Recognising the patterns

Pattern recognition is an innate ability humans possess, something that first evolved to help our ancient ancestors recognise threats: Was the grass moving because of the wind, or because a predator was stalking them? Is that fruit good to eat, or will it kill us? Over time, it also helped our ancestors to recognise kin versus strangers, which is useful for a whole lot of reasons. It’s the modern expression that can cause problems, and the video above helps to show why that is.

The video, the latest in the “TED Ed” series of lessons from the TED Talks folks, talks about the mathematics behind patterns. They exist everywhere, and, as the video tells us, the probability that a pattern exists can be predicted.

However, just because humans see a pattern, that doesn’t mean the pattern reveals any actual meaning. Instead, it can be random coincidence.

I come up against this frequently. Whenever we talk about humans’ social behaviour—whether politics, pop culture fandom, religion, anything at all—we have to be very careful to limit our attempt to understand that behaviour to what we can verify empirically. A single study or opinion poll, however interesting it may be, is only a starting point until more data is collected to support the conjecture.

So, an election opinion poll by itself tells us very little. This is why political scientists urge observation of trends, rather than specific poll results. A well-constructed “poll of polls” (such as the work done by Nate Silver’s Five Thirty Eight) can be useful. Combining the data from reputable polls makes it easier to see real patterns, rather than imagined (or hoped for) ones.

The vast majority of us don’t believe in or subscribe to conspiracy theories, the starting point for the video, but the video’s starting device helps explains how otherwise reasonable people can get sucked into them and into anti-intellectualism like climate change denial and anti-vaxxers, to cite just two highly prominent examples. The people who believe conspiracy theories probably aren’t crazy or simple: They’re merely following their genetic tendency to see patterns where there is actually randomness, and to ascribe particular meaning to those patterns despite any the lack of any corroborating evidence.

Pattern recognition allowed our ancient ancestors to survive and procreate so that we could evolve. But pattern recognition sometimes threatens to undo all we’ve achieved over the millennia as we fall for dangerous ideas.

This is why we must always challenge our assumptions, including the belief in a pattern or its meaning. In an election year, that can be a very difficult thing to do, no matter how intelligent or rational we may be. But it’s the only way we can know if the grass is moving because of the wind, or because a predator is stalking us.