Working memory and prefrontal cortex dysfunction: specificity to schizophrenia compared with major depression

Abstract

Background

A large number of studies suggest the presence of deficits in dorsolateral prefrontal cortex function during performance of working memory tasks in individuals with schizophrenia. However, working memory deficits may also present in other psychiatric disorders, such as major depression. It is not clear whether people with major depression also demonstrate impaired prefrontal activation during performance of working memory tasks.

Methods

We used functional magnetic resonance imaging to assess the patterns of cortical activation associated with the performance of a 2-back version of the N-Back task (working memory) in 38 individuals with schizophrenia and 14 with major depression.

Results

We found significant group differences in the activation of dorsolateral prefrontal cortex associated with working memory performance. Consistent with prior research, participants with schizophrenia failed to show activation of right dorsolateral prefrontal cortex in response to working memory tasks demands, whereas those with major depression showed clear activation of right and left dorsolateral prefrontal cortex as well as bilateral activation of inferior and superior frontal cortex.

Conclusions

During performance of working memory tasks, deficits in prefrontal activation, including dorsolateral regions, are more severe in participants with schizophrenia (most of whom were recently released outpatients) than in unmedicated outpatients with acute nonpsychotic major depression.

Introduction

A growing literature on cognitive deficits in schizophrenia and their neurobiological correlates suggests that people with schizophrenia have deficits in working memory (WM) that are associated with abnormalities in the activation of the prefrontal cortex, particularly the dorsolateral prefrontal cortex. However, only a few studies have examined whether such WM deficits are specific to schizophrenia or are also present in people with other psychiatric disorders. For example, research on the neurobiological underpinnings of major depression suggest that prefrontal cortex dysfunction may also be present in this disorder, which may in turn contribute to WM deficits. As such, the goal of our study was to use functional magnetic resonance imaging (fMRI) to compare the patterns of cortical activation associated with the performance of both verbal and nonverbal WM tasks in people with schizophrenia and people with major depression.

Typically WM is defined as the ability to temporarily maintain and manipulate information online (Baddeley and Della Sala 1996). A large number of studies have demonstrated that people with schizophrenia demonstrate performance deficits on a wide range of WM tasks (Barch et al 1998, in press; Cohen et al 1999, Gold et al 1997, Goldberg et al 1998, Gooding and Tallent 2001, Park and Holzman 1992, Park and Holzman 1993, Stone et al 1998, Wexler et al 1998. Moreover, people with schizophrenia demonstrate abnormalities in the activation of prefrontal cortex during performance of WM tasks. In particular, several studies suggest a disturbance in the activation of dorsolateral regions of the prefrontal cortex (PFC; i.e., middle frontal gyrus, Brodmann’s area 46/9; Andreasen et al 1992, Barch et al 2001, in press; Berman et al 1992, Berman et al 1986, Carter et al 1998, Menon et al 2001, Perlstein et al 2001, Volz et al 1997, Weinberger et al 1988, Weinberger et al 1986, Weinberger et al 1996. The most frequent pattern of abnormal dorsolateral PFC activation is a reduction in activation among people with schizophrenia compared with healthy control participants, although a few studies have actually found greater activation of PFC in people with schizophrenia when they perform WM tasks Callicott et al 1998, Callicott et al 2000, Manoach et al 1999. Furthermore, some studies have identified relationships between the degree of structural abnormality in dorsolateral PFC and cognitive disturbances in people with schizophrenia Baare et al 1999, Seidman et al 1994. Thus, a wealth of literature supports the presence of dorsolateral PFC deficits in schizophrenia and their role in cognitive deficits in this disorders.

Several recent reports suggest that people with other psychiatric disorders, such as unipolar major depression, also display WM deficits Landro et al 2001, Merriam et al 1999, Pelosi et al 2000, Sweeney et al 1998 However, these deficits may not be as severe as those found in schizophrenia Goldberg et al 1993, Merriam et al 1999, and a number of investigators have not found such WM deficits in major depression Cohen et al 1999, Grant et al 2001, Sweeney et al 2000, Zakzanis et al 1998 or in bipolar disorder Gooding and Tallent 2001, Park and Holzman 1992. It has been suggested that such WM deficits in major depression also reflect disturbances in prefrontal function Merriam et al 1999, Purcell et al 1997, Sweeney et al 1998. However, structural and resting blood flow studies in mood disorders have typically identified deficits in more orbital and medial regions of prefrontal cortex, as opposed to dorsolateral PFC (Drevets 2000), although a few studies have identified abnormalities in dorsolateral regions Dolan et al 1993, Medved et al 2001. To our knowledge, only one previous study has examined brain activation during cognitive challenge in major depression, finding greater impairment in dorsolateral PFC activation during performance of the Wisconsin Card Sorting Task in patients with schizophrenia compared with major depression (Berman et al 1993). However, brain activation during performance of a specific WM task has not yet been compared among people with schizophrenia and those with major depression.

The goal of this study was to compare brain activation, using fMRI, and behavioral performance in people with schizophrenia and with unipolar major depression during performance of both verbal and nonverbal WM tasks. We predicted that participants with schizophrenia would show significantly impaired WM performance and impaired activation of dorsolateral PFC compared with participants with major depression. The inclusion of both verbal and nonverbal WM tasks also allowed us to determine whether differences between diagnostic groups might vary as a function of material type.