1 |
Using language processing and speech analysis for the identification of psychosis and other disorders
|
|
|
|
In: Biol Psychiatry Cogn Neurosci Neuroimaging (2020)
|
|
BASE
|
|
Show details
|
|
2 |
Language as a Biomarker for Psychosis: A Natural Language Processing Approach
|
|
|
|
In: Schizophr Res (2020)
|
|
BASE
|
|
Show details
|
|
3 |
S19. ANALYZING NEGATIVE SYMPTOMS AND LANGUAGE IN YOUTHS AT RISK FOR PSYCHOSIS USING AUTOMATED LANGUAGE ANALYSIS
|
|
|
|
BASE
|
|
Show details
|
|
4 |
24.2 NATURAL LANGUAGE PROCESSING STUDIES OF PSYCHOSIS AND ITS RISK STATES
|
|
|
|
BASE
|
|
Show details
|
|
5 |
T35. SPEED OF FACE PROCESSING PREDICTS PSYCHOSIS IN AT-RISK YOUTHS
|
|
|
|
Abstract:
BACKGROUND: Face processing deficits characterize schizophrenia, including in prodromal stages. Baseline face processing deficits predict psychosis onset in clinical high-risk (CHR) youths, specifically in fear/anger processing (Corcoran, 2015). Here, we evaluate the predictive value for psychosis onset, specifically of speed of face processing in CHR youths. METHODS: In a cohort of 49 CHR patients (of whom 7 later transitioned to psychosis), and 31 healthy controls, we examined reaction times for face processing using the UPenn battery, including the Emotion Recognition (ER-40; Kohler 2005), Emotion Discrimination (EMODIFF), Emotional Acuity (PEAT), Facial Identification, as well as motor praxis. RESULTS: CHR converters had significantly slower reaction time for face emotion recognition (FER) (p=.008), but not discrimination, acuity or facial identification. They also had slower motor praxis and greater baseline prodromal symptom severity (p<.001), but these did not account for slowing in FER. CHR converters had significantly slower reaction for all emotions, including anger (p=.008), fear (p=.005), happiness (p=.006) and sadness (p<.001), but not for neutral. DISCUSSION: The finding of slower reaction times in converters, specifically for the ER-40, which entails verbal naming of emotions, suggests language disturbance may underlie this association, consistent with our previous studies that FER accuracy and linguistic features also predict conversion. Motor praxis did not account for this association, suggesting they are separate components of risk. Present studies include fMRI of FER in CHR, which will inform future remediation strategies augmented by neurostimulation.
|
|
Keyword:
Poster Session I
|
|
URL: https://doi.org/10.1093/schbul/sbz019.315 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6455642/
|
|
BASE
|
|
Hide details
|
|
6 |
Prediction of psychosis across protocols and risk cohorts using automated language analysis
|
|
|
|
BASE
|
|
Show details
|
|
|
|