Many of us spent increasing amounts of time viewing digital media. A large body of prior research shown that such viewing emotionally engaging media elicits strong autonomic, visceral, and neural responses that vary as a function of media category, with emotional content eliciting stronger responses than emotionally neutral content. The most widely used collection of standardized emotional digital pictures is the International Affective Picture System, developed and curated by Peter Lang and Margaret Bradley here at the University of Florida.
Measuring how the brain and the body respond to emotional media such as pictures or videos is of interest not only for basic neuroscience research, but also for clinical intervention and diagnostic procedures in a variety of psychiatric disorders. In this group of research projects, we leverage rapidly evolving generative AI technology, along with multimodal neuroimaging (simultaneous EEG and fMRI) and advanced computational modeling to characterize—in time and space—the neural code of how emotional responses to digital media unfold. In addition, we establish a generative AI architecture that generates emotional pictures on demand, according to user specifications.