- Email: [email protected]
Impulsivity is associated with behavioral decision-making deficits
Available online at www.sciencedirect.com
Psychiatry Research 158 (2008) 155 – 163 www.elsevier.com/locate/psychres
Impulsivity is associated with behavioral decision-making deficits Ingmar H.A. Franken ⁎, Jan W. van Strien, Ilse Nijs, Peter Muris Institute of Psychology, Erasmus Affective Neuroscience Lab, Erasmus University Rotterdam, The Netherlands Received 11 August 2006; received in revised form 23 February 2007; accepted 3 June 2007
Abstract Impaired decision-making is a key-feature of many neuropsychiatric disorders. In the present study, we examined task performance in a healthy population consisting of those whose scores indicated high and low impulsivity on several behavioral decision-making tasks reflecting orbitofrontal functioning. The measures included tasks that assess decision-making with and without a learning component and choice flexibility. The results show that subjects high on impulsivity display an overall deficit in their decision-making performance as compared with subjects low on impulsivity. More specifically, subjects with high impulsivity show weaknesses in learning of reward and punishment associations in order to make appropriate decisions (reversal-learning task and Iowa Gambling Task), and impaired adaptation of choice behavior according to changes in stimulus–reward contingencies (reversal-learning task). Simple, non-learning, components of reward- and punishment-based decision-making (Rogers DecisionMaking Task) seem to be relatively unaffected. Above all, the results indicate that impulsivity is associated with a decreased ability to alter choice behavior in response to fluctuations in reward contingency. The findings add further evidence to the notion that trait impulsivity is associated with decision-making, a function of the orbitofrontal cortex. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Reversal learning; Decision-making; Impulsivity; Frontal functions; Reward
1. Introduction Decision-making is a cognitive function concerned with the process of reflecting on the consequences of a certain choice (Bechara, 2005). In recent years, decisionmaking functions have become a major research topic within neuropsychology, cognitive psychology, neuroscience, and economics. One of the topics of interest is the role of emotions on decision-making. In contrast to earlier theories that viewed decision-making as a rational ⁎ Corresponding author. Institute of Psychology, Erasmus University Rotterdam, Woudestein T12-35, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. Fax: +31 10 4089009. E-mail address: [email protected] (I.H.A. Franken).
choice, it is now believed that human decision-making is mainly based on emotions, especially on the expected hedonic outcome of the choice (Cabanac, 1992).1 In some classical studies, Bechara et al. (1994, 2000a) demonstrated that patients with frontal lobe damage have problems with emotional decision-making: they often pursue actions that bring some kind of immediate reward, despite severe long-term consequences such as the loss of job, home, and family. Several psychiatric and neurological conditions have been associated with such specific disturbances in emotional decision-making (for 1 Recent studies highlight the importance of explicit strategy and knowledge in decision-making, especially in the Iowa Gambling Task (Maia and McClelland, 2004).
0165-1781/$ - see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2007.06.002
156
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
reviews, see Rahman et al., 2001; Brand et al., 2006). For example, impaired decision-making has been reported in relation to addiction (Hester and Garavan, 2004), frontal lobe dementia (Rahman et al., 1999), borderline personality disorder (Bazanis et al., 2002), attention deficit hyperactivity disorder (Toplak et al., 2005), eating disorders (Cavedini et al., 2004), obsessive-compulsive disorders (Cavedini et al., 2002a), pathological gambling (Cavedini et al., 2002b), and disruptive behavior disorders (Ernst et al., 2003). One topic that needs to be resolved is whether these deficits are specific for clinical populations (Rogers, 2003). As it is well known that the personality trait of impulsivity is a major ingredient of several psychiatric (Moeller et al., 2001), personality (Rogers, 2003), and neurological disorders (Miller, 1992), including the aforementioned disorders. In the burgeoning neuropsychopharmacological literature on decision-making, the personality trait of impulsivity has been frequently suggested to be associated with weaknesses in decision-making. That is, impulsive persons display a decreased reflection on the consequences of their choice. These suggestions are in keeping with the notion that real-life decision-making involves choices that are based on expected but uncertain rewards and penalties, and that optimal choices are based on well-considered strategies. As such, it seems plausible to assume that impaired decision-making reflects a variety of impulse control problems (Morgan et al., 2006). Two studies addressed the relationship between impulsivity and decision-making in healthy subjects. A first study by Franken and Muris (2005) examined the link between decision-making, as indexed by the Iowa Gambling Task, and individual differences in functional and dysfunctional impulsivity and the impulsivity-related trait of reward sensitivity. The researchers found no relation between Iowa Gambling Task performance and dysfunctional impulsivity scores. However, reward sensitivity, a trait associated with impulsivity, was significantly related to the Iowa Gambling Task score. Surprisingly, this link was positive, suggesting that a higher level of an impulsivity-related trait is associated with better decisionmaking. In another study, also using the Iowa Gambling Task, Zermatten et al. (2005) found that decision-making was influenced by the impulsivity-related trait of ‘lack of premeditation’. Higher scores on ‘lack of premeditation‘ were positively linked to disadvantageous decision-making. Thus, while there is certain clinical evidence for the detrimental effects of impulsivity on decision-making, studies in non-clinical samples have yielded less convincing results. This may well have to do with the fact that these studies have only employed one specific test of
decision-making, the Iowa Gambling Task, so it remains to be seen whether different findings will emerge if other behavioral tests of decision-making are used. In this study, we focus on three decision-making tasks reflecting related but distinct aspects of decisionmaking: the Iowa Gambling Task (Bechara et al., 1994), which measures affective decision-making and includes a learning component, the Rogers Decision-Making Task (Rogers et al., 1999a), which also taps affective decision-making but does not include a learning component, and a probabilistic reversal-learning task (O'Doherty et al., 2001), which assesses affective decision-making with a learning component as well as the ability to alter choice behavior in response to fluctuations in reward contingency. In more detail, the Rogers Decision-Making Task taps explicit decisionmaking strategies without a reward-based learning element. This means that for each trial, all relevant information is presented to the participant; the participant does not need previously learned information in order to make a correct decision. The Iowa Gambling Task, on the other hand, measures decision-making strategies in which the participant has to learn to discriminate the advantageous choices from the disadvantageous choices. This means that the Iowa Gambling Task, unlike the Rogers Decision-Making Task, requires reward-based learning capacity (Fellows, 2004). The reversal-learning task (Rolls, 1999), measures not only reward-based learning but is also designed to index the adaptation of behavior according to changes in stimulus–reward contingencies (i.e., reversal learning; Clark et al., 2004), a capacity which is regarded as a basic requirement for normal social and emotional behavior (Cools et al., 2002; Kringelbach and Rolls, 2003; Rolls, 1996). Emotion-related visual reversal-learning tasks reflect the ability to alter choice behavior in response to fluctuations in reward contingency (Rahman et al., 1999). To summarize, all the tasks used in the present study reflect decision-making under risk. The Rogers Decision-Making Task only measures reward-based decision-making. The Iowa task adds a reward-based learning aspect to the decision-making process, whereas the probabilistic reversal task adds both a reward-based learning aspect and a reversal aspect2 that measures adaptive decision-making skills. Neuroimaging studies, which employed decisionmaking tasks without a learning element similar to the 2
It can be argued that the Iowa task also comprises a reversallearning aspect as the disadvantageous decks offer higher gains at the beginning of the task compared with decks C and D. However, this reversal aspect is minimal compared with the reversal-learning task.
