Working memory computerised task and cognitive abilities

International audienceWorking memory is currently measured with complex span paradigm, in which memoranda have to be remembered, while performing concurrent processing. This type of paradigm had been demonstrated to be much more correlated with high cognitive abilities when the processing is time-constrained (i.e., paced by experimenter) than when timing depends on participant’s responses. The aim of our study was to further investigate the role of time-pressure generated by processing on the relation between working memory performance and cognitive abilities. Unlike traditional computer-paced paradigms, in which process timing is determined a priori and identical for all participants, in our paradigm the pace was adjusted to participants’ reaction time. A counting span task was used, in which young adults were told to remember letters, while counting dotes on screen (target dotes were mixed with distractive dotes). Before the counting span task, mean times to count the 5, 6, 7, 8 or 9 target dotes were collected. These durations were used to display dots to process in the complex span task. As usually observed with computer-paced span task, the adjusted computer-pace condition gave rise to lower memory performance than a condition paced by participants (i.e., screen with dots remain till participants give their answer). Nevertheless, the correlations of adjusted-pace condition with cognitive abilities (i.e., verbal, numerical and abstract abilities) were weaker than those of participant-pace condition. These findings questioned the assumption that complex span task in which processing task is paced by computer are greater predictor of high-level cognitive abilities