Inference complexity and the logic bias effect in conditional reasoning.

This article investigates the limits of intuitive logic within the hybrid dual processing model, which posits that human reasoning involves both intuitive (type 1) and analytical (type 2) processes. Using a conditional reasoning (CR) task, the study manipulated inference complexity—defined by inference type (modus ponens [MP] vs. modus tollens [MT]) and conclusion wording (normal vs. contrary)—to examine how complexity affects the "logic bias" effect, where implicit logical processing interferes more with belief-based responding than vice versa. Across two experiments with college students, results showed that the logic bias decreases as inference complexity increases and is absent for the most complex inference type (MT/Normal), suggesting that more complex inferences rely less on intuitive logic and more on analytical processing. Additionally, individual differences revealed that higher analytical thinking disposition (measured by the actively open-minded thinking scale) is associated with a stronger logic bias, whereas greater working memory capacity relates to reduced interference from off-task processing, indicating variability in reliance on intuitive logic among individuals.