Understanding self-report response bias in high-functioning autism

Assessment of self-report response bias, such as random responding, lack of insight/self-reflection, malingering or, conversely, socially desirable responding, should be integral parts of any forensic psychological assessment. However, many of the tools that are used are not designed or specifically validated for use for people who have High Functioning Autism (HFA). For the purposes of this thesis, HFA refers to those people with a diagnosis of Autism who have average to above average cognitive abilities. This thesis aims to address this gap in the evidence base. Chapter 1 provides a general introduction into the assessment of self-report response bias and the current challenges being faced in relation to using existing psychometric measures. Chapter 2 presents a systematic review of the literature on how self-report response bias has been assessed in forensic contexts in the UK over the last 10 years. The findings highlight that the UK seems to favour the Paulhus Deception Scales (PDS: Paulhus, 1998), which is different to the measures used in other parts of the world. Chapter 3 examines the psychometric properties of the PDS and considers its use in forensic contexts, with the focus on UK samples. Chapter 4 presents an empirical study that aimed to establish a normative data set for the PDS and Structured Inventory of Malingered Symptomatology (SIMS: Widows & Smith, 2005) with a High Functioning Autistic community adult sample. This study provides some early evidence that alternative cut-off scores should be used with this population, as part of a wider holistic assessment of response style and bias. Chapter 5 concludes the thesis with a summary of main findings and recommendations for future research and practice