Three experiments are reported, which are based upon the Wason four-card selection task inspection time paradigm, in which subjects solve computer-presented trials while using a mouse to indicate the card currently under consideration. Evans (1996) had shown that selected cards were inspected for longer than non-selected cards, and this was taken as support for the existence of pre-conscious heuristic processes that direct attention towards relevant aspects of a problem. However, Roberts (1998b) suggested that this inspection time effect is artefactual, due to task format induced biases. Experiment 1 utilized a "change" task: Cards were presented either as selected or not selected, and subjects changed these where necessary. This demonstrated an association between card selection and inspection time independently of one between the act of response and inspection time. Experiment 2 utilized a standard selection task, but subjects either responded within 2 s of each card presentation, or made selections with no time pressure. The curtailment of thinking time increased matching behaviour—more cards matching the terms in the rules were selected—and was replicated in Experiment 3 using a within-subjects design. Overall, the data support Evans' heuristic-analytic framework albeit with some caveats.