Probability mismatch and template mismatch: A paradox in P300 amplitude?

State anxiety influences P300 and P600 event-related potentials over parietal regions in the hollow-mask illusion experiment

The hollow-mask illusion is an optical illusion where a concave face is perceived as convex. It has been demonstrated that individuals with schizophrenia and anxiety are less susceptible to the illusion than controls. Previous research has shown that the P300 and P600 event-related potentials (ERPs) are affected in individuals with schizophrenia. Here, we examined whether individual differences in neuroticism and anxiety scores, traits that have been suggested to be risk factors for schizophrenia and anxiety disorders, affect ERPs of healthy participants while they view concave faces. Our results confirm that the participants were susceptible to the illusion, misperceiving concave faces as convex. We additionally demonstrate significant interactions of the concave condition with state anxiety in central and parietal electrodes for P300 and parietal areas for P600, but not with neuroticism and trait anxiety. The state anxiety interactions were driven by low-state anxiety participants showing lower amplitudes for concave faces compared to convex. The P300 and P600 amplitudes were smaller when a concave face activated a convex face memory representation, since the stimulus did not match the active representation. The opposite pattern was evident in high-state anxiety participants in regard to state anxiety interaction and the hollow-mask illusion, demonstrating larger P300 and P600 amplitudes to concave faces suggesting impaired late information processing in this group. This could be explained by impaired allocation of attentional resources in high-state anxiety leading to hyperarousal to concave faces that are unexpected mismatches to standard memory representations, as opposed to expected convex faces.

The clash of expectancies: Does the P300 amplitude reflect both passive and active expectations?

We investigated the clash of different types of expectations by studying event-related potentials and behavioural correlates of passive and active expectations. In the three tasks we used, target stimuli could either confirm or disconfirm both expectations, or confirm one while disconfirming the other. Depending on the task, expected events were related to either increased or decreased P300 amplitude. This was contingent on whether the expected item was probable given the context, or whether it was a target match. This suggests that the effect of expectancy on the P300 amplitude can be direct or indirect. In the case of a direct effect, the evidence suggests that most likely only unconscious, automatic estimation of stimulus probability is reflected in the size of P300 amplitude. However, active, conscious expectancies can influence the P300 activity via indirect routes, by influencing stimulus significance, by leading to the closure of perceptual epoch, or by changing the level of difficulty of processing. Such indirect effect of expectancy can have an opposite direction to the effect of automatically formed expectancies-that is, expected stimuli could be related to larger P300 amplitudes.

Effects of stimulus repetition, duration, and rise time on startle blink and automatically elicited P300

Intense auditory stimuli of sudden onset evoke not only startle blinks but also an event-related potential component resembling classic P300, even when subjects have no assigned task. To more closely examine the relationship of this P300 to startle, event-related potentials and eyeblink were recorded from 16 young adults in three paradigms designed to produce wide variation in startle amplitude: an Habituation series of 30-ms, 105dBA white noise bursts, a Duration paradigm which presented 105dB noise bursts for 3, 10, 30, or 90 ms, and a Rise Time paradigm which varied the rise/fall times (3, 15, 30, and 45 ms) of 110dBA, 1000-Hz tone bursts. Subjects received two runs of each paradigm. Only on the final Duration and Rise Time runs were stimuli explicitly task relevant; on those runs subjects rated verbally, midway in each 8.4-s interstimulus interval, the disturbingness of the prior sound. Although even the briefest noise bursts evoked parietal P300 as well as startle blink, P300 did not behave like startle. P300 habituated more slowly than did blink amplitude, was more responsive to sustained noise than were blink, N110, and P190, and most importantly, did not show the sensitivity to stimulus rise time manifested by these measures. These findings suggest that the amplitude of automatically elicited P300 is not governed by the same mechanisms as startle amplitude, but behaves more like a defense response.

Stimulus probability and motor response in young and old adults: an ERP study

The effects of responding hand and stimulus probability were investigated in young and old subjects in an RT task in which both rare and frequent stimuli required a response. It was found that the effect of stimulus probability was less pronounced in old subjects than in young, and that the latency of P3 was longer in the elderly, although their RTs were not different from young subjects. ERPs for right hand responses were larger than for left hand responses; this difference was discernible already in the P2 and N2 peaks of the ERP. A tentative explanation is offered for these large and unexpected hand differences. An interpretation in terms of an age-related decrease in resources is proposed for the increased P3 latency and the decreased probability effect on P3 amplitude in old subjects.