My take on the Mary’s Room thought experiment

For those of you who aren’t familiar with the thought experiment:

Mary’s room

Mary is a brilliant scientist who is, for whatever reason, forced to investigate the world from a black and white room via a black and white television monitor. She specializes in the neurophysiology of vision and acquires, let us suppose, all the physical information there is to obtain about what goes on when we see ripe tomatoes, or the sky, and use terms like ‘red’, ‘blue’, and so on. She discovers, for example, just which wavelength combinations from the sky stimulate the retina, and exactly how this produces via the central nervous system the contraction of the vocal cords and expulsion of air from the lungs that results in the uttering of the sentence ‘The sky is blue’. […] What will happen when Mary is released from her black and white room or is given a color television monitor? Will she learn anything or not?

There seems to be a paradox. On one hand, Mary knows everything there is to know about seeing color, yet our intuition says she’ll learn something new once released into the world of color. How can she know everything and then subsequently learn something new? The thought experiment misleads us into this sort of conclusion because our idea of knowledge and learning is fuzzy.

What does it mean for humans to know something? In a broad sense, we know a system when we have an internal model of that system that enables us to predict its behaviors for any set of inputs. Sort of like an accurate city map that enables us to predict where everything is inside the city. We learn something about a system when we get new information that causes us to modify our internal model of the system to maintain its accuracy. If we learned that a new road was built in our city, we’d have to include it in our city map in order to maintain its accuracy. We know everything about a system when no new information could necessitate modifying our internal model in order to improve its accuracy.

But like city maps, our internal models are only representations of reality, not exact replicas. They’re inherently missing details that exist in reality. A city map wouldn’t include a representation of a fly on a sidewalk. And leaving that detail out doesn’t make it a worse map of the city. Similarly, our internal models of systems in the world leave certain details out (evolution designed it that way). But it’s still meaningful to say we completely know a system even if our models don’t represent every detail of reality.

Mary is said to know everything about visual processing. However, her knowledge is based on an internal model that doesn’t represent every detail of reality. For instance, her model may not be representing how the nervous system works on a quantum scale. But despite this fact, we can still say she knows everything about visual processing. Learning about the nervous system on a quantum scale wouldn’t require her to modify her visual processing model because including representations of it wouldn’t affect the accuracy of her predictions.

When Mary first encounters a red rose, she’ll experience redness for the first time and feel like she has learned something new. Seeing a red rose activated a subconscious process that created a representation of reality and made that representation available for conscious attention. She had a new experience when focusing her conscious attention on her subconsciously produced representation of the color red.

How does this new representation of reality, the experience of redness, play into Mary’s understanding of visual processing? The answer is, it doesn’t. The new experience doesn’t lend new information that requires her to update her visual processing model in order to improve it. In a meaningful way, Mary didn’t learn anything new about visual processing. Her existing models of her brain and visual processing could’ve predicted her brain’s exact changes upon seeing a red rose for the first time – without incorporating into her models a representation of her new redness experience.

Mary cannot activate the subconscious process that generates the redness experience simply by having an accurate conscious model of the process. But activating that subconscious process and experiencing its productions isn’t a necessary condition for understanding it. Just like how you can maximally understand something by having a model that doesn’t include a representation of every detail of reality.

When Mary sees a red rose for the first time, she might feel as if she’s learned something new, but her understanding of reality will remain unchanged.

Bye Felipe misses the problem

There’s a new internet trend where women publicly shame men who send them abusive or hostile messages on online dating sites like OkCupid and Tinder. Here are some examples from Alexandra Tweten’s instagram byefelipe:



When I first saw these screenshots, I thought they were pretty funny. There’s entertainment value in seeing grown men drop the civilized-person charade to unleash their inner 13-year-old self who calls you a “fucking asshole gaylord” when you beat him at Halo.

But the more I saw, the less funny it got.

I stopped laughing at the tantrum-throwing man-children and started sympathizing instead. After seeing hundreds of these screenshots, on byefelipe and similar blogs, I suspected it’s a symptom of a common problem men have. Alexandra thinks it’s misogyny and a toxic sense of entitlement:

After seeing these disturbing messages grouped together, it’s easy to come to the conclusion that our society has a misogyny problem… the thing that drives men to brutally injure women who ignore them are all connected to the sense of toxic entitlement some men possess.

Misogyny is commonly defined as a hatred, dislike, or mistrust of women. And yes, if these men were misogynists, their messages would probably look a lot the one’s in Alexandra’s collection.

