Background: Childhood adversity, dissociation and adult attachment have all been implicated in the development of hallucinations or ‘voice-hearing’. Testing psychological models in relation to subclinical phenomena, such as proneness to hallucinations in non-clinical samples, provides a convenient methodology to develop understanding of the processes and mechanisms underlying clinical symptoms. Aims: This paper investigates the relative contribution of childhood adversity, dissociation and adult attachment in explaining hallucination proneness in a non-clinical sample. Methods: Students and staff with no previous contact with secondary care at the University of Manchester were recruited. Participants completed a series of self-report measures: the Launay‒Slade Hallucination Scale (LSHS), the Relationship Scale Questionnaire (RSQ), the Childhood Trauma Questionnaire (CTQ), the Dissociative Experiences Schedule (DES II) and the Positive and Negative Affect Schedule (PANAS). Results: As hypothesized, insecure attachment, childhood adversity and dissociative symptoms were correlated with hallucination proneness. Multiple regression analysis, controlling for confounds of age and negative affect, indicated that the RSQ, CTQ and DES II predicted hallucination proneness. Only DES II and RSQ avoidant attachment were significant independent predictors in the final model. Conclusions: This study provides further evidence to support the idea that attachment and dissociation are important psychological mechanisms involved in voice-hearing proneness. Further testing is required with a clinical population.