Does insecure attachment lead to psychosis via dissociation? A systematic review of the literature.

PURPOSE: Insecure attachment may constitute a vulnerability factor for psychosis, and dissociation may be a key mechanism in the development of auditory hallucinations specifically. While there is good evidence for the role of these processes in isolation, it is unclear whether dissociation accounts for the association between insecure attachment and psychosis. This systematic review takes a theory-driven approach to examine proposed causal relationships across the clinical and nonclinical literature. METHODS: We searched five databases (PubMeD, Web of Science, PsycINFO, CINAHL and ETHOS) for published and unpublished research examining attachment, dissociation and psychosis. Two independent reviewers extracted the data and assessed the quality of all included studies. RESULTS: We identified 242 potential articles and included 13 in the final review (2096 participants). We found that (1) disorganised attachment was consistently associated with dissociation and inconsistently associated with voices and paranoia, (2) dissociation was associated with voices and paranoia, and these links were stronger in clinical samples, and (3) dissociation played a role in the impact of insecure attachment on voice hearing and paranoia in clinical groups. CONCLUSIONS: This is the first review to synthesise the research examining attachment, dissociation, and psychosis. The evidence is consistent with proposed causal hypotheses and raises conceptual and measurement issues, for example, the need to clarify the relative contributions of different insecure attachment styles, and utilise behavioural/observational measures to strengthen study designs. Most importantly, we need experimental and longitudinal studies to confirm causal links and targets for treatment.

Does disorganised attachment lead to auditory hallucinations via dissociation? An experimental study with an analogue sample.

OBJECTIVES: Auditory hallucinations (such as hearing voices) are common in clinical and non-clinical populations. Many people who hear voices also report early adversity and have an insecure attachment style. Current cognitive models suggest that dissociation mediates an association between disorganised attachment and auditory hallucinations, but this has not been tested experimentally. DESIGN: We recruited a non-clinical analogue sample highly predisposed to auditory hallucinations and utilised an experimental design to examine the impact of disorganised attachment imagery on hallucinatory experiences, and whether dissociation mediates an expected association. METHODS: Participants completed self-report measures of state auditory hallucinations and dissociation before and after random allocation to secure or disorganised attachment conditions. RESULTS: Attachment imagery did not affect auditory hallucinations. Both secure and disorganised attachment conditions increased state dissociation. Secure attachment imagery reduced paranoia, but state dissociation did not mediate this effect. An exploratory analysis found that trait dissociation fully accounted for the association between trait-disorganised attachment and hallucinatory experience while controlling for paranoia. CONCLUSIONS: Secure attachment imagery reduces paranoia but not auditory hallucinations and the impact on paranoia is not mediated by dissociation. Secure attachment imagery may be useful in reducing fears and distress associated with voices, rather than the frequency or severity of hallucinations. Disorganised attachment may increase hallucinatory experiences for people vulnerable to dissociation. Trait dissociation should be assessed in clinical settings and addressed where indicated as a means of targeting vulnerability to distressing voices.