Probed recall for serial order deficits in short-term memory in schizophrenic patients.

Background: Patients with schizophrenia frequently display problems in tasks demanding working memory. In a previous study, we examined short-term memory (STM) for serial order by having participants recall lists of letters from the first item to the last item in the order in which they were presented, and we examined the types of errors made (e.g., omissions, intrusions and movements; [Neuropsychology 15 (2001) 128]). We found that the disproportionate errors schizophrenic patients made were omissions at the end of six-item lists, a finding we suggested might reflect patients' longer output times, which adds to information maintenance demands. If this is the case, we predicted that the group difference in the terminal positions could be eliminated through the use of a probed recall paradigm.

Method: In the current study, 26 schizophrenic patients and 33 control participants were tested on a probed recall task that was similar to our previous serial recall task except that instead of recalling the whole sequence of letters, participants were probed as to which letter appeared in a specific position in the sequence.

Results: We found that when participants were probed for later positions, recall was equivalent in the groups (i.e., recency), but disproportionately worse in patients for earlier positions.

Conclusions: We suggest that schizophrenic patients' limited STM span for serial order is not attributable to a selective deficit in memory for serial order. Rather, we propose that it may be explicable in terms of impaired information maintenance and thus this becomes evident in conditions involving longer sequences of stimuli.