitinglist condition, an attention trainer or an individually adapted parietallobe betatrainer.
Results: Using qEEG analysis we extracted ERP and ERD/ERS and constructed an averaged database. Main ERP components provided positional estimates using LORETA. Based on error percentages and response times we extracted databases with better-than-average and worse-than-average performers. Differences in ERP power correlated with performance and with task difficulty. Loci were brodmann area 7, left and right side. Using wavelet ERD and ERS analysis, we found alpha desynchronization and beta synchronization in abovementioned areas. The group of good performers had notably stronger beta synchronization than the group of bad performers. Furthermore individual ERD/ERS frequencies were used as a basis for individually adapted neurofeedback.
Conclusions: Based upon qEEG pretesting results, neurofeedback training is now in progress. ERP analysis has shown a correlation between localized ERP power and test performance. Results from qEEG posttests following individually adapted neurofeedback will be presented.
Affiliation: (1) Department of Biomedical Engineering, Science and Research Center, Azad University. (2) Department of Medical Physics, Tarbiat Modares University. (3) Iranian Telecommunication Research Center (ITRC).
Background: Surface and needle electrodes can detect electromyogram signals, in which the principal components are Motor Unit Action Potentials (MUAPs). Electromyogram decomposition is an approach for studying neuro-muscular diseases. There are many methods
for this purpose, such that this subject is an active research area in the field of biomedical signal processing. Surface electromyogram (sEMG) decomposition has some benefits, but there are also some problems. In this paper, an intelligent system is designed and implemented, combining three channels of sEMG processed with conventional methods, Maximum Voluntary Contraction, the electrode position and the muscle anatomical knowledge, to obtain a better decomposition with respect to other methods.
Methods: This Method is divided into three parts: (1) Surface EMG Modeling to get the most reliable signal and use for the validation process, (2) Array Electrode Analysis for optimal electrode placement, and (3) Decomposition process that includes intelligent system.
Three conventional methods are used for channel decomposition of sEMG, each having its benefits and difficulties. Many parameters (such as thresholds) influence their performance and must be adjusted optimally. These methods are as follows: a-MINIMUM DISTANCE METHOD Because of MUAP variations, the minimum distance method can be viewed as
one of the bestdecomposition algorithms. b-UOFC METHOD In this method, the DB wavelet transform up to 10 levels is used. The feature vector is the energy (or standard deviation) of these levels. For the classification stage, a modified version of UOFC(Unsupervised Optimal Fuzzy (Clustering) is carried out. c -STBC METHOD The STBC algorithm is the k-means modified clustering algorithm. This algorithm uses shape and temporal information to achieve optimum cluster number. Because of the fusion function, the best structure for the intelligent
system is proposed as a fuzzy structure. The system can be seen divided into the following sections: 1-PREPROCESSING STAGE, 2-FUZZIFICATION STAGE, 3-FUZZY RULE STAGE, 4-FUSION & DEFUZZIFICATION STAGE and 5-FINAL FUSION STAGE.
To validate the system and compare its performance with those of the other studied methods, five MUAPs in a long time surface EMG were simulated and used.
Results: In the first preprocessing stage, only three MUAPs were detected and the other two appeared as noise. It is noticeable that the total performance of the decomposition system is
improved through using the intelligent system. One of the major problems in sEMG decomposition is the MUAPs overlap. The intelligent system overcomes this problem by an inverse mapping solution. In other words, there are some active segments in any channel, recognized as an outlier (because of overlap, they don\'t belong to any MUAPs). But in the other channel, these belong to the MUAPs. In this way, we can present a better approach with the minimum computational cost (compared to other methods), for MUAPs\'overlaps.
Conclusions: Experimental results show the superior performance of the intelligent system, which has solved the most important problem of decomposition, i.e. the MUAPs\' overlap.
Improvement in resolving superimposition comes about from using information from different spatial channels, and the proposed intelligent system presents an effective way to make use
of this information. In the other words, intelligent system can overcome superimposition problem by an inverse mapping solution. In such a way, we can present an appropriate solution with the minimum computational cost (with respect to other methods) for MUAPs\' overlap.
Affiliation: (1) Azad University,Science & Research Division,Tehran,Iran
(2) Tarbiat Modarres University,Tehran,Iran
Background: The amplitude of the H-reflex has been known to have considerable variability even if the intensity of electrical stimulation to the afferent nerves is held constant from quantity and modeling vision. Classical statistical measures investigate static properties of the
monosynaptic reflex variability. But some seemingly irregular biological signals are not random in time and have long-range time correlation. So analysis of the dynamic properties by fractal analysis method might provide new insights to the underlying mechanisms of the H-reflex variability after bi-polar and tri-polar TENS.
Methods: The effects of electrical stimulation were evaluated using extracted fractal component by Coarse Graining Spectral Analysis (CGSA) and Higuchi's algorithm to calculate the percentage of random fractal components, spectral exponent and fractal dimension on the sequences of H-reflex, M-wave, H/M, H-M and H+M amplitudes in soleus and gasterocnemius
muscles after bi-polar and tri-polar TENS. Four healthy non-athletic volunteer men between 20 to 40 years old participated in this study. They were tested in two groups and four steps.
Results: Using CGSA, mean of percent fractal, which shows random fractal in all time series were more than 70%. Also these sequences approximately have followed the type power spectrum, which described power-law decay. Mean of the spectral exponent was more than 0.6 for all sequences. So a powerful time correlation was found.
Conclusions: Changes in percent of fractal and spectral exponent values were the criterion to compare the data after bi-polar and tri-polar TENS application in each experimental group. Statistical analysis showed significant difference between bi-polar and tri-polar TENS in for H+M sequence (P=0.043) and H/M sequence (P=0.075) in soleus muscle. Significant difference in spectral exponent in H+M sequence of soleus muscle after bi-polar and tri-polar TENS showed the behaviour of a slow twitch muscle motor neuron pool in spinal cord after these two types of TENS.
Affiliation: (1) Department of Psychology, Goldsmiths College, University of London.
(2) Division of Neuroscience & Mental Health, Imperial College London. (3) Trinity College of Music.
This is a pilot for a larger study with the aim of extending with novice musicians the findings of Egner and Gruzelier (2003) with elite musicians. They demonstrated professionally significant gains in artistry in music performance following alpha/theta training, but not with SMR or beta 1training, nor with aerobic exercise or mental skills/rehearsal training or the Alexander technique. Here are presented the results of 12 novice singers from London music colleges who were randomly assigned in equal numbers to ten sessions over two months of alpha/theta (A/T) training or SMR training. The study and analysis are ongoing. Results are presented for pre and post training assessment of music performance, attention, memory, mood and processes associated with creativity. There was evidence of significant within and between session learning in increasing the theta/alpha ratio (p<0.001 & p<0.047), but not in elevating the SMR/theta ratio. Despite the latter limitation semantic cued memory increased in the SMR group in support of Vernon et al (2003) (p<0.049, one tailed). Otherwise there were several suggestive differential effects advantaging the A/T group over the SMR group in music performance, creativity and attention, the latter in the direction of the results of Egner and Gruzelier (2004) though not reaching significance in their study. The Test of Variables of Attention (TOVA) showed an increase in sensitivity (d') with A/T (p<0.04) and the reverse with SMR (Group x Time p<0.04), largely due to a reduction in omission errors with A/T and the opposite mean change in the SMR group (G x T, p<0.066). There was also a reduction in RT variability (p<0.012). Support for associations with creativity followed improvement in flexibility on the Guildford Alternative Uses Test (p<0.055), and the rule breaking subscale of the Adaptor/Innovator Test (p<0.022). The Baddeley Sentence Checking Test which involves working memory and reasoning was also advantaged (p<0.047) by A/T training. Finally blind lay evaluations from video clips of expressiveness, confidence and stage presence disclosed improvement following A/T (all p<0.001) in contrast to SMR training (G x T, all p<0.002).
Affiliation: (1) Department of Physiological Psychology, University of Salzburg, Austria. (2) Department of Sport Science and Kinesiology, University of Salzburg, Austria.
Background: The goal of this experiment was to analyze frontal midline theta in the shooter's aiming period to get a better understanding of the neural mechanisms associated with sustained, focused attention. Haufler et al., (2000) reported a more pronounced theta band activity during the pre shot phase in expert shooters. Besides a variety of cognitive functions as working memory (Onton, Delorme, & Makeig, 2001; Gevins et al.1997); short term or recognition memory (Burgess & Gruzelier 1997), frontal midline theta, is additionally related to sustained (Kubota et al. 2001) and internalized attention (Aftanas & Golocheikine, 2001), processes relevant in shooting.
Methods: EEG was recorded during the aiming period of eight expert shooters and ten novices. For every subject the 40 best shoots were selected for analysis. Power of the frontal midline theta has been calculated in 500 msec. time intervals for the 3 seconds, preceding the shot (except the last 500 msec.). In addition LORETA has been calculated to analyze differences in the source of the frontal midline theta signal.
Results: The results indicate a significantly stronger increase of frontal midline theta during the last second preceding the shot for the experts. Additionally differences in the source of the frontal midline theta signal have been found in the anterior cingulate cortex and the medial frontal areas.
Conclusions: The data clearly indicate that experienced marksmen exhibited a stronger activation of the frontal midline theta, thus, are able to better focus during the aiming process. Additionally the differences between experts and novices assessed by LORETA analyses support the view that frontal midline theta is generated by activation in medial frontal and anterior cingulated areas. A possible application of this finding might be by using neurofeedback for training in sportsmen that have to keep sustained attention for a long time interval.
Affiliation: Imperial College medical faculty, London. UK.
In a series of controlled studies we have examined the efficacy of self-hypnosis training for beneficial influences on the immune system, and importantly the translation to health including immune competence at exam time and a chronic viral illness (Gruzelier, 2002,a,b for review). Comparisons have also been made with a Japanese nontouch healing method, Johrei, which may be conceptualized as an energy healing intervention. Johrei was found to be more effective than a self-hypnosis/CBT package or a relaxation control in buffering the effects of the stress of medical school exams on the typical stress-induced decline in natural killer cell (NKC) counts. In fact counts significantly increased (p<0.001) with Johrei (Naito et al, 2003), and there were some subtle benefits to mood (Laidlaw et al, 2002). Here we extend these interventions to early stage HIV infection where there are proven declines in CD4 lymphocytes, deterioration of mood and electrophysiological anomalies (Baldeweg et al, etc). We found that after a 4 month intervention period, the decline in CD4 counts (ul/month) which best characterises untreated HIV infection was reversed in all Johrei patients (X = +17, +8 to +28, N = 10), leading to a difference (p<0.0.03) from the Hypnosis/CBT group (X = -9, -17 to -1, N = 15) and a difference (p<0.49) from the wait-list control group (X = -20 to -3, N = 13). There were no accompanying benefits to viral load, cortisol, DHEA-S, NKC, and a range of psychological questionnaires: state anxiety, perceived stress, locus of control, impact event scale, sleep quality. Attrition tended to be higher in the Johrei group. Examining the CD4 gradient to include 12 months before and 12 months from the intervention, and to include wait-list controls who were subsequently randomized to Hypnosis/CBT (N = 27) or Johrei (N = 19), and compared with matched controls blindly selected from the data base (N = 44), there was a weak Group x Gradient effect (p<0.095), with only the Johrei group showing a significant improvement (p<0.04). With the longer time estimate of 12 months, the CD4 decline was less manifest occurring in 27% controls, 20% Hypnosis and 7% Johrei. The HIV benefits, more apparent during the 4 month intervention period than over 12 months, are merely suggestive, but when considered with the exam stress study (Naito et al, 2003) are certainly worthy of further investigation.
Affiliation: (1) School of Physics, University of Sydney. New South Wales 2006, Australia.
(2) The Brain Dynamics Centre, Westmead Millennium Institute Westmead Hospital and Western Clinical School of the University of Sydney Westmead, New South Wales 2145, Australia. (3) Faculty of Medicine, University of Sydney New South Wales 2006, Australia.
Background: The physiological or clinical significance of quantitative measures of EEG activity is almost always assessed on the basis of post hoc correlations with independent measures. A new level of interpretation is added when EEG spectra can be predicted by a model describing the main physiological properties of the brain responsible for EEG generation.
Methods: A large-scale continuum model of EEG generation that averages properties over a few tenths of a millimetre, and incorporates excitatory and inhibitory neural populations in the cortex and thalamus, is used to relate empirical EEG spectra to underlying physiology. Parameters such as connection strengths between neural populations, average rise and decay rates of the cell body potential upon stimulation, and the average transmission time for signals between cortex and thalamus, are determined by fitting model predictions to empirical spectra. The stability of model parameters and qEEG measures (band powers, total power, alpha peak frequency, spectral entropy) within individuals is explored, and the ability of theoretical spectra to distinguish between healthy individuals is investigated and compared with that of empirical spectra.
Results: Model parameters are shown to have a reproducibility that is slightly lower than that of traditional qEEG measures, although this depends on the chosen parameterization. Decreasing the skew of parameter distributions leads to improved reproducibility. Parameters are determined that are among the most reliable indices of EEG activity.
Conclusions: These findings can be used to estimate the certainty with which clinical information can be derived from changes in qEEG measures or model parameters, and to determine the extent and nature of information on underlying physiology that can be derived using the current version of our model of brain electrical activity.
Affiliation: (1) Department of Physiological Psychology, University of Salzburg, Austria.
Background: A recent model established by Herrmann et al. (2004) tries to explain the functional importance of evoked as well as induced human 40Hz-gamma oscillations. Early phase-locked gamma might reflect a matching mechanism, whereas late non-phase-locked gamma represents a utilization process.
Methods: Subjects were shown real and scrambled objects while the EEG was recorded. EEG data processing focused on a combination of gamma power and ERP (event-related potentials) analysis. We investigated the P1 and N1 components, as well as a late component peaking at around 230ms. Additionally, we performed BESA source analysis to unravel the generators of the different ERP-components.
Results: ERPs showed a strong task-related difference in a component peaking at 230ms (C230). For this component, BESA analysis revealed three sources, two in the inferior temporal lobe of both hemispheres and one in a fronto-medial region. Induced gamma power was enhanced for objects compared to scrambled objects in the region of fronto-medial source within 200-300ms. This effect was preceded by enhanced evoked gamma activity for objects but not scrambled objects in an early time window of 100ms post-stimulus in all of the three sources.
Conclusions: Our findings are in line with the MUM model by Herrmann et al. (2004) which predicts that the 'utilization' of the encoded trace is reflected by induced gamma in a later time window, whereas evoked gamma reflects a matching process with long-term memory in an earlier time window. Our conclusion is that the matching process takes place in inferior temporal regions, and utilization in fronto-medial areas that probably represent the cingulate gyrus.
Affiliation: University of Groningen.
It is known that spectral features of the EEG itself and the mutual information between features and condition exhibit a clear time structure during the motor imagery used to control a BCI. Though using this time structure might boost performance, most BCIs use instantaneous classification. Among the few current approaches are time-dependent Neural Nets and Hidden-Markov Models (HMM). We used data from five subjects of which three were from the BCI Competition. We compared HMM performance to a base line of instantaneous classifiers (Linear model and kNN). We also created HMMs with- and without the ability to use the time structure. By comparing these variants of the HMM we describe the temporal nature of data from EEG-based BCI trials. A Flat HMM without time structure resembles a mixture model of Gaussians; a second variant discards information by estimating the observation distributions from all the data rather than from data per movement condition. Here, only the transition probabilities are estimated from disjoint subsets corresponding to conditions. With the limited number of subjects differences between the HMM and base line methods were too small to attain significance. Performance on real data also varied very little between the variants of the HMM. However, difference in performance was highly significant on artificial data with a known time structure. The small differences we found indicate that a HMM is not superior to an instantaneous classifier for this type of data. From the comparison of the performance of HMM variants on real data and on artificial data we conclude that the data is mostly linearly separable, and that the temporal structure in the EEG pattern does not appear to be informative. However, we expect that a modelling approach using HMMs may still have benefits, e.g. in rejecting non task-related brain states.
Affiliation: (1) Department of Psychology, Goldsmiths College, University of London.
(2) School of Medicine, Chang Gung University, Department of Physical Medicine and Rehabilitation, Chang Gung Memorial Hospital, Taiwan. (3) First Department of Neurology, Chang Gung Memorial Hospital.
Acupuncture therapy has been practiced in Chinese medicine for thousands of years using acupuncture needles to stimulate specific acu-points. Acupuncture therapy has also been applied to patients with stroke and traumatic brain injury (TBI), believing that patients may be more cognitively alert and motorically aroused in affected limbs. This follows evidence that for three decades studies have shown that early implementation of cognitive training and treatment strategies involving cognitive stimulation may boost cognitive function recoveries in TBI. Traditionally, the effectiveness of rehabilitation treatments is often assessed by behavioral evaluation, and index of activities of daily living (ADL). However, the results obtained from these assessments often lack significant correlations with treatment effectiveness. Because the resulting faster recovery may be due to axonal and collateral sprouting, and stimulation of the reticular system, by using a neurophysiological approach we hope to understand the correlation between brain cognitive function recovery and neuroplasticity in more detail. Here we report on pilot studies using transcutaneous acupuncture simulation on both the normal subjects and patients with traumatic brain injury showing increased electrophysiological activities (P300 to the subject's name) in the frontal lobes. The acu-points: Nei-Guan and Her-Gu (dorsal surface of the thumb joints) were selected because of associations with increased arousal and alertness. In patients the increase in P300 amplitudes could last for at least 30 minutes after the cessation of acupuncture stimulation. Rehabilitation treatments were implemented during the period of acupuncture stimulation. Results showed that the acupuncture group had more significant functional independent measures (FIM) score improvements than control patients. Formal controlled studies are planned at Goldsmiths College and Hommerton and Blackheath brain injury units.
Affiliation: (1) Institute of Biocybernetics and Biomedical Engineering PAS, Warsaw, Poland; (2) Institute of Physics, Maria Curie-Sklodowska University, Lublin, Poland
Background: Respiratory disturbances in stuttering can influence lung ventilation and gas exchange leading even to hyperventilation or hypoventilation. A new breath regulation method based on visual feedback (VF) with using on-line registered end-tidal CO2 was implemented in computer system for therapy of respiratory disturbances in stuttering. The VF therapy method was preliminary tested.
