Pitting base rate driven heuristics against conditional reasoning in multivariate contingency assessment

Contingency assessment is a major module of adaptive cognition and a prominent topic of ecological rationality. Virtually all influential theories assume that contingency estimates between Y and X are inferred from subjective conditional probabilities of focal Y levels given different X levels, p (...

Full description

Saved in:
Bibliographic Details
Main Authors: Fiedler, Klaus (Author) , Kutzner, Florian (Author)
Format: Article (Journal)
Language:English
Published: 2024
In: Journal of experimental psychology. Learning, memory, and cognition
Year: 2024, Volume: 50, Issue: 12, Pages: 1901-1917
ISSN:1939-1285
DOI:10.1037/xlm0001407
Online Access:Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1037/xlm0001407
Verlag, lizenzpflichtig, Volltext: https://www.webofscience.com/api/gateway?GWVersion=2&SrcAuth=DOISource&SrcApp=WOS&KeyAID=10.1037%2Fxlm0001407&DestApp=DOI&SrcAppSID=EUW1ED0DBEpocO5YVuvwyTZwTxtPr&SrcJTitle=JOURNAL+OF+EXPERIMENTAL+PSYCHOLOGY-LEARNING+MEMORY+AND+COGNITION&DestDOIRegistrantName=American+Psychological+Association
Get full text
Author Notes:Klaus Fiedler, Florian Kutzner
Description
Summary:Contingency assessment is a major module of adaptive cognition and a prominent topic of ecological rationality. Virtually all influential theories assume that contingency estimates between Y and X are inferred from subjective conditional probabilities of focal Y levels given different X levels, p ( Y focal | X different levels ) . Yet, conditional probabilities are cognitively demanding, as Yfocal must be assessed separately for all levels of Xdifferent level. Pseudocontingencies (PCs) afford an alternative mechanism relying on base rates. In a PC, the more frequent level on one attribute appears contingent on the more frequent level on another attribute. When PCs are manipulated orthogonally to conditional probabilities, the former dominate the latter (Fiedler, 2010). PC dominance is shown in Experiments 1 and 1a to be particularly striking when a multivariate task setting calls for the assessment of all k<middle dot>(k - 1)/2 pairwise contingencies between k attributes. Experiment 2 shows that contingency judgments are dissociated from evaluative conditioning.
Item Description:Gesehen am 27.02.2025
Physical Description:Online Resource
ISSN:1939-1285
DOI:10.1037/xlm0001407