I have been playing with the idea of doing a PhD for a looong time. This note centralizes my thoughts on the topic.

Another way to think about this question: would applying What makes for a high quality education? to myself look “formal PhD program”-shaped?

Lay of the land

Current situation

The answer to this question is relative to my current situation, which is fairly unusual and could be described as “self-funded independent researcher.” Some specifics worth calling out:

I have full control over my time, outside of parenting/family responsibilities
- I currently work ~4-6h for 6 days, so ~30h/week
Self-motivated to work hard on things that interest me
Live in Pasadena CA (near Caltech, my alma mater)
No financial pressures for the foreseeable future
- This may change once kids are older (in 5+ years)
Studied applied math/CS in undergrad, strong industry software experience
Pro open science & publishing in non-traditional venues
No interest in working in academia long-term
Unclear if I would get past the PhD application gate now given my distance from academia & the current funding climate¹

Topic of study

My chosen topic of study is intelligence. My two primary quest(ions):

How Do We Learn?
- Current formulation: what are the learning rules that give rise to the incredible plasticity of the brain, while still reliably producing important behavior?
How Might We Learn?
- Current formulation: how can we advance & improve education as a cultural technology?

My approach to (1) so far has been through more theoretical/computational lenses. I don’t think there is a selective benefit to being in a PhD program for this kind of approach (the story would be very different if I wanted to do eg experimental neuroscience). Though there is a practical constraint: a novel analytical framework for the brain will likely need bespoke experiments to test, not just reanalysis of existing datasets. This is a strong argument for collaboration with an experimentalist, if not a formal program. Can’t sponsor data collection alone.

My approach to (2) so far has mostly been reading, thinking, and discussing with others. I don’t think I would pursue a PhD in education in any scenario. My plan is to start a microschool for my/friends’ kids in the future, so I am not as actively focused on this question at the moment.

Parts of a PhD program I am a fan of

I like the idea of:

Having an advisor/mentor
Likeminded peers to cook with, both at your school and at conferences

This boils down to having a high quality social environment & structures for getting feedback. It feels difficult to replicate this well in an independent research setting (eg Andy Matuschak mentions this often), and would likely be the main reason I would pursue a PhD.

Parts of a PhD program I don’t like as much

I don’t like the idea that:

Grant funding may impact what kinds of science are pursued
Active training on strategic thinking around what questions to pursue may be under-emphasized
Extreme variability in all aspects based on advisor
Any and all bureaucratic requirements

This boil downs to external pressures to not be able to pursue the kind of questions I want & too much structure/support from eg a hands-on advisor potentially leading to not developing independent thinking skills as much.

On the former, this seems like a real possibility but probably is strongly dominated by your choice of advisor.

On the latter, I am pointing at the notion that New ideas require thinking through things yourself before learning what others think. I also wonder if there is a homogenizing effect on the problems PhD students choose to work on? This is too speculative/premature to really take into account much though.

Conclusion

Continue doing independent research for now.

Prioritize reaching out to PhD students/postdocs/profs more as I work through my own questions. See if natural collaborations emerge from these discussions & if I want to work more closely with a specific lab. Evaluate how I am feeling about my social environment & feedback structures in 6 months or so, and revisit the broader question at that time.

Continuing to do research myself will also give me a better sense of how I feel about doing research & a better chance of getting accepted to a program later if I decide that’s what I want.

Separately, I will write out a curriculum & progression for what my ideal personal PhD program would look like. Even though the next N years will certainly not go this way & I would probably consider it a failure if things actually followed a static plan, this is still a very worthwhile exercise – Planning is priceless, plans are worthless.

Influences on my thinking here:

How can I allow my thinking to accrete over time?
Free the PhD
Reimagining the PhD
Flounder mode
Cultivating a state of mind where new ideas are born
How the book A Brief History Of Intelligence was written – following curiosity, emailing academics
Ongoing conversations with friends