Wim De Neys

Dual Processing in Reasoning Two Systems but One Reasoner

Human reasoning has been characterized as an interplay between an automatic belief-based system and a demanding logic-based reasoning system. The present study tested a fundamental claim about the nature of individual differences in reasoning and the processing demands of both systems. Participants varying in working memory capacity performed a reasoning task while their executive resources were burdened with a secondary task. Results were consistent with the dual-process claim: The executive burden hampered correct reasoning when the believability of a conclusion conflicted with its logical validity, but not when beliefs cued the correct response. However, although participants with high working memory spans performed better than those with lower spans in cases of a conflict, all reasoners showed similar effects of load. The findings support the idea that there are two reasoning systems with differential processing demands, but constitute evidence against qualitative individual differences in the human reasoning machinery.

Dual Processing in Reasoning

Human reasoning has been characterized as an interplay between an automatic belief-based system and a demanding logic-based reasoning system. The present study tested a fundamental claim about the nature of individual differences in reasoning and the processing demands of both systems. Participants varying in working memory capacity performed a reasoning task while their executive resources were burdened with a secondary task. Results were consistent with the dual-process claim: The executive burden hampered correct reasoning when the believability of a conclusion conflicted with its logical validity, but not when beliefs cued the correct response. However, although participants with high working memory spans performed better than those with lower spans in cases of a conflict, all reasoners showed similar effects of load. The findings support the idea that there are two reasoning systems with differential processing demands, but constitute evidence against qualitative individual differences in the human reasoning machinery.

Smarter Than We Think When Our Brains Detect That We Are Biased

Human reasoning is often biased by stereotypical intuitions. The nature of such bias is not clear. Some authors claim that people are mere heuristic thinkers and are not aware that cued stereotypes might be inappropriate. Other authors claim that people always detect the conflict between their stereotypical thinking and normative reasoning, but simply fail to inhibit stereotypical thinking. Hence, it is unclear whether heuristic bias should be attributed to a lack of conflict detection or a failure of inhibition. We introduce a neuroscientific approach that bears on this issue. Participants answered a classic decision-making problem (the “lawyer-engineer” problem) while the activation of brain regions believed to be involved in conflict detection (anterior cingulate) and response inhibition (lateral prefrontal cortex) was monitored. Results showed that although the inhibition area was specifically activated when stereotypical responses were avoided, the conflict-detection area was activated even when people reasoned stereotypically. The findings suggest that people detect their bias when they give intuitive responses.

Bias and Conflict A Case for Logical Intuitions

Human reasoning has been characterized as often biased, heuristic, and illogical. In this article, I consider recent findings establishing that, despite the widespread bias and logical errors, people at least implicitly detect that their heuristic response conflicts with traditional normative considerations. I propose that this conflict sensitivity calls for the postulation of logical and probabilistic knowledge that is intuitive and that is activated automatically when people engage in a reasoning task. I sketch the basic characteristics of these intuitions and point to implications for ongoing debates in the field.Human reasoning has been characterized as often biased, heuristic, and illogical. In this article, I consider recent findings establishing that, despite the widespread bias and logical errors, people at least implicitly detect that their heuristic response conflicts with traditional normative considerations. I propose that this conflict sensitivity calls for the postulation of logical and probabilistic knowledge that is intuitive and that is activated automatically when people engage in a reasoning task. I sketch the basic characteristics of these intuitions and point to implications for ongoing debates in the field.

Automatic–heuristic and executive–analytic processing during reasoning: Chronometric and dual-task considerations

Human reasoning has been shown to overly rely on intuitive, heuristic processing instead of a more demanding analytic inference process. Four experiments tested the central claim of current dual-process theories that analytic operations involve time-consuming executive processing whereas the heuristic system would operate automatically. Participants solved conjunction fallacy problems and indicative and deontic selection tasks. Experiment 1 established that making correct analytic inferences demanded more processing time than did making heuristic inferences. Experiment 2 showed that burdening the executive resources with an attention-demanding secondary task decreased correct, analytic responding and boosted the rate of conjunction fallacies and indicative matching card selections. Results were replicated in Experiments 3 and 4 with a different secondary-task procedure. Involvement of executive resources for the deontic selection task was less clear. Findings validate basic processing assumptions of the dual-process framework and complete the correlational research programme of K. E. Stanovich and R. F. West (2000).Human reasoning has been shown to overly rely on intuitive, heuristic processing instead of a more demanding analytic inference process. Four experiments tested the central claim of current dual-process theories that analytic operations involve time-consuming executive processing whereas the heuristic system would operate automatically. Participants solved conjunction fallacy problems and indicative and deontic selection tasks. Experiment 1 established that making correct analytic inferences demanded more processing time than did making heuristic inferences. Experiment 2 showed that burdening the executive resources with an attention-demanding secondary task decreased correct, analytic responding and boosted the rate of conjunction fallacies and indicative matching card selections. Results were replicated in Experiments 3 and 4 with a different secondary-task procedure. Involvement of executive resources for the deontic selection task was less clear. Findings validate basic processing assumptions of the dual-process framework and complete the correlational research programme of K. E. Stanovich and R. F. West (2000).

