Information from Lay-Language Summaries is Embargoed Until the Conclusion of the Scientific Presentation
187—Decision Making: Prefrontal Cortex
Sunday, November 10, 2013, 8:00 am - 12:00 noon
187.28: Knowing what you want and how to get it: Interaction between the orbital prefrontal cortex or prelimbic cortex and the amygdala in goal-directed behavior in rhesus monkeys
Location: Halls B-H
E. C. FIUZAT, S. E. V. RHODES, *E. A. MURRAY; Lab. of Neuropsychology, NIMH, NIH, BETHESDA, MD
Abstract Body: Goal-directed behavior is based on knowledge of the relationship between actions and their corresponding goal and knowledge about the incentive value of the goal. Using an action-outcome task devised for macaques, we found that the orbital prefrontal cortex (PFo) and the amygdala are essential for goal-directed behavior (Rhodes et al., 2013, J. Neurosci., 33: 3380). Specifically, we found that monkeys with bilateral excitotoxic lesions of either the PFo or the amygdala showed disrupted devaluation effects. Results following bilateral aspiration lesions of prelimbic cortex (PL) were equivocal. To build on these findings, and to investigate whether the amygdala and either of the prefrontal areas must interact to guide goal-directed behavior, we examined the effect of surgical crossed-disconnections of the amygdala and PFo and, separately, of the amygdala and PL. Using the same methods as Rhodes et al. (2013), rhesus monkeys (Macaca mulatta) were trained to perform two different instrumental responses (tap and hold) on a touch-sensitive screen for two different food rewards. To determine whether these two instrumental responses were goal directed, monkeys were tested after sensory-specific reinforcer devaluation. In brief one of the two food outcomes was prefed to satiety to lower its value, and then monkeys were tested for their willingness to produce the two different actions. If the behavior were goal directed, then devaluation of a specific food reward would result in reduction in the response that produced that goal (i.e., that particular reward outcome). Our main measure was the number of responses performed in the test sessions run immediately after reinforcer devaluation. Responses were of two types: 1) responses associated with the devalued food and 2) responses associated with the nondevalued food. In sessions following selective satiation, unoperated controls (N=12) were on average almost twice as likely to produce the nondevalued relative to the devalued response (4.6 devalued and 8.6 nondevalued responses). Monkeys with crossed lesions of the amygdala and PL (Amyg x PL; N=4) performed similarly to controls, showing clear devaluation effects (5.2 devalued and 9.0 nondevalued responses). By contrast, monkeys with crossed lesions of the amygdala and PFo (Amyg x PFo; N=2) did not show such a difference. On average, these monkeys performed 6.9 devalued responses and 7.9 nondevalued responses in sessions following reinforcer devaluation. These preliminary data suggest that functional interaction of the amygdala with PFo, but not PL, is essential for goal-directed behavior.
Lay Language Summary: The amygdala is a part of the limbic system that helps to shape behavior in response to a variety of good and bad stimuli. It shares elaborate communication channels with the prefrontal cortex- the brain’s control center for planning and decision making. Recent studies indicate that the amygdala assigns value to rewards and adjusts that value as circumstances change. It is also thought that the amygdala along with the orbital prefrontal cortex (a section of the prefrontal cortex) interact to guide reward-based behavior. It is hypothesized that the amygdala and the orbital prefrontal cortex are important for guiding behavior after changes in the value of a goal or reward. Of special interest is the way in which value is linked to particular choices or actions. Goal-directed behavior is a “vital component of our behavioral repertoire, enabling us to control our environment in the service of our desires and needs” (Dickinson and Balleine, 1994). To assess how value is linked to actions we trained monkeys to perform two different actions to earn two different food rewards. Specifically, we trained six rhesus macaques to perform two different responses, tap and hold, on a touch-sensitive screen for two different food rewards, peanuts and skittles respectively. We prefed the monkeys one of the food rewards to satiety. While monkeys were still sated, we placed them in the testing apparatus and gave them the option of performing either the tap or the hold responses. Prefeeding temporarily devalues one of the food rewards. Accordingly, if the behavior is goal directed, animals will perform fewer responses that produce that food. Control monkeys, as expected, were on average twice as likely to choose the response associated with the higher value (non-sated) food. Monkeys with brain lesions of the amygdala in one hemisphere and brain lesions of the orbital frontal cortex in the opposite hemisphere did not respond similarly. They performed an almost equal number of taps as holds, despite the drop in value of one of the rewards. These types of crossed disconnection lesions are immensely useful because they produce a specific disruption of the functional interaction of those two structures, while at the same time preserving function of the individual structures in one hemisphere. Our research indicates that the amygdala and the orbital prefrontal cortex work together to help us determine the relationship between our actions and our current goals. Monkeys with dysfunction in these two areas cannot make the switch to a new action once a reward (goal) loses value. This is important in helping scientist understand how rewards can sway attention or learning and help people make choices. Understanding this neural circuitry may help to develop new therapies for those suffering from depression, anxiety disorders, and provide insight into reward-seeking behaviors like addiction. Dickinson A, Balleine B (1994) Motivational control of goal-directed action. Anim Learn Behav 22:1-18.
Neuroscience 2013 (43rd annual meeting of the Society for Neuroscience)Exit