This seems like a great reason to be cranky at the movie… this sort of thing would definitely make me angry at a movie, and I would absolutely endorse that response. Why would you not?
Because almost every book or movie out there has something in it, somewhere, that would annoy me, and I like enjoying books and movies. Because the movie’s content is there anyway and me being cranky about it does absolutely nothing about the underlying culture that makes it normal for nerds to be made fun of in an action movie. Once I notice that I’m cranky and realize why, if I want to do something productive about it I can mention to my friends that I don’t think nerds should be characterized that way, or I can write a blog post about it...without being cranky. Emotions are a signal. Once I’ve decoded that signal, I don’t need the negative emotion anymore, unless it makes me more effective at doing something, which is rarely the case.
Re: having a “mysterious annoying emotion that [you] don’t endorse”: I’m not sure I understand this experience. Could you elaborate a bit?
We live in a world that’s practically designed to play on our emotions. TV ads basically designed to make me insecure about my appearance/coolness/etc so I’ll want to buy their product? I don’t need to walk around with lingering insecurities; I thought I looked fine and was cool enough two minutes ago, and an ad by people whose job is to manipulate me shouldn’t change that. Lots more people, in more complicated social relationships, than the ancestral environment? Sometimes people do things that annoy me, but I know that the thing they did was harmless and I don’t think it should annoy me, and whereas if they were a family member I’d ask them not to do it for my sake, they’re actually just an acquaintance, and I don’t want to go around feeling annoyed. Or sometimes I make a faux pas and I apologize and take all the necessary steps to fix it–I don’t want to keep feeling guilty after this, but I often do. Sometimes I have to make a choice between two different things I want to do, and I pick the one I want more, but then I’m sad about the one I didn’t pick–I don’t want to keep feeling sad and enjoy the thing that I did pick less.
Do we have any evidence that these “flash-intuitions”, these “gut” judgments, are actually accurate? How accurate are they?
They’re based off data that I could in theory notice and analyze logically–the way someone’s skin feels, the way someone looks when they’re breathing, etc. It just doesn’t feel like logical analysis anymore, it feels like “oh crap, X.” I think that a lot of intuition could be logical reasoning, it’s just learned to the point that it happens fast at a subconscious level by pattern-recognition. And they don’t have to be anywhere close to 100% accurate to be helpful. Usually the first thing I do with a gut feeling is get more data. “This patient seems worse...okay, I’ll check all their vital signs again.” Sometimes I’m right and something has changed. Sometimes I’m wrong. I’d still rather have the gut feelings than not. I don’t know how much data there is on this.
Hm. Thank you for taking the time to explain; I definitely appreciate it. Your experience and values seem to differ from mine in a number of ways; that does seem to be what’s behind the OP’s advice being of different utility to us.
As for the bit about accuracy of gut feelings: I take your point about them being a good signal to investigate further. I do remain quite dubious about the use of the gut feelings directly, in place of explicit reasoning. I would very much like to see some data about this.
The advantage of gut feelings and intuition lie with their ability to synthesize years of experience and thousands of variables into one answer within less than a second.
When is this necessary?
During a conversation, someone watching your face is going to be observing how you react (even in the smallest possible ways) as they speak. You don’t have an hour, five minutes, or even two seconds to decide how to present yourself; they’re going to judge you based on that instantaneous reaction (or a lack of one, including a delayed reaction or straight face.)
Anyone who is a natural “people person”—the kinds of people who can get almost anything they want from anyone around them, who make great salesmen or politicians—is going to need to be able to continuously react “properly”, and that means intuitive judgments.
Same thing with any kind of games/sports, or literally any other situation where a quick reaction is required and not immediately responding will doom you no matter what.
The domain of “behaving in such a way as to gain and maintain an advantageous position in social interactions” is very different from other domains, like “diagnose and repair difficulties with computer equipment”, “diagnose and treat afflictions in human beings”, “understand mathematical concepts”, and almost anything else. It seems to me that the domain of social interactions with other humans is in fact a unique domain, not properly comparable (in the context of the current discussion) to anything else.
Certainly gut feelings are key in social interactions; in fact, the most charismatic, likable, and socially successful people do what they do largely unconsciously, and are almost entirely unable not only to explain their technique to others, but even to recognize there is a technique, that other people are not as skilled at using.
My question is, how accurate are gut feelings in other domains — especially those domains where there is are objectively right answers and wrong answers, and where it is possible and even easy, in principle, to compare the answer you get from your gut feeling to the actual right answer? In the treatment of computers and people, in math, in science, in engineering? (Answering this question requires data!)
What’s more: I really don’t think that this
The advantage of gut feelings and intuition lie with their ability to synthesize years of experience and thousands of variables into one answer within less than a second.
is an accurate characterization of where gut feelings in successful social interactors come from. It’s been my experience that such instinctual social success is largely innate. Oh sure, it may be honed, but saying that the gut feeling is a synthesis of years of experience is just almost certainly not what’s going on there. More likely it’s a naturally great ability to model others, to respond (unconsciously) to nonverbal cues, etc.
Players are often required to make decisions with no time whatsoever to plan. For example, you might find yourself surrounded by enemies with no warning.
You need to know whether to run on foot, to teleport away, or to fight.
The difference between reacting in a third of a second and a fourth of a second could mean life or death.
