Home > Psychology, Religion, Science > Fundamentalism, Spirituality, and IQ

Fundamentalism, Spirituality, and IQ

Why are some people religious? Putting aside arguments over whether the claims of particular religions have factual validity or not (up-front declaration: I’m an atheist), the existence of wildly varying degrees of religious sensibility in every society studied is quite perplexing for psychologists, sociologists and anthropologists. Some colleagues and I have just had a paper published (PDF here) which attempts to provide a piece of this puzzle, focusing on the relationship between religious beliefs and general intelligence, or IQ.

For some time, psychologists and neuroscientists have made attempts to explain religion, ranging from the ill-fated ‘God Helmet‘ (the idea that religious experiences might be a form of temporal lobe epilepsy) to ideas about how religions bind us into communities, to approaches involving hypothesised psychological mechanisms such as the hyperactive agent-detection device (HADD; an evolutionarily advantageous ability which would see us ascribing ‘human-like agency [to the] environment that might not actually exist’ (p. 31), and misinterpreting, say, the rising of the sun as an intentional act which must therefore have some kind of supernatural actor). None of these theories are necessarily mutually exclusive – it’s clear a future psychological theory of religion will need to explain evidence from a wide variety of perspectives.

Of course, to many people, any attempt to scientifically explain their deeply-held beliefs will be highly offensive, but probably the most controversial research in this area has involved IQ. A wide range of recent studies have looked at connections between cognitive ability and religiosity. To take just three examples: this study found that atheism rates in 137 countries correlated highly (r = .60) with the country’s average IQ, while this one showed in a sample of students that high-IQ individuals are more likely to question religious beliefs, and this one found that atheist adolescents had higher IQs than their moderately religious peers, who in turn were higher in IQ than religious fundamentalists. There are plenty of other studies, too.

Could your score on tests like this predict how religious you are? (Yes!)

A common response to this kind of research is summed up by atheist blogger extraordinaire, PZ Myers, in this rather unhelpful and misguided post. Myers seems to misunderstand why we bother collecting scientific data when he says: ‘I’ve known lots of religious people who really are brilliant, and I also know lots of atheists who were sincerely religious once upon a time…’. Ironically, given he often criticises pseudoscientists for basing their arguments on anecdote, Myers seems to be saying that his personal experiences invalidate several statistical studies on large samples. Huh. Not so impressive.

What is impressive, to get back to the studies, is the consistency of the data. In every study to date (that I’ve seen), higher religious beliefs have been found to be associated with lower scores on IQ tests. Now, I know some people, like Myers, just hate IQ tests, possibly because they’ve read The Mismeasure of Man (which, as an aside, is looking less and less credible as time goes on), but IQ really is a part of mainstream psychological science, and I can assure you you’d find it very difficult to make an evidence-based argument against it having validity (I always like to point IQ-skeptics – and the far more irritating IQ-deniers – towards this Nature Reviews Neuroscience paper, which I think makes a pretty convincing case that IQ is both real and useful).

While those previous studies on religion and IQ are consistent in their results, they do have a number of limitations, outlined in our paper. For instance, we now know that religion isn’t a binary value (it isn’t just that people are religious, or they aren’t). Papers using the statistical technique of factor analysis have found that people vary on several different facets of religiosity, some reflecting the more fundamentalist, literal-minded types of religion, and others reflecting more ‘spiritual’ elements. Most IQ-religion papers don’t examine these. Others have used relatively small sample sizes, or – like one of those linked above – rely on aggregate data, which can be inaccurate. Finally, IQ isn’t the whole picture – no previous studies have looked at how personality factors, like openness to experience (which does seem to be linked to religion) might mediate the relationship between IQ and religious belief.

To get around these problems, we – that is, Gary Lewis, Tim Bates, and I – used data from the MIDUS, a huge-scale survey of thousands of adults in the US which asked them questions on almost everything, from their socioeconomic status to their occupational information, from their health behaviours to how well they sleep, and also included IQ and personality tests (it really is a goldmine). Thus, our study had the advantages of a large sample size (around 2300 people), and a wide range of sub-scales of religiosity (specifically: religious mindfulness, spirituality, religious support, religious identification, private practice, and fundamentalism – see the paper for definitions and sample questions). Controlling for level of education, trait openness to experience, age, and sex, we looked at the strength of the relationship between IQ and each of the religious variables in turn.