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
Rogers Decision-Making Task, show that the orbitofrontal cortex is involved in this type of decision-making (Rogers et al., 1999b). Furthermore, neurological patients with prefrontal damage show significant deficits on all these types of decision-making tasks (Mavaddat et al., 2000; Rogers et al., 1999a). Neuroimaging studies employing decision-making tasks with a learning element, such as the Iowa Gambling Task, show medial prefrontal activity during this task (Fukui et al., 2005; Northoff et al., 2006). In addition, neurological patients with prefrontal damage show deficits on this task (for a review, see Bechara, 2004; Dunn et al., 2006). Several neuroimaging studies among healthy subjects show that the prefrontal cortex is also activated during a probabilistic reversal task (Cools et al., 2002; Kringelbach and Rolls, 2003; Remijnse et al., 2005). And again, deficits in reversal learning are also observed after prefrontal cortex damage (Hornak et al., 2004; Rahman et al., 1999; Rolls et al., 1994). To summarize, a wealth of data suggests that the orbitofrontal cortex is an essential structure for adequate decisionmaking (Clark et al., 2004; Ridderinkhof et al., 2004). In the present study, two main research questions were addressed. First, we examined whether healthy subjects with high-impulsivity scores would have impaired decision-making performance compared with subjects with low-impulsivity scores. We expected that subjects with high impulsivity would have an impaired overall performance on the Rogers Decision-Making Task, the Iowa Gambling Task, and the reversal-learning task. However, in line with the nature of impulsivity, it would be expected that subjects with high impulsivity would have shorter deliberation times on these decision-making tasks. Second, we explored which specific aspects of decision-making would be associated with impulsivity. 2. Methods 2.1. Participants and procedure A sample of 70 undergraduate psychology students (20% males) participated in the present study. The mean age of the sample was 20.8 years (S.D. = 2.9). The group was median-split into a high impulsiveness group (n = 30; high-impulsives) and a low impulsiveness group (n = 40; low-impulsives) based on their score (range = 0–17; median = 4) on the impulsiveness scale of the I7 questionnaire (Eysenck et al., 1985). The low-impulsives had a mean I7 impulsiveness score of 2.3 (S.D. = 1.4) and the mean score of the high-impulsives was 9.5 (S.D. = 3.5). After participants provided informed consent, they were asked to complete the questionnaires and then the decision-making tasks were administered in the following
157
order: reversal-learning task, Rogers Decision-Making Task, Iowa Gambling Task. The study was approved by the ethical committee of the Erasmus Medical Centre. 2.2. Instruments 2.2.1. I7 Impulsiveness Scale The Dutch version of the 19-item Impulsiveness Scale of the I7 questionnaire (Eysenck et al., 1985; Lijffijt et al., 2005) was used as the measure of impulsivity. In this questionnaire, impulsiveness is regarded as acting without first considering the possible consequences. The impulsiveness scale contains items such as ‘Do you often buy things on impulse?’, and ‘Are you an impulsive person?’ Respondents answered the items on a dichotomous (i.e. ‘yes’ and ‘no’) response scale. The I7 is a frequently employed questionnaire to assess the personality trait of impulsivity (see Lijffijt et al., 2005). The scale has good psychometric properties, with good reliability and validity. Cronbach's alpha of the Dutch version was reported to be 0.80 (Lijffijt et al., 2005); in this study, alpha was 0.72. 2.2.2. Positive and Negative Affect Scales In order to compare both impulsivity groups on current affect, the Positive and Negative Affect Scales (PANAS; Watson et al., 1988) were administered. The PANAS is a 20-item bidimensional mood inventory. The Positive Affect scale reflects the extent to which a person feels enthusiastic, active, and alert (Watson et al., 1988), whereas the Negative Affect scale is an index of subjective distress and unpleasurable engagement that subsumes a variety of aversive mood states, including anger, contempt, disgust, guilt, fear, and nervousness (Watson et al., 1988). Psychometric properties of the PANAS scales are adequate, with good reliability and validity (Boon and Peeters, 1999; Watson et al., 1988). Cronbach's alpha of the Dutch version was reported to be 0.89 and 0.86 for the PA and NA scales, respectively (Boon and Peeters, 1999). 2.2.3. Rogers Decision-Making Task The Rogers Decision-Making task was identical to the task as described by Rogers et al. (2003). However, the present version consisted of 40 trials randomly presented within two blocks. On each trial, participants had to choose one of the two presented gambles in order to collect as many points as possible. The gambles were represented in the form of a histogram. The height of this histogram indicated the probability of gaining, and the possible amounts of gains were indicated in green above the histogram. The possible amounts of losses
158
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
were indicated in red beneath the histogram (see Rogers et al., 2003). For each trial, one gamble acted as the control gamble, which consisted of a 50% probability of winning 10 points and a 50% probability of losing 10 points. The other simultaneously presented gamble was the “experimental gamble”. This trial type varied on three factors: the probability of winning, high (75%) vs. low (25%); the possible gains, large (80 points) vs. small (20 points); and the possible losses, large (80 points) vs. small (20 points). This procedure resulted in eight different gambling situations, and because the experimental gamble appeared randomly on the left or the right of the screen, this resulted in a total of 16 different gambling situations. Participants had to choose one of the two gambles by pressing the numbers ‘1’ or ‘2’ on a keyboard. Three indices of decision-making were derived from the Rogers Decision-Making Task. First, we took the overall proportion of choices of the experimental over the control gamble. Furthermore, following Rogers et al. (2003), we also included two types of “all or nothing” trials. In the “gains only” trial type, participants had to choose between a guaranteed win of 40 points or a 50% chance of winning 80 points. In the “losses only” trial type, participants had to choose between a guaranteed loss of 40 points or a 50% chance of losing 80 points. For the ‘gains only’ and ‘losses only’ trials, the dependent measure was the proportion of choices on which volunteers chose the uncertain outcome (experimental trial). Third, we also recorded the time needed to make a decision (deliberation time). After choosing, participants received feedback on the amount of points they won or lost and the total amount of points they had gathered so far. 2.2.4. Iowa Gambling Task In the present study we used the computerized version of the Iowa Gambling Task (Bechara et al., 1994), which consisted of 100 successive trials in which subjects were required to choose a card from one of four decks. The participants were instructed to try to gain as much money as possible by drawing cards from one of four decks (A, B, C, D). The decisions to choose from the decks are motivated by reward and punishment schedules inherent in the task. Two of the decks (A and B) are disadvantageous, producing immediate gains (large rewards), but these are accompanied by larger losses in the long run (larger punishments). The C and D decks are advantageous: here, gains are modest but more consistent and losses are smaller. A net score ((C + D) − (A + B)) was computed, with a higher score indicating that a subject is more often choosing advantageous decks.
2.2.5. Probabilistic reversal-learning task This task was similar to the one used by O'Doherty et al. (2001), except for the fact that we used a fixed number of trials (i.e. 100), and in our task a reversal took place after five correct choices. Subjects had to choose one of two stimuli (S+ and S−, which were easily discerned geometrical figures) that were presented simultaneously on the computer screen above each other. The S+ (advantageous) stimulus had the following properties: a reward–punishment ratio of 70:30, a reward range of 80–250 points, and a punishment range of 10–60 points. The S− (disadvantageous) stimulus had the following properties: a reward–punishment ratio of 40:60, a reward range of 30–60 points, and a punishment range of 250–600 points. Accordingly, persistent selection of S+ resulted in an overall gain, and persistent selection of S− resulted in an overall loss. When starting, subjects were unaware whether the upper or the lower geometrical figure was the S+ or S−. They had to learn by trial and error which one was the advantageous stimulus. Every time the participants chose the S+ stimulus five times in a row, there was an S+ and S− reversal (which the participant was unaware of). Refer to O'Doherty et al. (2001) for a detailed description. Scores derived from the probabilistic reversal task were the total number of reversal contingencies, the total number of errors, and the mean deliberation time of the choices (Swainson et al., 2000; Evers et al., 2005). Furthermore, a win-stay and lose-stay analysis was conducted (Budhani and Blair, 2005). To change choice after large gain can be regarded as disadvantageous choice behavior, and change after large loss could be regarded as advantageous choice behavior. This resulted in two additional scores: the proportion of choices that can be classified as “stay after large win” and the proportion of choices that can be classified as “stay after large loss”. 2.3. Data analysis Before analyzing the decision-making variables, we examined the characteristics of the two groups. Age, affect (PANAS score) and gender differences between the two groups were analyzed using Students t-test or a Fischer's Exact test. To answer the main research questions, the groups were compared on their behavioral decision-making performance by a multivariate analysis of variance (MANOVA) with a Bonferroni correction for multiple comparisons. Additionally, in order to explore the relationships among various decision-making measures, Pearson correlation coefficients of these measures were calculated. Further, to test which of the behavioral
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163 Table 1 Mean values (standard deviations) of low- and high-impulsives on several indices of decision-making behavior
IGT net score RDMT proportion of risky choices RDMT proportion of risky choice on “all or nothing trials” RDMT mean deliberation time (ms) PRT stay after large gain proportion PRT stay after large loss proportion PRT number of reversals PRT number of errors PRT mean deliberation time (ms)
Low-impulsives
High-impulsives
13.6 (30.7) 0.52 (0.15)
− 0.9 (23.5) a 0.58 (0.13)
0.43 (0.23)
0.50 (0.16)
3192.7 (1423.5)
2402.5 (1123.6) a
0.85 (0.14)
0.79 (0.13) a
0.27 (0.16)
0.36 (0.17) a
7.3 (3.6) 34.9 (9.3) 620.2 (330.1)
5.2 (3.0) a 39.6 (8.4) a 547.4 (190.2)
Notes. N = 70. IGT = Iowa Gambling Task, RDMT = Rogers DecisionMaking Task, PRT = probabilistic reversal task. a Significant differences in MANOVA.
measures was the most powerful predictor of self-reported impulsivity, a stepwise multiple regression analysis was employed including all behavioral measures as independent variables, and self-reported impulsivity as the dependent variable. 3. Results
159
Subsequent pairwise comparisons indicated that highimpulsives had lower scores on the Iowa Gambling Task compared to low-impulsives, F(1,69) = 4.4, P b 0.05, η2 = 0.06. In Fig. 1, the mean scores of both impulsivity groups over the five blocks on the Iowa Gambling Task are shown. On the Rogers Decision-Making Task, highimpulsives had shorter deliberation times, F(1,69) = 4.4, P b 0.05, η2 = 0.09. High-impulsives did not made overall more risky choices on the Rogers Decision-Making Task F(1,69) = 1.9, P = 0.09, η2 = 0.04. On the probabilistic reversal task, high-impulsives reached a lower number of reversals F(1,69) = 6.3, P b 0.05, η2 = 0.09, had a higher proportion of stay after large loss choices, F(1,69) = 5.0, P b 0.05, η2 = 0.07, and a lower proportion of stay after large gain choices, F(1,69) = 5.0, P b 0.05, η2 = 0.06, and had a higher number of errors, F(1,69) = 4.7, P b 0.05, η2 = 0.07. Regression analyses showed that the Iowa Gambling Task score was the most powerful predictor of self-reported impulsivity and explained 27% of the variance of self-reported impulsivity (standardized β = −0.27, partial r = −0.27, P b 0.05). The Pearson correlation coefficients among the three decision-making questionnaires are displayed in Table 2. Besides obvious correlations among sub-scores within tasks, we also observed a significant positive correlation between the net score of the Iowa Gambling Task and the number of reversals on the reversallearning task (r = 0.24, P b 0.05). Further, the deliberation times on the Rogers Decision-Making Task and
3.1. Group characteristics The groups of low-impulsives and high-impulsives were equal with regard to gender ratio (23% vs. 17% males; P = 0.76), mean age (20.7 vs. 21.0 years, respectively; P = 0.62), mean positive affect score (33.4 vs. 34.8, respectively; P = 0.22), and mean negative affect score (19.4 vs. 18.4, respectively; P = 0.55). 3.2. Decision-making Mean scores of both groups on the decision-making variables are displayed in Table 1. There was an overall effect of impulsivity on the behavioral measures of choice behavior, Wilks' Lambda = 0.85, F(10,57) = 2.1, P b 0.05, η 2 = 0.27, indicating that high-impulsives displayed generally impaired decision-making performance as compared to low-impulsives.3 3
We conducted an extreme group analysis with top and bottom 20% impulsivity scores. This analysis yielded a similar pattern of results on the individual tasks effect on decision-making. However, due to the loss of power, some of the differences dropped below the significance level.
Fig. 1. Mean net score (advantageous–disadvantages choices) of both groups on the Iowa Gambling Task over the five blocks.
