Assessing the stability of thematic and taxonomic preferences across explicit and implicit measures

Assessments of similarity between objects has shown to draw upon both taxonomic and thematic properties. While cross-task preferences have been demonstrated (Mirman & Graziano, 2012), the current experiment aimed to examine the reliability of such preferences across an extended range of explicit and implicit measures of similarity. In a within-subjects design, 50 participants completed three established measures assessing preferences for taxonomic or thematic relations; a free sort task, a triad task and the Visual World Paradigm, with a further implicit measure developed based upon the single category Implicit Association Task. Preferences were calculated on the basis of choices made on the sorting and triad task, competitor viewing time on the VWP, and response time on the IAT. Across all measures, consistent preferences were not found. Furthermore, no significant correlations were found between the magnitude of preferences for the four measures including no correlations between the two explicit or the two implicit measures. In contrast to previous research demonstrating reliable cross-task preferences, performance on the tasks used here argue against stable individual differences in taxonomic and thematic processing and suggest that, for most people, the use of each processing pathway is flexible and determined by both context and goals.