Chapter

Spatial Cognition VI. Learning, Reasoning, and Talking about Space

Volume 5248 of the series Lecture Notes in Computer Science pp 378-393

Modeling Cross-Cultural Performance on the Visual Oddity Task

  • Andrew LovettAffiliated withQualitative Reasoning Group, Northwestern University
  • , Kate LockwoodAffiliated withQualitative Reasoning Group, Northwestern University
  • , Kenneth ForbusAffiliated withQualitative Reasoning Group, Northwestern University

* Final gross prices may vary according to local VAT.

Get Access

Abstract

Cognitive simulation offers a means of more closely examining the reasons for behavior found in psychological studies. This paper describes a computational model of the visual oddity task, in which individuals are shown six images and asked to pick the one that doesn’t belong. We show that the model can match performance by participants from two cultures: Americans and the Mundurukú. We use ablation experiments on the model to provide evidence as to what factors might help explain differences in performance by the members of the two cultures.

Keywords

Qualitative representation analogy cognitive modeling oddity task