An experiment was conducted involving 99 adult participants (median age = 23) completing a battery of ability tests (cognitive, perceptual speed and psychomotor) and 54 trials of a keyboard-based text editing task. Analyses were run at both the aggregate and the individual level in order to overcome limitations of a single approach. A power function accounted for 97% of variance in the effect of practice on group level task performance. The circumplex pattern of correlations held for the pattern of inter-trial task performance correlations highlighting the emergent nature of learning and change. Individual differences in strategy use and abilities made large and unique contributions in predicting task performance. Strategy use was logged for every trial and patterns of shift and stability over time were categorised, modelled and related to predictors (i.e., ability and prior knowledge) and outcomes (i.e., task performance). Findings were used to build on theories of individual differences in strategy shift and task performance. The practical relevance of these findings for training, selection, and job design will be highlighted.
Paper presented at the 7th Industrial and Organisational Psychology Conference (IOP) and 1st Asia Pacific Congress on Work and Organisational Psychology (APCWOP), Adelaide, Australia, 28 June - 01 July 2007