Neural mechanisms of interpretation shifting and memory formation in ambiguous social contexts

PROJECT SUMMARY In our daily lives, we frequently encounter ambiguous information. Navigating this ambiguity—generating interpretations, adapting them in response to new information, and encoding them in memory—plays a pivotal role in shaping our thoughts and behaviors. Biases in interpretation and memory are both risk factors and symptoms for major depressive and anxiety disorders, with clinical strategies like cognitive reappraisal— deliberate, targeted changes in how someone interprets a situation—aiming to mitigate these biases. However, past work has either focused on isolated parts of this process (e.g., how someone forms an interpretation without examining their likelihood of shifting it) or used simple perceptual stimuli (e.g., visual illusions) where individuals alternate between a small number of definitive percepts. Real-world scenarios, particularly social ones, are inherently more subjective and thus, are more likely to reveal biases associated with conditions like depression or anxiety. Despite their importance, social scenarios have received limited attention in this research context, possibly because of the difficulty in balancing experimental control with ecological realism. This proposal aims to address these gaps by combining functional magnetic resonance imaging (fMRI) with a novel behavioral task paradigm that mimics real-life social ambiguities. Leveraging advancements in natural language processing (NLP), my task involves presenting participants with ambiguous social scenarios, collecting their subjective interpretations, and then exposing them to alternative views from other participants. NLP enables the objective comparison of a virtually unlimited number of interpretations in order to generate alternatives in real time that are tailored to how the participant just interpreted an image. Past work has demonstrated that multivariate neural patterns reflect differences in percepts and interpretations across subjects as well as changes in percepts within subjects, even when sensory input is held constant. My study will investigate where and how shifts in neural patterns to the same sensory information (namely, ambiguous social photographs) predict changes in interpretations (Aim 1) and the encoding of these interpretations into memory (Aim 2). I anticipate that I will find distinct, yet complementary, neural substrates supporting these behaviors. Ultimately, this research will yield a mechanistic model explaining where, how, and why the same sensory information can lead to different subjective interpretations and how these interpretations are encoded into memory. This model will contribute to our fundamental understanding of these cognitive processes and will also generate testable hypotheses for therapeutic interventions aimed at normalizing interpretational and memory biases related to subjective information.