The failed simulation effect and its implications for the optimization of extracurricular activities

Cal Newport’s book How To Become a Straight-A Student The Unconventional Strategies Real College Students Use to Score High While Studying Less (that I blogged about recently) discusses a concept that Newport calls the failed simulation effect. Newport:

The Failed-Simulation-Effect Hypothesis If you cannot mentally simulate the steps taken by a student to reach an accomplishment, you will experience a feeling of profound impressiveness.

Newport, Cal (2010-07-20). How to Be a High School Superstar: A Revolutionary Plan to Get into College by Standing Out (Without Burning Out) (p. 182). Crown Publishing Group. Kindle Edition.

Newport gives the following example in his book:

Playing in a rock band doesn’t generate the Failed-Simulation Effect. You can easily simulate the steps required for that accomplishment: buy an instrument, take lessons, practice, brood, and so on. There’s no mystery. By contrast, publishing a bestselling book at the age of sixteen defies simulation. “How does a teenager get a book deal?” you ask in wonderment. This failure to simulate generates a sense of awed respect: “He must be something special.”

Newport, Cal (2010-07-20). How to Be a High School Superstar: A Revolutionary Plan to Get into College by Standing Out (Without Burning Out) (pp. 182-183). Crown Publishing Group. Kindle Edition.

On the basis of this insight, Newport’s bottom line for people looking for accomplishments in the high school extracurricular realm is:

Pursue accomplishments that are hard to explain, not hard to do.

My impression is that Newport is broadly correct as far as college admissions advice goes: activities that are hard to simulate seem more impressive, and therefore improve one’s chances at admission (ceteris paribus). But impressing admissions committees isn’t the only goal in life. In this post, I explore the question: how aligned is this advice to the other things that matter, namely, direct personal value (in the form of consumption and human capital) and social value?

Understanding the question

I’m interested in exploring how closely the following are related for a given accomplishment (with the exception of (1) and (2), the rest measure the value created in some respected; see also this page):

  1. Hardness: The amount of skill, effort, or character strength needed.
  2. Impressiveness (primarily in the context of signaling quality to colleges): The degree to which people (particularly college admissions officers) are impressed.
  3. Human capital: The useful knowledge and skills acquired in the context of the accomplishment.
  4. Social value: The benefit to the world from one’s accomplishment. Note that social value could be direct or indirect, mediated through later accomplishments that rely on the human capital and networking gains from the activity.
  5. Consumption: The fun or excitement of the accomplishment.
  6. Networking: Getting connected to people in the course of accomplishing.

Newport’s insight is that hardness and impressiveness aren’t as closely correlated as we might want to believe. But how closely is impressiveness related to the items (3)-(6)? That’s what we want to explore here. But first, a bit about how hardness relates to the remaining items. For brevity, we will not discuss (5) and (6) further in the post. Instead, we will concentrate our energies on how (1) and (2) relate with (3) and (4).

Where do hardness and impressiveness differ?

Hardness and impressiveness aren’t completely uncorrelated. A few moments of introspection should reveal that that’s the case: if impressive things were easy to do, many people would be doing them, and they would cease to be impressive. But Newport’s central insight is that hardness and impressiveness aren’t as correlated as they seem on the surface. There are things that are quite hard to do but don’t seem impressive because they are mainstream and follow a standard path. There are other things that seem more impressive than their actual hardness warrants.

Consider a 2 X 2 matrix

  Not impressive Impressive
Not hard Not hard, not impressive (e.g., watching TV) Impressive but not hard (somewhat innovative activities, the sort that Newport wants to encourage more of)
Hard Hard but not impressive (e.g., reaching ranked third in high school academics, learning a difficult musical instrument) Hard and impressive (e.g., becoming a really really good music player, getting a medal in a national math or sports contest)

Note that the “hard but not impressive” characterization is relative. Being ranked third in high school academics is impressive. But it’s a lot less impressive to elite colleges (the colleges that Newport’s audience wants to get  into) relative to the amount of effort it takes to achieve. Similarly, learning a difficult musical instrument is somewhat impressive, but not as impressive as it is hard.

We restrict attention in this post to hard activities that people seriously consider doing, rather than random hard stuff people may do for dares or bets (like staying up for 100 hours at a stretch).

