Scientists create AI language models that identify signs of schizophrenia

WASHINGTON DC: Scientists have created new tools based on AI language models that may identify tiny signs in the speech of individuals with schizophrenia.

The study, published in PNAS, seeks to explore how automated language analysis may assist clinicians and scientists in diagnosing and assessing psychiatric illnesses.

Currently, mental diagnosis is nearly exclusively reliant on conversations with patients and people close to them, with blood tests and brain scans playing only a minor role.

This lack of clarity, however, precludes a more comprehensive understanding of the causes of mental disease and the monitoring of treatment.

The researchers invited 26 people with schizophrenia and 26 control participants to complete two verbal fluency tasks in which they were asked to name as many words as they could from the category "animals" or begin with the letter "p" in five minutes.

The scientists utilised an AI language model that has been trained on enormous quantities of internet material to represent the meaning of words in a comparable fashion to humans to examine the answers supplied by participants.

They investigated whether the AI model could predict the phrases that people spontaneously recalled and whether this predictability was impaired in patients with schizophrenia.

They found that the answers given by control participants were indeed more predictable by the AI model than those generated by people with schizophrenia and that this difference was largest in patients with more severe symptoms.