Publication Date


Document Type


First Advisor

Lilly, Michelle M.||Magliano, Joseph P.

Degree Name

Ph.D. (Doctor of Philosophy)

Legacy Department

Department of Psychology


Post-traumatic stress disorder; Cognitive psychology; Memory


Cognitive theories of PTSD assume that increases in peritraumatic anxiety and dissociation disrupts standard encoding processes, namely with an attentional bias toward perceptual information and away from conceptual information. However, this assumption currently lacks supporting evidence using moment-to-moment markers of encoding processes. This project explores if one specific encoding process, event segmentation, is impacted during a stressful event and if such an impact affects memory for the event. Event segmentation is an encoding process that involves chunking streams of continuous spatiotemporal information into discrete units. In studies measuring event segmentation during non-stressful events, findings suggest that segmentation ability positively predicts event memory. To date, no studies have assessed the impact of stress on event segmentation and subsequent memory. This project measured moment-to-moment event segmentation during a non-stressful film and a stressful film using a paradigm that requires participants to indicate boundaries between "meaningful units of activity" while watching each film. The aim of this project is to use the event segmentation task as a measurable, non-invasive, moment-to-moment marker of one encoding process during a stressful experience to provide an analog test of a core assumption within cognitive theories of PTSD. Specifically, unsystematic event segmentation was used as a marker for anxiety- and dissociation-impacted encoding based on the premise that any factor that affects attentional engagement in an experience is expected to decrease the systematicity of segmentation. Attentional engagement allows event segmentation to be influenced by (a) changes in the physical environment and (b) prior knowledge. This project was motivated by four hypotheses, all conceptually based on cognitive theories of PTSD and the empirical literature on event segmentation. Hypothesis 1 proposed that segmentation of stressful experiences is less systematic when compared against the segmentation of everyday, non-stressful experiences. Hypothesis 2 proposed that affective and dissociative responses within stressful experiences result in reduced systematicity of segmentation. Hypothesis 3 proposed that reduced segmentation systematicity results in poor voluntary memory (recall and recognition). Hypothesis 4 proposed that reduced systematicity of segmentation is a conduit (i.e., mediator) through which anxiety and dissociation have a negative effect on voluntary memory. The final sample included 73 mixed-gendered NIU students (predominantly freshmen) with no sexual assault histories or symptoms of PTSD. Planned analyses indicated no support for any of the four hypotheses. However, data analyses revealed numerous significant effects. Each significant effect was in the opposite direction of predictions. Most critically, and opposite of the direction predicted by Hypothesis 1, the stressful film resulted in higher segmentation scores than the non-stressful film. The remaining hypotheses assumed stressful experiences would diminish (not enhance) segmentation scores; therefore, it was not surprising to discover the remaining analyses produced results that trended (often significantly) in the opposite direction of predictions. Other unexpected significant effects included positive effects of anxiety and dissociation on segmentation systematicity (opposite of Hypothesis 2), a negative effect of segmentation systematicity on recognition of the stressful film (opposite of Hypothesis 3), and mediating effects of high (not low) segmentation systematicity on the negative relationship between anxiety and dissociation on event memory (opposite of Hypothesis 4). One possibility is that anxiety and dissociation may enhance segmentation systematicity in a manner that is consistent with theories of PTSD that emphasize increased sensitivity to perceptual features during traumatic events. Research on event segmentation has consistently showed that segmentation happens when there is perceptual change. The finding that segmentation was negatively correlated with memory performance for the stressful film but was positively correlated with memory performance for the non-stressful film lends credence to this interpretation. Given the unexpected nature of the findings, this interpretation should be considered with caution and would benefit from replication. Although the significant findings were unexpected, the results suggest measuring anxiety- and dissociation-induced change to at least one encoding process (i.e., event segmentation) is possible within the laboratory. This is the first known study to demonstrate that encoding during an analog traumatic event can be measured moment-to-moment and has interpretable implications on memory. The results of this study lend credence to future research exploring the relationship between segmentation and memory in the context of traumatic experiences.


Advisors: Michelle M. Lilly; Joseph P. Magliano.||Committee members: Christopher A. Kurby; Holly K. Orcutt; Alan Rosenbaum; David P. Valentiner.


162 pages




Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type