But “misogyny” is unnecessarily used here as a label for a more common and less complex behavior that many people exhibit. As I’ve alluded above, young boys and immature men often have temper issues that result in a bout of low-grade insults. Whether they’re denied a chance at having sex or making the game-winning headshot, these men who can’t control their anger will eventually make it known.

Labeling this sore loser behavior popping up in online dating as “misogynistic” might be technically correct, but it distracts us from what’s actually going on:

There’s a population of men whose egos are constantly battered by frequent rejection, all the while feeling hopeless, frustrated, and confused throughout the beating. With their egos in critical condition, they desperately look for a quick relief. And blaming women who reject them does the job well. It absolves them from taking personal responsibility for their failures, which helps their ego live to see another day.

They’ll believe they’ve been playing the game right all along, and that they’re only being rejected because she’s a huge bitch, highly superficial, low class, or just clueless about social/dating norms. But these are the smarter and more mature sore losers. Most don’t even bother rationalizing their anger, they just feel a strong impulse to harm that which they hate.

And that’s when verbal abuse comes into play.

They transform into a shitty emotionally abusive machine that fires out schoolyard bully insults, the lowest form of insults. A good insult requires careful observation of your target’s flaws and sufficient communication skills and self-awareness to express your observations and feelings about them.

The schoolyard bully insult is a lot easier to manufacture:

  1. Identify the stereotype that fits your target best
  2. Select several of the most insulting cliches associated with said stereotype
  3. Tell your target that your chosen cliches apply to them

Little boys playing Xbox Live and Bye Felipe man-children are doing this same strategy.

With this perspective, it’s hard to make the case that the Bye Felipe phenomenon is evidence of a misogynistic culture. More appropriately, it’s evidence of immature, frustrated, and confused men playing a game they suck at. It’s similar to what inspired the Xbox Live boy to eloquently call his opponent a “fucking asshole gaylord”. It’s weak evidence the boy is homophobic, and strong evidence he’s just an immature sore loser.

We expect adults, and especially adult males, to be mature and in control of their emotions. So when adults use schoolyard bully insults, we’re biased to assume they mean what they say. When in reality, they aren’t optimizing for self-expression and instead optimizing for hurting others and minimizing cognitive energy. Even if they were optimizing for self-expression, they’re being too contradictory to be taken seriously. For instance, based on the conversations Bye Felipe men have moments before being rejected, you would assume they’re interested in the woman they later claim are too fat, ugly, or slutty to date.

Alexandra admits there’s a larger problem at play that she can’t identify:

I have been asked multiple times, “What’s the answer to this? What can these dating sites do to curb this problem?” And I struggle to answer, because this is just a symptom of a larger problem

Alexandra, a self-proclaimed feminist writer since 2010, is right that what she’s experiencing is a symptom of a larger problem, but she’s too blindsided by her feminist perspective to realize that the problem is unrelated to misogyny and self-entitlement. The problem is that immature, low self-esteem men who don’t understand women have started online dating.

The problem is that immature, low self-esteem men who don’t understand women have started online dating. These men should be getting help instead of being publicly shamed.

Bye Felipe doesn’t address the underlying issues, it just showcases the symptoms. It’s analogous to posting videos of people with Tourette’s syndrome cursing women in public and blaming it on the patriarchy without acknowledging the Tourette’s.

15 Dating and Relationships Lessons From OKCupid Users In Their 70s and 80s

OKCupid’s oldest users unintentionally impart their dating and relationship wisdom in their match questions, where they tend to leave thoughtful, candid explanations to their answers. Below are 15 dating and relationship lessons taken from both men and women in their 70s and 80s:

1. Being “good” at sex is largely determined by whether you’re into it


2. Faking orgasms might be common practice, even amongst men


3. Having multiple sex partners doesn’t make you incapable of being monogamous


4. You can get love from sex-less relationships


5. Your openness with your feelings can depend on how comfortable your partner makes you feel


6. Put more effort into break-ups of long-term relationships 


7. You can’t keep every promise – including the promise to always love someone


8.  Our dates don’t just evaluate us on our abilities, but on us as a whole


9. Physical attractiveness may depend more on love and affection when we’re older


10. The drive for affection and intimacy we’ve experienced in our youth can be just as strong, if not stronger, as we age


11. Whether it’s doing drugs, watching TV, or eating fatty foods, don’t judge your date by their vice of choice; judge them by how well they can do them in moderation


