What is "neuroscience"?

At the end of December when everyone was out of the lab on vacation the Journal of Neuroscience twitterers ran an episode of Ask Me Anything, Neuroscience. I had responded to an earlier teaser on this and asked the acting Editor in Chief of the Journal of Neuroscience the question which titles this post, figuring she should know. Obviously, I shaded the question....a little.

She replied:

..which is fascinatingly imprecise. Particularly for an EIC who has to decide categorically what is and is not appropriate material for the Journal she Edits. If we were talking about the range of investigation covered by the presentations at the annual meeting of the Society for Neuroscience, this would be a great answer. The breadth of science at that meeting is tremendous and I can buy that it covers almost everything "to do with neurons". This is not the case for the Journal of Neuroscience. Which should probably be re-named the "Journal of Some Neuroscience but not other Neuroscience".

As you will recall, Dear Reader, I have observed on more than one occasion that as a wee graduate student trainee I realized this fact with some dismay. I was outraged! How can this type of science be okay and this other type of science not, when the only difference is the techniques involved?!??, I wondered. How can these people not see that the Emperor's New Clothes are not better, more precise or more mechanistically insightful results, they are just different levels of analysis?

Over the *cough*cough*decades this attitude has turned to bemusement, particularly as the Journal of Neuroscience's JIF has slid inexorably* down (currently 6.3) into just-barely-above-the-herd levels (25th in the Neuroscience category). Just ahead of such titles as Glia and Brain Behavior and Immunity. It is behind the Journal of Pineal Research, ffs! Yes, yes, JNeuro still punches above its JIF in reputational terms with the cognoscenti but there are many JIF-equivalent-or-better journal options. And after all, we all realize that the JIF still rules where it counts- when people aren't assessing the science from an informed perspective. So the cost to those who do that other type of science involving neurons that is not acceptable for JNeuro has lessened considerably. The gains of sneaking one into the JNeuro have likewise lessened. Better to try at a less technique-limited venue that has a higher JIF

There was followup from the JNeuro twitter intern:

and a related reply from the acting EIC.

Also particularly amusing given the place that "shows mechanism" holds in the mind of the average bio-scientist type, most certainly including neuroscientists, these days. I'd like to see an accounting of how many J Neuro articles in a given year reasonably qualify as "New observation without mechanism". I'm betting the number is so low as to falsify this claim in any reasonable mind.

Then later there was this claim during an unrelated exchange:

Which I think is bizarre buck-passing for an Editor or Associate Editor of a Journal to engage in. At the least, it illustrates how and why it is bogus to claim "New observation without mechanism" is welcome-- if one only selects reviewers who will not buy this for a second then where are we? Also, I am curious if AEs use the presumption of what reviewers might say to desk-reject said manuscripts. See also, the above comments about what qualifies as "neuroscience" and whether or not certain approaches and techniques are ruled in/out at this particular journal. Speaking as a reviewer, I try to follow the Editorial lead in the sense that "appropriate for this journal" has to be recommended, I rely on what they have actually been publishing**.

In closing, I'll point out that I write this for the current version of younger-me. Those of you who aspire some day to publish in J Neuro, because you are a proud neuroscientist and proud member of the Society for Neuroscience. You who bring your posters to the Annual Meeting and then notice, chillingly, that science like yours never seems*** to get published in J Neuro. Have a heart. Leave your Imposter Syndrome behind. There are many so-called "more specialized" (that's meant to be an insult when reviewers or AEs say that, btw) journals which have better JIFs. Get your work published there. Keep coming to the SfN meeting and chatting with the folks who appreciate what you do.

Keep on with the science that satisfies you.

And feel free to snicker about those people who do cell biology accidentally in neurons and call themselves neuroscientists.

*all snark aside, I do lament this. J Neuroscience is a great journal and resisted Glamming it up and JIF chasing in response to the invention of Neuron and Nature Neuroscience. It is unfortunate it is being punished for this. And of course, before the aforementioned baby Glams, it really did shine as a pinnacle for a Society published journal.

**Unless, of course, I am engaging in a rather intentional pushback along the lines of what the JNeuro EIC is suggesting, i.e., putting my marker down that I think the journal in question should be publishing a certain kind of paper.

