What if not thinking about decisions made you better at decision-making?
respect your subconscious
Okay hear me out on this one. You’re deciding on your next apartment and you have four options in front of you. You’ve just learned a bunch about each of them: price, square footage, number of rats that live in the walls. Now, imagine that God has given you three different ways you can go about making this decision: (1) you can make a decision right now, immediately after having learned about the apartments; (2) you can take some time to intentionally think about it before deciding; or (3) you can take some time before deciding, but you’re not allowed to actively think about it—you’ll be distracted with some other task, like solving a math problem or performing an improv skit. Which approach would you take?
Everyone agrees that there is a limit to how much you should think about decisions—hence the irritating term ‘analysis paralysis’—but few people have claimed that no thinking is the best choice of all. That is, until two researchers at the University of Amsterdam did some experiments in the early 2000’s and came to the conclusion that maybe we shouldn’t think about complex decisions at all.
Let me explain. The researchers—Ap Dijksterhuis and Loran F. Nordgren—put a group of survey participants into the scenario I just described. They gave subjects a long list of information about four apartments—much more information than you can hold in your head at once—and then subjects had to decide which apartment they preferred, either (1) immediately, or (2) after a few minutes of careful deliberation, or (3) after a few minutes of completing an unrelated, cognitively demanding task. The apartment descriptions were intentionally crafted so that one of the apartments was “best”—more positive traits than all the others. Surprisingly, those who did a completely random task for a few minutes and then decided performed better than the others—they made the right choice 59% of the time, while those who spent that same amount of time actively thinking about it only got it right 47% of the time, and those who didn’t have any time to think or do other tasks got 36%.
This is a strange outcome. It makes sense that having a few minutes to think results in better decision-making than having to decide immediately (47% vs 36%). But why would taking a few minutes to do something completely different—without getting any time to actually think about the problem—result in even better choices? The researchers cited this experiment as evidence of something they call “unconscious thought.” In their view, while the participants were directing their attention elsewhere (a difficult memory test), their brain was subconsciously organizing all the information they had previously learned and was determining the best decision, to be delivered to the conscious mind in the form of an inexplicable intuition.
Neat. Let’s talk about caveats though. I’m not a fan of coming to grand sweeping conclusions about human behavior based on the survey results of a bunch of college students. This is a criticism of experimental psychology in general: human behavior is extremely complex and sensitive to all kinds of factors that are impossible to control, so experimental results are notoriously difficult to replicate. Also, there are a number of questions we can ask about this particular experimental setup. Are we sure that the apartment we crafted to have “the most positive features” was objectively the “correct” choice? Different people value different things in an apartment at different times (maybe the extra spicy chicken wings you had last night are making you especially concerned about the number of bathrooms in your future apartment). Also, the timeframes for the experiment were quite short (3 minutes of either deliberation or distraction), which doesn’t reflect the amount of time people usually take to think through a decision.
In fairness, Dijksterhuis and friends conducted several experiments to address these concerns. They tried shorter and longer timeframes, and found that providing more time actually increases the benefits of the unconscious thought. They also conducted an experiment involving a real-life decision to purchase a poster, and showed that unconscious thought resulted in greater satisfaction a month after the original purchase. In yet another experiment they showed that the benefits of unconscious thought only occur when the subject is told ahead of time that they would have to make a decision. In other words, there is some intentionality to the unconscious thought process: it only does background work relevant to your conscious objectives.
We can go back and forth about the details of their various experiments and the critiques of them. I’m more interested, though, in the ethos behind Dijksterhuis et al’s work. It’s a good corrective for our tendency to imagine that the best way to make decisions in complex situations is to consciously and intensively deliberate on them. We’ve all had the experience of working hard on a problem for hours with no result, and then suddenly having the solution come to us in a flash of inspiration when we’re not even thinking about it. One view is that these moments of inspiration are merely a result of giving our mind a chance to reset—perhaps to get out of “local minima” in the solution space and consider alternative approaches. But maybe there’s a more active, creative, and even intelligent process going on outside our awareness, which makes itself apparent to us once its results have been computed.
Here’s an interesting point the authors make: human working memory—our capacity to consciously hold disparate pieces of information in our head—is estimated at about seven items, and the “processing speed” of our conscious mind is pegged at somewhere between 10-60 bits per second. But the rest of our brain—all the processing that occurs subconsciously or unconsciously—is estimated at more than ten million bits per second. Admittedly, estimates of the brain’s total capacity are extremely difficult to make, but there is little doubt that the amount of information we can hold under conscious attention is orders of magnitude smaller than all the processing our brain does in the background.
The researchers’ ultimate point is not that subconscious mental processing is objectively superior to the conscious alternative in every respect, but that they have different strengths. Conscious thought is much more precise and amenable to logical rules (hence why you can’t subconsciously calculate the result of 27x14). But unconscious thought is better at processing large amounts of information, is more “divergent”, and more associative. Conscious thought is useful—necessary, in fact—for the initial step of gathering all the information we need to make a decision. Once it has been gathered, though, perhaps it’s best to leave the integration and synthesis of all that information to the unconscious. This feels like a reversal of the way people usually think about it:
A major reason that people distrust intuition is the belief, which is often implicitly held, that intuitions are snap judgments that arrive in consciousness with little or no prior information processing. However, such a belief may not be justified. In many cases, intuitions may well be the result of extensive unconscious thought. Intuitions are the summary judgments the unconscious provides when it is ready to decide.
What I take away from the unconscious thought research is a greater admiration for all the processing our brains are doing outside our awareness. When I originally learned about the unconscious, in high school, it was in the context of Freud. The message was something like “you're driven by these vile egotistical impulses (usually relating to your mom) and you have no control over them”, which I didn’t really like at the time and still don’t. But now when I think about the subconscious it's more: look at this vast beautiful vista, with so much in it—all your memories and experiences, all your knowledge, the wisdom of all of your ancestors in one form or another, and the depth of all the love you have for everyone. It’s all there, and maybe it’s chugging along in the background, working on the questions that you struggle with the most, preparing to spring forth the answers when you’re ready to receive them.
This idea of unconscious thought being less transparent/understandable, yet stronger at addressing fluidly defined problems is highly reminiscent of the ongoing shift/debate of relying on human written code (explainable, possibly error prone, clearly handles precisely bounded problem spaces) versus generative AI/LLMs/neural nets (black box, more precise when given more time to optimize weights, can identify difficult-to-articulate patterns).
Even more reason to trust yourself!