Poster Session E, Monday, March 25, 2:30 – 4:30 pm, Pacific Concourse
The Influence of Reasoning Ability and Relational Cueing in Solving Relational Match-to-Sample Problems
Matthew J. Kmiecik1, Alex D. Martin1, Lauren M. Kim1, Rudy Perez1, David M. Martinez1, Ekarin E. Pongpipat1, Daniel C. Krawczyk1,2; 1The University of Texas at Dallas, 2The University of Texas Southwestern Medical Center
Performing relational comparisons is considered among the most intelligent cognitive capacities. Animal studies have demonstrated that chimpanzees and crows are capable of relational reasoning with second-order relationships (e.g., analogies); however, it has been hypothesized that only humans are capable of reasoning with third-order relationships. Exploring this process in humans, we were interested in understanding the relationship between relational reasoning performance in varying performers given different instructions. To investigate this, participants were presented relational match-to-sample problems that varied in relational complexity: perceptual, analogical, and system mappings. Problem types were presented with minimal instruction, no practice, and randomly across four blocks. The participants received feedback after each attempt and an instructional manipulation was given to facilitate relational thinking. Performance types (high vs. low) were defined to understand individual differences in reasoning ability. Results showed that performance decreased with greater relational complexity; however, the participants’ rate of learning the three relational structures depended on performance level and whether the instructional manipulation was given. Specifically, high performers more accurately solved analogical mappings and demonstrated a curvilinear learning rate across block, such that system mapping performance decreased before increasing across the experiment. High performers also had higher visual working memory scores and better verbal and scene analogy performance. The instructional manipulation was a much weaker effect than performance type but resulted in participants learning the analogical mapping structure at a faster rate. Humans’ ability to learn complex relational comparisons with minimal instruction/feedback is associated with working memory ability and improved by relational cueing.
Topic Area: THINKING: Reasoning