We’re not good at large numbers. Our brains are adapted to living in groups of perhaps 150 individuals. City living is a very recent innovation, and our psychological mechanisms struggle to cope with it. One way in which we may go astray is through the misapplication of heuristics that worked well for our ancestors, but which misfire in very large groups. Suppose you learn that there is a person in your group prone to violence without provocation. If you live in a group of 150 individuals, you need to be on high alert: your paths will cross. But if you live in a city of 5 million people, you really shouldn’t worry (unless you have some reason to think the violent individual lives in your street). In fact, on learning that there is such an individual you will probably feel more anxiety than you should – not as much as you would if you knew your paths would cross, but more than you rationally ought to.In general, we are bad at understanding low-probability risks. One result is that we attend far more to risks that are novel or which stem from deliberate actions by others than those that come from disease or which we are familiar with. We worry far more about Ebola than the flu, though the flu is likely to kill far more people and carries a higher risk of a pandemic (in Africa, malaria kills many more people than does Ebola). We worry much more about terrorism than about food poisoning, though the risk of dying from food poisoning is much greater than the risk of dying from terrorism.
These facts make more wonder about the effects of viral videos on our social cognition. Here’s one, more or less at random: a young man on a suburban Brisbane train racially abusing a security guard. There have been a number of such incidents recently, in Australia and elsewhere: a man or woman filmed hurling racist abuse at others, captured on someone’s mobile phone and uploaded to YouTube or Facebook. There is a perception that racism is rising in Australia, but it might just be more visible: the ubiquity of mobile phones, and the fact that these videos are horribly fascinating, ensures that though these incidents are not more common than in the past, we notice them more (on the other hand, it is possible that their visibility promotes copy cats).
So far as I know, there is no good research on what effect these viral videos have on our social cognition: on our sense of security, trust, and faith in government. It is plausible to think that because we are not good at understanding probabilities, these videos have a bigger effect on our perceptions than they should. Its not merely that we tend to vastly overestimate the representativeness of one incident; its also the fact that since sense perception words only over short distances, seeing a video may greatly increase the effect: our brains may be primed to treat everything we see and hear as something that is representative of our very local environment. Recent politics in many countries has been characterised by a growing distrust in government and opposition, and a search for alternatives. I wonder whether a sense that our local environments are more dangerous than they were (a sense in stark opposition to actual crime figures, which are trending down) may not be playing a role in our rejection of the status quo.