Success in this situation, assuming it’s possible, is dependent on your experience in similar situations and your instinctual reaction. Since you do not have the time to think, your decision is almost guaranteed to be imperfect, but any improvement in it is highly beneficial.
Obviously, the same would often apply in war or in certain crisis situations.
You mention lots of fields (computers, math, science, engineering) where your argument is almost tautological: in a case where you have time to reconsider each decision, a slow but reliable and precise method is better than a snap judgment. Yes, I would agree with you, and I would also agree that logical thinking is better than intuitive thinking in many, many situations.
Are you suggesting that the ability to model others or respond to nonverbal cues is innate, rather than learned? I would definitely disagree, though proving it would be difficult. I suspect that it’s a matter of internalizing the results of numerous actions and reactions in different situations. In my experience, it’s often developed by people who travel lots or are otherwise exposed to tons of different people in a situation where being friendly and getting on their good side is very helpful. Some of them, pretty bad at socializing before they were in such a situation (and really gave it the necessary effort to learn).
I disagree, however, when you say that being socially successful is innate.
The difference between reacting in a third of a second and a fourth of a second could mean life or death.
Success in this situation, assuming it’s possible, is dependent on your experience in similar situations and your instinctual reaction. Since you do not have the time to think, your decision is almost guaranteed to be imperfect, but any improvement in it is highly beneficial.
I do not play sports, but I did spend several years doing high-end raiding (mostly as a main tank) in World of Warcraft, which I think fits your criteria. Raid play is fast-paced and demanding, with necessary reaction times measured in fractions of a second.
I would not characterize good play in a WoW raid as based on intuition. Here is, basically, the process for beating a new, challenging raid boss:
Go in, try the boss. The entire raid dies horribly, of course.
Meticulously, exhaustively analyze the combat log. Note down all observations made of boss behavior. Correlate data.
Brainstorm solutions, based on raid leader’s and key raid members’ comprehensive, minutely detailed knowledge of game mechanics.
Make detailed plan. The plan implicitly includes generally correct play from all raid members; note that for almost all classes in WoW, optimal play means following detailed algorithms for ability usage, often worked out at length by top “theorycrafters”, who are often people with advanced degrees in physics and mathematics (no, I am not exaggerating) — plus, of course, extensive experience, to the point where playing correctly is at the level of muscle memory.
Attempt to execute plan. Correct execution demands precise, down-to-the-second performance from all raid members.
If successful: yay! If failure: proceed to step 2. Repeat until victory.
If this is an intuition-based approach, then I don’t know what “intuition” means.
You mention lots of fields (computers, math, science, engineering) where your argument is almost tautological: in a case where you have time to reconsider each decision, a slow but reliable and precise method is better than a snap judgment. Yes, I would agree with you, and I would also agree that logical thinking is better than intuitive thinking in many, many situations.
Of course logical thinking is better when you have time to use it. I’m not asking whether it’s better. I’m asking whether “gut judgments” are accurate, and how accurate they are.
Basically, I see many people claiming that in “crunch time” scenarios, you have no choice but to apply the gut judgment. Ok. But my question is: if you later go back and apply logical reasoning to the (by now, perhaps, irrelevant) problem, does it turn out that your gut judgment was right? How right? How often? Etc.
I would not characterize good play in a WoW raid as based on intuition.
On a PvE server, on in PvE in general—yes, raid bosses are basically a puzzle that you figure out and then execute to the best of your ability. But take a PvP server, say you’re assembling for the raid and are attacked. This is the fight where you have half a second to realize what someone is trying to do to you and come up with a counter.
I hesitate to say that you have to act on your intuition in a PvP fight, probably a better word is memorized (mostly subconsciously) patterns based on experience—that’s what drives your actions.
On a PvP server, if you’re assembling for a (serious) raid and are attacked, you sigh, say “goddamnit… jerks”, and then res as fast as possible in a way that will get you to the raid ASAP. (And that’s back when you couldn’t just teleport directly to the instance from wherever.)
“Memorized patterns based on experience” is a good characterization (often they’re even memorized consciously). Although, there is a nontrivial element of intuition in competitive (arena) PvP, where your opponent’s psychology is an important factor.
On a PvP server, if you’re assembling for a (serious) raid and are attacked, you sigh, say “goddamnit… jerks”, and then res as fast as possible in a way that will get you to the raid ASAP.
That rather depends on your guild. “Screw the raid, we’ve got faces to melt!” is not an uncommon response :-)
I’m asking whether “gut judgments” are accurate, and how accurate they are.
I have very little experience with WoW, so it’s interesting to hear how deliberate and reasoned a high-level raid is. I have a little experience with sports, combat, and combat sports.
It’s pretty surprising that our brains handle abstractions as well as they do. It’s not at all surprising that they can process and integrate sensory information as fast as they can, because that trait is crucial to survival for most animals.
When Kevin Durant fakes a pass and then shoots from 30 feet away, he’s doing something he’s done thousands of times before. It’s a pattern. But he’s adjusting that pattern for many things that weren’t present in practice, and no two shots are exactly alike. His brain is calculating a trajectory much faster than any of us could with pencil and paper, and his cerebellum is “answering” hundreds of individual questions about muscle opposition that our roboticists might not be able to coordinate at all. He misses some shots, of course. But insofar as a made shot counts for accuracy or right judgment, he probably has better accuracy in much less than a second than anybody could achieve with reflection.