12. Both passion and dedication are necessary for a better relationship


13. You can have multiple loving relationships simultaneously


14. Your jealousy is a sign of an unhealthy relationship 


15. Old people are funny


Ethicality of Denying Agency

If your 5-year-old seems to have an unhealthy appetite for chocolate, you’d routinely take measures to prevent them from eating it. Any time they’d ask you to buy them some, you’d probably refuse their request, even if they begged. You might make sure that any chocolate in the house is well-hidden and out of their reach. You might even confiscate chocolate they already have, like if you forced them to throw out half their Halloween candy. You’d almost certainly trigger a temper tantrum and considerably worsen their mood. But no one would label you an unrelenting tyrant. Instead, you’d be labeled a good parent.

Your 5-year-old isn’t expected to have the capacity to understand the consequences to their actions, let alone have the efficacy to accomplish the actions they know are right. That’s why you’re a good parent when you force them to do the right actions, even against their explicit desires.

You know chocolate is a superstimulus and that 5-year-olds have underdeveloped mental executive functions. You have good reasons to believe that your child’s chocolate obsession isn’t caused by their agency, and instead caused by an obsolete evolutionary adaptation. But from your child’s perspective, desiring and eating chocolate is an exercise in agency. They’re just unaware of how their behaviors and desires are suboptimal. So by removing their ability to act upon their explicit desires, you’re denying their agency.

So far, denying agency doesn’t seem so bad. You have good reason to believe your child isn’t capable of acting rationally and you’re only helping them in the long run. But the ethicality gets murky when your assessment of their rationality is questionable.

Imagine you and your mother have an important flight to catch 2 hours from now. You realize that you have to leave to the airport now in order to make it on time. As you’re about to leave, you recalled the 2 beers you recently consumed. But you feel the alcohol left in your system will barely affect your driving, if at all. The problem is that if your mother found out about your beer consumption, she’d refuse to be your passenger until you completely sobered up – as she’s done in the past. You know this would cause you to miss your flight because she can’t drive and there are no other means of transportation.

A close family member died in a drunk driving accident several years ago and, ever since, she overreacts to drinking and driving risks. You think her reaction is irrational and reveals she has non-transitive preferences. For example, one time she was content on being your passenger after you warned her you were sleep deprived and your driving might be affected. Another time she refused to be your passenger after finding out you had one cocktail that hardly affected you. She’s generally a rational person, but with the recent incident and her past behavior, you deem her incapable of having a calibrated reaction. With all this in mind, you contemplate the ethicality of denying her agency by not telling her about your beer drinking.

Similar to the scenario with your 5-year-old, your intention is to ultimately help the person whose agency you’re denying. But in the scenario with your mother, it’s less clear whether you have enough information or are rational enough yourself to assess your mother’s capacity to act within her preferences. Humans are notoriously good at self-deception and rationalizing their actions. Your motivation to catch your flight might be making you irrational about how much alcohol affects your driving. Or maybe the evidence you collected against her rationality is skewed by confirmation bias. If you’re wrong about your assessment, you’d be disrespecting her wishes.

I can modify the scenario to make its ethicality even murkier. Imagine your mother wasn’t catching the plane with you. Instead, you promised to drive her back to her retirement home before your flight. You don’t want to break your promise nor miss your flight, so you contemplate not telling her about your beer consumption.

In this modified version, you’re not actually making your mother better off by denying her agency – you’re only benefiting yourself. You just believe her reaction to your beer consumption isn’t calibrated, and it would cause you to miss your flight. Even if you had plenty of evidence to back up your assessment of her rationality, would it be ethical to deny her agency when it’s only benefiting you?

Ambiguous Communication is a Social Skill

Imagine you made dinner plans with your acquaintance, Jill, a few days ago. One hour before it was time to meet, you receive a message from her saying “Sorry! I have to bail on dinner tonight. Stressed out from work. Maybe next week!”. Your goal is to express that you’re skeptical that her excuse warrants the last minute cancellation and that you’re somewhat upset about it. You could either respond with:

  1. “I’m kind of upset. Just being stressed out from work doesn’t sound like a good excuse for making such a last minute cancellation.”
  2. “No problem, next week sounds good”

Option 1 gives Jill little room to interpret your message to mean something other than what you intended to communicate. This unambiguous message pretty much guarantees you’ll achieve your goal, but you’d be naive to ignore the risk involved.

