This study describes our method of rescripting early memories linked to recurrent negative imagery in patients with social phobia. Further, we looked at the effect of this procedure on the distress, vividness and meaning of patients’ early memories linked to their current negative imagery. We also looked at the effect on social cognitions and other symptoms of social phobia. One session of memory rescripting significantly reduced the distress and vividness of patients’ traumatic social memories as well as their meaning within session and 1 week later. Further, the procedure led to a significant within session reduction in the distress and vividness of patients’ negative images. One week later, there was a trend for patients to rate their images as less distressing, but not less vivid. Finally, on a measure of social cognitions (the SCQ), patients reported having significantly fewer negative social concerns in the previous week (e.g., “I am weird/different;” “people will stare at me”) and reported believing them much less strongly. They also showed significant improvement on the SPWSS, which measures the severity of components of social phobia such as, anxious affect, avoidance, self-focused attention, and anticipatory and postevent processing.
As expected, our sample of patients reported high levels of vividness and distress in relation to their images and memories. This is interesting because vividness has been a cornerstone variable in studies of mental imagery. Yet little is known about its relationship to valence. Experimental studies in non-clinical populations indicate that these variables may be related. For example, Bywaters, Andrade, and Turpin (2004)
demonstrated that images of pictures rated as extremely valenced (positively or negatively) were also rated as being more vivid. Several studies in clinical populations suggest high levels of vividness and distress for images associated with current concerns. Pratt, Cooper, and Hackmann (2004)
showed that spontaneous and self-generated spider images in people with high spider anxiety were rated as more vivid and distressing than those of participants with low spider anxiety, whilst butterfly images in the two groups did not differ. Studies of body dysmorphic disorder (Osman, Cooper, & Hackmann, 2004
) and bulimia nervosa (Somerville, Cooper, & Hackmann, this issue
) indicated that whilst images of the self in these disorders were not more frequent than in control groups they were more vivid and distressing. Finally, Hunt et al. (2006)
report that blind observer ratings of horror and distress associated with the content of phobic imagery were strongly correlated with participants’ own ratings of the vividness of these same images.
In this study, vividness and distress of both images and memories fell following the intervention. In two studies of intrusive images of memories in PTSD, ratings of vividness, distress, frequency and sense of “nowness” fell steadily in synchrony over several weeks. This followed the start of reliving, restructuring and rescripting of the traumatic memories during treatment. There had been no changes in these variables during the period of assessment (Hackmann et al., 2004
; Speckens, Hackmann, Ehlers, & Clark, 2006
). The stepped improvement as the weeks went by in these two studies suggests that there may have been further and possibly more stable improvements in the present study on social phobia, had there been more sessions devoted to work on the disturbing memories.
This preliminary, uncontrolled study, like the results of a subsequent controlled study reported elsewhere (Wild et al., in press
), supports the view that as well as attending to current symptoms of social phobia it may be worthwhile to look at the possible origins of the problem, and to target key memories that carry toxic meanings for the patient, as they appear to provide input to recurrent images of the self in this disorder. This could be important as the images themselves have been shown to trigger anxiety and safety behaviours, and may thus play a role in maintenance (e.g., Hirsch et al., 2004
The efficacy of imagery rescripting in this study is consistent with a growing body of research that has used the procedure in cognitive-behavioural treatments for different disorders, including bulimia nervosa (i.e., Cooper et al., 2007
; Ohanian, 2001
), borderline personality disorder (i.e., Giesen-Bloo et al., 2006
), and PTSD (e.g., Smucker & Neiderdee, 1995
). Rescripting the identified social phobia related traumatic memories in this study led to dramatic reductions in patients’ beliefs associated with the memories and images, as well as in social phobia cognitions and other symptoms.
There are several aspects of our intervention that may have been helpful. The procedure described here incorporates components of reliving, verbal restructuring and rescripting of the event in imagery. These elements are also components of the protocol for a recently reported effective treatment for PTSD (Ehlers et al., 2005
). Imaginal rescripting has also been utilised in isolation, in cases of PTSD where patients had not responded to prolonged exposure (Grunert et al., 2003
). Of particular interest are the results of studies like that carried out by Arntz et al. (this issue)
, which compared imaginal exposure with and without rescripting of memories, and furnishes us with a glimpse of ways in which the effectiveness of various components of the treatment may be examined.
In our study the procedure included asking the patient to vividly evoke the memory three times in total, describing it from various perspectives in the present tense. This repetition is similar to reliving in various CBT programs for PTSD, in which patients are asked to repeatedly go through the traumatic memory. It has been suggested that reliving is effective in PTSD because it allows for reflection and spontaneous cognitive change, as well as habituation (Hackmann, 2005
). Foa and Rothbaum (1998)
have also observed that reliving allows the patient to come to the possible conclusion that the original negative evaluations were not consistent with all the available evidence. Thus, the reliving aspect of our intervention may have helped patients to come to new conclusions about the earlier event and its current implications.
