Computational Strategies for Balancing Trade-offs between Risk and Effort during Walking

Understanding factors that contribute to fall risk has broad implications for improving the health and quality of life for several populations with neuromotor impairments, including people post-stroke, people with Parkinson’s disease, and amputees. There are also significant medical costs associated with falls: In the United States alone, falls among people who are at least 65-years old incur costs of approximately $50 billion.

Conventional approaches to reducing the risk of falling focus on building strength or training people in more effective strategies to recover from loss of balance. However, these approaches fail to address the role of the choices that people make. For example, people may fall because they have difficulty judging risk accurately, overestimate their functional capacity, or tend to be risk tolerant.

Developing methods to assess individual differences in risk perception and risk preference during walking could lead to a new class of fall-reducing interventions that improve a person’s ability to identify and avoid potential risks that push them beyond their physical capacity.

Current theoretical models of how humans select features of their walking pattern primarily aim to minimize an energy-related cost in a risk-neutral context. The overall objective of the present work is to extend the current risk-neutral theory to a more generalizable, and clinically relevant, risk-sensitive theory of locomotion. The investigators will use a novel approach to examine how people balance trade-offs between effort and risk during walking through the use of a risk-sensitive decision-making framework.

They will also determine if models of risk- and effort-based decision-making in other domains explain age-dependent differences in how people choose between routes associated with varying levels of physical risk and effort. In addition to identifying fundamental principles of behavior selection during walking, the work will also provide high school students with practical experiences working at the intersection of experimental motor control, biomechanics, and game development.

The investigators will collaborate with a local high school in Los Angeles to develop project-based activities focused on statistical thinking and biomechanics that align with the Common Core and Next Generation Science Standards. These activities will be made available for the public to use through the USC Viterbi K-12 Stem Center, on the principal investigator’s website, and via Github.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.