Causal Learning With Local Computations

• Published in: Journal of experimental psychology. Learning, memory, and cognition memory, and cognition, 2009-05, Vol.35 (3), p.678-693
• Main Authors: Fernbach, Philip M, Sloman, Steven A
• Format: Article
• Language: English
• Subjects: Association Learning; Bayesian Statistics; Biological and medical sciences; Causal Analysis; Causal Models; Causality; Cognition. Intelligence; Cues; Decision Making; Experimental psychology; Experiments; Feedback; Fundamental and applied biological sciences. Psychology; Heuristics; Human; Humans; Hypotheses; Imagination; Inference; Inferences; Information processing; Knowledge of Results (Psychology); Learning; Learning. Memory; Memory; Mental Recall; Motion Perception; Orientation; Practice (Psychology); Problem Solving; Psychological Theories; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Reasoning. Problem solving; Statistical Probability
• ISSN: 0278-7393; 1939-1285
• DOI: 10.1037/a0014928

The authors proposed and tested a psychological theory of causal structure learning based on local computations . Local computations simplify complex learning problems via cues available on individual trials to update a single causal structure hypothesis. Structural inferences from local computations make minimal demands on memory, require relatively small amounts of data, and need not respect normative prescriptions as inferences that are principled locally may violate those principles when combined. Over a series of 3 experiments, the authors found (a) systematic inferences from small amounts of data; (b) systematic inference of extraneous causal links; (c) influence of data presentation order on inferences; and (d) error reduction through pretraining. Without pretraining, a model based on local computations fitted data better than a Bayesian structural inference model. The data suggest that local computations serve as a heuristic for learning causal structure.