Controlled semantic cognition as measured by a pictorial mouse-tracking task in young adults

Student thesis: Master's ThesisMaster of Science by Research (MScR)

Abstract

Holding knowledge of concepts and their associated functions is essential for completing tasks in everyday life. How we manipulate that knowledge appropriately across contexts is known as semantic control. Research has begun to explore how semantic control declines in later life and is thought to be important for other cognitive skills that require flexible understanding of conceptual knowledge, such as producing coherent speech. However, past studies have focussed on a specific set of lexical tasks to measure semantic control, so little is known about the specificity of nonverbal semantic abilities. Two experiments utilised a novel, pictorial, two-alternative forced choice task to assess conflict in semantic decision-making. Performance was assessed via mouse-tracking for greater sensitivity to the decision-making process and was compared with coherence of speech and domain-general cognitive control. Both experiments were conducted with undergraduate students. Experiment 1 manipulated conflict using graded ratings of semantic relatedness of stimuli. As conflict increased, participants produced significantly larger geometric mouse-tracking scores, indicating they had deviated more towards the incorrect response. Mouse-tracked performance on the semantic matching task (SMT) predicted speech coherence, suggesting participants who were more successful in resolving conflict between concepts also produced more on-topic speech, in line with previous research. Experiment 2 manipulated conflict in the SMT through graded semantic similarity ratings, which focussed on taxonomic overlap of stimuli. Once again, there was a significant effect of conflict: participants produced mouse trajectories with greater curvature on trials with higher conflict. However, there was no significant relationship between semantic control and coherence of speech. Neither experiment showed any association between semantic control and verbal fluency, cognitive flexibility, or response inhibition. The discrepancy between the current results and previous research is discussed, with reference to different task requirements and the potential dissociation between taxonomic and thematic processing.
Date of Award22 Mar 2022
Original languageEnglish
Awarding Institution
  • University of Bristol
SupervisorJosie Briscoe (Supervisor)

Cite this

'