Elsevier

NeuroImage

Volume 24, Issue 1, 1 January 2005, Pages 253-259
NeuroImage

Rapid communication
Functional activity related to risk anticipation during performance of the Iowa gambling task

https://doi.org/10.1016/j.neuroimage.2004.08.028Get rights and content

Risk anticipation is an important cognitive/emotional component of decision making. The Iowa Gambling Task [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], which is the most widely used “risk-anticipation task” in clinical studies, has been demonstrated to be sensitive to lesions involving the ventromedial prefrontal cortex or amygdala. However, the critical neural circuitry involved in this complex task has not yet been fully clarified even in healthy subjects. Using a 3-T scanner, we performed an event-related functional magnetic resonance imaging study in 14 healthy subjects performing the task. The statistical parametric mapping showed that the risk anticipation component (risky decisions minus safe decisions) exclusively activated the medial frontal gyrus. Furthermore, we found a significant interindividual correlation between the task performance and the magnitude of brain activity during risky decisions. These results indicate that the Iowa Gambling Task does recruit the neural circuitry that is critical in decision making under uncertainty, particularly when subjects perceive the risk of their decision.

Introduction

Making advantageous decisions by weighing differently on their risks and benefits based upon previous experiences is an important ability for survival; however, this process would rely not only on cognitive but also on emotional substrates. The Iowa Gambling Task (IGT) was developed to simulate real-life financial decisions (Bechara et al., 1994). In contrast to other tasks in which all necessary information is available for making each decision (e.g., Elliott et al., 1999, Paulus et al., 2001, Rogers et al., 1999), the IGT is based on a long exploratory learning process to evaluate long-term risk anticipation in decision making. Moreover, IGT performance is strongly influenced by emotional factors related to rewards and penalties. As a result of these unique properties, the IGT is considered invaluable in investigating various neuropsychiatric conditions (Bechara and Damasio, 2002, Bechara et al., 2001, Best et al., 2002, Blair et al., 2001, Cavedini et al., 2001, Cavedini et al., 2002, Grant et al., 2000, Mazas et al., 2001, Mintzer and Stitzer, 2002, Mitchell et al., 2002, Petry, 2001, van Honk et al., 2002).

Despite the above mentioned properties, the neural underpinning of the IGT has not yet been fully clarified. Poor performance of this task has been associated with lesions involving the ventromedial prefrontal cortex (Bechara et al., 1994, Bechara et al., 1996, Bechara et al., 1999) or amygdala (Bechara et al., 1999, Bechara et al., 2003). Recent lesion studies suggest the involvement of more extensive structures including the dorsolateral prefrontal cortex for the IGT (Clark et al., 2003, Manes et al., 2002). Evidence from neuroimaging studies is limited. Using positron emission tomography (PET), Ernst et al. (2002) demonstrated widespread activations including the orbitofrontal, dorsolateral, prefrontal, and anterior cingulate cortices in healthy subjects. However, PET evaluates the summation of brain activity over a period of time (typically >1 min), during which the IGT requires multiple cognitive components (e.g., working memory) and should elicit complex emotional reactions. Hence, the most important component of the IGT, namely, anticipation of the long-term consequences associated with risky decision making (Bechara et al., 1996), may not be appropriately highlighted by PET.

In the present experiment, an event-related functional magnetic resonance imaging (fMRI) was used to assess neural responses to risk anticipation during the Iowa Gambling Task. We hypothesized that (1) specific regions of the prefrontal cortex would show increased activation during risky decision making, and (2) neural activity in these regions would correlate with the task performance.

Section snippets

Subjects

Nineteen volunteers (16 men and 3 women) originally participated in this study. However, five subjects could not identify the advantageous decks after task completion. This observation is different from previous literature reporting that healthy subjects should be able to identify the advantageous decks correctly in a natural environment (Bechara et al., 1996, Bechara et al., 1997). This discrepancy probably resulted from the study-specific factors such as the task performance under the MRI

Task performance

One subject failed to press a button twice during the 3.5-s selection periods, and these two trials were excluded from the analyses. Reaction times did not differ significantly between the risky decisions (1.16 ± 0.38 s) and safe decisions (1.13 ± 0.40 s). According to the previous report (Bechara et al., 1994), we subdivided the 100 trials into five blocks consisting of 20 trials each. For each block, we counted the number of disadvantageous selections (Deck A and B) and advantageous

Discussion

In this study, we used an event-related fMRI to dissociate neural activity related to decision making during the IGT performance. We found new evidence showing the significance of the medial prefrontal cortex for the IGT performance: (1) only the medial prefrontal cortex showed significantly differential activations between the risky and safe decisions; (2) the net scores of the task was significantly correlated with the magnitude of the medial prefrontal activity during risky decisions.

The IGT

Acknowledgments

We thank Toshihiko Aso, Makiko Yamada, Mitsunari Abe, and Ayako Kato for their support in the fMRI data acquisition; Shin-ichi Urayama and Chihiro Namiki for the help with the data processing; and Robert Turner for the helpful comments on the methodological issues in the study design. This study was in part supported by Grant-in-Aid for Young Scientists (B) (15700257) to T.H. from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.

References (44)

  • M. Ernst et al.

    Decision-making in a risk-taking task: a PET study

    Neuropsychopharmacology

    (2002)
  • S. Grant et al.

    Drug abusers show impaired performance in a laboratory test of decision-making

    Neuropsychologia

    (2000)
  • R.D. Lane et al.

    Neuroanatomical correlates of pleasant and unpleasant emotion

    Neuropsychologia

    (1997)
  • M.Z. Mintzer et al.

    Cognitive impairment in methadone maintenance patients

    Drug Alcohol Depend.

    (2002)
  • D.G. Mitchell et al.

    Risky decisions and response reversal: is there evidence of orbitofrontal cortex dysfunction in psychopathic individuals?

    Neuropsychologia

    (2002)
  • D.G. Norris et al.

    An investigation of the value of spin-echo-based fMRI using a Stroop color-word matching task and EPI at 3 T

    NeuroImage

    (2002)
  • N.M. Petry

    Substance abuse, pathological gambling, and impulsiveness

    Drug Alcohol Depend.

    (2001)
  • M.P. Paulus et al.

    Prefrontal, parietal, and temporal cortex networks underlie decision-making in the presence of uncertainty

    NeuroImage

    (2001)
  • A. Bechara et al.

    Failure to respond automatically to anticipated future outcomes following damage to preforontal cortex

    Cereb. Cortex

    (1996)
  • A. Bechara et al.

    Deciding advantageously before knowing the advantageous strategy

    Science

    (1997)
  • A. Bechara et al.

    Dissociation of working memory from decision making within the human prefrontal cortex

    J. Neurosci.

    (1998)
  • A. Bechara et al.

    Different contributions of the human amygdala and ventromedial prefrontal cortex to decision-making

    J. Neurosci.

    (1999)
  • Cited by (198)

    • Common and distinct brain activity associated with risky and ambiguous decision-making

      2020, Drug and Alcohol Dependence
      Citation Excerpt :

      Among the regions demonstrating convergent activity, the ACC has been associated with a wide range of risky-DM tasks. Depending on the type of risky-DM task at hand, ACC involvement has been linked with perceptions and anticipation of risk, reward prediction, loss avoidance, and learning the consequences of risky behaviors (Alexander and Brown, 2011; Fukui et al., 2005; Fukunaga et al., 2012; Magno et al., 2006). For example, ACC activity increases with the increasing probability of an “explosion” during the BART (i.e., as risk increases) (Fukunaga et al., 2012).

    View all citing articles on Scopus
    View full text