Methods: 24 stuttering persons and fluent speakers (aged 14-37) participated in the study. Sound and expired CO2 signals were recorded while speaking without and with VF to control breathing as well as during rest respiration, before each utterance. The factor of breath ergonomics during speaking (FE) and stuttering intensity were determined.
Results: The FE factors obtained for stuttering persons were significantly lower than those for fluent speakers (p<0.001). But, stuttering intensity during speaking with visual feedback was lower than during speaking without VF. The factor FE was significantly higher for speaking with VF than without VF (p<0.01), too. Besides, the factors FE obtained by stuttering persons during speaking with VF didn't differ statistically from that obtained for both kinds of fluent speakers utterances. Conclusions: ETCO2-based visual feedback method applied by stuttering persons to control breath during speaking decreased stuttering intensity and increased the ergonomics of breathing during speaking.
Affiliation: Psychological Institute, Clinical and Applied Psychology, University of Bonn.
Background: Stimulation of arterial baroreceptors has effects on neural activity unrelated to cardiac autonomic regulation. Baroreceptor stimulation has been reported to exert an inhibitory effect on the startle reflex, motor activity, electrocortical activity, pain sensitivity, and to affect the sleep-wakefulness-cycle. This study addressed the following: (1) Can we enhance baroreceptor reflex sensitivity (BRS) by breathing manoeuvres? (2) Is there a relationship between increased BRS and cortical processing of an aversive noise?
Methods: Twenty-five female normotensive students (age=25.6 ± 7.1) were exposed to an aversive white noise while performing three breathing techniques: (1) thoracal breathing, (2) abdominal breathing, (3) pressure breathing (expiration against slightly closed glottis and lips). Subjective measures of loudness, annoyance and task difficulty were assessed, in addition to ECG, beat-to-beat finger blood pressure (Finapres) and respiration. BRS was calculated non-invasively as the slope of the linear interrelationship between systolic blood pressure (SBP) and the following interbeat interval (IBI). Partial correlation of BRS and perceived annoyance was computed and corrected for perceived loudness.
Results: BRS differed significantly between the breathing patterns F(2,72(=7.01, p < 0.05). The correlation of annoyance and baroreceptor sensitivity - after loudness was partialed out - was statistically significant for abdominal breathing pattern (r = -.47, p < 0.05).
Discussion: It is possible to enhance BRS using a breathing technique known and widely used in most relaxation trainings as abdominal breathing. Subjects perceived an aversive noise as less annoying while performing abdominal breathing. The cortical inhibiting effect of this technique was probably mediated via an enhanced BRS. This study yields a rationale for the well known instruction for coping of acute stress and anger situations: breathe slowly in and slowly out.
Affiliation: Institute of the Human Brain of Russian Academy of Sciences, St. Petersburg, Russia. Norwegian University for Science and Technology, Trondheim, Norway
Transcranial direct current stimulation (tDCS) is a method of activation and inhibition of the cortex by injecting low amplitude electrical direct currents (DC) through electrodes placed on the scalp. Therapeutic currents are usually so small (between 200 mcA and 1000 mcA) that people can not discriminate those currents from placebo; however the effects are noticeable both in behavior and in EEG and PET correlates. For example, research shows that anodal (cathodal) currents induced by positive (negative) potentials activate (inhibit) cortical networks. These activation (inhibition) effects are expressed in changes in EEG, event related potentials (ERPs), positron emission tomography (PET) and responses to magnetic stimulation.
The lecture presents an overview of literature and the results of my own research in the field of transcranial direct current stimulation. In Russia at the Institute of Experimental medicine this field was initiated in 1970th by Prof. Natalia Bechtereva and her colleagues. Nowadays this technique is applying in the Institute of the Human Brain of Russian Academy of Sciences in clinical setting for treatment of ADHD children.
Background: The initial intent of the DSM was to provide diagnostic detail, and from diagnosis would follow proper treatment. This has not turned out to be the reality, with behavioural diagnostic proliferation, and treatment non-specificity reflecting clinical reality in many situations.
Methods: The linkage of genetics and behaviour has to take the intermediate step of the genetic expression. This intermediate step between genetics and behaviour is seen electrophysiologically as the phenotype.
Results: A listing of genetically linked EEG findings and other patterns commonly seen will be presented as a list of candidate phenotypes, with the responses to medication and to neurofeedback both indicated.
Conclusions: Behaviour alone doesn't predict proper therapeutic approach, though the electrophysiological phenotype allows powerful prediction of effective medication and neurotherapy approach.
Ph.D. (3), & Marco Congedo, Ph.D. (2, 4).
Affiliation: (1) Harold Abel School of Psychology, Capella University, Minneapolis, MN. (2) Nova Tech EEG, Inc, Knoxville, TN, (3) Department of Psychosocial and Community Health, University of Washington, Seattle, (4) National Institute for Research in Informatics and Automated Systems (INRIA), Grenoble, France.
Purpose of Study: Chronic fatigue syndrome (CFS) is an illness characterized by profound fatigue lasting at least 6 months accompanied by disturbances of sleep, mood, musculoskeletal pain, and other symptoms (Fukuda et al., 1994). Investigators have sought, mostly unsuccessfully, to characterize these complaints by linking them to specific changes on objective measures of brain function. EEG methods previously have been shown that CFS patients had higher theta power than healthy control subjects (Billiot et al., 1997). In another, spectral analysis suggested that physical activity-induced EEG signal changes might serve as physiological markers for CFS (Siemionow et al., 2004). Although not diagnostic, abnormalities were observed that distinguished the ill and healthy groups. Recent findings conducting a co-twin control study of monozygotic twins discordant for CFS, 21 pairs of twins examined with quantified EEG indicated that the magnitude of the delta and theta bands at Fz, Cz and Pz was significantly higher in CFS twins than their healthy co-twins (Budzyski, Budzyski, Fischer, Goldberg, Ashton, Buchwald, 2004).In the present analysis, we used 18 of the 21 same datasets to extend our findings using low-resolution electromagnetic tomography (LORETA) (Pascual-Marqui, 1995, 1999; Pascual-Marqui et al., 1994). LORETA is an inverse solution technique that estimates the distribution of electrical neuronal activity in 3-dimensional space. It has been extensively used in electrophysiological research (Isotani et al., 2001;Lubar, Congedo and Askew, 2003) and evaluated independently in several laboratories (Fuchs et al., 1999; Worrell et al. 2000; Winterer et al.,2001). Our objectives were to investigate if specific LORETA features or band parameters are associated with CFS, and thus, could differentiate monozygotic twins with CFS from their healthy co-twins.
Methods: Participants. Twins were recruited through the Chronic Fatigue Twin Registry. A more comprehensive description of the CFS twin registry can be found elsewhere (Buchwald, Herrell, Ashton, Belcourt, Schmaling, Goldberg, 1999). Participant Diagnosis. To determine diagnoses, the Diagnostic Interview Schedule, Version III-A (DIS) (Robins and Helzer, 1985), was administered via telephone interview to Registry participants. Twins were classified as meeting CFS criteria according to their responses to a CFS symptom checklist and the diagnoses generated by the Diagnostic Interview Schedule. The same inclusion and exclusion criteria and review processes were applied to the fatigued and healthy twins. Participant Selection. From the twin registry, 22 sets of monozygotic twins discordant for CFS were chosen for a 7-day in-person evaluation based on registry information and additional telephone screenings. Twins were required to 1) be at least 18 years of age; 2) be reared together; 3) be discordant for CFS (1 twin met the Centers for Disease Control and Prevention CFS criteria, the other was healthy); 4) discontinue alcohol, caffeine, and all medications known to affect sleep or cognition at least 2 weeks prior to evaluation; and 5) travel to Seattle together. One set of twins did not complete the EEG evaluation and the data from 3 others cold not be used in this analysis. Zygosity was confirmed by DNA analysis. Medical records for the past 5 years were reviewed independently by an experienced physician, a psychologist, and an infectious disease specialist to confirm medical status and CFS diagnosis.EEG Data Collection. EEGs on each twin pair were obtained back-to back on a Lexicor Neurosearch 24. Researchers were blinded to which twin had CFS. The EEG was sampled with 19 electrodes in the standard 10-20 International placement referenced to linked ears. Data were collected for 100 2-second epochs in each of 3 conditions: eyes open, eyes closed, and eyes closed while performing a mental mathematics task (serial 7s). Eyes closed task data were then transported into the Eureka3! (Nova Tech EEG, Inc.) software, plotted, and carefully inspected for manual artifact-rejection. Average cross-spectral matrices were computed for bands delta, theta, alpha1, alpha2, beta1, beta2, beta3, beta4, and beta5. EEG Data Collection. EEGs on each twin pair were obtained back-to back. Researchers were blinded to which twin had CFS. The EEG was sampled with 19 electrodes in the standard 10-20 International placement referenced to linked ears. Data were collected for 100 2-second epochs in each of 3 conditions: eyes open, eyes closed, and eyes closed while performing a mental mathematics task (serial 7s). Eyes closed task data were then transported into the Eureka3! software, plotted, and carefully inspected for manual artifact-rejection. Average cross-spectral matrices were computed for bands delta, theta, alpha1, alpha2, beta1, beta2, beta3, beta4, and beta5. LORETA Variables. For each twin, cross-spectral matrices were computed and averaged over 4-second epochs resulting in one cross-spectral matrix for each twin and for each of the discrete frequencies within each band. The LORETA-Key package (Pascual-Marqui, Michel, and Lehmann, 1999) was used to compute LORETA current density in the frequency domain directly from the average cross-spectral matrix (Frei, Gamma, Pascual-Marqui, Lehmann, Hell, and Vollenwider, 2001). The solution space is restricted to cortical gray matter using the digitized probability atlas of the Brain Imaging Center at the Montreal Neurological Institute (Collins, Neelin, Peters, and Evans, 1994), divided in 2394 voxels measuring 7 x 7 x 7 mm.Data Analysis. For each frequency band and subject, the current density modules at each voxel were smoothed with a 10.5 mm 3dimensional moving average filter, normalized, and finally log-transformed. To compare the current density power between the CFS and healthy twins, we used the randomization-permutation multiple comparison t-sum approach (Congedo, Finos, and Turkheimer, 2004). We performed one test for each of the 9 frequency band-pass regions. For the whole data set (2394 x 9 variables), voxel-by-voxel paired t-tests were computed. A threshold of significance was then computed by the t-sum method. For all bands, we tested the hypothesis that the mean LORETA current density power of the 2 groups differed by subtracting the values for each CFS twin from that of their healthy co-twin (Congedo, Finos, and Turkheimer, 2004).
Results: Two of the 13 frequency band parameters examined differed significantly between the CFS and healthy co-twins. All reported results are indicating the corrected p-value as p<0.05. As demonstrated in Figure 1, LORETA current source density in the delta (2-3.5 Hz) band was higher in the CFS twins than among the healthy twins in the left uncus and parahippocampal gyrus (Brodmann areas 28, 36, 38 and 20). Figure 2 illustrates that theta also was higher in the CFS group in the cingulate gyrus (Brodmann areas 24 and 32) and right precentral gyrus of the frontal lobe (Brodmann areas 6 and 8). The maximum t-statistic, or maximum t-value across the entire volume was 3.61 and 6.08 for delta and theta respectively.
Discussion: In conclusion, objective measures of theta and delta current source density distinguished affected and non-affected twins in this co-twin control study of monozygotic twins discordant for CFS. One aspect of this study that deserves emphasis is the fact that the participants were carefully chosen illness discordant twins. Of additional relevance, in a study of older adults, 11% of "normal" control subjects exhibited an abnormal EEG (Leuchter et al., 1993), underscoring the value of exceedingly well-matched controls such as the ones we used in this study. Because twin studies are especially well suited to the study of illnesses for which the appropriate comparison groups are not clearly defined, they offer an alternative approach to examine possible patterns of brain activity in CFS. Future research should focus on larger groups of patients and control subjects. If alterations in LORETA patterns in CFS can be confirmed, this technique could find clinical applications.
Affiliation: (1) University of Seville, Spain. (2) Center for Brain Injury Rehabilitation (CRECER), Seville, Spain.
Background: Acquired Brain Injury (ABI) is a medical problem, with diverse etiology, commonly encountered today in the general community. Different neurophysiological methods have been developed to study ABI patients. Quantitative EEG (QEEG) has been proposed as a clinical diagnostic test to identify, localize, measure, and classify acquired brain injuries. This study describes an objective and quantitative discriminant algorithm that would allow classification of ABI patients according to their functional state.
Methods: \"Eyes closed\" QEEG recordings were conducted from 19 scalp locations in 50 patients with ABI (mean: 20 months after insult). Functional state of these patients was measured through Functional Independence Measure (FIM) and Functional Assessment Measure (FAM) scales. Patients were separated into three groups: complete dependency, moderate dependency and independence. EEG spectral analyses and data reduction were conducted, yielding a total of 9 QEEG variables.
Results: The discriminant function obtained between complete dependency and independence showed global classification accuracy of 96.6%, sensitivity of 93.3%, and specificity of 100%. Both positive and negative predictive values were over 90%, and cross-validation of this function showed an accuracy of 89.7%. The discriminant scores correlated significantly with all FIM+FAM subscales and were used to test against patients in the moderate independence group.
Conclusions: The algorithm obtained seems to be useful for the functional classification of patients with traumatic brain injury and stroke.
Affiliation: Institute of Psychophysiology and Rehabilitation
Background: Sleep stages might be seen as a natural situation of modifications in central nervous system functional status followed by the changes in autonomic control are dependent on patient's psychoemotional status and cardiovascular function level. The goal was an investigation of sleep quality and autonomic control in coronary artery disease (CAD) patients in relation to their functional state.
Methods: The contingent was 50 healthy subjects and 1086 CAD patients. Without anxiety and depression were 465 patients, with anxiety - 229; with depression - 64, and with both, anxiety and depression - 171 patients.Objectively sleep was evaluated using polysomnography: total sleep time (TST), sleep efficiency (SE), wakefulness after sleep onset (WASO), stages 1, 2, 3, and 4, REM sleep and body movements. Subjectively sleep was assessed using Pittsburgh sleep quality questioner. Hospital Anxiety and Depression Scale was used for psychological testing. Autonomic control was evaluated using heart rate (HR) variability analysis.
Results: CAD patients, as compared with healthy subjects, demonstrated reduced TST, SE, non-REM sleep, and REM sleep and increased WASO. Subjective sleep quality was worse for CAD patients as well. Worsening of sleep quality was following lowering functional state in CAD patients. Hypertension was followed by reduced TST, SE, and non-REM sleep, while diabetes and heart failure - additionally by a reduced REM sleep. Objective sleep quality was worsening in CAD patients with depression, while only subjective one - in patients with anxiety. CAD patients demonstrated lowered ability of autonomic HR control restoration during sleep in parallel to worsening functional status and sleep quality. Patients with depression and/or anxiety, as compared to those without, had significantly lower HR variability and higher mean HR frequency.
Conclusions: Coronary artery disease is disturbing sleep structure decreasing total sleep time and sleep stages 3-4 as well as reducing restoration of autonomic control during the night, especially in patients displaying depression/anxiety.
Affiliation: Brain Research Laboratories, New York University School of Medicine, New York, NY (1).
Nathan S. Kline Institute for Psychiatric Research, Orangeburg, NY (2).
Age related changes in the EEG have long been reported. Departures from age expected changes have been observed in mild cognitive impairment and dementia, the magnitude of which correlate with the degree of cognitive impairment. Such abnormalities include increased delta and theta activity, decreased mean frequency and changes in coherence. Longitudinal studies of mild cognitively impaired elderly (MCI) have reported differences in the QEEG profiles of those patients who go on to deteriorate and those that do not. Using positron emission tomography (PET) and magnetic resonance imaging (MRI), abnormalities in the hippocampus, the entorhinal cortex and the temporal neocortex have been shown to predict conversion to dementia in MCI patients. In this study, we focus on normal elderly (n=44) that present with only subjective complaints of cognitive decline but with no objective evidence of cognitive dysfunction, in order to determine the utility of QEEG in predicting future (within 7 years) cognitive decline or conversion to dementia. QEEG features were extracted from artifact-free eyes closed resting EEG, recorded from the 19 regions of the standardized International 10/20 System, compared to age expected normal values and expressed as Z-scores. High prediction accuracy (approximately 95%) was demonstrated. Source localization algorithms which identify the mathematically most probable underlying generators of the scalp recorded EEG found abnormalities in the baseline recordings of those subjects who go on to decline or convert to dementia in the same regions identified by other imaging methods including MEG and MRI.
Affiliation: Dept of Physiological Psychology, University of Salzburg
Background: The aim of this talk is to highlight the relation of cognitive performance specifically to changes in the EEG alpha and theta frequency band. While the neural efficiency model as proposed by Haier et al. (1992) or Neubauer et al. (1995) postulates a weaker neural activation in more intelligent subjects, the model of Klimesch predicts a stronger activation for these subjects.
Methods: In two experiments EEG was recorded while subjects either had to solve the
RAVEN matrices or a verbal analogy test. Band power changes in individually adjusted theta and alpha bands were calculated for a time interval in which actual problem solving took place.
Results: The results for the upper alpha band indicate that in the RAVENs task only for easy problems more intelligent subjects exhibited a weaker activation (neural efficiency). For solving of more complex problems, however, only more intelligent subjects were able to increase activation significantly. In the verbal analogies task, on the other hand, more intelligent subjects exhibited a stronger upper alpha activation in the left hemisphere. In the theta band, however, in both experiments more intelligent subjects exhibited a significantly stronger activation.
Conclusions: The fact that in both experiments more intelligent subjects displayed a stronger activation in the theta band clearly contradicts the neural efficiency hypothesis for this frequency range. The results for the upper alpha band however, can be interpreted consistently by the alpha inhibition hypothesis (Doppelmayr et al. 2005). As demonstrated repeatedly by Klimesch (1997) upper alpha is strongly related to semantic memory. If access to semantic memory is necessary for solving a specific task, more intelligent subjects will exhibit a stronger upper alpha activation. If, on the other hand, no semantic processing is needed more intelligent subjects are able to specifically inhibit the alpha frequency band.
