Those last weeks I am very much involved with designing experimental environments. I want to develop a business plan for investing in smart cities, and a good business plan could do with the understanding of how can human behaviour change in various experimental conditions. Thus, how are people likely to behave when living in a smart city? How is their behaviour going to be different from those living in a classical, non-smart (dumb?) city? Behaviour is what we do in response to stimuli from our environment. OK, so now I begin by defining what I do. When I take on defining something apparently so bloody complex that my head is turning at the very thought of defining it, the first thing I do is to structure what I do, i.e. I distinguish pieces in the whole.
What kind of categories can I distinguish in what I do? One of the first distinctions I can come up with is by the degree of recurrence. There are things I do so regularly that I even don’t always notice I do them. When repeating those actions, I practically fly on automatic pilot. Walking is one of them. I breathe quite systematically as well, as I think about it. I drive my car almost every day, along mostly repetitive itineraries. As I sail onto the waters of the incidental, and the shore of mindless recurrence progressively vanishes behind me, I cross that ring of islands, like the rings of the Saturn, where big things happen just sometimes, yet they happen in many, periodically spaced sometimes. These are summer holidays, Christmases, or wedding anniversaries. As I sail past the reef of those reassuringly festive occasions, I enter the hardly charted waters of whatever happens next: these are things that I subjectively perceive as absolutely uncertain, and still which happen with a logic I cannot grasp on the happening.
So, here we are with one distinction inside our behaviour: the steady stream of routine, decorated with the regular patches of periodical, ritualized, big events, and all that occasionally visited by the hurricanes of the uncertain. When people live in an urban environment (in any environment, as a matter of fact), their living is composed of three behavioural types: routines, cyclical actions, and reactions to what they perceive as unpredictable. If living in smart cities is about to change our urban lives, it should have some kind of impact upon our behaviour, thus on routines, cyclical events and emergencies. How can it happen and how can we experiment with it?
Another intuitive distinction about our behaviour is that between freedom and constraint. I perceive some of my actions as taken and done out of my sheer free will, whilst I see some others as done under significant constraint. I know that the very concept of free will is arguable, yet I decided to rely on my intuition, and this not a good moment to back off. Thus, I rely and I distinguish. There are actions, which I perceive as undertaken and carried out freely. Trying to be logical, now, I interpret that feeling of freedom as being my own experience of choice. In some situations, I am experiencing quite a broad repertoire of alternative paths to take in my action. A lot of alternatives means that I don’t have enough information to figure out one best way of doing things, and I am entertaining myself with my own feeling of uncertainty. Freedom is connected to uncertainty, but not just to uncertainty. If I can do things in many alternative ways, it means nobody tells me to do those things in one, precise way. There are no ready-made recipes for the situation, or relevant social norms, in my local culture. On the other hand, my highly constrained behaviour corresponds to situations tightly regulated by social norms.
When I have two different distinctions, I can make a third one, two-dimensional, this time. In the most obvious and the least elaborate form it is a table, as shown below:
|Free behaviour (no detailed social norms)||Constrained behaviour (normatively regulated)|
|Routine behaviour||Modality #1||Modality #2|
|Cyclical behaviour||Modality #3||Modality #4|
|Emergency behaviour||Modality #5||Modality #6|
A normal, fully sane person would leave that table as it is, but I am a scientist, and I have inside me that curious ape, that happy bulldog, and the austere monk. I just need some maths to have something for rummaging in. I just have convert my table into a manifold, with those two nice axes. Maybe I could even trace an indifference curve in it, who knows? Anyway, I need converting modalities into numbers. The kind of numbers I see here are probabilities. The head of the table, namely the distinction between freedom and constraint can be translated into the probability that any given piece of my behaviour (yours too, by the way) is regulated by an unequivocal social norm. It is more fun than you think, as a matter of fact, as we have lots of situations when there are many social norms involved and they are kind of conflicting. I am driving, in order to pick my kid from school, and suddenly I drive over a dog. I should stop and give emergency care to the dog, but then I will not pick up my child from school at time. Of course, at the end of the day, we can convert all such dilemmas into the Hamletic “to be or not to be”, which really narrows down the scope of available options. Still, real life is complicated.