160
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
Table 2 Correlations among the behavioral measures of decision-making
1. IGT net score 2. RDMT proportion of risky choices 3. RDMT proportion of risky choice on “all or nothing trials” 4. RDMT mean deliberation time 5. PRT stay after large gain proportion 6. PRT stay after large loss proportion 7. PRT number of reversals 8. PRT number of errors 9. PRT mean deliberation time
1
2
3
4
5
6
7
8
– − 0.21 − 0.02 0.17 0.13 − 0.15 0.24⁎ − 0.18 − 0.03
– 0.17 −0.17 −0.11 0.20 −0.14 0.20 −0.07
– − 0.02 − 0.01 0.07 0.00 − 0.06 − 0.13
– − 0.19 − 0.04 − 0.05 0.06 0.27⁎
– − 0.22 0.61⁎⁎ − 0.39⁎⁎ − 0.03
– − 0.33⁎⁎ 0.68⁎⁎ 0.10
– − 0.77⁎⁎ − 0.18
– 0.26⁎
Notes. N = 70. IGT = Iowa Gambling Task, RDMT = Rogers Decision Making Task, PRT = Probalistic Reversal Task. ⁎P b 0.05; ⁎⁎P b 0.01.
the Reversal Task were positively correlated (r = 0.27, P b 0.05). 4. Discussion The current study examined the relation between behavioral decision-making and the personality trait of impulsivity. It was found that high-impulsives displayed a general deficit in their decision-making abilities as compared to low-impulsives. That is, high-impulsives had sub-optimal performance on various decision-making indices. The effect size of the overall performance deficit was large (Cohen, 1988). More specifically, impairments were found on several measures: High-impulsives displayed a decreased performance in learning of reward and punishment associations in order to make appropriate decisions, and impaired adaptation of choice behavior according to changes in stimulus–reward contingencies. The results on simple, non-learning, components of reward- and punishment-based decision-making were less clear; these seemed relatively unaffected as compared with the tasks that require reward-based learning strategies. Above all, the results indicate that impulsivity is associated with a decreased ability to alter choice behavior in response to fluctuations in reward contingency. Interestingly, impulsives display lowered deliberation times on the Rogers Decision-Making Task, and it would be tempting to conclude that high-impulsives act before considering the consequences of their choice. However, the performance on Rogers' task is equal for both groups, suggesting that impulsives have a more efficient decision-making strategy on this specific task. The present study demonstrated that high-impulsives have lower scores on the Iowa Gambling Task net score compared with low-impulsives. In addition, the results of regression analyses showed also that the Iowa Gambling Task net score was the most powerful
predictor of self-reported impulsivity. The observed association between impulsivity and Iowa Gambling Task performance is in line with the study of Zermatten et al. (2005) in which the authors found a negative relation between the impulsivity-related trait of ‘lack premeditation’ and the Iowa Gambling Task score. However, this outcome is at variance with other studies among healthy populations (Franken and Muris, 2005). In contrast to the present study, in that study no correlation between decision-making, as measured by the Iowa Gambling Task and impulsivity could be observed. A fact that may explain this divergence is that in the Franken and Muris study a gambling task with a progressive schedule of increased delayed punishments was used (punishments increased progressively as subjects selected more cards from the disadvantageous decks, see Bechara et al., 2000b) resulting in a somewhat different task that may not be related to impulsivity. Further, in both studies different questionnaires were employed for measuring impulsivity (i.e., the Dickman Impulsivity Inventory vs. the I7). It is conceivable that different impulsivity questionnaires tap different behavioral aspects of impulsivity. In addition, the observed association between impulsivity and decision-making is also at variance with studies among psychiatric patient populations which did not observe a relation between impulsivity and behavioral decision-making (Bazanis et al., 2002; Jollant et al., 2005). Clearly, more research on which aspects of decision-making are related to various indices of impulsivity is needed. The present results contribute to the discussion about the inconsistency of the definition of impulsivity (Evenden, 1999a,b). Roughly, a distinction can be made between definitions stressing the inhibition of an activated response, as is measured using Go–Nogo tasks, and definitions stressing the integration of reward/punishment information in order to choose between two or more response options, as is measured using the current decision-
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
making tasks. Interestingly, in a study of Lijffijt et al. (2004), the authors found no difference between low- and high-impulsives on behavioral measures of inhibitory motor control using the same impulsivity measure as in the current study. This suggests that the I7, and similar impulsivity questionnaires tapping trait impulsivity, reflect problems with the integration of reward/punishment information and not the inhibition of motor control. Besides the distinction between response inhibition and integration of reward/punishment integration, other subclassifications of impulsivity have been suggested. For example, Barratt (1985) distinguishes three different aspects of impulsivity: cognitive impulsiveness which is involved in making quick cognitive decisions, motor impulsiveness which is involved in acting without thinking, and non-planning impulsiveness which involved a lack of “looking into the future” or planning. The presently used impulsiveness scale, the I7, reflects poor behavioral control and inability to delay gratification of which the latter aspect is obviously close to the behavioral decision-making instruments used in the present study. Using multifactorial impulsivity scales, it would be interesting to study which specific facets of impulsivity relate to specific neuropsychological mechanisms of decision-making. Further, these results show also that reversal-learning tasks tap a somewhat different aspect of impulsivity and may be a useful addition to the behavioral measures of impulsivity. Our additional correlational analyses among the behavioral tasks show that, although all measures tap decision-making, these do not measure equal aspects. Except for the small correlations between the Iowa Gambling Task net score and the number of reversals on the reversal-learning task, and the small correlation between the deliberation times on the Rogers Decision-Making Task and the Reversal Task, there were no significant correlations observed between the various decision-making tasks. This suggests that these measures indeed tap different aspects of choice behavior. Note that the correlation between the Iowa Gambling Task net score and the number of reversals on the reversal-learning task suggests that these tasks tap, at least partly, a similar concept. This supports Clark et al.'s (2004) notion that parallels exist between reversal learning and decision-making. The present results also add new evidence to the notion that impulsivity is associated with frontal lobe deficits (King et al., 2003; Miller, 1992; Spinella, 2004), more specifically, since both gambling tasks and reversal learning are associated with orbitofrontal functioning, suggesting a specific site of action. Probably, the association between impulsivity and executive functions is due to common neuroanatomical pathways. It is known
161
that both impulsivity and decision-making functions have common prefrontal neural substrates. The present results indicate that persons scoring high on the I7 impulsivity questionnaire display sub-optimal decisionmaking strategies, which are probably modulated by the ventral prefrontal cortex (Clark et al., 2004). It is interesting to consider these results in the light of the discussion of whether decision-making deficits that characterize drug addiction (Clark and Robbins, 2002) are the result of drug-induced changes in the prefrontal cortex (e.g. Schoenbaum et al., 2006) or are already present before the use of drugs. Although the present study is not specifically designed to resolve this question, this can only be done by prospective designs; it shows that decision-making deficits can be present in impulsive, normal individuals. Because it is known that drug addiction is associated with augmented levels of impulsivity (Dawe and Loxton, 2004) which results in drug use (Sher et al., 2000), the current results suggest that an individual with drug addiction may have preexisting impulsivity associated decision-making deficits. Again, it must be kept in mind that this issue can only be resolved in studies with longitudinal designs. The present study has several limitations. We employed a sample of young students with an overrepresentation of females. However, we do not think that this is critical for the conclusions since it is not plausible to assume that the relationship between behavioral decision-making and self-reported impulsivity is different for other healthy populations. A further limitation is that we have no data on psychiatric morbidity in this population. However, previous studies among student populations show that these numbers are typically very low. In addition, the reliability of the behavioral decisionmaking tasks employed in the present study is unknown. Clearly, future studies should address this important topic. Lastly, impulsivity status was based upon a median-split rather than extreme values or clinical cutoff points. However, it must be noted that clinical cutpoints for impulsivity have not yet been developed. To summarize, the results indicate that impulsivity is associated with a decreased ability to alter behavior in response to fluctuations in emotional significance to stimuli. The findings add further evidence to the notion that trait impulsivity is associated with decision-making, a function of the orbitofrontal cortex. Acknowledgements This work was supported by the Netherlands Organization for Scientific Research (NWO). The authors thank Dr. A. Bechara for providing the Iowa
162
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
