I have previously written about “PhD-level intelligence,” a term popularized by AI companies but also in fashion with people who like to think all success in life is due to how intelligent people are. And I have expressed in no uncertain terms that I think this is a nonsensical concept. What sets people with PhDs apart is not their intelligence but other qualities required to successfully complete a PhD program. Here, I want to discuss various abilities and character traits that are as important as or even more important than IQ in the journey to getting a PhD. And there may be more. I’m addressing the ones I consider the most critical.
Executive function
Executive function is the ability to organize tasks and activities to achieve a specified goal. People with a deficit in executive function have difficulty performing the correct tasks in the correct order at the correct time, even if they can perform all of the tasks in principle. As an example, compare the morning routine of a person with and without executive-function deficit. The person with executive-function deficit will have to be reminded about every single task, such as “brush your teeth,” “wash your face,” “put on your pants,” “eat breakfast,” “pack your bag,” “put on your shoes,” and so on. By contrast, the person with good executive function can just be told “we meet downstairs at 8am” and they know what to do when and will be ready on time.
Because a PhD is meant to be an independent body of work, lack of executive function can severely hinder progress. At the same time, supervisors will often compensate for students’ lack of executive function by telling students exactly what to do when. I have seen a fair share of PhD students with some form of executive-function deficit that nevertheless managed to graduate eventually.1 But I have also seen a PhD student get totally derailed and ultimately leave the program due to their lack of executive function. All things equal, students with better executive function will perform better in a PhD program.
Fearlessness/lack of anxiety
If you listen to professors talk about their strongest students, there is one adjective that is mentioned more often than any other. That adjective is “fearless.” It seems all professors know how important it is for success in research to not be held back by excessive fear, worry, or anxiety.
There are two types of anxiety we can talk about. One is an actual anxiety disorder, anxiety so severe that it interferes with regular daily functioning. That is obviously not helpful. But even people without any sort of clinical anxiety disorder will commonly be hindered by fears or worries. Most people are intimidated by projects that are very big, very ambitious, or require skills or knowledge they don’t have. And those projects make the best PhD theses. Yes people in PhD programs are motivated by knowledge acquisition and learning, and yet even they don’t really want to learn things completely outside of their comfort zone. For example, people like me who are experts at math, computing, or data analysis don’t normally want to learn how to do real-world experiments, even if that is what would be required to significantly move a project forward. The normal computational scientist says “to really solve this problem, I would have to learn how to culture bacteria and then do about a thousand hours of tedious wet-lab experiments; I guess I’ll do something else.” The fearless super-PhD says “to really solve this problem, I will have to learn how to culture bacteria and then do about a thousand hours of tedious wet-lab experiments; I guess that’s what I’ll do over the next year; how hard can it be?”
Mental resilience
Stuff is going to go wrong. Repeatedly. Over and over again. And then even when things go right, when you manage to obtain some meaningful research results, you write them up, submit for publication, and the paper gets torn apart by the reviewers. Or you submit a grant proposal and it gets rejected, and you revise and resubmit only to receive another rejection, and another, and another. Being a scientist is a constant exercise in rejection, disappointment, and failure. It can be discouraging for the best of us. To succeed in this type of environment requires exceptional mental resilience. You have to be able to bounce back from the punches, get up again and again, and keep pushing forward. Eventually things will work out—as long as you don’t get too discouraged along the way.
Discipline
One of the biggest differences between undergraduate and graduate education is undergraduate education can be compared to a sequence of short sprints whereas graduate education is more akin to an ultramarathon. During their undergraduate, the highly intelligent students can succeed without ever working particularly hard or having to develop strong discipline. They can just engage in short sprints of high-intensity work, the occasional all-nighter or weekend to cram for an exam, and that’s perfectly sufficient to get good grades. However, this strategy doesn’t work at all in a PhD program.2 PhD projects are just too big to be dispatched in quick bursts of effort. What is typically required is sustained effort over months and possibly years, and this requires the discipline to show up every day and put in a meaningful amount of work, to stick to a routine for an extended period of time. Students who lack this discipline will find it difficult to succeed in a PhD program.3
Motivation and drive
A PhD is a five-year project where you’re mostly on your own. Yes you have a supervisor, but their primary purpose is guidance and advice. And even the best supervisor is unlikely to check in with you more than once or twice a week. So on most days, it’s up to you to make sure you work and make progress. You need to be self-motivated and want to work on your PhD project. If you need somebody else to supervise you and check what you’re doing every step of the way you’re not PhD material.
