Intrusive images and memories in major depression

Abstract

Individuals with current major depression were interviewed to investigate the prevalence of distressing intrusive mental imagery among depressed patients and study the phenomenology of these intrusions. Of the 39 currently depressed patients, 17 experienced some form of repetitive intrusive imagery (i.e., either an intrusive memory or image), with intrusive memories being more common than images. The intrusive imagery was experienced as highly uncontrollable and interfered significantly with patients’ everyday lives. The intrusions were experienced with a sense of ‘nowness’, as well as physical and emotional re-experiencing. Despite high levels of re-experiencing, levels of dissociation were very low. The intrusive imagery was in some patients part of a wider network of key defining autobiographical memories, consistent with the idea that it is likely to play a significant role in maintaining the patient's depressive mood. Interventions targeting these intrusions could potentially result in a positive shift in depressed mood.

Introduction

In recent years, there have been a number of studies showing that intrusive mental imagery is characteristic of a number of anxiety disorders including health anxiety (Wells & Hackmann, 1993), social phobia (Hackmann, Clark, & McManus, 2000; Hackmann, Surawy, & Clark, 1998), and agoraphobia (Day, Holmes, & Hackmann, 2004). Unlike autobiographical memories, which typically consist of detailed visual scenes explicitly linked to an original experience, these images typically consist of material that is abstracted from or represents an imaginal extension of an actual experience. As a result, the person may not necessarily be aware of the connection between the image and the original event (e.g., Wells & Hackmann, 1993). To date, however, there has been little investigation of this kind of intrusive imagery in major depression, a condition that is often comorbid with anxiety disorders.

Several studies have shown that patients with depression, like those with posttraumatic stress disorder (PTSD: Ehlers, Hackmann, & Michael, 2004; Ehlers et al., 2002; Hackmann, Ehlers, Speckens, & Clark, 2004; Reynolds & Brewin, 1998), frequently experience high levels of intrusive visual memories (Brewin, Hunter, Carroll, & Tata, 1996; Kuyken & Brewin, 1994; Reynolds & Brewin, 1999). Typically, depression severity is related to the frequency of intrusions and the level of avoidance of the memories (Brewin, Watson, McCarthy, Hyman, & Dayson, 1998a; Kuyken & Brewin, 1994). More recently, Starr and Moulds (2006) found the negative interpretations of intrusive memories to be correlated with depression severity and cognitive avoidance in a depressed undergraduate sample. Critically, the presence of these intrusions predicts the course of disorder even when initial symptoms are controlled for, suggesting that they are an important maintaining factor (Brewin, Reynolds, & Tata, 1999; Brewin, Watson, McCarthy, Hyman, & Dayson, 1998b).

These findings are consistent with the claims of many social psychologists (e.g., Srull & Wyer (1990), Srull & Wyer (1993)) that knowledge about the self does not just exist in a generalised, semantic form (c.f., trait knowledge about the worthlessness or unloveability of the self: Beck, 1976), but in the form of episodic memories of specific autobiographical events. Such events (e.g., a child being told by a parent that he or she was not wanted) may form ‘turning points’ (Pillemer, 1998) that help to define or provide evidence for the conclusions about the self at which depressed patients may have arrived. Apart from their theoretical importance in understanding the nature of negative self-representations in depression, the presence of intrusive memories (and possibly images) may also provide an opportunity to apply novel forms of therapy such as imagery rescripting (Smucker, Dancu, Foa, & Niederee, 1995) to depressed patients.

The content of intrusive memories in depression consists mainly of family illness and death, personal injury or assault, and interpersonal crises (Brewin et al., 1996). Whereas memories of death, illness, or injury to family members, and interpersonal problems, are more common in depression, memories of personal illness, injury, or assault are more common in PTSD (Reynolds & Brewin, 1999). Reynolds and Brewin (1999) found few qualitative or quantitative differences in the experience of intrusive memories between a matched sample of depressed and PTSD patients. For example, both groups reported some degree of reliving past experiences, as well as accompanying physical sensations. However, the measures were not detailed and may have been insufficiently sensitive to detect such differences that were present. For example, the measure of reliving did not specifically enquire about the ‘sense of nowness’ that is a characteristic of intrusive memories in PTSD (Hackmann et al., 2004). Only the prevalence of intrusive memories was found to be somewhat lower within the depressed group. A small number of patients in each group described their most frequent intrusion as a verbal or visual cognition concerning an event that had not actually happened but might conceivably do so, and to our knowledge this (Reynolds & Brewin, 1998) is the only report suggesting that depressed individuals may experience repetitive intrusive images (as opposed to memories).

Intrusive memories in PTSD, particularly those that contain an element of re-experiencing in the present, are considered to be underpinned at least partly by dissociative mechanisms (e.g., Bremner, Vermetten, Southwick, Krystal, & Charney, 1998), and PTSD is associated with elevated scores on the Dissociative Experiences Scale (Carlson & Putnam, 1993). In contrast, there has been little if any research into the relationship between dissociative tendencies and major depressive disorder. Starr and Moulds (2006) used the Response to Intrusions Questionnaire (Clohessy & Ehlers, 1999) to assess the extent to which mildly depressed individuals use dissociation to control their memories, but found no relationship between the negative meanings of intrusive memories and dissociation. The possible existence of elevated levels of dissociation in major depression, and the relationship of dissociation to intrusive images and memories, remains untested.

A final issue that has not previously been studied concerns the dynamic aspects of intrusive images and memories in depression. Is it the case, for example, that only those memories that can be elicited within a single research interview are likely to intrude in the future, or may they be replaced by other intrusive memories? If this is the case, can we expect the old and new intrusions to be related and, if so, in what way? These questions are relevant both to current research methods (i.e., are one-time assessments of intrusions adequate or likely to underestimate their importance) and to theoretical questions (i.e., do intrusions form an interconnected network of related representations?). A subset of the patients taking part in this study underwent imagery rescripting designed to reduce the frequency of their most prominent intrusion, and this created an opportunity to see whether these were replaced with alternative images and memories.

The current study had four main aims, all important for understanding in more detail the nature of these potential therapeutic targets. The first was to investigate whether depressed participants reported intrusive images, and to describe the frequency of such images relative to intrusive memories. Finding that images were common would raise questions about whether both functioned equally as maintaining factors and whether both were equally appropriate therapeutic targets. Finding that images were uncommon would add to knowledge about the psychological mechanisms that distinguish anxiety and depression. Patients with a primary diagnosis of PTSD, who would be expected to experience frequent intrusive memories, were excluded, but the existence of other comorbid anxiety disorders was recorded to see whether they would account for the existence of intrusive imagery.

Second, we wished to study the qualities and impact of intrusions, and any associated emotions, in greater detail than had been done before, in order to gather clues about possible differences between depression and PTSD, the other disorder in which intrusive memories are prominent. Qualities included vividness, sense of nowness, and emotional re-experiencing, and impact included level of interference, uncontrollability, and distress caused.

The third aim was to assess levels of chronic dissociative experiences in a depressed sample, and to investigate whether these experiences were related to the existence of intrusive images and memories. Again, this provides a potential point of contrast with PTSD and could point up differences in the phenomenology of the two disorders. It is possible that intrusions require different therapeutic techniques if they are or are not accompanied by dissociation.

The fourth aim was to report whether the content of intrusions changed with psychological treatment, and whether additional memories began to intrude that had not been evident at assessment. The appearance of additional memories would suggest that therapists employ additional careful assessments before concluding that therapy had been successful.