task, Manon de Long for writing a Dutch version of the Iowa task software, and Saskia van Dantzig for assisting with the learning reversal and decision-making tasks. References Barratt, E.S., 1985. Impulsiveness subtraits: arousal and information processing. In: Spence, J.T., Izard, C.E. (Eds.), Motivation, Emotion and Personality. Elsevier North Holland, pp. 137–146. Bazanis, E., Rogers, R.D., Dowson, J.H., Taylor, P., Meux, C., Staley, C., Nevinson-Andrews, D., Taylor, C., Robbins, T.W., Sahakian, B.J., 2002. Neurocognitive deficits in decision-making and planning of patients with DSM-III-R borderline personality disorder. Psychological Medicine 32, 1395–1405. Bechara, A., 2004. The role of emotion in decision-making: evidence from neurological patients with orbitofrontal damage. Brain and Cognition 55, 30–40. Bechara, A., 2005. Decision making, impulse control and loss of willpower to resist drugs: a neurocognitive perspective. Nature Neuroscience 8, 1458–1463. Bechara, A., Damasio, A.R., Damasio, H., Anderson, S.W., 1994. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 50, 7–15. Bechara, A., Damasio, H., Damasio, A.R., 2000a. Emotion, decision making and the orbitofrontal cortex. Cerebral Cortex 10, 295–307. Bechara, A., Tranel, D., Damasio, H., 2000b. Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions. Brain 123, 2189–2202. Boon, M.T., Peeters, F.P., 1999. Affectieve dimensies bij depressie en angst (Dimensions of affectivity in depression and anxiety). Tijdschrift voor Psychiatrie 41, 109–113. Brand, M., Labudda, K., Markowitsch, H.J., 2006. Neuropsychological correlates of decision-making in ambiguous and risky situations. Neural Networks 19, 1266–1276. Budhani, S., Blair, R.J.R., 2005. Response reversal and children with psychopathic tendencies: success is a function of salience of contingency change. Journal of Child Psychology and Psychiatry 46, 972–981. Cabanac, M., 1992. Pleasure: the common currency. Journal of Theoretical Biology 155, 173–200. Cavedini, P., Riboldi, G., D'Annucci, A., Belotti, P., Cisima, M., Bellodi, L., 2002a. Decision-making heterogeneity in obsessive–compulsive disorder: ventromedial prefrontal cortex function predicts different treatment outcomes. Neuropsychologia 40, 205–211. Cavedini, P., Riboldi, G., Keller, R., D'Annucci, A., Bellodi, L., 2002b. Frontal lobe dysfunction in pathological gambling patients. Biological Psychiatry 51, 334–341. Cavedini, P., Bassi, T., Ubbiali, A., Casolari, A., Giordani, S., Zorzi, C., Bellodi, L., 2004. Neuropsychological investigation of decision-making in anorexia nervosa. Psychiatry Research 127, 259–266. Clark, L., Robbins, T.W., 2002. Decision-making deficits in drug addiction. Trends in Cognitive Sciencess 6, 361–363. Clark, L., Cools, R., Robbins, T.W., 2004. The neuropsychology of ventral prefrontal cortex: decision-making and reversal learning. Brain and Cognition 55, 41–53. Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences, 2 ed. Erlbaum, Hillsdale, NJ. Cools, R., Clark, L., Owen, A.M., Robbins, T.W., 2002. Defining the neural mechanisms of probabilistic reversal learning using
event-related functional magnetic resonance imaging. Journal of Neuroscience 22, 4563–4567. Dawe, S., Loxton, N.J., 2004. The role of impulsivity in the development of substance use and eating disorders. Neuroscience and Biobehavioral Reviews 28, 343–351. Dunn, B.D., Dalgleish, T., Lawrence, A.D., 2006. The somatic marker hypothesis: a critical evaluation. Neuroscience and Biobehavioral Reviews 30, 239–271. Ernst, M., Grant, S.J., London, E.D., Contoreggi, C.S., Kimes, A.S., Spurgeon, L., 2003. Decision making in adolescents with behavior disorders and adults with substance abuse. American Journal of Psychiatry 160, 33–40. Evenden, J.L., 1999a. Impulsivity: a discussion of clinical and experimental findings. Journal of Psychopharmacology 13, 180–192. Evenden, J.L., 1999b. Varieties of impulsivity. Psychopharmacology 146, 348–361. Evers, E.A.T., Cools, R., Clark, L., van der Veen, F.M., Jolles, J., Sahakian, B.J., Robbins, T.W., 2005. Serotonergic modulation of prefrontal cortex during negative feedback in probabilistic reversal learning. Neuropsychopharmacology 30, 1138–1147. Eysenck, S.B.G., Pearson, P.R., Easting, G., Allsopp, J.F., 1985. Age norms for impulsiveness, venturesomeness and empathy in adults. Personality and Individual Differences 6, 613–619. Fellows, L.K., 2004. The cognitive neuroscience of human decision making: a review and conceptual framework. Behavioral and Cognitive Neuroscience Reviews 3, 159–172. Franken, I.H.A., Muris, P., 2005. Individual differences in decisionmaking. Personality and Individual Differences 39, 991–998. Fukui, H., Murai, T., Fukuyama, H., Hayashi, T., Hanakawa, T., 2005. Functional activity related to risk anticipation during performance of the Iowa Gambling Task. NeuroImage 24, 159–253. Hester, R., Garavan, H., 2004. Executive dysfunction in cocaine addiction: evidence for discordant frontal, cingulate, and cerebellar activity. Journal of Neuroscience 24, 11017–11022. Hornak, J., O'Doherty, J., Bramham, J., Rolls, E.T., Morris, R.G., Bullock, P.R., Polkey, C.E., 2004. Reward-related reversal learning after surgical excisions in orbito-frontal or dorsolateral prefrontal cortex in humans. Journal of Cognitive Neuroscience 16, 463–478. Jollant, F., Bellivier, F., Leboyer, M., Astruc, B., Torres, S., Verdier, R., Castelnau, D., Malafosse, A., Courtet, P., 2005. Impaired decision making in suicide attempters. American Journal of Psychiatry 162, 304–310. King, J.A., Tenney, J., Rossi, V., Colamussi, L., Burdick, S., 2003. Neural substrates underlying impulsivity. Annals of the New York Academy of Sciences 1008, 160–169. Kringelbach, M.L., Rolls, E.T., 2003. Neural correlates of rapid reversal learning in a simple model of human social interaction. NeuroImage 20, 1371–1383. Lijffijt, M., Bekker, E.M., Quik, E.H., Bakker, J., Kenemans, J.L., Verbaten, M.N., 2004. Differences between low and high trait impulsivity are not associated with differences in inhibitory motor control. Journal of Attentional Disorders 8, 25–32. Lijffijt, M., Kenemans, J.L., Caci, H., 2005. Validation of the Dutch translation of the I7 questionnaire. Personality and Individual Differences 38, 1123–1133. Maia, T.V., McClelland, J.L., 2004. A reexamination of the evidence for the somatic marker hypothesis: what participants really know in the Iowa gambling task. Proceedings of the National Academy of Sciences of the United States of America 101, 16075–16080. Mavaddat, N., Kirkpatrick, P.J., Rogers, R.D., Sahakian, B.J., 2000. Deficits in decision-making in patients with aneurysms of the anterior communicating artery. Brain 123, 2109–2117.
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163 Miller, L.A., 1992. Impulsivity, risk-taking, and the ability to synthesize fragmented information after frontal lobectomy. Neuropsychologia 30, 69–79. Moeller, F.G., Barrat, E.S., Dougherthy, D.M., Schmitz, J.M., Swann, A.C., 2001. Psychiatric aspects of impulsivity. American Journal of Psychiatry 158, 1783–1793. Morgan, M.J., Impallomeni, L.C., Pirona, A., Rogers, R.D., 2006. Elevated impulsivity and impaired decision-making in abstinent ecstasy (MDMA) users compared to polydrug and drug-naive controls. Neuropsychopharmacology 31, 1562–1573. Northoff, G., Grimm, S., Boeker, H., Schmidt, C., Bermpohl, F., Heinzel, A., Hell, D., Boesiger, P., 2006. Affective judgment and beneficial decision making: Ventromedial prefrontal activity correlates with performance in the Iowa Gambling Task. Human Brain Mapping 27, 572–587. O'Doherty, J., Kringelbach, M.L., Rolls, E.T., Hornak, J., Andrews, C., 2001. Abstract reward and punishment representations in the human orbitofrontal cortex. Nature Neuroscience 4, 95–102. Rahman, S., Sahakian, B.J., Hodges, J.R., Rogers, R.D., Robbins, T.W., 1999. Specific cognitive deficits in mild frontal variant frontotemporal dementia. Brain 122, 1469–1493. Rahman, S., Sahakian, B., Cardinal, R., Rogers, R., Robbins, T., 2001. Decision making and neuropsychiatry. Trends in Cognitive Sciences 5, 271–277. Remijnse, P.L., Nielen, M.M.A., Uylings, H.B.M., Veltman, D.J., 2005. Neural correlates of a reversal learning task with an affectively neutral baseline: an event-related fMRI study. NeuroImage 26, 609–618. Ridderinkhof, K.R., van den Wildenberg, W.P., Segalowitz, S.J., Carter, C.S., 2004. Neurocognitive mechanisms of cognitive control: the role of prefrontal cortex in action selection, response inhibition, performance monitoring, and reward-based learning. Brain and Cognition 56, 129–140. Rogers, R.D., 2003. Neuropsychological investigations of the impulsive personality disorders. Psychological Medicine 33, 1335–1340. Rogers, R.D., Everitt, B.J., Baldacchino, A., Blackshaw, A.J., Swainson, R., Wynne, K., Baker, N.B., Hunter, J., Carthy, T., Booker, E., London, M., Deakin, J.F., Sahakian, B.J., Robbins, T.W., 1999a. Dissociable deficits in the decision-making cognition of chronic amphetamine abusers, opiate abusers, patients with focal damage to prefrontal cortex, and tryptophan-depleted normal volunteers: evidence for monoaminergic mechanisms. Neuropsychopharmacology 20, 322–339.
163
Rogers, R.D., Owen, A.M., Middleton, H.C., Williams, E.J., Pickard, J.D., Sahakian, B.J., Robbins, T.W., 1999b. Choosing between small, likely rewards and large, unlikely rewards activates inferior and orbital prefrontal cortex. Journal of Neuroscience 19, 9029–9038. Rogers, R.D., Tunbridge, E.M., Bhagwagar, Z., Drevets, W.C., Sahakian, B.J., Carter, C.S., 2003. Tryptophan depletion alters the decisionmaking of healthy volunteers through altered processing of reward cues. Neuropsychopharmacology 28, 153–162. Rolls, E.T., 1996. The orbitofrontal cortex. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 351, 1433–1443. Rolls, E.T., 1999. The Brain and Emotion. Oxford University Press, New York. Rolls, E.T., Hornak, J., Wade, D., McGrath, J., 1994. Emotion-related learning in patients with social and emotional changes associated with frontal lobe damage. Journal of Neurology, Neurosurgery and Psychiatry 57, 1518–1524. Schoenbaum, G., Roesch, M.R., Stalnaker, T.A., 2006. Orbitofrontal cortex, decision-making and drug addiction. Trends in Neurosciences 29, 116–124. Sher, K.J., Bartholow, B.D., Wood, M.D., 2000. Personality and substance use disorders: a prospective study. Journal of Consulting and Clinical Psychology 68, 818–829. Spinella, M., 2004. Neurobehavioral correlates of impulsivity: evidence of prefrontal involvement. International Journal of Neuroscience 114, 95–104. Swainson, R., Rogers, R.D., Sahakian, B.J., Summers, B.A., Polkey, C.E., Robbins, T.W., 2000. Probabilistic learning and reversal deficits in patients with Parkinson's disease or frontal or temporal lobe lesions: possible adverse effects of dopaminergic medication. Neuropsychologia 38, 596–612. Toplak, M.E., Jain, U., Tannock, R., 2005. Executive and motivational processes in adolescents with Attention-Deficit-Hyperactivity Disorder (ADHD). Behavioral and Brain Functions 1, 8. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS Scales. Journal of Personality and Social Psychology 54, 1063–1070. Zermatten, A.M., Van der Linden, M.P., d'Acremont, M.M., Jermann, F.M., Bechara, A.P., 2005. Impulsivity and decision making. Journal of Nervous and Mental Disease 193, 647–650.