The verbal restructuring component deliberately prompted patients to consider evidence for and against the meaning of their early memory and image. This helped them to reevaluate how they perceived the event at the time (e.g., were they generally perceived as inferior or unlikable or did it just feel that way?), and how accurate the meaning was in the present (e.g., how do other people actually view them now?). Some patients realised that they were never really worthy of rejection, it just felt like that. Others appreciated that they were rejected in the past, but that this was no longer the case: the meaning of the memory was out of date. In addition, verbal restructuring helped patients to take a less personal view of the event, to see it as meaning (for example) something about the bullies, rather than about themselves.
Finally the patients engaged in imagery rescripting of the early memory. This typically involved imagining entering the scene as an empowered adult who could take effective action, and then subsequently experiencing compassion and nurturance as the younger person. Evoking this vivid imagery from new perspectives is likely to have been another important ingredient in the intervention. Patients may have seen more clearly and felt more strongly that their younger selves were acceptable and deserving of comfort rather than criticism, and that the event had implications about others (e.g., the bullies or critical adults) rather than themselves. Further, some may also have interpreted other people's intentions as more benign than they had previously thought. Hackmann (2005)
suggests that experiencing empowerment and compassion in imagery helps patients to reevaluate themselves and the behaviour of others, which can soften the sense of threat.
Overall the process of reviewing and rescripting the traumatic memory reduced the severity of social phobia as reflected in the mean item rating of the SPWSS and also the total belief and frequency scores on the SCQ. It is possible that the procedure led patients to update the meaning of the earlier event and to no longer see it as something toxic with future implications. This may have led them to reevaluate other social concerns which may account for why they had fewer in the week following rescripting and why they believed them less strongly. The SPWSS measures patients’ perceived severity of social anxiety in the previous week as well as their avoidance, self-focused attention, and anticipatory and postevent processing. Regarding change on this measure, it is possible that having learned to see the event differently and to master the event in imagery (in that they intervened, and then soothed the younger self) may have encouraged patients to engage in more social situations, and to do so differently: attending to them more fully once in them, as well as worrying about them less before and afterwards. This study did not look at the direct impact of rescripting on specific processes (e.g., self-focused attention, avoidance, pre- and postevent processing), or the effects on mood, and future research could do this to determine how rescripting mediates change in social phobia symptoms.
This study is a small, exploratory study with several limitations. Firstly, there was no control session. Thus, it is unclear whether imagery rescripting led to change or whether discussion of the early memory without rescripting would have led to change by itself. However, in their recent study, Wild et al. (in press)
, found that compared to a control session (involving discussion of the memories), imagery rescripting led to significantly greater change in the distress and vividness of patients’ images and memories as well as symptoms of social phobia, both within session and one week later. These results were similar to those in a study of rescripting in bulimia nervosa (Cooper et al., 2007
Secondly, the sample size is small, particularly after 1 week, in this naturalistic, exploratory study. However, despite this, rescripting did lead to significant within session change, and much significant change when tested after 1 week, which is encouraging, and may stimulate others to do more ambitious studies. Thirdly the intervention was brief, and it might be interesting to try a longer intervention and examine the stability of the results over a longer period of time, as in the study of depressed patients with intrusive memories (Wheatley et al., this issue
The design of our study does not allow for analysis of the three separate components of our procedure, i.e., reliving, verbal restructuring and imagery rescripting. Thus, it is not possible to conclude which component is the most important or effective. Future studies could compare these components separately and in combination. For example, does the largest shift come from evoking the memory as in the reliving component? Or, is it updating the memory via restructuring and rescripting that shifts beliefs and emotion most effectively? Does verbal restructuring enhance the chance of providing the patient with an alternative and less toxic view of the past? Or, is it sufficient to just prompt for change in the memory image, by asking the patient to view it from another perspective? One could also examine whether interventions that bring about belief change are necessary for symptom reduction, or whether creating other kinds of competing images that contain some of the old elements, but with new features and positive affect might be sufficient to decrease symptomatology (Brewin, 2006
Finally, one could explore whether identifying and transforming memories that the patient sees as supporting negative beliefs could enhance collaboration and response to other aspects of treatment. Imagery rescripting provides a powerful emotional experience for patients, enhancing compassion for the younger self and reducing the sense of helplessness. It often seems to bring about greater willingness to experiment with dropping safety behaviours and reducing avoidance, as the patient has made the meta-cognitive shift towards seeing the distressing self image as being the product of their own mind, rather than mirroring reality. As one patient most eloquently expressed this shift:
It is as if I have been looking at myself in a mirror for years, and it is only now that I see that it was the mirror that was flawed and not me.