Some rudimentary efforts to do so have, on occasion, been made. While wholesale bots (i.e., no real-time human control at all) are totally incapable of performing at the level required to beat high-end raid bosses, certain simple, repetitive parts of the process can be automated with add-ons and macros.
There are two issues here: desirability and difficulty.
Desirability: if you automated those parts, then there wouldn’t be a game. No one wants to just theorycraft for a while and then sit there and watch while things happen automatically. Theorycraft is the metagame. The parts where you actually execute the plan are the gameplay. And the gameplay is fast-paced, exciting, adrenaline-rush-generating, skills-demanding, and cool-looking. The excitement of the gameplay is what WoW raiders live for.
Or at least, most WoW players take this stance. Knowing this, Blizzard has consistently banned any game add-ons that go too far in automating things. There’s a fine line, and sometimes it shifts, and sometimes it’s blurry, but the intent is clear: thou shalt play the game yourself, not write code that will play the game for you. (As with all commandments, precise interpretation is a longer discussion.)
Difficulty: The reason you can’t actually fully automate the steps in question (unless, perhaps, you are the game/boss designer, and have access to all the internal game variables) is largely because:
Positioning (i.e., location and movement of characters in the game world) matters a lot. (The reasons why are several, and probably boring, but take my word for this.)
Timing matters a lot. Which is to say, not only must character ability usage be timed correctly with respect to the behavior of game elements (monsters, environmental events, etc.), but players must also time their actions with respect to, and in response to, what other players are doing.
There are many variables that go into correct play. Combinatorial explosion would make automating this a daunting task. For a human, learning a boss strategy, or a play technique, is faster than devising and implementing an algorithm to execute it. To a human, you can just say “kite that mob over there, then release it when I yell on voice chat”, and (if he’s a skilled player) he won’t need to be told twice. Writing code to do this… is likely possible, but not easy.
Are you suggesting that the ability to model others or respond to nonverbal cues is innate, rather than learned?
Yes, I am not only suggesting but saying it explicitly (but see caveat below). Huge, obvious case in point: the autism spectrum. People on the spectrum (such as myself) have little to no ability to perceive nonverbal cues or (non-explicitly; again see caveat) model others.
Even for neurotypical (that is, non-autistic) people, there is a range of ability in this area.
I would definitely disagree, though proving it would be difficult. I suspect that it’s a matter of internalizing the results of numerous actions and reactions in different situations. In my experience, it’s often developed by people who travel lots or are otherwise exposed to tons of different people in a situation where being friendly and getting on their good side is very helpful. Some of them, pretty bad at socializing before they were in such a situation (and really gave it the necessary effort to learn).
Caveat to the above: I think these skills are innate in most people; that there is a range of ability, with the autism spectrum at one of that range and naturally charismatic, socially apt people on the other; but that the skills can be learned, with effort, as explicit skills.
For instance, autistic people can train themselves to recognize nonverbal cues and social signals; but this is not a matter of simply unconsciously perceiving the cues/signals/situations and just “knowing” their meaning, as it is for most people; rather it is a case of consciously paying attention and looking for things; and the meanings of these cues and signals must be looked up, researched, and memorized. In other words, a logic-based approach to compensate for lack of an intuitive ability.
It is probably also the case that neurotypicals who are not on the extreme positive end of the social ability spectrum, but do not lack the innate intuitive ability, can train their ability in the manner you mention. I would not know, of course, but it seems plausible enough, and consistent with what I’ve heard.
When I was 13 years old, I was a heavily logic-dominant thinker, and I was terrible at reacting under pressure–I found this out when I started taking the required classes to become a lifeguard. I think this is mainly because, even though I could reason through what I was supposed to do, I was misinterpreting the nervousness of social pressure and people watching me perform as uncertainty about what to do. I also tended to be so occupied by thinking things through that I would have “tunnel vision”–my method wasn’t fast enough to flexibly adapt when I thought a situation was one thing and it turned out to be something different.
In first year nursing school, I had gut feelings, and they were screaming at me all the time. I ignored them–justifiably, because they were pretty useless. I didn’t yet have what they call “clinical judgement”, which AFAICT consists of your intuition knowing what details to work from. Four years ago I didn’t really know what it looked like when someone was having trouble breathing–now I could list probably 10 little details to look for. But the mental process isn’t a checklist down those ten items with yes or no for each and making an aggregate score–it’s “this person looks okay” or “crap, this person doesn’t look okay.” And this happens even if I’m not asking myself the question–I look at a patient and my brain pings me that something is wrong. I think the main limitation that my 13-year-old self had to work under was that I ignored my gut feelings, so I frequently didn’t notice new information that didn’t make sense–if it didn’t fit into the mental model I’d made of what was going on, it got filtered out. Intuition is good at noticing confusion. Logical thinking tries to suppress confusion by fitting details into a model even if they don’t fit very well, and it doesn’t answer questions that aren’t asked, either.
Moral of the story: it takes time and effort to train gut feelings. They don’t come from nowhere.
Your experience and values seem to differ from mine in a number of ways; that does seem to be what’s behind the OP’s advice being of different utility to us.
I take by this that you don’t have the experience of it feeling like your brain’s being hijacked into having an emotion that you don’t want?