Psychiatry Research 158 (2008) 155 – 163 www.elsevier.com/locate/psychres
Impulsivity is associated with behavioral decision-making deficits Ingmar H.A. Franken ⁎, Jan W. van Strien, Ilse Nijs, Peter Muris Institute of Psychology, Erasmus Affective Neuroscience Lab, Erasmus University Rotterdam, The Netherlands Received 11 August 2006; received in revised form 23 February 2007; accepted 3 June 2007
Abstract Impaired decision-making is a key-feature of many neuropsychiatric disorders. In the present study, we examined task performance in a healthy population consisting of those whose scores indicated high and low impulsivity on several behavioral decision-making tasks reflecting orbitofrontal functioning. The measures included tasks that assess decision-making with and without a learning component and choice flexibility. The results show that subjects high on impulsivity display an overall deficit in their decision-making performance as compared with subjects low on impulsivity. More specifically, subjects with high impulsivity show weaknesses in learning of reward and punishment associations in order to make appropriate decisions (reversal-learning task and Iowa Gambling Task), and impaired adaptation of choice behavior according to changes in stimulus–reward contingencies (reversal-learning task). Simple, non-learning, components of reward- and punishment-based decision-making (Rogers DecisionMaking Task) seem to be relatively unaffected. Above all, the results indicate that impulsivity is associated with a decreased ability to alter choice behavior in response to fluctuations in reward contingency. The findings add further evidence to the notion that trait impulsivity is associated with decision-making, a function of the orbitofrontal cortex. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Reversal learning; Decision-making; Impulsivity; Frontal functions; Reward
1. Introduction Decision-making is a cognitive function concerned with the process of reflecting on the consequences of a certain choice (Bechara, 2005). In recent years, decisionmaking functions have become a major research topic within neuropsychology, cognitive psychology, neuroscience, and economics. One of the topics of interest is the role of emotions on decision-making. In contrast to earlier theories that viewed decision-making as a rational ⁎ Corresponding author. Institute of Psychology, Erasmus University Rotterdam, Woudestein T12-35, P.O. Box 1738, 3000 DR Rotterdam, The Netherlands. Fax: +31 10 4089009. E-mail address: [email protected] (I.H.A. Franken).
choice, it is now believed that human decision-making is mainly based on emotions, especially on the expected hedonic outcome of the choice (Cabanac, 1992).1 In some classical studies, Bechara et al. (1994, 2000a) demonstrated that patients with frontal lobe damage have problems with emotional decision-making: they often pursue actions that bring some kind of immediate reward, despite severe long-term consequences such as the loss of job, home, and family. Several psychiatric and neurological conditions have been associated with such specific disturbances in emotional decision-making (for 1 Recent studies highlight the importance of explicit strategy and knowledge in decision-making, especially in the Iowa Gambling Task (Maia and McClelland, 2004).
0165-1781/$ - see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2007.06.002
156
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
reviews, see Rahman et al., 2001; Brand et al., 2006). For example, impaired decision-making has been reported in relation to addiction (Hester and Garavan, 2004), frontal lobe dementia (Rahman et al., 1999), borderline personality disorder (Bazanis et al., 2002), attention deficit hyperactivity disorder (Toplak et al., 2005), eating disorders (Cavedini et al., 2004), obsessive-compulsive disorders (Cavedini et al., 2002a), pathological gambling (Cavedini et al., 2002b), and disruptive behavior disorders (Ernst et al., 2003). One topic that needs to be resolved is whether these deficits are specific for clinical populations (Rogers, 2003). As it is well known that the personality trait of impulsivity is a major ingredient of several psychiatric (Moeller et al., 2001), personality (Rogers, 2003), and neurological disorders (Miller, 1992), including the aforementioned disorders. In the burgeoning neuropsychopharmacological literature on decision-making, the personality trait of impulsivity has been frequently suggested to be associated with weaknesses in decision-making. That is, impulsive persons display a decreased reflection on the consequences of their choice. These suggestions are in keeping with the notion that real-life decision-making involves choices that are based on expected but uncertain rewards and penalties, and that optimal choices are based on well-considered strategies. As such, it seems plausible to assume that impaired decision-making reflects a variety of impulse control problems (Morgan et al., 2006). Two studies addressed the relationship between impulsivity and decision-making in healthy subjects. A first study by Franken and Muris (2005) examined the link between decision-making, as indexed by the Iowa Gambling Task, and individual differences in functional and dysfunctional impulsivity and the impulsivity-related trait of reward sensitivity. The researchers found no relation between Iowa Gambling Task performance and dysfunctional impulsivity scores. However, reward sensitivity, a trait associated with impulsivity, was significantly related to the Iowa Gambling Task score. Surprisingly, this link was positive, suggesting that a higher level of an impulsivity-related trait is associated with better decisionmaking. In another study, also using the Iowa Gambling Task, Zermatten et al. (2005) found that decision-making was influenced by the impulsivity-related trait of ‘lack of premeditation’. Higher scores on ‘lack of premeditation‘ were positively linked to disadvantageous decision-making. Thus, while there is certain clinical evidence for the detrimental effects of impulsivity on decision-making, studies in non-clinical samples have yielded less convincing results. This may well have to do with the fact that these studies have only employed one specific test of
decision-making, the Iowa Gambling Task, so it remains to be seen whether different findings will emerge if other behavioral tests of decision-making are used. In this study, we focus on three decision-making tasks reflecting related but distinct aspects of decisionmaking: the Iowa Gambling Task (Bechara et al., 1994), which measures affective decision-making and includes a learning component, the Rogers Decision-Making Task (Rogers et al., 1999a), which also taps affective decision-making but does not include a learning component, and a probabilistic reversal-learning task (O'Doherty et al., 2001), which assesses affective decision-making with a learning component as well as the ability to alter choice behavior in response to fluctuations in reward contingency. In more detail, the Rogers Decision-Making Task taps explicit decisionmaking strategies without a reward-based learning element. This means that for each trial, all relevant information is presented to the participant; the participant does not need previously learned information in order to make a correct decision. The Iowa Gambling Task, on the other hand, measures decision-making strategies in which the participant has to learn to discriminate the advantageous choices from the disadvantageous choices. This means that the Iowa Gambling Task, unlike the Rogers Decision-Making Task, requires reward-based learning capacity (Fellows, 2004). The reversal-learning task (Rolls, 1999), measures not only reward-based learning but is also designed to index the adaptation of behavior according to changes in stimulus–reward contingencies (i.e., reversal learning; Clark et al., 2004), a capacity which is regarded as a basic requirement for normal social and emotional behavior (Cools et al., 2002; Kringelbach and Rolls, 2003; Rolls, 1996). Emotion-related visual reversal-learning tasks reflect the ability to alter choice behavior in response to fluctuations in reward contingency (Rahman et al., 1999). To summarize, all the tasks used in the present study reflect decision-making under risk. The Rogers Decision-Making Task only measures reward-based decision-making. The Iowa task adds a reward-based learning aspect to the decision-making process, whereas the probabilistic reversal task adds both a reward-based learning aspect and a reversal aspect2 that measures adaptive decision-making skills. Neuroimaging studies, which employed decisionmaking tasks without a learning element similar to the 2
It can be argued that the Iowa task also comprises a reversallearning aspect as the disadvantageous decks offer higher gains at the beginning of the task compared with decks C and D. However, this reversal aspect is minimal compared with the reversal-learning task.