You might make Jill feel uncomfortable or guilty for knowing you think negatively of her. She may also resent you for undermining what may be a legitimate excuse in her mind. Since she’s only your acquaintance, she may feel you’re being too nosy and don’t deserve a more detailed explanation. On the other hand, she may not respond negatively at all, and possibly respect you more for enforcing your standards and being a direct communicator. But since you don’t know her well, nor have you developed a high level of trust with her, option 1 is a high-stakes gamble.

Option 2 lets you avoid the risk of option 1 entirely. But it guarantees you won’t achieve your goal. There’s virtually no chance Jill will interpret your message to mean what you intended.

Both options 1 and 2 seem suboptimal. You’re forced to either go all in or fold. Fortunately, there’s a third option that lets you figuratively “stay in the game” without risking it all. This approach uses the social skill of ambiguous communication, i.e., crafting and sending ambiguous messages that blur the meaning and intention behind what you say and offers you and your communication partner competing interpretations to choose from.

When you communicate unambiguously, you’re introducing a single, high-evidence, and rigidly-defined interpretation of your message. If you can’t confidently predict your partner’s reaction to the message’s interpretation, then you’re risking negative consequences that may be irreversible or hard to undo.

When you communicate ambiguously, you’re sending a message that can be interpreted by you and your partner in more ways than one. With enough uncertainty in any one particular interpretation, both of you can benefit.

With ambiguous communication, you have a third option:

3. Ambiguous message: “Wait until next week? That’s going to cause me too much stress” or “You should buy a stress ball for next week”

Option 3 is like a combination of options 1 and 2. You reduce the risk of option 1 because the ambiguity leaves Jill with some uncertainty about whether the interpretation that you’re upset and discredit her excuse is the right one. The ambiguity also frees you from committing to that interpretation too.

If you find that Jill’s reaction to your intended interpretation is causing damage, you could convince her that a different interpretation was the true one all along without having to contradict your past self. For instance, Jill might interpret “Wait until next week? That’s going to cause me too much stress” to mean that you don’t care about her well-being. If she communicates that sentiment, you can deny her interpretation and convince her of a different interpretation – like that you just wanted to see her this week and not next week.

Besides reducing risk, option 3 also gives you a chance to achieve your goal. Jill is likely to correctly interpret your message to mean that you’re upset and discredit her excuse. And she might bring up the subject in conversation on her own accord. She might, however, make the wrong interpretation (intentionally or unintentionally) and start a conversational thread that won’t help you reach your goal. For instance, if you say “You should buy a stress ball for next week”, she might think you’re trying to offer genuine advice and initiate a conversation about stress-management strategies.

In summary, here’s how Jill and you benefit from your ambiguous message:

How Jill benefits from your ambiguous messages:

  • Reduced suffering from insults
  • Given opportunity to choose an interpretation they’re more comfortable with
  • Less obligated to meet expectations they don’t want to meet
  • Less obligated to make decisions in the moment
  • Can use plausible deniability for their benefit

How you benefit from sending ambiguous messages:

  • May be given the benefit of the doubt
  • Opportunity to deny damaging interpretations
  • Opportunity to choose a new interpretation without contradicting your past self
  • Less likely to say something hurtful
  • Less likely to appear to be overreacting

Expressing your thoughts and feelings to others can be risky. You need to develop a certain level of trust before candidly divulging your thoughts and feelings. And you need to understand them well enough to know what hurtful things you should avoid saying. Without trust and familiarity you’re more likely to accidentally say something that puts you and your conversation partner at risk of harsh psychological pain. Fortunately, you can reduce your risk of committing these blunders by making your messages more ambiguous.

More examples of ambiguous communication in practice

When reading these examples, try imagining what the outcomes of saying the unambiguous and ambiguous messages might be. Then decide which outcome you think is better.

Your new friend from school scheduled your next lunch hangout 3 weeks from today. You want tell her that scheduling a casual hangout so far in advance is suspicious/odd behavior. You don’t want to make her uncomfortable by making her feel obligated to explain something she may not be comfortable talking about.

Friend: “How about June 27th?”
You: “Wow, I’ve never had to whip out my calendar for a lunch appointment”

You’re trying to tell your co-worker that he’s missing out on building relationships with other co-workers by never joining company happy hours. You don’t want to make him feel pressured to attend.

Co-worker: “I don’t think I’ll go this time”
You: “You should at least make an appearance, people are forgetting you work here!”

On their first day on the job, your employee took a lunch break that was 15 minutes too long. You want to tell him that his tardiness isn’t acceptable. At the same time, you don’t want them feeling like they’re in big trouble on their first day.

