Validation of a child-friendly version of the monetary incentive delay task

doi:10.1093/scan/nss057 SCAN (2013) 8, 720 ^726 Validation of a child-friendly version of the monetary incentive delay task Sarah M. Helfinstein,1 Michael L. Kirwan,2 Brenda E. Benson,3 Michael G. Hardin,4 Daniel S. Pine,3 Monique Ernst,3 and Nathan A. Fox5 1 Imaging Research Center, Department of Psychology, and Department of Neurobiology, University of Texas at Austin, Austin, TX 78712, 2Center on the Developing Child, Harvard University, Cambridge, MA, 02138, 3National Institute of Mental Health, Bethesda, MD, 20814, 4Department of Psychology, Stanford University, Palo Alto, CA, 94304 and 5Department of Human Development, University of Maryland, College Park, MD, 20742, USA Keywords: Reward; childhood; fMRI INTRODUCTION Over the last decade, considerable research has examined the neural correlates of reward processing in humans (Knutson et al., 2001b, Kirsch et al., 2003; Ernst et al., 2004; Delgado et al., 2005, Dreher et al., 2009), particularly how the brain responds to cues indicating the potential to receive a reward (Knutson et al., 2001a; Hommer et al., 2003; Ernst et al., 2004; Scheres et al., 2007). Research has found that individual differences in neural response while anticipating a potential reward relate to temperament [e.g. behavioral inhibition, (Guyer et al., 2006) and psychopathology including depression (Smoski et al., 2009), anxiety (Guyer et al., 2011) and ADHD (Ernst et al., 2003; Scheres et al., 2007)]. There is also evidence of developmental change in these responses: several studies have shown that the neural correlates of reward change from childhood to adolescence to adulthood (Bjork et al., 2004; Ernst et al., 2005; Galvan et al., 2006; Bjork et al., 2010, Cohen et al., 2010; Van Leijenhorst et al., 2010a,b). The ‘Monetary Incentive Delay’ (MID) task (Knutson et al., 2000) has been used in a large number of brain-imaging studies (Knutson and Cooper, 2005). This task extends work from Schultz and colleagues implicating the striatum in the neural response to anticipation of reward delivery. The paradigm is used commonly in imaging studies examining adults and mid-to-late adolescents. In the task, subjects see a cue that indicates the amount of monetary reward or loss at stake on a given trial, followed by a target to which they must make a speeded response, followed, finally, by feedback indicating whether or not they received the reward or lost money. Several studies using the MID task (Knutson et al., 2000, 2001a, b, 2003) have found a cluster of brain regions that consistently show increased activation to cues Received 21 December 2011; Accepted 30 April 2012 Advance Access publication 7 May 2012 The authors thank Emily Barrios for her assistance in data collection, and Alan Hoofring for his artistic contributions in creating the stimuli used in the task. This work was supported by the National Institute of Mental Health-Intramural Research Program, and by a grant from the National Institutes of Health to NAF [MH078105]. Correspondence should be addressed to Sarah M. Helfinstein, 3925-B W Braker Ln, Austin, TX 78759, USA. E-mail: indicating the potential for receipt of larger incentives. This cluster includes striatum, thalamus, medial prefrontal cortex, posterior cingulate cortex and insula. The MID task is not designed for use with young children. The incentive cues are abstract symbolic representations that can be difficult for young children to understand, and, like most psychological experimental paradigms, it is not built to engage the interest of youngsters. In order to study the developmental trajectory of reward processing from childhood through adolescence and adulthood, it would be helpful to have a version of the MID task that can be used effectively with children. With this goal, we designed a child-friendly version that paralleled the MID task structure for reward trials, but provided a coherent story line and the engaging graphics that would keep young children invested in the task. ~ To test the validity of this taskthe pinata taskwe examined both behavioral performance on the task and neural activation in the scanner with children between the ages of 8 and 14 years. For the behavioral study, we predicted that subjects would show a relation between incentive type and both reaction time and hit rate, with higher incentives producing faster reaction times and higher hit rates. For the imaging study, we predicted that a parametric analysis would demonstrate a relation between anticipation of cue incentive magnitude and activation. Specifically, we expected a similar pattern of regional engagement as seen in adults studied with the MID task: activation of the striatum, thalamus, medial prefrontal cortex and insula to anticipation of larger incentives. METHODS Participants Study #1Behavior Participants were 12 8-year-old children (five female), 14 10-year-old children (eight female), 15 12-year-old children (nine female) and 12 14-year-old adolescents (three female). All subjects were free from serious neurological or medical problems and had never received long-term medication. Two subjects (one 8-year old and one 12-year ß The Author (2012). Published by Oxford University Press. For Permissions, please email: The monetary incentive delay (MID) task (Knutson, 2000) is an imaging paradigm used to measure neural activity of incentive receipt anticipation. The task reliably elicits striatal activation and is commonly used with both adult and adolescent populations, but is not designed for use with children. In the ~ current article, we present data on the newly designed pinata task a child-friendly analog of the MID task. We demonstrate the task can be used successfully in children to study the neural correlates of anticipatory incentive processing. Results from a behavioral study and a neuroimaging study are reported. In Study #1, a sample of 8- to 14-year-old children demonstrates expected behavioral effects: subjects responded most quickly and most accurately on trials with greater potential rewards; older children displayed faster reaction times than younger. In Study #2, 8- to 12-year-old children showed neural activation patterns consistent with those seen in adults in the MID task: activation was modulated by cue incentive value in ~ reward-processing regions, including the striatum, thalamus, mesial prefrontal cortex and insula. Study results suggest that the pinata task is a valid analog of the MID task, and can be used to assess neural correlates of reward processing in children as young as 8–9 years of age. Monetary incentive delay task for kids SCAN (2013) old) failed to complete the entire task, and a third (a 14-year old) had reaction times >2 s.d. outside the mean for his age group; these subjects were excluded from analysis. Participants were recruited both via age-targeted mailing lists and at university-based summer camps. All participants provided informed assent, and participants’ guardia (...truncated)