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
Rogers Decision-Making Task, show that the orbitofrontal cortex is involved in this type of decision-making (Rogers et al., 1999b). Furthermore, neurological patients with prefrontal damage show significant deficits on all these types of decision-making tasks (Mavaddat et al., 2000; Rogers et al., 1999a). Neuroimaging studies employing decision-making tasks with a learning element, such as the Iowa Gambling Task, show medial prefrontal activity during this task (Fukui et al., 2005; Northoff et al., 2006). In addition, neurological patients with prefrontal damage show deficits on this task (for a review, see Bechara, 2004; Dunn et al., 2006). Several neuroimaging studies among healthy subjects show that the prefrontal cortex is also activated during a probabilistic reversal task (Cools et al., 2002; Kringelbach and Rolls, 2003; Remijnse et al., 2005). And again, deficits in reversal learning are also observed after prefrontal cortex damage (Hornak et al., 2004; Rahman et al., 1999; Rolls et al., 1994). To summarize, a wealth of data suggests that the orbitofrontal cortex is an essential structure for adequate decisionmaking (Clark et al., 2004; Ridderinkhof et al., 2004). In the present study, two main research questions were addressed. First, we examined whether healthy subjects with high-impulsivity scores would have impaired decision-making performance compared with subjects with low-impulsivity scores. We expected that subjects with high impulsivity would have an impaired overall performance on the Rogers Decision-Making Task, the Iowa Gambling Task, and the reversal-learning task. However, in line with the nature of impulsivity, it would be expected that subjects with high impulsivity would have shorter deliberation times on these decision-making tasks. Second, we explored which specific aspects of decision-making would be associated with impulsivity. 2. Methods 2.1. Participants and procedure A sample of 70 undergraduate psychology students (20% males) participated in the present study. The mean age of the sample was 20.8 years (S.D. = 2.9). The group was median-split into a high impulsiveness group (n = 30; high-impulsives) and a low impulsiveness group (n = 40; low-impulsives) based on their score (range = 0–17; median = 4) on the impulsiveness scale of the I7 questionnaire (Eysenck et al., 1985). The low-impulsives had a mean I7 impulsiveness score of 2.3 (S.D. = 1.4) and the mean score of the high-impulsives was 9.5 (S.D. = 3.5). After participants provided informed consent, they were asked to complete the questionnaires and then the decision-making tasks were administered in the following
157
order: reversal-learning task, Rogers Decision-Making Task, Iowa Gambling Task. The study was approved by the ethical committee of the Erasmus Medical Centre. 2.2. Instruments 2.2.1. I7 Impulsiveness Scale The Dutch version of the 19-item Impulsiveness Scale of the I7 questionnaire (Eysenck et al., 1985; Lijffijt et al., 2005) was used as the measure of impulsivity. In this questionnaire, impulsiveness is regarded as acting without first considering the possible consequences. The impulsiveness scale contains items such as ‘Do you often buy things on impulse?’, and ‘Are you an impulsive person?’ Respondents answered the items on a dichotomous (i.e. ‘yes’ and ‘no’) response scale. The I7 is a frequently employed questionnaire to assess the personality trait of impulsivity (see Lijffijt et al., 2005). The scale has good psychometric properties, with good reliability and validity. Cronbach's alpha of the Dutch version was reported to be 0.80 (Lijffijt et al., 2005); in this study, alpha was 0.72. 2.2.2. Positive and Negative Affect Scales In order to compare both impulsivity groups on current affect, the Positive and Negative Affect Scales (PANAS; Watson et al., 1988) were administered. The PANAS is a 20-item bidimensional mood inventory. The Positive Affect scale reflects the extent to which a person feels enthusiastic, active, and alert (Watson et al., 1988), whereas the Negative Affect scale is an index of subjective distress and unpleasurable engagement that subsumes a variety of aversive mood states, including anger, contempt, disgust, guilt, fear, and nervousness (Watson et al., 1988). Psychometric properties of the PANAS scales are adequate, with good reliability and validity (Boon and Peeters, 1999; Watson et al., 1988). Cronbach's alpha of the Dutch version was reported to be 0.89 and 0.86 for the PA and NA scales, respectively (Boon and Peeters, 1999). 2.2.3. Rogers Decision-Making Task The Rogers Decision-Making task was identical to the task as described by Rogers et al. (2003). However, the present version consisted of 40 trials randomly presented within two blocks. On each trial, participants had to choose one of the two presented gambles in order to collect as many points as possible. The gambles were represented in the form of a histogram. The height of this histogram indicated the probability of gaining, and the possible amounts of gains were indicated in green above the histogram. The possible amounts of losses
158
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
were indicated in red beneath the histogram (see Rogers et al., 2003). For each trial, one gamble acted as the control gamble, which consisted of a 50% probability of winning 10 points and a 50% probability of losing 10 points. The other simultaneously presented gamble was the “experimental gamble”. This trial type varied on three factors: the probability of winning, high (75%) vs. low (25%); the possible gains, large (80 points) vs. small (20 points); and the possible losses, large (80 points) vs. small (20 points). This procedure resulted in eight different gambling situations, and because the experimental gamble appeared randomly on the left or the right of the screen, this resulted in a total of 16 different gambling situations. Participants had to choose one of the two gambles by pressing the numbers ‘1’ or ‘2’ on a keyboard. Three indices of decision-making were derived from the Rogers Decision-Making Task. First, we took the overall proportion of choices of the experimental over the control gamble. Furthermore, following Rogers et al. (2003), we also included two types of “all or nothing” trials. In the “gains only” trial type, participants had to choose between a guaranteed win of 40 points or a 50% chance of winning 80 points. In the “losses only” trial type, participants had to choose between a guaranteed loss of 40 points or a 50% chance of losing 80 points. For the ‘gains only’ and ‘losses only’ trials, the dependent measure was the proportion of choices on which volunteers chose the uncertain outcome (experimental trial). Third, we also recorded the time needed to make a decision (deliberation time). After choosing, participants received feedback on the amount of points they won or lost and the total amount of points they had gathered so far. 2.2.4. Iowa Gambling Task In the present study we used the computerized version of the Iowa Gambling Task (Bechara et al., 1994), which consisted of 100 successive trials in which subjects were required to choose a card from one of four decks. The participants were instructed to try to gain as much money as possible by drawing cards from one of four decks (A, B, C, D). The decisions to choose from the decks are motivated by reward and punishment schedules inherent in the task. Two of the decks (A and B) are disadvantageous, producing immediate gains (large rewards), but these are accompanied by larger losses in the long run (larger punishments). The C and D decks are advantageous: here, gains are modest but more consistent and losses are smaller. A net score ((C + D) − (A + B)) was computed, with a higher score indicating that a subject is more often choosing advantageous decks.
2.2.5. Probabilistic reversal-learning task This task was similar to the one used by O'Doherty et al. (2001), except for the fact that we used a fixed number of trials (i.e. 100), and in our task a reversal took place after five correct choices. Subjects had to choose one of two stimuli (S+ and S−, which were easily discerned geometrical figures) that were presented simultaneously on the computer screen above each other. The S+ (advantageous) stimulus had the following properties: a reward–punishment ratio of 70:30, a reward range of 80–250 points, and a punishment range of 10–60 points. The S− (disadvantageous) stimulus had the following properties: a reward–punishment ratio of 40:60, a reward range of 30–60 points, and a punishment range of 250–600 points. Accordingly, persistent selection of S+ resulted in an overall gain, and persistent selection of S− resulted in an overall loss. When starting, subjects were unaware whether the upper or the lower geometrical figure was the S+ or S−. They had to learn by trial and error which one was the advantageous stimulus. Every time the participants chose the S+ stimulus five times in a row, there was an S+ and S− reversal (which the participant was unaware of). Refer to O'Doherty et al. (2001) for a detailed description. Scores derived from the probabilistic reversal task were the total number of reversal contingencies, the total number of errors, and the mean deliberation time of the choices (Swainson et al., 2000; Evers et al., 2005). Furthermore, a win-stay and lose-stay analysis was conducted (Budhani and Blair, 2005). To change choice after large gain can be regarded as disadvantageous choice behavior, and change after large loss could be regarded as advantageous choice behavior. This resulted in two additional scores: the proportion of choices that can be classified as “stay after large win” and the proportion of choices that can be classified as “stay after large loss”. 2.3. Data analysis Before analyzing the decision-making variables, we examined the characteristics of the two groups. Age, affect (PANAS score) and gender differences between the two groups were analyzed using Students t-test or a Fischer's Exact test. To answer the main research questions, the groups were compared on their behavioral decision-making performance by a multivariate analysis of variance (MANOVA) with a Bonferroni correction for multiple comparisons. Additionally, in order to explore the relationships among various decision-making measures, Pearson correlation coefficients of these measures were calculated. Further, to test which of the behavioral
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163 Table 1 Mean values (standard deviations) of low- and high-impulsives on several indices of decision-making behavior
IGT net score RDMT proportion of risky choices RDMT proportion of risky choice on “all or nothing trials” RDMT mean deliberation time (ms) PRT stay after large gain proportion PRT stay after large loss proportion PRT number of reversals PRT number of errors PRT mean deliberation time (ms)
Low-impulsives
High-impulsives
13.6 (30.7) 0.52 (0.15)
− 0.9 (23.5) a 0.58 (0.13)
0.43 (0.23)
0.50 (0.16)
3192.7 (1423.5)
2402.5 (1123.6) a
0.85 (0.14)
0.79 (0.13) a
0.27 (0.16)
0.36 (0.17) a
7.3 (3.6) 34.9 (9.3) 620.2 (330.1)
5.2 (3.0) a 39.6 (8.4) a 547.4 (190.2)
Notes. N = 70. IGT = Iowa Gambling Task, RDMT = Rogers DecisionMaking Task, PRT = probabilistic reversal task. a Significant differences in MANOVA.
measures was the most powerful predictor of self-reported impulsivity, a stepwise multiple regression analysis was employed including all behavioral measures as independent variables, and self-reported impulsivity as the dependent variable. 3. Results
159
Subsequent pairwise comparisons indicated that highimpulsives had lower scores on the Iowa Gambling Task compared to low-impulsives, F(1,69) = 4.4, P b 0.05, η2 = 0.06. In Fig. 1, the mean scores of both impulsivity groups over the five blocks on the Iowa Gambling Task are shown. On the Rogers Decision-Making Task, highimpulsives had shorter deliberation times, F(1,69) = 4.4, P b 0.05, η2 = 0.09. High-impulsives did not made overall more risky choices on the Rogers Decision-Making Task F(1,69) = 1.9, P = 0.09, η2 = 0.04. On the probabilistic reversal task, high-impulsives reached a lower number of reversals F(1,69) = 6.3, P b 0.05, η2 = 0.09, had a higher proportion of stay after large loss choices, F(1,69) = 5.0, P b 0.05, η2 = 0.07, and a lower proportion of stay after large gain choices, F(1,69) = 5.0, P b 0.05, η2 = 0.06, and had a higher number of errors, F(1,69) = 4.7, P b 0.05, η2 = 0.07. Regression analyses showed that the Iowa Gambling Task score was the most powerful predictor of self-reported impulsivity and explained 27% of the variance of self-reported impulsivity (standardized β = −0.27, partial r = −0.27, P b 0.05). The Pearson correlation coefficients among the three decision-making questionnaires are displayed in Table 2. Besides obvious correlations among sub-scores within tasks, we also observed a significant positive correlation between the net score of the Iowa Gambling Task and the number of reversals on the reversallearning task (r = 0.24, P b 0.05). Further, the deliberation times on the Rogers Decision-Making Task and
3.1. Group characteristics The groups of low-impulsives and high-impulsives were equal with regard to gender ratio (23% vs. 17% males; P = 0.76), mean age (20.7 vs. 21.0 years, respectively; P = 0.62), mean positive affect score (33.4 vs. 34.8, respectively; P = 0.22), and mean negative affect score (19.4 vs. 18.4, respectively; P = 0.55). 3.2. Decision-making Mean scores of both groups on the decision-making variables are displayed in Table 1. There was an overall effect of impulsivity on the behavioral measures of choice behavior, Wilks' Lambda = 0.85, F(10,57) = 2.1, P b 0.05, η 2 = 0.27, indicating that high-impulsives displayed generally impaired decision-making performance as compared to low-impulsives.3 3
We conducted an extreme group analysis with top and bottom 20% impulsivity scores. This analysis yielded a similar pattern of results on the individual tasks effect on decision-making. However, due to the loss of power, some of the differences dropped below the significance level.
Fig. 1. Mean net score (advantageous–disadvantages choices) of both groups on the Iowa Gambling Task over the five blocks.
