Home > Education, Psychology, Science > What Dungeons & Dragons Tells Us About Your Brain

What Dungeons & Dragons Tells Us About Your Brain

Allow me to be deeply geeky for a moment here: it seems Dungeons & Dragons was only half right.

Those slavering, lonely sad sacks among you who are familiar with creating characters in this role-playing, roll-dicing game will know that, when you generate your in-game avatar, you are confronted with a list of attributes (such as Strength, Wisdom, and Charisma) from which you choose if you’d like to be a charmingly pathetic weakling, a musclebound oaf, or anywhere in between. You then pick a class (such as Fighter, Wizard, or Thief), which describes how you’ll deal with your exciting adventures – hack ‘n’ slash with a big axe, or cower at the back with Magic Missile.

So the overall message is this: people vary a lot, but you can still put them into a taxonomy of ‘classes’, ways in which they’re comfortable with dealing with the world. And now, as I hit rock bottom in the ‘tortured metaphor to open a blog post’ stakes, I will assess how close  a fit the D&D system is to real human psychology.

Here is a dungeon.

First, let’s have a look at those attributes. Surely it’s silly and insulting, since we know that people are fickle and changeable, to try to classify them on spectra of arbitrary traits? Some people think it is. There’s a school in psychology known as the ‘idiopathic’ approach to personality, which states that people’s behaviour comes from random, ever-changing, and unknowable causes, and we shouldn’t try to wrestle such a morass of complexity into little boxes. But is that really the case? Surely everyone admits that there are underlying aspects of personality, upon which you can make predictions about behaviour? Your worrisome friend, for instance, is going to go into meltdown every time he has a job interview, whereas you, much more phlegmatic, take it in your stride.

Most psychologists in the field of Individual Differences accept this is true, to differing extents. Back in the 1930s, the wonderfully-named psychologists Allport & Odbert collected all the terms in the English Language that they could find which described people’s personality. They found a whopping 18,000(!). Grouping these into themes, making them into questionnaires, and asking people to rate themselves on each term, formed the basis of the science of personality.

Once you have a big load of data showing how people rate themselves, you can do some statistics. A technique called factor analysis is used to draw out underlying correlations between all the terms – in other words, finding which descriptive words tend to stick to other ones in people’s personalities. These bunches of terms can then be given names, and classified as factors. The current consensus (though by no means accepted by all psychologists) is that 5 factors come out. These are (with descriptors taken from here):

  1. Openness -appreciation for art, emotion, adventure, unusual ideas, curiosity, and variety of experience.
  2. Conscientiousness – a tendency to show self-discipline, act dutifully, and aim for achievement; planned rather than spontaneous behavior.
  3. Extraversion – energy, positive emotions, urgency, and the tendency to seek stimulation in the company of others.
  4. Agreeableness – A tendency to be compassionate and cooperative rather than suspicious and antagonistic towards others.
  5. Neuroticism – A tendency to experience unpleasant emotions easily, such as anger, anxiety, depression, or vulnerability.

So, once you’ve got your factors, on which people may rate very high, very low, or anywhere in between, are there objective ways of assessing how close a fit they are to reality? Indeed there are – if you can make predictions about people’s lives on the basis of their personality traits, you’ve got a good case that they aren’t just imaginary.

In a review in 2007, Brent Roberts and colleagues analysed a mountain of data from personality studies, and concluded that you can indeed make successful predictions about major life events such as divorce, employment, and even mortality, on the basis of some of these personality factors. People high in conscientiousness, for instance, have a lower chance of divorce than people high in extraversion, and those high in neuroticism and agreeableness tend to keep their jobs for longer. While this stuff may seem obvious, it’s actually critically important – showing that the science of personality has a firm grasp on reality.

There is, of course, a lot more work to do. For example, psychologists have begun to link these factors to genetics. An interesting new idea is to rephrase the 5 factors in terms of reward-responsiveness (for instance, high extraversion could be a tendency to see interaction with others as a reward in and of itself). Another major question regards why these personality factors evolved in the first place. Personality psychologists clearly have their work cut out.

The future looks bright for personality psychology, and at least the idea behind the Dungeons & Dragons attribute system (though definitely not the names or concepts of the D&D attributes themselves) is vindicated. But what about the class system? Can we reliably put people into groups based on the strategy they use to interact with the world?

Ask most schoolteachers, and they’ll say yes. This is because the concept of ‘learning styles’ pervades teacher training courses, and it goes almost unquestioned that it is possible to group children into a taxonomy of ‘visual’, ‘auditory’, ‘kinaesthetic’ (and so on) learners, and that specific types of learning will work only for each specific subtype of child.