Affiliation: Department of Psychology, Goldsmiths College, University of London.
The psychological significance of theta activity can be seen to span both the provinces of cognitive and affective neuroscience. Relevant to the former, processes include working memory, recognition memory, mental calculation, conscious awareness, and spatial navigation. Relevant to the latter are processes which on the whole were examined earlier and include the waking to sleep transition - hypnogogia, meditation, hypnosis, and positive mood states. Animal and human research will be integrated highlighting the conjunction of the different avenues of research on limbic structures and long range functional integration in the brain.
Affiliation: (1) University of Salzburg, Austria. (2) University of Tuebingen, Germany.
Background: During detection of a visual target a template of it needs to be hold in working memory (WM) in order to perform a memory match. Here the question is addressed whether a shift of spatial attention amplifies this memory match, and how this is manifested in the brain. Attention is associated with oscillatory activity at gamma frequency (40 Hz), and theta activity (around 5 Hz) reflects WM processes. Recent evidence suggests that gamma and theta oscillations might be phase coupled during memory processes in the EEG. Here we investigate whether a shift of spatial attention leads to increased phase synchronisation of theta and gamma activity.
Methods: A Posner task was run in which a cue, directing attention to one visual hemifield, was presented. The cue could be valid or invalid. Subsequently, one of two peripheral targets was presented in the left or right visual field. Subjects had to indicate which target was shown. EEG was recorded during the task. In response to a target, induced gamma amplitude was increased at posterior parietal cortex contralateral to the stimulated hemifield. Using a beamformer algorithm, bilateral neural sources of posterior gamma activity were localised for each subject individually. Scalp EEG was transformed into source space. Gabor expansion was used to obtain frequency information of source activity, and 40:n Hz phase synchronisation was calculated on a single trial basis.
Results: Targets elicited higher phase synchronisation between 40 Hz and oscillatory activity at theta frequency for the contralateral posterior source compared to ipsilateral between 100 and 200 ms poststimulus. This phenomenon was only found when spatial attention was shifted to the hemifield where the target was presented, i.e. in the valid cue condition than in the invalid condition.
Conclusions: Here we show that a match between a visual target and its memory template is reflected by increased gamma:theta phase coupling in the posterior parietal cortex. We could further show that attention does not only amplify early visual processing as it was often discussed, but it even enhances the matching of visual information with WM contents.
Affiliation: (1) Department of Psychology, University of Wales, Swansea, UK. (2) Department of Psychology, Virginia Polytechnic Institute and State University, USA
Introduction: Although much evidence suggests that participants who are highly susceptible to hypnotic induction show measurable frequency-specific differences in EEG power between the hypnotic and waking states, the idea that hypnosis represents a distinct state of consciousness remains controversial. For many years it has been known that the bispectral analysis of the EEG (a measure of dependency between EEG frequencies) is one component of an index of 'level of consciousness' that is used in a patented and clinically reliable measure of depth of anaesthesia. Despite this, however, bispectral analysis of the EEG has not previously been used in hypnosis research. The aim of this study was to apply bispectral analysis of the EEG in an attempt to resolve the state-non-state debate in hypnosis.
Methods: Participants were young healthy students who had previously been screened for hypnotic susceptibility using the Stanford Hypnotic Susceptibility Scale-C. EEG was recorded from 11 high susceptible and 12 low susceptible participants during rest with eyes closed both before and after hypnotic induction. Bispectral analysis was calculated for each individual in the waking and hypnosis conditions for every frequency pair from ~1Hz to 100Hz at 1Hz intervals. Analysis was by Partial Least Squares analysis (PLS), a method that can be used to identify those features of a multivariate time series that maximally co-vary with the experimental design.
Results: Bispectral values were generally low but were significantly greater for the high susceptible group than the low susceptible group, especially at low frequencies. PLS analysis showed significant changes in the bispectrum between the waking and hypnosis conditions only for the high susceptible groups and these changes were widely distributed across electrodes and frequency pairs. However, when the bispectrum was normalised by the power at each frequency the difference between conditions was severely attenuated.
Conclusion: Although, as expected, the bispectrum discriminated reliably between the waking and hypnosis conditions only for the high susceptible group, the differences primarily reflect between-condition differences in the power spectrum rather than a change in the non-linear-dependency between different frequencies per se. This suggests that the bispectral analysis adds little to the state-non-state debate that is not already provided by traditional Fourier analysis.
Affiliation: University of Kuwait.
Background: Neurophysiological evidence and EEG findings are used to illustrate how the physical properties of thalamocortical neurons cause oscillatory electrical activity and frequency response curves with a resonance peak in the EEG alpha band. Human EEG data confirms that with conditions promoting the linear and synchronous activity of thalamocortical circuits, these circuits, considered in aggregate, behave like other simple oscillators. The dynamic character of any simple oscillator can be fully described by just two parameters. The "damping ratio" parameter indicates the rate at which an oscillator loses energy. This determines the duration of oscillatory activity following a transient disturbance and also determines the sharpness of the resonance peak in an oscillator's frequency response curve. The second parameter, natural frequency, indicates the rate at which energy is transferred between system elements. This determines the frequency of free oscillation as well as the frequency of a resonance peak if the oscillator is "driven" at different frequencies. These two parameters provide a comprehensive description of the functional character of thalamocortical circuits that can be related directly to the physical properties of excitatory and inhibitory neurons. The psychological significance of individual differences in thalamocortical damping ratios and natural frequencies is confirmed by data which demonstrates that these parameters are related in a strong, systematic, and theoretically meaningful fashion to differences on all the major and generally accepted dimensions of personality and cognitive performance. These results open up the prospect of important advances in the understanding, diagnosis and treatment of psychiatric illness.
Affiliation: (1) University of Seville, Spain, (2) Center for Brain Injury Rehabilitation, (CRECER), Seville, Spain.
Background: Patients with severe Acquired Brain Injury (ABI) suffer from different cognitive, behavioral and psychophysiological sequelae. To treat these patients, different methods have been examined. Neurofeedback (NFB) is a computerized operant conditioning procedure whereby an individual modifies his/her EEG amplitude, frequency and connectivity patterns using external rewards. Studies on severe ABI patients have shown that it is very common for these patients to show alterations in Theta/low-beta ratio in their QEEG patterns.
Methods: Here, we empirically introduce this procedure for the rehabilitation of individuals with severe ABI. Patients were trained on a theta/low-beta protocol. An ABA design (A = NFB condition; B = mock condition) was utilized for a total of 15 sessions for each condition.
Results: We observed a trend towards normalization not only in trained frequency bands, but in all altered bands. Behavioral results showed an improvement in sustained attention, mood, and psychosocial adjustment as well as fewer sleep problems and less fatigue.
Conclusions: Our results show that this procedure allows clinicians to monitor the progress of rehabilitation and possibly shortens the length of treatment.
Affiliation: Practice for Psychotherapy for Children and Adults, & EEG-TRAIN - EEG-Biofeedback Trainingszentrum Hannover / Germany
Background: The actual demand in ongoing psychotherapy in combination with neurotherapy needs a unique and spontaneously structured procedure with flexible, adaptive use of multimodal clinical elements, among them EEG-Biofeedback, cognitive therapy, behavioral management and stress-management. This study describes the micro-steps in progress in a multimodal therapy of a 58 year old male with heart attack and 2 cerebro-vascular traumas - strokes - in 1997 with different psychosomatic impairments as spastic semiparese, impaired deep sensitivity left, and a variety of comorbidities.
Methods: The patient is treated with elements of cognitive-behavioral psychotherapy, Hemoencephalography (HEG) and Neurofeedback with frequency-band adaptation. About 80 sessions were carried out, usually 2 sessions per week. The treatment is still ongoing. Means of assessement and ongoing treatment-control were 16-sites MiniQ with statistical analysis, Continuous Performance Test (QIK), ongoing Symptom Tracking, and individual ratings. From these findings the different modules of ongoing treatment were chosen: cognitive methods and behavioral shaping for stress-management, mood control, and motivation; inter- and intrahemisperic Neurofeedback in bipolar placements with low and high frequency inhibits and shifting in reward-frequency-band and localisations. One module consisted of training of different sites to reprogram the impaired left site by activating mirror neurons. Main issue is to show the differential progress in micro-steps for different parameters.
Results: Central Neurofeedback could stabilise the arousal, specific intrahemispheric training on the sensuomotor strip right worked for relaxing overtensed parts of the left site of the body,
frontal/prefrontal training reactivated motivation remarkably, training of sites for mirror neurons initiated differentiation of right-left movements, parietal/occipital training right worked for reactivation of sensory emotions left site. Cognitive methods produced ananchoring of these neurophysiological proceedings with cognitive-emotional concepts and plans of action, as well as with controlling over-reactions in thoughts, emotions and body movements.
Conclusions: The progress in a complex Neuro-Psychotherapy requires focus to multimodal microsteps for careful management of continuous process. The study shows the effectivity of the multimodal adaptive procedure including Neurofeedback in a severe impaired stroke-patient with distinct enhancement of quality of life, as long as careful attention is focused on pragmatic and limited possibilities. Further considerations need to highlighten the concept "of the next microstep" and to incorporate Neurotherapy into cognitive treatment-concepts.
Gert De Mulder (1), Stefan Sunaert (3), Silvia Kovacs (3), & Paul Van de Heyning (1)
Affiliation: (1) Dept of Neurosurgery and Otorhinolaryngology, Antwerp University Hospital, Belgium. (2) Dept of Psychology, Wales University, Swansea, UK. (3) Dept of Radiology, Louvain University Hospital, Belgium.
Background: Sensory information processing in all the sensory cortices is associated with synchronized 40 Hz or gamma-band frequency firing. Synchronization of separate gamma-band activities, present in different corticothalamic columns, is proposed to bind distributed neural gamma activity into one coherent sensory percept. 40 Hz thalamocortical dysrhythmia has been proposed as a pathophysiological process present in neurogenic pain, tinnitus, Parkinson\'s disease, or depression. Trancranial magnetic (TMS) and Intracranial Electrical Stimulation (IES) can be used to suppress the dysrhythmic firing associated with pain, tinnitus and musical hallucinosis.
Methods: Quantitative EEG is performed in patients with uni- and bilateral tinnitus, pain, and musical hallucinosis in a soundproof area. Spectral analysis with specific attention to theta and gamma band firing is analyzed in 10 patients. After successful TMS, twelve patients with tinnitus were selected for auditory cortex implantation and five patients underwent an implantation of an extradural electrode overlying the primary somatosensory cortex aimed at alleviating unilateral phantom perceptions. Two patients with musical hallucinosis are investigated using only TMS.
Results: Patients demonstrate a spontaneous hotspot in the high beta/gamma range within the auditory cortex, contralateral to the side to which the tinnitus is referred or bilaterally in bilateral tinnitus. Tinnitus severity correlates to the current density asymmetry ratio of gamma activity as recorded in the auditory cortices. Results from electrical stimulation via extradural implanted electrodes demonstrated a highly significant tinnitus suppression both for pure tone tinnitus and neuropathic pain, and musical hallucinosis can be suppressed by TMS.
Conclusions: These preliminary results suggest that QEEG might be a method to objectivate phantom perceptions, by looking for its neural correlate in the respective sensory cortex. The hypothesis that gamma-band activity is the electrophysiological counterpart of sensory perception is supported. Targeting this area of abnormal firing is capable of suppressing its associated phantom perceptions.
Affiliation: (1) University of Washington, Seattle, USA., (2) Train Your Brain, Los Angeles, CA., (3) Private Practice, Truckee, CA., (4) Fullerton College, Fullerton, CA., (5) EEG Institute, Woodland Hills, CA.
Background: There is a growing interest in the potential use of neurofeedback (NF) for the treatment of chronic pain. However, these are limited published data concerning the effects of current NF treatment protocols on pain.
Methods: To examine the efficacy of a NF protocol that included slow-wave activity training at T3-T4, pre- and post-session 0-10 numerical ratings of pain intensity and other symptoms were assessed in a consecutive series of patients with Complex Regional Pain Syndrome who were treated with NF. To help determine if the immediate benefits associated with NF training last beyond individual treatment sessions, a patient with chronic pain associated with multiple sclerosis is currently receiving NF training that includes slow-wave training at T3-T4 (10-13 Hz), P3-P4 (6.5 - 11.5 Hz), and CZ-FZ (8-10 Hz).
Results: Of the 14 CRPS seen, statistically significant (p < .05) and large effects (d â¥ .80) were noted for reductions in pain intensity and perceived muscle tension, and increases in feelings of relaxation. Statistically significant and moderate effects (d.50) were found for decreases in muscle spasms and increases in feelings of well-being. The individual patient being seen over time reports consistent and substantial reductions in pain from before to after each treatment session, and that the reductions in pain that occur with NF treatment last from hours to days after the sessions. Moreover, the time of reduced pain after each session is increasing as treatment progresses.
Conclusions: The findings suggest that NF treatment is associated with immediate improvements in pain and overall feelings of relaxation and well-being. The long-term outcomes for the patient currently being seen, as well as findings from any additional patients seen, will be presented, and used as a basis for discussion concerning the use of NF for chronic pain treatment.
Affiliation: ICREA- Universidad Politecnica de Barcelona, Spain, and Dept. of Computer Science, University College London (UK)
The objective of this Symposium is to introduce 'presence research' to those interested and involved in the study of consciousness. Presence research originated with the attempt to understand and exploit the feeling of being and acting at a remote location in the context of teleoperator systems, and has been extended to the understanding of the processes involved in being in the place represented in a virtual reality.
Presence is a multidisciplinary area of research including computer science, psychology, neuroscience, and others. From a technical point of view it investigates the characteristics of a virtual environment that are crucial to generate a sense of being and acting within these virtual places. From the neuroscientific point of view it investigates how computer generated virtual sensory stimuli may be integrated within the human perceptual system to generate an artificial sense of reality. It is out contention that the study of presence provides a useful tool for the study of consciousness. Conceptually virtual environments allow the restriction the study of consciousness to a limited setting, giving us a more constrained frame of reference. This approach also provides an adequate scope within which to explore the neurological basis of consciousness.
Technically virtual environment systems provide us with a (virtual) reality that can be manipulated to explore situations in a way that is impossible in the real world. A particular example is the study of how brain-computer interfaces may be used for the manipulation of events within virtual environments and the consequences of that for the human participant.
This Symposium will provide an overview of the field, and also go in depth into specific topics. Some of these ideas have been presented in Sanchez-Vives and Slater (2005) Nature Reviews Neuroscience 6:332.
Presence in Virtual Reality
This talk will introduce virtual reality systems, the concept of presence and its measurement. Evidence that presence in virtual environments does occur will be presented, and applications in the realm of psychotherapy will be discussed.
Affiliation: Instituto de Neurociencias de Alicante, Universidad Miguel Hernandez-ConsejoSuperior de Investigaciones Cientificas, San Juan de Alicante (Spain).
The understanding of the phenomenom of presence in relation to sensory processing and perception will be discussed, with special emphasis on studies of self-perception and spatial processing. The talk will also examine the potential use of virtual reality for the study of consciousness.
Affiliation: Dept. of Computer Science, University College London (UK)
Brain-computer-interface (BCI) allows people to control a computer system using only their thoughts. We have integrated the EEG-based Graz-BCI system with the highly-immersive Cave-like virtual-reality (VR) system in UCL, London. We have carried out a set of experiments related to the question: can people use their thoughts to control VR, and if so, what is it like? In this talk we will present a high-level overview of the system and the results from our experiments, carried out over the last two years.
Presented by James Halper
Affiliation: Brain Research Laboratories, Department of Psychiatry, New York University School of Medicine and Nathan Kline Psychiatric Research Institute (ERJ & LSP).
We consider that the normative neurometric age regression equations that describe the power spectrum of the resting quantitative electroencephalogram (QEEG) define the Ground State of the brain. In disorders of brain function, the brain is displaced from this Ground State. In neurometric QEEG analysis, all QEEG variables are rescaled as Z-scores relative to the corresponding mean and standard deviation of the normative data. The pathological displacement can be conceptualized as a Brain State Vector (BSV) in a multidimensional signal space, scaled in standard deviations along each dimension. The length of the BSV represents the severity of the neuropathology and the direction of the BSV defines the quality of the abnormality. An ideal way to measure the BSV is to compute the Mahalanobis Distance across sets of neurometric Z-scores.
BSV trajectories of Mahalanobis Distances can be used to guide treatment. During the recent past, we have carried out a series of interventions in neurological and psychiatric patients based on the hypothesis that those pharmacological or physiological treatments that will most benefit the patient are those that will shorten the BSV and return it to within normal regions of the signal space, namely a hypersphere of radius 2.5 standard deviations centered at the origin of the space, as well as those which decrease the incidence of sharp wave, spikes or other epileptiform activity in the raw EEG.
These interventions have been performed in neurological patients in coma after respiratory arrest, cardiac arrest, or hypoxia as well as in psychotic patients manifesting hallucinations. Case histories will be presented illustrating the successful application of this method in patients with a variety of brain dysfunctions.
Affiliation: (1) Nova Tech EEG, Inc, (2) France Telecom R&D, (3) Capella University, (4) Q-Metrx,Inc, (5) Southwest College of Naturopathic Medicine, (6) Southwest College Research Institute, (7) Laboratory of Neurobiology of Action Programming, (8) Russian Academy of Sciences, (9) Institute of the Human Brain, (10) Norwegian University of Science and Technology.
This roundtable will illustrate the basic physical and mathematical principles behind the standardized Low-Resolution Electromagnetic tomography (sLORETA) method for source localization and feedback. The exposition will stress an intuitive rather than formal understanding of principles. Presentation of literature on sLORETA with emphasis on population patterns where feedback may be utilized and sLORETA of components and ERP based subtypes with feedback implications. The combination of Independent Component Analysis (ICA) and sLORETA provides a powerful tool for studying brain electromagnetic dynamics both off-line and on-line. The relation to current real-time methods such as Neurofeedback (NF) and Brain Computer Interface (BCI) will be explored.
