Assessing metacognition during or after basic-level and high-level cognitive tasks? A comparative study in a...

This study explored in a non-clinical sample whether metacognitive awareness assessments measured during ("on line") and after ("end line") a neuropsychological task lead to comparable results in tests exploring basic-level or high-level cognitive functions. Short-term memory and working memory tests (forward and backward digit recall of the WAIS-III) were used to measure basic-level cognitive function. A social cognition test, the French adaptation of the Faces Test, was used to assess high-level cognitive function through recognition of facial emotions. For these two tests, we explored "on line" metacognitive awareness using a method based upon Koriat and Goldsmith's protocol. After each answer, participants were asked to rate their level of confidence in the correctness of their response. Persons had also to rate their confidence in their answer only once, at the end of the neuropsychological test, in order to explore "end line" metacognitive awareness. They were then asked "do you feel you have passed this test?" and had to rate their feeling of success on a 4-point Likert-type scale ("no", "rather not", "rather yes", "yes"). No association was found between "on line" and "end line" metacognitive awareness scores on memory tests. Poor "end line" metacognitive awareness was associated with lower "on line" metacognitive awareness score in the social cognition test. It might be of interest to assess both "on line" and "end line" metacognitive awareness in persons with schizophrenia to better take into accounts the multi-faceted structure of metacognition.