160
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
Table 2 Correlations among the behavioral measures of decision-making
1. IGT net score 2. RDMT proportion of risky choices 3. RDMT proportion of risky choice on “all or nothing trials” 4. RDMT mean deliberation time 5. PRT stay after large gain proportion 6. PRT stay after large loss proportion 7. PRT number of reversals 8. PRT number of errors 9. PRT mean deliberation time
1
2
3
4
5
6
7
8
– − 0.21 − 0.02 0.17 0.13 − 0.15 0.24⁎ − 0.18 − 0.03
– 0.17 −0.17 −0.11 0.20 −0.14 0.20 −0.07
– − 0.02 − 0.01 0.07 0.00 − 0.06 − 0.13
– − 0.19 − 0.04 − 0.05 0.06 0.27⁎
– − 0.22 0.61⁎⁎ − 0.39⁎⁎ − 0.03
– − 0.33⁎⁎ 0.68⁎⁎ 0.10
– − 0.77⁎⁎ − 0.18
– 0.26⁎
Notes. N = 70. IGT = Iowa Gambling Task, RDMT = Rogers Decision Making Task, PRT = Probalistic Reversal Task. ⁎P b 0.05; ⁎⁎P b 0.01.
the Reversal Task were positively correlated (r = 0.27, P b 0.05). 4. Discussion The current study examined the relation between behavioral decision-making and the personality trait of impulsivity. It was found that high-impulsives displayed a general deficit in their decision-making abilities as compared to low-impulsives. That is, high-impulsives had sub-optimal performance on various decision-making indices. The effect size of the overall performance deficit was large (Cohen, 1988). More specifically, impairments were found on several measures: High-impulsives displayed a decreased performance in learning of reward and punishment associations in order to make appropriate decisions, and impaired adaptation of choice behavior according to changes in stimulus–reward contingencies. The results on simple, non-learning, components of reward- and punishment-based decision-making were less clear; these seemed relatively unaffected as compared with the tasks that require reward-based learning strategies. Above all, the results indicate that impulsivity is associated with a decreased ability to alter choice behavior in response to fluctuations in reward contingency. Interestingly, impulsives display lowered deliberation times on the Rogers Decision-Making Task, and it would be tempting to conclude that high-impulsives act before considering the consequences of their choice. However, the performance on Rogers' task is equal for both groups, suggesting that impulsives have a more efficient decision-making strategy on this specific task. The present study demonstrated that high-impulsives have lower scores on the Iowa Gambling Task net score compared with low-impulsives. In addition, the results of regression analyses showed also that the Iowa Gambling Task net score was the most powerful
predictor of self-reported impulsivity. The observed association between impulsivity and Iowa Gambling Task performance is in line with the study of Zermatten et al. (2005) in which the authors found a negative relation between the impulsivity-related trait of ‘lack premeditation’ and the Iowa Gambling Task score. However, this outcome is at variance with other studies among healthy populations (Franken and Muris, 2005). In contrast to the present study, in that study no correlation between decision-making, as measured by the Iowa Gambling Task and impulsivity could be observed. A fact that may explain this divergence is that in the Franken and Muris study a gambling task with a progressive schedule of increased delayed punishments was used (punishments increased progressively as subjects selected more cards from the disadvantageous decks, see Bechara et al., 2000b) resulting in a somewhat different task that may not be related to impulsivity. Further, in both studies different questionnaires were employed for measuring impulsivity (i.e., the Dickman Impulsivity Inventory vs. the I7). It is conceivable that different impulsivity questionnaires tap different behavioral aspects of impulsivity. In addition, the observed association between impulsivity and decision-making is also at variance with studies among psychiatric patient populations which did not observe a relation between impulsivity and behavioral decision-making (Bazanis et al., 2002; Jollant et al., 2005). Clearly, more research on which aspects of decision-making are related to various indices of impulsivity is needed. The present results contribute to the discussion about the inconsistency of the definition of impulsivity (Evenden, 1999a,b). Roughly, a distinction can be made between definitions stressing the inhibition of an activated response, as is measured using Go–Nogo tasks, and definitions stressing the integration of reward/punishment information in order to choose between two or more response options, as is measured using the current decision-
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
making tasks. Interestingly, in a study of Lijffijt et al. (2004), the authors found no difference between low- and high-impulsives on behavioral measures of inhibitory motor control using the same impulsivity measure as in the current study. This suggests that the I7, and similar impulsivity questionnaires tapping trait impulsivity, reflect problems with the integration of reward/punishment information and not the inhibition of motor control. Besides the distinction between response inhibition and integration of reward/punishment integration, other subclassifications of impulsivity have been suggested. For example, Barratt (1985) distinguishes three different aspects of impulsivity: cognitive impulsiveness which is involved in making quick cognitive decisions, motor impulsiveness which is involved in acting without thinking, and non-planning impulsiveness which involved a lack of “looking into the future” or planning. The presently used impulsiveness scale, the I7, reflects poor behavioral control and inability to delay gratification of which the latter aspect is obviously close to the behavioral decision-making instruments used in the present study. Using multifactorial impulsivity scales, it would be interesting to study which specific facets of impulsivity relate to specific neuropsychological mechanisms of decision-making. Further, these results show also that reversal-learning tasks tap a somewhat different aspect of impulsivity and may be a useful addition to the behavioral measures of impulsivity. Our additional correlational analyses among the behavioral tasks show that, although all measures tap decision-making, these do not measure equal aspects. Except for the small correlations between the Iowa Gambling Task net score and the number of reversals on the reversal-learning task, and the small correlation between the deliberation times on the Rogers Decision-Making Task and the Reversal Task, there were no significant correlations observed between the various decision-making tasks. This suggests that these measures indeed tap different aspects of choice behavior. Note that the correlation between the Iowa Gambling Task net score and the number of reversals on the reversal-learning task suggests that these tasks tap, at least partly, a similar concept. This supports Clark et al.'s (2004) notion that parallels exist between reversal learning and decision-making. The present results also add new evidence to the notion that impulsivity is associated with frontal lobe deficits (King et al., 2003; Miller, 1992; Spinella, 2004), more specifically, since both gambling tasks and reversal learning are associated with orbitofrontal functioning, suggesting a specific site of action. Probably, the association between impulsivity and executive functions is due to common neuroanatomical pathways. It is known
161
that both impulsivity and decision-making functions have common prefrontal neural substrates. The present results indicate that persons scoring high on the I7 impulsivity questionnaire display sub-optimal decisionmaking strategies, which are probably modulated by the ventral prefrontal cortex (Clark et al., 2004). It is interesting to consider these results in the light of the discussion of whether decision-making deficits that characterize drug addiction (Clark and Robbins, 2002) are the result of drug-induced changes in the prefrontal cortex (e.g. Schoenbaum et al., 2006) or are already present before the use of drugs. Although the present study is not specifically designed to resolve this question, this can only be done by prospective designs; it shows that decision-making deficits can be present in impulsive, normal individuals. Because it is known that drug addiction is associated with augmented levels of impulsivity (Dawe and Loxton, 2004) which results in drug use (Sher et al., 2000), the current results suggest that an individual with drug addiction may have preexisting impulsivity associated decision-making deficits. Again, it must be kept in mind that this issue can only be resolved in studies with longitudinal designs. The present study has several limitations. We employed a sample of young students with an overrepresentation of females. However, we do not think that this is critical for the conclusions since it is not plausible to assume that the relationship between behavioral decision-making and self-reported impulsivity is different for other healthy populations. A further limitation is that we have no data on psychiatric morbidity in this population. However, previous studies among student populations show that these numbers are typically very low. In addition, the reliability of the behavioral decisionmaking tasks employed in the present study is unknown. Clearly, future studies should address this important topic. Lastly, impulsivity status was based upon a median-split rather than extreme values or clinical cutoff points. However, it must be noted that clinical cutpoints for impulsivity have not yet been developed. To summarize, the results indicate that impulsivity is associated with a decreased ability to alter behavior in response to fluctuations in emotional significance to stimuli. The findings add further evidence to the notion that trait impulsivity is associated with decision-making, a function of the orbitofrontal cortex. Acknowledgements This work was supported by the Netherlands Organization for Scientific Research (NWO). The authors thank Dr. A. Bechara for providing the Iowa
162
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163