Joan Romano, PhD (1), Joseph Barber, PhD (1), Diana Cardenas, MD (1), George Kraft, MD (1), Amy Hoffman, MS (1), Dave Patterson, PhD (1)
Affiliation: (1) Department of Rehabilitation Medicine, Box 356490, University of Washington, Seattle, WA, 98195-6490, USA.
Background: Chronic pain in a significant problem in many individuals with disabilities, but is often refractory to traditional biomedical treatment. Case studies suggest that hypnosis has the potential to benefit at least some of these individuals.
Methods: In the current study, adults with chronic pain and a disability (spinal cord injury, multiple sclerosis, acquired amputation, neuromuscular disease, and cerebral palsy) are treated with up to 10 sessions of hypnotic analgesia.
Results: Preliminary findings from the first 33 study participants showed statistically significant pre-to post-treatment decreases in characteristic pain intensity that is maintained at three-month follow-up. Significant improvements were also observed in pain unpleasantness and perceived control over pain. Concentration of treatment (e.g., daily vs. up to weekly), pre-treatment outcome expectancy, and initial response to treatment during the first session were not significantly associated with long-term outcome. Hypnotizability was also not associated with treatment outcome in the group as a whole, although preliminary analyses suggest that participants with high and low hypnotizability responded differently to the different analgesia suggestions. By September 2006, data will be collected on over 77 additional subjects (over 110, total).
Conclusions: Analyses will determine which of the findings from the first subjects replicate, help determine the extent to which hypnotic analgesia for the treatment of chronic pain in persons with disabilities may be helpful, and provide preliminary findings concerning the predictors of hypnotic analgesia treatment outcome.
Affiliation: Living Health Center for Research and Education, Istanbul Turkey
Background: Schizophrenia is sometimes considered as most devastating of the mental illnesses because its onset is early in a patient's life and its symptoms can be destructive to patient and to the patient's family and the friends. EEG studies of schizophrenic patients indicate a higher number of patients with abnormal records decreased alpha activity, increased delta and theta activity and possibly more left sided abnormalities, and some coherence abnormalities. Schizophrenia affects 1 in a 100 people at some point during their lives and while there is no cure, it is treatable with antipsychotic drugs. The clinical Antipsychotic Trials for Interventions Effectiveness (CATIE) raise important questions about relapse, treatment resistance, minimizing adverse effects and improving treatment adherence. About 74 % of the patients discontinued the first medication prescribed within a year and they will have a relapse. It shows an enormous need for developing better methods of delivering treatments and better ways to manage the disease. What CATIE and other research reinforce is that we have an enormous need in treating patients with schizophrenia. We still don't have sufficient impact on negative symptoms, cognitive dysfunction, nonadherence to treatment or comorbid substance abuse and/or post -concussive syndrome. We thought neurofeedback treatment may address positive symptoms, negative symptoms, cognitive dysfunction, nonadherence to treatment or comorbid substance abuse and/or post -concussive syndrome while the patient is on or not on antipsychotics.
Methods: Most of the patients were diagnosed with chronic schizophrenia before coming to our center and did not improve in their symptoms with antipsychotic medications. Neurometric analysis of the QEEG of the patients suggested chronic schizophrenia consistent with the clinical judgement of the Author. All 40 patients received QEEG analysis with the Nxlink data bank at baseline and at the end of the treatment. All patients were taken off their medications during the duration of the half life of their medications, (for example, if they were on risperdal, the half life of the medication is 6 days, so they were off medication for 6 days before QEEG recording, and then drugs were reinstated). They also received the Scale for the Assessment of Positive Symptoms (SAPS), Scale for the Assessment of Negative Symptoms (SANS) and Schizophrenia Syndrome Rating Scale (SSRS) at the baseline and at the end of the treatment. 12 out of 40 patients also received the TOVA and MMPI at baseline and at the end of the treatment. Responses to SAPS and SANS and SSRS were analyzed to evaluate the effectiveness of neurofeedback. All patients received between 40-120 neurofeedback training sessions. NF protocols were chosen according to their QEEG neurometric analysis. 7/ 40 patients dropped the treatment between 17 and 37 sessions of neurofeedback treatment and 2/40 did not respond. All the patients came 6 days a week for two neurofeedback sessions a day. Each session for 30 minutes and the second session was given after at least a 45min.break.
Results: 31 schizophrenic patients improved after neurofeedback treatment. 21 schizophrenic patients who received neurofeedback training showed significant improvement based on QEEG analysis with the Nxlink data bank, SSRS, SAPS, SANS. An additional 10/12 schizophrenic patients who received neurofeedback training showed significant improvement based on QEEG analysis with Nxlink data bank, SSRS, SAPS, SANS as well as the TOVA and MMPI results. The patients who showed coherence abnormalities in their QEEG responded better and quicker to neurofeedback training. Antipsychotic medications may increase coherence abnormalities in the brain but coherence abnormalities can be corrected with neurofeedback treatment to get a better outcome. NF treatment may increase adherence to treatment and reduce possible side effects of antipsychotics in schizophrenia. Further study with controls is warranted.
Affiliation: (1) Dept of Neurosurgery and Otorhinolaryngology, Antwerp University Hospital, Belgium. (2) Dept of Radiology, Louvain University Hospital, Belgium
Background:The auditory system has two different pathways supplying auditory information to the cortex: the tonotopical lemniscal system and the non-tonotopic pathways also known as the extralemniscal or non-classical pathways. Hyperactivity of tonic and burst firing in the lemniscal and extralemniscal system respectively are correlated to tinnitus. The extralemniscal system connects to the other sensory systems, whereas the lemniscal system does not. It has been shown that median nerve stimulation is capable of altering tinnitus perception in somatic tinnitus. As the C2 nerve has direct anatomical links to the auditory pathways it can be expected that SS stimulation of the C2 nerve or dorsal column stimulation at the C2 level can be used clinically to suppress tinnitus.
Methods: 7 patients (3 women, 4 men) suffering intractable somatic tinnitus were treated with C2 stimulation. Under local anesthesia two subcutaneous 4 channel electrical leads are implanted via a midline approach over the left and right occipital nerves. Stimulation is applied at different frequencies (6, 10, 12, 18, 40 Hz) and VAS scores are compared before and after SS stimulation.
One healthy volunteer undergoes an fMRI while sound is presented and electrical stimulation applied in order to find a pathophysiological mechanism explaining the tinnitus suppressing effects.
Results: Average tinnitus suppression for pure tones is 87.5%, whereas white noise/narrow band noise can only be suppressed for 28%, a statistically significant difference (mann-Whitney z=2.48, p=0.013).
Analysis of the fMRI data demonstrates that C2 stimulation at 6 Hz has a different effect on the brain than at other stimulation frequencies. At 6 Hz not only auditory and somatosensory activation is seen, caused by the simultaneously applied stimuli, but also activation of the autonomic and limbic system, as well as the reward system.
Discussion and Conclusion: This small pilot study suggests that C2 stimulation could become a valuable treatment option for somatic tinnitus in selected patients with pure tone tinnitus. The interesting fMRI data merit further investigation to elucidate the pathophysiological mechanism of tinnitus attenuation in somatosensory stimulation.
Affiliation: (1) Brainquiry B.V., Nijmegen, The Netherlands. (2) Brain Resource Company B.V., Nijmegen, The Netherlands. (3) EEG Resource Institute, Nijmegen, the Netherlands
Real-life neurofeedback is task-related neurofeedback: the feedback on brain activity is provided on the moments and in the setting in which the effects should occur. In this, real-life neurofeedback deviates from regular neurofeedback in the clinic, where training is done in a clinical environment sitting behind a computer screen. Therefore, real-life neurofeedback does not rely on the assumption of carry-over effects from the clinical session to daily practice. Furthermore, the real-life approach circumvents the operant conditioning method used in regular neurofeedback. Adversely, real-life neurofeedback operates through classical conditioning, where the task itself and the task-setting serve as the conditioned stimulus.
In general, the basis for the real-life neurofeedback training sessions is an assessment session, in which the task-related brain profile (the differences in brain activity for good vs. bad events) is quantified for each subject individually. The profile that belongs to the good events reflects an optimal mental state for the task at hand. The training incorporates feedback on this optimal mental state: the task should only be executed when the feedback cues the user the optimal mental state has been reached. In a scientific study it was established that golf players show differences in brain activity for good puts and for missed puts and that real-life neurofeedback training improved their success rate by 20 %, on average, in as little as three training sessions (effectively, this is an improvement of 40% since players already started on 50% baseline).
In summary, real-life neurofeedback training enhances learning and speeds up the effect of feedback, resulting in a decrease of the minimum number of training sessions compared to regular neurofeedback and leads to more specific learning for the specific task at hand (state dependent learning).
Affiliation: (1) EEG Resource Institute, Nijmegen, the Netherlands. (2) Brain Resource Company B.V. / Brainquiry B.V., Nijmegen, The Netherlands
Neurofeedback training requires considerable effort of clients in terms of time, money and energy, particularly for clients who live more than an hour away from their therapy practice. Tele-neurofeedback is a new approach to in order to facilitate the treatment process for these clients. In this approach, the clients handle their own hardware at home, using software installed on their own computer, while being real-time supervised by their therapist via the internet. In this study, we describe the procedures that were developed in order to secure a professional treatment, based on initial vis-a -vis training. Second, we will report on the conditions that were found to be crucial for this type of training in order to be successful, the major one being undisturbed concentration. Thirdly, we will inventory results of this approach in terms of type of complaints, duration of therapy, decrease of complaints and type of protocols, in comparison with regular NFB treatment. Preliminary results suggest that in particular the type of complaints are specific, in the sense that over time less children with AD(H)D were offered tele-neurofeedback. Further developments in this field of NFB treatment will be discussed.
Affiliation: University of California, Los Angeles, USA.
A convergence of recent and earlier findings focused on the potential impact of mechanisms underlying EEG rhythmic patterns on the mammalian learning process has provided the basis for a compelling model addressing the utility of EEG operant conditioning as a means of altering brain organization. If correct, this model can provide a guide to improved training methods and software logic in this field. This convergence includes evidence concerning the process of long-term potentiation and related synaptic plasticity, as well as a consideration of the critical role of reward as a basis for learning consolidation. Studies in animals and humans have shown that bursts of dominant frequency rhythmic activity occur systematically following trained responses, and behave systematically in relation to the learning process. This EEG pattern is termed "post-reinforcement synchronization". Neurophysiological findings suggest that this pattern results from a brief suppression of arousal with reward. The discrete EEG oscillations that reveal this suppression in turn function to stabilize and consolidate the neurocircuitry underlying the rewarded behavioral response. Detailed evidence will be presented in support of these conclusions.
[Posters will remain up during conference]Neurofeedback and enhanced performance: The plethora and the paucity.
Children with Attention Deficit/Hyperactivity Disorder (ADHD) typically show excess theta and reduced beta waves in their EEGs compared to control children, but little is known of the modulation of these parameters as attention fluctuates.
Data were previous recordings from 13 children with ADHD (never medicated with psycho-stimulants) and 10 controls, all 6-9 years old and having at least 5 errors in the Go/no-Go task Three analysis techniques, namely spectral amplitude (AMPL), coherence (COH) and a new method called Dominant Phase Amplitude (DPA), were applied to 0.5 s segments just preceding stimulus presentation that were classified into correct or incorrect trials (i.e. adequate or inadequate attention). Permutation tests (Achim, 1995, 2001) with alpha =0.01 were applied separately by frequency band but collectively to all channels or channel pairs.
Over all 23 children, the EEG preceding correct and incorrect trials differed for the following band-method combinations: 4 Hz-DPA, 12 Hz-AMPL, 14 Hz-AMPL, 14 Hz-COH, 18 Hz-AMPL (amplitude before errors was elevated at 4 Hz and reduced at 14 and 18 Hz). Although these differences tended to be significant in the ADHD group but not in the control group, all corresponding group x condition interactions yielded larger p values than the non significant condition effect within the control group. Looking specifically for group x condition effects, two band-methodcombinations were identified on which the ADHD children could differ from controls, but their patterns were not convincing (one was clearly due to a single outlier).
The lack of effect in controls could be due to low power, since they typically had much fewer error trials. Finding within-subject effects that parallel previously documented group differences suggests that these EEG markers reflect state rather than trait, but evidence is still lacking whether the inattention of children with ADHD differs qualitatively from that of controls.
This paper describes the application of the mixed general linear model statistical analysis of the quantitative information from an electroencephalogram (qEEG). The modeling is similar to regression, which builds a regression or “best-fit” model for the data structure, but, in addition, provides for correlations between observations. A linear mixed model states that data consists of two parts, fixed effects and random effects. Fixed effects determine the expected values of the observations, while andom effects account for random deviations from these expected values both between and within individuals. If errors are independent between subjects, deviations from expected values may also be modeled by modeling the covariance structure of residuals. The term ‘repeated measures’ here refers to data with multiple observations from one specific source. It is reasonable to assume that observations from the same individual, are correlated, if only slightly, in some measurable way. Consequently, statistical analysis of repeated measures data gives a more accurate prediction capability when the issue of covariation between these measures is addressed.
With advances in technology and recently available mixed model methodology, e.g., the mixed procedure (Mixed PROC) of the SAS® system, the covariance structure can more easily be incorporated into the statistical model. In our approach we disregarded potential random effects that would not be specific to single individuals. Thus, by absorbing potentia within-subject random effects into the covariance matrix, we can work with the simplified model.
The use of a mixed procedure for modeling data structures provides a more accurate and objective method of analysis and yields quantifiable equations for testing predictions. Essentially, this method allows the physiological pattern of each individual, not related to any other variable, to be represented and accounted for within the model.
Advances in our understanding of the importance of functional specialisation in the neocortex have been accompanied by a growing realisation of the need for mechanisms to co-ordinate activity between different brain regions. Unfortunately, our knowledge about how this co ordination emerges lags well behind our knowledge of functional specialisation. It seems likely, however, that healthy functioning of the brain is characterised by the formation of transient functional alliances between different brain regions mediated by synchronous oscillatory activity. Recent developments in signal analysis have begun to make it possible to test these ideas in humans using electroencephalogram (EEG) measures of functional connectivity. The aim of this study was to apply some of these recently developed methods to help identify abnormalities of functional co ordination in the brains of patients with schizophrenia. EEG recordings from 35 un-medicated patients with schizophrenia (15 positive and 20 negative syndrome) were compared with 24 healthy controls during Eyes Open and Eyes Closed resting Conditions. The patient groups not only showed lower overall levels of connectivity across all frequency bands than the controls but also showed a different pattern of connectivity. In particular, patients showed a less clustered pattern of connectivity and lower inter hemispheric communication. The implications of these finding and the strengths and limitations of the methodological methods used will be discussed.
The design and analysis of neurofeedback experiments poses several problems related to (among others) the definition of the learning process, the treatment of extra-cranial artifacts, and the group averaging of individual learning curves.
A suitable theoretical and statistical framework will be delineated. First, the kinds of learning process that can be shown as a consequence of the neurofeedback treatment will be defined. Second, appropriate experimental designs to uncover them will be explained. Third, the use of single-subject permutation-randomization analysis of covariance (ANCOVA) will be shown to handle artifacts (treated as covariates), while quantifying the experimental evidence for all kinds of learning process as previously defined. Finally, the parametric combination of p-values will be proposed as a flexible method to derive group-based statistical inference.
The proposed framework is shown to be valid in a wide range of experimental designs where learning curves are of concern, and also powerful, specific, rapid, and flexible. The statistical method is the same for all learning processes and is easily understood by non-specialized audience. All necessary computations can be performed with freeware software. All www resources will be provided.
After over 30 years of research in neurofeedback, there are not reference methods for the analysis of neurofeedback data. There is no consensus even on what a neurofeedback learning process is or should be. The aim of this work is to propose a framework capable to categorize previous research according to the kind of learning process investigated and to offer a suitable method for the treatment of experimental/clinical data.
(1) Brain Research and Neurophysiology Laboratory, Department of Psychology, The University of Tennessee, Knoxville, (2) Nova Tech EEG, Inc., Knoxville, Tennessee, (3) Lexicor Medical Technology, Inc. Boulder, Colorado
Last year in Udine we presented a pilot study (n=3) on LORETA Neurofeedback. In the meanwhile we completed the study adding three subjects. With this presentation we aim to report new results and to discuss future research directions. The Neurofeedback based on Low-Resolution Electromagnetic Tomography (LORETA) has been conceived to overcome some limitations of the current neurofeedback paradigm, namely, the limited information provided by a single or a small number of electrodes placed on the scalp, and the choice of the reference electrode. By means of inverse solutions techniques such as (LORETA) spatially delimited brain activity can be evaluated in intracortical tissue. The task is to feed back in real-time LORETA current density.
Six individuals were trained to improve brain activation (suppress low Alpha (8-10 Hz) and enhance low Beta (16-20 Hz) current density) in the anterior cingulate gyrus cognitive division (ACcd). Three participants took part of six experimental sessions, and three of 20 experimental sessions, each lasting approximately 30 minutes. Permutation-randomization analysis of covariance was performed on learning curves of the target activity during the sessions and on a set of randomized trial subministered at the end of the treatment. Randomized trials consisted of eight two-minutes periods (trials) for which participants were asked to try to obtain as many rewards as they could (4 "plus" trials) or as few rewards as they could (4 "minus" trials). The order of trials was decided at random. We tested the hypothesis that the Beta/Alpha ratio increased over sessions (learning curves) and that participants acquired volitional control over their brain activity so to be able to obtain more rewards during the plus condition as compared to the minus condition (randomized trials).
We performed single-subject analysis and we obtained group (n=6) inference by means of parametric p-value combination (additive and multiplicative methods). For the randomized trials we found strong evidence of volitional control of the Beta/Alpha ratio (additive: p=0.012; multiplicative: p=0.002). For the learning curves we found weaker, yet substantial, evidence of increased Beta/Alpha ratio over session (additive: p=0.152; multiplicative: p=0.007). Single-band analysis revealed that both effects were driven by the Beta band.
