Vigilance in schizophrenia and its disruption by impaired pre-attentive selection: a dysintegration hypothesis.

Introduction: The study determined, simultaneously, whether major deficits of schizophrenia (sustained and selective attention, slow information processing, slow motor responding) are independent or related to each other.

Methods: An auditory vigilance task (Pigache Attention Task, PAT) required a button-press to targets during four 5-minute subtests (slow diotic, fast diotic, slow dichotic, fast dichotic, analogous to four versions of the continuous performance test). Twenty schizophrenics on the first test-occasion of a double-blind, placebo controlled, crossover study were compared to 11 healthy subjects. Also, all 28 fortnightly test occasions were analysed to quantify the schizophrenia deficits more precisely and the PAT was evaluated in a larger group of 86 healthy subjects.

Results: Schizophrenics were significantly impaired on all task parameters versus healthy subjects. The patients' errors were independent and additive (grand mean components: basic task 36%, including a 9% time-on-task component; speed increment 26%; dichotic increment 38%). Errors, latencies, and psychosis severity were mutually correlated.

Conclusions: The performance of all subjects confirmed a quantitative Vigilance Decision Model. The PAT impairments in schizophrenia suggested that rival options (e.g. thoughts) redeployed or suppressed attention away from the task , indicating a dysfunction of pre-attentive selection processes. Brain mechanisms are discussed and a new dys integration hypothesis of schizophrenia is proposed.