Computational mechanisms underlie motor decision-making and skill learning.
Richard Ivry, Professor
Psychology
Applications for Spring 2025 are closed for this project.
Aims: While humans are remarkably adept at adjusting their movements to operate in uncertain environments, motor control remains a major challenge for AI systems. This project aims to study the computational rules underlying human motor control to improve algorithms for human-like robots:
1. We will conduct longitudinal studies to observe how human participants develop complex motor skills over time and apply computational models to capture this development process.
2. We will examine how humans rapidly adapt motor commands to new contexts. For example, individuals quickly adjust their gait when transitioning from walking on grassy terrain to muddy terrain. We will study the computational and neural mechanisms underlying this process.
Role: Duties: We are looking for a research assistant to help with:
1. Coding computer games for skill training using JavaScript.
2. Supervising the longitudinal experiments.
3. Building behavioral and neural network models for motor control.
Qualifications: Qualifications: We are seeking candidates majoring in Computer Science, Data Science, Cognitive Science, or related fields.
Candidates with experience in machine learning, particularly in RNNs and transformers, are preferred.
Day-to-day supervisor for this project: Tianhe Wang, Ph.D. candidate
Hours: 9-11 hrs
Biological & Health Sciences Education, Cognition & Psychology