The ways instructions affect argument evaluation remain largely unknown. There may thus be uncovered ways to improve people's argumentation as well as matters better left unsaid. We examined how accurate and biased argument evaluation differ between instructions that emphasise facts, objectivity, or conscience. Data from adult Finns (N = 1600) showed that highlighting objective evaluation of arguments was connected to lower myside bias and to greater discernment between strong and weak arguments than when considerations of conscience or factual information were emphasised. The clearest differences in argument evaluations were between the instructions emphasising objectivity and conscience. We also analysed whether the cognitive correlates of argument evaluation varied depending on the instructions and found no differences. Instead, higher actively open-minded thinking, scientific reasoning skills, and cognitive reflection correlated systematically with better ability to separate strong from weak arguments. They were also never positively related to myside bias in argument evaluation.
The omission bias refers to the tendency to judge harmful omissions as less immoral than equally harmful commissions. Rooted in the philosophical doctrine of doing and allowing, omission bias has often been treated as a principle in moral decision-making. Although extensively replicated among Western samples, little research has directly examined whether omission bias and its underlying processes generalise across cultures. The present study employed a 2 (China vs. US) x 2 (Commission vs. Omission) between-subjects design with 266 American and 275 Chinese participants. Participants evaluated scenarios in which either a commission or an omission led to the same harmful outcome. The omission bias in morality perceptions generalised across cultures, although not for all scenarios. Intentionality and causality perceptions were significant predictors of immorality, robust across cultures and scenarios. Scenario-specific reversals of omission bias reflected shifts in intentionality and causality. These findings provide cross-cultural evidence for an attributional asymmetry account of omission bias.
People sometimes view it as justified for others (and for themselves) to consciously adopt beliefs that are inconsistent with the available evidence, provided that the belief is thought to promote a moral good. In the present research we examine whether this phenomenon is observed regardless of people's level of commitment to epistemic rationality. Two studies show that how much people value epistemic rationality predicts more favourable evaluations of other people's evidence-based beliefs, and more unfavourable evaluations of beliefs formed based on non-evidential considerations (Study 1-2). The only circumstance where a strong commitment to epistemic rationality did not favour the evidence-based belief was when it required racial stereotyping of a man based on local crime statistics. This exception was not attributable to elevated Care/harm concerns, did not depend on whether the target was black or white (Study 3 A), or on whether stereotyping was based on skin colour or clothing (Study 3B).
We explore intuitive monitoring and control within a Brunswikian framework where intuition refers to probabilistic processes rooted in perception and inductive experience, while analysis involves deterministic processing of facts and symbols, in turn implying operational criteria in terms of different response distributions. In the context of the bat-and-ball problem, we test the claim that intuition uses contextual cues to monitor the analytic calculations. Across two experiments, we show that participants' responses fall into three categories: (i) the normative algorithm, (ii) a simple response bias, and (iii) responses where contextual cues are taken into consideration by switching the cost of the bat and the ball when such a shift is intuitively plausible. The results suggest that, although it may be difficult to elicit intuitive control in a maths quiz like the bat-and-ball problem, the choice of analytical calculation is affected by the intuitive plausibility of the output in the predicted direction.
This research examines how accuracy motivation shapes numerical estimation by altering the mental representation of the number scale. We demonstrate that accuracy instructions lead individuals to adopt a finer-grained scale, resulting in the use of more precise estimates. This shift appears to be driven by a belief that using a finer-grained scale is necessary for producing accurate estimates, even though greater resolution alone does not guarantee greater accuracy. Across four studies, we show that accuracy instructions consistently increase numerical precision without improving objective estimation accuracy. We also identify an unintended consequence of this shift: adopting a finer-grained scale paradoxically constrains the magnitude of adjustments from initial estimates, even when larger revisions are warranted. These findings reveal a novel unintended consequence of emphasising accuracy, offering new insights into the mechanisms underlying anchoring effects and the broader cognitive processes involved in numerical judgement and decision-making.
