Abstract
Behavioural studies suggest that voluntary and involuntary attention may operate using different mechanisms. The present study examined this issue in EEG activity induced by voluntary and involuntary attention. Participants performed a face-discrimination task under conditions of voluntary or involuntary attention. A peripheral cue preceded the face in both tasks, but for involuntary attention it was non-predictive, while for voluntary attention the face appeared more often in the cued location (70%). Both sessions included target-absent trials allowing evaluation of cue-related activity in isolation. Behaviorally, both voluntary and involuntary conditions produced better performance on valid than invalid trials (RT). Spectral analysis of the EEG revealed increases in gamma-band response, but the pattern was different for voluntary and involuntary conditions. For target absent-trials there was greater cue-related activity in the voluntary than the involuntary condition that peaked about 150 ms after cue onset. This finding is consistent with greater use of the cue during voluntary attention. Similar gamma band increases were recorded for the face targets in both attention conditions. On valid trials, this activity was significantly less in the voluntary compared to the involuntary condition, presumably because subjects shifted their attention to the cued location in the voluntary attention condition. On invalid trials the relative magnitude of the face-related activity reversed and there was greater power in the voluntary session. That is expected because in the predictive session, subjects have shifted their attention to the cued location which in these trials is erroneous and an additional shift of attention is then required. Thus, gamma band activity reveals a difference in cue-related activity for voluntary and involuntary attention and an interaction with trial type in target-related activity between these two types of attention.