Multiple-Cue Probability Learning

Synonyms

Function learning; Lens model; Probabilistic categorization; Weather prediction task

Definition

Multiple-Cue Probability Learning (MCPL) is an experimental paradigm concerned with how well people can learn imperfect relationships between cues and outcomes. In a typical MCPL task, participants are shown an array of cues each of which predicts a particular outcome with some probability; usually this probability is less than unity, mirroring the imperfect nature of cues in the natural environment. The cues are usually instantiated as simple perceptual stimuli, which can be either discretely (often binary) valued, such as color – a given cue might be a red or green, for instance – or stimuli can be comprised of continuously valued dimensions – such as bars of different lengths. The former case with discrete cues is typically referred to as nonmetric multiple-cue probability learning (NMCPL), and the latter case with continuous cues is termed metricmultiple-cue probability...