I guess something that’s atypical about me for a LWer is that I’m very agreeable and somewhat of a conformist. I don’t like to bother other people. Acting on frustration or anger would often make me a bother to other people. Even when I’m in the right, I can fix the situation more effectively from a standpoint of not being angry. My angry self might say things that my later non-angry self would regret, and I’ve gotten pretty good at not doing that.
Very interesting about the training of gut feelings. A bit from my own experience:
I worked for a number of years in tech support positions, where I was often called upon to do PC maintenance/repairs/troubleshooting. After a while, I definitely developed an intuition about what might be wrong with a computer, given some set of symptoms, and often put that intuition to good use in the diagnosis/repair process.
However, one critical advanced skill I learned was not to trust that intuition too much. That is: a machine is brought in for repairs; symptoms are provided; I think “aha, it sounds like a motherboard problem”. Certainly, when going through diagnostic procedures, I should then be on the lookout for confirming evidence. But one of the most serious errors a technician might make in this situation is not being sufficiently thorough in checking the other possibilities for what might be wrong. Other problems might (perhaps more rarely) lead to the same symptoms; furthermore and even more insidiously, the provided symptoms might give no indication whatever about some other, largely unrelated problem.
Astute readers of Less Wrong may recognize such a failure as, in large part, good old confirmation bias.
Edit: And note that the bulwark against making such errors is to have a rigorous diagnostic procedure, follow it, and get into the habit of reasoning about the situation explicitly (bouncing ideas off another person helps with this). In other words: a logic-based approach.
Intuition is good at noticing confusion.
It might (for some people?) be more useful to think of the “noticing confusion” feeling as a distinct thing, rather than simply calling it “intuition”. Certainly I, for one, experience it as a specific mental sensation, so to speak.
I take by this that you don’t have the experience of it feeling like your brain’s being hijacked into having an emotion that you don’t want?
I’m not sure… that is to say, I’m not sure what you mean, exactly, so I’ll attempt to describe something I experience, and maybe you can tell me if it’s the same as what you’re talking about.
Sometimes (though more rarely, these days) I will have a certain sort of negative emotional response, which I would describe as anxiety; it generally comes with restlessness and inability to concentrate. Naturally, this is not an emotion I ever want to be having. I have identified several specific sorts of situations that trigger this.
I don’t know that I’d describe it as feeling like my brain’s being hijacked, but only because it seems logically questionable; “hijacked” implies there’s some agent that is the hijacker. I usually have a limited ability to suppress this feeling by analyzing what I think is causing it (usually it’s worry about certain specific sorts of outcomes/events), and attempting to reason about the likelihood of such an outcome, and what I can do to prevent it. So such an emotion is annoying, but not really mysterious (except insofar as I don’t actually know all the details of my own psychology, but then, all my emotions are mysterious in that broader sense).
I don’t know that I’ve ever had any other sort of emotion that I would say I didn’t want to be having. For example, usually when I feel anger, it’s in situations when I think that it is appropriate to feel anger. In the case of frustration, though, you might be right; frustration might be one emotion that’s dispensable once it’s served its role as a signal.
Even when I’m in the right, I can fix the situation more effectively from a standpoint of not being angry. My angry self might say things that my later non-angry self would regret, and I’ve gotten pretty good at not doing that.
I have, on occasion, said things in anger that led to escalation of conflict, but I don’t think I’ve regretted saying them, since I was in the right, and felt that I was both correct and morally entitled to my comments.
When I said that we seemed to have some different values, I meant that I don’t think there’s anything inherently wrong with being angry, if the anger is justified.
However, one critical advanced skill I learned was not to trust that intuition too much. That is: a machine is brought in for repairs; symptoms are provided; I think “aha, it sounds like a motherboard problem”. Certainly, when going through diagnostic procedures, I should then be on the lookout for confirming evidence. But one of the most serious errors a technician might make in this situation is not being sufficiently thorough in checking the other possibilities for what might be wrong.
Very true. Confirmation bias and not looking hard enough for a diagnosis is a big issue in medicine, too. I’m not sure if there’s a difference between health care practitioners who were originally logic-dominant thinkers or originally intuition-dominant thinkers, or whether both struggle and have to learn the other skill anyway.
A difference is that when you’re working, you have time to be as slow, thoughtful, and deliberate as you want when figuring out a problem. Obviously it’s better to reason things through as well as noticing intuitions, but System 2 (roughly, explicit reasoning) is slow and effortful and puts a heavy load on working memory, and System 1 (roughly, intuitions) is fast and doesn’t fill up working memory. My younger self wanted to reason though everything logically–and as a result, because nursing is a profession where you’re always working under time constraints, I was always a step behind everyone else, always took longer to get started at the beginning of the day, always stayed an hour past the end of a shift to finish charting. I don’t think this is because I’m a “slow” thinker–I finish written exams in half the time that it takes most of the other nursing students. Also, in my experience having a load on working memory increases confirmation bias–I don’t know if this has been studied, although it wouldn’t be a hard study to do. I’m more curious about things that don’t make sense now.
And note that the bulwark against making such errors is to have a rigorous diagnostic procedure, follow it, and get into the habit of reasoning about the situation explicitly (bouncing ideas off another person helps with this). In other words: a logic-based approach.
Modern medicine makes use of checklists a lot. I think this is awesome. I don’t need any urging to use them; I was making personal checklists on my work sheet way before I knew this was already a thing. And “if in doubt, ask someone else to come have a look” is pretty universal too. Also not something I need urging to do.
