The amygdala, OFC and MFC contribute to learning, in part, by evaluating feedback. In an effort to extend our understanding of the physiological mechanisms underlying affective processing, we developed an fMRI paradigm to reveal blood-oxygen-level dependent (BOLD) responses to visual images that signal reward. Each week, subjects learned to associate images of novel objects with a high or low probability of water reward. Areas responding to the value of recently learned reward-predictive images included MFC area 10m/32, ventrolateral prefrontal cortex area 12, and inferior temporal visual cortex (IT). The amygdala and OFC, each thought to be involved in value encoding, showed little such effect. Instead, these two areas primarily responded to visual stimulation and reward receipt, respectively. Our findings demonstrate the importance of ventrolateral prefrontal cortex, MFC, and IT in representing the values of recently learned visual images. Tastes and their associated values drive food consumption and influence choice behavior. To understand the neural mechanisms underlying reward-guided decision making and value learning, we characterized the gustatory system in subjects with fMRI while measuring licking to control for oral movements, and to assess preferences. To identify taste-responsive cortex, we delivered small quantities of sucrose, citric acid, or distilled water in random order without any predictive cues (e.g., visual stimuli) to subjects while using event-related fMRI. In addition, we used an MRI-compatible lick sensor to measure subjects licking during the scans. fMRI signals associated with licking in the absence of fluid delivery were used to mask responses to fluid delivery/receipt and associated licking. Licking in the absence of fluid delivery and fluid receipt times for each tastant were incorporated into a general linear model to analyze fMRI data. By contrasting BOLD responses to sweet and sour tastes with those from distilled water, we identified taste responses in primary gustatory cortex area G, an adjacent portion of the anterior granular and dysgranular insular cortex, and area 12o. Choice tests run outside the scanner revealed that all three monkeys strongly preferred sucrose over citric acid or water. BOLD responses in the ventral striatum and amygdala reflected monkeys preferences, with greater BOLD responses to sucrose than citric acid. Finally, we examined the influence of hydration level by contrasting BOLD responses to receipt of fluids when monkeys were thirsty and after ad libitum water consumption. BOLD responses in area G and area 12o in the left hemisphere were greater following full hydration. By contrast, BOLD responses in portions of medial frontal cortex were reduced after ad libitum water consumption. These findings highlight brain regions involved in representing taste, taste preference and the motivational aspects of tastes. Increasingly, changes in pupil size are being used as an index of affective state. Under appropriate conditions, pupil size can be correlated with behavioral events with good temporal resolution, making it a useful measure of learning. To probe the contribution of OFC to learning, we trained intact animals and those with OFC lesions on Pavlovian trace conditioning of stimulus-reward associations. The visually presented conditioned stimulus (CS+) was followed by a fluid reward delivered 500 ms after the stimulus was turned off. The CS was followed by an unfilled interval and no reward was delivered. Pupil size was measured with an eye-tracking device. Controls exhibited an increased pupil size to the CS+, compared to the response to the CS-, within a couple of training sessions (mean = 2) and continued to show the conditioned pupil response in anticipation of reward across at least 4 consecutive sessions. By contrast, animals with OFC lesions required more sessions (mean = 14 sessions) to acquire the conditioned autonomic response and three out of four failed to sustain it for 4 consecutive sessions even with an extended period of training (48 sessions on average). We conclude that the OFC is necessary for learning Pavlovian autonomic responses to cues that predict positive events. An earlier study showed that amygdala inputs made a significant contribution to the stimulus-outcome coding of neurons in OFC. In that study, subjects learned a small set of stimulus-reward associations. When these associations were well learned, and highly familiar, neuronal activity was recorded in OFC. In a follow up study, we recorded from OFC neurons while subjects were engaged in learning of stimulus-outcome associations. As before, subjects performed a task that required them to evaluate two stimuli and then choose one to receive the reward associated with that option. Four main findings emerged. First, amygdala lesions slowed the acquisition and use of stimulus-reward associations. Further analyses indicated that this impairment was due, at least in part, to ineffective use of negative feedback to guide subsequent decisions. Second, the activity of neurons in OFC and MFC rapidly evolved to encode the amount of reward associated with each stimulus. Third, amygdala removal reduced the encoding of stimulus-reward associations during the evaluation of different stimuli. Reward encoding of anticipated and received reward after choices were made was not altered. Fourth, amygdala lesions led to an increase in the proportion of neurons in MFC, but not OFC, that encoded the instrumental response that monkeys made on each trial. These correlated changes in behavior and neural activity following amygdala lesions strongly suggest that the amygdala contributes to the ability to rapidly learn stimulus-reward associations through shaping encoding within OFC and MFC.
