Are the electroencephalograms mainly rhythmic? Assessment of periodicity in wide-band time series.

To test the hypotheses that (i). electroencephalograms (EEGs) are largely made up of oscillations at many frequencies and (ii). that the peaks in the power spectra represent oscillations, we applied a new method, called the period specific average (PSA) to a wide sample of EEGs. Both hypotheses can be rejected. Although the principal peaks in the two spectra agree most of the time, quite often a peak in the power spectrum accompanies no periodicity peak and some periodicity peaks have no power spectral peak. The Fourier spectrum is not a reliable indication of rhythms. EEG samples from patients during waking, sleeping and seizure states, and volunteer healthy subjects doing cognitive tasks quite often show no significant rhythms, on an arbitrary, common sense definition. When clear rhythms are seen, they involve one or two, rarely up to four or five simultaneous non-harmonically related frequencies. Rhythms are special cases; most of the power spectrum most of the time is nonrhythmic. "Good" rhythms usually have quite narrow peaks, with frequency modulation of <5%, strengths of >2.5 up to >10 times the expectation from chance, and they often show fine structure by being quite local and brief. Most rhythms are quasisinusoidal but others are sharp-cornered recurrent events with <50% duty cycle. In the face of wide variability, we do not report any systematic differences in periodicity among EEGs from different parts of the brain or different brain states or species; it will take many more exemplars of each state, species or brain part to establish characteristic features. The PSA method may be the best so far proposed to demonstrate and quantify periodicity in wide-band time series with noise, but it has serious limitations. Discussion leads to the conclusion that it is time for a new paradigm or metaphor for brain waves.

Temporal fluctuations in coherence of brain waves.

As a measure of dynamical structure, short-term fluctuations of coherence between 0.3 and 100 Hz in the electroencephalogram (EEG) of humans were studied from recordings made by chronic subdural macroelectrodes 5-10 mm apart, on temporal, frontal, and parietal lobes, and from intracranial probes deep in the temporal lobe, including the hippocampus, during sleep, alert, and seizure states. The time series of coherence between adjacent sites calculated every second or less often varies widely in stability over time; sometimes it is stable for half a minute or more. Within 2-min samples, coherence commonly fluctuates by a factor up to 2-3, in all bands, within the time scale of seconds to tens of seconds. The power spectrum of the time series of these fluctuations is broad, extending to 0.02 Hz or slower, and is weighted toward the slower frequencies; little power is faster than 0.5 Hz. Some records show conspicuous swings with a preferred duration of 5-15s, either irregularly or quasirhythmically with a broad peak around 0.1 Hz. Periodicity is not statistically significant in most records. In our sampling, we have not found a consistent difference between lobes of the brain, subdural and depth electrodes, or sleeping and waking states. Seizures generally raise the mean coherence in all frequencies and may reduce the fluctuations by a ceiling effect. The coherence time series of different bands is positively correlated (0.45 overall); significant nonindependence extends for at least two octaves. Coherence fluctuations are quite local; the time series of adjacent electrodes is correlated with that of the nearest neighbor pairs (10 mm) to a coefficient averaging approximately 0.4, falling to approximately 0.2 for neighbors-but-one (20 mm) and to < 0.1 for neighbors-but-two (30 mm). The evidence indicates fine structure in time and space, a dynamic and local determination of this measure of cooperativity. Widely separated frequencies tending to fluctuate together exclude independent oscillators as the general or usual basis of the EEG, although a few rhythms are well known under special conditions. Broad-band events may be the more usual generators. Loci only a few millimeters apart can fluctuate widely in seconds, either in parallel or independently. Scalp EEG coherence cannot be predicted from subdural or deep recordings, or vice versa, and intracortical microelectrodes show still greater coherence fluctuation in space and time. Widely used computations of chaos and dimensionality made upon data from scalp or even subdural or depth electrodes, even when reproducible in successive samples, cannot be considered representative of the brain or the given structure or brain state but only of the scale or view (receptive field) of the electrodes used. Relevant to the evolution of more complex brains, which is an outstanding fact of animal evolution, we believe that measures of cooperativity are likely to be among the dynamic features by which major evolutionary grades of brains differ.

EEG coherence has structure in the millimeter domain: subdural and hippocampal recordings from epileptic patients.

