Can you think productively while in action? Or, once you’ve practiced what to do, is it better to let everything run offline? This week, I’ll try to say something by way of response to these questions, drawing on my book Thought in Action: Expertise and the Conscious Mind (OUP, 2016). Thanks John for the invite! And since I imagine that some of you, like me, are a bit groggy after the Thanksgiving weekend of feasts (and family feuds), I propose we start with a jolt:
How fast can you think?
A call comes through to the triage desk of a large hospital in the New York City metropolitan area: a pregnant woman with multiple abdominal gunshot wounds is due to arrive in three minutes. Activating a trauma alert, the head nurse on duty, Denise, requests intubation, scans, anesthesia, surgery, and, due to the special circumstances, sonography and labor and delivery. The emergency medical service team slides the gurney into the hospital, and the trauma center staff starts in with Denise coordinating and overseeing the entire process. How is it possible to think about so much so quickly?
When Denise (not her real name) recounted this story to me, she was enrolled in my philosophy of science course for nursing doctoral students. In this course, amid discussions of Karl Popper’s criterion for demarcating science from nonscience and Thomas Kuhn’s theory of scientific revolutions, I would hear stories from her and other students about nursing: about newborns in the critical care unit; about patients who would, against all orders, remove their EKG monitors to use the toilet; about the travails of discussing the benefits of blood transfusions with desperately ill Jehovah’s Witnesses. These stories often reminded me of how little urgency exists in philosophy—one doesn’t feel much time pressure to solve the problem of free will when it has been open for the past two thousand years—as well as how little tragedy. The nurses in my classes learned fairly quickly that they should bring tissues for me. The day I heard Denise’s story was not an exception: neither mother nor baby were able to be saved. Their stories also sometimes illustrated how much thought goes into even the quickest decisions.
Let us now step away from the bustle of the emergency department and consider the conscious mind. Did Denise need to think consciously about what she had to do? Did she consciously decide to request the labor and delivery team? Or, after the call came in, were her actions unconscious and automatic? What is the role of consciousness in emergency decision making?
Unconscious, reflex-like decisions are fast. Neural signals that result in action can travel along myelinated nerves at upwards of 120 meters per second. A runner at the starting block, after hearing the pistol, pushes off in less time than it takes to blink an eye. Compare this to the leisurely 0.5 – 2 meters per second pace at which pain signals travel along unmyelinated nerves. But how fast can one follow through a conscious train of reasoning, one that leads, for example, from the weather forecast to a decision about what clothes to wear? When we have time, we weigh the options—if I bring the light jacket, I’ll be somewhat cold outside but won’t need to carry it when I’m inside—and arrive at what, upon consideration, seems to be the best decision. Yet what happens when, as in Denise’s case, there is a starting pistol for our decisions? Do we go on autopilot, grab the galoshes and run?
To probe the speed limits of conscious thought, I conducted an experiment. I had four accomplished chess players—two masters, one national master and one (retired) international master—think aloud, saying what came into their minds, if anything, as they were playing a game of lightning chess, which is a variation of chess that allows a mere one minute per player for an entire game. After I explained the task, my participants all expressed doubts; they think all the time, they said, but they assumed that thinking aloud while playing would likely slow them down too much. Indeed, the first player I was scheduled to test made such a convincing case for this, I almost skipped over the one-minute game trials and moved on to the five-minute ones. However, the five-minute trials were not necessary. As soon as they were paired with their online opponents, clocks started ticking and they started talking, very rapidly and more or less nonstop, about their reasoning processes as they were playing.
To be sure, much of what they said was elliptical. For example, at one point during a game I heard: “if I play B6, he plays F3; is that the idea? B6, F3; what about C5, D5—D5? D5? If he plays D5 I get into a Leningrad-y thing; but I don’t want a Leningrad-y thing . . .” Nonetheless, such comments indicate that my chess players were engaged in rapid, conscious deliberations.
How did this affect their games? The results were what would be expected given the pairings: when paired against slightly weaker players, they won; against slightly stronger ones, they lost. And in terms of their own personal evaluations of their games, they all felt that they performed at least as well as they usually do, with one commenting, “I think it actually helped my game.” Another said, “I never think I play well, but I played like I play.”
