Overgeneralized autobiographical memory and future thinking in combat veterans with posttraumatic stress disorder

Abstract

Background

Studies show that individuals with Posttraumatic Stress Disorder (PTSD) tend to recall autobiographical memories with decreased episodic specificity. A growing body of research has demonstrated that the mechanisms involved in recalling autobiographical memories overlap considerably with those involved in imagining the future. Although shared autobiographical deficits in remembering the past and imagining the future have been observed in other clinical populations, this has yet to be examined in PTSD. This study examined whether, compared to combat trauma-exposed individuals without PTSD, those with combat-related PTSD would be more likely to generate overgeneralized autobiographical memories and imagined future events.

Method

Operation Enduring/Iraqi Freedom (OEF/OIF) veterans with and without PTSD were presented with neutral word cues and were instructed to generate memories or imagine future autobiographical events. Responses were digitally recorded and were coded for level of episodic specificity and content related to combat trauma.

Results

Individuals with PTSD were more likely to generate overgeneral autobiographical memories and future events than individuals without PTSD, and were more likely to incorporate content associated with combat when remembering the past or thinking about the future.

Limitation

Limitations of the study include a cross-sectional design, precluding causality; the lack of a non-trauma exposed group, relatively small sample, and almost all-male gender of participants, limiting the generalizability to other populations.

Conclusion

These findings suggest that individuals with PTSD show similar deficits when generating personal past and future events, which may represent a previously unexamined mechanism involved in the maintenance of PTSD symptoms.

Introduction

Considerable progress has been made toward identifying cognitive (Conway & Pleydell-Pearce, 2000), neural (Rubin, 2005), and functional (Bluck, 2009) properties of autobiographical memory across a wide range of populations. The study of autobiographical memory has become especially pertinent to psychopathology research, and in particular Posttraumatic Stress Disorder (PTSD), a psychiatric disorder characterized by unwanted distressing autobiographical memories (American Psychiatric Association [DSM-IV-TR], 2000). Some have even argued that PTSD is a response to a traumatic autobiographical memory rather than to a traumatic event per se (Rubin, Berntsen, & Bohni, 2008).

One consistent finding is that individuals with PTSD retrieve less specific autobiographical memories than those without PTSD, producing what is often referred to as “overgeneralized memories” (for a review see Moore & Zoellner, 2007). That is, when individuals with PTSD are provided with a word cue and are asked to recall a personal memory, they tend to retrieve categorical information (e.g. repeated time points, events that extend over more than one day) rather than unique moments within a distinct event. Overgeneralized autobiographical memory has been documented in Vietnam veterans with PTSD (McNally et al., 1995, McNally et al., 1994), cancer survivors (Kangas, Henry, & Bryant, 2005) and injured individuals with acute stress disorder (Harvey, Bryant, & Dang, 1998). Significantly, it is not merely a consequence of trauma exposure (Moore & Zoellner, 2007), it appears to emerge independent of depression (McNally et al., 1995), and is not viewed as a marker of overall psychopathology, inasmuch as it is not related to other anxiety disorders (e.g. Wenzel et al., 2002, Wilhelm et al., 1997). Overgeneralized autobiographical memory has been implicated as an important mechanism underlying the pathogenesis of PTSD as it appears to contribute to the onset and maintenance of the disorder (Bryant et al., 2007, Harvey et al., 1998).

In this paper, we explore whether the overgeneralization observed in the autobiographical memories of those with PTSD may extend to other autobiographical temporal time points, namely the construal and imagination of future episodes. Converging research from clinical, neuroscientific, and basic behavioral studies demonstrate that the underlying structures and processes involved in generating autobiographical memories overlap considerably with those involved in imagining one’s autobiographical future. For example, recent brain imagining studies reveal that regions in the medial prefrontal cortex, posteromedial parietal cortex, and the medial temporal lobes (including regions in the parahippocampal cortex and hippocampus) are similarly engaged when individuals generate past and future personal events (e.g., Addis et al., 2007, Szpunar et al., 2007; for reviews see Schacter et al., 2008, Szpunar, 2010). Studies with amnesiac populations report that individuals with damage to the hippocampus have difficulty producing autobiographical memories and imagining future episodes (Klein et al., 2002, Tulving, 2002). Developmentally, the ability to recall past and imagine future events emerges and declines in parallel across the life-span (Addis, Wong, & Schacter, 2008; Atance & O’Neill, 2005).

In terms of overgeneralized autobiographical memories, studies with clinical populations have found that individuals who recall autobiographical memories with less episodic specificity show similar patterns when imagining the future. For example, Williams et al. (1996) found that compared to controls, suicidally depressed patients generate both personal memories and future events with less episodic specificity. Similar findings have also been observed among schizophrenic (D’Argembeau et al., 2008) and complicated grief patients (Maccullum & Bryant, 2011). Although overgeneralized autobiographical memory is well-documented in PTSD, studies have yet to examine whether they extend to imagined future events.

In addition to investigating overgenerality in PTSD, the present study examined the extent to which PTSD influenced the content of imagined future events. Schacter and Addis (2007) hypothesize that because autobiographical memories are reconstructed, they possess a flexibility that enables the construction and simulation of imagined future events (constructive episodic simulation hypothesis). Moreover, it has been demonstrated that predictions about the future are based on the ease with which past events are recalled (Tversky & Kahneman, 1973). Therefore episodic simulations are likely to draw on the contents of those memories most accessible at that time. In light of these findings, Schacter, Addis, and Buckner (2008) predict that because anxious individuals show increased memory accessibility for negative past events, they are likely to construct negatively biased simulations for the future. Such biases in bidirectional ‘time travel’ indicate that people with PTSD are more likely to recall and imagine trauma-related events.

In sum, compared to combat veterans without PTSD, we predicted that (1) PTSD participants would retrieve autobiographical memories and imagine future events with less episodic specificity and (2) PTSD participants would include greater combat-related trauma content in their autobiographical memories and imagined future events.