Development of a Rasch-calibrated emotion recognition video test for patients with schizophrenia

Area of Science:

• Psychiatry
• Psychology
• Neuroscience

Background:

• Schizophrenia patients often exhibit emotion recognition (ER) deficits impacting social functioning.
• Existing ER measures lack comprehensiveness, reliability, and validity, hindering accurate assessment.
• A need exists for a robust tool to evaluate ER abilities in schizophrenia.

Purpose of the Study:

• To develop and validate the Computerized Emotion Recognition Video Test (CERVT).
• To provide a comprehensive, reliable, and valid assessment of ER in schizophrenia patients.
• To ensure the CERVT is unbiased by gender.

Main Methods:

• Candidate CERVT items from 8 basic emotion domains were selected.
• Rasch analysis was employed to validate selected items.
• 269 schizophrenia patients and 177 healthy adults participated in validation.

Main Results:

• 217 CERVT items were finalized after removing misfit items and adjusting for differential item functioning.
• The CERVT demonstrated good model fits and unidimensionality across its 8 domains.
• The test exhibited good to excellent reliability (0.84-0.93) and acceptable validity, unaffected by gender.

Conclusions:

• The CERVT is a comprehensive, reliable, and valid tool for assessing ER in schizophrenia.
• The developed test provides individualized scores for 8 basic emotions.
• CERVT offers a gender-unbiased assessment, crucial for understanding social cognition in schizophrenia.