Introduction: Reason has always been considered an essential feature of our species, leading us to tremendous progress in the evolutionarily very short time of our existence. On the other hand, everyday life and research demonstrate we are prone to various misconceptions and errors in reasoning and decision-making, and we often do not behave in a way one would deem rational. Rationality is a more encompassing concept than intelligence, with various abilities, skills, tendencies, and knowledge structures indicative of Type 2 processing (i.e., analytical thinking) being related to its underpinnings. There is ample evidence of the beneficial effects of analytical thinking both in laboratory research as well as when it comes to important beneficial psychological and behavioural outcomes in everyday life. Moreover, two constructs and their respective measures proved to be promising predictors of accuracy on a wide range of tasks indicative of rational thinking, as well as epistemically founded beliefs (i.e., lesser tendency to epistemically unfounded beliefs) - cognitive reflection test (CRT) and actively open-minded thinking (AOT) (for a review see for example Pennycook, Fugelsang & Koehler, 2015a; Stanovich et al., 2016). On the other hand, democracy advances through open discussion, questioning, and debating opposing stances, even when allowing for absurd outbursts and stupidity, such as when established facts are challenged, at least in theory. Ideally, such an inclusive process leads to consensus and ultimately, society converges on the best available evidence. However, we are also witnessing the adverse effects of strong ideological and worldview polarization over a number of scientific and empirical issues. Although the human propensity for tribalism has evolutionary utility, sacrificing epistemic accuracy in favour of signalling affiliation and loyalty to groups with which we share the same values, ideas, and beliefs can result in serious consequences for collective well-being (Clark et al., 2019; Clark & Winegard, 2020). The most obvious example is probably the persistent, damaging, and increasing gap between Republicans and Democratsin the USA regarding climate change (Dunlap et al., 2016). A similar ideological gap has recently been observed over the COVID-19 coronavirus pandemic, as well (Kerr et al., 2021). Croatia is by no means short of its share of strong worldview and ideological divisions, easily observed in everyday context along the left-right, or colloquially red and blue, Partisans and Ustashas, spectrum on various topics. Interpreting scientific facts and related empirical evidence represents the foundation for shaping social policies and decision-making, with significant implications for both individuals and society as a whole in their daily lives. Therefore, investigating the role of analytical thinking on the acceptance of epistemically facts represents a valuable research endeavour. Within this dissertation, we aimed to examine the relationship between analytical reasoning and the acceptance of scientific and related forms of empirical evidence with regard to the congruency ofsuch facts with religious and socio-political orientations. Asfar as we know, no research has been conducted in Croatia explicitly addressing this problem. In our investigation, we focused on two promising measures of propensity for analytical thinking – CRT and AOT, and a number of topics we presumed might be subject to polarization along ideological and worldview lines, from evolutionary theory, homosexuality and vaccinating girls against HPV to the so-called Istanbul Convention, as well as two issues that we presumed should not be subject to polarization (antibiotics and obesity). We approached this problem by testing the postulates of the expressive rationality account. The expressive rationality account places primary emphasis on identity-protective motivated reasoning as a way of processing information that rationally advances the individual's goals (Kahan, Peters, et al., 2012). According to this perspective, the problem is not that people do not think or are bad at reasoning, but that they do not always use their cognitive capacities to converge their beliefs to scientific knowledge and best available evidence, and rather use them to form, express, and defend beliefs that are indicative of their worldviews and sacred values (Kahan, 2013; see also Stanovich, 2021a) or to form persuasive arguments (Mercier, 2016; Mercier & Sperber, 2011, 2017). Moreover, individuals who possess the capacity and inclination for the demanding Type 2 processing should be even more skilled in identity-protective reasoning, i.e., aligning their beliefs according to their ideological or cultural groups with which they identify and systematically filtering information that corresponds to their worldview (Kahan, 2013, 2017d; Stanovich, 2013). The direct hypothesis following this reasoning is that the individuals most adept at analytical thinking will also be the most polarized. Namely, they will easily recognize situations of political and social significance and selectively use their critical reasoning capacities in interpreting, i.e., accepting or rejecting scientific and empirical facts, depending on whether they are aligned with their worldview orientations. This very assumption is in direct contrast to the bounded rationality thesis. Within the bounded rationality account, motivated