I'm a computational PhD student in the field -- I think you'd most enjoy "On Intelligence" by Jeff Hawkins, an inspiring book written by a hacker himself!
If you'd like a very excellent, fascinating title on mathematics and art and mind, try "Godel, Escher, Bach" by Hofstadter. Another fun, classical book with more of an AI bend is "Society of Mind" by Marvin Minsky.
For more classical neuroscience, "Phantoms of the Brain" by Ramachandran, well-known in perception research, is good. I haven't read it, but "In Search of Memory" is supposed to be excellent by Kandel, a memory researcher superstar. (I've found that many of the books by non-scientists about the brain are often shallow, claim to know too much, and use the same familiar stories over and over.) For neuroscience at a lower level (i.e. the biology of neurons, neural networks, cells), I'm not aware of a good popular account.
I also recommend picking up a copy of Mind magazine from your bookstore! And if you can get a copy of a Scientific American magazine special on the brain (every few years I believe), those are usually very well written as well :-)
I second that. "On Intelligence" is a great book, and although I like physics and CS more than biology, it has seriously caused me to think of pursuing a career in neuroscience, simply because it shows how little we truly know about ourselves.
If you're looking for something a bit more digestable, you could try "Biological Psychology: An Introduction to Behavioral, Cognitive, and Clinical Neuroscience", which is more appropriate for a survey level (100, 1000, depending on your school) course.
That's just a reference volume. It's fine for what it is, but really as a launching off point for specific topics. And reading it cover to cover would get old, fast.
I'd second principles of neural science, though that's with the expectation that the reader has the appropriate background in basic chemistry, biology, and maybe anatomy. Out of the dozens of books I have on neurosci, that's probably the single most comprehensive.
I worked at a neuroscience lab for a bit. I was focusing on EEG, the signals, oscillations, and emergent patterns in the brain. For this, they highly recommended I read Buzsaki's Rhythms of the Brain: http://www.amazon.com/Rhythms-Brain-Gyorgy-Buzsaki/dp/019530... .
Could you be more specific? Are you looking for a populist read, a textbook, or an anthology of review articles? Straight up neuroscience is usually lower-level or a more cognitive/psychology high-level slant too? Research oriented or general principles or abstract ideas?
Your purposes might also help. There's been good work done on Bayes and datamining and I think some has made it into a book format.
I need a textbook - I'm looking for something that would be given to first year Neuroscience students. I would prefer something low-level, but it would be nice if the book ties high-level concepts to the low-level concepts.
I've wanted to do a Ph.D. for some time, but neither pure CS nor pure Maths (I have a masters in the latter) excites me anymore. Neuroscience interests me, but I want to make sure that I'm not deluding myself; therefore I'd like to work through a proper introductory text before making the next move.
EDIT: Listened to my inner grammar Nazi and fixed capitalization & sentences.
I used the Gazzaniga when I taught Intro to Cognitive Neuroscience even as I wasn't super impressed (which is a common complaint - tolerance, not true acceptance). I know some folks like the Banich book instead.
If, with your Math background, you're interested in neural computation and modeling there are also some good books along those lines. Just let me know and I'll hunt down the links.
Happy to help. Focus on ions and channels if you're interested in neuron functions. That's a typical hangup for those new to biology. Some knowledge of the vascular system also helps. Otherwise, it mostly comes down to classifying cell types then brain and body regions.
I'll also note that if you apply to a "Neuroscience" grad program, you're probably going to be doing biology experiments all day, or psychology experiments if you pursue cognitive neuroscience.
If you want to use CS and Math, the best you can do in a neuroscience program is data analysis. Hence, I would recommend AI or Mathematical Psychology (though the latter is about modeling high-level events).
I agree on the first three and disagree on the last two points.
I know neuroscientists that do few biology experiments, if at all. There are other ways to get data if you don't collect it yourself. I also know folks in Psychology programs that don't do psychology experiments, based on the same principle. For instance, different types of modeling allow for contributions that coincide with data but as theoretical and analytic tools.
I wouldn't recommend Mathematical Psychology (too much theory, not enough application). I don't know enough about AI programs to comment.
I should have clarified further ... by and large, if you apply to a program in "Neuroscience," you are applying to a program in "Neurobiology." For example, MIT has a huge Brain and Cognitive Sciences option, with a beautiful new building dedicated to Neuroscience. However, a few years ago I was told they only accepted one new student who applied only for computational work that year, since they have very few professors who would support computational work alone.
The fact is that there are very few professors working on "Computational Neuroscience," which is basically modeling experimental results. (I consider this data analysis because most of your time is spent analyzing experimental data. But, its intention is to not only explain given data but to predict the results of future experiments.)
Hence, if you apply to a neuroscience graduate program, there will usually only be one or two faculty members who do computational work. (There are some exceptions -- e.g. UCSD, Caltech.) Many programs won't even take you if you're only interested in computational work because if you somehow don't get along with the professor or it doesn't work out, you won't have a backup lab.
AI programs are divorced from experiments, but they get at the "bigger issues" and are satisfying for CS people -- whereas computational neuroscience models very complex small systems. Unfortunately, computational neuroscience also seems to be struggling -- many biologists discredit it, saying that its only real breakthrough has been Hodgkin and Huxley's model of the action potential!
You might want to look slightly outside the Neuroscience department. Math and CS departments also do quite a bit of computational neuroscience.
Here at NYU, we've got at least 5 computational neuroscience people within the courant institute (math+CS depts), and there are several more in the actual neuroscience department. Plus we have a computational biology Ph.D. program within cims.
I know Rutgers has the BioMAPS computational biology program, and I think princeton has something similar.
I'm at MIT. I don't think that's true. In my lab alone (neuroimaging), there are two students focusing on computational issues. True it's of data, but that's the challenge to any scientist - explaining the data in front of you and coming up with ways to collect better data. Modeling is just one form of that. Computationalists, therefore, have much to add.
I suppose that our main disagreements revolve about the ambiguous definition of "neuroscience," which is nowadays an all-encompassing term which has seemed to broaden to mean "absolutely any work which even mildly studies any aspect of the nervous system or behavior." In this respect, there are many opportunities for computation people.
In neurobiology (the original "neuroscience"), the opportunities may be more limited. I was moreso discussing this latter definition in my posts, mainly because this is what many neurobiology-centric academic departments are called.
As you can see in my reply, I wish there was a simple answer. But I'm afraid the field is so young there really isn't a great general purpose read. The more popular ones leave me wanting. The more research oriented ones are a slog. The problem is, I think, that the field is so young relative to something like physics so the general themes are hard to discern for everyone involved - generalists and specialists - to relay out to a wide audience.
If you'd like a very excellent, fascinating title on mathematics and art and mind, try "Godel, Escher, Bach" by Hofstadter. Another fun, classical book with more of an AI bend is "Society of Mind" by Marvin Minsky.
For more classical neuroscience, "Phantoms of the Brain" by Ramachandran, well-known in perception research, is good. I haven't read it, but "In Search of Memory" is supposed to be excellent by Kandel, a memory researcher superstar. (I've found that many of the books by non-scientists about the brain are often shallow, claim to know too much, and use the same familiar stories over and over.) For neuroscience at a lower level (i.e. the biology of neurons, neural networks, cells), I'm not aware of a good popular account.
I also recommend picking up a copy of Mind magazine from your bookstore! And if you can get a copy of a Scientific American magazine special on the brain (every few years I believe), those are usually very well written as well :-)