Cognitive ability, thinking dispositions, and instructional set as predictors of critical thinking ☆ Robyn Macpherson, Keith E. Stanovich ⁎ University of Toronto, Canada Received 25 September 2006; received in revised form 12 May 2007; accepted 18 May 2007 Abstract This study examined the predictors of belief bias in a formal reasoning paradigm (a syllogistic reasoning task) and myside bias in two informal reasoning paradigms (an argument generation task and an experiment evaluation task). Neither cognitive ability nor thinking dispositions predicted myside bias, but both cognitive ability and thinking dispositions were significant predictors of the ability to overcome belief bias in the syllogistic reasoning task. However, instructional set (either decontextualizing or non-directive instructions) had a significant effect on myside bias in the argument generation task, as well as a marginal effect on the syllogistic reasoning task. On the latter, and to some extent on the former task, instructional set interacted with cognitive ability. The debiasing effect of decontextualizing instructions was particularly large for those participants in the lowest quartile of cognitive ability. © 2007 Elsevier Inc. All rights reserved. Keywords: Individual differences; Belief bias; Instructional set; Myside bias; Critical thinking; Cognitive ability 1. Introduction In the educational literature on critical thinking, the ability to decouple prior beliefs and opinions from the evaluation of evidence and arguments is deemed to be a skill of paramount importance (Baron, 1991, 2000; Ennis, 1987, 1995; Norris & Ennis, 1989; Paul, 1984, 1987; Perkins, 1995; Stanovich, 1999, 2004; Sternberg, 1997, 2001, 2003; Wade & Tavris, 1993). In the empirical literature, this skill is often operationalized as the ability to avoid myside bias and belief bias. Myside bias is the tendency to evaluate evidence, generate evidence, and test hypotheses in a manner biased toward one's own opinions. Strong myside bias effects have been demonstrated in numerous studies (Baron, 1991, 1995; Greenhoot, Semb, Colombo, & Schreiber, 2004; Kuhn, 1991; Nussbaum & Kardash, 2005; Perkins, 1985; Perkins, Farady, & Bushey, 1991; Stanovich & West, 2007; Toplak & Stanovich, 2003). The evidence is Learning and Individual Differences 17 (2007) 115 – 127 www.elsevier.com/locate/lindif ☆ This research was supported by grants from the Social Sciences and Humanities Research Council of Canada and the Canada Research Chairs program to Keith E. Stanovich. ⁎ Corresponding author. Department of Human Development and Applied Psychology, University of Toronto, 252 Bloor St. West, Toronto, Ontario, Canada M5S 1V6. Tel.: +1 416 923 6641 2447. E-mail address: kstanovich@oise.utoronto.ca (K.E. Stanovich). 1041-6080/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.lindif.2007.05.003