Intuition as Recognition

Posted: 27-04-2016 16:02:53 UTC

Here’s a story. A fire commander tries to put out a small fire in a normal one-floor house. He hoses it with water, which has oddly little effect. He tries again; same result. “Strange”, he thinks. Taking a few steps back he decides to confer with his team. In that moment, something feels amiss. He can’t give a reason for his disquiet. But he orders his team out of the house anyway. Shortly after, the floor where they stood collapses: the fire had started in the basement, unknown to them.

Intuition can sometimes seem magic. We marvel at stories like the fire commander’s. Cases where fast, accurate, and compelling intuitions save lives. Impressive, too, are the chess grandmasters who can intuit how a match will unfold. Or the neonatal nurse who can sense a baby has a dangerous infection well before tests return. Intuitive judgements, like these, seem superhuman. Precognitive even.

Perhaps because they can seem so inspired, such judgements are widely admired. A strong presumption that we should heed our gut’s voice exists — particularly in business and sporting circles.

Is this presumption defensible?

A moment’s reflection tells us: likely not. Uncritical support of intuition is dumb. Intuition can save lives. But it can also break them. People do all sorts of stupid things on a hunch. They abandon good jobs suspecting they are the next Elon Musk; repent at leisure after trusting their gut in marriage; squander their retirement fund feeling lucky at casinos. And so on and so sad. After such mistakes, we often wish we’d muzzled intuition. We regret not deliberating well beforehand. Which raises the question — when is intuition trustworthy?

There is, it turns out, a robust answer here. Intuitive decision making has been studied for over 40 years, and we now have a clear idea about how it works and when it’s reliable.

Most researchers agree that intuition is a type of judgement. Gerd Gigerenzer describes it as “a judgement that appears quickly in consciousness, whose underlying reasons we are not fully aware of, and is strong enough to act upon.” Such judgements are usually triggered by a cue. If the cue is reliably linked to an outcome, a pattern of events will spring to mind. Knowledge of this pattern is intuition’s basis.

Consider chess for example. From a given configuration of pieces (cue) you can predict fairly reliably how a player will respond (outcome). Play enough to internalise this pattern of moves and intuition will suggest a good counter.

Intuitive judgement, then, is simply a form of memory — hardly occult. Herbert Simon put the point crisply: “The situation has provided a cue; this cue has given the expert access to information stored in memory; and the information provides the answer. Intuition is nothing more and nothing less than recognition.”

Or it is in the case of skilled intuition, at least. Unfortunately, however, intuition doesn’t always arise this way.

As Gigerenzer has shown, sometimes intuitions reflect not expertise but heuristics. These are cognitive rules of thumb, shortcuts we use when deciding under uncertainty. An inexperienced stock picker, for example, might buy stock according to the recognition heuristic: ‘go with the one you recognise’.

Heuristics like this have value. However, we often misapply them and hence decide poorly. Intuition’s opaque nature deepens the problem. Introspection won’t reveal whether an intuitive thought is based on skill or an imperfect heuristic. There’s no mental alarm bell here, no subjective marker distinguishing the two.

Hence our problem: when to trust?

The answer involves largely discounting mental life. Instead, we need to consider our environment. Two conditions must be met for reliable intuition.

One is sufficient regularity. Your context must have knowable regularities, common patterns of action. Psychologists say such contexts have ‘high-validity’. Consider chess again. Here, cues (patterns of pieces) reliably lead to specific outcomes (the player’s response). Cues are highly valid predictors, in other words. Fire fighters and chess players occupy very different contexts. High-validity marks both, however.

The second condition is prolonged practice. You must have gained knowledge of regularities through deliberate practice. A grandmaster can read a chess game at a glance, but it takes years to develop this skill. 10,000 hours of intense practice is a good estimate for many activities. Sound intuition exacts a steep price.

In sum, just so long as your environment is regular and you have a chance to learn, skilled intuitions will probably emerge. This may seem like a bland and academic finding. In fact, however, it is quietly subversive.

I say this because many professional environments are not regular at all. Stock pickers, clinical psychologists, political pundits — all work in contexts where there is much noise but little signal. The conditions for skilled intuition imply that these professions’ members will often suffer an illusion of validity. They believe they are both experienced and skilled. Really, though, they are simply the former¹.

This isn’t blameworthy; obviously, it’s wrong to fault someone for not being skilled in an environment precluding skill. Puncturing their delusion, however, seems fair. Without valid cues, accurate intuitions are never more than lucky hits. Praising these professionals’ intuitive judgement is therefore incoherent.

Slightly shocked? Well then. You probably have a lingering belief that intuition is magical. Don’t be mugged by mysticism. Remember: intuition is always unreliable in the absence of stable regularities.