Reflective function refers to the capacity to reflect on the mind of self and others in the context of the attachment relationship. Reflective function has been shown to be an important correlate of a variety of disorders, including borderline personality disorder. Current measures available to assess this construct in youths are either time- and/or labor-intensive or require experimental testing conditions; therefore, a self-report measure is needed for quick assessment of reflective function capacity in youths. The aim of the current study was to investigate whether the the newly adapted 46-item Reflective Function Questionnaire for Youths (RFQ-Y; Sharp et al., 2009) could provide a reliable and valid assessment of reflective function in adolescents. A sample of inpatient adolescents aged 12–17 (N = 146) was recruited. Adequate internal consistency was established for the RFQ-Y. Findings also demonstrated significant positive associations with a criterion measure of reflective function, with an experimental-based assessment of mentalization, and relations with empathy, supporting criterion and convergent validity of the RFQY. As expected, the RFQ-Y also showed strong negative relations with borderline personality disorder features as assessed by both self- and parent-report. In addition, adolescent patients who scored above clinical cut-off on self-reported BPD features were found to have significantly poorer reflective function compared to adolescents without BPD features. These findings support the notion that reflective function can be validly and reliably assessed in adolescent populations (Ha et al., 2013).
We conducted item response theory (IRT) analyses to refine the Reflective Function Questionnaire for Youth (RFQ-Y) Scale B. Data from a non-clinical sample of young people (n = 737; ages 18-25) was used to derive a shortened version of the RFQ-Y. Results were replicated in a clinical sample of inpatient adolescents (n = 467; ages 12-17), resulting in a five-item measure, thereafter named the RFQY-5. The RFQY-5 item set was then scrutinized for construct validity against the original 23-item RFQY item set in a randomly selected sample of 100 inpatient adolescents not included in the IRT replication, and 186 healthy adolescents drawn from the community. Results showed that the RFQY-5 performed similarly as the long version in terms of associations with criterion variables, and outperformed the longer version in discriminating between inpatient and community-dwelling adolescents who differed in their levels of borderline traits. The study provides evidence in support of the use of the RFQY-5 in research and clinical settings (Sharp et al., 2021).