Jeffrey Starns

People have an amazing ability to remember specific events. In response to a vague cue (e.g., "Did you travel last summer?"), we can retrieve detailed information in seconds even though it must be isolated from the multitude of other experiences that we could potentially remember. In my research, I test theories of how memory works. I work with theories that are expressed as computational models, which means that anyone can figure out what the theory predicts by solving a set of equations or running a computer program.

I am also interested in failures of memory, especially illusory memories. People sometimes retrieve a compelling memory of an event that did not actually occur; for example, they might swear that they fell off of a horse as a child when this actually happened to one of their siblings. Many of the models that I use assume that illusory memories are an unavoidable "side effect" of the systems and processes that support accurate memory.

Although most of my projects investigate memory, I also explore how people evaluate evidence when they are challenged to make quick decisions. This decision-making work focuses on models that can account for response times as well as accuracy.

Finally, I am currently developing methods for teaching people how to properly update their beliefs when they encounter new information, a process known as Bayesian Reasoning.

Laboratory Website: 

Dynamic Memory Modeling Lab