Neural Correlates of Obstacle-Avoidance Planning

Introduction: A basic principle of cognitive neuroscience is that the brain directs movement, but it is unclear how this is done. Currently, we know very little about the neural computations mediating simple actions like reaching across a table, much less about more complex ones like playing tennis or violin. I want to understand the brain mechanisms that underlie this movement preparation. My goal is to combine computational analyses with empirical tests in order to propose neural computations that give rise to motion planning. High-density EEG, fMRI, behavioral studies, and computational modeling will be used to study the neural correlates of action. Using mathematics and neuroscience methods in tandem has shown me that neuroimaging and computation-driven prediction and testing, together, have the potential to shed light on movement planning to an extent not possible for each technique independently. My proposal expands on past research and on a series of pilot studies that will form the core of my Senior Project in neuroscience. I will examine the temporal dynamics of the lateralized readiness potential (LRP) during movements around obstacles. The LRP is a stimulus-evoked event-related potential (ERP) that occurs after one hand is preferentially primed to move. The LRP is often examined in precuing paradigms (Rosenbaum, 1980), in which advance information (the precue) is given about anatomical (e.g. hand, finger) or functional (e.g. direction, force) parameters of the response. When the hand is precued, the LRP occurs in the foreperiod before the response signal, and the LRP amplitude is greater than in uninformative cue conditions (De Jong et al., 1988). Furthermore, the foreperiod LRP amplitude is greater when all functional parameters about the movement are precued, in addition to the anatomical parameters (Leuthold et al., 2004). Most LRP studies involve simple responses. What happens when the movement is more complex? We constantly navigate through cluttered environments where there is no direct path to our goal. Thus, it is crucial to determine how an obstacle's presence affects the underlying motor program. Models of motor planning explain planning mechanisms for avoiding obstacles, and neuroimaging can assess the viability of the models' core predictions. I will focus on two influential models of planning that account for findings in behavior and single neurons, but have never been extended to neural findings at the cognitive level. The Posture-Based Movement Planning (PBMP) model (Vaughan, 2001) proposes that a goal posture to the target is planned first, followed by …