Schizophrenic subjects show aberrant fMRI activation of dorsolateral prefrontal cortex and basal ganglia during working memory performance

Abstract

Background: Working memory (WM) deficits in schizophrenia have been associated with dorsolateral prefrontal cortex (DLPFC) dysfunction in neuroimaging studies. We previously found increased DLPFC activation in schizophrenic versus normal subjects during WM performance (Manoach et al 1999b). We now have investigated whether schizophrenic subjects recruit different brain regions, particularly the basal ganglia and thalamus, components of frontostriatal circuitry thought to mediate WM.

Methods: We examined regional brain activation in nine normal and nine schizophrenic subjects during WM performance using functional magnetic resonance imaging. Subjects performed a modified version of the Sternberg Item Recognition Paradigm that included a monetary reward for correct responses. We compared high and low WM load conditions to each other and to a non-WM baseline condition. We examined activation in both individual subjects and averaged group data.

Results: Relative to normal subjects, schizophrenic subjects exhibited deficient WM performance, at least an equal magnitude of right DLPFC activation, significantly greater left DLPFC activation, and increased spatial heterogeneity of DLPFC activation. Furthermore, only the schizophrenic group activated the basal ganglia and thalamus, even when matched for task performance with the normal group.

Conclusions: Aberrant WM performance and brain activation in schizophrenia may reflect dysfunction of frontostriatal circuitry that subserves WM. Future studies will elucidate the contribution of the anatomical components of this circuitry to WM deficits.

Introduction

Working memory (WM) is the process of actively holding information “on-line” and manipulating it in the service of guiding behavior (Baddeley 1992). It is a temporary store whose contents are continually updated, scanned, and manipulated in response to immediate information-processing demands. WM is a critical building block of cognition, and it is impaired in schizophrenia (Park and Holzman 1992). WM deficits have been demonstrated in medicated and unmedicated schizophrenic patients (Carter et al 1996), persist throughout the course of illness (Park et al 1999), and are relatively resistant to pharmacotherapy (Goldberg and Weinberger 1996).

The participation of the dorsolateral prefrontal cortex (DLPFC) in WM is well established Friedman and Goldman-Rakic 1994, Petrides et al 1993b and most neuroimaging studies of WM in schizophrenia demonstrate aberrant DLPFC activation Manoach et al 1999b, Weinberger and Berman 1996. However, the neural circuitry underlying WM deficits in schizophrenia is not well understood. Working memory deficits may arise from primary DLPFC dysfunction or from a dysregulation of the DLPFC by other cortical or subcortical structures. The DLPFC projects to the striatum and receives projections back from the basal ganglia via the thalamus (Alexander et al 1986). This frontostriatal circuitry is thought to participate in WM D’Esposito and Grossman 1996, Houk and Wise 1995. Like the DLPFC, the striatum shows metabolic activation during WM performance in nonhuman primates (Levy et al 1997). In addition, during the delay period of delayed-response tasks, striatal neurons exhibit sustained activity that closely resembles that of the DLPFC (Apicella et al 1992). Finally, lesions and dysfunction of the basal ganglia in both human and nonhuman primates produce impairments on delayed response tasks Battig et al 1960, Partiot et al 1996. In schizophrenia, aberrant prefrontal cortex activation has been associated with decreased metabolic rate in the basal ganglia Buchsbaum et al 1992, Siegel et al 1993 and with a failure to suppress blood flow to the striatum during WM performance (Rubin et al 1991). These findings suggest that dysfunction of frontostriatal circuitry may underlie WM deficits.

The primary goal of our study was to investigate whether schizophrenic subjects show aberrant activation of the subcortical components of frontostriatal neural circuitry, specifically the basal ganglia and thalamus, during WM performance. We also expected to replicate our previous findings of increased DLPFC activation and a relation between better WM performance and increased DLPFC activation in schizophrenic subjects (Manoach et al 1999b). Finally, based on their association with WM performance in our study of the SIRP in normal subjects (Manoach et al 1997) and numerous other WM studies Cohen et al 1997, Jonides et al 1998, Smith et al 1998, we expected both groups to activate the supplementary motor area, lateral premotor and motor areas, and the intraparietal sulcus.

We used the Sternberg Item Recognition Paradigm (SIRP; Sternberg 1966) and fMRI to examine task-related differences in regional brain activity in normal and schizophrenic subjects. The SIRP is a continuous performance, choice reaction time (RT) task that reliably activates the DLPFC in both normal and schizophrenic subjects Manoach et al 1997, Manoach et al 1999a, Manoach et al 1999b, Rypma et al 1999. RT is a linear function of the number of items held in WM (WM load; Sternberg 1966), and accurate responses are predicated upon the internal representation of these items. We compared a high WM load condition to a baseline task to identify group differences in regional activation associated with WM. We also compared the high WM load condition to a low WM load condition to insure that our findings in the first comparison could not be attributed to qualitative differences in the baseline task. Finally, because DLPFC activation is found to be related to WM performance Braver et al 1997, Callicott et al 1999, we examined group differences in activation when task performance was comparable. Because individuals with schizophrenia have WM deficits, matching groups by either selecting normal subjects for deficient performance or schizophrenic subjects for normal performance results in unrepresentative samples. Instead, we matched groups for performance by comparing them at different levels of WM load. The performance of schizophrenic subjects in the low WM load condition was comparable to that of normal subjects in the high WM load condition. For our matched performance group comparison, we contrasted the regional activation of schizophrenic subjects in the low WM load versus the baseline comparison to that of normal subjects in the high WM load versus the baseline comparison.