Researchers at Columbia University recorded young people already considered at high clinical risk for psychosis, then used automated software to measure how coherent and complex their speech was, without a clinician's judgment involved in the measurement.
Among 34 young people, five of whom later developed psychosis, the automated measures of semantic coherence and sentence complexity predicted which of the participants would go on to develop psychosis with 100 percent accuracy in this small sample, outperforming predictions based on clinicians' own interviews with the same young people.
A perfect score in a small, single-site study is a strong signal to pursue, not a finished test; larger and more diverse follow-up work was needed to see whether the accuracy would hold, and later studies by other groups have shown real but more modest results. It remains an active research tool rather than something used to diagnose patients today.
Read the original study
Bedi et al., npj Schizophrenia, 2015 · doi.org/10.1038/npjschz.2015.30
How someone spoke predicted what came next.
Source
- Bedi, G., Carrillo, F., Cecchi, G. A., Slezak, D. F., Sigman, M., Mota, N. B., Ribeiro, S., Javitt, D. C., Copelli, M., & Corcoran, C. M. (2015). Automated analysis of free speech predicts psychosis onset in high-risk youths. npj Schizophrenia, 1, 15030. https://doi.org/10.1038/npjschz.2015.30