Newport wants to shift people from the “not impressive” column to the “impressive” column, and notes that there are plenty of activities in the top right quadrant.

What qualitative attributes characterize activities in the top right quadrant, and their very opposite, namely, activities in the bottom left quadrant? Some observations (based on Newport):

  1. Standard versus nonstandard: Activities that a lot of people are already doing don’t seem that impressive, even if they are hard. And some relatively hard activities are a standard part of people’s academic and extracurricular experience. Learning AP BC calculus and writing a rudimentary mobile app may be of roughly equal hardness. But a lot of people are doing the former since it is part of the standard path. Learning how to play the violin may be about as hard as doing research in a marine biology lab. But the former is a relatively standard extracurricular activity that many people do because it’s what they are supposed to do, or because their parents force them to do it.
  2. Outward-facing: Things that seem like they serve larger numbers of people seem impressive. Mastering one’s learning of a subject is less impressive than doing something that reaches out to many people. But this could have more to do with the “Convincing people” point I make below.
  3. Convincing people: Activities that involve changing other people’s minds seem prima facie more impressive. Learning the violin doesn’t require convincing anybody of anything. You just sit down and learn (or take lessons). Getting somebody to publish your book, on the other hand, requires convincing a publisher that your book is worth publishing. And getting people to buy the book requires convincing buyers that the book is worth buying. Creating an online community or successful marketing/lobbying for a nonprofit both have the flavor of convincing people. It makes sense that having convinced people is a convenient indicator of having done something impressive. In a sense, the evaluation of impressiveness has been outsourced to the other people already convinced.
  4. Discrete original projects: I’m not too sure of this, but people do seem to be somewhat biased in favor of discrete, distinctive projects with clearly identified names or a distinctiveness of identity. “I created a popular website with 1000 pages of information about topic X” sounds more impressive than “I wrote 1000 Quora answers about X” even if it’s the case that the latter activity generates more pageviews in the long run.

Hardness, impressiveness, and human capital

Now that we’ve identified some general points of divergence between hardness and impressiveness, we can consider the question: how do hardness and impressiveness differ in terms of the extent to which they correlate with human capital acquisition (i.e., the acquisition of knowledge and skills that have long-term utility)? As before, we restrict attention to hard activities that people seriously consider doing, rather than random hard stuff people may do for dares or bets (like staying up for 100 hours at a stretch). Let’s look at the four potential sources of divergence and compare based on those:

  1. Standard versus nonstandard: The “hard but not impressive” cluster comprises the more standard activities, whereas the “not hard but impressive” cluster comprises more nonstandard activities. So, this consideration boils down the question to: do standard activities produce more human capital than nonstandard activities? My answer is (very guardedly) mildly in favor of standard activities. Although much of school learning is wasteful, the standard subjects still have the benefit of several years of curriculum development that provides a certain bare minimum of quality. Nonstandard stuff exhibits higher variance. I suspect that the typical nonstandard activity is worse for building human capital than the typical standard activity. But I also think there’s more scope for doing really well on the human capital end by picking a really good nonstandard activity. Another consideration in favor of nonstandard is that there’s a large supply of people who can do the standard stuff, so that the marginal value of adding another person with standard skills is high, whereas the nonstandard stuff could involve building rare, specialized skills.
  2. Outward-facing: My guess is that at the high school level, the most high-value activities (from the human capital perspective) tend to involve learning about the world (not limited to what’s in school syllabi) rather than creating products. This isn’t a hard-and-fast rule, but it does point in the direction of impressive activities being less valuable from the human capital perspective than hard activities.
  3. Convincing people: This argues in favor of impressiveness. The skill of convincing people is an important one, and the act of convincing people also requires one to do a better job overall with presentation and background knowledge. This is good preparation for later life, where one needs to often suggest new things and convince people of them.
  4. Discrete original projects: Impressiveness favors discrete original projects. I think this is an argument in favor of impressiveness being better at building human capital, but a very weak one. People acquire valuable skills in the process of creating their own original projects that they wouldn’t when contributing to existing projects (for instance, creating your own website means you have to learn about website creation and getting traffic). On the other hand, participating in existing projects makes it easier to calibrate your learning, get feedback, and improve.

