General Information & Editors

Current Issue

Archives

Subscriptions

Author Guidelines

Advertising Information

Books for Review

Contact Us

Parametric vs. Non-Parametric Statistics of Low Resolution Electromagnetic Tomography (LORETA)

R. W. Thatcher, D. North and C. Biver

ABSTRACT
This study compared the relative statistical sensitivity of non-parametric and parametric statistics of 3-dimensional current sources as estimated by the EEG inverse solution Low Resolution Electromagnetic Tomography (LORETA). One would expect approximately 5% false positives (classification of a normal as abnormal) at the P < .025 level of probability (two tailed test) and approximately 1% false positives at the P < .005 level.

EEG digital samples (2 second intervals sampled 128 Hz, 1 to 2 minutes eyes closed) from 43 normal adult subjects were imported into the Key Institute’s LORETA program. We then used the Key Institute’s cross-spectrum and the Key Institute’s LORETA output files (*.lor) as the 2,394 gray matter pixel representation of 3-dimensional currents at different frequencies. The mean and standard deviation *.lor files were computed for each of the 2,394 gray matter pixels for each of the 43 subjects. Tests of Gaussianity and different transforms were computed in order to best approximate a normal distribution for each frequency and gray matter pixel. The relative sensitivity of parametric vs. non-parametric statistics were compared using a “leave-one-out” cross validation method in which individual normal subjects were withdrawn and then statistically classified as being either normal or abnormal based on the remaining subjects.

Log10 transforms approximated Gaussian distribution in the range of 95% to 99% accuracy. Parametric Z score tests at P < .05 cross-validation demonstrated an average misclassification rate of approximately 4.25%, and range over the 2,394 gray matter pixels was 27.66% to 0.11%. At P < .01 parametric Z score cross-validation false positives were 0.26% and ranged from 6.65% to 0% false positives. The non-parametric Key Institute’s t-max statistic at P < .05 had an average misclassification error rate of 7.64% and ranged from 43.37% to 0.04% false positives. The non-parametric t-max at P < .01 had an average misclassification rate of 6.67% and ranged from 41.34% to 0% false positives of the 2,394 gray matter pixels for any cross-validated normal subject.

In conclusion, adequate approximation to Gaussian distribution and high cross-validation can be achieved by the Key Institute’s LORETA programs by using a log10 transform and parametric statistics, and parametric normative comparisons had lower false positive rates than the non-parametric tests.

Cognitive Impairment in Major Depressive Disorder Revealed by Event-Related Potential N270

Wei Mao, Yuping Wang and Dequan Wang

ABSTRACT
Major depressive disorder (MDD) is a mood disorder that is often accompanied by the impairment of cognitive functions. The aim of this study was to investigate the feasibility of N270 as an index for evaluating the cognitive impairment in MDD patients. Twenty-five patients with MDD diagnosed according to DSM-IV and 25 age matched normal controls performed a matching task while event-related potentials (ERPs) were recorded from their scalp. There were two kinds of stimulus pairs in this study: match condition, the second stimulus (S2) in a pair was identical to the first one (S1); mismatch condition, S2 conflicted with S1 in the color attribute. Subjects were required to press a button in the match condition and to press another button in the mismatch condition.

A negative ERP component, N270, which was considered to reflect conflict processing activity in human brain, was evoked by S2 of the mismatch condition. The patient group showed a delayed and smaller N270 than the control group. The prolongation of its peak latency was significant at P3 and P4 electrodes, and the reduction of its peak amplitude was significant at F3, F4, P3 and P4 electrodes. The amplitude of P300 elicited in the match condition was decreased in the patient group at P3 and P4 electrodes, but its latency did not differ from the control group. The results indicate that MDD patients as a group showed cognitive decline. N270 is a sensitive index in revealing cognitive impairment.

EEG and Seizures in Autistic Children and Adolescents: Further Findings with Therapeutic Implications

John R. Hughes and Michelle Melyn

ABSTRACT
The goal of this study was to investigate the incidence of epilepsy and also the EEG findings among children with autism (A), a devastating disorder, and to compare these data to an EEG control group. EEGs were quantified as to the degree of epileptiform activity and also slow wave abnormalities. Abnormal EEGs were found in 75% of the 59 A children and 82% of their 151 EEGs; 46% had clinical seizures. Nearly all children with seizures had epileptiform activity, but almost 20% of those with spike discharges did not have clinical attacks. Slow wave abnormalities were more frequent and of a greater degree of severity in the A group, compared to controls. One new finding was that a greater variability of EEG abnormalities was found in the A group. Therapeutic implications, based on these latter findings, are discussed.