Employee: [Comes inside the office]
You: “[Look at watch] Let’s get a smaller lunch next time, yeah?”

When not to use ambiguous communication

You shouldn’t have to use ambiguous communication with people who you know and trust. Direct and precise communication is more efficient and exchanging information and grants you more express power. Otherwise you’re at risk of practicing the kind of counter-productive passive communication people frown upon.

How to understand people better

I’ve been taking notes on how I empathize, considering I seem to be more successful at it than others. I broke down my thought-patterns, implied beliefs, and techniques, hoping to unveil the mechanism behind the magic. I shared my findings with a few friends and noticed something interesting: They were becoming noticeably better empathizers.

I realized the route to improving one’s ability to understand what people feel and think is not a foreign one. Empathy is a skill; with some guidance and lots of practice, anyone can make drastic improvements.

I want to impart the more fruitful methods/mind-sets and exercises I’ve collected over time.

Working definitions:
Projection: The belief that others feel and think the same as you would under the same circumstances.
Model: Belief or “map” that predicts and explains people’s behavior.

Stop identifying as a non-empathizer

This is the first step towards empathizing better—or developing any skill for that matter. Negative self-fulfilling prophecies are very real and very avoidable. Brains are plastic; there’s no reason to believe an optimal path-to-improvement doesn’t exist for you.

Not understanding people’s behavior is your confusion, not theirs.

When we learn our housemate spent 9 hours cleaning the house, we should blame our flawed map for being confused by his or her behavior. Maybe they’re deathly afraid of cockroaches and found a few that morning, maybe they’re passive aggressively telling you to clean more, or maybe they just procrastinate by cleaning. Our model of the housemate has yet to account for these tendencies.

People tend to explain such confusing behavior with stupidity, creepiness, neurosis or any other traits we associate with the mentally ill. With Occam’s Razor in perspective, these careless judgers are statistically the mentally ill ones. Their model being flawed is much more probable than their housemate going insane.

Similar to the fundamental attribution error, this type of mistake is committed more often with people we dislike. A good challenge is to try understanding confusing behavior from individuals or sub-cultures you dislike. You’ll find yourself disliking them a bit less if you’re doing it right.

Another challenge is to try and find the appeal in popular attractions/entertainment you dislike. For instance, if you dislike music videos, try watching a few until you get the “Aha” moment. Yes, that’s what it should feel like when you get it right.
As you’re able to explain more behaviors, your model of people becomes more robust, making you an overall better empathizer.

Projection works, but not for resolving confusion

People’s intuition for how someone’s feeling is normally accurate—with more ambiguous cases—intuition needs conscious support. Unfortunately, most rely too heavily on the “put yourself in their shoes” mantra. You are not always like most people and can react very differently in the same circumstances. There’s already an inclination to project and putting yourself in their shoes rarely overturns initial judgments. If you’re confused about someone’s behavior, it most likely means projection hasn’t worked so far.

Instead, build accurate models of people and figure out whether your model would’ve predicted such behavior. If not, gather reliable evidence proving what the person actually felt and tweak your model accordingly. Hopefully this is starting to sound a lot like the scientific method.

Understand yourself better

As mentioned above, projection normally works well (which is probably why humans are so inclined to do it). Projection, however, isn’t useful if you can’t predict your own reactions in another’s situation.

Catch yourself next time you experience an emotional reaction and try figuring out what network of beliefs caused it. As a personal anecdote, I tried to uncover the beliefs causing me to procrastinate on my work. I narrowed down the portions of work I had an emotional reaction to and discovered I believed I either didn’t have the skill or knowledge to complete the task. Now, when I try to explain other’s procrastination, I ask what part of the work they are having willpower issues with and determine their self-efficacy for those tasks. I was surprised to learn that others had the same beliefs causing their procrastination. Understanding yourself well can lend more non-trivial competing hypotheses.

Caveat: If you’re very different from most people, then understanding yourself better won’t be as helpful. In this case, I’d suggest finding someone more typical to be your proxy. Get to know them well enough to the point where your proxy model can explain/predict behaviors in other typical people.

Put others in YOUR shoes, that’s how they’re empathizing with you

We often find our empathy skills lacking when trying to explain others’ reactions to our own behaviors. We normally consider how we’d perceive our own behaviors coming from another person before acting—making questions like “Why did he think I didn’t want to see him last night?” or “Why was she so offended by my jokes?” hard to figure out off projection alone.
Use the fact that most people project to your advantage. If someone’s trying to empathize with you, they’ll most likely project i.e. put themselves in your shoes.

Imagine a man and woman on a date at a fancy restaurant and just about finished eating their meals. The waiter drops off the bill and the woman glances at the bill. She says enthusiastically, “Wow great food and for a great price too!” The man pays for the bill and moments later his mood shifts, becoming noticeably sadder and quieter. The woman knew he’s more passive than her, but still confused by his behavior.

As it turns out, the man imagined himself describing food as having a “great price” and realized he’d say that about cheap food. The man brought her to the fancy restaurant hoping to impress her, but felt his attempt failed. The woman didn’t think the food was cheap, she thought it was reasonably priced given how good it tasted and the restaurant’s upscale reputation. If she thought the food was cheap, she’d explicitly say so. Since she knows he’s more passive, she could’ve inferred the man believes others are more or less as passive as he is. Thinking back to the incident, she should’ve considered how people would interpret her statement as if she had a reputation for being passive.

One lesson I’ve learned from this technique is that considerate people are more sensitive to inconsiderate behavior. Because they closely monitor their own behaviors, they tend to assume others are about as equally conscientious. When they determine someone’s behavior to be inconsiderate, they are more likely to interpret the behavior as a sign of dislike or apathy rather than obliviousness.

Knowing others are projecting can help you learn more about yourself too. For instance, if you’re confused as to why your friends always ask “Is everything’s ok?” when you feel fine, consider that your friends may be observing certain behaviors they themselves would exhibit when uncomfortable. And maybe you are, in fact, uncomfortable, but aren’t consciously aware of it.

The simplest explanation is usually correct

As you develop your mental model of people, you’ll notice models share a lot in common. For instance, primitive motives like attraction, attention and status can explain the same behaviors exhibited in many people. These “universal” components to your models often yield more likely hypotheses. People are obviously more typical than they are not.

Try to pick out which behaviors are consistently explained by the same mechanism in your models. For instance, it’s helpful to know that most submissive/dominant behavior is done out of status disparities, not some idiosyncratic personality trait. Your knowledge of how people interact with status disparities will offer a powerful starting hypothesis.

As you continue to merge your models together, you’ll be that much closer to a unifying theory of people!

Build models of people, like a scientist

Start developing models of individuals and groups, which predict their behaviors under certain circumstances. Like a scientist, when the model proves to have low predictive value, tweak them until they do. Combining your models is a good approach.
Say you’re having trouble understanding why your brother does everything his new “friend” tells him to do. He’s never acted like that towards anyone before; your model of your brother is missing something. Fortunately, you’ve seen such behavior before, explained by a different model, the one of your co-worker. That model made you realize that, like your co-worker, your brother finds his new friend much higher status and feels lucky receiving his attention. Not only did you strengthen your brother model, you’ve also collected more evidence that such behavior is more likely status-related and less likely person-specific, making all your models more robust.

Experience more

If I tried imagining what a professional soccer player feels like scoring a winning goal, I’d use my memory of the time I scored the winning goal at a pick-up soccer game and multiply my euphoria by some factor. Imagining what emotions someone would feel under circumstances you’ve never experienced isn’t easy. Your best approximation may depend on a similar circumstance you have experienced. Therefore, experiencing more means being a better empathizer.

Empathy checklist

Here’s a short checklist of the different techniques to use whenever you’re confronted with confusing behavior. Run through the list until you feel confident about your conclusion.
Put yourself in their shoes

  • Think of times you’ve been in a similar situation and explain your reaction
  • Can the behavior be explained by a more “universal” model than a person-specific one?
  • How are they empathizing with you, given they are projecting?
  • How are they empathizing with you, given what you know about how they perceive others?
  • What successful model have you used to explain similar behavior for similar people?
  • Is your conclusion affected by your attitude towards the subject?

Biases of Intuitive and Logical Thinkers

Any intuition-dominant thinker who’s struggled with math problems or logic-dominant thinker who’s struggled with small-talk knows how difficult and hopeless the experience feels like. For a long time I was an intuition thinker, then I developed a logical thinking style and soon it ended up dominating — granting me the luxury of experiencing both kinds of struggles. I eventually learned to apply the thinking style better optimized for the problem I was facing. Looking back, I realized why I kept sticking to one extreme.

I hypothesize that one-sided thinkers develop biases and tendencies that prevent them from improving their weaker mode of thinking. These biases cause a positive feedback loop that further skews thinking styles in the same direction.