Anyway, I am passing now to scaling numerically the side of my table, as a probability, and I am bumping against a problem: if I translate the recurrence of anything as a probability, it would be the probability of happening in a definite period of time. Thus, it would be a binomial distribution of probability. I take my period of time, like one month, for example, and I just stuff each occurrence in my behaviour into one of the two bags: “yes, it happens at least once in one month” or “no, it doesn’t”. The binomial distribution is fascinating for studying the issue of structural stability (see Fringe phenomena, which happen just sometimes), but in a numerical manifold it gives just two discrete classes, which is not much of a numerical approach, really. I have to figure out something else and that something else is simply recurrence, understood as the cycle of happening, like every day, every three days, every millennium etc.
And so I come up with that nice behavioural graph in PDF, available from the library of my blog . See? Didn’t I tell you I would make an indifference curve? This is the red one in the graph. It is convex, with its tails nicely, assymptotically gliding the long of the axes of reference, so it is bound to be an indifference curve, or an isoquant. The only problem is that I haven’t figured out, yet, what kind of constant quantity it measures. It will come to me, no worries. Still, for the moment, what comes is the idea that on the two tails of this curve I have somehow opposite patterns of behaviour, mostly as for their modifiability. On the bottom right tail, where those ritualized routines dwell, I can modify human behaviour simply by modifying one simple rule, or just a few of them. From now on, I tell those people (or myself) to do things in way B, instead of way A, and Bob’s my uncle: with any luck, and with a little help from Mr Selten and Mr Hammerstein (1994) those people (or me) will soon forget that the rule has ever been changed. On the opposite, upper left tail of that curve, I have things happening really just sometimes, and virtually no rules to regulate human behaviour. How the hell can I modify behavioural patterns in these whereabouts? Honestly, nothing sensible comes to my mind.
Smart cities mean lots of digital technologies. I have just watched a short video, featuring a robot (well, a pair of automated arms fixed to the structure of a bar), which can prepare hundreds of meals, like a professional cook, imitating the movements of a human. Looks a bit scary, I can tell you, but this is what a smart city can look like: some repetitive jobs done by robots. Besides robots, what can we have in a smart city, in terms of smart technologies? GPS tracking, real-time functional optimization (sounds complicated, but this is what you have, for example, in those escalators, which suddenly speed up when you step onto them), personal identification, quick interpersonal communication, and the Internet of things (an escalator can send emails to a cooling pump, which, in turn, can get friends, via social media, among the local smart energy grids). These technologies can take the functional form of: robots (something moving), mobile apps in a phone, in a pair of glasses etc. (something that makes people and things move), and infrastructure (something that definitely shouldn’t move). In their smart form, these things can optimize energy, and learn. We use to call the latter capacity Artificial Intelligence. I think that it is precisely the learning part that can affect our lives the most, in a smart city. We, humans, are kind of historically used to be learning faster than our environment. We are proudly accustomed to figure out things about things before those things change. In a smart city, we have things figuring out things about us, and at an accelerating pace.
In one of my previous updates (see Smart cities, or rummaging in the waste heap of culture ) I made those four hypotheses about smart cities. Good, now I can reappraise those four hypotheses in terms of human behaviour. We behave the way we behave because we have learnt to do so. In a smart city, we will be behaving in the presence of technologies, which can possible learn faster and better than us. Now, keeping in mind that table and that graph, above, how can the coexistence with something possibly smarter than us modify our patterns of behaviour? Following the logic, which I have just unfolded, modification of behaviour can start in the bottom right area of my graph, or with Modality #2 in the tabular form, and then it could possibly move kind of along the red curve in the graph. Thus, what I previously wrote about new patterns observable in handling money, in consuming energy, in rearranging the geography of our habitat, and finally in shaping our social hierarchies, means that smart cities, and their inherently intelligent technologies can impact our behaviour first and most of all by creating and enforcing new rules for highly recurrent, ritualized actions in our life.
I am consistently delivering good, almost new science to my readers, and love doing it, and I am working on crowdfunding this activity of mine. You can consider going to my Patreon page and become my patron. If you decide so, I will be grateful for suggesting me two things that Patreon suggests me to suggest you. Firstly, what kind of reward would you expect in exchange of supporting me? Secondly, what kind of phases would you like to see in the development of my research, and of the corresponding educational tools?
 Hammerstein, P., & Selten, R. (1994). Game theory and evolutionary biology. Handbook of game theory with economic applications, 2, 929-993.