I don’t know that I’d describe it as feeling like my brain’s being hijacked, but only because it seems logically questionable; “hijacked” implies there’s some agent that is the hijacker.
I don’t literally mean that. It’s just what it feels like.
I have, on occasion, said things in anger that led to escalation of conflict, but I don’t think I’ve regretted saying them, since I was in the right, and felt that I was both correct and morally entitled to my comments.
Even when this is the case, I don’t find that anger helps me get what I want. Then again, being agreeable, a lot of what I want is “not to be in conflict anymore.” Also, I think some people kind of enjoy the powerful feeling that anger gives them. Whereas I find the feeling of anger aversive.
It seems we mostly agree about the usefulness and applicability of gut feelings, as well as their limitations. (Of course, if someone else is aware of any research about their accuracy, I am still interested in seeing it.)
One way I would summarize the ideal setup is: during “downtime”, use logic-based reasoning to come up with a rigorous and easy-to-apply procedure; during “crunch time”, use intuition to generate probable avenues of investigation and likely candidates for diagnosis and solution; supplement with the pre-developed procedure to guard against biases and ensure correct usage of intuition-derived data.
I may have had an explicit belief that my own intuitions were wrong most of the time. I don’t think I had a belief that following intuitions period was bad; I always admired people who seemed to be able to do so and get good results.
Because almost every book or movie out there has something in it, somewhere, that would annoy me, and I like enjoying books and movies. Because the movie’s content is there anyway and me being cranky about it does absolutely nothing about the underlying culture that makes it normal for nerds to be made fun of in an action movie. Once I notice that I’m cranky and realize why, if I want to do something productive about it I can mention to my friends that I don’t think nerds should be characterized that way, or I can write a blog post about it...without being cranky. Emotions are a signal. Once I’ve decoded that signal, I don’t need the negative emotion anymore, unless it makes me more effective at doing something, which is rarely the case.
We live in a world that’s practically designed to play on our emotions. TV ads basically designed to make me insecure about my appearance/coolness/etc so I’ll want to buy their product? I don’t need to walk around with lingering insecurities; I thought I looked fine and was cool enough two minutes ago, and an ad by people whose job is to manipulate me shouldn’t change that. Lots more people, in more complicated social relationships, than the ancestral environment? Sometimes people do things that annoy me, but I know that the thing they did was harmless and I don’t think it should annoy me, and whereas if they were a family member I’d ask them not to do it for my sake, they’re actually just an acquaintance, and I don’t want to go around feeling annoyed. Or sometimes I make a faux pas and I apologize and take all the necessary steps to fix it–I don’t want to keep feeling guilty after this, but I often do. Sometimes I have to make a choice between two different things I want to do, and I pick the one I want more, but then I’m sad about the one I didn’t pick–I don’t want to keep feeling sad and enjoy the thing that I did pick less.
They’re based off data that I could in theory notice and analyze logically–the way someone’s skin feels, the way someone looks when they’re breathing, etc. It just doesn’t feel like logical analysis anymore, it feels like “oh crap, X.” I think that a lot of intuition could be logical reasoning, it’s just learned to the point that it happens fast at a subconscious level by pattern-recognition. And they don’t have to be anywhere close to 100% accurate to be helpful. Usually the first thing I do with a gut feeling is get more data. “This patient seems worse...okay, I’ll check all their vital signs again.” Sometimes I’m right and something has changed. Sometimes I’m wrong. I’d still rather have the gut feelings than not. I don’t know how much data there is on this.
Hm. Thank you for taking the time to explain; I definitely appreciate it. Your experience and values seem to differ from mine in a number of ways; that does seem to be what’s behind the OP’s advice being of different utility to us.
As for the bit about accuracy of gut feelings: I take your point about them being a good signal to investigate further. I do remain quite dubious about the use of the gut feelings directly, in place of explicit reasoning. I would very much like to see some data about this.
The advantage of gut feelings and intuition lie with their ability to synthesize years of experience and thousands of variables into one answer within less than a second.
When is this necessary?
During a conversation, someone watching your face is going to be observing how you react (even in the smallest possible ways) as they speak. You don’t have an hour, five minutes, or even two seconds to decide how to present yourself; they’re going to judge you based on that instantaneous reaction (or a lack of one, including a delayed reaction or straight face.)
Anyone who is a natural “people person”—the kinds of people who can get almost anything they want from anyone around them, who make great salesmen or politicians—is going to need to be able to continuously react “properly”, and that means intuitive judgments.
Same thing with any kind of games/sports, or literally any other situation where a quick reaction is required and not immediately responding will doom you no matter what.
The domain of “behaving in such a way as to gain and maintain an advantageous position in social interactions” is very different from other domains, like “diagnose and repair difficulties with computer equipment”, “diagnose and treat afflictions in human beings”, “understand mathematical concepts”, and almost anything else. It seems to me that the domain of social interactions with other humans is in fact a unique domain, not properly comparable (in the context of the current discussion) to anything else.
Certainly gut feelings are key in social interactions; in fact, the most charismatic, likable, and socially successful people do what they do largely unconsciously, and are almost entirely unable not only to explain their technique to others, but even to recognize there is a technique, that other people are not as skilled at using.
