The role of automatization in complex inductive reasoning: evidence from three experiments

Automaticity plays a central role in facilitating higher-order cognitive processes by reducing demands on working memory. Across three experiments, we examined how the quality and stability of automatized knowledge affect performance in complex inductive reasoning tasks. Participants learned arbitrary symbol–meaning associations until they met predefined accuracy and speed criteria. Experiment 1 showed that, once automatization was achieved, individual differences in learning trajectories (e.g., number of trials or errors) no longer predicted performance in a symbol-based reasoning task. Experiment 2 demonstrated the durability of automatization, as performance on a complex symbol recombination task remained stable after a 30-day interval. Experiment 3 contrasted participants who learned symbol meanings accurately but without speed (non-automatized group) with those who reached both accuracy and speed thresholds (automatized group). Only the automatized group showed superior reasoning performance, particularly under increasing task complexity. These findings provide converging evidence that automatized access to learned knowledge—defined by both accuracy and speed—is essential for efficient complex reasoning, and highlight the cognitive cost of non-automatized retrieval even when accuracy is high.