There are dozens of different versions of this idea, some more pseudoscientific than others. The worst types will include bullshit about  neuro-linguistic programming (the only stuff you’ll hear about NLP that isn’t bullshit is people telling you ‘it doesn’t work’) as well as nonsense about ‘right brain’ and ‘left brain’ personalities (you know, the pervasive myth that you can ‘unlock’ your right brain’s creativity). For now though, we’ll look only at the basic taxonomic theory.

If this claim were correct (or even approaching correct), it would be A Big Deal, and so the Association for Psychological Science tasked a group of neutral researchers, led by Harold Pashler, to assess the evidence. Their report, published in the excellent journal Psychological Science in the Public Interest, won’t exactly thrill proponents of the ‘learning styles’ hypothesis. Its conclusion?

The contrast between the enormous popularity of the learning-styles approach within education and the lack of credible evidence for its utility is, in our opinion, striking and disturbing. If classification of students’ learning styles has practical utility, it remains to be demonstrated.

Ouch. So what was it that made up their minds?

Here’s a hypothetical experiment. Take a group of 100 kids, and split them into two groups, depending on whether they’re ‘visual’ or ‘kinaesthetic’ learners. We can do this on the basis of criteria outlined by whichever ‘learning styles’ proponent we’ve picked, but it’s important we don’t tell the teachers, or the kids themselves, which group is which – let’s keep our experiment double-blinded. We’re going to teach the kids some basic maths – addition and subtraction. In week 1, give the ‘visual’ group an old-school (no pun intended) lesson with chalk and blackboard, and the ‘kinaesthetic’ group wooden blocks which they’re taught to pile up, physically doing the maths, hands-on. Then, give them a maths test. In week 2, give the kinaesthetic learners the blackboard lesson, the visual leaners the blocks, then test all them again.

Imagine we got the following results: both groups learned amazingly well in the first week, and abysmally in the second. The graph of their performance, then, looks a bit like this:

Test score with different learning styles

Here’s the problem: no such results exist. Pashler and colleagues note that not a single study fits this criteria: showing that a particular subset of kids learn better with their own method, and worse with another. This is the type of evidence that we’d need to really recommend these ‘learning styles’ ideas, and it simply isn’t there.

Here is a dragon.

The authors hasten to point out that they’re not saying that people don’t have preferences. Those kinaesthetic kids might have hated the visual lesson, for example. But the important thing is how these children have learned. And nobody has yet shown that you can assign kids taxonomic groups which correspond in any way to the reality of their brains’ handling of new information.

So – sorry, Dungeons & Dragons. You might have made a good call about the attributes, but the class thing is a bit off. Sure, people have professions in their lifetimes, and some people are good at some things others aren’t. But when it comes to psychology, our strategies for learning things aren’t so personalised. It’s nice to think that our kids might all have individualised styles of learning, and that every child has a chance to be clever in their own way, but the reality is that learning seems to be unitary, and it seems to be based on one spectrum: that of IQ. Inventing pleasant-sounding but evidence-free systems as ‘learning style’ proponents do is definitely not helping. What are needed are improved teaching methods across the board, teaching useful, applicable knowledge in ways appropriate to each individual subject, not to each individual pupil. Pashler et al. again:

Given the capacity of humans to learn, it seems especially important to keep all avenues, options, and aspirations open for our students, our children, and ourselves. Toward that end, we think the primary focus should be on identifying and introducing the experiences, activities, and challenges that enhance everybody’s learning.

Now, where did I leave my d20…?


Pashler, H., McDaniel, M., Rohrer, D., & Bjork, B. (2009) Learning Styles: Concepts and Evidence. Psychological Science in the Public Interest 9 (3), 106-116.

Roberts, B. W., Kuncel, N. R., Shiner, R., Caspi, A., & Goldberg, L. R. (2007) The power of personality: The comparative validity of personality traits, socioeconomic status, and cognitive ability for predicting important life outcomes. Perspectives on Psychological Science 2 (4), 313-345.

  1. Mickey
    April 1, 2010 at 7:03 pm

    “Deeply geeky” indeed. Fascinating though.

    Also, should I regret never going to the Edinburgh Dungeon? I love cheesy tourist traps, but I heard nothing but bad things about it. Is it just a bunch of lame sound effects and Burke & Hare puns?

    • Stuart Ritchie
      April 1, 2010 at 9:19 pm

      Thank you very much! And… I don’t think I’ve ever been there myself. I did go to the York Dungeon when I was a kid, I’m pretty sure. I just don’t find waxworks scary, no matter what silly contorted facial expressions they may have…

  2. Ned
    April 2, 2010 at 12:14 pm

    Excellent read. I may gently poke people towards this if they mention any of this ‘kinaesthetic learner’ gubbins.

    • Stuart Ritchie
      April 2, 2010 at 1:08 pm

      Cheers! It is remarkably widespread. And don’t even get me started on ‘multiple intelligences’…

  1. April 25, 2010 at 12:03 am
  2. May 14, 2010 at 4:14 pm
  3. May 23, 2010 at 3:23 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: