Gray’s impulsivity dimension: A distinction between Reward Sensitivity versus Rash Impulsiveness

Abstract

According to Gray’s reinforcement sensitivity theory of personality there are individual variations in the sensitivity of basic brain and behavioral systems that respond to punishing and reinforcing stimuli. Originally, Gray employed the term “impulsivity” for the personality trait that reflects the Sensitivity to Reward. However, there is growing doubt whether sensitivity for reward and impulsivity refer to one and the same trait. The present study addresses the hypothesis that Sensitivity to Punishment constitutes one personality dimension, whereas measures of Sensitivity to Reward actually are tapping two separate dimensions, one actually referring to Reward Sensitivity and another that pertains to Rash Impulsiveness. For this purpose, a second-order principal component analysis on the subscales of questionnaires that intend to measure aspects of these personality traits was carried out. Results provide further evidence for the notion that sensitivity to reward indeed consists of the two hypothesized factors: Reward Sensitivity and Rash Impulsiveness. Specific links between Reward Sensitivity and Rash Impulsiveness with other constructs such as Eysenck’s personality factors, Negative and Positive affect, and Anhedonia, further demonstrate the distinctiveness of both dimensions.

Introduction

According to Gray’s reinforcement sensitivity theory of personality (RST; Gray, 1987, Pickering et al., 1995, Pickering and Gray, 2001) or the revised version of this account (Corr, 2004, Gray and McNaughton, 2000), there are individual variations in the sensitivity of basic brain and behavioral systems that respond to punishing and reinforcing stimuli. The two basic systems discussed in the RST are the Behavioral Inhibition System (BIS) and the Behavioral Approach System (BAS). According to the original RST, BIS is activated by conditioned stimuli associated with punishment or the termination of reward. BAS is activated by stimuli associated with reward or termination of punishment in order to guide the organism to appetitive stimuli. The RST is based on the principle that individual differences in personality reflect the variation in sensitivity toward stimuli associated with punishment and reward, respectively BIS and BAS.

Originally, Gray employed the term “impulsivity” for the personality trait that reflects BAS sensitivity. Briefly, he assumed that individuals with heightened impulsivity are more sensitive to signals of reward compared to low impulsivity individuals. Recently, however, there is growing doubt whether sensitivity for reward and impulsivity refer to one and the same trait. That is, impulsivity typically pertains to behaviors that are rash and spontaneous. In contrast, Reward Sensitivity does not necessarily refer to rash and spontaneous behaviors (Dawe & Loxton, 2004). This is nicely illustrated by a recent study of Franken and Muris (2005) who demonstrated that a BAS-Reward Sensitivity scale was a significant predictor of performance on the Iowa gambling task, a behavioral measure of reward-related decision making. More precisely, good performance on the Iowa gambling task was found to be positively related to Reward Sensitivity as indexed by the Reward Responsiveness subscale of the BIS/BAS Scales. In contrast, impulsivity, as measured using the Dickman Impulsivity Inventory (DII), did not predict Iowa task performance.

Several questionnaires have been developed to assess Sensitivity to Punishment (BIS), Reward Sensitivity (BAS), and the closely associated trait of impulsivity. The BIS/BAS scales (Carver & White, 1994) tap BIS and three dimensions of BAS: Reward Responsiveness, Fun Seeking, and Drive. The Sensitivity to Punishment and Sensitivity to Reward Questionnaire (SPSRQ; Torrubia, Avila, Molto, & Caseras, 2001) is a measure consisting of only two subscales, one of which measures Sensitivity to Punishment, whereas the other assesses Sensitivity to Reward. Further, the Tridimensional Personality Questionnaire (TPQ; Cloninger, 1987) may also be used to assess aspects of the RST. Briefly, this scale contains three scales: Novelty Seeking and Reward Dependence, which of course reflect BAS sensitivity, and Harm Avoidance, which can be regarded as an equivalent of BIS. Finally, the DII (Dickman, 1990) is known as a specific impulsivity index, and assesses two aspects of this BAS-related construct (Jackson and Smillie, 2004, Smillie and Jackson, in press): Functional and Dysfunctional Impulsivity. Taken together, this brief review of BIS/BAS-related instruments indicates that there is considerable agreement on the fact that there seems to be only one factor reflecting sensitivity for punishment. This is also supported by empirical data showing that there are high correlations among traits that reflect sensitivity for punishment, such as the BIS scale, trait anxiety, Harm Avoidance, and Neuroticism (see Dawe, Gullo, & Loxton, 2004).

Yet, it remains highly doubtful whether Sensitivity to Reward and impulsivity refer to one unitary construct. In two recent papers, Dawe and colleagues (Dawe et al., 2004, Dawe and Loxton, 2004) propose a two-component model, with one factor pertaining to Reward Sensitivity and the other referring to Rash Impulsiveness. Although these two components seem to be related, they are thought to represent separate systems, each referring to their own neural substrates. The Reward Sensitivity component reflects a heightened sensitivity to unconditioned and conditioned rewarding stimuli, and as such bears strong similarity to Gray’s BAS. The Rash Impulsiveness component refers to the tendency to engage in rash, spontaneous behavior and is characterized by cognitive features that make the individual prone to disregard risk and neglect future consequences (Dawe et al., 2004).

A factor analytic study investigating the structure of the personality factors as defined in Eysenck’s, Gray’s, and Cloninger’s taxonomies (Zelenski & Larsen, 1999) revealed a solution consisting of three factors that were named reward sensitivity, impulsivity-thrill seeking and punishment sensitivity. These factors were comparable with those suggested by Dawe and Loxton (2004). The Zelenski and Larsen study further demonstrated that Reward Sensitivity predicted positive emotions, punishment sensitivity predicted negative emotions, while impulsivity-thrill seeking was less successful in predicting emotions. The present study further investigates the hypothesis that sensitivity to punishment consists of one personality dimension, whereas sensitivity to reward contains two separate dimensions, one actually referring to Reward Sensitivity and another pertains to Rash Impulsiveness (Dawe & Loxton, 2004). This issue was examined by performing a second-order principal component analysis on the subscales of questionnaires that intend to measure aspects of these biologically based personality traits (i.e., BIS/BAS scales, SPSRQ, TPQ, and DII). Further, to test the divergent and convergent validity of the predicted Sensitivity to Punishment and Sensitivity to Reward factors, relations to Eysenck’s personality traits, and scores of Positive and Negative Affect and Anhedonia (the inability to gain pleasure from normally pleasurable experiences) were also examined. It was predicted that the sensitivity to punishment dimension would be positively linked to Neuroticism and Negative Affect (Carver & White, 1994). In addition, it was expected that Rash Impulsiveness would be positively related to Eysenck’s dimension of extraversion (Corr et al., 1997, Revelle, 1997) and that Reward Sensitivity would be positively associated with Positive Affect (Carver and White, 1994, Jorm et al., 1999, Larsen and Ketelaar, 1991) and negatively with Anhedonia (Le Pen, Gaudet, Mortas, Mory, & Moreau, 2002).