Intra-dimensional/extra-dimensional set-shifting performance in schizophrenia: Impact of distractors

Abstract

Background

We sought to determine if a representative group of young chronic patients with schizophrenia would demonstrate selective impairments in set shifting processes of the CANTAB Intra-dimensional/extra-dimensional (IDED) task. We predicted that patients would have prominent difficulties with Compound Discrimination (C_D) (stage of the task in which irrelevant stimuli are introduced) and Extra-Dimensional Shifting (EDS) (stage of the task in which a new stimulus dimension must be attended) on the basis of the results of cortical hypodopaminergic states in subhuman primates (for C_D) and effects of dorsolateral prefrontal cortical lesions on set shifting and prior results in schizophrenia (for EDS).

Methods

We administered the IDED to 36 patients and 26 healthy controls. Additionally, we administered the Wisconsin Card Sorting Test (WCST), another test of set shifting, and a Continuous Performance Test (CPT) type task of attention to patients with schizophrenia in order to investigate which cognitive components accounted for performance difficulties at different stages of the IDED task.

Results

Patients had selective difficulties on C_D and EDS stages of the task. In schizophrenic patients early stages of the task involving the introduction and establishment of attentional set were correlated to CPT performance, while later set shifting stages were correlated with WCST categories attained.

Conclusion

We found evidence that patients with schizophrenia were susceptible to introduction of unreinforced irrelevant stimuli at the C_D stage, such that the previously rewarded target stimuli no longer held hegemony as a representation. This type of processing failure may reflect difficulties in stabilizing a representation and is consistent with effects of prefrontal hypodopaminergia in primates. Secondly, “survivors” of this stage experienced marked difficulties on EDS-stage, suggestive of classic prefrontal failures.

Introduction

A synopsis of findings from neuropsychological, postmortem, ligand and functional imaging studies strongly implicates the prefrontal cortex in the pathogenesis of schizophrenia. For example, patients with schizophrenia perform on the same level or worse than patients with frontal lobe lesions on cognitive tasks probing prefrontal function (Pantelis et al., 1999), and imaging studies show abnormal physiologic activity in the prefrontal cortex of patients with schizophrenia during performance of such tasks as compared to normal controls (Weinberger et al., 2001). The neural mechanisms underlying these prefrontal dysfunctions are not completely understood. Evidence points towards afferentation abnormalities in ascending pathways, cytoarchitectural anomalies, cellular pathology of the prefrontal cortex itself, and/or abnormalities in neuromodulatory actions in the prefrontal cortex, including prefrontal hypodopaminergia (Weinberger et al., 2001). Notions of prefrontal hypodopaminergia in schizophrenia have recently received support based on (1) the effects of COMT val allele on frontal cortical function and its association with schizophrenia (Egan et al., 2001, Goldberg et al., 2003) (2), post mortem findings of a decrease in dopamine terminal afferentation in prefrontal cortex of patients with schizophrenia (Akil et al., 1999) and (3), increases in availability of D1 receptors in the prefrontal cortex of patients with schizophrenia (Abi-Dargham et al., 2002, Abi-Dargham and Moore, 2003), presumably as a compensatory upregulation secondary to chronic understimulation.

The current study aims to characterize cognitive deficits in schizophrenia based on assumptions about the effects of hypodopaminergia on prefrontal function. As an indicator of prefrontal functioning, we chose the ability to shift attention between and within different stimulus categories since impairments in attentional set-shifting have been demonstrated (1) in patients with frontal lobe lesions (Owen et al., 1991, Pantelis et al., 1999), (2) in patients with schizophrenia (Elliott et al., 1995, Goldberg et al., 1987, Pantelis et al., 1999, Weinberger et al., 1986), and (3) in clinical populations with known or putative dopaminergic pathologies, such as patients with Parkinson's disease (Downes et al., 1989, Lange et al., 1992, Owen et al., 1992). Moreover, some of the cognitive processes involved in shifting attention have been shown to be affected by direct manipulation of the prefrontal dopaminergic circuitry. For example, catecholaminergic (dopamine and noradrenaline) depletion of the prefrontal cortex (but not the caudate nucleus; see also Collins et al., 2000) of the common marmoset using 6-OHDA lesions impairs the ability to develop an attentional set and leads to increased susceptibility to distraction from task-irrelevant stimuli (Crofts et al., 2001). Similarly, computational models of dopamine effects on signal/noise ratios in the prefrontal cortex point to susceptibility to distraction as a hallmark of hypodopaminergic states at the cortical level (Durstewitz et al., 2000, Seamans et al., 1998). In concert these two observations provide the impetus for the present study: That schizophrenic patients will be prominently compromised during set shifting tasks in which relevant and irrelevant distractor stimuli must be distinguished before a correct decision can be made to maintain or shift set.