When People Are More Logical Under Cognitive Load

The present study introduces dual task methodology to test opposing psychological processing predictions concerning the nature of implicatures in pragmatic theories. Implicatures routinely arise in human communication when hearers interpret utterances pragmatically and go beyond the logical meaning of the terms. The neo-Gricean view (e.g., Levinson, 2000) assumes that implicatures are generated automatically whereas relevance theory (Sperber & Wilson, 1986/1995) assumes that implicatures are effortful and not automatic. Participants were presented a sentence verification task with underinformative sentences that have the potential to produce scalar implicatures like Some oaks are trees. Depending on the nature of the interpretation of Some (logical or pragmatic) the sentence is judged true or false. Executive cognitive resources were experimentally burdened by the concurrent memorization of complex dot patterns during the interpretation process. Results showed that participants made more logical and fewer pragmatic interpretations under load. Findings provide direct support for the relevance theory view.

Working memory and everyday conditional reasoning: Retrieval and inhibition of stored counterexamples

Two experiments examined the contribution of working memory (WM) to the retrieval and inhibition of background knowledge about counterexamples (alternatives and disablers, Cummins, 1995) during conditional reasoning. Experiment 1 presented a conditional reasoning task with everyday, causal conditionals to a group of people with high and low WM spans. High spans rejected the logically invalid AC and DA inferences to a greater extent than low spans, whereas low spans accepted the logically valid MP and MT inferences less frequently than high spans. In Experiment 2, an executive-attention-demanding secondary task was imposed during the reasoning task. Findings corroborate that WM resources are used for retrieval of stored counterexamples and that people with high WM spans will use WM resources to inhibit the counterexample activation when the type of counterexample conflicts with the logical validity of the reasoning problem.

Belief inhibition during thinking: Not always winning but at least taking part

Human thinking is often biased by intuitive beliefs. Inhibition of these tempting beliefs is considered a key component of human thinking, but the process is poorly understood. In the present study we clarify the nature of an inhibition failure and the resulting belief bias by probing the accessibility of cued beliefs after people reasoned. Results indicated that even the poorest reasoners showed an impaired memory access to words that were associated with cued beliefs after solving reasoning problems in which the beliefs conflicted with normative considerations (Experiment 1 and 2). The study further established that the impairment was only temporary in nature (Experiment 3) and did not occur when people were explicitly instructed to give mere intuitive judgments (Experiment 4). Findings present solid evidence for the postulation of an inhibition process and imply that belief bias does not result from a failure to recognize the need to inhibit inappropriate beliefs, but from a failure to complete the inhibition process. This indicates that people are far more logical than hitherto believed.

Inference suppression and semantic memory retrieval: Every counterexample counts

Reasoning with conditionals involving causal content is known to be affected by retrieval of counterexamples from semantic memory. In this study we examined the characteristics of this search process in everyday conditional reasoning. In Experiment 1 we manipulated the number (zero to four) of explicitly presented counterexamples (alternative causes or disabling conditions) for causal conditionals. In Experiment 2, using a generation pretest, we measured the number of counterexamples participants could retrieve for a set of causal conditionals. One month after the pretest, participants were presented a reasoning task with the same conditionals. The experiments indicated that acceptance of modus ponens linearly decreased with every additionally retrieved disabler, whereas affirmation of the consequent acceptance linearly decreased as a function of the number of retrieved alternatives. Results for denial of the antecedent and modus tollens were less clear. The findings show that the search process does not necessarily stop after retrieval of a single counterexample and that every additional counterexample has an impact on the inference acceptance.

Bats, balls, and substitution sensitivity: cognitive misers are no happy fools

Influential work on human thinking suggests that our judgment is often biased because we minimize cognitive effort and intuitively substitute hard questions by easier ones. A key question is whether or not people realize that they are doing this and notice their mistake. Here, we test this claim with one of the most publicized examples of the substitution bias, the bat-and-ball problem. We designed an isomorphic control version in which reasoners experience no intuitive pull to substitute. Results show that people are less confident in their substituted, erroneous bat-and-ball answer than in their answer on the control version that does not give rise to the substitution. Contrary to popular belief, this basic finding indicates that biased reasoners are not completely oblivious to the substitution and sense that their answer is questionable. This calls into question the characterization of the human reasoner as a happy fool who blindly answers erroneous questions without realizing it.