Neurons in the orbitofrontal cortex (OFC) encode the sensory properties, magnitude and subjective value of expected and received rewarding outcomes. In addition, the activity of OFC neurons reflects anticipatory arousal, as measured by pupil diameter. OFC neuronal activity is also modulated by attention. It is not well understood yet how gaze and/or attentional modulation of neuronal activity in OFC guides behavior, and in what circumstances OFC neuronal activity is necessary. To investigate the causal contribution of OFC to visual exploration of valuable objects, we trained animals with bilateral excitotoxic OFC lesions and controls on a preferential viewing task. The experiment occurred in two stages: stimulus-reward association training and preferential viewing test. The stimulus-reward training established thirty high-value images and thirty low-value images. When the training stage was completed, we administered a preferential viewing test. On each trial, two images -- one high-value image and one low-value image -- were presented simultaneously and the animals freely viewed them for 4 s. In this stage, an intermediate amount of fluid was delivered at the end of every trial. Both groups showed a consistent preference for the high-value images. During the preferential viewing test, when both a high- and a low-value image were available, animals looked at the high-value image first on 80% of trials. In addition, they viewed the high-value images 3.7 times longer and made 3.5 times more fixations relative to the low-value images. There was no group difference on any of these measures. Our findings indicate that, at least when stimulus-value associations are acquired postoperatively, OFC is not necessary for guiding the eyes to valuable images. For decades, it has been known that lesions of OFC and surrounding white matter produce blunted affective responses to threatening stimuli. Recent evidence involving restricted (excitotoxic, fiber-sparing) damage to OFC has shown contradictory effects of either no involvement of OFC or heightened affective responses, but these studies differ with respect to the species studied and the extent of OFC damage. To address these discrepancies and further examine the putative role of OFC in emotional regulation, we studied animals with restricted bilateral excitotoxic lesions targeting either lateral OFC or medial OFC. The performances of these groups were compared to that of a group of unoperated controls on defensive responses to the presentation of a rubber snake, rubber spider, and neutral objects for comparison. Both subtotal lesion groups showed heightened defensive responses as indexed by latency to retrieve a food reward in the presence of the spider and snake. Surprisingly, the group with medial OFC lesions showed greater food-retrieval latencies even in the presence of neutral objects, compared to unoperated controls. Our data indicate that medial and lateral OFC make dissociable contributions to emotional regulation. In addition, the findings strongly suggest that any emotional blunting associated with OFC damage is due to a disruption of fibers of passage (that is, fibers passing nearby OFC), as opposed to damage to cell bodies residing in OFC. Lesion studies suggest dissociable functions of medial prefrontal cortex (MFC) and orbitofrontal cortex (OFC), with MFC being essential for social cognition and OFC being essential for value-based decision making. Although bilateral amygdala damage also results in impairments in these domains, it is not known whether the dissociable functional roles of MFC and OFC critically depend on interactions with the amygdala. To test this possibility, we compared the performance of animals with crossed surgical disconnection of the prelimbic cortex, a subregion of MFC, and amygdala (PL x AMY) and animals with surgical disconnection of the OFC and amygdala (OFC x AMY), to a group of controls (CON). All animals were assessed for food-retrieval latency while viewing videos of social stimuli (a test of social interest) and object choices based on current food value (devaluation task, a test of value-based decision making). Compared to the CON group, group PL x AMY, but not group OFC x AMY, showed significantly reduced latencies to reach for a reward in the presence of videos of conspecifics, indicating reduced social valuation and/or reduced social interest. In a test of value-based decision making, however, the opposite pattern was observed; group OFC x AMY, but not group PL x AMY, displayed severe deficits on object choice following selective satiation. These data indicate that MFC and OFC interact with the amygdala to subserve distinct behavioral contributions in the domains of social cognition and decision making, respectively. Adaptive behavior requires animals to learn from experience. Ideally, learning should both promote choices that lead to rewards and reduce choices that lead to losses. Because the ventral striatum (VS) contains neurons that respond to aversive stimuli and aversive stimuli can drive dopamine release in the VS, it is possible that the VS contributes to learning about aversive outcomes, including losses. To examine the role of the VS in learning from gains and losses, we compared the performance of animals with VS lesions and controls on a reinforcement learning task. In the task the animals gained or lost tokens, which were periodically cashed out for juice, as outcomes for choices. They learned over trials to choose cues associated with gains, and not choose cues associated with losses. We found that animals with VS lesions had a deficit in learning to choose between cues that differed in reward magnitude. By contrast, animals with VS lesions performed as well as controls when choices involved a potential loss. We also fit reinforcement learning models to the behavior and compared learning rates between groups. Relative to controls, the animals with VS lesions had reduced learning rates for gain cues. Therefore, in this task, the VS plays a specific role in learning to choose between rewarding options. Additional information about learning can be gleaned from probabilistic reward tasks, in which animals learn to choose between three probabilistically rewarded images presented on a touch screen. In one such task, one of the three options was replaced with a novel image that had not yet been associated with reward. Animals therefore had to choose between exploring the novel option or exploiting their existing knowledge about the two familiar options that remained available, where the availability of reward was known. Impairments in optimizing rewards via novelty exploration present as either increased novelty seeking or novelty aversion. We found that animals with VS lesions were impaired on the probabilistic three-arm bandit task implemented on a touchscreen. They showed increased novelty seeking as indexed by choice and reaction times upon the first presentation of a novel stimulus, especially when the overall probability of reward was greatest prior to the presentation of a novel image. The results suggest that the VS may be specialized to guide choices based on information about reward likelihood.
Showing the most recent 10 out of 20 publications