My question is, how accurate are gut feelings in other domains — especially those domains where there is are objectively right answers and wrong answers, and where it is possible and even easy, in principle, to compare the answer you get from your gut feeling to the actual right answer? In the treatment of computers and people, in math, in science, in engineering? (Answering this question requires data!)
What’s more: I really don’t think that this
is an accurate characterization of where gut feelings in successful social interactors come from. It’s been my experience that such instinctual social success is largely innate. Oh sure, it may be honed, but saying that the gut feeling is a synthesis of years of experience is just almost certainly not what’s going on there. More likely it’s a naturally great ability to model others, to respond (unconsciously) to nonverbal cues, etc.
How about sports and fast paced games?
Players are often required to make decisions with no time whatsoever to plan. For example, you might find yourself surrounded by enemies with no warning.
You need to know whether to run on foot, to teleport away, or to fight.
The difference between reacting in a third of a second and a fourth of a second could mean life or death.
Success in this situation, assuming it’s possible, is dependent on your experience in similar situations and your instinctual reaction. Since you do not have the time to think, your decision is almost guaranteed to be imperfect, but any improvement in it is highly beneficial.
Obviously, the same would often apply in war or in certain crisis situations.
You mention lots of fields (computers, math, science, engineering) where your argument is almost tautological: in a case where you have time to reconsider each decision, a slow but reliable and precise method is better than a snap judgment. Yes, I would agree with you, and I would also agree that logical thinking is better than intuitive thinking in many, many situations.
Are you suggesting that the ability to model others or respond to nonverbal cues is innate, rather than learned? I would definitely disagree, though proving it would be difficult. I suspect that it’s a matter of internalizing the results of numerous actions and reactions in different situations. In my experience, it’s often developed by people who travel lots or are otherwise exposed to tons of different people in a situation where being friendly and getting on their good side is very helpful. Some of them, pretty bad at socializing before they were in such a situation (and really gave it the necessary effort to learn).
I disagree, however, when you say that being socially successful is innate.
I do not play sports, but I did spend several years doing high-end raiding (mostly as a main tank) in World of Warcraft, which I think fits your criteria. Raid play is fast-paced and demanding, with necessary reaction times measured in fractions of a second.
I would not characterize good play in a WoW raid as based on intuition. Here is, basically, the process for beating a new, challenging raid boss:
Go in, try the boss. The entire raid dies horribly, of course.
Meticulously, exhaustively analyze the combat log. Note down all observations made of boss behavior. Correlate data.
Brainstorm solutions, based on raid leader’s and key raid members’ comprehensive, minutely detailed knowledge of game mechanics.
Make detailed plan. The plan implicitly includes generally correct play from all raid members; note that for almost all classes in WoW, optimal play means following detailed algorithms for ability usage, often worked out at length by top “theorycrafters”, who are often people with advanced degrees in physics and mathematics (no, I am not exaggerating) — plus, of course, extensive experience, to the point where playing correctly is at the level of muscle memory.
Attempt to execute plan. Correct execution demands precise, down-to-the-second performance from all raid members.
If successful: yay! If failure: proceed to step 2. Repeat until victory.
If this is an intuition-based approach, then I don’t know what “intuition” means.
Of course logical thinking is better when you have time to use it. I’m not asking whether it’s better. I’m asking whether “gut judgments” are accurate, and how accurate they are.
Basically, I see many people claiming that in “crunch time” scenarios, you have no choice but to apply the gut judgment. Ok. But my question is: if you later go back and apply logical reasoning to the (by now, perhaps, irrelevant) problem, does it turn out that your gut judgment was right? How right? How often? Etc.
On a PvE server, on in PvE in general—yes, raid bosses are basically a puzzle that you figure out and then execute to the best of your ability. But take a PvP server, say you’re assembling for the raid and are attacked. This is the fight where you have half a second to realize what someone is trying to do to you and come up with a counter.
I hesitate to say that you have to act on your intuition in a PvP fight, probably a better word is memorized (mostly subconsciously) patterns based on experience—that’s what drives your actions.
On a PvP server, if you’re assembling for a (serious) raid and are attacked, you sigh, say “goddamnit… jerks”, and then res as fast as possible in a way that will get you to the raid ASAP. (And that’s back when you couldn’t just teleport directly to the instance from wherever.)
“Memorized patterns based on experience” is a good characterization (often they’re even memorized consciously). Although, there is a nontrivial element of intuition in competitive (arena) PvP, where your opponent’s psychology is an important factor.
That rather depends on your guild. “Screw the raid, we’ve got faces to melt!” is not an uncommon response :-)
Dirty casuals :p
I have very little experience with WoW, so it’s interesting to hear how deliberate and reasoned a high-level raid is. I have a little experience with sports, combat, and combat sports.
It’s pretty surprising that our brains handle abstractions as well as they do. It’s not at all surprising that they can process and integrate sensory information as fast as they can, because that trait is crucial to survival for most animals.
When Kevin Durant fakes a pass and then shoots from 30 feet away, he’s doing something he’s done thousands of times before. It’s a pattern. But he’s adjusting that pattern for many things that weren’t present in practice, and no two shots are exactly alike. His brain is calculating a trajectory much faster than any of us could with pencil and paper, and his cerebellum is “answering” hundreds of individual questions about muscle opposition that our roboticists might not be able to coordinate at all. He misses some shots, of course. But insofar as a made shot counts for accuracy or right judgment, he probably has better accuracy in much less than a second than anybody could achieve with reflection.
