ABSTRACT
It is unclear whether witness mental illness and special measures used with witnesses in court impacts juror decision-making. Participants (N = 204) from the general public and student population completed a measure assessing attitudes towards mental illness before reading a mock trial vignette where witness mental illness (depression, schizophrenia, no mental illness) and the special measure used in court (screen, intermediary, no special measure) were manipulated. Participants were then instructed to formulate judgements about the witness testimony provided (reliability, competency, credibility) and their likelihood of finding the defendant guilty. The findings showed that witnesses with depression were perceived as significantly more competent than witnesses with schizophrenia, or with no mental illness. Witnesses with depression were also perceived as significantly more competent than witnesses with schizophrenia when a screen was used in court. There was however no difference in competency ratings for witnesses with depression versus those with schizophrenia when no special measure was used, or when an intermediary was used, although witnesses with depression were still viewed more favourably overall. These findings suggest that some awareness of these biases is needed in court. Improving clarity about why special measures are used in court might also go some way towards addressing this issue.
Acknowledgements
The authors would like to thank Alex Jones for discussion regarding the statistical analyses
Disclosure statement
No potential conflict of interest was reported by the authors.
Data availability statement
Data supporting the findings of this study have been submitted to this Journal.
Notes
1 To calculate mean scores for the schizophrenia and depression conditions, data was collapsed across all three special measure conditions (screen, intermediary, no special measure).
2 The authors also ran the full analyses without the inclusion of the ‘attitudes towards mental illness’ covariate. All findings returned from this analysis were identical to those outlined here where the covariate was included in the analyses.