Hardness, impressiveness, and (direct) social value

How do the “impressive but not hard” activities compare with the “hard but not impressive” activities in terms of the
direct value they produce for society? We’ll do a point-by-point comparison similar to that for human capital, but first, a little digression.

Although many hard activities are not valuable, it is almost always the case that valuable activities are at least somewhat hard. The logic is similar to the logic for hard and impressive activities described earlier in the post. Namely, if valuable activities were easy to do, they would already have been done to the extent where they either became hard at the margin or lost value at the margin.

PayPal co-founder Max Levchin credits this insight to co-founder Peter Thiel (see here). Levchin recounts that, back when PayPal was in its infancy, he was enamored by the idea of using elliptic curve cryptography to speed up some aspects of PayPal’s secure transactions. Elliptic curve cryptography uses some pretty cool math and offers interesting implementation challenges. But it turned out that the speedup offered wasn’t really helpful with the things that PayPal needed to do. Levchin learned from Thiel that hardness isn’t the source of value. On the other hand, things that are valuable are almost always bound to be hard, because if they were easy, they’d have already been accomplished. Indeed, Levchin’s new company, named Hard Valuable Fun, builds on this insight.

As before, we restrict attention to hard activities that people seriously consider doing, rather than random hard stuff people may do for dares or bets (like staying up for 100 hours at a stretch). Now, let’s compare hard and impressive activities in terms of their social value:

  1. Standard versus nonstandard: The “hard but not impressive” cluster comprises the more standard activities, whereas the “not hard but impressive” cluster comprises more nonstandard activities. So, this consideration boils down the question to: do standard activities produce more direct social value than nonstandard activities? I think the general answer is a resounding no. Standard activities are largely focused on building human capital or signaling quality (to colleges and others), rather than on the creation of direct social value. This is true even for standard extracurriculars, such as learning musical instruments. Even the standard extracurriculars billed as socially useful, such as volunteer work by US students in Columbia, often produce negligible social value (see Jonah’s post on volunteering for a more in-depth discussion). Note that the indirect social value created through human capital acquisition might still be huge for some activities that build human capital, but that is not what we’re trying to assess in this part of the post. Nonstandard activities exhibit higher variance, but could at least in principle be chosen for higher social value. Another consideration in favor of nonstandard is that there’s a large supply of people who can do the standard stuff, so that the marginal value of getting something nonstandard accomplished may be higher on account of more low-hanging fruit.
  2. Outward-facing: Impressive activities tend to be outward-facing. And creating direct social value generally requires being at least somewhat outward-facing. So, this consideration points in favor of impressiveness over hardness.
  3. Convincing people: This argues in favor of impressiveness. Creating positive change usually requires convincing people at some level. This could be direct suasion, or it could be attracting people to visit one’s website or buy one’s book or use one’s products in some other capacity.
  4. Discrete original projects: Impressiveness favors discrete original projects. I think this is an argument in favor of impressiveness being better at creating direct social value, but a very weak one, and there are many counterexamples. Creating your own website may seem more impressive than just writing a bunch of Quora answers, but the latter may get read a lot more.

Below, I summarize what I’ve said about hardness, impressiveness, human capital, and direct social value:

Consideration Human capital consideration points in favor of hardness or impressiveness? Direct social value consideration points in favor of hardness or impressiveness?
Standard versus nonstandard Hardness (but weak) Impressiveness (but weak)
Outward-facing Hardness (but weak) Impressiveness
Convincing people Impressiveness Impressiveness
Discrete original projects Impressiveness (but very weak)

Impressiveness (but very weak)

Overall, it seems that a shift towards impressiveness would perform better in terms of direct social value and slightly worse in terms of human capital. But the variation between different choices of activities overwhelms the general comparison of hardness and impressiveness. In other words, there are probably a lot of activities within the impressive category (at varying levels of hardness) that perform well on the human capital and direct social value dimensions. One just needs to be know to look for them.

Any thoughts on the above would be appreciated.

PS: I’m planning to do another post (or posts) on how people in high school and early college, or others in a similar age group, can select side projects and execute them well.

Cross-posted at LessWrong and Quora

Leave a Reply