(Warning: Started out as discussion of intelligence, turned into a screed about physics and impostor syndrome, and is generally more grumpy than it needs to be. Take with a grain of salt.)

The typical model of intelligence looks something like this:

[[Resources]] -> [[Intelligence]] -> [[Result]]

There's a quantity we call "intelligence" that takes external resources and turns them into results we care about. My opinion is that this is a reasonable but impoverished model. It's the simplest way of decomposing the world into internal and external capabilities. Each gets reduced to a single summary statistic- for resources, you typically use money as the measure; for intelligence, IQ. Output is then some monotonically increasing function of these two variables. This is reasonable in the case that they are the only variables you know about.

IQ is the simplest possible quantity you can use for predicting output, but it is not sufficiently detailed to say anything about causality. Like many other predictive variables, it's only useful insofar as you don't have more specific information about the domain. If, for instance, you want to hire someone who can fix your company's problems but you have no idea what those problems are, selecting based on IQ makes sense. But if you have domain-specific knowledge about your problems, you can to pick a person whose specific strategies will help. (See this discussion of why anecdotal reasoning can be better than statistical reasoning.) The problem with "intelligence" is not that it's meaningless, but rather that it's a single summary statistic of an infinitely huge space.

We should view intelligence as a collection of strategies that either have or do not have adaptive fit to particular situations. This doesn't necessarily jive with standard measurements of intelligence. For instance, intelligence is in large part measured by capacity for articulating language well and for being able to think on an abstract level. But there are many tasks for which thinking gets in the way. Modelling social interactions explicitly can take the value out of them. Being trained in jumping to meta-level discussion is especially pernicious. If you can't stay on an object level, you can't do anything. Abstractions are only useful insofar as they cash out in concrete terms- otherwise you become addicted to thoughts about thoughts because you've spent your whole life being rewarded (by yourself and others) for having cool insights. But this addiction to "insight porn" can be maladaptive in cases where reality is inherently messy. (That is to say, almost all the time.) Focusing on language skills as a signal of intelligence creates a sinkhole of thought where the energy of potential creation goes to die. Instead of learning something and then going on to apply that to a problem, you theorize and write essays about it because that's what all of your favorite thinkers do.
As someone who was a physics major at MIT, I'm often told that physics is supposed to teach you a way of thinking about the world that will enable you to solve any problem. But does it? Most problems that one works on as a physicist (at least in general undergraduate classes) are ones that are analytically tractable- that have some universal solution. As we know though, competition is inherently intractable. Complicated coevolving nonlinear systems are much more ubiquitous than ones with simple, tractable causal structure. Studying physics encourages a mindset of simplified models, but attempts at simplification of complex systems either fail, or succeed at the cost of other values.

The physics mindset of extracting the most important abstraction out of a complex system is only useful at the beginning of a cycle of competition. Once things get moving, the specific details in the problem take over.

You see this all the time when people in the "hard" sciences talk about the "soft" sciences. Physicists complain that biologists or psychologists lack simple, central organizing models that explain the core phenomena in a way that you can derive the dynamics from first principles. But they forget that physics is restricted to the simplest possible physical systems. Say what you will about the difficulty of learning quantum mechanics, but modelling a single electron in the ground state of a hydrogen atom is way simpler than modelling even a single cell in the human body.

Perhaps even more importantly, it encourages a mindset of thinking as opposed to doing. Physicists tend to think more like philosophers than engineers, because most of what they do is mental refactoring, as opposed to a constant loop of action and feedback with their environment. Programmers constantly get feedback from a compiler, but the loop on theoretical physics tends to be quite long (in cosmology there are years between updated measurements of the CMB to test theories of the early universe).

Physicists can also fall into the trap of taking ideas too seriously. My favorite characterization of this problem is treating reasoning as a memetic immune disorder. People most of the time default to socially learned behaviors that can work in everyone's favor even if they seem illogical when you analyze them. But because you are encouraged to trust the results of your analysis, you're more likely to do things that are dumb in reality.

It doesn't help that most reasoning in physics is based on eternal laws and invariant properties. I think that this tendency toward inviolable, universal world models leaves physicists particularly prone to impostor syndrome.


People resist the idea that skills are useful only relative to particular environments because there's a slippery slope towards saying that everyone is a special snowflake that has their own unique abilities. "Everyone is a winner" rightfully leaves a bad taste in your mouth. This perspective is damaging because it places value on the presupposed uniqueness of an individual. There are about 7 billion people in the world. There are not 7 billion unique niches for economically or socially useful activities. As such, there will always be other people competing for more or less the same role, and all but one will not be the best. And you can't ignore the people who are better than you either. Because of the way social networks form, your friends are on average better than you in every way
The idea that your value derives from some unique property of your self is what leads to impostor syndrome (at least in my experience). If I'm supposedly so unique, why can I observe evidence of my non-uniqueness all around me? For every component of my self that I value, I can identify both people that are better than me at it and times when I have sabotaged the development of that component. This uniqueness assumption is reinforced in the rhetoric used in combating impostor syndrome. Everyone supposes that you actually have a unique self hiding under there and that all this negativity is just some temporary mental block that's getting in the way of you seeing that*. 
But this "block" is actually a result of taking this model seriously and noticing how it doesn't seem to correspond to reality. Every time you fail to measure up to your supposed value, every time you run into someone who is better at your special skill than you are, a crack appears in this facade of uniqueness. The reality is that you don't have an inherent value. Your value is a function of "you" and "the environment," neither of which are fixed. 

(Forgive me: this is an easy way to sound deep without providing a more useful framework for thinking about problems of value.)

People who think they have inherent value are just ones that have good fit with the environment that they work in. The assumption of inherent value is psychologically protecting and very nice. You don't need to waste time wondering whether or not you should be doing what you're doing. When things are going as expected, you can just treat value as an invariant that holds for the system. And then things change. And then you try and hold on to this invariant, but it never existed in the first place. It was just a summary statistic.

You're not an impostor, but you're not real either. The very notion of being an impostor implies an inherent reality from which you deviate. That's the insight that you really need to get over impostor syndrome: seeing through the idea that there is a single true self. Any model of self that makes an attempt at incorporating all of your actual thoughts and actions will inevitably run into counterexamples. Then it's forced between suppressing that data or using it to revise your self-model into something horrible and maladaptive.

People graduating from college talk about being "a real person" or entering "the real world" as if either of these things existed. The behavior patterns that you have to take on to get by are different as a working person in your mid-twenties than a student in your early twenties, but that doesn't make the world of work more real. It has a different set of constraints on adaptive behavior, and thus gives people a different sense of constructed reality. But it is not itself "more real."

The world has regularities, and so do you, but it is impossible to tell in advance whether or not a particular trait of yours is a truly "inherent" property, or whether it's just a motif that happens to be recurring a lot in this movement of your life. 

As T.S. Eliot argues in his essay Tradition and the Individual Talent, we place too much value on the uniqueness of artists:
One of the facts that might come to light in this process is our tendency to insist, when we praise a poet, upon those aspects of his work in which he least resembles anyone else. In these aspects or parts of his work we pretend to find what is individual, what is the peculiar essence of the man. We dwell with satisfaction upon the poet’s difference from his predecessors, especially his immediate predecessors; we endeavour to find something that can be isolated in order to be enjoyed. Whereas if we approach a poet without this prejudice we shall often find that not only the best, but the most individual parts of his work may be those in which the dead poets, his ancestors, assert their immortality most vigorously.
Instead of focusing on being invariant, true selves, we should focus on channeling the truth. I want to write more on this, but it'll have to wait for later since I have already spent too much time on this post. Instead I'll just leave with this quote from Elizabeth Gilbert's great TED talk on genius in art:
I think that allowing somebody, one mere personto believe that he or she is like, the vessel,you know, like the font and the essence and the sourceof all divine, creative, unknowable, eternal mysteryis just a smidge too much responsibility to put on one fragile, human psyche.It's like asking somebody to swallow the sun.

* For more on the social dynamics that result from pinning value to uniqueness, see Venkatesh Rao's laws of status illegibility, as explored through The Office.