Field of Science

In West and East, it's easier to remember things that are a bit weird

Look around the world, and you'll find that most gods and magical entities are surprisingly similar to regular people, but with one or two magical powers. The same goes for most works of fiction - your typical superhero is, in most respects, a pretty regular guy.

At first sight, the Judaeo-Christian god seems to be an exception to this rule. But if you look at how Christians often relate to their god, never mind how it is portrayed in the Bible, it's not really such an anomaly.

Back in 2002, Pascal Boyer proposed that this was not an accident.  He suggested that mundane, everyday objects are instantly forgettable, and that really weird stuff is just to hard to remember. What really stands out, and what our brains intuitively latch onto, are things that deviate only a bit from the normal. He called these 'minimally counterintuitive ideas' (MCI).

Over the past 10 years, research into this idea has produced some support, but also some experimental results that didn't fit the predictions.

The most recent experiment has taken advantage of the virtual reality world of Second Life. By using a virtual environment, they able not only to create a set-up that would be impossible in the real world, but to study people from different cultures and people who don't typically participate as subjects (i.e. people other than students).

The experimenters, Ryan Hornbeck at the University of Oxford, UK, and Justin Barrett (now at the Fuller Theological Seminary in California) created a kind of virtual museum containing 18 objects. Half of these were everyday (such as a ball hitting a wall) and half had something weird about them (such as a parrot that disappears).

They lead their study subjects (50 native-English speakers living in the West and 50 native-Chinese speakers living in Asia) round this museum, and let them briefly view each object. Afterwards they tested them on how many they could remember. After a while (up to 15 days later) they were invited to be tested again.

As shown in the graphic, the Westerners were more likely to remember the 'minimally counterintuitive' objects than the intuitive objects - whether tested immediately or after a period of days. For the Chinese speakers, there was no difference in immediate recall, but there was a difference in the delayed recall.

They found that the longer the period until the second test, the more likely it was that the intuitive items would be forgotten. That's what you would expect - but that's not what they found for the MCI objects. For these, memorisation seemed to be constant, whatever the delay.

That suggests that MCIs that actually get remembered are less likely to be forgotten, compared with intuitive memories.

Intriguingly, they also found that age had an effect.The older participants were equally good at remembering MCI and intuitive objects. Younger participants, however, were significantly better at remembering the MCI objects.

The researchers were intrigued by why that might be. What they end up suggesting is that, for young adults who are still learning about the world, it pays off to devote a lot of mental energy to memorising and assessing things that deviate from the expected.

By later adulthood, however, "the most important exceptions are likely to have already been encountered". As a result, anything that turns up that looks weird and out of place to an older person is likely just to be a one-off aberration, and so not worth paying too much attention too.

A sobering thought!
Hornbeck, R., & Barrett, J. (2013). Refining and Testing “Counterintuitiveness” in Virtual Reality: Cross-Cultural Evidence for Recall of Counterintuitive Representations International Journal for the Psychology of Religion, 23 (1), 15-28 DOI: 10.1080/10508619.2013.735192

Creative Commons License This article by Tom Rees was first published on Epiphenom. It is licensed under Creative Commons.


  1. Did the researchers pose a possible explanation for the difference between East and West in intuitive recall?

    Also, in the graphic it looks like the Westerners remembered more overall. Did they address that issue?

  2. It's certainly an interesting experiment but I don't see how it has any relationship to your opening gambit.

  3. @B.T. Newberg - no, they say that there were no significant differences between Western and Eastern participants, and also say that no differences were hypothesized.

  4. @Jayarava, the basic idea is that MCI concepts are more culturally transmissible, since they are at the sweet spot of being different enough to be memorable but not so different as to be too complex to memorize.

    So they were testing whether slightly odd things were easier to remember (they weren't) and easier to recall (they were). So this supports the idea that the reason we have gods that are much like humans but with a few distinctive features is that they are 'minimally counterintuitive'

  5. Hi Tomas,

    I think there are better and stronger arguments and evidence for why we might have created gods mainly like us (at least like us socially). Apophenia for instance. Or our being social empathetic animals, with strong preferences for people that look like us.

    Mainly... because many of the Egyptian gods, for example, were slight odd in having animal heads. Assyrian gods were slightly odd in having wings (as were the Hebrew angels based on them). Tibetan wrathful deities are extremely odd, as are many of the Indian deities. Viṣṇu takes many forms: turtle, boar, dwarf amongst them. Lots of cultures have gods in animal forms.

    The more I think of it the more I don't think many cultures did have gods "much like us". Most were quite counterintuitive in form, even when their social relations are similar to ours. It's really only the Greeks (and the peoples they influence) that go in for humaniform gods in a big way.

    I think the research is interesting, but I'm not convinced of the case you are making. It might contribute in those minority of cases where gods are roughly human, but other factors are surely much more important.

  6. You know, I'm inclined to agree with you. It's an interesting hypothesis, but I don't think it's well supported by the data,and then there's all the points you make. It would be interesting to see some research to see if there is any consistency in the way gods are represented around the world. Maybe the MCID hypothesis is actually a non starter.

  7. Not sure that egyptian, etc, etc gods are a point against MCI ideas.
    As I understood it, MCI doesn't just mean physical appearance, but rather an overall look at the idea.

    I'm no expert in ancient Egyptian religion, but their gods still had personalities like us, likes and dislikes. They could hear and talk(after a fashion), they eat, they fight, they can get sick, etc, etc...

    Sure they might have a jackal ead, but they're not all THAT different from us. Just ... enough to be a god. Enough to catch your attention and help them lodge in memory and be worth talking about

  8. Kassul the distinction Tomas draws repeatedly is us and "slightly odd things" or images that were "minimally counter intuitive". You can certainly split hairs on this if you wish, but I think this (interesting) effect has little explanatory power for human gods and that we already have much better explanations. In the end Tomas agrees.

  9. I think the one problem the theory has is in definition. It talks about "minimally" counterintuitive, but what exactly that means when it comes to gods is pretty much open to speculation. The idea is that if gods were any more or less strange than they are, they would be harder to remember, but I haven't seen that tested directly.

  10. I'd like to see more experiments (like this one) that use people from different parts of the world. I think for too long many test results reflect the reactions/behaviour of Western college kids.

    Now, I may not be reading the graph right, but I'm not sure all of your observations are correct. You write, "For the Chinese speakers, there was no difference in immediate recall, but there was a difference in the delayed recall." It appears that there was only a difference in the delayed recall of MCI, but not Intuitive. Maybe that's what you meant, though it was unclear to me.

    You also write, "They found that the longer the period until the second test, the more likely it was that the intuitive items would be forgotten." But this doesn't appear to be the case with the Chinese, does it?

    Anyways, thanks for blogging about this. Food for thought.

    @ Jayarava, you write " Or our being social empathetic animals, with strong preferences for people that look like us." I think there is something to this, and it reminds me of Gene Roddenberry's insistence on making most alien species be humanoid. Of course, it was easier to film, but he also believed people would be able to relate better.

  11. Why is everyone discussing this study as if there were no confidence intervals displayed in the graph? The differences that were found were nowhere near being statistically significant. This study found nothing of any scientific value, so why discuss it?

  12. I noticed the same thing baopu81 did. The Chinese subjects in the delayed response actually remembered the Intuitive better than the MCI, the reverse of how the westerners performed.


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