Yes. This is exactly right, and true in WoW as well.
So, I realize this is off-topic, but I’m really curious: wouldn’t it be easier to automate steps 1, 5 and 6?
Some rudimentary efforts to do so have, on occasion, been made. While wholesale bots (i.e., no real-time human control at all) are totally incapable of performing at the level required to beat high-end raid bosses, certain simple, repetitive parts of the process can be automated with add-ons and macros.
There are two issues here: desirability and difficulty.
Desirability: if you automated those parts, then there wouldn’t be a game. No one wants to just theorycraft for a while and then sit there and watch while things happen automatically. Theorycraft is the metagame. The parts where you actually execute the plan are the gameplay. And the gameplay is fast-paced, exciting, adrenaline-rush-generating, skills-demanding, and cool-looking. The excitement of the gameplay is what WoW raiders live for.
Or at least, most WoW players take this stance. Knowing this, Blizzard has consistently banned any game add-ons that go too far in automating things. There’s a fine line, and sometimes it shifts, and sometimes it’s blurry, but the intent is clear: thou shalt play the game yourself, not write code that will play the game for you. (As with all commandments, precise interpretation is a longer discussion.)
Difficulty: The reason you can’t actually fully automate the steps in question (unless, perhaps, you are the game/boss designer, and have access to all the internal game variables) is largely because:
Positioning (i.e., location and movement of characters in the game world) matters a lot. (The reasons why are several, and probably boring, but take my word for this.)
Timing matters a lot. Which is to say, not only must character ability usage be timed correctly with respect to the behavior of game elements (monsters, environmental events, etc.), but players must also time their actions with respect to, and in response to, what other players are doing.
There are many variables that go into correct play. Combinatorial explosion would make automating this a daunting task. For a human, learning a boss strategy, or a play technique, is faster than devising and implementing an algorithm to execute it. To a human, you can just say “kite that mob over there, then release it when I yell on voice chat”, and (if he’s a skilled player) he won’t need to be told twice. Writing code to do this… is likely possible, but not easy.
Not exactly. Yeah, I know this isn’t WoW.
Yeah, my comments were WoW-specific. Roguelikes are very different.
Yes, I am not only suggesting but saying it explicitly (but see caveat below). Huge, obvious case in point: the autism spectrum. People on the spectrum (such as myself) have little to no ability to perceive nonverbal cues or (non-explicitly; again see caveat) model others.
Even for neurotypical (that is, non-autistic) people, there is a range of ability in this area.
Caveat to the above: I think these skills are innate in most people; that there is a range of ability, with the autism spectrum at one of that range and naturally charismatic, socially apt people on the other; but that the skills can be learned, with effort, as explicit skills.
For instance, autistic people can train themselves to recognize nonverbal cues and social signals; but this is not a matter of simply unconsciously perceiving the cues/signals/situations and just “knowing” their meaning, as it is for most people; rather it is a case of consciously paying attention and looking for things; and the meanings of these cues and signals must be looked up, researched, and memorized. In other words, a logic-based approach to compensate for lack of an intuitive ability.
It is probably also the case that neurotypicals who are not on the extreme positive end of the social ability spectrum, but do not lack the innate intuitive ability, can train their ability in the manner you mention. I would not know, of course, but it seems plausible enough, and consistent with what I’ve heard.
More on gut feelings:
When I was 13 years old, I was a heavily logic-dominant thinker, and I was terrible at reacting under pressure–I found this out when I started taking the required classes to become a lifeguard. I think this is mainly because, even though I could reason through what I was supposed to do, I was misinterpreting the nervousness of social pressure and people watching me perform as uncertainty about what to do. I also tended to be so occupied by thinking things through that I would have “tunnel vision”–my method wasn’t fast enough to flexibly adapt when I thought a situation was one thing and it turned out to be something different.
In first year nursing school, I had gut feelings, and they were screaming at me all the time. I ignored them–justifiably, because they were pretty useless. I didn’t yet have what they call “clinical judgement”, which AFAICT consists of your intuition knowing what details to work from. Four years ago I didn’t really know what it looked like when someone was having trouble breathing–now I could list probably 10 little details to look for. But the mental process isn’t a checklist down those ten items with yes or no for each and making an aggregate score–it’s “this person looks okay” or “crap, this person doesn’t look okay.” And this happens even if I’m not asking myself the question–I look at a patient and my brain pings me that something is wrong. I think the main limitation that my 13-year-old self had to work under was that I ignored my gut feelings, so I frequently didn’t notice new information that didn’t make sense–if it didn’t fit into the mental model I’d made of what was going on, it got filtered out. Intuition is good at noticing confusion. Logical thinking tries to suppress confusion by fitting details into a model even if they don’t fit very well, and it doesn’t answer questions that aren’t asked, either.
Moral of the story: it takes time and effort to train gut feelings. They don’t come from nowhere.
I take by this that you don’t have the experience of it feeling like your brain’s being hijacked into having an emotion that you don’t want?
I guess something that’s atypical about me for a LWer is that I’m very agreeable and somewhat of a conformist. I don’t like to bother other people. Acting on frustration or anger would often make me a bother to other people. Even when I’m in the right, I can fix the situation more effectively from a standpoint of not being angry. My angry self might say things that my later non-angry self would regret, and I’ve gotten pretty good at not doing that.
Very interesting about the training of gut feelings. A bit from my own experience:
I worked for a number of years in tech support positions, where I was often called upon to do PC maintenance/repairs/troubleshooting. After a while, I definitely developed an intuition about what might be wrong with a computer, given some set of symptoms, and often put that intuition to good use in the diagnosis/repair process.
However, one critical advanced skill I learned was not to trust that intuition too much. That is: a machine is brought in for repairs; symptoms are provided; I think “aha, it sounds like a motherboard problem”. Certainly, when going through diagnostic procedures, I should then be on the lookout for confirming evidence. But one of the most serious errors a technician might make in this situation is not being sufficiently thorough in checking the other possibilities for what might be wrong. Other problems might (perhaps more rarely) lead to the same symptoms; furthermore and even more insidiously, the provided symptoms might give no indication whatever about some other, largely unrelated problem.
Astute readers of Less Wrong may recognize such a failure as, in large part, good old confirmation bias.
Edit: And note that the bulwark against making such errors is to have a rigorous diagnostic procedure, follow it, and get into the habit of reasoning about the situation explicitly (bouncing ideas off another person helps with this). In other words: a logic-based approach.
It might (for some people?) be more useful to think of the “noticing confusion” feeling as a distinct thing, rather than simply calling it “intuition”. Certainly I, for one, experience it as a specific mental sensation, so to speak.
I’m not sure… that is to say, I’m not sure what you mean, exactly, so I’ll attempt to describe something I experience, and maybe you can tell me if it’s the same as what you’re talking about.
Sometimes (though more rarely, these days) I will have a certain sort of negative emotional response, which I would describe as anxiety; it generally comes with restlessness and inability to concentrate. Naturally, this is not an emotion I ever want to be having. I have identified several specific sorts of situations that trigger this.
I don’t know that I’d describe it as feeling like my brain’s being hijacked, but only because it seems logically questionable; “hijacked” implies there’s some agent that is the hijacker. I usually have a limited ability to suppress this feeling by analyzing what I think is causing it (usually it’s worry about certain specific sorts of outcomes/events), and attempting to reason about the likelihood of such an outcome, and what I can do to prevent it. So such an emotion is annoying, but not really mysterious (except insofar as I don’t actually know all the details of my own psychology, but then, all my emotions are mysterious in that broader sense).
I don’t know that I’ve ever had any other sort of emotion that I would say I didn’t want to be having. For example, usually when I feel anger, it’s in situations when I think that it is appropriate to feel anger. In the case of frustration, though, you might be right; frustration might be one emotion that’s dispensable once it’s served its role as a signal.
I have, on occasion, said things in anger that led to escalation of conflict, but I don’t think I’ve regretted saying them, since I was in the right, and felt that I was both correct and morally entitled to my comments.
When I said that we seemed to have some different values, I meant that I don’t think there’s anything inherently wrong with being angry, if the anger is justified.
Very true. Confirmation bias and not looking hard enough for a diagnosis is a big issue in medicine, too. I’m not sure if there’s a difference between health care practitioners who were originally logic-dominant thinkers or originally intuition-dominant thinkers, or whether both struggle and have to learn the other skill anyway.
A difference is that when you’re working, you have time to be as slow, thoughtful, and deliberate as you want when figuring out a problem. Obviously it’s better to reason things through as well as noticing intuitions, but System 2 (roughly, explicit reasoning) is slow and effortful and puts a heavy load on working memory, and System 1 (roughly, intuitions) is fast and doesn’t fill up working memory. My younger self wanted to reason though everything logically–and as a result, because nursing is a profession where you’re always working under time constraints, I was always a step behind everyone else, always took longer to get started at the beginning of the day, always stayed an hour past the end of a shift to finish charting. I don’t think this is because I’m a “slow” thinker–I finish written exams in half the time that it takes most of the other nursing students. Also, in my experience having a load on working memory increases confirmation bias–I don’t know if this has been studied, although it wouldn’t be a hard study to do. I’m more curious about things that don’t make sense now.
Modern medicine makes use of checklists a lot. I think this is awesome. I don’t need any urging to use them; I was making personal checklists on my work sheet way before I knew this was already a thing. And “if in doubt, ask someone else to come have a look” is pretty universal too. Also not something I need urging to do.
I don’t literally mean that. It’s just what it feels like.
Even when this is the case, I don’t find that anger helps me get what I want. Then again, being agreeable, a lot of what I want is “not to be in conflict anymore.” Also, I think some people kind of enjoy the powerful feeling that anger gives them. Whereas I find the feeling of anger aversive.
It seems we mostly agree about the usefulness and applicability of gut feelings, as well as their limitations. (Of course, if someone else is aware of any research about their accuracy, I am still interested in seeing it.)
One way I would summarize the ideal setup is: during “downtime”, use logic-based reasoning to come up with a rigorous and easy-to-apply procedure; during “crunch time”, use intuition to generate probable avenues of investigation and likely candidates for diagnosis and solution; supplement with the pre-developed procedure to guard against biases and ensure correct usage of intuition-derived data.
Does this sound like a fair summary?
Just curious, did you have any explicit beliefs that made you ignore your intuition?
I may have had an explicit belief that my own intuitions were wrong most of the time. I don’t think I had a belief that following intuitions period was bad; I always admired people who seemed to be able to do so and get good results.