follow CCP

Recent blog entries
popular papers

Science Curiosity and Political Information Processing

What Is the "Science of Science Communication"?

Climate-Science Communication and the Measurement Problem

Ideology, Motivated Cognition, and Cognitive Reflection: An Experimental Study

'Ideology' or 'Situation Sense'? An Experimental Investigation of Motivated Reasoning and Professional Judgment

A Risky Science Communication Environment for Vaccines

Motivated Numeracy and Enlightened Self-Government

Making Climate Science Communication Evidence-based—All the Way Down 

Neutral Principles, Motivated Cognition, and Some Problems for Constitutional Law 

Cultural Cognition of Scientific Consensus

The Tragedy of the Risk-Perception Commons: Science Literacy and Climate Change

"They Saw a Protest": Cognitive Illiberalism and the Speech-Conduct Distinction 

Geoengineering and the Science Communication Environment: a Cross-Cultural Experiment

Fixing the Communications Failure

Why We Are Poles Apart on Climate Change

The Cognitively Illiberal State 

Who Fears the HPV Vaccine, Who Doesn't, and Why? An Experimental Study

Cultural Cognition of the Risks and Benefits of Nanotechnology

Whose Eyes Are You Going to Believe? An Empirical Examination of Scott v. Harris

Cultural Cognition and Public Policy

Culture, Cognition, and Consent: Who Perceives What, and Why, in "Acquaintance Rape" Cases

Culture and Identity-Protective Cognition: Explaining the White Male Effect

Fear of Democracy: A Cultural Evaluation of Sunstein on Risk

Cultural Cognition as a Conception of the Cultural Theory of Risk

« WSMD? JA! Two trends in 1--nuclear power risk perception | Main | Coming up this week: Religiosity & Rationality Panel »

Cognitive Science of Religion & acceptance of evolution

Rational reconstruction of remarks at this eventSlides here.

1. The presentations today are united by a common framework: the Cognitive Science of Religion (CSR). CSR uses the concepts and methods of the decision sciences to enrich scholarly understanding of religious behavior (Barrett 2007; van Slyke 2016).

The CSR framework, however, does not uniquely determine any particular account of religiosity. Indeed, in my remarks I will describe a pair of opposing CSR research programs, and then use CSR to conduct a simple empirical test of their relative plausibility.

2. The first CSR program can be called the “bounded rationality theory of religion” (BRTR)*.  Drawing on dual process reasoning theories, BRTR attributes religiosity to an overreliance on heuristic, “System 1” information processing.  From childhood onward, we encounter functional objects—from pencils to automobiles, from roller skates to intercontinental ballistic missiles, etc.— that were purposefully constructed by an intelligent agent to advance her aims.  It is therefore quite natural (so to speak) to infer that functional objects in nature—from the stripes of zebras to the opposable thumbs of human beings—are likewise the inventions of a purposive, intelligent agent (e.g., God).

Overcoming this intuition is hard.  To do so, we must employ conscious, effortful “System 2” reasoning. This taxing form of information processing not only interrogates the System 1 intuition of intelligent design in nature; it also motivates us to attend to scientific evidence, which documents the contribution that mindless natural processes, tempered by natural selection, make to the order we observe in the cosmos (Shtulman, 2014; Bloom & Weisberg, 2007; Shtulman  & Calabi, 2012; Gervais 2015).

3. The second CSR program is the “expressive rationality theory of religion” (ERTR). ERTR resists the equation of religion with irrationality. Forming accurate perceptions of facts is not the only thing that people do with their reason; they also use it to secure and protect their status in groups, the members of which are united by their shared cultural values.  Religion is a font of such groups.  If their goal is to convey their membership in and loyalty to a religious community, it is perfectly rational for people to engage information in a manner that induces beliefs commonly associated with such a group (Kahan & Stanovich 2016; cf. Yonker et al. 2016).

People can be expected, moreover, to use both System 1 and System 2 reasoning to advance this end (cf. Yonker et al. 2017). Rather than use their cognitive proficiency to resist intuitions supportive of their religious identities, people can be expected to use their capacity for conscious, effortful information processing to ferret out evidence supportive of their groups’ positions and to explain away evidence that threatens to undermine the same. This is the same process (“Motivated System 2 Reasoning, or MS2R) that explains why the citizens who are most proficient in critical reasoning are also the most polarized on politically contested facts relating to climate change, gun control, nuclear power, etc.