Approximate Entropy in the Electroencephalogram During Wake and Sleep

Naoto Burioka, Masanori Miyata, Germaine Cornélissen, Franz Halberg, Takao Takeshima, Daniel T. Kaplan, Hisashi Suyama, Masanori Endo, Yoshihiro Maegaki, Takashi Nomura, Yutaka Tomita, Kenji Nakashima and Eiji Shimizu

ABSTRACT
Entropy measurement can discriminate among complex systems, including deterministic, stochastic and composite systems. We evaluated the changes of approximate entropy (ApEn) in signals of the electroencephalogram (EEG) during sleep.

EEG signals were recorded from eight healthy volunteers during nightly sleep. We estimated the values of ApEn in EEG signals in each sleep stage. The ApEn values for EEG signals (mean ± SD) were 0.896 ± 0.264 during eyes-closed waking state, 0.738 ± 0.089 during Stage I, 0.615 ± 0.107 during Stage II, 0.487 ± 0.101 during Stage III, 0.397 ± 0.078 during Stage IV and 0.789 ± 0.182 during REM sleep. The ApEn values were found to differ with statistical significance among the six different stages of consciousness (ANOVA, p<0.001). ApEn of EEG was statistically significantly lower during Stage IV and higher during wake and REM sleep.

We conclude that ApEn measurement can be useful to estimate sleep stages and the complexity in brain activity.

High Frequency EEG Activity During Sleep: Characteristics in Schizophrenia and Depression

Janet L. Tekell, Robert Hoffmann, William Hendrickse, Robert W. Greene, A. John Rush and Roseanne Armitage

ABSTRACT
Previous studies indicate that high frequency power (>20Hz) in the electroencephalogram (EEG) are associated with feature binding and attention. It has been hypothesized that hallucinations and perceptual abnormalities might be linked to irregularities in fast frequency activity. This study examines the power and distribution of high frequency activity (HFA) during sleep in healthy control subjects and unmedicated patients with schizophrenia and depression.
This is a post-hoc analysis of an archival database collected under identical conditions. Groups were compared using multivariate analyses of covariance (MANCOVA) using group frequency by stage analysis. A multiple regression analyzed the association between HFA power and clinical symptoms.

Schizophrenic (SZ) and major depressive disorder (MDD) patients showed significantly greater high frequency (HF) power than healthy controls (HC) in all sleep stages (p<0.0001). SZs also exhibited significantly greater HF power than MDD patients in all sleep stages except wakefulness (W) (p<0.05). In all groups, gamma (35-45Hz) power was greater in W, decreased during slow wave sleep (SWS) and decreased further during rapid eye movement (REM). Beta 2 (20-35 Hz) power was greater in W and REM than in SWS. Only positive symptoms exhibited an association with HF power.
Elevated HFA during sleep in unmedicated patients with SZ and MDD is associated with positive symptoms of illness. It is not clear how HFA would change in relation to clinical improvement, and further study is needed to clarify the association of HFA to the state/trait characteristics of SZ and MDD.

Structure of the Spontaneous All Night Sleep in Epileptics with Polysomnography

W. W. Wang, Q. H. Xie and X. Wu

ABSTRACT
Twenty epileptics and 11 healthy controls were investigated with polysomnography. Epileptics with interictal discharges had significantly longer total recording time and longer REM latency than the controls (P<0.05). Epileptics had a more fragmentized sleep than controls. Sleep efficiency of patients with interictal discharges was lower than that of patients without them (P<0.05). Among the patients with interictal discharges, those with partial seizures had a significantly higher arousal number than those with generalized seizures (P<0.05). The number of NREM stage shifts in patients with partial seizures tended to be higher than in those with generalized seizures.

EEG Coherence Analysis in Never-Medicated Patients with Panic Disorder

Akira Hanaoka, Mitsuru Kikuchi, Ryutarou Komuro, Hiroshi Oka, Tomokazu Kidani and Shunsuke Ichikawa

ABSTRACT
To investigate the functional abnormalities in the central nervous system (CNS) of patients with panic disorder (PD), we compared the electroencephalography (EEG) coherence values in 18 never-medicated PD patients with those in age-matched normal control subjects, and examined the relationships between EEG coherence values and both the duration of disease and the severity of panic attacks. EEG data were recorded in the resting state.
The PD patients had lower coherence values with significant differences in F3-F4, C3-C4, P3-P4, F7-T5, and F8-T6. There were positive correlations for the higher alpha band between coherence values and both the duration of disease and the severity of panic attacks.

