Jan Berkhout

Contamination of scalp EEG spectrum during contraction of cranio-facial muscles

Abstract Autospectral and coherence values were determined for three EEG derivations, two craniofacial muscles, and the biceps during separate 90 sec periods of strong contraction. Spectra of contracting frontalis and masseter muscles showed peak power between 30 and 60 c/sec, with peak biceps power during contraction centered around 50 c/sec. Over the course of contraction, all three muscles showed an increase in the proportion of low frequency power (below 40 c/sec) and a decrease in power above that frequency. During contraction of the cranio-facial muscles, there were generalized increases in EEG power at all frequencies from vertex-frontal, vertex-mastoid and vertex-temporal derivations. However, these increases were minimal below about 14c/sec unless one electrode of a derivation was located directly over the contracting muscle. Coherences between EEG derivations showed considerable increases during muscle contraction only at frequencies above 14c/sec. In addition, coherence between EEG and EMG derivations were large only when one EEG electrode was located over the muscle. These results indicate that there is very little direct, linear EMG infiltration of the EEG signal below 14 c/sec during tonic contraction of individual muscles unless the EEG electrode is extremely close to a contracting muscle.

Alterations of the human electroencephalogram induced by stressful verbal activity

Abstract Certain auto- and cross-spectral components of the EEG appear to be characteristic of responses to verbal stimuli. Employing a discriminant-analysis procedure applied to spectral parameters, it proved possible to separate subjectively stressful from non-stressful verbal stimuli, and to determine distinctive EEG responses to verbal stimuli of similar stress value differing only in semantic content. The EEG components characteristic of these response states were consistent over small populations, and the criteria developed for their identification proved valid over several different subjects without requiring individual calibration. Other, individual-specific characteristics of the EEG were observed consistently to parallel the cyclical occurrence of several constituent epochs in a twenty-item question-answer sequence.

Temporal Stability and Individual Differences in the Human EEG: An Analysis of Variance of Spectral Values

Normalized power spectra were determined for each of a series of 30 five-and ten-second epochs of eight-channel EEG recordings, taken from 47 subjects under conditions of rest and perceptual task stress.

Simian eeg activity related to problem solving during a simulated space flight

EEG electrode stimulated simian mental activity in problem solving during simulated space flight, discussing skull implantation and EEG recordings of hippocampus activity

Comparative frequency distributions of large and small amplitude rhythms of the human electroencephalogram

Abstract A base line crossing interval tabulating procedure was used to assess the rhythmic activity of a group of single channel occipital bipolar recordings in two discontinuous amplitude ranges. One hundred seconds of record from each of 70 subjects were analyzed. The subjects included 38 normal individuals and 32 with miscellaneous cerebral pathology. Slowing of the large amplitude distribution was noted in the subjects with cerebral pathology. This contrasted strongly with the more static behavior of the small amplitude activity of these subjects. Large amplitude activity at or below 8 c/sec meeting the strict persistence and homogeneity requirements used was found only among the pathological group. The two amplitude ranges had distinctive frequency distributions and appeared to be independently generated. Activity levels of the two ranges were also independent at frequencies where the two distributions overlapped. It was concluded that it would be advisable to incorporate some form of amplitude threshold restrictions into routine frequency analysis.