It turned out that higher IQ scores were significantly associated with lower scores on five of the six measures of religiosity – all of them except spirituality. Thus, the more intelligent you are, the less likely you are to be religious – except if you’re just ‘spiritual’. Higher IQ was most negatively associated with fundamentalism (the tendency to agree with statements like ‘the Bible is the word of God’).  The other associations were interesting, too – education level related very negatively to fundamentalism, while people high in openness were much more likely to report being ‘spiritual’, but less likely to be fundamentalists. Older people were more likely to report higher instances of prayer and identification with a  religious in-group, while females were more likely to be spiritual, but no more likely to be fundamentalists, than males.

How do we explain the IQ-related results? We think it’s most likely that higher-IQ individuals will tend to question the stories and claims of religions, which is perhaps why they’re particularly unlikely to believe the Bible is the final word. For skeptics, familiar with ‘New Age’ spirituality, our results might make some sense, too; those of average intelligence, but very high (some would say too high) openness to new experiences, might get involved with quasi-religious practices we’d characterise as ‘woo’ – crystals, distance healing, that sort of thing (maybe our paper is empirical confirmation of the old adage that if you’re too open-minded, your brain will fall out…). A future study we’re planning will look at how IQ and openness relate to beliefs in alternative medicine.

So, think of it this way: A set of basic cognitive tests which include ‘repeat back this list of numbers/words from memory‘ and ‘how many words can you think of that start with the letter B?‘ are a statistically significant predictor of what someone’s beliefs will be about the meaning of life and the origin of the universe. I think that’s a pretty fascinating finding, and worthy of further investigation, with the following two caveats. First, our study was almost entirely on Christians, meaning that one should be cautious in generalising the findings to other religions.

Second – and this isn’t really a caveat, but an important finding from the study – only a small part of the variance in religious belief was explained by IQ. That is, as mentioned above, we’ll have to look further than just IQ for our future super-theory of religious belief, and shouldn’t expect such a complex, multi-faceted phenomenon to be explained by intelligence alone. Our study provides new insights about which particular forms of religious belief are related to IQ, bringing us a teeny bit closer to an understanding of the psychology of religion.

  1. September 5, 2011 at 7:19 pm

    Congratulations on getting your paper published, Stu! I’ve never been more glad for the inclusion of an abstract so I can dodge all the statistical interpretation 😉

    Do you think education about woo and new-age spiritual mumbo jumbo could, over time, lead to a similar correlation in line with other measures of religiosity?

    • Stuart Ritchie
      September 5, 2011 at 8:57 pm

      Hope so! Watch this space for future research on just that 😉

  2. September 11, 2011 at 7:34 pm

    Excellent paper, excellent summary. Love your work Stuart, well done.

  3. Sean
    September 12, 2011 at 4:09 pm

    An interesting post. I’ll admit that it tends to confirm what I’d thought already, which is that intelligence is helpful but not sufficient for breaking out of religion.

    One thing that surprised me was in the Big Five meta-analysis you linked. I hadn’t expected such a strong link between conscientiousness and religiosity. Do you have any more information about that?

    • Stuart Ritchie
      September 13, 2011 at 1:18 am

      Afraid not, but the intelligence-personality-religion connection is something we’ll definitely be doing further work on in future, so watch this space!

  4. September 12, 2011 at 5:22 pm

    Good stuff. Keep it up.

  5. September 12, 2011 at 6:34 pm

    Fundamentalismo, Religiosidad, e Inteligencia (Fundamentalism, Spirituality, and IQ) https://timeoutofmindblog.wordpress.com/2011/09/05/fundamentalism-spirituality-and-iq/

  6. pwnzsaurus
    September 13, 2011 at 12:17 am

    Eh, the entire study is founded on an ad populum fallacy. Why? Because it seems to imply that either 1) religion is invalid because morons believe it, or 2) that only morons are religious.

    How does this work? Quite simply: by tracking trends in any part of the population, finding what the most intelligent among us believe, and pretending that that has some significant bearing on the validity of those people’s beliefs.

    If you had asked the majority of scientists and intellectuals 300 years ago whether there was a god, they’d reply affirmatively. And yet even though Leibniz and Newton were “highly” religious — and extraordinarily intelligent — those facts have no bearing on the truth of the matter.

    It’s akin to arguing that “all academicians (with high IQ’s) believe in inherent motion” and pretending that you’ve thereby proven that inherent motion exists.

    • Stuart Ritchie
      September 13, 2011 at 1:16 am

      Oh dear. You did actually read the article, yes? You don’t even have to get past the first two sentences to see that the question is ‘why are people religious?’ and is explicitly not ‘are the specific claims of religion(s) true?’. Just read those first two sentences, and then look back at what you said. You really have made yourself look rather silly.

