The mismatch negativity during natural sleep: intensity deviants.

Auditory event-related potentials (ERPs) were recorded from 7 subjects who slept for a single night in the laboratory. An 'oddball' sequence of brief tone pips, differing in intensity, was used. Frequently occurring 70 dB 'standards' were presented with infrequent 80 dB intensity increment deviants and 60 dB intensity decrement deviants. The probability of each deviant was 0.1. Stimuli were presented in a random sequence every 600 ms while subjects were awake but inattentive and during stages 2 and REM of sleep. During wakefulness, the intensity increments elicited a broad fronto-central negativity with two discernable peaks. The first, peaking at approximately 120 ms, showed a polarity inversion at the mastoid and likely represented a summation of the N1 wave and the mismatch negativity (MMN). The second, peaking at approximately 330 ms, may have reflected an enhanced N2b component. In REM sleep, the increment deviants elicited a small amplitude 100-200 ms negativity but its amplitude was not significantly larger than the baseline level. It was followed by a larger and significant 300-450 ms negativity but this was considered too delayed to represent the MMN. The decrement deviants elicited a small amplitude, but statistically non-significant, MMN-like wave during both wakefulness and in REM sleep. A MMN-like wave was absent in stage 2 sleep.

Preparatory brain potentials in major depressive disorder.

1. Psychomotor slowing in depression is frequently reflected by delayed reaction times (RT). 2. The role of central arousal mechanisms in response slowing was examined by comparing scalp-recorded slow negative potentials of depressed patients with normal controls in two separate studies. 3. Varying fore-warned RT conditions elicited contingent negative variation (CNV) waveforms and the resultant mid-point amplitudes of these waveforms together with orienting (O-wave), expectancy (E-wave) and post-imperative negative variation (PINV) component amplitudes and sensory evoked responses (N1, P2) were compared between groups. 4. RTs were significantly slowed in depressed patients and the patient group exhibited consistently larger PINV amplitudes. Depending on the RT condition, patients also exhibited larger mid-point CNV amplitudes and smaller N1 and P2 amplitudes.

Event-related potentials and selective attention in major depressive illness.

Depressive episodes have been frequently characterized by deficits in information processing efficiency which are particularly evident when required to sustain or focus attention. As cerebral event-related potentials (ERPs) have been shown to reflect various aspects of selective attention and attentional dysfunction, this study attempted to examine ERPs in depressed patients performing a selective auditory attention task. Twenty-nine patients with a diagnosis of major depressive disorder (DSM-III) and 15 normal, non-psychiatric controls served as experimental subjects. Auditory potentials were recorded from the vertex of subjects who listened selectively to a series of tone pips in one ear and ignored concurrent tone pips to the other ear. Tone pips were delivered at short (320-500 ms) interstimulus intervals and subjects were required to detect, within the attended ear, rare 'target' tones of a different pitch than the more frequent 'standard' tones. In addition to behavioral indices of 'hits' and 'false alarms', ERP-derived measures included N1 amplitudes to attended and ignored stimuli, 'coefficients of attention' as calculated from N1 amplitude ratios and the latency onset and amplitude of the 'negative difference' (Nd) wave resulting from the subtraction of attended and ignored waveforms. Behavioral measures indicated that depressed patients were as efficient as controls in task performance and in fact they exhibited a significant left ear advantage in the additional task. Although a significant 'N1 effect' was observed with attended tones eliciting larger amplitudes than unattended tones, ERP measures of selective attention did not tend to differentiate the two groups.

Subjective and brain-evoked responses to electrical pain stimulation: effects of cigarette smoking and warning condition.