We examined in two studies how analytic thinking and valuing epistemic rationality contribute to political belief bias - an inclination to judge politically congruent conclusions as valid (vs. invalid) or to judge politically incongruent conclusions as invalid (vs. valid), regardless of their actual validity. Study 1 established that people exhibit political belief bias, although unexpectedly it was only observed among liberals (vs. conservatives). As hypothesised, analytic thinking was associated with less political belief bias. In Study 2, we measured political belief bias by assessing participants' political beliefs about each conclusion directly, rather than relying on their general political orientation as a proxy. Analytic thinking as well as how much people value epistemic rationality predicted less political belief bias. These findings highlight the importance of analytic thinking as well as motivation to form beliefs based on logic and evidence in mitigating political belief bias.
Representational Change Theory has proposed two processes, constraint relaxation (CR) and chunk decomposition (CD), that are responsible for restructuring during insight problem solving (Knoblich et al., 1999). If these processes are independent (& Ouml;llinger et al., 2008), then improvements in matchstick arithmetic performance due to instructions that prompt either process should be limited to problems that require that process. In Study 1, CD instructions only improved performance on CD problems while CR instructions only improved performance on CR problems. A combined instruction condition did not outperform the single instruction conditions, and in some cases decreased solution rates. Study 2 varied whether solutions directly matched the instructions and found some evidence that CD instructions generalised to CD solutions that did not directly match instructions. However, CR instructions did not increase solutions on CR problems, and single instructions sometimes decreased solutions on problems requiring another process. Although these results are most consistent with hypotheses that these processes are independent and competitive in the context of matchstick arithmetic problems, issues with the specificity of the hints used in the instructional conditions are discussed.
We report the results of five experiments that identify situations in which counterfactual and causal explanations contain the same or different content. For stories about an exceptional event, participants focused on the exceptional event when they created counterfactual explanations but not as often when they created causal explanations in Experiment 1. For stories about a controllable choice, they referred to the controllable choice when they created counterfactual explanations but not as often when they created causal explanations in Experiment 2. For stories about actions and inactions, they focused on the same content when they created counterfactual and causal explanations in Experiment 3. Experiments 4a and 4b showed that the focus on actions in explanations is vulnerable to the order in which information about actions and inactions is provided. The results are consistent with the proposal that people construct explanations by simulating a set of counterfactual possibilities. The emphasis in causal explanations on explaining what happened, and in counterfactual explanations on explaining what could have happened differently, leads to the selection of different sorts of causes from the counterfactual possibilities.
Chess is a game of strategic thinking and time management, where a player can lose a game on time despite making all the best moves. Finding the best move is a deliberate and energy-intensive process in a game where players are often under time pressure. Therefore, players who can balance this trade-off will have a significant advantage. The current study explores such instances where winning is contingent on how well players balance their accuracy under time pressure. We found that winning players, compared to their opponents, followed a more adaptive decision strategy-they made more theoretical best moves (i.e., accurate moves) in highly critical positions. However, the accuracy difference between the opponents was very similar in less critical positions. We conclude that winning players have a better understanding of when and how to allocate their limited resources efficiently, even when controlling for differences in skill levels, compared to their opponents.
According to Inferentialism, a conditional is true precisely if there is a compelling argument for its consequent starting from its antecedent conjoined with contextually given background knowledge. As a direct consequence of this view, the truth of a conditional's component parts does not guarantee the truth of the conditional, given that it is compatible with there not being an argument from the conditional's antecedent to its consequent. Thus, Inferentialism invalidates the principle called Centering, according to which the joint truth of phi and psi guarantees the truth of "If phi, psi". This distinguishes Inferentialism from all standard semantics of conditionals, all of which validate Centering. Here, we report the results from a study that registered (i) participants' truth evaluations of a set of conditionals; (ii) their truth evaluations of the component parts of those conditionals; (iii) their judgements of the strength of the inferential connection between antecedent and consequent for each conditional; and (iv) their metacognitive confidence in the inference strength judgements. The data from (i)-(iii) allowed us to test Centering and to investigate whether possible violations of the principle can be explained in inferentialist terms. The data from (iv) allowed us to make some progress on testing predictions derived from Hypothetical Inferential Theory, the psychological face of Inferentialism. Our findings show that almost 60% of the relevant responses violated Centering; that participants obeyed Centering only when there was a strong inferential link between true antecedent and true consequent; and that the best models were those that involved inference strength and metacognitive confidence, while the Centering model was weak.