Curiosity
The fundamental driving force for a PhD student or working scientist should be curiosity. While you can make it through a PhD program with just discipline and the sheer determination to do what needs to be done, chances are it won’t be enjoyable. Instead, if every day you wake up deeply curious about why things work the way they do you will naturally be drawn to doing research, and you will make progress.
Some of the best research comes from deep curiosity, from asking questions nobody else has ever thought about. Most people are not that curious. They don’t even think about the types of obscure topics that can absolutely capture the mind of an engaged scientists. It takes a tremendous amount of curiosity to see all the unsolved puzzles in the world.
Empathy and social skills
To be successful in science, you frequently need to engage with other people, be it for a collaboration, because you need some of their materials, because they review your papers or grants, or simply because they are your intended audience for papers or talks. While this is a wide range of different possible interactions, they all require somewhat similar skills. Most important is the skill of being able to put yourself into the shoes of the other person, having some empathy for their perspective, and treating them with respect and understanding.
Similar to the trait of executive function, people can survive in science without strong social skills. And yet, the highest performers will frequently be quite adept at social interactions. You don’t lead a large collaboration if you can’t interact with or manage people. And similarly, you don’t build a reputation for being a strong writer or excellent invited speaker unless you can engage with your audience and meet them where they are. And finally, many situations in science require the careful navigation of adversarial or political situations, be it responding to critical reviewers, criticizing eminent colleagues in the field, negotiating complicated authorship questions, or competing for scarce resources.
A sense of morality/ethics
I strongly believe you cannot long-term be successful in science if you don’t have a sense of morality and ethics. If you have no compunction about making up data or otherwise committing scientific fraud you may appear to be getting ahead initially but sooner or later things will unravel. Good science requires commitment to good ethics, and once you give up on the ethics you’ll quickly stop producing good science, as incentives other than the objective truth will come to the fore.
Most people feel a visceral discomfort about lying or being deliberately dishonest. They may confuse themselves about their data, or let hope or wishful thinking get the better of them, but they are not deliberately perpetrating fraud. The exception are sociopaths, who feel no such discomfort. They are fortunately relatively rare, 1–3% of the general population. And I don’t think science is a good place for sociopaths. They do better in environments where they can make claims without having to provide evidence, where there isn’t a strong culture of generating life-long track records of performance, and where they can routinely find new marks rather than having to deal with the same handful of experts over and over again. The rare high-profile cases of scientific fraud notwithstanding, most scientists don’t commit fraud and most people who are inclined to commit fraud will not do well in science.
In conclusion
As you can see, there are numerous abilities and character traits critical for obtaining a PhD and unrelated to intelligence. In fact, some of them may be anti-correlated to intelligence.4 On the topic of whether AIs have any of these abilities at the level of a PhD, that’s a discussion for another time. For today, suffice to say I wouldn’t expect an AI to have a sense of morality or ethics. AIs will happily lie to you or even encourage you to commit suicide. They certainly have no compunctions about committing scientific fraud.
Executive-function deficit is common in people on the Autism spectrum, who also are often drawn to PhD programs. So some form of executive-function deficit is not uncommon in PhD students.
Remember: You’re not in a PhD program to get good grades anyways. Grades are not the point.
This is also why often the students that did best in undergrad don’t make great PhD students. The required skillsets are different.
The literature on this topic is somewhat inconclusive. However, there is some evidence that high intelligence is a risk factor for various mental disorders, that intelligence seems to have co-evolved with anxiety, and that intelligence can lead to more rumination and worry.
I am sharing this post with my grad students! This is one reason why we tend to value research experience more than stats (GPA + GRE) in our program’s admissions. Plenty of students can do the work but not all have the non-academic qualities to go the distance.
Mental resilience yes. I wonder if there's an argument to be made that AI is more resilient than the average doctoral student though. It doesn't really care about Reviewer 2....