Personality and responses to appetitive and aversive stimuli: the joint influence of behavioural approach and behavioural inhibition systems
Introduction
Gray’s (1987) theory (reinforcement sensitivity theory (RST)) of individual differences posits two major separable neurological systems involved in the regulation of personality and emotion. The behavioural approach (BAS) and the behavioural inhibition systems (BIS) facilitate the expression of appetitive (impulsive-sensation seeking traits) and aversive motivation (anxiety traits), respectively (Pickering & Gray, 1999). The BAS is sensitive to conditioned reward stimuli and responds with appetitive motivation (i.e., approach behaviour) when such cues are encountered. In contrast the BIS responds with aversive motivation (i.e., inhibitory behaviour) to conditioned cues signalling punishment. This model is regarded as an influential account of the neuropsychological determinants of personality and has been applied to a wide range of phenomena.
Recently however, inconsistent findings regarding associations between measures of personality and behavioural responses to reward and punishment has led to important modifications of RST. Firstly, Corr (2002a) has argued that, rather than separable as first hypothesised, the two motivational systems may jointly influence behaviour (the joint subsystems hypothesis). Secondly, it is suggested that associations between self-report measures of personality and actual reactions to motivational stimuli (in particular, appetitive) are influenced by individual reinforcement expectancies (Corr, 2002b) and the particular self-report instrument utilised (Corr, 2001). The primary aim of this paper is to provide an empirical examination of the joint subsystems hypothesis. The role of reinforcement expectancies and the utility of particular self-report BIS/BAS instruments will also be examined.
Corr’s (2001) modification of RST to include the joint subsystems hypothesis represents an attempt to account for inconsistencies in the evidence on associations between self-report personality and behavioural responses to reward and punishment. In particular, a number of studies have questioned the link between impulsive traits and responsivity to reward (e.g., Corr et al., 1995; Pickering et al., 1997; Zinbarg & Mohlman, 1998). Studies have reported no association between impulsivity and reward sensitivity (e.g., Pickering et al., 1997; Zinbarg & Mohlman, 1998) whilst others have found anxiety to be negatively associated with appetitive motivation (see Corr, 2002a; Larsen & Ketelaar, 1991). Impulsivity and anxiety also appear to display theoretically important interactions (Corr, 2001; Zinbarg & Mohlman, 1998). Gray’s original separable systems approach cannot easily explain such data as appetitive and aversive motivational systems are considered independent processes.
In contrast, the joint subsystems hypothesis predicts that responses to appetitive stimuli may be strongest in individuals with both high BAS and low BIS functioning. Similarly, aversive responses to punishment cues will be highest in individuals with a combination of high BIS and low BAS (Corr, 2001, Corr, 2002a). According to Corr, data consistent with the joint subsystems hypothesis is expected when experimental situations contain weak mixed reward and punishment cues which results in activation of both the BAS and the BIS. In contrast, data in line with the separable subsystems hypothesis is expected when experimental characteristics are such that strong single reinforcement (i.e., appetitive or aversive) stimuli are employed. Given that it is difficult in experimental situations to entirely eliminate cues for punishment (Corr, 2002a; Pickering et al., 1997) the joint subsystems contention is appealing and helps to explain the inconsistent data concerning interactive BIS/BAS effects (see Matthews & Gilliland, 1999, for a review).
To date however, only one experimental investigation of the joint subsystems hypothesis exists. Corr (2002a) reported data consistent with the joint subsystems hypothesis as high anxiety and low impulsivity participants displayed the most electromyographic (EMG) reactivity to aversive stimuli (unpleasant slides) while low anxiety and high impulsivity participants demonstrated the lowest levels of EMG assessed startle. Similarly, high impulsivity and low anxiety was associated with increased levels of disinhibition on an information-processing task, indicating that anxiety (BIS) antagonised the approach tendencies of the high impulsivity (BAS) participants.
A number of specific self-report indices of BIS/BAS functioning exist (e.g., Carver & White, 1994; Torrubia, Avila, Molto, & Caseras, 2001). The most reliable and valid of these is the Sensitivity to Punishment Sensitivity to Reward Questionnaire (SPSRQ; Torrubia et al., 2001) which demonstrates clear associations with appetitive and aversive motivational behaviour (e.g., Avila, 2001). Corr (2001) proposed that BIS/BAS functioning can be indexed by the Eysenck Personality Questionnaire (EPQ; Eysenck, Eysenck, & Barrett, 1987). Gray’s BIS/BAS dimensions represent a 30-degree rotation of Eysenck’s E and N factors, such that BIS is closest to N and BAS is closest to E (Gray, 1987). Furthermore, BIS inclines away and BAS toward the psychoticism factor (P; Corr, 2001). Thus, Corr (2001) proposes that EPQ-derived BAS and BIS should be calculated as: , [where x is the number of E items]. To date, no studies have examined the validity of these EPQ-derived BIS/BAS indices and their relationship with specifically designed measures (e.g., SPSRQ).
