Reliability and validity studies of the Facial Discrimination Task for emotion research

The Facial Discrimination Task (FDT) (Erwin, R.J., Gur, R.C., Gur, R. E., Skolnick, B., Mawhinney-Hee, M., Smailis, J., 1992. Facial emotion discrimination: I. Task construction and behavioural findings in normal participants. Psychiatry Research 42, 231-240.) consists of standardized black-and-white photographs of Caucasian actors exhibiting happy, sad, and neutral faces. Originally designed for brain-imaging research in emotion recognition in schizophrenia and major depression, it has since been successfully employed in emotion recognition studies on mental retardation and psychosomatic disorders. This article presents new basic psychometric data from three studies with a total of 401 college undergraduates. Content validity, item reliability (test-retest, item-total correlation, item difficulty) and test reliability (internal consistency) were established. Happy and sad items were easier to agree upon than neutral ones. In general, happy items had the highest validity, highest test-retest reliability, and highest item-total correlations. Recognition errors of neutral items were biased toward negative affect. Advantages and limitations of the FDT for clinical research applications are discussed.