Field of Science

Why adults don't believe in god... perhaps!

Earthworms tunnel underground to aerate the soil - right? Go on, admit it... you were tempted to to answer yes to that one! OK well perhaps not, but these kinds of backwards connections between facts are surprisingly common. Generally, if a phenomenon (e.g. 'earthworms') have a functional consequence (aerating the soil), we have an inbuilt predisposition to believe that the phenomenon exists for the purpose of the consequence. Goats need mountains to climb. Therefore mountains exist to provide goats with something to climb on!

It's more common in children than in adults, and in kids it even has a label - 'promiscuous teleology' - coined by Deborah Kelemen at Boston University. Kelemen's latests study explores the question of why adults are less likely than children to give teleological explanations.

Basically, the test subjects (121 university students) were flashed a series of questions that were either teleological ('Water condenses to moisten the air') or physical ('Flowers wilt because they get dehydrated'), with a few questions thrown in just to make sure people were paying attention ('Zebras have black stripes because they eat coal', 'Children wear gloves to keep their hands warm').

They had to say whether the statement was true or false. The catch was that they only had a limited time to decide. I've pulled out two of the results in this graph. In the 'No time pressure' condition, they had 5 seconds. In the 'Time pressure' condition, they only had 3.2 seconds.

As you can see, there's a pretty clear result. Even with plenty of time to think about it, well educated adults succumb to incorrect teleological explanations pretty often. But when you put these people under time pressure, the acceptance rate jumps.

Kelemen reckons that this is because adults actually retain their childhood predilection for teleology, but they repress it consciously. They've been taught the real explanations, and when they have time to think it through, their education shows. But their gut instinct is for design.

And what does this have to do with a belief in a creator god? Well, Kelemen actually didn't find any correlation between a predilection for teleology and belief in god. But I have several problems with the analysis.

Firstly, all 'belief in god' was treated the same - but some people believe in a creator god, and some in a distant god. You would expect teleological beliefs to correlate not with just any religious beliefs, but only with belief in an active, creator god. Secondly, all the teleological responses - hurried and unhurried - were lumped. You would only expect the unhurried, conscious responses to match with a conscious belief in god.

Rather more interesting is the evidence that, as children grow up, their belief in god declines (see, for example, Francis 1989 for the UK and Uecker et al 2007 for the USA). If adults lose teleological beliefs by suppressing them as a result of education, is this perhaps at least a partial explanation of this loss of faith?

Deborah Kelemen, Evelyn Rosset (2009). The Human Function Compunction: Teleological explanation in adults Cognition DOI: 10.1016/j.cognition.2009.01.001

Francis, L.J. (1989). Measuring attitudes towards Christianity during childhood and adolescence. Personality & Individual Differences, 10, 695-698

Uecker, Jeremy E., Regnerus, Mark., Vaaler, Margaret L. (2007). Losing My Religion: The Social Sources of Religious Decline in Early Adulthood Social Forces, 85, 1667-1692


  1. When I was 11, a teacher asked me "Why is there oxygen in water?"

    The correct answer, in her mind, was "So fish can breathe".

  2. I think this has less to do with belief in gods and more to do with solipsism, or the limits of the imagination under stress. Picturing independent agency in anything other than the self usually requires deliberate conscious thought. But Kelemen is definitely on to something, and kudos to her for an intriguing experiment!

    Here's some promiscuous teleology for you. God exists because people believe in him!

  3. Perhaps, but they had some pretty good controls - including the 'bad teleology' questions. Anyway, 'More research needed', as they say!

  4. My gut feeling is in favor of the hypothesis that we retain our teleological leanings but suppress it in adulthood (especially since I just got done reading SuperSense, heh...)

    That said, a comment on this study: I wonder how much of the time pressure effect could have been due to imprecise parsing rather than actual acceptance of a teleological explanation. For instance, if I were to read in a hurry, I would have difficulty distinguishing between the sentences "water condenses to moisten the air" and "water condenses which moistens the air". i.e. when the teleological vs. physical distinction comes down to a single word, time pressure may have prevented the study participants from fully parsing what they were agreeing to.

    Who knows, hard to say...

  5. As I recall fro this study (its a while since I read it...), the questions were semantically similar but some were logically incorrect. So it was really a question of screening out the duff logic. Which of course is more difficult when you're put under a time pressure.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS