https://www.frontiersin.org/articles/10.3389/fpsyg.2017.02191/full
The Negative Relationship between Reasoning and Religiosity Is Underpinned by a Bias for Intuitive Responses Specifically When Intuition and Logic Are in Conflict
Richard E. Daws and Adam Hampshire*
The Computational, Cognitive and Clinical Neuroimaging Laboratory (C3NL), Imperial College London, London, United Kingdom
It is well established that religiosity correlates inversely with intelligence. A prominent hypothesis states that this correlation reflects behavioral biases toward intuitive problem solving, which causes errors when intuition conflicts with reasoning. We tested predictions of this hypothesis by analyzing data from two large-scale Internet-cohort studies (combined N = 63,235). We report that atheists surpass religious individuals in terms of reasoning but not working-memory performance. The religiosity effect is robust across sociodemographic factors including age, education and country of origin. It varies significantly across religions and this co-occurs with substantial cross-group differences in religious dogmatism. Critically, the religiosity effect is strongest for tasks that explicitly manipulate conflict; more specifically, atheists outperform the most dogmatic religious group by a substantial margin (0.6 standard deviations) during a color-word conflict task but not during a challenging matrix-reasoning task. These results support the hypothesis that behavioral biases rather than impaired general intelligence underlie the religiosity effect.
Introduction
The relationship between religiosity and intelligence has been an important topic amongst scientists and the public for some time (Harris, 2004; Dennett, 2006; Hitchens, 2007; Dawkins, 2008). Early evidence from the twentieth century suggested that religiosity and intelligence negatively correlated amongst college students (Howells, 1928; Sinclair, 1928). Subsequently, Argyle (1958) concluded that intelligent students are less likely to be religious. More recently, scientists have shown a striking paucity of religious belief (Ecklund et al., 2016), particularly within the elites of the National Academy of Sciences (Larson and Witham, 1998) and the Royal Society (Stirrat and Cornwell, 2013).
Psychometric population studies have now firmly established that religiosity influences cognitive style (Shenhav et al., 2012), and that religiosity and intelligence negatively correlate (Verhage, 1964; Pargament et al., 1998; Nyborg, 2009; Gervais and Norenzayan, 2012; Pennycook et al., 2013, 2014; Razmyar and Reeve, 2013; Zuckerman et al., 2013). Furthermore, it has been reported that IQ and disbelief in God correlate at r = 0.60 across 137 countries (Lynn et al., 2009).
The cognitive sciences are establishing a mechanistic understanding of the religiosity effect. For example, it has been seen that religious background modulates visual attention (Colzato et al., 2008). Lesion studies have demonstrated that ventro-medial prefrontal cortex lesion patients have elevated scores of religious fundamentalism (Asp et al., 2012). Experimental studies have demsontrated that increases in religious fundamentalism relate to increases in memory recall accuracy and higher rates of false-positives in a memory task (Galen et al., 2009). Religious fundamentalism has also shown modest positive correlations with life satisfaction (Carlucci et al., 2015) and negative correlations with cognitive flexibility (Zhong et al., 2017) and openness (Saroglou, 2002; Carlucci et al., 2011, 2015).
Dual-process models (Evans, 2008) assert that cognition is composed of intuitive and logical information processing. Individual differences in cognitive style have been related to the propensity to engage logical processes during problem solving (Stanovich and West, 1998). Meanwhile, recent experimental evidence has demonstrated a link between religiosity and cognitive style (Gervais and Norenzayan, 2012; Pennycook et al., 2014). From this, a prominent hypothesis has emerged which suggests that the religiosity effect is underpinned by cognitive-behavioral biases that cause poorer detection of situations in which intuition and logic are in conflict (Pennycook et al., 2014). Put simply, religious individuals are less likely to engage logical processes and be less efficient at detecting reasoning conflicts; therefore, they are more likely to take intuitive answers at face value and this impairs performance on intelligence tests. More broadly, from the perspective of this “dual-process” hypothesis, religious cognition is facilitated and hallmarked by intuitive decision making (Norenzayan and Gervais, 2013; Morgan, 2014; Oviedo, 2015).
It can be predicted from this hypothesis that the religiosity effect should be particularly disadvantageous for handling problems with counterintuitive answers; however, as a cognitive-behavioral bias, rather than reduced cognitive capacity per se, it follows that religiosity may not affect all tasks that involve reasoning. Reasoning tasks without intuitively obvious but logically correct answers may engage religious individual's latent ability to resolve complicated problems.
Here, we apply a novel combination of analyses to data from two Internet-cohort studies with detailed sociodemographic questionnaires and performance data from multiple cognitive tasks. Critically, these cohorts are large enough for the religiosity effect to be reliably examined in relation to, and while factoring out, a range of potentially confounding sociodemographic factors.
In study 1, we test four predictions of the dual-process hypothesis. (1) The religiosity effect should be greatest for reasoning latent variables as resolved via factor analysis. (2) The religiosity effect should be greatest for reasoning tasks designed to involve conflict resolution. (3) The religiosity effect should be in addition to, and not dependent on, other sociodemographic variables. (4) The pattern of the religiosity effect across tasks should differ qualitatively from those observed for other sociodemographic factors relating to latent reasoning ability.
In study 2, we replicate the findings of study 1 and test the further predictions that religious dogmatism mediates the religiosity-reasoning relationship at the levels of individuals (5) and religious groups (6). Finally, we test whether conversion to, or apostasy from, a religious group predicts cognitive performance (7).
