What Makes a Good Reasoner?: Brain Potentials and Heuristic Bias Susceptibility

What Makes a Good Reasoner?: Brain Potentials and Heuristic Bias Susceptibility Wim De Neys (wim.deneys@univ-tlse2.fr) CNRS, Universite de Toulouse, France Nikolay Novitskiy (nikolay.novitskiy@psy.kuleuven.be) Lab Experimental Psychology, University of Leuven, Belgium Jennifer Ramautar (J.Ramautar@nin.knaw.nl) Netherlands Institute for Neuroscience, Amsterdam, the Netherlands Johan Wagemans (Johan.wagemans@psy.kuleuven.be) Lab Experimental Psychology, University of Leuven, Belgium Abstract Human reasoning is often biased by intuitive heuristics. A key question is why some people are less susceptible to this bias than others. It is debated whether the bias results from a failure to monitor one’s intuitive conclusions for conflict with logical considerations or from a failure to inhibit the tempting intuitions. This results in different views on the role of individual differences in executive monitoring and inhibition capacity for sound reasoning. The present study presents a new approach to address this issue. After an initial reasoning screening a group of the most and least biased reasoners were invited for an EEG study in which neural markers of their executive monitoring (ERN amplitude) and inhibition (N2 amplitude) skills were recorded. Results indicated that biased reasoners were characterized by less developed inhibition but not monitoring capacity. Findings support the view that monitoring one’s intuition for conflict during thinking is a flawless and undemanding process suggesting that even the poorest reasoners at least detect that they are biased. Keywords: Decision-making; Reasoning; EEG Introduction Decades of reasoning and decision-making research showed that human thinking is often biased (Evans, 2008; Kahneman, 2002). In general, human reasoners seem to have a strong tendency to base their judgment on fast intuitive impressions rather than on more demanding, deliberative reasoning. Although this intuitive or so-called “heuristic” thinking can be very useful, it will sometimes cue responses that conflict with traditional normative logical or probabilistic considerations and bias our decision- making. Whereas it is well established that human judgment is often biased, the nature of this bias is far less clear. Some influential authors have argued that the widespread heuristic bias can be attributed to a failure to monitor our intuition (e.g., Kahneman & Frederick, 2002). Because of lax monitoring, people would simply fail to detect that the intuitive response conflicts with normative considerations. However, others have argued that there is nothing wrong with the monitoring process (e.g., Epstein, 1994; Houde, 2007; Sloman, 1996). According to these authors, people have little trouble in detecting that their intuitive response is biased. The problem, according to this view, is that people’s intuitive beliefs are so tempting that they fail to discard them. Thus, people “behave against their better judgment” (Epstein, 1994) when they give an unwarranted heuristic response: They detect that they are biased but simply fail to block the biased response. In sum, according to this flawless detection view, biased decisions are attributed to an inhibition failure rather than a conflict monitoring failure per se. The debate on the nature of heuristic bias results in opposing views on the interpretation of individual differences in bias susceptibility. Although the vast majority of educated adults are typically biased when solving classic reasoning and decision-making tasks, some people do manage to reason correctly and refrain from giving the tempting but unwarranted heuristic response. Individual differences in executive control capacity (as measured with general working memory or intelligence test) are widely cited as the cause of this reasoning performance variability (e.g., De Neys, 2006; De Neys & Verschueren, 2006; Evans, 2008; Stanovich & West, 2000). However, conflict monitoring and inhibition are both considered key executive processes and the precise contribution of each component as possible mediator of reasoning performance has not been established. Bluntly put, it is not clear what makes a good reasoner: Having a superior monitoring capacity, having a superior inhibition capacity, or a combination of both. The two views on heuristic bias make differential predictions here. According to the lax monitoring view, people are mainly biased because of inefficient monitoring. Hence, one can expect that good reasoners will be primarily characterized by superior executive monitoring skills. Good reasoners will be better at monitoring their intuitively cued conclusions for conflict with more normative considerations and will be more likely to detect that their initial response is biased. However, the flawless monitoring view conceives monitoring during thinking as a quite undemanding process by entailing that even the most biased reasoners are successful at it (De Neys, Moyens, & Vansteenwegen, 2010; Franssens & De Neys, 2009). Hence, given the postulated minimal demands of the monitoring process during thinking, one can predict that individual differences in