Wei Ji Ma, Baylor College of Medicine

A lot of what we know about human decision making and perception is based on data obtained from psychophysics experiments in which subjects choose between two alternatives on each trial. It is usually assumed that their responses are independent across trials. However, some of our data suggest that a response on one trial biases a subject to make the same response on the next trial. Such sequential dependencies have not been studied extensively. The aims of this project are to:
1. Review previous literature on the topic
2. Gain an understanding of the kinds of statistical dependencies that exist in sequential decisions by analyzing data from previous psychophysical experiments
3. Quantify how strong or serious these effects are
4. Think about possible ways to reduce sequential effects in experiments
The main requirement for this project is familiarity with basic statistics. Familiarity with Matlab would be useful, but not necessary.