      • pwnzasaurus
        October 22, 2011 at 8:16 pm

        Oh dear. You didn’t actually read, or understand, any of what I wrote, did you?

        The article has to do with the “relationship” between IQ and religious belief. The implication, or conclusion, that most people would draw from the study — and that you seem to want to draw — is that people who are religious are unintelligent, or, conversely, that unintelligent people are religious. Hence, people “are” religious because they’re unintelligent.

        That aside, the study is flawed, again, in two respects:

        1) It’s necessarily focused on an ad populum fallacy. Why? Because it takes a belief that is popular and widely held by academics, correlates it with their IQ, and tries to draw some causation out of that correlation.

        2) If the correlation between IQ and religion has any bearing on either of the two, the study shoots itself in the foot. Why? Because you can point to any time in history where academics believed some thing, draw a correlation between the thing and the academics’ IQ, and pretend you’ve found a cause for it. Whether it’s religion, Aristotelian physics, or Cartesian dualism.

        Although you throw in a caveat at the end (“[we] shouldn’t expect such a complex, multi-faceted phenomenon [religion] to be explained by intelligence alone”), if you truly believed that were true, then your entire study is a moot point, isn’t it?

      • pwnzasaurus
        October 22, 2011 at 8:21 pm

        Before you leave an angry reply, let’s answer one simple question first:

        What was the purpose of this study? What did it intend to show?

        If it intends to show that, in the modern era, most unintelligent people are religious, you’re doing nothing to reveal the “psychology” of religion, you’re merely pointing out an historical fact.

        However, if it intends to show some causation between intelligence and religious belief, the study fails for the reasons above.

      • pwnzasaurus
        October 22, 2011 at 11:12 pm

        Let me clarify something, and kind of negate my last two responses:

        Now that I’ve re-read the article, and my response, my original response was directed at the “inferences” that would be made from a study like this.

        If you’re trying to prove causation through perpetual correlation, that’s fine. E.G., throughout time and history unintelligent people have tended to be religious, therefore if you meet an unintelligent person, they’re “probably” religious.

        However, the mistake (not made in this article) is in assuming that that has: 1) any bearing on the validity of religious belief in general (i.e., the belief in a god), or 2) that somehow the propensity the “unintelligent” have for being religious somehow makes religion itself “unintelligent”.

        Hence, the intelligence of religious people has no bearing on either religion or intelligence.

        So I take back my first response to your response, because I had forgotten what you had said in the article.


  7. September 13, 2011 at 12:21 am

    Now that is interesting . . .

    • Stuart Ritchie
      September 13, 2011 at 1:17 am

      Glad you thought so, thanks!

  8. September 13, 2011 at 3:23 am

    > ‘repeat back this list of numbers/words from memory‘ and ‘how many words can you think of that start with the letter B?‘ …

    For what it’s worth, I’d do terrible on those kinds of verbal tests. I’m also an atheist and quite intelligent in non-verbal ways (written words are much better for me). The problem is that “statistically significant associations” don’t really tell you much about individuals.

    • Stuart Ritchie
      September 13, 2011 at 11:24 am

      Well, yes they do – the statistically significant associations between most of those IQ tests (the positive manifold) tells us that there’s probably one factor underlying most people’s intellectual abilities. Take a look at the paper I linked to in the post that explains the g factor (see Box 1 especially): http://www.larspenke.eu/pdfs/Deary_Penke_Johnson_2010_-_Neuroscience_of_intelligence_review.pdf

      This isn’t to say that everyone is either good at all tests or bad at all tests; it just means that in general, people who are good at one type of test will be good at all of them.

    • September 14, 2011 at 4:28 pm

      I’m the same – I think mostly in non-verbal ways, and sometimes struggle with word tests – even though I still usually do quite well in them compared to the average, but not as well as I should. But I also find them very boring, and tend not to work as hard at them, so that might be a reason.

  9. Mahina_D
    September 13, 2011 at 10:53 am

    I would be interested to see what kind of correlation there is between emotional IQ and religiosity. Any plans to look into that?

  10. September 14, 2011 at 4:20 pm

    If these findings are correct – and I have no reason to doubt them, then it is obviously so because God designated it thus! ;}

    • September 14, 2011 at 4:23 pm

      I understand that 40% of Americans believe in the literal truth of Genesis. Well that correlates quite nicely with the number of Americans of below average intelligence, doesn’t it?