***Yes, there will be the occasional paper that gets into a given journal. And you will think "aha, we have something very similar so let's submit!". Give it a try for sure. But don't be too amped when you get desk-rejected. Often enough you will find out that the relationships between the editorial staff and the authors is slightly closer than you enjoy. Shrug and move on. Or, if a PI and you DGAF about your reputation at that particular journal, write a pointed inquiry to the AE to see what they say. I had one of these at a journal that rhymes with Serebral Kortex awhile ago. The new editorial staff tried to slam the old editorial staff and basically said, well that would never get in anymore. I was amused. And we published that paper somewhere else and moved on. As one does.

42 responses so far

  • JJ says:

    But JNeuro is great because everybody reads it. Although, for a behavioral scientist like me, it has gotten more and more difficult to pass the editors. Still possible though.

  • JJ says:

    AND I don't care about the impact factor... even if my dean does

  • The Other Dave says:

    Quit obsessing. It's a magazine. Do you agonize over the fact that Rolling Stone publishes articles that are not about the music industry? How does "Blow your man's mind with these simple bedroom tricks!" be described as 'Cosmopolitan'?

    And anyway, I have never ever seen a promotion & tenure letter or grant review talk about JIF. People recognize journals by name. J. Neurosci. has an excellent reputation, as do other journals that are top-notch for their area. I think targeting JIFs might be counterproductive.

  • pinus says:

    JNeuro has not just gotten tougher for behavior people. I more or less gave up on it because I would get such bullshit reviews for papers that would then sail in to journals with higher impact.

  • drugmonkey says:

    Interesting. Perhaps they should heed Leia's advice to Vader? "The tighter you squeeze...."

  • qaz says:

    TOD - Rolling Stone and JNeurosci serve very different purposes. I know that you know better than to equate journalism (whatever the quality) with scientific communication. JNeurosci is the society journal for the Society for Neuroscience and is supposed to be the place to report neuroscience results for the scientific community.

  • drugmonkey says:

    What is neuroscience, qaz?

  • becca says:

    Neuroscience *is* psychology done by people with more slicers and lasers.
    Neuroscience *should be* the biological study of how biological study is possible. /metawaffles

  • DWG says:

    Neuroscience is the science of the brain, central nervous system and its constituent parts, and interactions of those parts along developmental and evolutionary dimensions. These interactions may occur at multiple scales: molecular, cellular, networks, systems, and communities - but all have as a common basis the brain and its behavioral interactions with the environment and/or other brains. Neuroscience overlaps with psychology and cognitive sciences, but seeks mechanistic rather than conceptual representations underlying important behavioral phenomena or brain disorders.

    Or something like that.

  • drugmonkey says:

    "The brain and its interactions with the environment". I like that one.

  • Hasn't "What is neuroscience?" been the big question since Plato?

    In any case, I think that the tweet about mechanism not being necessary is subject to some potential misinterpretation. Neuroscience has to be mechanistic (or at least directed at a mechanism) at a biological level. If you were to draw a phylogenetic tree, that's where ancestral neuroscience would branch off of philosophy of mind.

    Now if J Neuroscience is calling for novel findings where the mechanism is not yet clear, I think that's pretty awesome. That's also a steep hill to climb though. "Insufficient mechanistic insight" is easy boilerplate criticism for any overworked grant or manuscript reviewer.

  • drugmonkey says:

    AB- my issue revolves more around what defines a "mechanism". And the notion that some levels of analysis lead to better "mechanistic insight" than others, particularly when I think they do not.

  • I see what you mean now. That being the case, isn't the statement about reviewers driving things kind of pragmatic? The steady state of any journal is going to hugely influenced by the trends present in the reviewer population, right?

    That's not to excuse any analytical biases that exist at J Neuroscience, because it (more than any other neuro journal) has a mandate to publish a broad swath of research and not be Neuron or Nature Neuroscience. But does any EiC have the editorial capital to free the scope of a big society journal from the influence of whatever is trendy in the field? And even if they could, should they?

    I guess I would answer that the EiC has to recognize and target specific problematic trends and biases. And that's where I think there is a problem with the "novel findings without mechanism" statement. That could be an unfunded mandate if it has to run up against the huge inertia/momentum of a reviewer population.

  • DWG says:

    Maybe "mechanistic" isn't the desired word. Perhaps "quantifiable" is better. For example, in pharmacology there are descriptive measures that are useful in characterizing neurotransmitter uptake. Such metrics may be apt quantitative descriptions of an underlying mechanism, but are indirect and based on the result of that process. Likewise, in brain imaging network analyses identify quantifiable relationships between nodes in a network model, but just as an equation describing uptake doesn't fully characterize the actions of a transporter, neither does a graph analysis in an fMRi experiment describe a mechanism of a cognitive process. Yet both of these would be useful descriptive tools used by neuroscientists.

    I would suggest that the difference between neuroscience and disciplines like psychology is not that there isn't such quantification, but in neuroscience the target of the quantification in neuroscience is the "gears" - in psychology and cognitive sciences the target is almost exclusively at the output of the machine.

  • At the end of December, when we hoped as many people as possible would be available, since experiments had wound down for the holiday and lots of people turn to Twitter, we ran an episode of Ask Me Anything, Neuroscience.

    In response to @drugmonkeyblog’s question about defining “neuroscience” I did suggest the tautology that it is: “science to do with neurons”. I think that any of us who are members of SFN want a broad and inclusive definition of neuroscience, and I didn’t realize that the question was with respect to The Journal of “Neuroscience” and how The Journal defines the word. As a result, I think there were a few things that I did not tweet clearly in my 140 character chunks.

    First, @drugmonkeyblog is correct that JNeuro still focuses on “mechanistic studies” and that, like all editorial identities, this has evolved over time as a result of editorial decisions, but also as a result of how reviewers (admittedly, selected by those editors) chose/choose to rank submitted manuscripts. Identity is not instantiated by a single decision. It is also something that is mutable, but becomes strongly imprinted. The particular definitions of “mechanistic” have changed a bit over the years I have been reading JNeurosci, but that identity is, indeed, one that is perpetuated by those of us who are editors (including me). Like any human endeavor, the standard is not applied perfectly, and is, indeed, “fascinatingly imprecise”, as is any definition of our adolescent/young adult field, but @drugmonkeyblog is correct that JNeuro publishes few observational studies. That’s why when I responded: “both” to the post from the interns asking if we publish:

    "Rogue JNeuroscience @JNeuroscience
    ‪#amajneuroscience‪ Because @drugmonkeyblog wants to know...Mechanistic vs top notch definitive studies on novel effects. Thoughts @MarinaP63‬"

    I was not referring to purely observational studies. I may not have interpreted, “Definitive studies on novel effects,” in the way that was meant. "Definitive studies" do well at all journals, including JNeurosci, but perhaps the problem is differing interpretations of what “definitive studies” means.

    I also said, and I think this was misunderstood, that eNeuro, in particular, is asking for submission of “New observations without mechanism”. It is true that these do not usually pass review at JNeurosci, and are currently discussed often in consultation sessions between Reviewing Editors and a Senior Editor as candidates for editorial rejection. Many papers that have historically not done well at JNeurosci are explicitly encouraged for submission to eNeuro, including purely biophysical studies of neuronal proteins expressed in non-neuronal cells, psychophysical studies without underlying neurobiological mechanisms, and some behavioral studies that do not explore the neuronal underpinnings of the behavior.

    Some of the information on process is in the editorial and an accompanying note from SFN (http://www.jneurosci.org/site/misc/JN_Combined_Editorial_and_Note.pdf) that came out in the 1/6/16 issue of JNeurosci that I began working on when I was appointed interim EiC in late September. This editorial tries to lay out the changes in the review process that have been going on throughout 2015.

    I hope to continue to answer any questions that are out there. This is a work in progress, but I feel strongly that the biggest strength of The Journal of Neuroscience is that it is a society journal, and as a result, I think its editors should be accountable to the members of SFN and communicate as clearly as possible. Also, at this time when all our modes of communication are changing rapidly, we have to choose to do experiments in review to remain relevant to our authors, reviewers and readers. At the same time, it is very important not to destroy what we have, and what we have built as a community of authors, readers, reviewers and editors, as we move into a more uncertain future than usual.

    * In response to your notes about IF, as you noted, I think a new and very important issue will be the cascading streams at Cell Press and Nature that will qualify as “non-specialized” for promotions committees because they don’t have “neuro” in their names. We don’t want to chase glamour, but we do want to be considered as a scientifically rigorous place to publish and find the strongest neuroscience articles (where “strongest” is even more difficult to define than “neuroscience”).

    ** All opinions expressed here are my own, and not those of my employer, SFN or The Journal of Neuroscience editorial board.

  • drugmonkey says:

    it (more than any other neuro journal) has a mandate to publish a broad swath of research

    I don't know if this is true. I don't know much about where the EiC of J Neuro derives her mandate. I agree that this should be the mandate. Part of my reason for prodding.

    But does any EiC have the editorial capital to free the scope of a big society journal from the influence of whatever is trendy in the field? And even if they could, should they?

