Hierarchical rule switching in prefrontal cortex Wako Yoshida a,b, ⁎, Hidefumi Funakoshi a , Shin Ishii a,c a Graduate School of Information Science, Nara Institute of Science and Technology, Japan b Wellcome Trust Centre for Neuroimaging, University College London, UK c Graduate School of Informatics, Kyoto University, Japan abstract article info Article history: Received 3 April 2009 Revised 2 December 2009 Accepted 3 December 2009 Available online 11 December 2009 Most real-world decision-making problems involve consideration of numerous possible actions, and it is often impossible to evaluate all of them before settling on preferred strategy. In such situations, humans might explore actions more efficiently by searching only the most likely subspace of the whole action space. To study how the brain solves such action selection problems, we designed a Multi Feature Sorting Task in which the task rules defining an optimal action have a hierarchical structure and studied concurrent brain activity using it. The task consisted of two kinds of rule switches: a higher-order switch to search for a rule across different subspaces and a lower-order switch to change a rule within the same subspace. The results revealed that the left dorsolateral prefrontal cortex (DLPFC) was more active in the higher-order switching, and the right fronto-polar cortex (FPC) was significantly activated with the lower-order switching. We discuss a possible functional model in the prefrontal cortex where the left DLPFC encodes the hierarchical organization of behaviours and the right FPC maintains and updates multiple behavioural. This interpretation is highly consistent with the previous findings and current theories of hierarchical organization in the prefrontal functional network. © 2009 Elsevier Inc. All rights reserved. Introduction In the real environment around us, there may be numerous possible behaviours available to us at any point, and it may be impossible to immediately make an appropriate decision by evaluat- ing all of them. To adapt to a dynamic environment, moreover, humans must seek candidate actions efficiently and select the best one within a limited time. Recent studies in the theoretical field suggest that adopting a hierarchical structure of candidate actions seem to exhibit superior performance in action selection and learning (Barto and Mahadevan, 2003; Wiering and Schmidhuber, 1998). In psychology, hierarchy has played a pivotal role in understanding organized, goal-directed behaviour, from early pioneering work (Milner, 1963; Newell and Simon, 1963) through to recent studies (e.g., Botvinick and Plaut, 2004; Schneider and Logan, 2006). From a neuroanatomical point of view, the organization of cortex is strongly hierarchical, and behavioural hierarchies in decision making has been proposed to be implemented within the prefrontal cortex (Badre, 2008; Botvinick, 2008; Koechlin et al., 2003; Wood and Grafman, 2003). Recently, theoretical studies have suggested that a hierarchical method derived from computational theory (i.e., hierarchical rein- forcement learning) might be implemented by human prefrontal cortex (Botvinick, 2008; Hazy et al., 2007); however, it remains unclear whether and how such a hierarchical mechanism might operate in the brain. Humans often select actions according to some sort of decision rule; he/she should consider switching between rules in response to environmental changes. The Wisconsin Card Sorting Task (WCST) (Grant and Berg, 1948) is one of the best-known tasks for studying such a rule switching process. In the WCST, the subject is required to discover a hidden correct rule from many possible rules using true/ false feedback given correspondingly to the selected rule. Since the correct rule often changes without notice, the subject should try a new rule if he/she receives a false feedback. Many imaging and lesion studies have shown that prefrontal cortex is closely involved in solving WCST (Berman et al., 1995; Goldberg et al., 1998; Hampshire and Owen, 2006; Konishi et al., 1998; Konishi et al., 2002; Lie et al., 2006; Monchi et al., 2001; Wang et al., 2001). However, although all of these tasks involved rule switch processes, different regions of the prefrontal cortex were reported as being engaged in rule switch functions and functional segregation of these regions has yet to be clarified. To identify brain regions involved in a hierarchical rule searching mechanism, in this study, we designed a Multi Feature Sorting Task whereby behavioural decision rules have a hierarchical structure. The rules in our task could be categorized into two levels: high level meta- rules and low level standard rules. Accordingly, this produces two kinds of rule switches: a higher-order meta-rule switch which changes a meta-rule to search for a rule belonging to the other meta-rule class and a lower-order rule switch to change rules within NeuroImage 50 (2010) 314–322 ⁎ Corresponding author. The Wellcome Trust Centre for Neuroimaging, Institute of Neurology, UCL, 12 Queen Square, London, WC1N 3BG, UK. Fax: +44 207 813 1445. E-mail address: w.yoshida@fil.ion.ucl.ac.uk (W. Yoshida). 1053-8119/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2009.12.017 Contents lists available at ScienceDirect NeuroImage journal homepage: www.elsevier.com/locate/ynimg