The reasons why one style might be overdeveloped and the other underdeveloped vary greatly. Genes have a strong influence, but environment also plays a large part. A teacher may have inspired you to love learning science at a young age, causing you to foster a thinking style better for learning science. Or maybe you grew up very physically attractive and found socializing with your peers a lot more rewarding than studying after school, causing you to foster a thinking style better for navigating social situations. Environment can be changed to help develop certain thinking styles, but it should be supplementary to exposing and understanding the biases you already have. Entering an environment that penalizes your thinking style can be uncomfortable, stressful and frustrating without being prepared. (Such a painful experience is part of why these biases cause a positive feedback loop, by making us avoid environments that require the opposite thinking style.)

Despite genetic predisposition and environmental circumstances, there’s room for improvement and exposing these biases and learning to account for them is a great first step.

Below is a list of a few biases that worsen our ability to solve a certain class of problems and keep us from improving our underdeveloped thinking style.

Intuition-dominant Biases

Overlooking crucial details

Details matter in order to understand technical concepts. Overlooking a word or sentence structure can cause complete misunderstanding — a common blunder for intuition thinkers.

Intuition is really good at making fairly accurate predictions without complete information, enabling us to navigate the world without having a deep understanding of it. As a result, intuition trains us to experience the feeling we understand something without examining every detail. In most situations, paying close attention to detail is unnecessary and sometimes dangerous. When learning a technical concept, every detail matters and the premature feeling of understanding stops us from examining them.

This bias is one that’s more likely to go away once you realize it’s there. You often don’t know what details you’re missing after you’ve missed them, so merely remembering that you tend to miss important details should prompt you to take closer examinations in the future.

Expecting solutions to sound a certain way

The Internship has a great example of this bias (and a few others) in action. The movie is about two middle-aged unemployed salesmen (intuition thinkers) trying to land an internship with Google. Part of Google’s selection process has the two men participate in several technical challenges. One challenge required the men and their team to find a software bug. In a flash of insight, Vince Vaughn’s character, Billy, shouts “Maybe the answer is in the question! Maybe it has something to do with the word bug. A fly!” After enthusiastically making several more word associations, he turns to his team and insists they take him seriously.

Why is it believable to the audience that Billy can be so confident about his answer?

Billy’s intuition made an association between the challenge question and riddle-like questions he’s heard in the past. When Billy used his intuition to find a solution, his confidence in a riddle-like answer grew. Intuition recklessly uses irrelevant associations as reasons for narrowing down the space of possible solutions to technical problems. When associations pop in your mind, it’s a good idea to legitimize those associations with supporting reasons.

Not recognizing precise language

Intuition thinkers are multi-channel learners — all senses, thoughts and emotions are used to construct a complex database of clustered knowledge to predict and understand the world. With robust information-extracting ability, correct grammar/word-usage is, more often than not, unnecessary for meaningful communication.

Communicating technical concepts in a meaningful way requires precise language. Connotation and subtext are stripped away so words and phrases can purely represent meaningful concepts inside a logical framework. Intuition thinkers communicate with imprecise language, gathering meaning from context to compensate. This makes it hard for them to recognize when to turn off their powerful information extractors.

This bias explains part of why so many intuition thinkers dread math “word problems”. Introducing words and phrases rich with meaning and connotation sends their intuition running wild. It’s hard for them to find correspondences between words in the problem and variables in the theorems and formulas they’ve learned.

The noise intuition brings makes it hard to think clearly. It’s hard for intuition thinkers to tell whether their automatic associations should be taken seriously. Without a reliable way to discern, wrong interpretations of words go undetected. For example, without any physics background, an intuition thinker may read the statement “Matter can have both wave and particle properties at once” and believe they completely understand it. Unrelated associations of what matter, wave and particle mean, blindly take precedence over technical definitions.

The slightest uncertainty about what a sentence means should raise a red flag. Going back and finding correspondence between each word and how it fits into a technical framework will eliminate any uncertainty.

Believing their level of understanding is deeper than what it is

Intuition works on an unconscious level, making intuition thinkers unaware of how they know what they know. Not surprisingly, their best tool to learn what it means to understand is intuition. The concept “understanding” is a collection of associations from experience. You may have learned that part of understanding something means being able to answer questions on a test with memorized factoids, or knowing what to say to convince people you understand, or just knowing more facts than your friends. These are not good methods for gaining a deep understanding of technical concepts.

