糖心TV

, Associate Professor, Department of Psychology will present Learning quickly, but sampling responses

Many everyday estimation tasks ask about discrete objects (How many paint buckets?) or require discrete responses (How many paint buckets are needed to paint a room?). We use Bayesian decision theory to characterize how people make discrete numerical responses, finding that participants learn quickly and accurately from experience but do not always choose the best response.

, Research Fellow, Department of Psychology, will present Sampling strategies: People are correct some of the time.

Do people use the correct strategy when combining probability information? We present an individual-level Bayesian analysis of data from a conjunction and disjunction estimation task (N=105). The correct strategy was used some of the time and many people sample strategies on a trial-by-trial basis.