Measurement of mentalizing: A systematic review and development of a construct validity framework.

Mentalizing-holding in mind, appraising and reappraising mental states-is evidenced across research disciplines as a key psychological capacity of developmental and clinical importance. Due to this interdisciplinary context, there are a broad range of mentalizing measures available. The primary aim of this review was to systematically review existing mentalizing measures. MEDLINE, PSYCHINFO, HAPI and PSYCHTESTS were searched to identify English-language, text-based, general adult population mentalizing measures. Mentalizing is established as a multidimensional concept that varies across relational contexts. However, explicit dimensional definition and systematic contextualisation of mentalizing measures is rare. The secondary aim of the review was to develop and pilot a novel framework for classifying the mentalizing dimensions and relational context operationalised by each measure. Following reference to the literature and expert discussion, a dimensional construct framework of mentalizing with 18 components was proposed and applied to existing instruments. 57 text-based measures were identified by the review. Inter-rater agreement using the novel framework was 75%. No measures completely overlapped on dimensions and relational context. Relational contexts of measures were largely mentalizing the self (52) and/or hypothetical unspecified others (33). The review demonstrates that measures of mentalizing can be compared on the basis of dimensions operationalised and relational context, increasing utility in measure selection. Divergent construct validity highlights a need for careful selection of appropriate measures on the basis of mentalizing components of interest. Increased clarity of construct validity could be used to illuminate conflicting findings in the mentalizing literature and to advance the development of new assessment instruments.