Infrahuman studies employing behavioral indices of pain reactivity have supported a central antinociceptive action of nicotine which appears to be selective and dependent on the class of pain elicited. Human investigations employing subjectively based ratings and judgments of pain intensity have been less conclusive regarding the painfulness of stimuli following nicotine/smoking administration. As the more objective brain-evoked potential (EP) measure has been shown to reflect pain intensity and to be sensitive to a variety of analgesics, this study attempted to examine, together with subjective responses, the effects of cigarette smoking on EPs to pain stimuli administered under varying warning conditions. Twelve male and twelve female smokers served as experimental subjects. In smoking and nonsmoking sessions, subjective intensity ratings (SR) and vertex EPs were assessed in response to electrical skin stimuli presented at a level 20% above pain threshold. Stimulation was either nonwarned or warned with warning conditions involving single or repeated presentations of electrical current at constant or increasing intensities 12 seconds prior to pain stimulation. SRs and peak-to-peak N1-P2 EP amplitudes were measured for each smoking session and warning condition. A significant condition effect was observed for SRs with increasing prepain warning stimulation resulting in the greatest pain ratings. Although smoking did not directly alter SRs or EPs to pain, smoking exposure, as measured by carbon monoxide, was found to be differentially correlated with EP alterations in male and female smokers depending on the warning condition.

The mismatch negativity to frequency deviants during the transition from wakefulness to sleep.

This study examines the changes in the Mismatch Negativity during the transition from a waking, conscious state to one of sleep and unconsciousness. Auditory event-related potentials were recorded from eight participants during the sleep onset period. A 1,000 Hz-standard stimulus was presented every 600 ms. At random, on 20% of the trials, the standard was changed to either a large 2,000-Hz or a small 1,100-Hz deviant. During wakefulness, the large deviant elicited a larger, long-lasting MMN than the small deviant. Following the large deviant during relaxed wakefulness and Stage 2 sleep, the MMN continued to be elicited although it was reduced in amplitude. No significant MMN was recorded for either deviant in Stages 1 and slow wave sleep. The loss of consciousness therefore appears to have a marked effect on the MMN.

The effect of sleep onset on event related potentials with rapid rates of stimulus presentation.

Amplitude changes to the auditory P1-N1-P2-N2 event-related potential (ERP) complex were measured during sleep onset (SO). A rapid rate of stimulus presentation (every 600 ms) was employed. Eight young adults were tested during alert wakefulness, relaxed wakefulness, and during Stages 1, 2, 3 and 4 of sleep. Subjects were repeatedly awakened and the procedure replicated to ensure a sufficient number of trials were presented within each of the sub-intervals. A large fronto-central P1-N1-P2-N2 complex was observed in the Alert Wakefulness condition. A negative slow wave that overlapped the P1-N1-P2 complex became increasingly attenuated as the subject became more drowsy and finally entered sleep. N1 thus became increasingly attenuated and P1 and P2 increasingly augmented during sleep onset. This difference attained statistical significance during Stage 1 of sleep. N2 did not increase in amplitude at SO, as has been reported in other studies.

Scalp topography of the auditory evoked K-complex in stage 2 and slow wave sleep.

During NREM sleep a very large amplitude wave-form, known as the K-complex, may be elicited upon presentation of an external stimulus. The present study compared the scalp distribution of a prominent negative wave peaking at about 550 ms and a later positive wave peaking between 900 and 1300 ms in stage 2 and slow wave sleep (SWS). Nine subjects spent a single night in the laboratory. They were presented with an 80 dB SPL 2000 Hz auditory tone pip every 15 s. The EEG was recorded from 29 electrode sites and referenced to the nose. A K-complex was elicited on 34% of trials in stage 2 and on 46% of trials in SWS. A negative wave peaking at 330 ms was larger on trials in which the K-complex was elicited than on trials in which it was not. The large amplitude N550 was readily observable on trials in which the K-complex was elicited but could not be observed on trials in which it was not. The N550 was bilaterally symmetrical and was maximum over fronto-central areas of the scalp in both stage 2 and SWS. It inverted in polarity at the mastoid and inferior parietal regions. The scalp distribution of N550 significantly differed between stage 2 and SWS. It showed a sharper decline in amplitude over parietal and posterior-inferior areas of the scalp in stage 2 compared to SWS. A later P900 was maximum over centro-frontal areas of the scalp and was also bilaterally symmetrical. It showed a significantly sharper decline in amplitude over widespread inferior areas during SWS. Because the scalp maps of the N550 and P900 are different in stage 2 and SWS, their intracranial sources must also be different.