Principal Investigator
Models for Behavioral Reaction Times & Decision Making
Ratcliff's diffusion model. Sample diffusion path pictures the decision-making process in one single trial. Once the sample path hits the boundary (upper for correct response, lower for error response), a decision is made. a
represents the distance between boundaries; z
is the starting point at which the diffusion process starts; Ter
is the non-decision time, e.g. encoding and motor response time; v
is the rate at which the diffusion process proceed.
Much of our past analysis and interpretation of our behavioral studies of the detection of amplitude modulations and detection of tones in noise has been based on signal-detection theory. However, these behavioral results provide a vast database of reaction times in two-choice decision tasks. We have taken advantage of modern models for reaction times and decision making, such as Ratcliff’s Diffusion Model, to study these tasks from a different perspective. The process by which subjects make decisions is a subject that is both little understood and highly disputed. Researchers in many fields are seeking answers in different ways, including philosophy, psychology, and neuroscience. We are seeking to correlate the activity of the brain with the decision making process.
The Diffusion Model provides an excellent fit to the behavioral distributions of reaction times, and thus it also provides a new framework for considering the task and the associated neural responses. In this project, we are trying to answer the following questions:
- How do the properties of sound stimuli affect decision-making?
- How are Reaction Times and brain activity related?
In other words, how are decisions related to these auditory tasks made? We hope to create a model that predicts reaction times using neural activity, specifically, the firing rate of neurons in the auditory midbrain (inferior colliculus) in response to amplitude modulated sounds.
« back to all projects
