Beyond common resources: the cortical basis for resolving task interference Robert Hester, a Kevin Murphy, a and Hugh Garavan a,b, * a Department of Psychology and Trinity College Institute of Neuroscience, Trinity College, Dublin, Ireland b Department of Psychiatry and Behavioral Medicine, Medical College of Wisconsin, Milwaukee, USA Received 27 January 2004; revised 12 May 2004; accepted 26 May 2004 Recent studies have suggested that declining inhibitory control observed during simultaneous increases in working memory (WM) demands may be due to sharing common neural resources, although it is relatively unclear how these processes are successfully combined at a neural level. Event-related functional MRI was used to examine task performance that required inhibition of varying numbers of items held in WM. Common activation regions for WM and inhibition were observed and this functional overlap may constitute the cortical basis for task interference. However, maintaining successful inhibitory control under increasing WM demands tended not to increase activation in these overlapping regions as might be expected if these common areas reflect common resources essential for task performance. Instead, increased activation was observed predominantly in unique, inhibition-specific regions including dorso- lateral prefrontal cortex. The finding that successfully maintaining weaker stimulus – response relationships in the face of competition from stronger, prepotent responses requires greater activity in these regions reveals the means by which the brain resolves task interference and supports theories of how top-down attentional control is implemented. D 2004 Elsevier Inc. All rights reserved. Keywords: Working memory; Task performance; Interference Introduction Kane, Engle, et al. (Engle et al., 1999; Kane and Engle, 2003; Kane et al., 2001) argue that the ability to control attention underlies the ability to inhibit irrelevant processing and to switch between competing tasks. This theory developed on some earlier work (Hasher and Zacks, 1988), including other models of attention that had also grouped these functions, for example, Baddeley (2001), attributes them to the central exec- utive component of his tripartite WM model and Burgess and Shallice (1998) to their supervisory attentional system. Within these models is the suggestion that the ability to control attention is influenced by and related to working memory (WM). For instance, Kane et al. argue that controlling attention is, among a number of things, the ability to maintain a stimulus or goal in the face of interference. As such, WM capacity provides an index of this ability due to the requirement of many WM tasks to simultaneously store and process information in the face of interference. Kane et al., along with other groups, have demonstrated a relationship between an individual’s ability to perform traditional executive tasks such as the Stroop (Kane and Engle, 2003) and antisaccade task (Kane et al., 2001) and their WM capacity (Baddeley et al., 2001; de Fockert et al., 2001; de Zubicaray et al., 2000; Mitchell et al., 2002; Roberts et al., 1994). For example, Roberts et al. (1994) demonstrated that antisaccade task performance declined as WM demands were increased. de Fockert et al. (2001) were able to demonstrate that with increasing WM demands, processing of irrelevant face informa- tion, measured via functional MRI, increased activity in the extrastriate and fusiform cortex. The suggestion from these studies is that WM plays a critical role in actively maintaining attentional priorities. As a consequence, when greater load demands are placed on WM, implementing these ‘attentional priorities’ suffers, and greater processing of irrelevant informa- tion occurs. Imaging studies have implicated the dorsolateral prefrontal cortex (DLPFC) in a number of roles important to both WM and the attentional control required for inhibition. For example, many studies examining executive function tests such as the Stroop, Eriksen Flanker, and Go/No-go tasks have identified this region, arguing that the DLPFC is crucial to the performance of executive tasks where a set of task rules must be maintained in the face of irrelevant information (MacDonald et al., 2000; Ullsperger and von Cramon, 2001; Zysset et al., 2001). A large body of research has also suggested that when required to maintain increasing WM loads, the level of activation in the DLPFC selectively increases (Braver and Bongiolatti, 2002; D’Esposito et al., 1999; Rypma et al., 2002; Veltman et al., 2003). Similarly, Rowe et al. (Rowe and Passingham, 2001; Rowe et al., 2000) argue that selecting a response from WM is associated with activation of the DLPFC, in particular Brodmann’s area 46. 1053-8119/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2004.05.024 * Corresponding author. Department of Psychology and Trinity College Institute of Neuroscience, College Green, Trinity College, Dublin, Dublin 2, Ireland. Fax: +353-1-671-2006. E-mail address: hugh.garavan@tcd.ie (H. Garavan). Available online on ScienceDirect (www.sciencedirect.com.) www.elsevier.com/locate/ynimg NeuroImage 23 (2004) 202 – 212