The experiment showed overall signs of volitional control and increased Beta/Alpha electrical activity in the target region (ACcd). Tomographic neurofeedback has never been done before and is very promising. Possible applications of the technique include the treatment of epileptic foci, the treatment of specific brain regions damaged as a consequence of traumatic brain injury, and in general of any specific cortical electrical activity. We are now designing a sort of 3-D brain voyager, an advanced interactive graphic representation of the brain to offer more direct and reliable neurofeedback.
A Fortune 1000 company has created a Peak Performance Center as part of their leadership development program. It was patterned after the program at the Center for Enhanced Performance at the United States Military Academy at West Point, the U.S. ‘s largest performance enhancement center, which was started by one of the authors (LC) in 1989.
The top 22 executives at Hillenbrand, Inc. participated in a 20 session program, which included training on the FreezeFramer; respiration, EMG, and skin conductance precedig EEG training to enhance concentration performed on the Peak Achievement Trainer. The executives participated in five hour-long sessions of Peak Achievement Training, consisting of individual coaching based on (with elaboration) the Peak Achievement Trainer's workbook, including about 35 minutes of practice on the InAll wideband suppression protocol per session. In the first and the last session, they were tested by measuring how long they could hold their InAll levels below 30 microvolts.
During the pre-test, the first trial averaged 19 seconds, ranging from 10 to 40. The first trial in the post test was more than twice as long, 44 seconds, with the range running from 25 to 65 seconds. The average duration of the best trial per session almost doubled from 65 seconds at the pre-test to 128 seconds at post test. The ranges were 18 to 180 and 48 to 220, respectively. This is clear evidence of very rapid learning.
This is clear evidence of very rapid learning. Response to te program was overwhelmingly positive, with notable improvements in their ability to attend in crucial meetings, engage in critical performance appraisals with employees, and deliver exceptional presentations. These executives have returned voluntarily for many additional sessions, primarily using the Peak Achievement Trainer's ConAlert protocol, and the program is being expanded to 39 more executives. They are writing an article for Harvard Business Review.
A psychiatric disorder can be viewed as a manifestation of a maladaptive brain organisation that arises from unfavourable interaction of genome and environment. At its core resides an abnormal cerebral network architecture with altered neural connectivity and communication. These changes lead to neuropsychological, cognitive and behavioural malfunctioning that give rise to the psychiatric symptoms on the subjective and behavioural level. Consequently, the diagnosis that relies exclusively on this most superficial level (e.g., ICD/DSM-classifications) is hampered by the sources of variability that occur between the different levels of manifestations (outlined in Fig.1 for psychotic illness).
There is, however, no need to rely exclusively on “surface” measures for classification of disease and diagnosis. It has become possible to directly map dysfunctional neural networks in the brain by recording abnormal magnetic brain activity. Focal electromagnetic slow waves, for instance, are generated in the surround of circumscribed pathological or lesioned brain areas. Abnormal focal spontaneous brain activity may occur as slow wave lesional activity, as epileptic spikes, and as seizure activity. Typical lesions producing a variety of abnormal waves include cerebral infarcts, contusions, local infections, tumors, developmental defects, degenerative defects or subdural hematomas. Abnormal slow waves not only result from neurological disease, but are also commonly observed in a variety of psychopathological conditions. We examined to what extent local clustering of magnetic slow wave generators was related to symptomatology and neuropsychological functioning in patients with DSM-IV diagnoses and to what extent such clusters mark dysfunctional brain areas.
Spontaneous magnetic brain activity was recorded with a whole-head neuromagnetometer in 82 schizophrenics, 80 patients with PTSD, 14 with affective disorders, 12 alcohol dependent, and 35 healthy subjects during 5-min periods of rest. For bandpass-filtered data in the delta (1.5 – 4Hz) and theta (4Hz – 8Hz) ranges, the origin of focal slow waves was determined (using a single equivalent current dipole model) and the number of generators was counted per voxel, i.e., a density function of focal generators was computed.
In most patients these density functions were circumscribed, limited to one or a few focal brain regions generating large amplitude slow waves. The localization of these foci was group-specific for both the delta and the theta bands. Consistently more slow waves were observed in left temporal and parietal regions in schizophrenic patients (Figure 2), and were correlated with more symptoms (temporal abnormality with hallucinations) and more perseverative errors in neuropsychological tests. We did not find differences in the focal clustering of delta and theta slow waves between schizophrenic patients with and without neuroleptic medication. A weak negative correlation observed between temporal theta activity and neuroleptic dosage might be mediated by severity of symptoms.
Depressive patients exhibited a suppression of frontal activity compared to any of the comparison groups. Less attenuation was apparent when patients had responded favourably to medication. In PTSD patients we generally observe left frontal islands of focal slow waves. The group-specific pattern of slow wave activity was not affected by mental activation.
In conclusion, the mapping of abnormal slow waves (ASWAM) is a potential diagnostic and prognostic tool for psychiatry as well as neurology. We conclude that focal slow wave activity, localized by means of magnetic source imaging is indicative of dysfunctional brain regions in psychiatric disorders and should be evaluated as a diagnostic and prognostic measure. Moreover, the mapping of dysfunctional neural networks may also help us to understand these disorders from a neuroscience perspective.
Changes in EEG coherence, and in theta and gamma EEG frequencies were reported in meditation. Since, however, scalp locations do not necessarily indicate directly underlying sources, only coherences computed between intracerebral model sources can reveal interpretable, functional connectivities between brain areas.
27-channel EEG was recorded from a long-term meditator while repeatedly performing three different meditations (Ch'an Buddhism), and during a control condition, in 4 identica, independent sessions. From the first 3 minutes during meditation and control, all artifact-free 2-second EEG epochs (N=2323) were analyzed. Intracerebral model sources were computed using LORETA for 2394 voxels; each voxel was assigned to the closest scalp electrode position, forming 27 cone-shaped brain regions of interest (ROIs). Theta (6.5-8 Hz) and gamma (35-44 Hz) coherences were computed between all pairs of scalp EEG electrodes and between all pairs of intracerebral LORETA ROIs. T statistics compared the coherences between control versus meditation.
During meditations compared to control, all significant changes in theta coherences were increases, while most in gamma were decreases, both for scalp EEG and for LORETA. But, the spatial distribution of scalp and LORETA coherences differed strongly (both for theta and for gamma). E.g., the ratio of significant coherence changes involving anterior compared to posterior locations was 95 to 30 for scalp EEG, but 12 to 17 for intracerebral LOREA (chi square = 13.3, p=0.003). Theta LORETA coherences increased in all three meditations. Gamma LORETA coherences predominantly showed decreases, but, consistent in the three meditations, had isolated increases in both temporal brain regions (anterior-posterior).
As expected, scalp computed EEG coherences differed from intracerebrally (LORETA) computed coherences. During meditation, the relevant LORETA results showed general coherence increases for theta, and decreases for gamma, but gamma also showed isolated increase in both temporal regions.
Subjects with Panic Disorder (PD) have high scores on alexithymia, a tendency to interpret ambiguous stimuli as threatening and an attentional bias toward threat-related cues. Several findings suggest that a dysfunction of temporo-limbic regions, in particular those of the right hemisphere, involved in emotional processing, might underlie these characteristics.
Alexithymia was evaluated by the Toronto Alexithymia Scale in 17 drug-free patients with DSM-IV PD and matched healthy controls (HC). In all subjects ERPs were recorded from 30 channels during a visual target detection task, in which stimuli with different emotional valence (neutral, erotic, threat-related and phobic) were used as distractors. The Low resolution Electromagnetic Tomography (LORETA) was used to identify cortical generators of the ERP P3a component for distractors.
Alexithymia was more frequent in PD subjects than in HC. Comparison between erotic and neutral distractors revealed a different pattern of activation in HC and PD subjects: an activation of the anterior cingulate, insula and medial frontal areas was observed in HC, while a reduced activation of the right parieto-temporal regions was found in PD subjects. In HC no difference in activation patterns was found when comparing threat-related and neutral stimuli, while in PD subjects a reduced activation of right temporal regions was observed when the same comparison was performed.
First the efficacy of the enhancement of alpha theta in elevating artistry to a professionally significant degree in conservatory students will be outlined, as it represents the first attempt to validate the protocol as a sole intervention. The implications will then be discussed of ancillary experiments designed to explore relaxation as a mediator, temporal changes in theta-alpha ratio within and between sessions, and post-training alterations of the topographical EEG. The relative influences of faster wave training on performance will also be outlined. A phenomenological analysis of the musician's experience will be reviewed, and together the results will be considered in the light of the historical development of alpha theta training and earlier applications aimed at treating anxiety and addiction. The limitations of our current state of knowledge about the applicability of this paradigm will be considered and open discussion will be encouraged. Current knowledge about the role of theta in psychological processing will be covered. Can other interventions such as autonomic biofeedback, meditation, hypnosis and energy medicine achieve similar goals? Comparative studies are underway. References will be provided.
From several EEG studies (Klimesch, Doppelmayr, Pachinger, & Ripper, 1997; Doppelmayr, Klimesch, Stadler, Pollhuber & Heine, 2002), there is good evidence that high power in the upper alpha band and low power in the theta band preceding a task is related to good cognitive performance. This study investigated the hypothesis whether an increased absolute alpha power or a decreased absolute theta power is capable of increasing cognitive performance.
Twenty-one healthy subjects where instructed to increase their absolute upper alpha ower, or decrease absolute theta power with neurofeedback training. Mental rotation tasks where presented before and after neurofeedback training.
After upper alpha training the subjects performed better on mental rotation tasks but not after theta training. Additionally it could be demonstrated that the pre-stimulus power in the upper alpha band increases after training.
This is well in line with other studies (Klimesch, Gerloff, & Sauseng, 2003) which showed that high upper alpha power in the pre stimulus interval is related to good cognitive performance. This study shows that neurofeedback training can be used to increase cognitive performance by way of those factors which are known to underlie good cognitive performance under normal conditions.
Analysis of phase synchrony over large cortical distances has become increasingly important in the study of normal and abnormal brain function. Deficiencies in long-range phase synchronization described by Bhattacharya (2001) have been tied to brain pathology. Recent research suggests that phase locking and phase coupling mechanisms may transfer information or functionally integrate large-scale neurocognitive networks. The present investigation examines the effects of long-term transcendental meditation (TM) prctice on EEG phase synchrony.
Nineteen channel EEG of 15 long-term practitioners of the TM technique is recorded during eyes closed resting conditions. Fifteen volunteers matched for gender, age and education served as Controls. Quantitative EEG analysis in standard delta, theta, alpha and beta bands is performed using Neurorep (NREP4) PHASE program.
Analysis of forty seconds of data for each subject yielded an average millisecond time delays (phase-lag) for each electrode pair at each frequency.
The fifteen TM subjects show more alpha (7-13 Hz) long-range (front-to-back) phase synchrony, as compared to the Control group. Though both groups showed more alpha phase synchrony than the younger normative database, the TM group had twice the number (177 vs. 87) of significantly shorter phase delays than the Controls(P<.05).
The results support a field theory of consciousness (John, 2001) where neuronal populations become more integrated at lower levels of excitation. Icreased alpha phase synchrony found here suggests that long-term TM practice enhances the functional integration of large-scale neuro-cognitive networks.
Bhattacharya, J. (2001) Reduced degree of long-range phase synchrony in pathological human brain. Acta Neurobiol. Exp. 61, 309-318
John, E.R. (2001) A field theory of consciousness. Consciousness and Cognition 10, 184-213
The Thought Translation Device (TTD) is a brain-computer-interface based on the self-regulation of slow cortical potentials (SCPs) and enables severely paralyzed patients to communicate using their brain potentials. However, many locked-in patients are no longer able to focus their gaze on a computers screen. Therefore an extended version of the TTD provides all feedback information necessary for brain computer communication over the auditory channel. To evaluate the system a first study with 3*18 healthy participants was carried out to test the performance of physiological regulation under three different feedback conditions: Visual, auditory, or combined visual-auditory feedback of slow cortical potentials. The training consisted of three sessions with 1500 trials with a random assignment of required cortical positivity or negativity. The results show that physiological regulation of SCPs can be learned with auditory and combined auditory and visual feedback although the performance is significantly worse than with visual feedback alone. In a next step two auditory communication paradigms are presented: An auditory letter selection program that tells a patient the letter sets with a computers' voice as well as the resulting response. For less skilled patients a question answering system was developed presenting questions and allowing them to answer them with brain responses. An example with a completely paralyzed patient shows that even with unreliable brain control significant answers can be obtained. The results of a first application to a completely paralyzed patient are presented.
Neurological disorders like cortico-subcortical stroke or degenerative neuromuscular diseases can lead to "locked-in" syndrome, a state of complete motor paralysis with intact sensory and cognitive functions. One of the most terrifying aspects of this syndrome is the loss of the ability to communicate. To re-establish communication in severely paralysed patients, we have developed an EEG-based brain-computer interface. Visual feedback of electrocortical activity and operant conditioning have been combined to train patients to operate this communication device by generating shifts in their slow cortical potentials. When a patient has achieved reliable control over his/her slow cortical potential shifts, these responses can be used to select or reject items (letters, commands, links in an internet browser) presented on a computer screen. This talk will highlight recent progress in brain-computer communication in the severely paralysed. This includes research on the cortical mechanisms underlying the voluntary production of slow cortical potential shifts. Functional magnetic resonance imaging in both healthy volunteers and patients with neurodegenerative diseases has shown that during the preparation interval preceding a required slow cortical potential shift, there were increased activations in widespread central and precentral regions. During the active phase, this was followed by distinct activation differences in the vicinity of the feedback electrode. Finally, future developments will be outlined including feedback and self-regulation of regional brain activity using real-time functional magnetic resonance imaging.
The question is addressed whether event-related potentials (ERPs) are generated by a superposition of evoked oscillations in the theta and alpha frequency range. We develop this hypothesis within the framework of an oscillatory phase resetting model for ERP generation. Furthermore, we investigate the question whether evoked theta and alpha have a similar functional meaning for memory as is known from event-related de/synchronization (ERD/ERS).
Our findings indicate that (i) theta and alpha show a significant increase in phase locking during the time window of early ERP components as compared to a prestimulus reference, (ii) the dynamics of event-related changes in evoked theta and alpha power obey the same principles as are known from event-related de-/synchronization research and (iii) latency measures of the P1-N1 complex are negatively correlated with individual alpha frequency. In addition, we have found that theta phase locking is larger during encoding than recognition and that good memory performers show a larger increase in theta and alpha phase locking during recognition in the time window of the N1.
The reported findings suggest that cognitive performance may be based at least in part by an interplay between the synchronous activation of three neuronal network systems, a working memory, attentional, and semantic memory system, each operating with a different frequency, the first in the theta (about 6 Hz), the second in the lower alpha (about 8 Hz) and the third in the upper alpha (about 12 Hz) frequency range. The implications of this theoretical framework are discussed by considering phase sensitive measures to analyze 'local' and 'large scale' integration processes between different neural networks.
Studies of human brain electric fields indicate that the "stream of consciousness" actually consists of identifiable, brief building blocks. This is based the observation that momentary landscapes of the spatial distribution of brain electric potential change in a discontinuous manner. Short periods of quasi-stable potential landscapes (~100 ms, called 'microstates') are concatenated by rapid transitions. Since different distributions of potential must have been produced by different neural populations, microstates can reasonably be assumed to incorporate different brain functions. In schizophrenic patients before medication, microstates of two classes (defined by potential landscapes) were shorter in duration than those of controls, suggesting precocious termination of certain classes of information processing in the patients. Moreover, sequencing of the microstates was different in the patients, suggesting deviations in "mental grammar" in schizophrenic disorder. The hypothesis that different microstates incorporate different brain functions was supported by microstate studies during spontaneous thoughts (prompted reports during the "stream of consciousness") and during reading of single nouns: microstates preceding reports of spontaneous, visual imagery differed from microstates preceding reports of spontaneous, abstract thought in the same way as microstates after reading an imagery-inducing noun differed from microstates after reading an abstract thought-inducing noun. In both experiments, LORETA functional tomography showed stronger right posterior activity for visual imagery, and stronger left anterior activity for abstract thought, regardless whether spontaneously occurring or reading-induced. These results suggest that it is promising to work towards a dictionary of the psychophysiological "atoms of thought", the building blocks of mentation.
The relative dearth of effective interventions for motor disorders in stroke patients and those with traumatic brain disease represents one of the most challenging problems of rehabilitation medicine and rehabilitation psychology. Although a broad variety of methods has been applied in different fields of rehabilitation only a few of them were rigorously tested by controlled studies and confirmed to represent successful treatment approaches whose therapeutic short-term effects positively transfer from the treatment environment into the real world of patients. A family of treatment methods, called Constraint-Induced Movement Therapy (CI-Therapy), fist suggested by Taub, represents one recent exception of this dissatisfactory condition. In a recent outcome study we tested this approach empirically in more than 70 patients with persistent movement disorders of the upper extremity following stroke and in 8 stroke children.
One basic idea of this approach hypothesizes that a large portion of movement disorders following stroke are not primarily related to the neurological damage induced by the stroke itself but to a conditioned suppression of movements initiated by a diaschises-related failure of limb functioning and by extensive compensatory use of the less-affected limb during the post-stroke period. As shown by our study and similar studies by others many stroke patients are able to reachive good movement functions with their stroke affected upper limb when systematically trained to overcome this learned suppression of motor function by intensive behavioral training of the affected limb for a period of 12 consecutive days and many hours of training per day. The talk will briefly summarize some of the initial animal and human experimental studies from which this approach was generated and then present some of its basic treatment principles. Furthermore, evidence from experimental studies on brain electrical activities and imaging of brain activities during voluntary and passive movements in stroke patients pre and post treatment will be presented indicating that positive treatment effects of CI-Therapy are associated with massive cortical plasticity and reorganization of cortex areas involved in motor control. It is hypothesized that the long lasting behavioral effects of CI-Therapy are based on treatment-induced cortical reorganization of such motor output areas.