From an outside perspective, it might seem that a one-minute-per-player match proceeds so rapidly that it intercepts conscious deliberation. Hands fly over the board or move the cursor on what appears to be mere impulse. Could it be that when the chess players in my experiment vocalized what was going on in their minds, it was the vocalization that elevated their thoughts and plans to the level of consciousness? Or is it that the appearance of automatic and effortless expertise in lightning chess, in nursing, and elsewhere, can be deceptive?
Daniel Kahneman, in his book Thinking, Fast and Slow, distinguishes what he calls “System 1” thinking, which he describes as automatic and intuitive—what you use when you add two and two—from “System 2,” thinking, which he describes as slow and analytical—what you use when you mentally compute seventeen times twenty-four. The calculations of the chess players in my study were System 2, but accelerated. Though perhaps not as fast as the blink of an eye, it seems that our complex, conscious deliberations can, when necessary, be fast. Or as the seventeenth philosopher Thomas Hobbes put it, “thought is quick.”
A Marine once told me that endless drills have allowed him to reload his weapons without any thought at all. Reloading a weapon, hitting the chess clock, picking up the call from the emergency service team—yes, all that is fast and automatic. But what about shooting? I wanted to ask the Marine that question, but, perhaps because of the large M4 carbine his arms, I was too intimidated. If someone suddenly points a gun at you in the midst of battle, one hopes that the impulse to respond gets carried along those myelinated nerves that serve as a super highway. Yet, at the same time, in war, as on the emergency room floor, one wants to be like the chess player: capable of justifying what you have done. Acting automatically precludes this.
There is an old saying, “look before you leap.” But if there really isn’t time to look beforehand, look while you leap and be ready to adjust plans mid-air.
“Yet, at the same time, in war, as on the emergency room floor, one wants to be like the chess player: capable of justifying what you have done. Acting automatically precludes this.”
Does it? It seems like this claim defines “automatic” actions as excluding or opposed to conscious thought. But that doesn’t seem right. We can differentiate at a minimum skilled and unskilled forms of automatic action; and I suspect the concept could do with even finer grains of distinction. Reflection doubtless enters this story, but that doesn’t imply we must assimilate any and all justifiable actions to the scope of discursive thought.
You make a good point. Thanks, yes, “automatic” need not be defined in that way. I think Ellen Fridland has some papers that present some good reasons to make fine grained distinctions between various kinds of “automatic” behavior, which I should definitely reread! However, you are right that I am using the term “automatic” to exclude conscious thought, which I think is also a standard way of using this term.
” However, you are right that I am using the term “automatic” to exclude conscious thought, which I think is also a standard way of using this term.”
I mis-stated my original worry. My quarrel isn’t with the term or its use, but with its application to contexts of human action.
Doubtless we can make sense of automatic behaviors which are truly involuntary, like the classic case of pulling one’s hand back from a hot stove. Assigning blame or other forms of responsibility to a “startle” or the like does seem out of place. But we can easily imagine that other cases — like your example of the trained soldier — wouldn’t be so clear cut. A well-trained and experienced soldier may have “automatic” responses on a battlefield. But this variety of automatic response seems unlike the reflex cases, and in ways interesting enough to cut against the claim that we can’t make sense of responsibility or justification.
The difficulty isn’t the opposition between automatic and conscious processing, as I originally put it, but in the belief that responsibility depends on conscious motivation and/or discursive articulation. I’m inclined to think this is true in many, but not all, cases of action. Consider the “autopilot” mode that can happen while driving over familiar territory, or while on a long-distance drive on long stretches of highway. There’s little or no conscious decision-making at work. But we don’t hold the driver any less responsible for an accident caused; nor does it seem right to say that the driver isn’t justified in the decision to turn right on Elm Street because that’s the way home.
Thanks for the thought-provoking post, in any case!
OK, Curious, thanks for the clarification. You’re making me think! I would agree that one can still be responsible for one’s actions that are performed on autopilot, however, I think that when you are really just going along and not making conscious decisions, then justification in the sense of being able explain why, say for driving, you went a certain speed, or turned a certain direction can be compromised. Yes, you can say that I turned left because I needed to get home. And maybe sometimes the nurse the emergency room could justify her actions in a similar manner. But those are not the ones that get questioned. It’s the ones that don’t clearly fit the mold and so the pat response, I did that because it works, isn’t going to cut it. Moreover, acting on autopilot, as I understand it–and here I’m getting into definitions again–just means not having a reason for your actions. Yet, experts, on my view, often do have reasons. The current world chess championship is a great example!
