51����

A third of the psychological tests used in US court proceedings aren’t generally accepted by experts in the field, a study has found.

“A clinician has the freedom to use whatever tool they want and it’s the wild west out there,” says Tess Neal at Arizona State University.

Neal’s team looked at the validity of 364 psychological assessments commonly used in US courts. Assessments were used in a range of circumstances, from parental custody cases to the determination of a person’s sanity or their suitability for a death sentence.

In a custody case, for instance, a psychologist might be asked to assess whether a parent is responsible enough to care for their child.

Neal’s team first looked at the huge range of psychological tests currently used in courts, according to 22 previous surveys of forensic mental health professionals. “There’s way more variety out there than we realised,” says Neal.

The researchers found that 60 per cent of the tests used in US courts hadn’t received generally favourable reviews of their scientific validity in widely accepted textbooks such as the Mental Measurements Yearbook. And 33 per cent weren’t broadly accepted by psychology experts, according to nine previously published reviews of the field.

The most problematic tests are usually those that are too subjective, says Neal. For instance, the second most common assessment used according to previous surveys was the Rorschach inkblot test, in which people are asked what images they see in abstract patterns. This has been widely criticised for letting clinicians interpret responses based on their own impressions of a person. “There are questions about its scientific underpinnings,” says Neal.

Another problematic personality test asks people to complete sentences where only the first few words are given, which again is thought to be too subjective.

Rarely challenged

In a further part to their study, the team searched a legal database of all US state and federal court cases from 2016 to 2018. They found that psychological tests are challenged in court in only 5 per cent of cases, and such challenges succeed only a third of the time. “Judges are supposed to screen out bad science,” says Neal. “They’re generally not even questioning it, never mind screening it out.”

No such wide-ranging and systematic review has been done in the UK. But Robert Forde, who has written a book called , says there are concerns in the UK too, particularly over decisions by parole boards, in which prisoners are assessed on their risk of reoffending before they can be released.

Parole boards tend to be influenced by the scores psychologists give prisoners on subjective factors, such as how much remorse or empathy they show, he says. “Those factors are prone to bias.”

In UK Family Court cases, a found that a fifth of psychologists who gave evidence weren’t qualified to do so, on the basis of their submitted curriculum vitae. One problem is that while some job titles such as “forensic psychologist” are legally protected, anyone can call themselves a psychologist.

Even when people are properly qualified, they may act as expert witnesses outside their field of expertise, says Ruth Tully of Tully Forensic Psychology, a consultancy firm based in Nottingham, UK.

Journal reference: Psychological Science in the Public Interest; DOI: 10.1177/1529100619888860