"Detecting noncredible symptomology in ADHD evaluations using machine learning"

Detecting noncredible symptomology in ADHD evaluations using machine learning https://www.tandfonline.com/doi/full/10.1080/13803395.2025.2458547

"Diagnostic evaluations for attention-deficit/hyperactivity disorder (ADHD) are becoming increasingly complicated by the number of adults who fabricate or exaggerate symptoms."

A natural consequence of having a behavioural defined condition with as yet, no objective biological test or tests. This is not a new topic by any means - Feigning ADHD? - but an incresingly important one. Most people don't fabricate or exaggerate but there will always be grifters, and in some cases grifters who either go 'shopping' for their diagnosis and/or utilise the various online resources that provide information on 'how to get a diagnosis' (typically, but not universally, wrapped in the neurospeak of neurobabble). Such an issue also aligns with the knowledge that some diagnostic providers are themselves not always 'accurate' in their diagnostic decisions: ADHD: Private clinics exposed by BBC undercover investigation https://www.bbc.co.uk/news/health-65534448

We've also seen this issue increasing in importance in research and particularly in amongst the various 'online' based psychology studies e.g. https://pubmed.ncbi.nlm.nih.gov/39582239/ where some studies don't even appear to have asked basic screening questions for their chosen cohorts (a sign of particularly bad science).

It looks however, like AI might be another tool in the arsenal to combat such 'noncredible' issues drawing on analysis of self-report responses during the assessment procedure. In this case, authors concluded: "unsupervised ML [machine learning] can effectively identify noncredible symptom reporting using scores from multiple symptom validity tests without predetermined cutoffs." As such technology gets more advanced, it will make it even easier to spot such noncredible situations.

But then the next question: why? Why would someone chose to feign symptoms to get a diagnosis? I know some people will talk about the 'advantages of a diagnosis' but this really isn't a credible answer for many occasions, outside that is, of potential access to some extremely powerful medicines. No, there's probably lots of other explanations which, drawing on the available studies on why people 'self-diagnose', means issues such as group belonging - particularly disability group belonging, the influence of social media and indeed, the presence of other psychopathology also exert potentially powerful influences.