How to prepare for a PhD in neuroscience

UPDATED 7/26/2009: Click more (or scroll to the end of the post) to see what the student ended up choosing as a major.

Yesterday, I received this email from a freshman preparing for a future in neuroscience:

Dear Neville

My name is […]. I’m a freshman in Biomedical Engineering at the [university in] Mexico City.

After graduation, I am very interested in pursuing the Brain and Cognitive Sciences graduate program. Given your experience, I would like to ask you for some advice.

Over the last few months, I have been thinking about pursuing a major in Electrical Engineering instead. My goal would be to have more engineering tools for my further studies in Neuroscience. Based on your courses, which focus do you think would be more useful to have as an undergraduate? Are there any courses which you would recommend I take to build a stronger background?

I greatly appreciate any guidance you could provide.

Although there are other places to find advice on preparing for a PhD (for instance, economist N. Greg Mankiw has a few advice posts including this one on preparatory math classes), I figured that my take might be unique enough to share it with others. Neuroscience, like other fields, is becoming ever more interdisciplinary and being “a biologist that studies the brain” is just not enough anymore.

Here’s what I wrote back:

I think you’re on a good path for applying to BCS. I think it’s better to major in EE or BE rather than neuroscience or psychology to prepare for a BCS PhD. It’s easy to pick up the neuroscience in graduate school and harder to develop basic hard science and quantitative skills later on. Between BE and EE, I think you will have to decide. Either one should potentially give you a good background. Think about which one is more exciting to you and which program has better instructors.

Here are key quantitative areas I’d recommend:

  • Linear algebra,
  • probability theory/stats,
  • differential equations,
  • signal processing (Fourier transform and linear systems analysis)

A good command of these topics will serve you well in graduate school and far beyond. These topics are probably more closely aligned with an EE background, but, again, I think BE could be a great major if you make sure to add these kinds of rigorous engineering/applied math courses. Additional helpful quantitative topics would be electricity and magnetism and basic stat mech. (The nervous system is, in part, electrical and neuroscience makes extensive use of diffusion equations.)

More broadly, a neuroscientist is a type of biologist. With the age of genetics and genomics upon us, I think it is great to know some biochemistry, genetics, and organic chemistry (roughly in that order of importance). And, after all of that, if your schedule allows, take some courses that are specific to neuroscience, ethology, or cognitive science. I found it very beneficial to take a medical school neuroanatomy course before graduate school, which was really my only neuroscience course pre-MIT. A first course that emphasizes such raw memorization will get you up to speed with the field and its specialized lingo quite well.

And, since you speak Spanish, I recommend this wonderful book by the most pre-eminent neuroscientist (Santiago Ramon y Cajal)… it’s full of great advice about doing science well: Los tónicos de la voluntad (Reglas y consejos sobre investigación científica) (In English, there is a recent translation: Advice for a Young Investigator.)

Feel free to add your own sound advice in the comments below.

Update: The student from Mexico City wrote me back yesterday (7/25).

I thought of what you told me, and using it along with other opinions (including mine), I decided to change my degree to EE. Thus, it will be easier now to have a deeper knowledge of the courses you recommended me, in which I’m very interested. Fortunately, while I was in BE I took some biology courses: genetics, molecular cell biology, and some anatomy and physiology, including the neuro part.

I’ve started to read the book you recommended me, I’ve found it very interesting and perfect for this time of my life! Thank you! I think it’s funny that, paradoxically, the english edition is the one that is in my university’s library (however, I have no problem with that 🙂

6 thoughts on “How to prepare for a PhD in neuroscience

  1. A good list. I would even go so far as to recommend an advanced probability/statistics course or a course in operations research (that’s what the decision theory course was called when i was in college). I mention this because i am increasingly convinced that stopping with an introductory course in statistics does more harm than good. A little machine learning doesn’t hurt either, but that can probably wait until grad school…

    Like

  2. Jeff, as someone who “re-learns” statistics every time I write a paper, I wholeheartedly agree. Also, machine learning is definitely becoming more and more important for scientific data analysis.

    Like

  3. A few things come to mind:
    1 – freshman year is awfully early to plan an entire career. Continue on your path, but keep an open mind.
    2 – Although quantitative courses are useful, to be a well rounded neuroscientist, you’ll need to look into courses in neuroscience, cell biology, biomedical engineering and psychology. Introductory level courses in each of these fields may give you insight as to which direction to take your career.
    3 – I always recommend taking time off between ugrad and grad school, especially if you’re interested in a PhD, to get a position as a research assistant in the field you’re looking at going into. There is nothing like working 40+ hours a week in a job which is close to what you want to get your PhD in to help you decide if grad school is right for you. Plus, it will give you additional letters of recommendation and an inside line to the research community you want to join.
    3b – before graduating, volunteer/join a lab on campus which is doing work like what you’re interested in. This will also help you decide what you want to do, and give you a chance to interact with fellow undergrads and graduate students to decide if your choice is right for you.

    Like

  4. I am such person who is going to pursue a PhD. program in neuroscience.

    I took some lesson in computational neuroscience before. I thinks that skill you mentioned before is very useful. Indeed, my master major is applied maths, and these courses are that easy for me.

    probability theory/stats, signal processing are widely used in encoding, decoding and learning.

    Linear algebra, differential equations are basic knowledges for understanding the neuronal network, such as Amari-Hopfiled network and so on.

    So I cannot agree with you anymore.

    Like

  5. I’m sorry, but I don’t agree with the stockpiling of quantitative skills and then waltzing into neuroscience like it’s some piece of cake. Actually, I resent that. That’s the last thing I need – a bunch of engineers entering the field thinking they ‘know’ neuroscience.

    Like

Leave a comment