  11. bpesta22@cs.com
    September 14, 2011 at 7:40 pm

    Thanks for citing my stuff– good paper too!

    B Pesta.

  12. NotAProphet
    September 15, 2011 at 10:12 am

    Maybe a leap, but this might go some way to explaining why those who believe most fervently in religion resort to using brute force and violence in an attempt to “defend” their beliefs and silence critics.

  13. tracy
    September 18, 2011 at 9:07 am

    Right there, in the very first paragraph, is the invalidation of the entire conclusion:
    “general intelligence”
    IQ only measures intelligence in as much as this intelligence relates to intellectualism and academia.
    There is no such thing as a reasonable, objective, test of intelligence, because the very nature of the word is subjective. Any researcher who claims “intelligence” at any degree of mathematical causation is delusional.
    The only thing that was demonstrated here is that scientists are less religious than the general population, woopdeedoo!!! Tell us something we didn’t know! Both being a scientist and being atheist are highly dependent on the very same educational pre-condition: critical thinking… So the claim to a randomised statistical approach for this poor study are unfounded, invalidating the entire conclusion.

    • Stuart Ritchie
      September 18, 2011 at 9:26 am

      I’m just going to assume this comment is a joke, including as it does all the standard, clichéd, ignorant critiques of intelligence research. Also it’s clear you didn’t read the post or the accompanying paper.

      • tracy
        September 19, 2011 at 4:52 pm

        The only cliché here is people still clinging to the long debunked notion that IQ is an efficient measurement of intelligence.

      • Stuart Ritchie
        September 19, 2011 at 5:30 pm

        Interesting. Could you refer me to where it was debunked, and by whom? I must’ve missed that. The hundreds of eminent scientists currently publishing IQ research in the top journals in the world must’ve missed it too. But I’m sure you’re right.

  14. September 18, 2011 at 7:26 pm

    I found this really fascinating and sometimes I do like PZ Myers’ blog but you nailed why that post of his was SO… bad. Lol. I had heard that IQ wasn’t to be trusted… But I never really looked into that… I basically believe you and this post. Found it on:

    Very interesting, honestly.

  15. September 20, 2011 at 7:08 am

    Oh tracy, you and I could have a lovely discussion about how mindblowingly difficult a task it’d be for someone to measure and compare our intelligence. However since neither of us respected experts in the field, mightn’t it be wise to defer to academia to advise whether or not that task is currently impossible?

  16. September 20, 2011 at 7:48 am

    Hi Stuart, care to comment on longitudinal studies in this area? do individuals traverse religious sub-scales as they age? as they are exposed to certain forms of education? other factors?

    • Stuart Ritchie
      September 20, 2011 at 9:42 am

      I’m not aware of any such studies – which is why we’re currently trying to find a good dataset to do one ourselves!

  17. Michelle
    September 24, 2011 at 5:27 pm

    Hey, I’m just wondering how you explain anomalies? I mean, if you are concluding, which you seem to be, that intelligence is a good predictor of who will be religious (especially fundamentally religious), then what do you think to the fact that you do meet plenty of people who are both fundamentally religious and very intelligent and well educated? I understand that the average is lower (although I think it might be interesting to do a study like this somewhere that religion, especially fundamentalism, is less socially acceptable, such as Britain and France and see if the results differ?) but, as I’m sure you know, from meeting religious people if nothing else, that not all are babbling idiots! I know that with this sort of thing you would never find there to be no anomalies, but do you have any ideas on why these anomalies might occur? How is it that as the general population get more educated, they get less religious, but for some it only increases their religiosity?

    I would also be interested to know a bit more detail (without purchasing the article, sorry!) what other questions you used to determine “fundamentalism”? Was it simply if the person held to the beliefs taught by a church (which?) or did you take the things The Bible teaches and ask whether people believed them and held true to them? You have given one example, would it be possible to hear a few more?

    I’d like to add the disclaimer here that I’m in no way an expert on any of these things, I just find this area of research interesting. As for debates as to whether IQ is a good measure of intelligence and the like, I’ll leave that to you guys (but will add the totally anecdotal evidence that I know a PhD student with an official IQ in the 70s, in a large part due to what is probably undiagnosed severe dyscalculia, but I do understand that IQ tests may well be useful as long as these sorts of things are taken into account).

  1. September 6, 2011 at 10:25 pm
  2. September 11, 2011 at 9:09 pm
  3. September 13, 2011 at 12:02 am
  4. September 17, 2011 at 11:52 am
  5. September 18, 2011 at 1:01 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: