Initiation and supervisory processes in schizophrenia and depression

Abstract

Clinical studies show that schizophrenic and depressive subjects have problems with daily life activities, and neuropsychological studies tend to explain these problems in terms of a dysexecutive syndrome. Verbal fluency and sentence arrangement are tasks considered to focus on two aspects of the dysexecutive syndrome known as initiation and supervision processes, respectively. In this study, we assessed performance in these two tasks in schizophrenia and depression. Twenty-six schizophrenic subjects (chronic schizophrenia, DSM IV definition) were compared with 26 control subjects balanced for sex, age and educational level, and 16 depressive subjects (major depression episode, DSM IV) were compared with 11 similarly balanced control subjects. Switching and clustering scores were evaluated during a semantic fluency task as two components underlying the initiation and organization processes. Capture errors specific to failure of the supervisory system and differences between the number of correct responses in two conditions (valid/invalid) were evaluated as indexes of the supervision process in a sentence arrangement task. In the semantic fluency task, switching scores were significantly lower in the schizophrenic and depressive subjects than in their respective controls. In the sentence arrangement task, only the schizophrenic subjects made significantly more capture errors than their controls and had significantly fewer correct sequences in invalid conditions than in valid conditions. This study shows a dissociation between supervision and initiation processes in two different psychiatric populations. Initiation is impaired, but supervision is preserved in depression, whereas both initiation and supervision are impaired in schizophrenia.

Introduction

Frontal lobe functions relate to behaviors and cognitions specifically associated with the frontal lobe (Stuss et al., 1997). These functions are required in daily life activities, particularly those involving inhibition of automatic actions or plan initiation. On the basis of clinical and neuroanatomical findings, the functions of the frontal lobes can be broadly subdivided into three levels (Stuss and Benson, 1986). In the first level, medial frontal structures influence motivation and drive, including the ability to initiate cognitive and motor activities. Second-level functions provide conscious control and direction of integrated behavior of total brain operations and are especially important in non-routine situations that require novel solutions (Norman and Shallice, 1986). A third proposed level consists of what is commonly called `self-awareness'. These three levels may be globally or separately impaired in diseases such as schizophrenia (Weingartner, 1986; Frith, 1992), fronto-temporal dementia, depression, and other conditions with frontal expression. In this study, we focused on frontal functions of the first and second levels (initiation and supervisory processes).

From a clinical point of view, impairment of initiation processes is described in terms of `poverty of speech' and `loss of interest' or apathy (Marin, 1990) and can be found in depression, schizophrenia, Parkinson's disease and other neurological disorders. From a cognitive point of view, verbal fluency is one of the tests used most frequently to assess initiation processes.

Several studies have indicated that schizophrenic and depressive patients produce fewer words than healthy subjects (Miller, 1975; Kolb and Whishaw, 1983; Caine et al., 1984; Hart et al., 1987; Gruzelier et al., 1988; Crawford et al., 1993; Franke et al., 1993; Trichard, 1995). These impairments have been explained by a subnormal capacity to initiate willed action (Frith, 1992; Stuss et al., 1992) and reduced access to semantic memory because of difficulties in organizing a lexical search (Allen et al., 1993). Recent studies (Robert et al., 1997aRobert et al., in press; Troyer et al., 1997) have demonstrated that two major components underlying verbal fluency performance can be distinguished: clustering (organizational search process in semantic memory) and switching (initiation of search strategies).

In contrast to initiation processes, control systems are simpler to describe from a cognitive point of view than from a clinical standpoint. In the Shallice model (Norman and Shallice, 1986; Shallice and Burgess, 1991), processes involved in the cognitive control of action can be divided into two levels. The first level is related to routine activities where selected learned triggering procedures are sufficient to carry out the task satisfactorily. The second level, known as the supervisory attentional system (SAS), is needed when routine operation is inadequate or when inappropriate schemata must be inhibited. A recent study suggests that routine and especially non-routine behavior are impaired in schizophrenia when investigated with a sentence arrangement task (Robert et al., 1997b). This impairment may be understood in terms of an inability to inhibit inappropriate routine schemata. From a clinical point of view, supervisory SAS impairment is described in terms of a `problem in stopping' (Lezak, 1995), i.e. breaking and modulating an ongoing action (disinhibition).

Clinical studies may be useful to distinguish between initiation and supervision. The aim of this study was to compare these two functions in schizophrenia, depression and health by using verbal fluency and sentence arrangement tasks.