Experts, mull it over: Only the novice unconscious mind matches conscious decisions

Beware of anyone ever telling you some straightforward rule about how the mind works. As any college student fast discovers, the simple rules to live by presented in Intro Psych – or any time a new aspect of the mind is discovered, really – remains a simple rule for a few years at most, before psychologists conduct the studies that turn a simple feature of the mind into a complex series of caveats and exceptions. Which is cool, if you’re into the science, but must be a special form of hell if you’re trying to figure out good advice about what to do when.

Today’s illustration of this phenomenon is the “deliberation without attention effect“, which I wrote about just a few months ago based on cool studies showing that we might make better choices about complex decisions if we distract ourselves instead of actively trying to work out the solution. It’s a very provocative idea, so naturally other researchers have put the study under a microscope trying to work out anything else that might be going on. A group of researchers led by Silvia Mamede focused on one important idea: whether it’s better to mull over a problem or let your unconscious mind tackle it might just depend on how familiar you are with these kinds of problems. What’s true for the novice might not be true for an expert.

One of the major contenders for explaining why unconscious thought might be better is the limited capacity of our working memory. Most of us just can’t keep all 12 features for each of 2 cars in mind well enough to make all the comparisons are work out which car is “better”. But experts are famously better able to remember details about their area of expertise because they seem to process them in a different way, allowing chess experts to easily recall snapshots of chess games in progress that would immediately vanish from the minds of novices (or even hobbyists like myself). And wouldn’t you be happier if you thought your physician would come to a better decision by mulling a problem over for a while than by turning to the latest New York Times crossword puzzle?

If you think doctors should be deliberate in their diagnoses, you can now start feeling relieved. Mamede and her colleagues presented a few dozen residents (doctors, but still in training at a teaching hospital) and another few dozen medical students, with medical cases that were either common presentations of relatively frequent conditions (“simple”, to the medical profession at least), or an atypical presentation or uncommon disease (“complex”, the kind that’s likely to get written up in a medical mystery feature). Sometimes the diagnosis had to be given immediately after the case was presented, sometimes after 8 minutes of solving anagrams, sometimes after 8 minutes of active problem-solving, and sometimes after 8 minutes weighing the different possibilities.

The first good news for doctors: All that training and experience pays off with the simple problems. Residents were at 80% accuracy even if given no time to even think about the case, and neither conscious problem-solving nor “deliberation without attention” improved on that. The more important news, though, is that these experienced residents were much better off trying to actively puzzle out the more complex problems. Their first, “gut instinct” diagnosis for the uncommon diagnoses was only correct 40% of the time, and remained that poor if they distracted themselves solving anagrams, but leapt closer to 60% when they put a little conscious thought into it. For experts, then, the conscious mind trumps the unconscious one at decision-making.

This doesn’t disprove “deliberation without attention”, though, not by a long shot. For novices, it’s still looking like the best strategy. The medical students – even in their fourth year, still at the start of their medical careers – did better overall when they let their unconscious minds do the lifting. With the simple cases, either conscious or unconscious thinking was the route to the correct diagnosis 65% of the time, beating out “gut instinct’s” rather dismal not-quite-50% accurate diagnosis. In the complex cases, though, conscious deliberation of the options actually hurt the medical students, tanking their diagnosis from the 20-25% range with instinct or unconscious deliberation down to around 15% with deliberate problem solving. This is what we call “overthinking the problem”, with potentially disastrous results.

Since newly-minted doctors have no way of knowing whether they face a simple or a complex case, by the numbers they might be better off taking up that crossword, which was better than a snap judgement for the simple cases and no worse than a snap judgment for the complex ones, than trying to juggle all the symptoms and history in their minds. Of course, this ignores the probability that these initial attempts at deliberate diagnosis, however incorrect they might turn out to be, are a necessary step to the expertise shown by the residents. So I’ll let the doctors make decisions however they see fit. If you aren’t trying to become an expert, though, you can probably let your unconscious mind handle the problem.

Which is a relief, as I really didn’t want to have to start making charts and decision matrices next time I’m in the market for a car. But I’m sure another study will come to light soon that refines that piece of advice even more.
Mamede, S., Schmidt, H., Rikers, R., Custers, E., Splinter, T., & Saase, J. (2010). Conscious thought beats deliberation without attention in diagnostic decision-making: at least when you are an expert Psychological Research, 74 (6), 586-592 DOI: 10.1007/s00426-010-0281-8


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s