Semester of Graduation

Spring 2020

Department

Psychology

Document Type

Thesis

Abstract

The current research examined if representations in LTM necessarily aid working memory (WM) performance and if increasing interference in LTM limits facilitation from LTM on WM. Remembering multiple objects from the same semantic category can create interference in LTM that may decreases the accessibility of LTM representations during a WM task. In two experiments, participants completed an initial study phase in which objects were categorically (i.e., semantically) related or unrelated. Participants then completed a change detection task that included both previously studied and unstudied objects. In Experiment 1, an object changed into another object from a novel category. We found no evidence of facilitation from LTM on WM performance. However, eye-tracking analyses suggested evidence of facilitation on encoding, through shorter dwell times on studied objects compared to unstudied objects. Furthermore, we found no effect of semantic-relatedness on accuracy or dwell times. Change detection was similarly accurate when the studied objects were all from different categories, as when the studied objects were from the same categories, demonstrating that interference in LTM did not affect WM. In Experiment 2, we attempted to increase reliance on LTM representations by increasing difficulty and interference in the WM task. The changed object on the post-change array came from the same category as the pre-change object. Like in Experiment 1, we did not find any evidence of LTM facilitation on WM performance. Once again, we found shorter dwell times on studied objects compared to unstudied objects. Additionally, as in Experiment 1, there was no effect of interference in LTM on WM performance. Change detection accuracy was similar between semantically-related objects and semantically-unrelated objects. Overall, the results from the current study demonstrate that LTM representations were not used to improve WM performance, but may have been used to facilitate the encoding processes.

Committee Chair

Beck, Melissa

DOI

10.31390/gradschool_theses.5104

Share

COinS