Dimensions of impulsive behaviour in abstinent alcoholics

Abstract

Impulsivity is a complex multidimensional behavioural construct that has been frequently implicated in the pathogenesis of addictive disorders. Both self-report and behavioural measures have been developed to assess its underlying dimensions but their interrelation is still poorly understood and few studies have employed a comprehensive set of measures within a given population to explore their interrelation.

In a stably abstinent alcohol-dependent population we explored the interrelations between two well known personality questionnaires (BIS-11 and SSS), and two behavioural measures reflecting different dimensions of impulsivity, behavioural disinhibition (Go/NoGo task) and delay discounting (DDT). In addition, we included the Iowa Gambling Task (IGT), a neurobehavioural measure of decision-making, a cognitive dimension that is closely related to impulsivity.

Strong correlations were found between the different subscales of the self-report measures. However, the correlations between the behavioural measures and the self-report measures were weak, suggesting that they both tap into different aspects of the concept. Finally, a principal component analysis on the data of the behavioural measures revealed that they all loaded on separate factors. These findings support the hypothesis that behavioural disinhibition and delay discounting are two independent dimensions of impulsivity and that decision-making is a third cognitive dimension, independent of both other measures.

Introduction

Alcoholism is a very heterogenic disorder and the individual differences with respect to its aetiology, course and severity are considerable. Thus, the identification of elements that constitute an increase in individual vulnerability and mediate the course of the disorder is of paramount importance. Impulsivity has been proposed as an important factor with respect to the initiation and course of addictive processes (Dawe et al., 2004, Dom, Hulstijn, et al., 2006). For example, prospective studies have shown that individual differences in childhood behavioural disinhibition predict early substance use during adolescence and substance use disorder in young adulthood (e.g., Masse & Tremblay, 1997).

Generally, impulsivity is used to describe a broad range of maladaptive behaviours including an inability to inhibit inappropriate action, insensitivity to delayed or uncertain consequences, perception of time as progressing more slowly than actual time, and the perseverance of negatively reinforced actions (McDonald et al., 2003, Moeller et al., 2001). On a conceptual level, there is now wide agreement that impulsivity consists of a number of different dimensions (Bechara, 2003, Dougherty et al., 2005, Evenden, 1999). Dawe et al., 2004, de Wit and Richards, 2004 define two broad dimensions. The first dimension can be labelled as “behavioural disinhibition or rash impulsiveness”. This includes problems both in response initiation and response inhibition. In this respect impulsive behaviour can be a consequence of responding prior to complete processing and evaluation of a stimulus (“acting without thinking”), or can result from a failure to inhibit an already initiated response (response inhibition). A second dimension has been labelled as “consequence sensitivity or impulsive decision-making” (de Wit & Richards, 2004). Here, impulsivity is defined as behavioural choices (“decisions”) that persist despite negative or less than optimal consequences, i.e., a preference for small immediate rewards over later, larger rewards (“delay discounting”) or a preference for larger immediate rewards coupled to later, uncertain larger punishments versus smaller immediate rewards associated with smaller, later punishments (“risk discounting”) (Monterosso, Erhman, Napier, O’Brien, & Childress, 2001).

Various measures have been developed to assess impulsive behaviour. Broadly these can be divided into self-report measures of personality that rely on an individual’s self-perception of their behaviour, and behavioural tasks that measure overt behaviour related to specific dimensions of impulsivity. Typically, both types of measurements do correlate only weakly, indicating that they measure different aspects of impulsivity (Dom, D’Haene, et al., 2006, Moeller et al., 2001, Reynolds et al., 2006).

There are few validated behavioural instruments available to measure impulsivity and it remains to be specified which aspect of the concept they tap in and to which degree these measures interrelate (Moeller et al., 2001, Reynolds et al., 2006). Only a few studies have used a comprehensive battery to assess simultaneously the different aspects of impulsivity within a given sample. Reynolds et al. (2006) used four behavioural tasks within a sample of healthy community recruited individuals. Factor analysis of their data indicated that these tasks loaded on two separate components, the first, labelled impulsive disinhibition (Stop Task and Go/NoGo Task) and second, impulsive decision-making (Delay Discounting Task and Balloon Analog Risk Task). In a sample of cocaine dependent patients, Monterosso et al. (2001) compared three behavioural decision-making tasks, i.e., Delay Discounting Task (DDT, Richards, Zhang, Mitchell, & de Wit, 1999), Iowa Gambling Task (IGT, Bechara, Damasio, Damasio, & Anderson, 1994), and the Roger Decision-Making Task (RDMT, Rogers et al., 1999). Impairments on all three tasks have been reported within substance abusers relative to controls (Bechara, 2003, Bjork et al., 2004, Dom, D’Haene, et al., 2006, Dom, Hulstijn, et al., 2006, Richards et al., 1999). These behavioural decision-making tasks have as communality that task performance involves an evaluation and choices (“decision-making”) between different consequent outcomes. Although the exact processes that are measured by these tasks remain to be specified, they all measure different aspects of decision-making, i.e., delay and risk-discounting.

In the current study we report on the relations between self-report and behavioural measures of impulsivity within a sample of stably abstinent alcohol dependent inpatients. The study consists of two separate analyses. First we explored the correlations among the self-report measures, behavioural measures, and the measures of problem severity. Next we used a principal component analysis to explore the component structure among the behavioural measures.

The Barratt Impulsivity Scale (BIS-11; Patton, Stanford, & Barratt, 1995) and the Sensation Seeking Scale (SSS; Zuckerman, Kolin, Price, & Zoob, 1964) were used as self-report measures of impulsivity. Impulsivity, as measured with the BIS is best described as a tendency to act rashly and without consideration of consequences. Sensation seeking is a tendency to seek out intense, novel forms of sensation and experiences, regardless of the risks involved. In a recent factor analysis, the BIS, the SSS and another widely used measure, the Eysencks’ I₇ Impulsiveness subscale (Eysenck, Pearson, Easting, & Allsopp, 1985), have been reported to load on the same factor, labelled by Dawe et al. (2004) rash-spontaneous impulsiveness. For the behavioural tasks we used a Go/NoGo task which has been widely used as a measure for response inhibition. In addition, two decision-making tasks were included, i.e., a Delay Discounting Task (Richards et al., 1999) and the Iowa Gambling Task (Bechara et al., 1994).

Based on the previous findings, described above, we hypothesized that the correlations among the self-report measures would be high but that the correlations between the self-report measures and behavioural measures would be only poor or modest. Furthermore, we expected that, with respect to the behavioural measures, the correlations between a measure of response inhibition (i.e., Go/NoGo task) and decision-making (i.e., DDT and IGT) would be poor, while correlations between the DDT and IGT would be high. Finally, we expected that separate inhibition and decision-making components would be identified from the principal components analysis.