Central questions in metacognitive research are how reliably people's confidence reflect success and how mental effort is regulated during cognitive tasks. Ackerman's Bird's-Eye View of Cue Integration methodology exposed confidence biases in a geometric problem-solving task. We replicated one experiment, incorporating eye-tracking to expose additional confidence predictors and insights into mental effort regulation. Fifty-four university students performed the task wearing eye-tracking glasses. The original confidence biases were replicated. Notably, the best predictive model for confidence included eye-tracking measures. Mean pupil diameter and number of visits to stimuli uniquely predicted success and confidence, highlighting the role of mental effort investment and attentional shifts. Individual-level analyses revealed that while the eye-tracking measures identified individuals with higher success, they did not reliably predict confidence or resolution (confidence associated with success). These findings underscore the potential of eye-tracking to provide valuable insights into metacognitive monitoring and mental effort regulation, and suggest eye-tracking potential utility when eliciting confidence ratings is impractical.
Knowledge affects human deductive reasoning, but the mechanisms by which this occurs are not fully understood. Example knowledge-the ability to generate and categorise specific examples of general possibilities-is proposed to play a central role, and individual differences in such knowledge are proposed to contribute to differences in deductive reasoning. To test these hypotheses, four studies investigated the role of example knowledge in adults' conditional reasoning about algebra. Individual differences in domain-specific knowledge predicted conditional reasoning about algebra when controlling for everyday and abstract conditional reasoning (Study 1), individuals spontaneously referred to examples during conditional reasoning about algebra (Study 2), individual differences in example knowledge predicted differences in conditional reasoning about algebra when controlling for general algebra knowledge and everyday conditional reasoning (Study 3), and training designed to increase example knowledge improved conditional reasoning about algebra (Study 4). The findings suggest that deductive reasoning in the domain of maths is in part a domain-specific skill, insofar as it depends on domain-specific knowledge, and that one cause of this knowledge-dependence is individuals' reliance on examples during reasoning.
From early adolescence, cognitive ability has been shown to predict reasoning performance. Recent studies indicate that adults with high cognitive ability tend to rely on accurate intuitions rather than deliberately correcting incorrect intuitions. However, it is unclear whether this pattern applies to adolescents. In the current study, we tested whether-as previously reported-younger adolescents are less likely to intuit correctly and whether cognitive ability is more predictive of correct intuitions or deliberate corrections in 7th graders and 12th graders. We used a two-response paradigm where participants gave a fast intuitive response under time pressure and cognitive load, followed by an unconstrained deliberate response. Results confirmed that younger adolescents produce fewer correct intuitions than older ones. Additionally, in younger adolescents, cognitive ability did not positively correlate with reasoning performance, while in older adolescents, it mainly predicted deliberation. The study suggests that the previously established association between sound intuiting and cognitive ability among adults emerges quite late in development.
Mathematical problem solving relies in part on higher-level domain-general skills, but contributions of such skills to problem solving are understudied in the field of mathematical cognition. This study sought evidence for domain-general contributions to maths problem solving of four such skills: logical reasoning, backward reference, planning/evaluating, and allocation of time. Participants completed problem solving tasks in two domains, probability and geometry. Concurrent engagement of each skill during each task was assessed. Individual differences in engagement of logical reasoning were correlated between tasks and associated with accuracy on each task; this pattern of results was not found for the other three skills. Qualitative analysis revealed that errors involving valid logic applied to false premises occurred in both tasks, but errors involving invalid logic occurred primarily on the geometry task. The findings are consistent with logical reasoning contributing to maths problem solving, but suggest that the mechanisms underlying such contributions may vary between mathematical domains and tasks.
