Morais,A.
Olsson,H.
Schooler,L.J.

This paper investigates whether the structure of people’s knowledge of causal relations between the features of categories predicts how they search for information in a categorization task. Participants were asked to draw a causal model that described how the symptoms of depression are causally related to one another, and to estimate the strengths of those relationships. Additionally, they were asked to categorize a series of patients as suffering from depression or not, after searching their symptoms. The results showed that the structurally more important a symptom was in a causal model, the more frequently and the earlier in search it was inspected. Also, a measure of feature importance that ignored causal strengths accounted for search behavior at least as well as the weighted version of the same measure.

Morais, A. S., Olsson, H., & Schooler, L. J. (2011). Does the structure of causal models predict information search? In B. Kokinov, A. Karmiloff-Smith, & N. J. Nersessian (Eds.), European perspectives on cognitive science. Sofia: New Bulgarian University Press. (Full text)

Publication year	2011
Document type:	In book
Publication status:	Published
External URL:
Categories:
Keywords:	causal modelsinformation search.