The aim of the present research project will to explore whether eye-tracking, together with temporal and kinematic indicators, could improve the detection of subjects demonstrating faking-good behaviour when responding to personality questionnaires. Two hundred and fifty volunteers will randomly assign to one of four experimental groups (honest unspeeded, faking-good unspeeded, honest speeded, and faking-good speeded). Participants will ask to respond to the MMPI-2 underreporting scales (L, K, S), the PPI-R Virtuous Responding (VR) validity scale, and the PAI Positive Impression (PIM) validity scale using a computer mouse. Eye-tracking data will be recorded. The collected data will include T-point scores on the L, K, S, VR, and PIM scales; response times on these scales; several temporal and spatial mouse parameters and eye-tracking features. These data will use to investigate the presence of significant differences between the two manipulated variables (honest vs. faking-good; speeded vs. unspeeded). Machine Learning analyses will be employed.
Many studies have focused on faking-bad behaviour and developed tools to facilitate its detection; such tools include the Structured Interview of Reported Symptoms-2 (SIRS-2), the Structured Inventory of Malingered Symptomatology (SIMS), and the Inventory of Problems-29. Faking-bad behaviour has received more research attention, perhaps because its welfare/social costs (in terms of, e.g. insurance compensation) are more easily recognizable; thus, the literature on the subject is not as rich and instruments to identify faking-good behaviour are lacking; for this reason, the present research project focused on faking-good behavior, specifically.
Furthermore, Based on the assumption that the more indicators considered, the harder it should be for fakers to successfully cheat, and considering the encouraging results obtained with the aforementioned studies, we believe that, to more accurately discover faking-good behaviours, eye-tracking could successfully be implemented together with reaction times, mouse-tracking and machine learning analyses in lie detection research. Indeed, taken together all these behavioural indexes could help the experts to have a more realistic and useful personality picture of subjects (e.g., employees, parents), reducing the costs of an unreliable assessment.
Overall, the results of this research project could be useful for improving the efficiency of personnel selection and forensic evaluations, drawing on direct and indirect assessment methods that can be applied in the majority of these contexts with high discrimination. The evidence-based findings could also improve the predictive and convergent validity of personnel selection and, for example, parenting skills assessment and lead to the development of a specific and brief personality questionnaire in the near feature.