Back in the early nineties when I was at Cambridge University I went to a talk given by Douglas Adams (of The Hitchhiker’s Guide to the Galaxy fame). The talk was primarily about technology but in his usual style he went off at a tangent on several occasions. One thing that he talked about has stuck in my mind to this day. He got onto the subject of virtual reality and simulation and strayed into the realms of psychology when he began to describe the parallels between computer simulations and mental models and how we use them to forecast future events.

What he had done was to give a name to a phenomenon I was intimately acquainted with. I remember it so well because I recognised an aspect of myself – I realised that I build mental models of people I interact with, refining the model as I become more familiar with a person. You can think of a model as a series of chains of cause and effect – if I do this then they will do that. I use that model to run through what I plan to say to them and how I plan to say it, and try to predict how they will react. This is all on a conscious level for me – there’s little or no instinctive component. That’s not to say it’s necessarily a slow process but it does introduce some delays into my social interactions while I think about the next move.

I’ve recently been thinking more about this, prompted by discussion of Theory of Mind – among other things – over on Journeys with Autism. There is a school of thought in psychology that considers people on the autistic spectrum lack a theory of mind (ToM); that is to say that they lack an awareness of others’ thoughts and feelings. I believe that my mental models are my ToM as I understand the concept. I also believe that my deficiencies in this area are primarily caused by my poor social skills: I don’t have the tools to actively experiment with people and thus refine my model so I just rely on passive observation. This naturally takes much longer to achieve comparable results when compared to somebody who is comfortable and adept in social interactions.

I’ve watched my wife when she starts talking to somebody that she has never met before. First she will observe them for a short time, somehow picking up non-verbal cues and forming an initial impression of the person in her mind. She will then approach them and start a conversation based on what she has observed. It appears to me that this initial conversation is a social tool that enables her to “get to know” the person quickly by gauging their reactions to her probing – something like a handshaking protocol in network communications. A little while later she’ll introduce me to them. I just respond neutrally to what they say – I start out very warily in social interactions because I have no mental model for this new person yet and no way of knowing how they will react to me – and probably get thought of as quiet and boring. I’m primarily observing them at this point, noticing how they respond to my wife and constructing a brand new mental model of them. This imposes quite a cognitive load on me and I can’t really cope with inventing something to say at the same time.

I have no idea how most people appear to handle this kind of thing so easily – it just seems to come naturally, instinctively, to them rather than being a learned skill as it is for me. It means that it takes me a long time to get to know somebody reasonably well – well enough to feel comfortable around them. I don’t form attachments easily or quickly. However I have found that my hand-crafted mental models can end up being more accurate than my wife’s intuition at predicting some people’s behaviour. Does this mean that I have a more complete theory of mind than my neurotypical wife? Is it like the difference between bespoke and off-the-peg? Or is it just more evidence that all theories and models have limitations and will depart from reality at some point because they seek to simplify and generalise in order to provide some insight into complex natural phenomena?