Mental imagery and bipolar disorders: Introducing scope for psychological treatment development?

Introduction

Bipolar disorder is a serious and disabling disorder with high risks of recurrence and completed suicide. The disorder exacts high direct and indirect costs to the patients, caregivers and society. While pharmacotherapy remains the mainstay of treatment, there is emerging evidence that cognitive behavioural therapy (CBT) could help provide an adjunctive treatment (Goodwin & Consensus Group of the British Association for Psychopharmacology, 2009). However, CBT may currently only be of value to a less chronic and recurrent sub-group of patients with bipolar disorders (Thase, Kingdon, & Turkington, 2014). There is thus a need to develop additional strategies for existing CBT, so as to enhance its efficacy in preventing mood recurrences in patients with bipolar disorders of various levels of severity and chronicity. Moreover for types of bipolar disorder where medication has weaker evidence, a psychological approach would be a particularly important avenue for future research (Thase et al., 2014), as it would for particular population groups (young people) and for areas where medication can be counter-indicated such as anxiety in bipolar disorder (Stratford, Blackwell, Di Simplicio, Cooper, & Holmes, 2014).

Mental imagery amplifies emotion, enhances conviction and promotes behaviour

Mental images are characterised by their subjective resemblance to sensory perception, as if, for example, ‘seeing with the mind’s eye’. They can involve multiple sensory modalities, including bodily sensations and feelings. In experiments comparing mental imagery versus verbal-semantic instructions when listening to negatively valenced scenarios, imagery led to a significantly greater increase in state anxiety than did verbal processing. A similar pattern of increased emotion for imagery relative to verbal processing instructions was also found for positive material (e.g. Holmes, Mathews, Dalgleish, & Mackintosh, 2006). In another experimental paradigm, compared to verbal thought, using mental imagery again led to a more powerful impact on emotion – and further the images were rated as being more ‘real’ (Mathews, Ridgeway, & Holmes, 2013).

Besides having a powerful effect on emotion (as reviewed elsewhere (Holmes & Mathews, 2010), mental imagery can also enhance the subjective conviction about the possibility of the occurrence of an imagined event. For example, it has been shown that imagining the symptoms of a disease increased the perceived likelihood of contracting the disease in the future. Imagining one’s own future behaviour such as voting increased the chance of enacting that behaviour subsequently. In summary, mental imagery appears to elicit powerful emotion, leads to a higher conviction of being about to experience an imagined event, and a greater likelihood of acting on the imagined event (Holmes & Mathews, 2010). As such, one could postulate that people experiencing vivid images would report having strong emotions, have difficulty distinguishing their inner emotional images from external reality, and be more inclined to carry out the acts depicted in their images than those not experiencing, or having less vivid, images.

Mental imagery and psychiatric disorders

Recently, there has been a call to consider mental imagery more actively in psychiatry – in both assessment and treatment (Di Simplicio, McInerney, Goodwin, Attenburrow, & Holmes, 2012). There is increasing evidence that mental imagery may be involved in triggering and maintaining mental disorders. For example, intrusive (image-based) memories or ‘flashbacks’ comprise the hallmark symptom of post-traumatic stress disorder (PTSD; American Psychiatric Association, 2013). Intrusive memories comprise vivid, sensory-perceptual emotional memories of distinct moments of a traumatic event. They intrude involuntarily into the mind, in contrast to voluntary episodic memory characterised by deliberate recall. They display a distinct neural pattern at encoding in experimental studies (Bourne, Mackay, & Holmes, 2013). Because of the powerful impact of these intrusive memories, patients experiencing them feel as if they are re-experiencing the trauma (feelings of ‘now-ness’ and ‘real-ness’) along with associated emotional states such as anxiety or disgust.

In addition to PTSD, intrusive and involuntary images have also been reported by patients suffering from agoraphobia, social phobia, specific phobias, obsessive–compulsive disorder, health anxiety and body dysmorphic disorders, etc. (Holmes & Mathews, 2010). Similar to the intrusive memories in PTSD, intrusive images in anxiety disorders are often linked to memories of adverse events that occurred in childhood or around the time of onset of the disorder. As well as past-oriented traumatic images, prospective (future-oriented) intrusive imagery has been reported. For example, patients with agoraphobia might ‘see’ images of impending social isolation and embarrassment (e.g. seeing oneself from the third-person perspective of having collapsed in a public area and being jeered at by a group of indifferent on-lookers). Patients with health anxiety may report prospective images of witnessing themselves being tortured and disfigured by the diseases that they worried about having contracted.

In addition to anxiety disorders, there are high rates of distressing and intrusive image–based memories of the past in depression. The content of such mental imagery is usually related to past interpersonal difficulties and personal loss, which reflects key negative themes that trigger and maintain depression. Furthermore, compared to non-depressed controls, patients with depression reported experiencing more negative prospective images, greater emotional impact of their prospective imagery, as well as impoverished positive imagery vividness (Morina, Deeprose, Pusowski, Schmid, & Holmes, 2011). A deficit of positive future-oriented images may contribute to a lack of optimism and a sense of hopelessness in depression (Blackwell et al., 2013).

Mental imagery susceptibility and bipolar disorders

Holmes, Geddes, Colom, and Goodwin (2008) proposed that mental imagery may play an important role in patients with bipolar disorders. For example, as positive images may amplify positive emotion and represent a positive goal to be attained, people experiencing positive images might thereby become elated, physiologically aroused and geared towards action in approaching the imagined positive goal. Such characteristics are reminiscent of elated mood, increased goal-directed activities and excitement – three key diagnostic criteria of manic episode in bipolar disorders in the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) classification (American Psychiatric Association, 2013). Therefore, repetitive or intense ‘overly’ positive imagery may contribute to promoting manic states. Similarly, negative images may contribute to escalating depression and anxiety. If patients with bipolar disorder were susceptible to thinking in terms of images and experiencing strong emotions (positive and negative) in response to such images, this may also account partially for their frequent and extreme ups and downs in emotional states and sub-syndromal mood instability during inter-episode periods. However, this tentative hypothesis remains to be tested directly.

Interestingly and in early stage work, Holmes et al. (2011) found that euthymic patients with bipolar disorders reported higher levels of the general use of imagery in daily life (a trait propensity to use imagery) and also reported a greater emotional impact of prospective imagery than did healthy volunteers. Hales, Deeprose, Goodwin, and Holmes (2011) found that in the context of suicidality, patients with bipolar depression had higher levels of these two types of mental imagery susceptibility than did patients with unipolar depression; moreover, bipolar patients were more preoccupied with their suicidal imagery and found it more compelling than those with unipolar depression. In another study, patients with bipolar disorders reported experiencing more vivid, exciting and pleasurable ‘flash-forwards’ (future images) at times of positive mood compared to patients with unipolar depression (Ivins, Di Simplicio, Close, Goodwin, & Holmes, 2014). Higher imagery susceptibility was also found in people with high hypomanic experiences (measured on the Mood Disorders Questionnaire; Malik, Goodwin, Hoppitt, & Holmes, 2014; Ng, Burnett Heyes, McManus, Kennerley, & Holmes, 2015), something we are following up in a familial risk study. Thus, there is to date some preliminary support that people with a bipolar phenotype are more likely to visualise their thoughts and also report stronger emotional arousal to their own visual images than people without a bipolar phenotype.

Emotional valence of mental images and the polarity of mood switch in bipolar disorders

If patients with bipolar disorders are susceptible to mental images, what other imagery characteristics might determine the direction of extreme mood swings unique to these disorders? Gregory, Brewin, Mansell, and Donaldson (2010) reported that intrusive negative prospective images were found during the most recent depressive episode, while intrusive positive prospective images were found during the most recent hypomanic episode in patients with bipolar disorders. Our research group is now following up this issue with a prospective study investigating changes in the emotional valence of prospective images in a group of patients with acute mania.

Behavioural activation system, mental imagery and bipolar disorders

The behavioural approach system (BAS) is conceptualised as a brain system that regulates affect, cognition and action to support the pursuit of goals or reward incentives. High BAS function is believed to relate to increases in confidence, energy and approach activity, with clear parallels with mania symptoms. An expanded BAS model of bipolar disorder has proposed that people with vulnerability to bipolar disorders present an overly sensitive BAS that is hypersensitive to goal- and reward-relevant cues (Alloy et al., 2009). This hypersensitivity can lead to excessive BAS activation in response to goal-striving or goal-attaining life events – possibly leading to hypomanic or manic symptoms, and to excessive BAS deactivation in response to life events involving loss, failure or frustration of personal goals – possibly leading to depressive symptoms (Alloy et al., 2009).

Recent studies have found that onset of hypomanic or manic episodes in at-risk individuals were more likely to be specifically preceded by goal-attaining ‘activating’ life events than positive or negative life events in general, in contrast to depressive recurrence being more likely to be preceded by negative ‘deactivating’ life events (Johnson et al., 2008). Given the special characteristics of mental images, they might be perceived as real events that have happened or are about to happen. For example, a powerful positive prospective image might then function synonymously as a BAS-relevant activating event, promoting an excessive activation of the hypersensitive BAS in patients with bipolar disorders. Interestingly, levels of BAS drive and reward responsiveness in patients with bipolar dis-orders were predicted by pleasure ratings associated with their positive mental images (Ivins et al., 2014). Future studies may attempt to elucidate the relationships between mental imagery, BAS sensitivity and symptoms characteristic of bipolar disorders.

Imagery-based CBT and bipolar disorders

If patients with bipolar disorders have high propensity to experience arousing and powerful emotional mental imagery, a predominant reliance on verbal techniques (as in traditional CBT for depression) might not be able to fully grasp the phenomenology associated with bipolar affective states. As a result, attempts at verbal refutation might fail to ‘hit the nail on the head’ and even strain the therapeutic alliance. An ‘imagery-based micro-formulation’ would be valuable to help the patients to make sense of the relationships between their highly emotional mental images and the amplification of mood symptoms (Ng, Krans, & Holmes, 2013). Sharing such an imagery-based micro-formulation with the patients might make them feel more understood, foster a sense of relief and hope and thus help to strengthen the therapeutic alliance with the therapists.

People with high versus low bipolar traits had an increased frequency of intrusive images even after viewing traumatic film footage (Malik et al., 2014). Patients with bipolar disorders might thus be particularly susceptible to developing intrusive and distressing images during their casual and regular encounter of positive and negative events in daily life. Preventative/protective strategies might be developed.

In therapy, maladaptive and intrusive images can be collaboratively examined and addressed. More adaptive meanings and conclusions can be incorporated into the images during imagery re-scripting work. For example, when a patient with mania experiences a positive image of possessing a red Ferrari car and passing by strangers with admiration, he may become elated, excited and impulsively act upon this image by purchasing a red Ferrari car without considering the financial ability to pay for it. Imagery re-scripting work may include discussion of the pros and cons of impulsive buying/speeding and also modify the image into one where the patient graciously declines getting on such a red Ferrari car. Alternatively, the patient might also be guided to replace such a mental image of excessive positive emotional valence with a safe, balanced, compassionate image with, for example, inter-personal intimacy, affection and warmth or an image of contentment with what they are (in this example, replacing the grandiose image with a safe and affectionate image of strolling along the street with the friends that they love; Ng et al., 2013). The experiential component of mental imagery techniques (e.g. experiencing the calm and content emotion of a compassionate image) may represent an added value to the often unsatisfactory process of reasoning with patients with bipolar disorders around the interpersonal consequences and responsibilities associated with ‘hyper-positive’ cognitions. It remains to be investigated whether a preferential use of imagery-based approaches within CBT benefits bipolar patients, and future research is needed. However, given the need to develop better psychological approaches for patients with bipolar disorders so far, and for anxiety within bipolar in particular, this clinical population appears the one where imagery-focused CBT could hold great added value in terms of treatment innovation.

Conclusion

The current editorial has provided some preliminary evidence that mental imagery can be problematic in bipolar disorders, and that mental imagery more generally is associated with amplified emotion. Intrusive imagery may contribute towards fuelling the rapid and sudden mood swings characteristic of bipolar disorders. Furthermore, the emotional valence of these intrusive images might be associated with the polarity of mood switch in bipolar disorders. We have also proposed that imagery-based CBT may aid therapy development for bipolar disorders and would be a most interesting avenue to explore further. However, the current evidence is very limited and more research studies are warranted. In clinical practice, one can at least enquire about problematic imagery in the assessment of patients with bipolar disorders and help better understand the psychological landscape of this group.