Primate amygdala neuron process the value of visual stimuli to guide decision-making and social interactions (Grabenhorst et al., 2019; Grabenhorst et al., 2023). However, it remains unclear whether different amygdala neurons are sensitive to distinct value components, such as reward probability and magnitude, and whether they integrate these components to decision variables, including risk and expected value. To address these questions, we recorded activity of amygdala neurons while monkeys viewed sequentially presented stimuli that provided separate information about the probability, magnitude, risk (variance), and expected value of upcoming rewards. Many amygdala neurons signalled reward probability in an abstract code that generalized across different visual cues. Some probability-coding neurons were insensitive to reward-magnitude information, thus signalling ‘pure’ probability rather than value. Other neurons signalled both sequentially cued probability and magnitude in a dynamic and flexible (reversible) manner, consistent with the integration of value components. In single neurons and population codes, such integration was reflected in explicit risk and expected-value signals. Population codes for these reward parameters were highly accurate and shown in behavioural comparisons to be suitable for guiding the monkeys’ economic choices. These findings reveal specializations in individual amygdala neurons for processing distinct value components and dynamically integrating them to decision variables. References Grabenhorst F, Baez-Mendoza R, Genest W, Deco G, Schultz W (2019) Primate Amygdala Neurons Simulate Decision Processes of Social Partners. Cell 177:986-998 e915.Grabenhorst F, Ponce-Alvarez A, Battaglia-Mayer A, Deco G, Schultz W (2023) A view-based decision mechanism for rewards in the primate amygdala. Neuron 111:3871-3884 e3814.