Emotion Inference in Artificial Social Intelligent Agents
Advances in computational cognitive modelling have enabled researchers to formulate and evaluate precise hypotheses regarding how individuals make social inferences (Ong et al., 2019). One of the key concepts underlying these studies is the theory of mind, which refers to the ability to attribute mental states (such as beliefs, desires, intentions, and emotions) to oneself and others, as it enables people to make predictions about others' behavior, understand their intentions, and coordinate their actions (Chen et al., 2021).
Despite recent advancements in computational cognitive modeling, emotions have often been neglected in these models (Ong et al., 2019). This signifies a significant limitation in comprehending a crucial aspect of human psychology, as emotions have a pivotal impact on social interactions, decision-making, and overall mental and emotional health.
One area where the theory of mind and computational cognitive modelling intersect is in the development of artificial social intelligence (ASI). ASI aims to create intelligent agents, that can interact with humans in a socially intelligent way (Williams et al., 2022).
The objective of this research endeavour is to construct a comprehensive framework, drawing inspiration from the work of Ong et al. (2019), aimed at facilitating the precise prediction and recognition of emotions expressed in written natural language. The proposed model employs a probabilistic methodology that entails the calculation of the likelihood associated with a particular emotion, based on variables encompassing actions, outcomes, beliefs, and desires present within the textual input.
The effectiveness of the model is going to be compared to a baseline language model text generator that does not use the emotion recognition algorithm however it is trained with various datasets.
It is hypothesized that the model with emotion inference algorithm provides better predictions of emotions in written natural language.
 D. C. Ong, J. Zaki, and N. D. Goodman, “Computational Models of Emotion Inference in Theory of Mind: A Review and Roadmap,” Topics in Cognitive Science, vol. 11, no. 2, pp. 338–357, Jul. 2018. doi:10.1111/tops.12371
 J. Williams, S. M. Fiore, and F. Jentsch, “Supporting Artificial Social Intelligence With Theory of Mind,” Frontiers in Artificial Intelligence, vol. 5, Feb. 2022. doi:10.3389/frai.2022.750763
 I. Delis, C. Chen, R. E. Jack, O. G. B. Garrod, S. Panzeri, and P. G. Schyns, “Space-by-time manifold representation of dynamic facial expressions for emotion categorization,” Journal of Vision, vol. 16, no. 8, p. 14, Jun. 2016. doi:10.1167/16.8.1