Advertisement

Systematic Overestimation of Reflection Impulsivity in the Information Sampling Task

      Impulsivity, a psychological construct comprising both motor and cognitive factors, is thought to underlie important interindividual differences in health and disease (
      • Evenden J.
      Impulsivity: A discussion of clinical and experimental findings.
      ). In particular, reflection impulsivity, which refers to the tendency to gather and evaluate information before decision making (
      • Kagan J.
      Reflection-impulsivity: The generality and dynamics of conceptual tempo.
      ), has been implicated in many psychiatric and neurological disorders (
      • Clark L.
      • Robbins T.W.
      • Ersche K.D.
      • Sahakian B.J.
      Reflection impulsivity in current and former substance users.
      ,
      • Tavares J.V.T.
      • Clark L.
      • Cannon D.M.
      • Erickson K.
      • Drevets W.C.
      • Sahakian B.J.
      Distinct profiles of neurocognitive function in unmedicated unipolar depression and bipolar II depression.
      ,
      • Zamarian L.
      • Benke T.
      • Brand M.
      • Djamshidian A.
      • Delazer M.
      Impaired information sampling in mild dementia of Alzheimer’s type but not in healthy aging.
      ). One of the standard tasks for measuring reflection impulsivity in healthy and clinical populations is the Information Sampling Task (IST), designed by Clark et al. (
      • Clark L.
      • Robbins T.W.
      • Ersche K.D.
      • Sahakian B.J.
      Reflection impulsivity in current and former substance users.
      ) and included in the widely used Cambridge Neuropsychological Test Automated Battery (CANTAB) (
      • Fray P.J.
      • Robbins T.W.
      • Sahakian B.J.
      Neuorpsychiatyric applications of CANTAB.
      ). In this CANTAB version of the IST, participants sample a variable amount of information about an uncertain outcome before making a decision. The amount of information sampled before the decision gives a measure of participants’ reflection impulsivity. In this correspondence, we show that the calculation of the IST’s main outcome measure, P(correct), is based on incorrect statistical inference, resulting in systematic overestimation of participants’ reflection impulsivity and potentially inflated type II error rates. This might affect the results of numerous recent psychopharmacological, neuropsychological, and psychiatric publications that have used the IST (
      • Tavares J.V.T.
      • Clark L.
      • Cannon D.M.
      • Erickson K.
      • Drevets W.C.
      • Sahakian B.J.
      Distinct profiles of neurocognitive function in unmedicated unipolar depression and bipolar II depression.
      ,
      • Zamarian L.
      • Benke T.
      • Brand M.
      • Djamshidian A.
      • Delazer M.
      Impaired information sampling in mild dementia of Alzheimer’s type but not in healthy aging.
      ,
      • DeVito E.E.
      • Blackwell A.D.
      • Clark L.
      • Kent L.
      • Dezsery A.M.
      • Turner D.C.
      • et al.
      Methylphenidate improves response inhibition but not reflection–impulsivity in children with attention deficit hyperactivity disorder (ADHD).
      ).
      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to Biological Psychiatry
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect

      References

        • Evenden J.
        Impulsivity: A discussion of clinical and experimental findings.
        J Psychopharmacol. 1999; 13: 180-192
        • Kagan J.
        Reflection-impulsivity: The generality and dynamics of conceptual tempo.
        J Abnormal Psychol. 1966; 71: 17-24
        • Clark L.
        • Robbins T.W.
        • Ersche K.D.
        • Sahakian B.J.
        Reflection impulsivity in current and former substance users.
        Biol Psychiatry. 2006; 60: 515-522
        • Tavares J.V.T.
        • Clark L.
        • Cannon D.M.
        • Erickson K.
        • Drevets W.C.
        • Sahakian B.J.
        Distinct profiles of neurocognitive function in unmedicated unipolar depression and bipolar II depression.
        Biol Psychiatry. 2007; 62: 917-924
        • Zamarian L.
        • Benke T.
        • Brand M.
        • Djamshidian A.
        • Delazer M.
        Impaired information sampling in mild dementia of Alzheimer’s type but not in healthy aging.
        Neuropsychology. 2015; 29: 353-367
        • Fray P.J.
        • Robbins T.W.
        • Sahakian B.J.
        Neuorpsychiatyric applications of CANTAB.
        Int J Geriatr Psychiatry. 1996; 11: 329-336
        • DeVito E.E.
        • Blackwell A.D.
        • Clark L.
        • Kent L.
        • Dezsery A.M.
        • Turner D.C.
        • et al.
        Methylphenidate improves response inhibition but not reflection–impulsivity in children with attention deficit hyperactivity disorder (ADHD).
        Psychopharmacology. 2009; 202: 531-539
        • Link W.A.
        • Barker R.J.
        Bayesian Inference: With Ecological Applications.
        Academic Press, Boston2009
        • Kras M.
        Impulsive traits and impulsive behavior associated with opioid dependence: Relationship to abstinence and heroin use.
        Victoria, Austalia, Doctoral Thesis, Monash University2015

      Linked Article

      • Reply to: Systematic Overestimation of Reflection Impulsivity in the Information Sampling Task
        Biological PsychiatryVol. 82Issue 4
        • Preview
          In our first publication on the Information Sampling Task (IST), published 10 years ago, we provided a simple equation, readily calculated in Excel (Microsoft, Redmond, WA), for estimating the level of uncertainty that a participant is willing to tolerate at the point when they commit to their decision (1). As noted by Bennett et al. (2) and others (3,4), estimating this level of uncertainty [termed P(correct)] is nontrivial, and is a problem well-suited to a Bayesian approach. Their new calculation incorporates valid information that is present in the ratio of colors in boxes that have already been opened.
        • Full-Text
        • PDF
      • Systematic Overestimation of Reflection Impulsivity in the Information Sampling Task: Age Dependency in Children
        Biological PsychiatryVol. 83Issue 2
        • Preview
          We read with great interest the letter recently published in Biological Psychiatry regarding the systematic overestimation of reflection impulsivity in the Information Sampling Task (IST) (1). The IST is a task in which the participants gather a variable amount of information prior to making a decision about an uncertain outcome (2). One of the main outcomes in this task is the P(correct), which estimates the level of uncertainty tolerated by the participant at the point of decision. Bennett et al.
        • Full-Text
        • PDF
      • The Choice of Prior in Bayesian Modeling of the Information Sampling Task
        Biological PsychiatryVol. 83Issue 12
        • Preview
          The Information Sampling Task (IST) was introduced in Clark et al. (1) as a measure of reflection impulsivity. Performance on the IST is evaluated through the average number of boxes opened for each trial of the test, whether the chosen color was in majority or not, and “the probability of the subject being correct at the point of decision [P(correct)]” (1). P(correct) is proposed calculated as a per-trial probability:
        • Full-Text
        • PDF