Date of Graduation

5-2026

Document Type

Thesis

Degree Name

Bachelor of Arts in Psychology

Degree Level

Undergraduate

Department

Psychological Science

Advisor/Mentor

Dr. Michelle Ramey

Committee Member

Dr. James Lampinen

Second Committee Member

Dr. Song Yang

Third Committee Member

Sophie Brady

Fourth Committee Member

Dr. Michelle Ramey

Abstract

Schemas are a type of semantic knowledge that are formed from generalizations about the world. Schemas are essentially expectations about how we expect the world to be, and they are helpful for understanding and categorizing information we encounter. Prediction errors, which occur when our predictions are violated, are caused by events that are inconsistent with our schemas. Current findings on prediction error’s effects on memory are contradictory, with some studies stating that the presence of a prediction error enhances memory for a context, while other research disagrees. A popular theory proposes that prediction error causes a boost in memory for the context surrounding where the prediction error occurs, but results are conflicting. This paper explores the reason why different studies on this question are finding different results and examines a potential solution to these contradictory findings. Specifically, this paper will explore the extent to which prediction error influences memory, including memory for the surrounding context. The stimuli used in our study were scenes of household rooms that each contained a target object inside of them, which were objects native to that area (i.e., pan in kitchen, coffee cup in living room). Participants first studied the scenes, then later completed a memory test for the scenes. This study manipulated two variables: the schema-congruency of the target object in the scene (whether it caused a prediction error) and the presence of that object during retrieval (whether the object was presented during the memory test for the scene). These factors were predicted to explain the disparity in the previous research. The results demonstrate that prediction error did not improve memory for target absent conditions (when the object is not presented during the memory test) and in fact weakened memory. For the target present condition, prediction error did improve memory. This research demonstrates that the prediction error enhancement for memory is only evident when the prediction error is re-presented in the stimuli at test, and in fact prediction error worsens memory when not re-presented. This suggests that, when participants are studying the scenes, the prediction error likely draws attention to itself causing the contextual surroundings of the scene to receive less attention. This finding can help to understand why prior research conflicts and has implications for better understanding eyewitness testimony with respect to lineup identification.

Keywords

Schemas; Prediction error; Memory; U-Shape Model

Download

Share

COinS