When intuition thinkers optimize for understanding, they’re really optimizing for a fuzzy idea of what they think understanding means. This often leaves them believing they understand a concept when all they’ve done is memorize some disconnected facts. Not knowing what it feels like to have deeper understanding, they become conditioned to always expect some amount of surprise. They can feel max understanding with less confidence than logical thinkers when they feel max understanding. This lower confidence disincentivizes intuition thinkers to invest in learning technical concepts, further keeping their logical thinking style underdeveloped.

One way I overcame this tendency was to constantly ask myself “why” questions, like a curious child bothering their parents. The technique helped me uncover what used to be unknown unknowns that made me feel overconfident in my understanding.

Logic-dominant Biases

Ignoring information they cannot immediately fit into a framework

Logical thinkers have and use intuition — problem is they don’t feed it enough. They tend to ignore valuable intuition-building information if it doesn’t immediately fit into a predictive model they deeply understand. While intuition thinkers don’t filter out enough noise, logical thinkers filter too much.

For example, if a logical thinker doesn’t have a good framework for understanding human behavior, they’re more likely to ignore visual input like body language and fashion, or auditory input like tone of voice and intonation. Human behavior is complicated, there’s no framework to date that can make perfectly accurate predictions about it. Intuition can build powerful models despite working with many confounding variables.

Bayesian probability enables logical thinkers to build predictive models from noisy data without having to use intuition. But even then, the first step of making a Bayesian update is data collection.

Combatting this tendency requires you to pay attention to input you normally ignore. Supplement your broader attentional scope with a researched framework as a guide. Say you want to learn how storytelling works. Start by grabbing resources that teach storytelling and learn the basics. Out in the real-world, pay close attention to sights, sounds, and feelings when someone starts telling a story and try identifying sensory input to the storytelling elements you’ve learned about. Once the basics are subconsciously picked up by habit, your conscious attention will be freed up to make new and more subtle observations.

Ignoring their emotions

Emotional input is difficult to factor, especially because you’re emotional at the time. Logical thinkers are notorious for ignoring this kind of messy data, consequently starving their intuition of emotional data. Being able to “go with your gut feelings” is a major function of intuition that logical thinkers tend to miss out on.

Your gut can predict if you’ll get along long-term with a new SO, or what kind of outfit would give you more confidence in your workplace, or if learning tennis in your free time will make you happier, or whether you prefer eating a cheeseburger over tacos for lunch. Logical thinkers don’t have enough data collected about their emotions to know what triggers them. They tend to get bogged down and mislead with objective, yet trivial details they manage to factor out. A weak understanding of their own emotions also leads to a weaker understanding of other’s emotions. You can become a better empathizer by better understanding yourself.

You could start from scratch and build your own framework, but self-assessment biases will impede productivity. Learning an existing framework is a more realistic solution. You can find resources with some light googling. One way is making sure you’re always consciously aware of the circumstances you’re in when experiencing an emotion.

Making rules too strict

Logical thinkers build frameworks in order to understand things. When adding a new rule to a framework, there’s motivation to make the rule strict. The stricter the rule, the more predictive power, the better the framework. When the domain you’re trying to understand has multivariable chaotic phenomena, strict rules are likely to break. The result is something like the current state of macroeconomics: a bunch of logical thinkers preoccupied by elegant models and theories that can only exist when useless in practice.

Following rules that are too strict can have bad consequences. Imagine John the salesperson is learning how to make better first impressions and has built a rough framework so far. John has a rule that smiling always helps make people feel welcomed the first time they meet him. One day he makes a business trip to Russia to meet with a prospective client. The moment he meet his russian client, he flashes a big smile and continues to smile despite negative reactions. After a few hours of talking, his client reveals she felt he wasn’t trustworthy at first and almost called off the meeting. Turns out that in Russia smiling to strangers is a sign of insincerity. John’s strict rule didn’t account for cultural differences, blindsiding him from updating on his clients reaction, putting him in a risky situation.

The desire to hold onto strict rules can make logical thinkers susceptible to confirmation bias too. If John made an exception to his smiling rule, he’d feel less confident about his knowledge of making first impressions, subsequently making him feel bad. He may also have to amend some other rule that relates to the smiling rule, which would further hurt his framework and his feelings.

When feeling the urge to add on a new rule, take note of circumstances in which the evidence for the rule was found in. Add exceptions that limit the rule’s predictive power to similar circumstances. Another option is to entertain multiple conflicting rules simultaneously, shifting weight from one to the other after gathering more evidence.