Sensation, Incentive Learning, and the Motivational Control of Goal-Directed Action

Excerpt

Recent analyses of goal-directed action have not only pointed to the importance of the learning processes through which actions and their consequences are encoded, but have also emphasized the performance factors that influence choice between actions and action initiation more generally (Dickinson and Balleine 2002; Hasselmo 2005; Balleine and Ostlund 2007). It is important to understand why this is the case. Although the value of acting—or not acting—can often appear to be obvious enough from an adaptive perspective, in fact the information that can be derived from the action-outcome association is not sufficient to determine a course of action; knowing that an action results in a particular outcome does not entail whether that action should be performed or not. Although it might appear adaptive to perform food-related actions more frequently when food deprived and less frequently when replete, this need not necessarily occur and often doesn’t, something well documented in cases of eating disorders (Davis et al. 2004; Morrison and Berthoud 2007). In fact, what determines whether a specific action will be both selected and subsequently initiated is not just the identity but also the evaluation of the outcome associated with the selected action.

In order to decide on a course of action, therefore, both the consequences and the value of the consequences of various alternative actions need to be specified. Therefore, establishing the determinants of a course of action, whether in psychological or neural terms, requires an account of (i) the learning processes through which action-outcome associations are encoded and (ii) the motivational and emotional processes that establish the value of the consequences or outcomes of actions. In this chapter both of these processes will be described further although, because there have been a number of recent reviews of the learning processes underlying goal-directed action (Balleine, Delgado, and Hikosaka 2007; Balleine, Liljeholm, and Ostlund 2009; Balleine and Ostlund 2007; Yin, Ostlund, and Balleine 2008; Balleine and O’Doherty 2010), I will mainly focus on evidence relating to the behavioral and neural bases of incentive learning, i.e., the process by which we and other animals assign value to the consequences or goals of goal-directed actions. To provide the basis for the presentation of current research on this issue I will focus first on evaluative conditioning, which constitutes the motivational basis of affective processes generally, and then turn to animal models of goal-directed action to describe how this basic evaluative process is elaborated into an incentive learning process in the service of this capacity. I will then take up the issue of the neural bases of incentive learning and describe some recent research on this issue in the final section.