These findings provide further evidence that PD patients have a lower degree of inter-hemispheric functional connectivity in the frontal region and intra-hemispheric functional connectivity in the bilateral temporal region, and that chronic condition or frequent panic attacks in PD patients may be related to the pathophysiological CNS changes.

Auditory and Visual P300 Evoked Potentials Do Not Predict Response to Valproate Treatment of Aggression in Patients with Borderline and Antisocial Personality Disorders

Roy R. Reeves, Frederick A. Struve and Gloria Patrick

ABSTRACT
In this study of patients with borderline personality disorder (BPD) or antisocial personality disorder (ASPD) hospitalized because of aggressive behavior, auditory and visual P300 evoked potentials were obtained prior to treatment with valproate. Eight ASPD patients (8 males, 0 females) and 11 BPD patients (2 males, 9 females) showed improvement, while in 7 patients with ASPD (7 males, 0 females) and 10 patients with BPD (2 males, 8 females), aggression was not improved. Differences in auditory and visual P300 latencies and amplitudes were not significant for either diagnosis, or for both diagnoses combined. These findings suggest that auditory or visual P300 evoked potentials may not be useful for predicting response of aggressive behavior to valproate treatment in patients with BPD or ASPD.

Auditory Hallucinations in Psychosis: Dysfunctional Cortical Connectivity in Neural Substrates of Central Auditory Processing and Episodic Verbal Memory

Alex A Sergejew, David Copolov, Mental Health Research Institute, Victoria, Australia
Gary Egan, Mental Health Research Institute, Victoria, Australia, and University of Melbourne, Australia

Introduction: Theories of the neurobiological basis of auditory hallucinations (AHs) in psychosis mainly conceptualise these phenomena as “inner speech” which is misattributed to an external agency.  However, the majority of neuroimaging studies of AHs report no evidence of activation of Broca’s area, an otherwise reliable concommitant of “inner speech.”  However, we have previously reported behavioral evidence of impaired interhemispheric transfer in a hallucinating patient group (McKay et al, American Journal of Psychiatry, 157: 759-766).

Methods: Brain activation associated with self-reported AHs was measured using PET in patients with psychosis (n=8) compared to perception of transient random human speech in non-hallucinating patients (n=7) and normal controls (n=8). Effective connectivity was also compared using both EEG and fMRI measures in two other groups of patients with and without AHs.

Results: Externally generated speech sounds elicited extensive bilateral activation of auditory cortical regions (Brodmann areas 40, 41, 42 and 22). In contrast, hallucinations were associated with a network of activation including bilateral auditory association cortex, left limbic regions and hippocampus, right medial frontal and right prefrontal regions.  Connectivity between left and right auditory association cortex appeared to be lower in patients with hallucinations.

Conclusions: The observed pattern of activation is most consistent with models of auditory hallucinations as mis-remembered episodic memory of speech (Copolov et al, Psychiatry Research: Neuroimaging, 122: 139-152). We formulated a new model of AHs in which AHs were related to reduced connectivity in neural substrates of episodic verbal memory and central auditry processing.

POSTER: Structural Changes of the Corpus Callosum in Subjects with Mild Cognitive Impairment

Philipp A. Thomann, Elmar Kaiser, Peter Schönknecht, and Johannes Schröder, University of Heidelberg, Germany
Torsten Wüstenberg, Humboldt-University of Berlin, Germany;
Frederik L. Giesel, Marco Essig, German Cancer Research Center, Heidelberg, Germany
Johannes Pantel, University of Frankfurt, Germany

Introduction: Previous studies demonstrate a significant atrophy of the corpus callosum (CC) in patients with Alzheimer’s disease (AD). If structural changes of the CC already occur in mild cognitive impairment (MCI) has not yet been investigated.

Methods: In the present study 21 subjects with mild cognitive impairment (MCI) (mean age 66.2 ± 0.75 years), 21 healthy controls (mean age 66.6 ± 0.6 years) and 10 age-matched patients with AD (mean MMSE 19.2 ± 3.85) were investigated using quantitative MRI. After drawing a horizontal line in the mid-sagittal T1 weighted slice from the most anterior to the most posterior point of the CC, it was divided in 5 parts (CC1-5) by constructing vertical lines of equal distance perpendicular to the horizontal line.

Results: The three groups did not differ significantly in intracranial volume. As expected, the CC was significantly smaller in patients with AD than in healthy controls and subjects with MCI. Atrophy was most severe in the rostral parts (CC1-3).

The subjects with MCI did not show significant difference in total callosal area compared to healthy controls, but significant smaller size in the two rostral segments (CC1 + CC2).

Conclusions: These findings support the hypothesis that MCI may represent a preclinical stage of AD.