Computation in Emotional Processing: Quantitative Confirmation of Proportionality Hypothesis for Angry Unhappy Emotional Intensity to Perceived Loss
- First Online:
- 157 Downloads
A computational model of emotion is derived (using minimalistic assumptions) to quantify how emotions are evolved to estimate the accuracy of an internally generated brain model that predicts the external world. In this model, emotion is an emergent property serving as a self-derived feedback that monitors the accuracy of the internal model via the discrepancy (error measure) between the (internal) subjective reality and (external) objective reality—reality-check subconsciously. Minimization of error (computed by the “gain” toward the desired outcome) will optimize congruency between internal and external worlds—resulting in happy emotion. Unhappy emotion is resulted from the discrepancy between internal and external worlds, which can serve as feedback for self-correction to minimize the “loss” (error) between desired and actual outcomes. Unhappiness provides the internal guide to self-identify whether the cause of error is due to input (sensory perception) error, output (motor execution) error, or modeling (internal model) error. Experimental validation of the hypothesis using the ultimatum game paradigm confirmed the inverse proportional relationship of anger to perceived gain (or direct proportionality to loss) that estimates the discrepancy between what we want and what we get. It also characterizes specific emotional biases by shifting the emotional intensity curve quantitatively.