Academic journal article Cognitive, Affective and Behavioral Neuroscience

Behavioral and Neural Predictors of Upcoming Decisions

Academic journal article Cognitive, Affective and Behavioral Neuroscience

Behavioral and Neural Predictors of Upcoming Decisions

Article excerpt

Although it is widely known that brain regions such as the prefrontal cortex, the amygdala, and the ventral striatum play large roles in decision making, their precise contributions remain unclear. Here, we used functional magnetic resonance imaging and principles of reinforcement learning theory to investigate the relationship between current reinforcements and future decisions. In the experiment, subjects chose between high-risk (i.e., low probability of a large monetary reward) and low-risk (high probability of a small reward) decisions. For each subject, we estimated value functions that represented the degree to which reinforcements affected the value of decision options on the subsequent trial. Individual differences in value functions predicted not only trial-to-trial behavioral strategies, such as choosing high-risk decisions following high-risk rewards, but also the relationship between activity in prefrontal and subcortical regions during one trial and the decision made in the subsequent trial. These findings provide a novel link between behavior and neural activity by demonstrating that value functions are manifested both in adjustments in behavioral strategies and in the neural activity that accompanies those adjustments.

Recent research in cognitive neuroscience has focused on understanding the neural mechanisms of decision making under uncertainty (Kahn et al., 2002; Krawczyk, 2002; Manes et al., 2002; Sanfey, Hastie, Colvin, & Grafman, 2003; Sanfey, Rilling, Aronson, Nystrom, & Cohen, 2003). Although existing neuropsychological and neuroimaging research has helped to identify the brain regions involved in decision making, questions remain regarding the precise functions of these regions. For example, substantial evidence from economic and cognitive decision making research has demonstrated that the outcome of one decision strongly influences the following decision (Erev & Roth, 1998; Mookherjee & Sopher, 1994; Sarin & Vahid, 2001), but how brain regions might mediate these trial-to-trial adjustments in behavioral strategies is unknown. If decision making is a dynamic process of choosing a decision, evaluating the outcome, and adjusting future behavior accordingly, brain regions involved in decision making should exhibit activity that reflects trial-totrial adjustments in behavior that occur as a result of the evaluation process. Thus, a more complete understanding of the neurobiological mechanisms of decision making must take into account how outcomes and neural processes during one trial affect decisions on subsequent trials.

Reinforcement learning theory provides a framework for assessing the predictive value of current outcomes for future decisions (Erev & Roth, 1998; Sarin & Vahid, 2001 ). Models based on reinforcement learning theory have been used copiously in the realms of machine learning and economic decision making (Barto, 1995; Sutton, 1992), and recently, researchers have utilized it when investigating the neural processes underlying reward-based learning (Barraclough, Conroy, & Lee, 2004; Egelman, Person, & Montague, 1998; O'Doherty, Dayan, Friston, Critchley, & Dolan, 2003; Paulus, Feinstein, Tapert, & Liu, 2004; Schultz, 2004). These researchers have demonstrated that principles of reinforcement learning theory can be used to predict activity in dopaminergic brain regions, such as the orbitofrontal cortex and the striatum, that correlate both with the predictive power of the cues associated with rewards (Friston, Tononi, Reeke, Sporns, & Edelman, 1994; Montague & Berns, 2002; O'Doherty et al., 2003; Schultz, 2004) and with changes in performance during learning (O'Doherty et al., 2003; Paulus et al., 2004). However, it remains unknown whether reinforcement learning theory can help explain the neural activity that underlies trial-to-trial adjustments in behavioral strategies.

A key prediction of reinforcement learning theory is that individuals use the outcomes of their decisions as reinforcements to update or adjust their decision strategies. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.