task, Manon de Long for writing a Dutch version of the Iowa task software, and Saskia van Dantzig for assisting with the learning reversal and decision-making tasks. References Barratt, E.S., 1985. Impulsiveness subtraits: arousal and information processing. In: Spence, J.T., Izard, C.E. (Eds.), Motivation, Emotion and Personality. Elsevier North Holland, pp. 137–146. Bazanis, E., Rogers, R.D., Dowson, J.H., Taylor, P., Meux, C., Staley, C., Nevinson-Andrews, D., Taylor, C., Robbins, T.W., Sahakian, B.J., 2002. Neurocognitive deficits in decision-making and planning of patients with DSM-III-R borderline personality disorder. Psychological Medicine 32, 1395–1405. Bechara, A., 2004. The role of emotion in decision-making: evidence from neurological patients with orbitofrontal damage. Brain and Cognition 55, 30–40. Bechara, A., 2005. Decision making, impulse control and loss of willpower to resist drugs: a neurocognitive perspective. Nature Neuroscience 8, 1458–1463. Bechara, A., Damasio, A.R., Damasio, H., Anderson, S.W., 1994. Insensitivity to future consequences following damage to human prefrontal cortex. Cognition 50, 7–15. Bechara, A., Damasio, H., Damasio, A.R., 2000a. Emotion, decision making and the orbitofrontal cortex. Cerebral Cortex 10, 295–307. Bechara, A., Tranel, D., Damasio, H., 2000b. Characterization of the decision-making deficit of patients with ventromedial prefrontal cortex lesions. Brain 123, 2189–2202. Boon, M.T., Peeters, F.P., 1999. Affectieve dimensies bij depressie en angst (Dimensions of affectivity in depression and anxiety). Tijdschrift voor Psychiatrie 41, 109–113. Brand, M., Labudda, K., Markowitsch, H.J., 2006. Neuropsychological correlates of decision-making in ambiguous and risky situations. Neural Networks 19, 1266–1276. Budhani, S., Blair, R.J.R., 2005. Response reversal and children with psychopathic tendencies: success is a function of salience of contingency change. Journal of Child Psychology and Psychiatry 46, 972–981. Cabanac, M., 1992. Pleasure: the common currency. Journal of Theoretical Biology 155, 173–200. Cavedini, P., Riboldi, G., D'Annucci, A., Belotti, P., Cisima, M., Bellodi, L., 2002a. Decision-making heterogeneity in obsessive–compulsive disorder: ventromedial prefrontal cortex function predicts different treatment outcomes. Neuropsychologia 40, 205–211. Cavedini, P., Riboldi, G., Keller, R., D'Annucci, A., Bellodi, L., 2002b. Frontal lobe dysfunction in pathological gambling patients. Biological Psychiatry 51, 334–341. Cavedini, P., Bassi, T., Ubbiali, A., Casolari, A., Giordani, S., Zorzi, C., Bellodi, L., 2004. Neuropsychological investigation of decision-making in anorexia nervosa. Psychiatry Research 127, 259–266. Clark, L., Robbins, T.W., 2002. Decision-making deficits in drug addiction. Trends in Cognitive Sciencess 6, 361–363. Clark, L., Cools, R., Robbins, T.W., 2004. The neuropsychology of ventral prefrontal cortex: decision-making and reversal learning. Brain and Cognition 55, 41–53. Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences, 2 ed. Erlbaum, Hillsdale, NJ. Cools, R., Clark, L., Owen, A.M., Robbins, T.W., 2002. Defining the neural mechanisms of probabilistic reversal learning using
event-related functional magnetic resonance imaging. Journal of Neuroscience 22, 4563–4567. Dawe, S., Loxton, N.J., 2004. The role of impulsivity in the development of substance use and eating disorders. Neuroscience and Biobehavioral Reviews 28, 343–351. Dunn, B.D., Dalgleish, T., Lawrence, A.D., 2006. The somatic marker hypothesis: a critical evaluation. Neuroscience and Biobehavioral Reviews 30, 239–271. Ernst, M., Grant, S.J., London, E.D., Contoreggi, C.S., Kimes, A.S., Spurgeon, L., 2003. Decision making in adolescents with behavior disorders and adults with substance abuse. American Journal of Psychiatry 160, 33–40. Evenden, J.L., 1999a. Impulsivity: a discussion of clinical and experimental findings. Journal of Psychopharmacology 13, 180–192. Evenden, J.L., 1999b. Varieties of impulsivity. Psychopharmacology 146, 348–361. Evers, E.A.T., Cools, R., Clark, L., van der Veen, F.M., Jolles, J., Sahakian, B.J., Robbins, T.W., 2005. Serotonergic modulation of prefrontal cortex during negative feedback in probabilistic reversal learning. Neuropsychopharmacology 30, 1138–1147. Eysenck, S.B.G., Pearson, P.R., Easting, G., Allsopp, J.F., 1985. Age norms for impulsiveness, venturesomeness and empathy in adults. Personality and Individual Differences 6, 613–619. Fellows, L.K., 2004. The cognitive neuroscience of human decision making: a review and conceptual framework. Behavioral and Cognitive Neuroscience Reviews 3, 159–172. Franken, I.H.A., Muris, P., 2005. Individual differences in decisionmaking. Personality and Individual Differences 39, 991–998. Fukui, H., Murai, T., Fukuyama, H., Hayashi, T., Hanakawa, T., 2005. Functional activity related to risk anticipation during performance of the Iowa Gambling Task. NeuroImage 24, 159–253. Hester, R., Garavan, H., 2004. Executive dysfunction in cocaine addiction: evidence for discordant frontal, cingulate, and cerebellar activity. Journal of Neuroscience 24, 11017–11022. Hornak, J., O'Doherty, J., Bramham, J., Rolls, E.T., Morris, R.G., Bullock, P.R., Polkey, C.E., 2004. Reward-related reversal learning after surgical excisions in orbito-frontal or dorsolateral prefrontal cortex in humans. Journal of Cognitive Neuroscience 16, 463–478. Jollant, F., Bellivier, F., Leboyer, M., Astruc, B., Torres, S., Verdier, R., Castelnau, D., Malafosse, A., Courtet, P., 2005. Impaired decision making in suicide attempters. American Journal of Psychiatry 162, 304–310. King, J.A., Tenney, J., Rossi, V., Colamussi, L., Burdick, S., 2003. Neural substrates underlying impulsivity. Annals of the New York Academy of Sciences 1008, 160–169. Kringelbach, M.L., Rolls, E.T., 2003. Neural correlates of rapid reversal learning in a simple model of human social interaction. NeuroImage 20, 1371–1383. Lijffijt, M., Bekker, E.M., Quik, E.H., Bakker, J., Kenemans, J.L., Verbaten, M.N., 2004. Differences between low and high trait impulsivity are not associated with differences in inhibitory motor control. Journal of Attentional Disorders 8, 25–32. Lijffijt, M., Kenemans, J.L., Caci, H., 2005. Validation of the Dutch translation of the I7 questionnaire. Personality and Individual Differences 38, 1123–1133. Maia, T.V., McClelland, J.L., 2004. A reexamination of the evidence for the somatic marker hypothesis: what participants really know in the Iowa gambling task. Proceedings of the National Academy of Sciences of the United States of America 101, 16075–16080. Mavaddat, N., Kirkpatrick, P.J., Rogers, R.D., Sahakian, B.J., 2000. Deficits in decision-making in patients with aneurysms of the anterior communicating artery. Brain 123, 2109–2117.
I.H.A. Franken et al. / Psychiatry Research 158 (2008) 155–163 Miller, L.A., 1992. Impulsivity, risk-taking, and the ability to synthesize fragmented information after frontal lobectomy. Neuropsychologia 30, 69–79. Moeller, F.G., Barrat, E.S., Dougherthy, D.M., Schmitz, J.M., Swann, A.C., 2001. Psychiatric aspects of impulsivity. American Journal of Psychiatry 158, 1783–1793. Morgan, M.J., Impallomeni, L.C., Pirona, A., Rogers, R.D., 2006. Elevated impulsivity and impaired decision-making in abstinent ecstasy (MDMA) users compared to polydrug and drug-naive controls. Neuropsychopharmacology 31, 1562–1573. Northoff, G., Grimm, S., Boeker, H., Schmidt, C., Bermpohl, F., Heinzel, A., Hell, D., Boesiger, P., 2006. Affective judgment and beneficial decision making: Ventromedial prefrontal activity correlates with performance in the Iowa Gambling Task. Human Brain Mapping 27, 572–587. O'Doherty, J., Kringelbach, M.L., Rolls, E.T., Hornak, J., Andrews, C., 2001. Abstract reward and punishment representations in the human orbitofrontal cortex. Nature Neuroscience 4, 95–102. Rahman, S., Sahakian, B.J., Hodges, J.R., Rogers, R.D., Robbins, T.W., 1999. Specific cognitive deficits in mild frontal variant frontotemporal dementia. Brain 122, 1469–1493. Rahman, S., Sahakian, B., Cardinal, R., Rogers, R., Robbins, T., 2001. Decision making and neuropsychiatry. Trends in Cognitive Sciences 5, 271–277. Remijnse, P.L., Nielen, M.M.A., Uylings, H.B.M., Veltman, D.J., 2005. Neural correlates of a reversal learning task with an affectively neutral baseline: an event-related fMRI study. NeuroImage 26, 609–618. Ridderinkhof, K.R., van den Wildenberg, W.P., Segalowitz, S.J., Carter, C.S., 2004. Neurocognitive mechanisms of cognitive control: the role of prefrontal cortex in action selection, response inhibition, performance monitoring, and reward-based learning. Brain and Cognition 56, 129–140. Rogers, R.D., 2003. Neuropsychological investigations of the impulsive personality disorders. Psychological Medicine 33, 1335–1340. Rogers, R.D., Everitt, B.J., Baldacchino, A., Blackshaw, A.J., Swainson, R., Wynne, K., Baker, N.B., Hunter, J., Carthy, T., Booker, E., London, M., Deakin, J.F., Sahakian, B.J., Robbins, T.W., 1999a. Dissociable deficits in the decision-making cognition of chronic amphetamine abusers, opiate abusers, patients with focal damage to prefrontal cortex, and tryptophan-depleted normal volunteers: evidence for monoaminergic mechanisms. Neuropsychopharmacology 20, 322–339.
163
Rogers, R.D., Owen, A.M., Middleton, H.C., Williams, E.J., Pickard, J.D., Sahakian, B.J., Robbins, T.W., 1999b. Choosing between small, likely rewards and large, unlikely rewards activates inferior and orbital prefrontal cortex. Journal of Neuroscience 19, 9029–9038. Rogers, R.D., Tunbridge, E.M., Bhagwagar, Z., Drevets, W.C., Sahakian, B.J., Carter, C.S., 2003. Tryptophan depletion alters the decisionmaking of healthy volunteers through altered processing of reward cues. Neuropsychopharmacology 28, 153–162. Rolls, E.T., 1996. The orbitofrontal cortex. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 351, 1433–1443. Rolls, E.T., 1999. The Brain and Emotion. Oxford University Press, New York. Rolls, E.T., Hornak, J., Wade, D., McGrath, J., 1994. Emotion-related learning in patients with social and emotional changes associated with frontal lobe damage. Journal of Neurology, Neurosurgery and Psychiatry 57, 1518–1524. Schoenbaum, G., Roesch, M.R., Stalnaker, T.A., 2006. Orbitofrontal cortex, decision-making and drug addiction. Trends in Neurosciences 29, 116–124. Sher, K.J., Bartholow, B.D., Wood, M.D., 2000. Personality and substance use disorders: a prospective study. Journal of Consulting and Clinical Psychology 68, 818–829. Spinella, M., 2004. Neurobehavioral correlates of impulsivity: evidence of prefrontal involvement. International Journal of Neuroscience 114, 95–104. Swainson, R., Rogers, R.D., Sahakian, B.J., Summers, B.A., Polkey, C.E., Robbins, T.W., 2000. Probabilistic learning and reversal deficits in patients with Parkinson's disease or frontal or temporal lobe lesions: possible adverse effects of dopaminergic medication. Neuropsychologia 38, 596–612. Toplak, M.E., Jain, U., Tannock, R., 2005. Executive and motivational processes in adolescents with Attention-Deficit-Hyperactivity Disorder (ADHD). Behavioral and Brain Functions 1, 8. Watson, D., Clark, L.A., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS Scales. Journal of Personality and Social Psychology 54, 1063–1070. Zermatten, A.M., Van der Linden, M.P., d'Acremont, M.M., Jermann, F.M., Bechara, A.P., 2005. Impulsivity and decision making. Journal of Nervous and Mental Disease 193, 647–650.