Prefrontal Cortex and Dynamic Categorization Tasks: Representational Organization and Neuromodulatory Control

Abstract

We present a computational model of the intradimensional/ extradimensional (ID/ED) task (a variant of the Wisconsin card sorting task) that simulates the performance of intact and frontally lesioned monkeys on three different kinds of rule changes (Dias et al., 1997, J Neurosci 17:9285-9297). Although Dias et al. interpret the lesion data as supporting a model in which prefrontal cortex is organized into different processing functions, our model suggests an alternative account based on representational content. A key aspect of the model is that prefrontal cortex representations are organized according to different levels of abstraction, with orbital areas encoding more specific featural information and dorsolateral areas encoding more abstract dimensional information. This representational scheme of the model is integrated with two additional key elements: (i) activation-based working memory representations controlled by a dynamic gating mechanism that simulates the hypothesized phasic actions of dopaminergic neuromodulation in prefrontal cortex, which acts to stabilize or destabilize frontal representations based on success in the task; and (ii) a weight-based associative learning system simulating posterior cortex and other subcortical areas, where the stimulus-response mappings are encoded. Frontal cortex contributes to the task via top-down activation-based biasing of task-appropriate features and dimensions in this posterior cortex system - this top-down biasing is specifically important for overcoming prepotent associations after a sorting rule reverses. The ability of the model to capture the double-dissociation observed by Dias et al. with orbital versus dorsolateral lesions supports the validity of these principles, many of which have also been useful in accounting for other frontal phenomena.

Anatomy of a decision: Striato-orbitofrontal interactions in reinforcement learning, decision making, and reversal.

“…In particular, we suggest that the OFC participates in decision making much in the same way as the dorsolateral PFC is thought to subserve cognitive control processes, via maintenance of goal-directed actions (Frank et al, 2001;Miller & Cohen, 2001;O'Reilly, Noelle, Braver, & Cohen, 2002;Schoenbaum & Setlow, 2001). Although dissociations have been found between the dorsolateral PFC and the OFC for the processes of working memory and decision making, respectively (Bechara et al, 1998), we suggest that the difference in these two regions is not in function but in the content of working memory representations (e.g., Goldman-Rakic, 1995).…”

“…Other computational models of OFC function have focused on the role of the OFC in reversal learning (Deco & Rolls, 2005), decision making (Wagar & Thagard, 2004), and attentional set shifting (O'Reilly et al, 2002). All of these models explore specialized mechanisms within the OFC that support such behavior.…”

Banishing the homunculus: Making working memory work

“…This research builds upon earlier work simulating many of the paradigmatic tasks thought to be characteristic of working memory and executive function, including: the Stroop effect (Cohen et al, 1990;O'Reilly & Munakata, 2000); the AX-CPT (Braver et al, 1995); the 1-2-AX (O'Reilly & Frank, submitted-a); the WCST (Rougier & O'Reilly, 2002); the ID/ED dynamic categorization task (O'Reilly, Noelle, Braver, & Cohen, 2002); and the Eriksen flanker task (Eriksen & Eriksen, 1974;Cohen, Servan-Schreiber, & McClelland, 1992;Yeung, Botvinick, & Cohen, 2004;Bogacz & Cohen, 2004). We also plan to simulate the ABCA/ABBA task (Miller et al, 1996), serial recall (phonological loop) (Burgess & Hitch, 1999), Sternberg task (Sternberg, 1966), and the N-Back task .…”

Learning representations in a gated prefrontal cortex model of dynamic task switching

“…Task-relevant information that should be maintained is a reliable predictor of reward, and should thus elicit dopamine firing, resulting in the updating of working memory (Braver & Cohen, 2000;O'Reilly et al, 1999), and the negative error signal should reset working memory representations. The net effect is to produce a form of trial-and-error search by activating and deactivating PFC representations (O'Reilly & Munakata, 2000;O'Reilly et al, 2002).…”

“…In this paper we present a biologically-based model performing a WCST-like task switching task that helps to illuminate the mechanistic role that the PFC plays in task switching. Specifically, our model is founded on the idea that the PFC is specialized for activation-based working memory (Braver & Cohen, 2000;Frank, Loughry, & O'Reilly, 2001;Miller & Cohen, 2001; O'Reilly, Braver, & Cohen, 1999;O'Reilly & Munakata, 2000;O'Reilly, Noelle, Braver, & Cohen, 2002). By representing information as maintained activation states, the PFC can contribute to task switching by rapidly updating these activation states in response to feedback.…”