Most antidepressant drugs lead to enhanced synaptic availability of the neurotransmitters serotonine and/or norepinephrine. However, affecting also other transmitters, e.g. acetylcholine, antidepressants cause peripheral autonomic dysfunction (e.g. dry mouth, tachycardia). Aim of our study was to objectify these autonomic dysfunction with respect to its consequences for biofeedback applications in psychosomatic and psychiatric patients.
Therefore, we applied simultanous recordings of ECG for assessment of heart rate variability (HRV), as well as skin blood flow and skin conductance level - indicating peripheral autonomic responses like inspiratory gasp response (IGR) and skin conductance response (SCR) - to patients under treatment with psychotropic drugs (amitriptyline, olanzapine, clozapine, fluoxetine, or hypericum extract; n=20 each).
We found that heart rate variability was reduced in all patients treated with amitriptyline, olancapine, or clozapine but not under treatment with fluoxetine, or hypericum. Exclusively in amitriptyline-, olancapine-, clozapine-treated patients 1) redilation of IGR was prolonged, indicating inhibition of norepinephrine re-uptake, and 2) in about 50% of these patients SCR was blocked completely, or reduced in the other 50% (due to anticholinergic effects).
We suggest that these autonomic dysfunction are due to various interactions of these drugs with neuronal structures and pathways (e.g. alpha 1-, alpha 2-, M2-, M 3-receptors, inhibition of norepinephrine re-uptake).
Applying biofeedback on psychiatric or psychosomatic patients, one should keep in mind that some psychotropic drugs interact with autonomic functions.
Direct brain-computer communication utilizes self-regulation of brain potentials to select letters, words or symbols from a computer menu. Selection of letters or words with brain potentials requires simultaneous processing of several tasks such as production of certain brain potentials at predefined time points simultaneously with processing of presented letter strings. This study addresses the question if the self-regulation of slow cortical potentials (SCP) automatizes with practice and can thus be considered as a skill comparable to motor or cognitive skills. Two nearly completely paralysed patients learned over several months to produce electrocortically negative and positive SCP by means of visual feedback. Improved performance and a reduction in performance variability were regarded as behavioural indicators for automaticity, while the topographic focalisation of cortical activation was considered as a neurophysiological indicator for automaticity. Different indicators of automaticity were expected to covary along with practice. In patient 1, performance measured as the percentage of correct SCP shifts increased simultaneously with the topographic focalisation of cortical activation. His performance became more stable with practice. For this patient the criteria for automaticity were all met. In patient 2, performance also inproved, but his cortical activity became topographically less focal. His performance was less stable than patient 1's.
Neurofeedback was systematically evaluated as an adjunctive treatment for drug and alcohol addiction at a residential center in a large-scale study of 121 subjects. Participants had principally abused methamphetamine, heroin, cocaine (including crack), and alcohol, either alone or in combination.
Two treatments were compared: 1) the control, which consisted of the standard treatment provided at the residential center, based on the Minesota Model incorporating sustained involvement in group work to support abstinence; 2) the experimental, which consisted of the control treatment plus neurofeedback. To minimize the possibility that treatment outcome differences were attributable to total time spent in therapy, the control participants were provided extra group and individual psychotherapy sessions, so that their hours of participation in treatment matched that of the experimental participants. Neurofeedback protocols included SMR-beta and alpha-theta training. The neurofeedback was individualized to the participant based on assessment data and response to the training protocols. Cognitive data were acquired pre- and post-treatment. Pre-post MMPIs were also obtained.
Results showed that the experimental group had significantly increased retention in therapy compared with controls. Continuous performance test data showed significant improvement in the experimental group, but no significant change in the control group. Analysisof other cognitive
data revealed a significant treatment interaction only for delayed memory performance. Changes in the MMPI were significant at the p=0.005 level for five of ten subscales: Hypochondriasis, Depression, Conversion Hysteria, Schizophrenia, and Social Introversion; there was a significant treatment interaction. Two additional scales, Psychopathic Deviate and Psychasthenia, showed significant change but no treatment interaction. Finally, one and three year follow-up data show a significant treatment interaction: experimentals were more than twice as likely as controls to sustain abstinence, irrespective of drug of choice.
This poster presentation describes the therapy of a 64 years old male who during the previous seven years had suffered two heart attacks and three cerebro-vascular accidents (strokes), leaving him with considerable functional impairment. In particular he had no expressive language skills, did not orient to anybody except his wife, had violent outbursts and was doubly incontinent. He was referred initially for neurofeedback to help with his double incontinence.
A combination of Haemoencephalography (HEG) and Neurofeedback was used. HEG was used initially to help with perfusion levels in the fontal cortex. Once these had stabilized, Neurofeedback was used to help with the incontinence. The presentation will describe in detail the rational for this approach.
HEG was carried out for twelve sessions. The mean perfusion levels rose from a score of 23.4 at assessment to a score of 83. after 10 sessions. This maintained for two further sessions. The use of HEG was successful in improving expressive language function and in contributing to a reduction in aggressive behaviour.
The use of HEG and Neurofeedback led to functional improvement in a markedly impaired individual. However, the pogress of the sessions highlighted the need for careful consideration of the use of Neurofeedback in such an impaired individual.
Chronic Fatigue is a central feature of CFIDS (USA)/ME (UK). Neurofeedback protocols, which involves inhibiting slow wave activity (2-8Hz) whilst recruiting faster wave activity (12-16Hz) in the frontal cortex, have been reported to be of help for this client group.
Results and Conclusions
This presentation will outline the characteristics of each client group and the rationale for Neurofeedback approaches. The implications for medical management and a refining of the diagnostic categories will also be discussed.
Urinary incontinence is the inability to control urination. A frequently used term is Over Active Bladder (OAB). The disorder affects up to ½ - 1% of the adult population but only about 10% of people seek treatment. Treatments conventionally used include medicine, surgery, behavioural therapy, absorbent products and devices. Some of these treatments help manage the problem. This paper describes two single case studies: The first that of a 19 year old female university undergraduate. The second, that of a 64 year old male who had suffered two heart attacks and three cerebro-vascular accidents (strokes) leaving him with considerable functional impairment. He had no expressive language skills, did not orient to anybody except his wife, had violent outbursts and was doubly incontinent.
Both subjects received weekly Neurofeedback Sessions. For the young lady, the protocol was guided by her QEEG and involved posterior alpha-wave training. The protocol for the male, involved a combination of Haemoencephalography (HEG) and Neurofeedback. HEG was used initially to help with perfusion levels in the frontal cortex. Once these had stabilised, Neurofeedback was used to help with the incontinence. The Neurofeedback protocol involved inhibiting hi-beta and theta-wave activity posteriorly, whilst recruiting lo-beta wave activity.
Subject 1: Prior to Neurofeedback, subject reported day-time incontinence occurring 2 – 3 times per week and bedwetting nightly.
Following three Neurofeedback sessions, the young lady reported no incidences of daytime incontinence. The total cessation of bedwetting took longer. Subject 2: HEG was carried out for twelve sessions. The mean perfusion levels rose from a score of 23.4 at assessment to a score of 83.2 after 10 sessions. This maintained for two further sessions. The use of HEG was successful improving expressive language function and in contributing to a reduction in aggressive behaviour. Faecal incontinence eased after 6 Sessions of Neurofeedback and urinary incontinence lessened.
These two single case studies demonstrate the effectiveness of Neurofeedback with incontinence. The second case highlights the need for the careful consideration of the use of Neurofeedback in such an impaired individual.
A large number of ERP connectivity and synchronization studies utilize measures of coherence and phase locking computed from pairs of multichannel scalp electric potential differences. ERP signals contain information on brain activity. However, a well-known fact about the EEG/ERP inverse problem insures that the electric potential difference at a certain scalp location does not necessarily reflect the activity of the underlying cortex. In this study, electric neuronal activity is estimated with standardized low resolution brain electromagnetic tomography (sLORETA) (Pascual-Marqui, Method Find Exp Clin 2002, 24D:5-12). These signals are then used as a basis for assessing, directly, functional intracortical connectivity.
Three-dimensional (3D) spatio-temporal signals of electric neuronal activity are estimated with s LORETA. This method is uniquely capable of exact (zero error) localization. In addition, it has the lowest spatial dispersion as compared to other published 3D linear, discrete, distributed EEG/MEG tomographies. sLORETA provides high time resolution signals of “virtually implanted electrodes” throughout the cortex. These 3D spatio-temporal signals are then analyzed in terms of independent components (Cardoso, Icassp'89:2109-2112), with independence being forced in the time domain, not in the spatial (cortical) domain.
In an hemifield visual ERP experiment, time lagged interhemispheric connections between the visual cortices are demonstrated. These results are remarkably similar to those produced by microstate segmentation modeling.
It may seem paradoxical that “independent” component analysis (ICA) should yield information on “connectivity”, since these two concepts are quite at odds with each other. The fact is that this implementation of ICA, independence is forced on the time series of activations, while the spatial, 3D, dimension contains the information on intracortical connections. Experimental validation has been presented for method, which proves to be a powerful tool for modeling the functional “wiring” of the brain.
One important feature of the brain is its ability to generate oscillations in alpha, beta and gamma bands. Different frequency components can display different reactivity patterns not only in the alpha, beta and gamma bands, but also within these bands. So for example upper and lower alpha component reactivities can show an antagonistic behaviour.
There is strong evidence that gamma activity has to do with the "binding function" and alpha activity with the "gating function". It is speculated that the beta rebound within the first second after movement or somatosensory stimulation is linked with a "resetting function" of networks in sensorimotor cortex.
Electrocorticographic data from subdural electrodes in candidates for epilepsy surgery show that short-lasting gamma oscillations between 60 - 90 Hz are very frequent in a motor task and always embedded in desynchronized alpha activity. The gamma bursts are very focused, whereas the alpha desynchronization is relatively widespread.
The discoveries in the field of neuroscience and biology in the last decade of the 20th century, in particular the superposition of different sensory images, the correlations of activities in different brain regions and the non-locality and persistence of memory, led to the idea that the high brain processes, such as memory and consciousness have a quantum origin. Consequently, many quantum models of memory have been proposed, some assigning special role of quantum rotators. However, the theoretical results obtained so far are of purely qualitative–conceptual nature, and therefore it is important to quantitatively analyze the quantum models.
Applying adiabatic quantum approach, we investigate the quantum aspect of neuronal processes in order to explore the hypothesis of the relation of memory with the interaction between quantum rotators and electromagnetic field. In particular, the possibility that memory is connected to the collective rotational energy levels of molecules is considered, analyzing the system consisting of molecular quantum rotators interacting with the cortical field.
The resulting probabilities of quantum transitions appeared to be dependent on frequency only, and not on the field’s strength, for the actual range of magnitude of the electric field in brain cells. The relevant number of water moleculs, related to probability of transitions that corresponds to maximum information entropy for the frequency of electric field the cytoplasm is exposed to, is shown to be close to the number of water molecules per cortical neuron, estimated from the cortex anatomy data.
The obtained dependence of quantum transitions suggests that the field frequency could be the information-bearing physical quantity; this may justify its adjustment as the basic mechanism of neuronal regulation. The agreement between the numbers of water molecules obtained theoretically and empirically could confirm the correctness of the approach.
(1) Pediatric Clinic, Faculty of Medicine, University of Skopje, Macedonia, (2) Macedonian Academy of Sciences and Arts, Skopje, Macedonia, (3) Republic Institute for Health Protection, Skopje, Macedonia
Measurements of blood lead levels, multimodal pshychophysiological assessment with biofeedback equipment (EEG, EMG, BVP, SC, Temp. and Resp.), cognitive psychologial tests (Raven) and investigation of graphomotor ability (Bender-Gestalt) on randomly selected children (n=31) from Veles have been performed.
The obtained results have shown slightly increased blood lead levels in tested children (mean value 16,51 ìg/dl), reflected correspondingly in the level of intelligence and graphomotor ability. In addition, the autonomous nervous system studied by peripheral biofeedback appeared to function normally, reacting adequately in stress situations. However, the EEG results have shown that only in two children the theta-beta ratio has normal values, while in others the obtained ratios correspond to increased attention deficit. Moreover, the two children with the highest theta-beta ratio showed pathological ADHD findings and also the highest blood lead levels (> 20 ìg/dl).
The suspected correlation between increased blood lead levels due to industrial pollution and changes in EEG, toward increased attention deficit in tested children has been cnfirmed, implying the need for corresponding health care and environmental response measures. Compared to other applied psychometric instruments, neurofeedback appeared to be the most sensitive and discriminative modality.
The central tenet of neurofeedback is that certain brain frequencies can be enhanced independently of others. There has been some doubt in the literature, however, as to whether or not theta waves are amenable to such alteration.
Visual stimulation (VS) using goggles has been put forward as a complement to neurofeedback, and several studies have documented its effectiveness in treating pain, as well as in producing altered states of consciousness and improving academic performance. Its effects on the EEG, however, are not clear, and contradictory findings have been reported.
Results will indicate whether apparent contradictions in the literature on VS and the EEG could be put down to direct stimulation of EEG equipment by VS goggles alone, or was due to individual differences in brain responsiveness to VS stimulation.
Long-range connectivity between anterior and posterior cortical areas may play a very important role in memory processes. We also know that EEG-theta oscillations (around 5 Hz) are functionally related to working memory whereas alpha oscillations (around 10 Hz) seem to reflect long-term memory functions. In the present experiments the role of theta oscillations in integrating distributed neural circuits and different memory functions is investigated.
Two experiments were un. In the first abstract visual patterns and corresponding labels had to be encoded and later the correct label had to be retrieved when a pattern was shown. Stable latency shifts of evoked theta activity between prefrontal and temporo-parietal electrode sites reflecting functional coupling of the underlying brain areas were investigated for memory encoding and retrieval. In the second experiment a task combining working memory and (semantic) long-term memory was run. The interplay between these two memory systems was studied by relating topographical dynamics of theta oscillations with EEG upper alpha power changes.
In the first experiment stable latency shifts of evoked theta activity between bilateral prefrontal and right temporo-parietal electrode sites, with anterior sites leading the posterior ones, were found during memory encoding. Coupling between left prefrontal and bilateral temporo-parietal areas was obtained during retrieval. These hemispheric differences might reflect different proessing demands. Whereas during encoding sequential processing of verbal and visuospatial information was required simultaneous access to this information was needed during retrieval. In the second experiment the following topographical dynamics were found for evoked theta oscillations: Theta waves were first spreading from anterior to posterior sites but then reversed their direction and spread from posterior to anterior. The time point of inversion of direction varied to a great extent between subjects but was related to the onset of upper alpha power decrease, reflecting activation of semantic long-term memory.
These results suggest that (i) storage functions of working memory depend on distributed prefrontal-temporal networks, (ii) theta oscillations seem to integrate neural circuits relevant for memory functions and (iii) that the interplay between EEG theta and alpha oscillations reflects integration of different memory systems.
Brain oscillations in specific frequency bandwidths have been argued to represent various aspects of information processing on distinct neuroanatomical levels. Whereas the lower frequency range (i.e., 1-8 Hz) is suggested to be implicated in the evolutionary older subcortical systems, higher frequencies (i.e., > 8 Hz) are argued to originate from thalamo-cortical and cortico-cortical interactions. In the present study the effects of testosterone on the coupling between the subcortical and cortical frequency bandwidths was investigated.
Sixteen healthy, right-handed volunteers were enrolled in a placebo controlled double-blind, within-subjects, cross-over design. On two testing days, which were separated by three weeks, between 9:00 and 10:00 a.m., participants received a single sublingually administration of 0.5 mg testosterone (with cyclodextrine as carrier) or placebo after which a baseline EEG recording was obtained (impedance: < 5 kOhm, sampling rate: 250 Hz, amplification: 20,000). Spectral power values (Hamming window) were averaged across all epochs within a single baseline and were then transformed to power density values for the delta (1-3 Hz), theta (4-7 Hz), alpha (8-13 Hz) and beta (13-30 Hz) frequency bands.
Paired-samples t-tests revealed a significant increase in delta power after T administration over the midfrontal electrode site [ t(16) = 2.13: p = 0.05 ]. Furthermore, whereas a significant Pearson’s correlation was observed for delta and beta oscillations in the placebo condition [ r(16) = 0.773: p = 0.0001 ], the latter coupling wa completely abolished after testosterone administration [ r(16) = 0.230: p = 0.4 ]. The difference between the two correlations was statistically significant [ Z(16) = 2.55: p = 0.01 ].
The observed increases in subcortical generated delta power, which was automatically followed by cortical-subcortical decoupling, suggests the capability of testosterone to induce decreases in communication between the cortical-subcortical areas.
University of California, Los Angeles, USA
Meaningful neurofeedback must depend ultimately on the principles of procedural learning as applied to operant conditioning of the EEG. This, in turn, is a function of the sequence of EEG events and related neuronal alterations that accompany the training process. Recent advances in the understanding of procedural learning have disclosed the importance of several relatively independent phases, including acquisition, stabilization, and consolidation. Neurofeedback attempts the training of specific EEG events, a process which must appropriately address these phases. This review will focus on a potential EEG marker for learning potentiation and stabilization, and suggest a relationship with the important consolidation function of sleep.
Recent developments in quantitative EEG measurement have provided several new measures of brain function that are amenable to the analysis of both the timing and localization of cortical responses to functional perturbations. This information in turn provides insights into the sequencing and interactions involved in both normal and abnormal cognitive functioning. This presentation will focus on both “event-related EEG responses” in a signal processing task with normal subjects, and regional correlation changes in EEG frequency modulation (“comodulation”) during specific state changes in clinical subjects. In the former, findings suggest that suppression of frontal cortex involvement follows initial signal detection and allows for efficient cognitive response. Sensory and motor EEG changes appear to be stimulus-bound, reflecting mainly sensory arousal and muscle activity. In the latter, functional changes associated with altered processing are fist reflected by dissociation in frontal areas, followed by reduced differentiation in temporal and parietal regions. These and other dynamics noted provide new clues for effective interventions with neurofeedback.
