Functional differences in face processing between the amygdala and ventrolateral prefrontal cortex in monkeys

- Amygdala neurons preferentially responded to facial expressions that are strongly negative.
- Emotional information of amygdala neurons showed a phasic peak immediately after face presentation onset.
- Ventrolateral prefrontal cortex neurons preferentially responded to facial expressions with multiple meanings.
- Emotional information of ventrolateral prefrontal cortex neurons showed a continuous elevation during face presentation.

The ability to categorize social information is essential to survive in a primate's social group. In the monkey brain, there are neural systems to categorize social information. Among these, the relationship between the amygdala and the ventrolateral prefrontal cortex (vlPFC) has recently gained focus with regard to emotion regulation. However, the processing of facial information and the functional differences in these two areas remain unclear. Thus, in this study, we examined the response properties of single neurons in the amygdala and vlPFC while presenting video clips of three types of facial emotions (aggressive threat, coo, and scream) in Macaca mulatta. Neurons in the amygdala were preferentially activated upon presentation of a scream facial expression, which is strongly negative, whereas the neurons in the vlPFC were activated upon presentation of coo, a facial expression with multiple meanings depending on the social context. Information analyses revealed that the amount of information conveyed by the amygdala neurons about the type of emotion transiently increased immediately after stimulus presentation. In contrast, the information conveyed by the vlPFC neurons showed sustained elevation during stimulus presentation. Therefore, our results suggest that the amygdala processes strong emotion roughly but rapidly, whereas the vlPFC spends a great deal of time processing ambiguous facial information in communication, and make an accurate decision from multiple possibilities based on memory.