Subdural recordings from 8 patients and depth recordings from 3 patients via rows of electrodes with 5-10 mm spacing were searched for signs of significant local differentiation of coherence calculated between all possible pairs of loci. EEG samples of 2-4 min were taken during 4 states: alertness, stage 2-3 sleep, light surgical anesthesia permitting the patient to respond to questions and electrical seizures. Coherence was computed for all frequencies from 1 to 50 Hz or 0.3-100 Hz; for comparisons the mean coherence over each of 6 or 7 narrower bands between 2 and 70 Hz was used. Whereas the literature supports the view that EEG coherence is usually substantial over many centimeters, the hypothesis here tested--and found to be well above stochastic expectations--is that significant structure occurs in the millimeter domain for EEG recorded subdurally or within the brain. In both the subdural surface samples and those from temporal lobe depth electrode arrays coherence declines with distance between electrodes of the pair, on the average quite severely in millimeters. This is nearly the same for all frequency bands. For middle bands like 8-13 and 13-20 Hz, mean coherence typically declines most steeply in the first 10 mm, from values indistinguishable from 1.0 at < 0.5 mm distance to 0.5 at 5-10 mm and to 0.25 in another 10-20 mm in the subdural surface data. Temporal lobe depth estimates decline about half as fast; coherence > or = 0.5 extends for 9-20 mm and > or = 0.25 for another 20-35mm. Low frequency bands (1-5, 5-8 Hz) usually fall slightly more slowly than high frequency bands (20-35, 35-50 Hz but the difference is small and variance large. The steepness of decline with distance in humans is significantly but only slightly smaller than that we reported earlier for the rabbit and rat, averaging less than one half. Local coherence, for individual pairs of loci, shows differentiation in the millimeter range, i.e., nearest neighbor pairs may be locally well above or below average and this is sustained over minutes. Local highs and lows tend to be similar for widely different frequency bands. Coherence varies quite independently of power, although they are sometimes correlated. Regional differentiation is statistically significant in average coherence among pairs of loci on temporal vs frontal cortex or lateral frontal vs. subfrontal strips in the same patient, but such differences are usually small.(ABSTRACT TRUNCATED AT 400 WORDS)

Dynamic properties of human visual evoked and omitted stimulus potentials.

Visual evoked potentials (VEPs) and omitted stimulus potentials (OSPs) are re-examined in scalp recordings from 19 healthy subjects. The principal finding is a distinction in form, latency and properties between OSPs in the conditioning stimulus range < 2 Hz, used in previous human studies, and those in the range > 5 Hz, used in previous studies of selected elasmobranchs, teleost fish and reptiles. We cannot find OSPs between 2 and 5 Hz. The high frequency ("fast," ca.6- > 40 Hz) and the low frequency ("slow," ca. 0.3-1.6 Hz) OSPs have different forms and latencies but both tend to a constant latency after the omission, over their frequency ranges, suggesting a temporally specific expectation. Fast OSPs (typically N120, P170-230 and later components including induced rhythms at 10-13 Hz) resemble an OFF effect, and require fixation but not attention to the interstimulus interval. Slow OSPs (usually P500-1100) require attention but not fixation; they are multimodal, unlike the fast OSPs. Based on cited data from fish and reptiles, fast OSPs probably arise in the retina, to be modified at each subsequent level. We have no evidence on the origin of slow OSPs. In both ranges not only large, diffuse flashes, but weak, virtual point sources (colored LEDs) meters away suffice. They are difficult to habituate. Both require very short conditioning periods. The transition from the single, rested VEP to the steady state response (SSR) at different frequencies is described. Around 8-15 Hz in most subjects larger SSRs suggest a resonance. Alternation between large and small SSR amplitude occurs around 4 Hz in some subjects and conditions of attention, and correlates with an illusion that the flash frequency is 2 Hz or is irregular. Jitter of the conditioning intervals greatly reduces the slow OSP but only slightly affects the fast OSP. Differences between scalp loci are described.

Changes in EEG power, acoustic evoked potentials and heart rate after mildly arousing non-acoustic priming stimuli in carp (Cyprinus carpio).

1. Changes in EEG power spectrum of carp to a priming non-acoustic stimulus followed by acoustic clicks were compared to those due to acoustic clicks delivered alone. Recordings were made from the telencephalon, midbrain and medulla. Acoustic evoked potentials (AEPs) to the clicks were also recorded. 2. EEG power changes to non-acoustic stimuli occurred over the whole 1-40 Hz frequency range and were regionally specific and consistent. 3. The changes in the EEG midfrequency 12-24 Hz power spectrum to non-acoustic stimuli were significantly correlated with changes in the AEP to subsequent clicks. An elevated medullary AEP amplitude and reduced duration were correlated with increased medullary EEG power and increased midbrain AEP duration. 4. Telencephalic EEG power changes were inversely related to changes in medullary and midbrain AEP amplitude.

Sustained potential shifts and changes in acoustic evoked potentials after presentation of a non-acoustic priming stimulus to carp (Cyprinus carpio).