According to the DSM-IV,Mental Retardation is significantly subaverage general intellectual functioning accompanied by significant limitations in adaptive functioning in at least two of the following skill areas: communication, self-care, home living, social/interpersonal skills, use of community resources, self-direction, functional academic skills, work, leisure, health and safety. Prior to this study we have seen very important and valuable effects of Neurofeedback in the study of children with Down Syndrome. This method is used by many clinicians to treat ADHD and GLDO cases so we waned to examine the effects of NF in Mentally Retarded.
All 16 subjects had neurofeedback training. Almost all subjects had increased delta and theta over the cortex according to QEEG. Some of the subjects were not able to read and write; not talk more than two or three words, some had illegible handwriting, almost all had very poor attention, concentration memory, academic failures, impulsive behavior, very poor social skills. One subject had a severe balance problem, two of them had lack of inhibition in social and sexual behavior. The approach is to examine the improvements of these problems according to the training.
16 subjects ranging 6-24 years old attending private learning centers were previously diagnosed with mental retardation at various university hospitals. Sessions were completed between 40-120 depending on the case. Evaluation measures included QEEG analysis, questionaire prepared by the center. NF trainings were performed by Lexicor Biolex software.NXLINK databank was used to determine ifthey have ADHD/ADD/GLDO with clinical judgement of the author. Lexicor QEEG signals were sampled at 128 Hz.
Electrodes were placed according to QEEG analysis at P3-T5, P4-T6, C3-C4, T3-T4, F3-F4, F7-T5 with a 0-4 delta reward and inhibits at 4-8 Hz and T3-T5, T4-T6 with a 12-15 Hz SMR reward and inhibits at 4-8 Hz.F3-Fz ,F4-Fz with 15-18 beta reward and inhibits at 4-8 Hz.
The present study was conducted with the following objectives: (1) to evaluate the ability of the power spectrum of brain electrical activity to predict performance on neuropsychological tests and school achievement tests and (2) to develop a objective EEG Brain Performance variables to estimate the cognitive health of the brain.
Neuropsychological tests (WISC-R) and school achievement tests (WRAT) and 19 channels of eyes closed EEG were obtained from 477 individuals ranging in age from 6 to 18.79 years of age. The results of power spectral analyses of the EEG were correlated with the subtests of the WISC-R and the WRAT using the Pearson product correlation analysis. The top 50 variables with the highest correlations were selected for entry into a step-wise multivariate regression analysis for each sub-test of the WISC-R and WRAT and separate regression equations were obtained, including the 95% confidence intervals. The predictive multivariate regression equations were independently validated by comparing the predicted full scale IQ versus the measured full scale IQ in the original sample of 477 subjects. Another independent validation of the regression equations involved predicting full scale IQ from 280 traumatic brain injured patients whose neuropsychological tests and whose EEG samples were obtained using a different EEG acquisition system at different locations and a different times.
The multivariate R with EEG as the independent variable and neuropsychological test performance as the dependent variable varied from 0.5 to 0.75 at P < .00001. The correlation between the predicted full scale IQ and the measured full scale IQ was 0.72. Independent validation of the regression equations using 280 TBI patients showed normal distributions of predicted IQ and correlations between predicted and observed that ranged from 0.58 to 0.82 (P < .00001).
Multivariate analyses of the EEG correlate with Neuropsychological test scores and school achievement with medium to high effect size. EEG coherence, EEG phase, ratios of power and power asymmetry were the strongest predictors of neuropsychological test performance and intelligence.
(1) Biofeedback Institute of AAPB CZ, Czech Republic, (2) Institute of Educational Psychological Counseling, Czech Republic, (3) EEG Biofeedback Center Chomutov, Czech Republic, (4) EEG Biofeedback Center Zatec, Czech Republic
ADHD and dyslexia were found to have the genetic cause, with markers diagnosable already in infancy. Other factors of brain dysfunctions (perinatal asphyxia, fevers etc.) are mostly gained in the early age. However, children with ADHD/LD enter the basic school without any specific treatment. Authors´ thesis, coming from their clinical experience, projectedthe preschool neurofeedback treatment as the solution. The study of was sponsored by the Ministry of Education grant.
Children visiting "special" kindergartens because of their behavior were chosen for the study. The common cause of their problems was ADHD, frequently coincided with speech disorders, learning disabilities, enuresis, and accident proness. Forty children, age four to seven years, completed the forty sessions training.
ERP response latencies significantly improved. IQ raised more than half of standard deviation; majority of children improved more than ten points in PIQ, one third in VIQ. The total number of thirty-five children improved their performance in the majority of neuropsychological tests. The children's communication skills improved either. The rating by caregiver improved as well. Half of the children was able to enter a regular basic school in the same year after the treatment, the second after one-year delay.
EEG biofeedback was proved as the method to prevent problems of ADHD/LD children in entering the school, the family problems caused by school problems, and the school system troubles. The results suggest neurofeedback could be recommended for the application in a preschool age. The results also open the question, why ADHD recovery is sought as late as after children enter the school, while it would be more reasonable to practice it as the prevention in the preschool age.
(1) Department of Applied Social Sciences, Canterbury Christ Church University College, UK, (2) Department of Cognitive Neuroscience & Behaviour, Imperial College London, Charing Cross Hospital, London, UK
Previous research has shown that training to enhance one spectral component, i.e., low beta (12-15Hz), whilst simultaneously inhibiting another, i.e., theta (4-8Hz), was associated with improved performance on a semantic working memory task. However, it remains unclear whether such improved performance was the result of enhanced semantic processing, improved attentional processing or an improvement in generic working memory. Hence, the aim of this study was to extend and clarify the previous research by examining the effect of neurofeedback training to enhance low-beta and inhibiting theta on measures of attention and working memory.
Twenty participants were randomly allocated to two groups, both of which attended ten sessions of neurofeedback training. Group 1 was given real feedback aimed at enhancing low-beta (12-15Hz) and inhibiting theta (4-8Hz), in contrast Group 2 received false feedback. To monitor participants awareness of group allocation performance questionnaires were administered at the end of each training session asking them to rate their ability to control their EEG. Pre and post neurofeedback training pariticpants from both groups also completed the attention network task (ANT) and an n-back working memory task.
There was no difference in mean performance ratings between the two groups regarding their perceived ability to control their EEG. Nevertheless, for those receiving real neurofeedback training there was no evidence of any change in either their low-beta or theta amplitudes or ratios (p>.1), this was despite a slight increase in points scored. From time1 to time2 both groups exhibited a reduction in commisison and omission errors when completing the n-back task, and a decrease in response latencies when completing the ANT.
Participants who received real feedback and those that received false-feedback both reported similar levels of control with regard to changes in their EEG. Hence, such a paradigm represents a useful method for teasing apart the possible benefits of neurofeedback training relative to possible placebo effects. However, those receiving real feedback failed to exhibit any change in their EEG. This highlights the need for clarification of some of the procedural aspects of neurofeedback training, such as threshold setting, before we can fully asses the cognitive and behavioural effects of neuruofeedback training.
The neurophysiological mechanisms of sleep spindle generation have been well established by numerous studies and involve variations in membrane potentials in both thalamic and corticothalamic networks. Traditionally, predominance of sleep spindles over central or centro-parietal areas was taken for granted. The interest in spindles has been enhanced by recent findings in sleep EEG research. Factors affecting sleep spindles will be briefly explained. Of particular interest are the topographic studies of sleep spindle incidenceor SMR/sigma power that seem to point to the existence of slow- and fast-spindle oscillations. Frontal spindles peaking at about 12 Hz have been found along with spindles of about 14 Hz more prominent over centro-parietal areas. These results have mainly been interpreted in relation to two functionally separated spindle generators. However, lower and higher frequencies of sleep spindles have also been attributed to a single mechanism: the duration of hyperpolarization-rebound sequence in thalamocortical neurons. Indeed, while long hyperpolarizations generate slower EEG frequencies, short hyperpolarizations cause faster EEG frequencies. Based on this and other considerations that will be pointed out, one could also assume that there is only one kind of sleep spindle, and that in some people there is also an anterior peak of alpha activity. Despite the growing body of knowledge, the definitive functional meaning of sleep spindles still remains to be elucidated. The issues that prevent definitive conclusions wil be reviewed. Furthermore, the thalamocortical system seems to be characterized by plastic mechanisms in which corticothalamic projections could play a fundamental role. Possible implications for EEG neurofeedback training will be addressed. It would also be of theoretical interest to investigate whether distinct neurofeedback training protocols (within and outside the sleep spindle frequency range) could induce different changes in spindle parameters, thereby shedding more light on the functional significance of sleep spindles.
A variety of functional brain imaging methods are being used to identify cerebral areas with abnormal activity and to select targets for possible intervention. These technologies have good temporal resolution, are relatively noninvasive, such as fluoro-deoxygluose positron emission tomography (FDG PET), functional magnetic resonance (fMRI), magnetoencephalography (MEG), and single photon emission tomography (SPECT). Nar infrared spectroscopy (NIRS) is one of the recently developed methodologies for continuous monitoring of alterations in oxygenated (oxy-Hb) and deoxygenated (deoxy-Hb) haemoglobin using near-infrared light, which penetrates biological tissues.
Normal subjects and patients with depression and anxiety (DSM-IV) were examined by NeuroBek-NIRS (NBCW-IR10), a multichannel near-infrared spectroscopy (ten channels) with Continuous Wave (CW) source which allows human subjects to observe and control changes of their own blood oxygen level-dependent (BOLD) response. NBS-RTsys 64-channel-resolved-optical-tomographic-imaging is used with the following technical parameters : gain of 10,000, filters pass-band between 0,3-30 Hz, 60 Hz “notch” filter of noise 2µv RMS. EEGs were recorded from the 16 electrode site of the international 10-20 system while the subjects were at rest with their eyes closed. New algorithms were developed to estimate alterations of oxyhaemoglobin (oxy-Hb), some deoxyhaemoglobin (deoy-Hb) and Total haemoglobin (HbT) during the performance of cognitive tasks. Plus, verbal repetition task, verbal fluency test, T.O.V.A and MicroCog were performed. Fronto-parieto-occipito-temporal (Fp1-Fp2-P3-P4-01-02-T3-T4) regions were measured. Skin conductance response (SCR) was examined too.
Significantly enhance SCR and oxy-Hb in fronto-orbital region during Microcog's task in depressed patients; deoxy-Hb decrease compared with the control group. Correlations at p<.01. between qEEG/ERPs and NBCW-IR10 were observed. Statistical significance was defined as p<0.005, difference statistically significant p<0.001.
The functional BOLD NIRS is a promising method for serial non-invasive bedside CBF measurements. This method can offer therapeutic possibilities in certain cases an alternative, a relay and a complement in aid of medicines.
Donald H. Bars, PhD
Research Consultant, University of Basel, Medical School, Psychiatric and Neurological clinics
The use of sensory evoked responses, evoked potentials (EP), can improve the effectiveness of therapeutic interventions. Evoked potential information is different than the data from other imaging techniques in that it registers responses that occur within the first 3 to 4 tenths of a second, post stimuli, most of which is an obligatory response to sensory reception. This decreases effects of other electrophysiological variables entering the results as more cognitive networks and environmental factors become involved in processing the presented stimuli.
This presentation will address several clinical research studies as the basis for gaining a beginning understanding of relationships between specific visual and auditory EP's and behaviors. A basics Background will allow an understanding of what an EP is, how an EP is created, and important considerations for clinical use. Several different paradigms will be presented and discussed, including how the created waveforms have been associated with a variety of brain functioning. Research will be presented concerning separation of explosive, out-of-control behaviors into several different physiological groups, a physiologically type of ruminating behavior and when combined with qEEG results can separate affective disorders and disorders of attention.
Marco Congedo, MA (1,2), David Joffe, BA (3) & Joel Lubar, PhD (1)
(1) Brain Research and Neurophysiology Laboratory, Department of Psychology, The University of Tennessee, Knoxville, (2) Nova Tech EEG, Inc., Knoxville, Tennessee, (3) Lexicor Medical Technology, Inc. Boulder, Colorado
A major limitation of current neurofeedback paradigma is the limited information provided by a single or a small number of electrodes placed on the scalp. A considerable improvement of the neurofeedback efficacy and specificity could be obtained feeding back brain activity of delimited structures. While traditional EEG information reflects the superposition of the electrical activity of a large number of neurons, by means of inverse solutions such as the Low-Resolution Electromagnetic Tomography (LORETA) spatially delimited brain activity can be evaluated in neocortical tissue.
Three individuals were trained to improve brain activation (suppress low alpha (8-10 Hz) and enhance Beta (16-20 Hz) current density) in the anterior cingulate gyrus cognitive division (ACcd). Participants took part of six experimental sessions, each lasting approximately 30 minutes. Six EEG baselines were recorded for each subject before the training started. Permutation test are conducted on baselines recordings before and after the beginning of the training. In addition a randomized trial was performed at the end of the treatment. During eight two-minutes periods (trials) participants were asked to try to obtain as many rewards as they could (4 "plus" trials) or as few rewards as they could (4 "minus" trials). The order of trials was decided at random. The hypothesis under testing was that participants acquired volitional control over their brain activity so to be able to obtain more rewards during the plus condition as compared to the minus condition.
An exact test of the randomized trials showed evidence of volitional control for two subjects (p=0.043 and p=0.1) and no evidence of volitional control for one of them (p=0.271). The combination of the three p-values provided an overall probability value for this experiment of 0.012 with the additive method and 0.035 with the multiplicative method. These results support the hypothesis of volitional control across the experimental group. Pre-post treatment brain changes are currently under analysis.
With as few as six training sessions, typically insufficient to produce any form of learning with scalp neurofeedback, the experiment showed overall signs of volitional control of the electrical activity of the ACcd. While the pilot study intended to test technical issues, we are now carrying out a larger study to test the ability to change intracranial electrical activity. Tomographic neurofeedback has never been done before. Possible applications of the technique are important and include the treatment of epileptic foci, the treatment of specific brain regions damaged as a consequence of traumatic brain injury, and in general of any specific cortical electrical activity.
Tobias Egner, PhD (1), Thomas F. Zech, PhD (2), & John H. Gruzelier, PhD (1)
(1) Department of Cognitive Neuroscience & Behaviour, Faculty of Medicine, Imperial College London, (2) Department of Clinical Psychology and Psychotherapy, Philipps-University of Marburg, Germany
Although important strides have been made towards establishing the clinical efficacy of neurofeedback of frequency components, a precise understanding of the impact of neurofeedback training on post-training spectral EEG topography, which is presumed to mediate any cognitive-behavioural effects, remains elusive. In order to elucidate the influence of commonly applied neurofeedback protocols on spectral EEG composition, two studies involving healthy participants were carried out.
In experiment 1, 22 subjects were trained on SMR, beta1, and alpha/theta protocols, with resting and on-task spectral EEG measures assessed before and after training. The specific associations between learning indices of each individual training protocol and changes in delta, theta, alpha, SMR, beta1, and beta2 absolute and relative activity levels in 11 scalp regions were assessed by means of partial correlation analyses. The results of experiment 1 served to generate hypotheses for experiment 2, where 26 subjects were randomly allocated to independent groups of SMR, beta1, and alpha/theta training. Spectral resting and on-task EEG measures again were recorded prior and subsequent to training, and analysed for significant changes within each group.
Only few associations between particular protocols and spectral EEG changes were found to be consistent across the 2 studies, and these did not correspond to expectations based on the operant contingencies trained. SMR training was found to be associated with reduced SMR levels in the relative resting EEG, while alpha/theta training was reliably associated with increased general slow wave (delta/theta) and reduced frontal beta band (beta1/beta2) activity in the relative resting EEG.
The results document that neurofeedback of frequency components does affect spectral EEG topography in healthy subjects, but that these effects do not necessarily correspond to either the frequencies or the scalp locations addressed by the training protocols, underline the complexity of the neural dynamics involved. The association between alpha/theta training and replicable reductions in frontal beta activity can be interpreted as empirical neurophysiological evidence supporting the conceptualisation of this training regime as reducing agitation and anxiety.
Sebern Fischer, MA
Many problematic behaviors are driven by fear, though there is no direct mention of fear in the DSM-IV. Fear or fear-based memory is a common element in a number of psychopathological conditions, including: Asperger's Syndrome, Post-Traumatic Stress Disorder, Generalized Anxiety Disorder, Reactive Attachment Disorder, Borderline Personality Disorder, Dissociative Identity Disorder, and many addictive disorders. The amygdala is responsible for generating fear. LeDoux (1996) and others postulate that the amygdala plays a central role in fear-based memory.
According to LeDoux, the low road from the amygdala leads directly to the orbitofrontal area of the brain. This relationship suggests the hypothesis that employing neurofeedback to train the amygdala may provide a mechanism for addressing fear-associated disorders.
A site, located below FP2 on the 10/20 system, was chosen to allow optimal access to the orbital frontal afferent pathway to the amygdala. This site has been designated as FPO2. Twenty-six individuals have been trained at this site in a clinical setting. The detail of session data will be discussed. Subjective reports were obtained following each session and logged in session notes.
Patients subjective reports included a reduction in fear, as well as a sense of calm and well-being, after training at FPO2. This change in affect appears to persist after the neurofeedback session.
In addition to the 26 patients treated by the author at the FPO2 site, a substantial body of anecdotal experience provided by other practitioners appears to confirm the efficacy of the protocol. Neurofeedback training at the FPO2 site provides a new protocol for clinically reducing fear level, as well as reducing the negative affect associated with fear-based memories.
LeDoux, J. (1996). The Emotional Brain. New York: Simon & Schuster.
Vera Grin-Iatsenko, PhD
Institute of the Human Brain, Russian Academy of Sciences
The effect of a EEG Biofeedback procedure consisting of rewarding enhanced EEG amplitudes of the sensorimotor (12-15 Hz) and beta 1 (15-18 Hz) rhythms on the physiological and electrophysiological parameters of attention was studied using the methods of scalp recording of evoked potentials in the bistimulus paradigm Go/NoGo and a physiological attention test (Test of Variables of Attention, TOVA). Eighty-five children with Attention Deficit Disorder were included in the study. EEG Biofeedback sessions significantly improved the attention parameters, behavior and school results in 71 (82,5%) children. In these cases a significant decrease in the amplitude of the N2 (after first and second stimuli) and increase in the amplitude of the P3 (after second stimuli) components in the frontal and central sites and improvement of the TOVA parameters were found.