cognition, i.e., myside bias, is conventionally considered one of the manifestations of miserly thinking we are prone to. According to this account, the very cause of misperception of scientific facts and related forms of evidence is the deficit of analytical thinking (failure to recognize the need to engage in analytical thinking and/or engaging superficial analytical processes that are reduced to serial associative cognition with a focal bias). Directly stemming from this is the hypothesis that the worldview and ideological polarization concerning scientific and empirical facts should decrease among individuals with a greater capacity and tendency of analytical thinking (Kahan, 2013; 2015a), i.e., should be greatest among individuals inclined to rely on Type 1 processing (Kahan, 2017d). Method: To examine the role of analytical thinking, i.e., CRT and AOT, and potential interactions with socio-political and religious worldviews, we conducted an initial study on a convenient sample consisting of 447 participants (after cleaning and imputation of missing data), and then on a quota sample (regarding age,sex, and place of residence) of 1568 (after cleaning and imputation of missing data) adult citizens of the Republic of Croatia. Missing data were relatively low across all the items, but to ensure that the power provided by the large sample was fully exploited, missing data were imputed using the relatively robust predictive mean matching method. In the preliminary study we also investigated psychometric characteristics of all instruments and shortened them to a practical level for use in the main study. To test our hypotheses, we conducted correlational analyses and regression analyses in which we controlled for basic sociodemographic factors (age, sex, and educational level). Results: The results showed that, in the Croatian context, ideology and religiosity effects on the perception of facts, and related forms of empirical evidence are present on a number ofscientific and social issues. That is, citizens generally show a greater tendency to agree with epistemically founded statements that are congruent with their worldview. Notably, the most noticeable differences between left-leaning and right-leaning citizens, and religious and non-religious citizens, were on the issue of evolution, homosexuality, the so-called Istanbul Convention, and HPV. These findings align with previous research suggesting that we are motivated reasoners, inclined to perceive and interpret information with a myside bias. However, it's worth noting that we were unable to identify issues on which right-leaning or religious citizens were more inclined to agree with epistemically based claims. This limitation underscores the need for further examination of the role of analytical thinking in this context. Overall, the evidence that CRT and AOT contribute to greater acceptance of scientific and related empirical facts that are generally not subject to polarization, without further qualification, is weak at best. Ultimately, this is not that surprising as the issues in question are more or less addressed by the school curriculum and various public health campaigns, and citizens with varying levels of analytical thinking are exposed to equal sources of knowledge and unanimous messages. Thus, their mindware is less likely to be systematically contaminated with distorted information. In such situations, it is possible that analytical thinking does not represent a crucial advantage or decisive role. From the perspective of dual-process theories (Pennycook, 2023; Stanovich, 2012; Stanovich et al., 2016), the tendency to be reflexive and actively open-minded does not necessarily result in greater epistemic accuracy if intuitive responses lead to the same outcomes. Of course, there is no ground to expect that analytical thinking in such situations will have a negative effect, a conclusion supported by our results. Regarding the main hypothesis, we found scarce evidence of an interaction between analytical thinking and socio-political and religious worldviews, which overall provide very little support for the expressive rationality account. If anything, the cases in which the interaction effect proved significant speak more in favour of bounded rationality, suggesting that ideological polarization abates among highly open-minded individuals. Furthermore, the important finding is that actively open-minded thinking showed a consistent positive effect in the acceptance of epistemically based claims, whereby AOT exhibited an independent predictive contribution (apart from only a few instances where an interaction effect was determined). Conclusion: Ultimately, we can conclude that while motivated reasoning was prevalent across most controversial topics, we did not find enough evidence that analytical thinking is harmful, i.e., that it exacerbates myside bias. In fact, especially regarding AOT, analytical thinking is associated with a greater agreement with scientific facts and similar empirical evidence, and in some (admittedly, rare) situations, the tendency to think with an actively open mind can even contribute to mitigating myside bias. Given the complex nature of the relationship between analytical thinking and the formation of epistemically rational beliefs, the findings and limitations of our research efforts should be understood as a starting point providing guidelines for further research.