Two prominent paradigms used in cognitive science to investigate processes of attentional set-shifting in humans are the Wisconsin Card Sorting Test (WCST) (Berg, 1948) and the intra-dimensional/extra-dimensional (IDED) subtask of the Cambridge Neuropsychological Test Automated Battery (CANTAB) (Downes et al., 1989, Sahakian and Owen, 1992). In the WCST, subjects are required to match a series of stimuli, each characterized by three different stimulus dimensions (shape, color, and quantity), to one of four templates corresponding to the stimulus in one of the three dimensions (shape, color, or quantity). After 10 consecutive correct sorts the relevant stimulus dimension changes and another one becomes relevant, requiring the subject to shift attention from one stimulus dimension to another one. The IDED begins with simple stimulus discriminations and reversals thereof in order to allow subjects to establish an attentional set (propensity) to a particular stimulus dimension (for example, shape elements).Stimulus complexities continuously increase, with distracting, irrelevant stimuli (for example, line elements) being added or superimposed to the stimuli of the relevant stimulus dimension. Only after an attentional set has been established does an attentional shift take place, requiring the subject to shift attention to first another exemplar of the same stimulus dimension (i.e. perform an “intra-dimensional” (ID) shift) and then to the previously irrelevant stimulus dimension (i.e. perform an “extra-dimensional” (ED) shift).

The ED shift of the IDED task has been considered equivalent to the WCST stimulus-contingency changes (Fray and Robbins, 1996), and based on this hypothesized relation impairments in ED performance have been interpreted as signs of perseveration, i.e. being “stuck” on a previously established attentional set (Elliott et al., 1995). Yet, as mentioned above, there is evidence that performance difficulties on tests of attentional set shifting under conditions of hypodopaminergia (Crofts et al., 2001, Seamans et al., 1998) might not only be caused by failure in performing an attentional shift per se, but also by impairments in other cognitive functions required to establish and maintain an attentional set such as susceptibility to distraction. Since the IDED allows the separate examination of the various cognitive functions required for and involved in attentional set-shifting such as set maintenance, stabilization of representations and sensitivity to distraction, the present study focused on task performance on the IDED task.

Prior work on the IDED task in schizophrenia consistently demonstrated difficulties of patients performing ED shifts (Elliott et al., 1995, Hutton et al., 1998, Joyce et al., 2002, Pantelis et al., 1999). Two studies found impairments at the pre ED-stages of the task (Elliott et al., 1995, Pantelis et al., 1999), attributing these deficits to either low IQ, signs of a basic, stimulus- and not category-specific form of perseveration (for failure at the early stages of the task concerned with establishing an attentional set), or impairment in generalizing a rule or concept formation (for failure at the ID stage of the task).

We hypothesized that schizophrenic patients would demonstrate large and selective deficits on the IDED task (1) already at the earlier stages of the task when distracting and/or non relevant stimuli are introduced as might be expected under hypodopaminergic states and based on the studies of Crofts et al. (2001), and Seamans et al. (1998), and accumulating work in schizophrenia, and (2) at the ED stage of the task, as based on effects of dorsolateral prefrontal cortical lesions on set shifting and prior results in schizophrenia (see above).We thus predicted that patients would have difficulty maintaining set when distractors were first introduced in the IDED task, i.e. at the so-called C_D (compound discrimination) stage of the IDED task (for a specific description of this stage, see methods section). In addition, in order to ascertain which cognitive component (attention or perseveration) accounted for performance (failure) at the different stages of the IDED task, we sought to determine the relationship of IDED task performance to performance on two other tasks widely used to measure attention and attentional set-shifting, the Continuous Performance Test (CPT) (Gordon et al., 1996) and the WCST (Berg, 1948). We predicted that early stages of the IDED task would be better explained by CPT task performance (and thus failure at this stage by excessive distractibility or impairments in sustained attention) and later stages of IDED performance by WCST performance (and thus failure at these stages by perseveration).

To our knowledge, this is the first time that the relationship between the IDED ED-stage and WCST has been directly examined.