4. Whether humans evolved from earlier species of primates is a central point of contention among (certain) religious groups, on the one hand, and secular ones, on the other.  Examining how this belief relates to religious identity and to cognitive reasoning proficiency thus furnishes an opportunity to test the relative plausibility of BRTR and ERTR.  .

In such a test, one administers to an appropriate sample (e.g., a large general population one) three types of measures.  The first consists of one or another cognitive proficiency assessment, such as the Cognitive Reflection Test (CRT) or the Ordinary Science Intelligence scale (OSI_2.0).  The second examines respondents’ religiosity (assessed with self-reported frequency of church attendance, frequency of prayer, importance of religion in life, etc.).  Finally one looks at the respondents’ “belief” in human evolution.

BRTR and ERTR.make opposing predictions about what we’ll see.  If, as BRTR conjectures, religious belief signifies over-reliance on heuristic System 1 reasoning, then acceptance of human evolution should increase as individuals become more proficient in, and more disposed to use, System 2 reasoning.  Religious respondents are much more likely than nonreligious ones to reject human evolution. But if BRTR is correct, then we ought to observe this gap shrink as religious respondents become progressively more proficient in cognition and hence progressively more likely to interrogate their intuition and more likely to consider the strength of the scientific evidence on humans’ natural history.

ERTR’s predictions are almost the opposite of BRTR’s.  If ERTR’s is correct that positions on human evolution are identity-expressive, then the gap between religious and nonreligious respondents should increase as religious and nonreligious respondents become more adept at alternatively crediting identity-affirming evidence and dismissing identity-threatening evidence on human evolution.  If that prediction is right, moreover, then we should expect these tendencies to largely cancel each other out when we investigate the impact of higher cognitive proficiency for the respondents as a whole.

The evidence clearly supports ERTR. As it predicted, religious and nonreligious subjects do not converge as their CRT scores increase. On the contrary, polarization increases.

The effect is even more dramatic when investigated in relation to scores on the OSI_2.0 assessment. Likely this is the case because CRT, as a result of its difficulty, fails to measure variance in half the population, whereas OSI_2.0 measures variance uniformly across the entire population.

In any case, because nonreligious subjects become more convinced and religious ones less as their cognitive proficiency increases, the net effect of higher CRT and OSI_2.0 scores is close to zero—another result that defies BRTR‘s, but not ERTR’s predictions.

Finally, we do a very compact experiment that corroborates these results.  Again, religious respondents are much less likely to accept human evolution than are nonreligious ones, a gap that only increases as cognitive reflection and science comprehension increase.  However, if we simply add to the evolution item (“Human beings, as we know them today, developed from earlier species of animals. (True/false)” the clause “According to the theory of evolution, . . .,” the gap between religious and nonreligious respondents nearly disappears, particularly among respondents who score the highest on cognitive proficiency scores (Kahan 2017). 

So religious respondents, it turns out, are as familiar as nonreligious ones with what science knows about human evolution.  All one has to do to coax that knowledge-revealing response out of them is remove the confound between an identity-affirming and a science-comprehending answer to the conventional evolution-acceptance question.   Nothing so extravagant as BRTR‘s dual-process theory explanation is needed to understand why religious and nonreligious respondents answer the standard item differently.

5.  But now for the meta-experiment: did you feel like you learned anything about religious belief or practice from this Cognitive-Science-of-Religion guided analysis of who believes what and why about human evolution? Do you think others would?

*I use the subscript “R” to emphasize that the “bounded rationality theory of religion” is related to but distinct from the forms of bounded rationality that generate group-based perceptions of risk and policy-related facts.


Barrett, J.L. Cognitive Science of Religion: What Is It and Why Is It Important? Religion Compass 1, 768-786 (2007).

Bloom, P. & Weisberg, D.S. Childhood origins of adult resistance to science. Science 316, 996-997 (2007).

Gervais, W.M. Override the controversy: Analytic thinking predicts endorsement of evolution. Cognition 142, 312-321 (2015).

