Neurocognitive modeling of emotional states using EEG and hidden markov models: A multidisciplinary approach

This interdisciplinary research cuts computational modeling and cognitive neuroscience approaches with the intention of studying dynamic emotional involvement with multimedia stimuli via HMM analysis of EEG data. In particular, the paper deals with advertisements that target excitement and love-type emotions, setting forth new paradigms for understanding the building and modulation of emotional experience across time in the human brain. EEG parameters such as amplitude, arousal, and frontal activation were studied as markers of neural reactions to emotionally arousing content. The neural markers are tracked over time to record the changes in emotional engagement. The HMMs use identifies hidden neural states and their probabilistic transitions, making the temporal description of neural dynamics during emotional processing rich and nuanced. The analytical approach provides identifiable neural patterns for excitement and love stimuli distinguished in terms of arousal, spectral amplitude, and hemispheric asymmetry in frontal activation. Due to these distinctions, we ascertain that the brain processes different affective tones distinctly, shedding light on the intricacies of emotion perception and its immediate brain counterpart. Using the results, a predictive HMM model is presented to model emotional changes when individuals are subjected to effective multimedia stimuli. The model serves as a bridge to further real-time developments in human-computer interaction, adaptive e-learning, immersive media conception, and affective UX (user experience) optimization. In other words, this enables the system to detect shifts in the user's emotions automatically and adapt content accordingly, representing truly affect-sensitive technologies. Amalgamating computational modeling with neurophysiological measurement, this study contributes to the birth of emotion-aware technology that can be dynamically responsive to the users' current affective state, thus harnessing engagement, personalization, and user satisfaction as opportunities. It builds on the interdisciplinary discourse between cognitive neuroscience, affective computing, and computational psychology to serve as a methodological guideline for future investigations into emotional dynamics and brain-computer interfaces (BCIs), as well as neuroadaptive technology. It makes a case for the relevance of temporal modeling in decoding emotional cognition and therefore advocates the continued employment of machine-learning approaches in brain activity and human affective behaviour studies.