Recently, Corr (2001) has indicated that studies employing Gray’s model to examine reactions to appetitive and aversive situations need to assess levels of subjective reward to ensure that manipulations of motivation (in particular appetitive; see Corr, 2002b) are effective. Specifically, it has been argued that participant-perceived reward needs to be equal to or greater than expected levels of reward for appetitive manipulations to be considered effective and for positive relationships between BAS traits and actual reactions to reward to be observed. The influence of reinforcement expectancies on associations between self-report measures and actual responses to reward as described by Corr, 2001, Corr, 2002b is yet to be fully investigated.
To this end, levels of BIS/BAS functioning and performance on two behavioural tasks assumed to tap into appetitive and aversive motivational processes was assessed. The Card Arranging Reward Responsiveness Objective Test (CARROT) provides an index of reward sensitivity (Powell, Al-Adawi, Morgan, & Greenwood, 1996) and is gaining popularity as a behavioural measure of the BAS and appetitive motivational processes (Al-Adawi & Powell, 1997; Kambouropoulos & Staiger, 2001). The Q-TASK (Newman, Wallace, Schmitt, & Arnett, 1997) assesses aversive motivation to a conditioned signal of punishment (loss of money). According to Pickering et al. (1997) the Q-TASK represents a valid measure of BIS processes due to its ability to condition participants prior to assessing the influence of the putatively punishing cue on response times. However, the validity of the joint subsystems hypothesis, which predicts an interactive effect of BIS/BAS functioning on performance is yet to be investigated with these two behavioural indices. The hypotheses tested were that (a) responses to reward on the CARROT would be highest in high BAS/low BIS participants; (b) responses to punishment on the Q-TASK would be highest in high BIS/low BAS participants.
Section snippets
Participants
Seventy-eight participants were recruited primarily from a university population; 30 males (mean age=23.83, SD=5.12) and 48 females (mean age=23.33, SD=7.32).
Design
With the exception of reinforcement expectancies (repeated measures, pre-post), a between subjects design was utilised, with scores on CARROT and Q-TASK compared between median-split BIS/BAS groups (EPQ-R and SPSRQ). Administration of the CARROT and Q-TASK was counterbalanced to eliminate possible order effects.
Specific measure: Sensitivity to Punishment Sensitivity to Reward Questionnaire (SPSRQ; Torrubia et al., 2001)
This questionnaire comprises
Descriptive statistics
Summary data for the personality questionnaires are displayed in Table 1. Scores on all personality measures were comparable to published norms. Median-split procedures yielded four groups (+IMP/−ANX, −IMP/−ANX, −IMP/+ANX, +IMP/+ANX). For EPQ-derived IMP (median=49.5)/ANX (median=17.00), the groups were constituted as follows: +IMP/−ANX, N=26; +IMP/+ANX, N=14; −IMP/−ANX, N=13, −IMP/+ANX, N=25. For SPSRQ measures, the groups were as follows: +IMP/−ANX, N=18; +IMP/+ANX, N=18; −IMP/−ANX, N=20,
Discussion
This study aimed to provide an empirical test of the joint subsystems hypothesis and the role of reinforcement expectancies in associations with self-report and behavioural measures of BIS/BAS functioning. Correlational analysis of self-report and behavioural responses yielded data consistent with central RST predictions in that positive relationships were observed between (a) BIS traits (EPQ-derived ANX and SP) and Q-TASK behavioural inhibition, and (b) SR and CARROT reward responsivity. Group
Acknowledgements
The authors wish to thank Andrew Cooper for providing access to an initial Q-TASK program which was programmed by John Winkelman at Ballarat University, Australia and to Natalie Loxton for her valuable help with additional programming.
References (21)
Testing problems in J.A. Gray’s personality theory: A commentary on Matthews and Gilliland (1999)
Personality and Individual Differences
(2001)J.A. Gray’s reinforcement sensitivity theory: Tests of the joint subsystems hypothesis of anxiety and impulsivity
Personality and Individual Differences
(2002)J.A. Gray’s reinforcement sensitivity theory and frustrative nonreward: A theoretical note on expectancies in reactions to rewarding stimuli
Personality and Individual Differences
(2002)- et al.
Personality and affective modulation of the startle reflex
Personality and Individual Differences
(1995) Perspectives on anxiety and impulsivity: A commentary
Journal of Research in Personality
(1987)- et al.
The personality theories of H.J. Eysenck and J.A. Gray: A comparative review
Personality and Individual Differences
(1999) - et al.
Behavioural inhibition system functioning in anxious, impulsive and psychopathic individuals
Personality and Individual Differences
(1997) - et al.
The sensitivity to punishment and sensitivity to reward questionnaire (SPSRQ) as a measure of Gray’s anxiety and impulsivity dimensions
Personality and Individual Differences
(2001) - et al.
The influence of smoking on reward responsiveness and cognitive functions: A natural experiment
Addiction
(1997) - et al.
Motivational deficits after brain injury: A neuropsychological approach using new assessment techniques
Neuropsychology
(1998)