Dysfunction of on-line representational, or 'working' memory is considered a core feature of schizophrenia that may underlie many of the disabling symptoms of the disorder. However, previous research on the topic has predominantly employed behavioural measurements, leaving the electrophysiological correlates of this impairment relatively uncharted. In the present study, event-related potentials (ERPs) were obtained from forty-eight patients with schizophrenia, and forty-six well controls as they performed a visuospatial delayed response (DR) working memory task. The task required participants to either touch the location of a visible target (perceptual trials), or its remembered position (memory trials) following a delay interval of 4 seconds. On 50% of trials a distractor stimulus identical to the target was introduced during the delay. The first part of the study indicated that, relative to controls, patients with schizophrenia experienced: a greater deterioration in accuracy on memory relative to perceptual trials; increased latency of sensory-attentional components Nl to P3b; and reduction of the Slow Wave (SW) memory effect, particularly over the right hemisphere. Although the working memory accuracy of patients with schizophrenia was unaffected by distractor presentation, they did not exhibit the frontal enhancement of SW memory effect evident in controls. The second part of the study revealed that whereas severity of both negative and disorganised symptoms correlated with reduced accuracy, ERPs were differentially associated with each symptom complex. Severity of negative symptoms correlated with reduced N1-P2 amplitude, and with a reduction of the SW memory effect under conditions of distraction. Conversely, severity of disorganised symptoms correlated with increased N2-P3b amplitude only. Positive symptoms were uncorrelated with any of the measures examined.
The final part of the study, based on a subgroup of the original sample, indicated several associations between DR behavioural and ERP indices and tests of frontal-executive function. For the schizophrenia group, two findings of particular interest were that (1) cognitive processes measured by Trails B : A contributed to the relationship between negative symptoms and SW memory activity during conditions of distraction; and (2) cognitive processes measured by the WCST contributed to the relationship between disorganisation and N2-P3b attentional activity. Li conclusion, the present study provides evidence that patients with schizophrenia manifested a disturbance in the timing of early sensory and attentional processing, as well as a reduction of working memory maintenance activity, that became more pronounced under conditions of distraction. Moreover, severity of negative symptoms was related to an additional reduction of on-line maintenance activity during conditions of distraction, while severity of disorganised symptoms was related to increased and inefficient activation during attention allocation. In line with these results, performance of the Trial Making Test (TMT), a test thought to engage attentional mechanisms of working memory, contributed to the relationship between negative symptoms and reduced SW memory activity during the distractor condition. Performance of the WCST, a test that may require allocation of attentional resources and response inhibition, contributed to the relationship between disorganised symptoms and N2-P3b attentional activity. Such results suggest that negative symptoms may be associated with an impairment of on-line maintenance of information during high load conditions in which attention must be deployed to suppress distraction, while disorganisation may be related to an inefficient attentional allocation strategy. Finally, positive symptoms were unrelated to working memory during any stage of processing, which may imply that positive symptoms and working memory dysfunction are separate features of schizophrenia.