Debiasing System 1: Training favours logical over stereotypical intuiting

Abstract

Whereas people’s reasoning is often biased by intuitive stereotypical associations, recent debiasing studies suggest that performance can be boosted by short training interventions that stress the underlying problem logic. The nature of this training effect remains unclear. Does training help participants correct erroneous stereotypical intuitions through deliberation? Or does it help them develop correct intuitions? We addressed this issue in four studies with base-rate neglect and conjunction fallacy problems. We used a two-response paradigm in which participants first gave an initial intuitive response, under time pressure and cognitive load, and then gave a final response after deliberation. Studies 1A and 2A showed that training boosted performance and did so as early as the intuitive stage. After training, most participants solved the problems correctly from the outset and no longer needed to correct an initial incorrect answer through deliberation. Studies 1B and 2B indicated that this sound intuiting persisted over at least two months. The findings confirm that a short training can debias reasoning at an intuitive “System 1” stage and get reasoners to favour logical over stereotypical intuitions.

Adieu Bias: Debiasing Intuitions Among French Speakers

“…In sum, these results confirm that the training improved reasoning performance, as early as the initial ‘intuitive’ stage. In this study, we replicated the sound intuiting effect found in previous debiasing studies (e.g., Boissin et al, 2021 , 2022 ). In other words, after the training intervention, reasoners were able to intuit the correct solution strategy and typically no longer required to correct an initial ‘erroneous’ response through deliberation.…”

“…However, strictly speaking, participants who select the singular very unlikely option (e.g., ‘an Oscar winner’ in the above example) do not violate the critical conjunction rule. As Boissin et al ( 2022 ) mentioned, given that we are interested in learning effects, selection of the very unlikely option can be considered a correct response. Hence, we considered answers on which the conjunction fallacy is avoided (i.e., unlikely and very unlikely answers) as correct answers.…”

“…Similarly, the time limit was set to 5 seconds for conjunction fallacy problems and 8 seconds for bat-and-ball problems. For all tasks, previous pretesting established that the time limits imposed a stringent time pressure that forced participants to respond significantly faster than in a traditional unconstrained, one-response test format ( Bago & De Neys, 2017 , 2019 ; Boissin et al, 2022 ). Note that the time limit and cognitive load were only applied during the initial response stage and not during the subsequent final stage in which participants were allowed to deliberate.…”

“…Recent successful debiasing studies have shown that a short training intervention can often help people to reason more accurately. This intervention consists of a plain-English explanation about the correct solution strategy and the typical biased response to a reasoning problem (e.g., Boissin et al, 2021 , 2022 ; Claidière et al, 2017 ; Franiatte et al, 2024 ; Hoover & Healy, 2017 ; Morewedge et al, 2015 ; Purcell et al, 2020 ; Trouche et al, 2014 ). Typically, in these studies, reasoners who received the intervention are able to produce correct responses to structurally similar problems afterwards.…”

“…Given that the ‘trained intuitor’ debiasing approach has important applied and theoretical implications, further validation is needed. However, a critical limitation of (debiasing) training studies is their predominant focus on (native) English speakers, who were tested with problem contents designed for an English-speaking test environment (e.g., Boissin et al, 2021 , 2022 , 2023a ; Franiatte et al, 2024 ; Hoover & Healy, 2017 ; Morewedge et al, 2015 ; Purcell et al, 2020 ). Such a limited scope of (most) psychological studies has been questioned by various scholars: They pointed out the lack of representation of diverse populations and languages, urging for greater inclusivity in scientific research (e.g., Arnett, 2008 ; Blasi et al, 2022 ; Huettig & Ferreira, 2023 ; Thalmayer et al, 2021 ).…”