Can’t wait to read this book!
Thanks Stefanie!
Hi Barbara, thank you so much for blogging this week.
I am just finishing up teaching a graduate seminar (and, supposedly, a book) on Anscombe’s /Intention/. Concerning Aristotle’s concept of practical deliberation, which she is concerned to develop and defend, Anscombe writes that “Generally speaking, it would be very rare for a person to go through all the steps of a piece of practical reasoning as set out in conformity with Aristotle’s methods” (p. 79), and indeed that “if Aristotle’s account were supposed to describe actual mental processes, it would in general be quite absurd. The interest of the account is that it describes an order which is there whenever actions are done with intentions …” (p. 80). (She also discusses the phenomenon of expert action, claiming that “In general … one does not deliberate about an acquired skill” (p. 54).) Her point, I take it, is that an agent’s reasoning about what to do is not always a (conscious or unconscious) process that precedes an action and gives rise to it, but rather a formal or structural feature of intelligent action itself — as I would put it, an agent’s practical reasoning is embodied in her standing, often implicit, on-the-fly grasp of what she is up to and why. This sounds a bit like your suggestion that thought in action can be a matter of being ready to adjust while in mid-air. Is that a plausible parallel to draw?
I like the way, John, you embed your criticism in a suggestion that perhaps a less reflective type of action is what I’m getting at in my final comment and perhaps even what I’ve meant all along. But that’s not quite right, even though it is very kind of you. Nonetheless, I think my view is compatible with Anscombe’s. But for a different reason. What I take Anscombe to mean by “acquired skills” are things like tying shoes, typing (for someone like me who’s not a typist), making coffee (for the non-barista), and so forth. She might be right that for such skills one does not deliberate over them. And the conclusions of a practical syllogism are like this as well. You’re hungry; you don’t need to reason through the syllogism before eating. However, I’m concerned with the type of skills that not only you’ve acquired through practice–like being able to eat with a knife and fork–but one’s you’ve acquired through around ten years of deliberate practice, such as those of the experienced nurse, professional athlete, performing artist and so forth, and are still aiming to improve (they are not just sitting on their laurels). These people, too, don’t reason through every part of their expert skill. Of course, they can’t. But I think that they generally engage in a level of reflective thought that coincides with most of their best performances and that relevant thoughts do not interfere with their performances. In sum, I think that there is a relevant difference between, say, the the type of skills exemplified by a swimmer during the Olympics and everyday acquired skills, and when I talk about expert action I mean actions similar (in the relevant way) to those the Olympic swimmer. Does that make sense?
Thanks for this. I really didn’t mean for my suggestion to embody a criticism! But let me think more on what you say about the difference between your view and Anscombe’s.
Pressure situations where immediate decisions are required, by definition, do not allow much thought. So how can we do it? The nurse, chess players and Marine are not novices. They are all highly trained and practiced in their disciplines. For these specialists, it is all system one thinking. The nurse is simply recalling a specific protocol when coordinating personnel and equipment for a gunshot wound. The chess Grandmaster has spent a lifetime studying chess opening, defenses, mid and end game strategy. He has developed a “batting eye” for chess positions and can instantly determine strengths and weaknesses. The Grandmaster will surely play speed chess to his level of mastery. The Marine has his practiced drills and memorized rules of engagement. The decision to shoot or not will be in accordance to those rules. Shooting accurately is different because it requires motor skills, thought does not. It seems accuracy decreases when someone shoots back. Except in the case of the Marine of course. I think considering the question of “Can you think productively while in action?” should begin when a person’s automatic and intuitive system 1 thinking fails or does not apply.
An interesting read. I would say that perhaps one has the ability to think quickly due to the commonly used patterns of their neurons firing. For example, the marine is able to reload his gun without thinking about it because he has learned this behavior and perfected it. The neural passage way allowing for this behavior to be carried out is fully functional and mapped out without a problem which would thereby increase the speed at which one could perform that action relative to an action they would have otherwise done more slowly.