1. Recordings were made from the region of the midbrain tectum and torus semicircularis of sustained potential shifts (SPS) to a non-acoustic priming stimulus and the change in subsequent acoustic evoked potentials (AEPs) to a train of six clicks after a long rest. 2. In the absence of priming stimuli (a jet of saline or water to the flank) the AEP to the first click in a train had the highest amplitude; with these stimuli it became the most attenuated. 3. The SPS to both non-acoustic stimuli was initially (ca 4 sec) negative, then became positive for a similar time period. 4. After saline jet the tectal and the torus AEP amplitude was significantly correlated with the torus SPS; after water jet, the tectal and the torus AEP durations were correlated with the SPS. 5. Application of alumina gel to the posterior telencephalic border caused elevation of the torus AEP amplitude after some 5 hr.

Coherence of compound field potentials reveals discontinuities in the CA1-subiculum of the hippocampus in freely-moving rats.

The ongoing micro-electroencephalogram was recorded with a chronically implanted comb-like array of 16 tungsten semi-microelectrodes 0.2 or 0.25 mm apart, spanning CA1 strata oriens, pyramidale and radiatum and into subiculum, in four behavioral states: walking, standing still, paradoxical and slow wave sleep and under scopolamine. Power, phase and coherence spectra were computed, the latter two for each of the 120 pairs, in frequency bands from 1 to 64 Hz. (1) Coherence is high for all frequencies within the same subfield, e.g. stratum radiatum, but falls with distance. Theta frequency (8 Hz), when prominent and widespread (during "theta states" walking and paradoxical sleep), shows the most widespread synchrony: coherence falls slowly, from 1.0 at 0.2 mm to 0.7 at c. 2 mm longitudinally within stratum radiatum; all other frequencies fall two or three times faster. (2) An abrupt drop in coherence occurs across field borders (CA1-subiculum) and between stratum oriens and radiatum, across a line just under stratum pyramidale, between high coherence regions on each side of the coherence discontinuity. A less extreme drop occurs in stratum radiatum 0.4 mm from the subiculum border, without obvious histological correlate. The discontinuities in coherence are stable through all four behavioral states as well as under scopolamine. (3) Phase profiles diagonally across CA1 and into subiculum show abrupt, local shifts of phase (up to 125) at these same levels. No gradual shift reaching 180 (phase reversal) occurs in the span of loci examined. (4) The theta power peak in theta states is not necessarily due to additional energy in that band; in some conditions it is mainly due to reduced power in other frequencies. Root mean square voltage is generally less in the high theta ("synchronized") than in the non-theta states. Only the theta peak correlates with a peak in coherence. (5) Significant microstructure in the dynamics of neuronal cooperativity distinguishes behavioral states and regions of the hippocampal cortex.

Lateral coherence of the electrocorticogram: a new measure of brain synchrony.

As one test of the idea that compound field potentials in higher centers have a fine structure, the horizontal extent of coherence (C) was studied on the brain surface, with many closely spaced semimicroelectrodes in rabbits and rats. On the average C tends to fall with distance (D) in the 0.5-10 mm range; apart from driven rhythms, C usually falls to noise level at D greater than 10 mm. A useful measure is D (mm) where C has fallen to 0.5 (DC = 0.5); for most F bands within the range 1-50 Hz this is usually 2.5-5 mm, averaging over the neocortex in both species. Synchrony for neural tissue should mean a degree of congruence in a population (not a 2-point correlation); decline of C with D can measure synchrony by reflecting the volume at or above a specified C. Sleeping and waking mammals, an invertebrate (Aplysia), a ray, and a reptile were compared in degrees of synchrony; this cannot be judged by eye and is found sometimes hardly different between high-voltage-slow and low-voltage-fast states. Aplysia has negligible synchrony; the ray and lizard may be intermediate. C maps show patchiness superimposed on the general decline with D; no obvious pattern between parts of the cortex is consistent among individuals. Factors influencing variance, repeatability and extent of significant C are assessed. Brain size, passive spread, electrode size (at least 1-100 microns) and closeness of contact with pia mater rarely contribute materially, even within 1 mm. C commonly falls moderately with frequency (F) from a maximum between 1 and 8 Hz, usually without consistent peaks except for special cases of driving rhythms, such as theta. Intracortically the distribution of C is more local, both radially and horizontally. Although it was not possible to say when the two electrodes were in the same lamina, most laminae are highly coherent with all others. One or two sharp radial discontinuities in C are common, often but not consistently in the middle layers. C shows no simple relation to distance. In spite of the prevalent high coherence between laminae, radially, C varies widely horizontally from low to high in the 0.1-1 mm range. C is regarded as one aspect of cooperativity in a cellular dynamic system with fine structure in the fractional millimeter and second range; so far we are observing it with severely distorting smoothing procedures.