Quantifying cognitive complexity: Evidence from a reasoning task

Isabel Arend, Roberto Colom, Juan Botella, Maria José Contreras, Victor Rubio, José Santacreu

Research output: Contribution to journalArticlepeer-review

28 Scopus citations

Abstract

There are some doubts about the nature of cognitive complexity. It has been proposed that the loadings on the first un-rotated factor can be taken as a way to quantify the cognitive complexity of a given task. However, the evidence is sparse. The present study tests 1968 participants in a computerized task that comprises linear syllogisms or three-term series problems. The correlation matrix is submitted to a factor analysis. The first un-rotated factor is taken as the vector of cognitive complexity. The vector of task difficulty was obtained after the proportion of participants that failed each syllogism. In addition to task empirical difficulty, three information processing models are taken as predictors of cognitive complexity. Then, regression analyses were carried out to predict cognitive complexity from the information processing (IP) models and task difficulty. Results show that the IP models and task difficulty predict cognitive complexity defined by the loadings on the first un-rotated factor. Therefore, it is concluded that those loadings can be taken as a way to quantify cognitive complexity.

Original languageEnglish
Pages (from-to)659-669
Number of pages11
JournalPersonality and Individual Differences
Volume35
Issue number3
DOIs
StatePublished - 1 Aug 2003
Externally publishedYes

Keywords

  • Cognitive complexity
  • Factor loadings
  • Information processing models
  • Linear syllogisms
  • Reasoning

ASJC Scopus subject areas

  • General Psychology

Fingerprint

Dive into the research topics of 'Quantifying cognitive complexity: Evidence from a reasoning task'. Together they form a unique fingerprint.

Cite this