Kahan, D.M. ‘Ordinary science intelligence’: a science-comprehension measure for study of risk and science communication, with notes on evolution and climate change. J Risk Res 20, 995-1016 (2017).

Kahan, Dan M. and Stanovich, Keith, Rationality and Belief in Human Evolution. Cultural Cognition/Annenberg Public Policy Center Working Paper No. 5. (September 14, 2016) Available at SSRN:

Shtulman, 2014; Bloom & Weisberg, 2007; Shtulman  & Calabi, 2012; Gervais 2015.

Shtulman, A. Science v. intuition: why it is difficult for scientific knowledge to take root. Skeptic (Altadena, CA) 19, 46-51 (2014).

Van Slyke, J.A. The cognitive science of religion (Routledge, 2016).

Yonker, J.E., Edman, L.R.O., Cresswell, J. & Barrett, J.L. Primed analytic thought and religiosity: The importance of individual characteristics. Psychology of Religion and Spirituality 8, 298-308 (2016).

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (6)

IMHO, you have to look more closely at who retains, past childhood, their religious identity and who drops it. In particular, for Christianity, there has been a big drop off in the formerly "mainstream" traditional and more intellectually liberal denominations membership. This is composed of people who retain their religious identity while also accepting science. People for whom the Bible is more of a historical document than directives straight from God. For whom the teachings of Jesus as to how to have a life well lived are more important than repentance and ultimate, post death salvation. This sort of religion seems to have less of an identity grip.

Does education re-enforce religious beliefs or is it true that among those who identify themselves as religious, there is an increasingly disproportionate number in the ones that remain as such that are fundamentalist rather than those in the traditional denominations as education levels increase?

Furthermore, what happens f you took a group of children raised as religious fundamentalists, and tracked them through adulthood? I would posit that a substantial number of the ones who went to college would drop that belief system. Especially if they got interested in science. An acceptance of science might undermine their childhood belief system. They might drop their religious identity entirely rather than go out and find some less fundamentalist group to identify with.

Additionally, most college students get very little exposure to science. So most people would not be giving their answers based on knowledge obtained by a course in evolutionary biology anyway. The answers essentially still have a lot of shaping by identity groups. Religious fundamentalists support their own institutions of education to even further limit outside ideological effects, and so you'd also have to include what it is you think a college education is. Is Liberty University analogous to Yale?

April 4, 2018 | Unregistered CommenterGaythia Weis

"So religious respondents, it turns out, are as familiar as nonreligious ones with what science knows about human evolution. All one has to do to coax that knowledge-revealing response out of them is remove the confound between an identity-affirming and a science-comprehending answer to the conventional evolution-acceptance question."

According to the theory of astrology, the positions of the planets in their orbits affects your love life. Yes or no?

According to Newtonian physics, the speed of light can be exceeded. Yes or no?

According to the theory of null hypothesis tests, if the probability of an observation assuming the null hypothesis is less than 5%, the hypothesis should be rejected. Yes or no?

According to the theory of "bounded rationality", polarisation will decrease with increasing scientific understanding. Yes or no?


Which of the above examples would you say are cases of "identity-affirmation"? How would you predict they would appear in your graphs?

(Seriously, Bayesians-vs-Frequentists does have a certain tribal vibe to it! But still...)


April 5, 2018 | Unregistered CommenterNiV

@Niv-- for normal person, none of those is identity-affirming/-threating. For you, on other hand, all are

April 5, 2018 | Registered CommenterDan Kahan

Intriguing! Can you explain further?

April 6, 2018 | Unregistered CommenterNiV

It would help greatly here if Dan - as far as I know the only one who can post graphics - would explain his terminology. Does "normal" refer to stable Gaussian (or at least Pareto-Levy) distributions? Or to some cultural variable, whose parameters vary over time?

Generally it would be great to enable graphics beyond links to .pdf files - probably too much trouble with the software, though. This is related:

April 8, 2018 | Unregistered CommenterEcoute Sauvage

I think this is a joke, but it appears to be related to religion:

April 9, 2018 | Unregistered CommenterEcoute Sauvage

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
Some HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <em> <i> <strike> <strong>