John Gruzelier, PhD & Tobias Egner, PhD
Department of Cognitive Neuroscience & Behaviour, Faculty of Medicine, Imperial College London
Biofeedback-assisted modulation of electroencephalographic (EEG) activity (neurofeedback), has recently been shown to improve cognitive performance and affect event-related brain potentials in healthy humans (Egner & Gruzelier, 2001, NeuroReport, 12, 4155-4160). These data encouraged a series of investigations into the potential of neurofeedback for enhancing normal function on a non-laboratory measure: music performance in conservatoire students.
In one study, participants were randomly allocated to either training on various neurofeedback protocols only, neurofeedback training plus physical and cognitive exercises, or a no-training control group. Music performances were video-recorded prior and subsequent to training, randomised, and evaluated by blind expert judges. Trends for improvement in musical performance were exclusive to the group that received neurofeedback training only, and improvements were highly correlated with learning to progressively raise theta (5-8 Hz) over alpha (8-11 Hz) band amplitudes (alpha/theta protocol). In a replication study, participants were randomly allocated to groups of alpha/theta neurofeedback, SMR neurofeedback, beta1 neurofeedback, physical exercise, cognitive exercise, or Alexander technique training. The results replicated those of the first study, as only the alpha/theta training group displayed significant music performance improvement.
These data demonstrate the potential for slow wave neurofeedback to improve a complex set of skills in a real-life context.
Jay Gunkelman, QEEGT, QEEGD
The ability to identify comorbidity in the qEEG requires knowledge of the presentation of the individual features, as well as the methods of display that demonstrate these features. This paper will attempt to use a case study to help identify comorbid features in head injury and bipolar disorder. A 45 year old female with a well documented history of bipolar disorder and addiction presented to us with a complaint of recent mild traumatic brain injury (MTBI). The left frontal trauma occurred when she had a motor vehicle accident, hitting a bridge abutment and being struck with the airbag, as well as a secondary crash into the guard rail on the other side of the highway, with a blackened eye and severe bruising on the left forehead.
She had a negative CT-Scan (normal). Her cognitive complaints were slowing and slurring of speech, word retrieval and vocabulary problems, and an inability to comprehend and attend while reading. Methods A clinical EEG and medically interpreted qEEG were performed. The qEEG database used was the Nx-Link database (E. Roy John), and there were additional processing done using the QND software to provide remontaging to a Laplacian montage, as well as to provide single Hz frequency resolution and spectral plots. The artifacted data was also processed using NovaTech's processing software and the LORETA algorithm developed by Roberto Pascal-Marqui (Key Institute, University of Zurich). Results The database showed elevated beta values (Z > 2.0), which is reported in bipolar disorder (L. S. Prichep, et al., 1990; L. S. Prichep and E. R. John, 1992), and the LORETA database showed excess beta (P>0.05), as a red coloration, with the exception of the left frontal area, which showed a statistical decrease from the mean, as a blue coloration, in contrast to the rest of the brain. The linked ear database showed no such left frontal beta deficit, though the remontaged Laplacian mapping of the spectral data did show the beta deficit.
NF and EEG/qEEG in mild traumatic brain injury (MTBI) have a dramatically increased literature (R.W. Thatcher, et al., 1997) with increasing attention to the application to bipolar disorder (S. Othmer, 2001), and an increasing literature in the neurophysiology of bipolar disorder (J. Hughes and E.R. John, 1999) also noted. This case illustrates both bipolar depression and MTBI, including an analysis with LORETA. Bipolar disorder's frontal beta excess is clearly contrasted with a decreased beta due to the head injury. A clear differentiation between white matter damage, which increases delta, and gray matter damage, which decrease both alpha (which recovers in time) and beta, which remains decreased, is discussed. The relative strengths and weaknesses of various qEEG measures for identification of these subtle differential findings will be discussed, including remontaging, absolute-versus relative power, coherence, discriminant analysis, and the use of LORETA.
Wolfgang Klimesch, PhD
Department of Physiological Psychology
University of Salzburg, Austria
Experiments from our laboratory suggest that event-related desynchronization (ERD) in the upper alpha band - a frequency band of 2 Hz width lying above individual alpha frequency (IAF) - reflects semantic long term memory processes, whereas event-related synchronization (ERS) in the theta band - a frequency band of 2 Hz width lying about 4 Hz below IAF reflects working memory processes. Recent findings supporting the suggested hypothesis will be reviewed briefly. In addition data will be presented showing that memory performance - and cognitive performance in general - does not only depend on the magnitude of ERD/ERS but on (tonic) resting or reference power as well. We found that good performance is related to two types of EEG phenomena (i) a tonic increase in alpha but a decrease in theta power, and (ii) a large phasic (event-related) decrease in alpha but increase in theta, depending on the type of memory demands. Based on these findings we recently performed a repetitive transcranial magnetic stimulation (rTMS) study (in collaboration with Ch. Gerloff, University of Tübingen). The aim was to improve cognitive performance by applying rTMS at IAF during a reference interval before the performance of a mental rotation task on a trial per trial basis. We predicted that rTMs at IAF would increase tonic and decrease phasic alpha power, thus leading to increased performance. Control conditions were rTMS at lower IAF (IAF - 3 HZ) and at 20 Hz. The results indicate that rTMS only at a frequency of IAF improved task performance and, concomitantly, increased tonic and decreased phasic alpha power. These findings indicate that the relationship between alpha oscillations and cognitive performance is causal rather than correlative in nature. Implications for biofeedback training will be discussed.
Roberto D. Pascual-Marqui, PhD, Michaela Esslen, PhD & Dietrich Lehmann, MD
The KEY Institute for Brain-Mind Research, University Hospital of Psychiatry, Zurich, Switzerland
A complex object such as the human brain is composed of a very large number of neurons, with a very large number of connections between neurons. Classical functional imaging (PET and fMRI) has emphasized the localization of function, but has practically neglected the role played by functional connections. Oscillations and synchronization, which emerge because of cortical interconnections, have been hypothesized to play an essential role in cognition and consciousness. This work presents some novel methods for the computation and imaging of cortical connectivity.
3D functional imaging of electric neuronal generators is performed with standardized low resolution brain electromagnetic tomography (sLORETA). This new method is uniquely capable of exact localization. In addition, it has the lowest spatial dispersion as compared to other published 3D linear, discrete, distributed EEG/MEG tomographies. Technical details and proof of properties for sLORETA are specified in (Pascual-Marqui RD, Method Find Exp Clin 2002, 24D:5-12). sLORETA provides high time resolution signals of "virtually implanted electrodes" throughout the cortex. Within any given time window, cortical connectivity is modeled in the following way: throughout the cortex, neuronal ensembles have the same dynamics of activation, except for location-specific scale factors and for location-specific latency shifts.
In visual event related potentials to stimulation with pictures of human faces, time lagged connections between primary/secondary visual cortices and the fusiform gyrus ("the face area") are demonstrated.
Validation has been presented for methods that solve an important problem in the neurosciences: the imaging of specialized brain areas and their interconnections
Jordan Pop-Jordanov, Prof. Dr. (1), Nada Pop-Jordanova, MD (2) & Natasa Markovska (2)
(1) Macedonian Academy of Science and Arts, (2) Department of Psychophysiology, Pediatric Clinic, Faculty of Medicine, Skopje, Macedonia
EEG biofeedback is ultimately based on mental-neural information flow, but all standard attempts to explain the mechanism of the subtle two-way interaction between non-material mental agencies and neural events were confronted with violation of the energy-matter conservation laws of physics. In this paper, the consecutive steps of biofeedback signal flow are investigated, considering the fundamental hypothesis of Eccles (Nobelist for Physiology and Medicine), which presumes that mental events cause neural events analogously to the probability fields of quantum mechanics. Applying an advanced quantum mechanical modeling, we calculated the probabilities for quantum transitions within neural molecules in cortical electric field. The results indicated the brain-wave frequency adjustment as an observable information-bearing physical mechanism. Therefrom, some basic rationales for biofeedback efficiency in treating neurological disorders are deduced. In particular, the procedure for attention deficit mitigation through brain wave frequency adjustment i.e. synchronization is studied. As an example the EEG biofeedback technique is applied to a group of ADD children resulting in an improvement of beta-theta ratio, as well as of the intelligence test scores and school grades.
Beverly Steffert, PhD
Birkbeck College, University London
Dyslexia has generally been regarded as a linguistic deficit involving the segmentation of words into phonemes (speech sounds) despite early research identifying abnormal eye movements in 75% of Dyslexics. However no investigation of visual acuity had ever identified any cause until the concept of scotopic sensitivity became known in the 80' s which identified cortical level deficits in the visual pathways. This paralleled known auditory sound discrimination deficits and causes difficulty in segmenting words at the visual level; the print appearing to move and blur, letters can "jump" out of sequence and reading is effortful, slow and tiring. Coloured (spectrum-specific) lenses have previously benefited visual dyslexics. The present research has been funded by an optometric lenses manufacturer to assess objectively any benefits to account for the subjective experience of benefit previously reported by coloured lenses wearers.
43 Subjects (7 to 13 years old) were recruited from an advertisement detailing symptoms of Visual Dyslexia and screened by an optician. Subjects were given coloured lenses (without any optical correction) according to their improvement in pattern perception under a spectrum of coloured light conditions. These measurements were independent of optically corrected lenses.
QEEGs were recorded under 5 conditions (eyes closed, eyes open, reading in daylight, reading in daylight with their coloured lenses, reading in mixed fluorescent and daylight with their coloured lenses). A range of psychometric and balance/co-ordination tests were also given with and without coloured lenses. QEEG data, reading test and balance measures were additionally collected from an control group of 12 normally progressing children with no reading problems.
The psychometric measures showed significant group differences (T-Tests) in "with coloured lenses" and "without coloured lenses" condition at the .01 level or better, for all children.
All children improved in one or other of the measures (working memory, speed of processing or literacy) either in terms of accuracy or of speed, but some showed greater overall improvement. This correlated with the optometric measures of visual improvement in the "with coloured lenses" condition as opposed to the "without coloured lenses" condition. QEEG data showed the visual Dyslexics had an abnormally high Alpha Peak in the without coloured lenses that attenuated when they were wearing their lenses. In other words the "reading without coloured lenses" condition was significantly different to the "reading with coloured lenses" condition in the 9 - 11 Hertz band over O1, O2, T5, T6, P3, P4, for these visual Dyslexics.
These results suggest that prior to wearing coloured lenses their occipital-parietal-temporal areas were not active enough for adequate reading since their "without coloured lenses" measures resembled a normal eyes closed alpha peak. Coloured lenses appear to enhance these reading. Children who suffer from visually related perceptual problems (rather than auditory/phonological deficits) that inhibit reading, are likely to benefit from coloured lenses. This is especially the case under fluorescent lighting. The implications for neurofeedback are discussed in terms of simultaneous activation of the occipital-parietal-temporal cortices.
M. Barry Sterman, PhD
School of Medicine, UCLA
The new Quantitative EEG metric termed "Comodulation" is based on the correlation of trends in spectral magnitudes across time between electrode pairs in relevant frequency bands and functional states. As such it discloses the orchestration of neuronal regulation by revealing the underlying regional integration of neural networks. It has been shown that disturbances of Comodulation in the unique dominant frequency of a given brain during the eyes-closed state can detect the frontal dysfunction of depression that has been documented by other brain imaging methods, as well as disturbed interactions between functional systems along the rostra-caudal axis of the cortex that are associated with somatic and behavioral issues. Further, this measure discloses disturbances in hemispheric interactions related to closed head injury. Focal lesions which corrupt local electrical activity are also identified with this metric. Further, its use clarifies some technical issues associated with EEG recording. The methodology, principles, and findings related to this new dimension of QEEG analysis will be reviewed.
Tanju Surmeli, MD
Living Health Center for Research and Education, Gayrettepe, Istanbul, Turkey
The purpose of this study is to evaluate whether QEEG guided bipolar Neurofeedback training is effective in developing speech, improve attention and concentration, improve learning, decrease behavioral problems or impulsivity, and balance problems of Down Syndrome kids.
Neurofeedback therapy designed to normalize abnormal QEEG scores was provided to determine the effectiveness of this approach.
Significant improvement (>50%) was noted in seven of the subjects. One subject dropped the study after eight sessions. On average, 40 sessions were required. All of the subjects were not able to have T.O.V.A test. All subjects parents reported that their kids were able to speak 4 or 5 meaningful words in a sentence following the treatment, have a meaningful conversation with their parents and the others, their attention and concentration improved dramatically, their behavior became more mature and controllable, no impulsivity seen, long and short term memory improved, and their balance problems improved. Now, they have a thinking brain and they can learn better and faster than before. Most of them have started going to normal school and their teacher reported that " their attention and memory better than their normal peers."
In this uncontrolled open trial of QEEG guided bipolar NF training, the majority of subjects with Down syndrome experienced substantial and rapid symptomatic improvement. Further study with controls and additional outcome measures is warranted. Changes in the approach to people with Down syndrome in the latter part of this century have resulted in a threefold increase in their life expectancy. The average age of death for a person with Down syndrome is in the mid-fifties. Changes with NF may increase in their life expectancy. Anecdotal reports indicate that NF can improve the symptoms of ADHD/ADD/GLDO in a rapid and cost effective way. We may apply this to some of Down syndrome kids who have ADHD/ADD or GLDO. Brain maps collected before, during, and at the conclusion of treatment would enhance our understanding of treatment effects of various Neurofeedback protocols, lead to more focused and productive research, and ultimately facilitate the development of more efficient treatment paradigms.
Michael Thompson, MD & Lynda Thompson, PhD
Biofeedback Institute of Toronto
Autistic spectrum disorder clients have primary deficits in their ability to interpret social communications (innuendo, abstract meaning), appropriately initiate and maintain social interactions, handle anxiety, shift mental set, and sustain external attention span and response control. These deficiencies correspond to patterns observed on the EEG that can include slowing at Pz, P4, T6, F4 and PF1 and differences in coherence and comodulation from normal data bases. High theta activity at Cz is also observed in association with the problems with attention span. This presentation will list symptoms of PDD (Autism) and Asperger's syndrome; describe typical cases; outline assessment and intervention; and give an overview of results of neurofeedback training with >60 cases, ages 5-51 including some with long-term follow-up.
David Vernon, PhD, Tobias Egner, PhD, Nick Cooper, MSc, Theresa Compton, BSc, Claire Neilands, BSc, Amna Sheri, BSc & John Gruzelier, PhD
Department of Cognitive Neuroscience & Behaviour, Imperial College London, Charing Cross Hospital, London
Based on research showing an association between theta (4-7Hz) and working memory, alpha (8-12Hz) and conceptual processing, and the sensorimotor rhythm (SMR) (12-15Hz) and attention, we examined whether training individuals to enhance one of these frequency components would differentially influence their cognitive performance on a range of tasks.
Forty individuals were randomly allocated to one of four groups. Three completed a course of neurofeedback training enhancing a particular component of their EEG (e.g., theta, alpha or SMR) whilst the fourth acted as a non-neurofeedback control group. The training consisted of 8 sessions of neurofeedback over four weeks. Pre and post training all groups completed a range of tasks measuring working memory, image rotation and attention.
Only those training the SMR rhythm exhibited changes in their SMR/theta and SMR/beta EEG ratios. This resulted from a decrease in both theta and beta amplitudes and a concurrent increase in SMR amplitude. This group also showed a significant improvement in working memory performance and limited improvement in attentional performance.
Eight sessions of neurofeedback training is sufficient for healthy individuals to exhibit increased SMR activity. Enhanced SMR activity is associated with improved performance on a semantic working memory task and an attention task. We discuss these results in terms of different frequency components influencing distinct cognitive processes.
Edwin Verstraeten, PhD & Raymond Cluydts, PhD
Dept. of Cognitive & Physiological Psychology
Vrije Universiteit/Free University Brussels, Belgium
EEG slowing and often a prolonged P300 latency have been demonstrated in sleep apnea. These results were mostly interpreted as evidence of cerebral damage due to intermittent hypoxemia. However, as reduced alertness also produces EEG slowing or longer P300 latencies, it seems that alertness and cognitive processes were significantly confounded.
Alpha power desynchronization was used because of its differential responsivity of 8-10 Hz and 10-12 Hz to non-specific alertness and task-related attention, respectively (see e.g., Klimesch's and Sterman's work). Recently, we have shown that desynchronization of 8-10 Hz seemed to be related to phasic alertness, whereas 10-12 Hz synchronization was associated with inaccurate attentional switching (Verstraeten & Cluydts, 2002, NeuroReport, 13, 681-684). Twenty sleep apnea patients and 24 age- and education-matched controls took part in this study. The EEG was recorded from AF3 during resting baseline and during attentional switching.
Control subjects showed a decreasing trend in 8-10 Hz power during attentional switching, which was significantly different from patients who did not show any task-related alpha reactivity response.
Sleep apnea patients demonstrate a blunted phasic prefrontal activation during attentional switching. It is assumed that dampening of the diffuse thalamocortical projection system has caused a reduced level of phasic alertness that can explain the attenuation of prefrontal activation ¾ without the need to presume brain damage.
The Society of Applied Neuroscience (SAN) is a nonprofit membership organization devoted to advancing neuroscientific knowledge and its innovative applications by empowering both scientists and practitioners in serving the public by optimising self-regulatory brain function.
Its membership is open to scientists and practitioners interested in an integrated approach with involves the neural, cognitive and behavioural levels of analysis.
SAN originated as a chapter of the International Society for Neuronal Regulation, an U.S. based organization, grew in numbers and turned into the Society of Applied Neuroscience through a membership referendum which also enlarged its scope. SAN is based in Europe and open to international membership.
SAN is registered in the Norwegian Register Brønnøysundregistrene as Organization Number 984-637-94.