P300 and the Daytime Consequences of Disturbed Nocturnal Sleep: Easy to Measure but Difficult to Interpret

IN 1965, SAM SUTTON PUBLISHED A BRIEF PAPER IN SCIENCE IN WHICH HE DESCRIBED A LATE POSITIVE COMPONENT IN AN AVERAGED AUDITORY evoked potential.1 Subsequently referred to as P300, this late component has become the most studied EEG phenomenon of the last 40 years. (A recent Medline search produced over 1200 hits for “P300 and EEG“ exclusive of articles published in psychology journals not listed in Pubmed and the numerous books and book chapters devoted to its study.) Sleep researchers have conducted relatively few studies of P300, yet there is a small and growing literature attempting to relate P300 during the day to nocturnal sleep or the lack of it. The paper by Devoto and colleagues in this issue is the latest in this growing series. What then is P300, and how can sleep researchers use it? P300 is an endogenous ERP component; that is, it reflects the higher processing of the psychological meaning of stimuli, rather than their “exogenous” sensory or perceptual properties such as loudness or brightness. It is elicited when subjects attend and discriminate stimulus events that differ from one another on some dimension. Often such discrimination occurs in the context of an “oddball” paradigm, in which P300 is usually elicited to low probability (target), task relevant stimuli.2 One notable feature of the P300 is that it can be elicited by a variety of stimuli or events from any stimulus modality; the only requirement is that the events have distinct onsets, and that they are classifiable into two or more categories. P300 can also be elicited when only a single stimulus is presented if the stimulus is rare and salient to the subject.3,4 In addition to standard visual or auditory stimuli, P300 has been produced in response to respiratory,5 somatosensory,6 olfactory,7 and even esophageal stimuli.8 The classification of P300 is made difficult due to the fact that several late positive components the overlap it. Several studies have reported that a long latency positive wave (labeled P3a9) can also be elicited by an unpredictable shift in an ongoing sequence of auditory stimuli under circumstances when no explicit task is given to the subject. Importantly, P3a is considered to be a different brain phenomenon from the classically described P300 (also referred to as P3b), with a shorter latency and fronto-central scalp distribution10 as compared to the ubiquitous centro-parietal distribution for the P3b.11 The P3a (or “novelty P3”) also seems to represent psychological events that are quite distinct from those of P3b. It has been linked to processes involved in the involuntary capture of attention by salient events12 and, in a different context, to the processing associated with the active inhibition of responses in so called “NO GO” paradigms.13 A number of factors have been suggested to account for the observed variations in P300 amplitude and latency. Amplitude can be independently effected by a number of factors including subjective probability, task complexity, stimulus complexity, stimulus value, expectancy and attention.14,15 P300 amplitude has also been shown to be under some genetic control,16,17 with a major field of study devoted to the investigation of reduced P300 in the children of alcoholics.18,19 Studies have generally found that P300 latency increases when there is difficulty in discriminating between stimuli, and is independent of response output processing as indicated by reaction time.20 This latency increase has been proposed to relate to the longer time for stimulus evaluation required when discriminations are difficult or ambiguous.21 P300 latency has also been shown to increase in the elderly22 and with drowsiness preceding sleep onset.23,24 P300 has been assessed in a variety of clinical populations (see reference 25 for review) and has unfortunately shown a marked lack of clinical specificity, with reduced amplitude and/or increased latency being a common finding in many disorders or syndromes. Studies conducted on obstructive sleep apnea syndrome (OSAS) patients during wakefulness have however reported mixed results, with some evidence for an increased P300 latency to visual26-29 and auditory stimuli,28,30,31 although no effect on auditory P300 latency has also been reported.26,27,32 Some studies have30,31 and some have not26-28,32 found OSAS patients to have reduced auditory P300 amplitudes, although there is 1 report of significant correlations between P300 amplitude and respiratory disturbance index , % stage 1 sleep and the maintenance of wakefulness test.33 Likewise improved nocturnal sleep following effective CPAP treatment has,30,31 and has not,27,29 been associated with decreased P300 latency relative to that seen prior to treatment. Neither of the studies that have evaluated respiratory somatosensory P300 in OSAS patients found a significant effect on amplitude or latency of the response relative to controls.32,34 Findings of the impact of experimental sleep fragmentation or deprivation on P300 have also been mixed. Morris et al35 reported reduced P300 amplitude and increased latency during 18 hours of sleep deprivation. Cote et al36 found no changes in P300 following two nights of sleep fragmentation, although they did report P300 and the Daytime Consequences of Disturbed Nocturnal Sleep: Easy to Measure but Difficult to Interpret