Background: Stigmatisation negatively affects those being stigmatised. Anti-stigma campaigns suggest that biogenetic explanations reduce attributions of personal responsibility towards those with a mental illness. This study focused on one experience: Voice-hearing, which is associated with mental illness, namely schizophrenia, but may be considered to be a common experience, as at least 10% of the general population will hear voices at some point in their lives. There are multiple theories on the mechanisms that underlie voice-hearing experiences. However, each theory has strengths and limitations and there is limited empirical evidence to support the view that one theory is superior to others. The primary aim of this study was to test whether attributions, emotional responses and behavioural intentions towards voice-hearers were sensitive to differential conceptualisations of voice-hearing. The secondary aims were (a) to test attribution theory and identify attributions and emotional mediators that influence behavioural intentions and (b) to explore the extent to which various socio-demographic characteristics influence attributions. Methodology: 1,004 members of the general public were recruited using online resources. They completed self-report questionnaires online. Attributions, emotional and behavioural responses were measured using an adapted attribution questionnaire, which also contained questions related to familiarity. Results: Analysis of Variance (ANOVAs) found that four out of five attributions were not sensitive to conceptualisations offered. The attribution of personal responsibility was sensitive to conceptualisations. Responses from the biogenetic conceptualisation were significantly lower that the responses from the cognitive conceptualisation. There was no statistical significance between conceptualisations and (a) emotional responses (b) behavioural intentions, or (c) behavioural outcomes. Mediation analyses indicated that there was an indirect relationship between four out of five attributions and behavioural intentions, which were mediated by emotional responses. Fear was the strongest predictor of coercive behaviours and was negatively associated with helping behaviour. The construct ‘pity’ appeared to be viewed as a negative emotion, as results for this construct were similar to those of anger and fear. Correlation analysis demonstrated a small-to-medium sized relationship between behavioural intentions and behavioural outcomes, which was added to an adapted attribution pathway model. Additional analyses found that stressful life experiences produced neither a direct nor an indirect relationship with behavioural intentions and appeared to be the least stigmatising of the six causal explanations. Further correlation analysis found that professional familiarity was associated with a reduction in attributions of dangerousness. However, there were only small associations between socio-demographic variables and attributions. 1314, RIP, UofN:4149204, UofL:06075465 Research Project Resubmission Page 9 of 193 similar to those of anger and fear. Correlation analysis demonstrated a small-to-medium sized relationship between behavioural intentions and behavioural outcomes, which was added to an adapted attribution pathway model. Additional analyses found that stressful life experiences produced neither a direct nor an indirect relationship with behavioural intentions and appeared to be the least stigmatising of the six causal explanations. Further correlation analysis found that professional familiarity was associated with a reduction in attributions of dangerousness. However, there were only small associations between socio-demographic variables and attributions. Conclusions: These finding are unique in that they extend research into stigmatising attributions towards voice-hearers and extend attribution theory. The biogenetic conceptualisations lowered attributions of personal responsibility, but mediation analysis suggests that attributions of personal responsibility were positively associated with increased social distance and coercive behavioural intentions. These results have clinical implications as the differences between medical and psychological approaches reflect the different conceptualisations offered to service-users within mental health services. The least stigmatising conceptualisation appeared to be Family Intervention. This was the only conceptualisation that utilised a bio-psycho-social model, Future research should focus on whether, in practice, this would be the least stigmatising conceptualisation used within mental health services. Further consideration should be given to its use in future anti-stigma campaigns.