Confidence ratings are frequently used to measure metacognition. However, a growing body of evidence has found that eliciting confidence ratings affects reasoning performance - a phenomenon known as reactivity. Additionally, recent research has suggested that there are individual differences in the direction and magnitude of reactivity effects. Here, in a preregistered experiment (N = 200), we examined whether eliciting confidence ratings during a fluid intelligence test (Raven's Progressive Matrices) affected performance, and whether this effect was moderated by participants' baseline metacognitive abilities captured using a perceptual discrimination task. We evaluated whether reactivity was moderated by 1) metacognitive sensitivity (the extent to which confidence judgments track objective performance), 2) metacognitive efficiency (the ratio of metacognitive sensitivity relative to cognitive performance), and 3) mean confidence on the perceptual discrimination task. Our findings suggest that only metacognitive sensitivity moderated reactivity, with participants who had relatively poor metacognitive sensitivity showing positive reactivity and those with relatively good metacognitive sensitivity not showing reactivity. This suggests that eliciting confidence ratings may have a compensatory effect - benefiting those who have poorer metacognition, which suggests they may be a useful intervention for promoting metacognitive processes.
Recent dual-process work suggests that sound responses to reasoning tasks are often generated intuitively. This has led to the hypothesis that deliberation contributes to sound response justification rather than mere response generation. However, direct evidence supporting this justificatory role of deliberation remains sparse. To test this hypothesis, three studies examined whether people could properly justify their responses to classic bias problems (base rate neglect, bat and ball, and risky choice problems) when responses were given intuitively or after deliberation. Participants provided fast, intuitive responses under time pressure in a first block, followed by unconstrained, deliberative responses to the same problems in a second block, with justifications required in both cases. Results show participants generally struggled to properly justify intuitive correct responses but tended to properly justify deliberative correct responses. These findings suggest that deliberation helps articulate reasons for intuitions, highlighting its role in the justification of decisions.
This article investigates metacognitive awareness in the domain of bullshit detection, focusing on whether overconfidence stems from cognitive limitations, the need to protect self-esteem, or dark personality traits. Across two preregistered studies with Slovak adults (N1 = 596, N2 = 433), we measured bullshit detection, overconfidence (overestimation and overplacement), self-rated detection ability, bullshitting frequency, cognitive abilities and dark traits. The results confirmed a Dunning-Kruger effect: individuals with poorer bullshit detection overestimated their abilities and believed they outperformed others. Overconfidence was negatively associated with cognitive performance and positively with self-esteem. Persuasive bullshitting was more frequent among individuals with lower bullshit detection, suggesting some may bullshit to compensate for cognitive limitations. Moreover, narcissism and Machiavellianism consistently predicted both persuasive and evasive bullshitting, indicating image-management motives. While cognitive abilities predicted metacognitive accuracy, dispositional traits were stronger predictors of self-reported bullshitting (another facet of metacognitive awareness). These findings underline the need to distinguish between different conceptualizations of bullshit and highlight the interplay of cognitive and motivational factors in metacognitive miscalibration.
Jumping to conclusions (JTC) is the tendency to reach a conclusion after reviewing only minimal evidence. However, research on the ways in which this individual difference functions in motivated and unmotivated contexts is mixed. In the present research, we investigated whether JTC would predict evidence search in a context where defensive motivated processing is likely to emerge. In Study 1 participants (N = 383) completed a JTC measure then engaged in an iterative search for information to evaluate a gun-control non-profit. In Study 2 we conceptually replicated Study 1 with Democrat and Republican non-profits. Both studies found main effects of defensive processing and JTC but no interaction between predictors, suggesting that JTC is not due to lack of motivation. Defensive motivation increases search length for those high in JTC; however motivation is not sufficient to close the gap between high and low JTC individuals.
