The Role of AI-Consultation in Malingering Detection during ADHD Assessments: An Insight in the Student Population

Bensink, Sebastiaan (2025) The Role of AI-Consultation in Malingering Detection during ADHD Assessments: An Insight in the Student Population. Master thesis, Psychology.

Text S.-Bensink---The-Role-of-AI-Consultation-in-Malingering-Detection-during-ADHD-Assessments.pdf Restricted to Repository staff only Download (764kB)

Abstract

The study researches the role of AI-consultation on malingering detection by two embedded performance validity tests (PVT), when compared to symptom criteria consultation. 92 psychology students were randomly assigned to one of three groups, Honest Responders (HR), Symptom Coached Responders (SC), or AI-Coached Responders (AIC), after which they received the instructions to perform to the best of their ability or feign ADHD according to specific instructions. Participants completed two common cognitive tests with embedded PVTs, the test Perceptual and Attention Functions – Selective Attention test (WAFS) of the Vienna Test System (VTS), and the Reliable Digit Span (RDS) of the Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV). The study aimed to evaluate if participants receiving AI-consultation prior to assessment were more successful in feigning ADHD while evading detection by PVTs, compared to participants receiving symptom criteria coaching. The AI-consultation was operationalised as a feigning guide, made by ChatGPT as a prompt. Data were analysed with non-parametric Kruskal-Wallis tests and post-hoc Dunn-Bonferroni tests. Significant differences were found between groups with large effect sizes. Post-hoc analyses showed that AIC participants demonstrated more cautious symptom amplification than SC participants, yet detection rates for both groups were similar, with 73.3% and 75.6% of simulators detected respectively. The observed specificity was 96.7%. The combination of the WAFS and RDS showed adequate sensitivity and excellent specificity. The study argues the potential role of AI-consultation in the future and underscores the importance of continuing research on validity testing.

Item Type:	Thesis (Master)
Supervisor name:	Fuermaier, A.B.M.
Degree programme:	Psychology
Differentiation route:	Clinical Neuropsychology (CN) [Master Psychology]
Date Deposited:	11 Feb 2025 15:01
Last Modified:	11 Feb 2025 15:01
URI:	http://gmwpublic.studenttheses.ub.rug.nl/id/eprint/4634

Actions (login required)

View Item