    They can certainly try to move the ship on a new heading. Should they? I believe that J Neuro should. Why? Basically because it is ridiculous to have a ginormous Society, host an Annual Meeting with a vast breadth of work and then say "Well, we're not going to publish [substantial fraction] of this work in our flagship journal because of levels of analysis reasons that have nothing to do with objective understanding".

    DWG- I'm not talking about non-quantifiable stuff. There are lots of things that can be quantified. Gears versus output? perhaps. But then why back in the day was some crude brain lesion considered TehAwesomeMechanism whereas selective pharmacological intervention (albeit systemically) considered NotNeuroscience? Why is expression of an immediate early gene (c-fos goes up all over the brain anytime you look at the mouse sideways, btw) better than a set of precise behavioral conditions? Nowadays, given the way things are running, the papers combine all sorts of techniques and perhaps the dividing lines are less clear cut but the principle remain.

  • drugmonkey says:

    JNeuro publishes few observational studies.

    This is incorrect. Science is never anything but observation. "This is what happened when we did X". Period, end of story. All else is inference and what the individual thinks they can conclude from the observation.

    the standard is not applied perfectly

    To be very clear, there are two issues bound up in this. First is whether the standard of "mechanistic" can sensibly be applied to Set A of techniques but not to Set B of techniques when they are both interventionist in the way that I think Marina is deploying "observational" in her dismissive comment above. That assumes some consistence about accepting manuscripts that come from Set A versus Set B. The second issue at hand is whether some papers using Set B get in where most do not and why that should be the case- in essence, is there an insider club that can get away with stuff that nobody else can. Two different things about standards being applied perfectly that should be addressed. IMO.

    behavioral studies that do not explore the neuronal underpinnings of the behavior.

    This is why I press. You state this like it is meaningful. Clearly it is to you. But if behavior and much of physiology is inherently a result of the function of neurons how is any of it not exploring the neuronal underpinnings? It obviously is, so therefore you need to do better if you want to take this issue seriously. We, I suspect, are all relying on the Justice Potter Stewart logic. I'd like to see us get closer to consistency.

    the biggest strength of The Journal of Neuroscience is that it is a society journal, and as a result, I think its editors should be accountable to the members of SFN

    I agree. As you wander around the SfN poster floor do you genuinely get the impression that the majority of the work presented there has an equal shot at publication in the Journal? If you review the membership roster of the SfN (dues paying membership roster, I'll note) do you think the main bulk of their primary interests and work has an equal shot at publication in the Journal? Should it? We're so proud of the vast diversity of the SfN when it comes to meeting time...why does that evaporate when it comes to the mission of the flagship Journal? Do you think, in your heart of hearts, that actually people who are interested in work that you find unqualified for the Journal shouldn't be at the Annual Meeting or on the Membership rolls at all?

    it is very important not to destroy what we have

    What does the journal "have"? How does one "destroy" it? From the outside one might assume that what the journal "had" was the fact that all of the best work in Neuroscience that wasn't Science or Nature material was published in J Neuro. Then Neuron and Nature Neuroscience were invented and stole that thunder clean away. The categorical snobbery resulted in the rise of journals such as Cerebral Cortex, Neuropsychopharmacology and Biological Psychiatry up past the JIF of J Neuro. This was potentially aided and abetted by a sense of insider-clubness that drives away even papers with the appropriate subject matter and techniques deployed (see comment upthread from pinus). So what is the essential nature of the Journal of Neuroscience that must not be destroyed? I am not sure I have a good idea of what this even is anymore.

    We don’t want to chase glamour, but we do want to be considered as a scientifically rigorous place to publish and find the strongest neuroscience articles (where “strongest” is even more difficult to define than “neuroscience”).

    You can't have your head in the sand about Glam and JIF. Every bloody EiC I've ever heard on the topic wants their journal to have "strong articles" and "the best work" and blah de blah. From practically the lowliest dump journal up into the high society-journal ranks (i.e., not Glam professional eds). right? How. Do. You. Do. This????

    "encouraging the membership to submit their best papers" is the usual strategy. It is a ridiculous joke. Submission behavior is driven by journal rank which is roughly correlated with JIF as we all know. JIF trends tend to inspire vicious or virtuous cycles depending on which way the JIF is trending. See Addiction Biology, see J Neuro. Opposite directions, opposite behaviors. In my view, J Neuro is at that dangerous cusp where it is basically a dump journal for Glam and Demi-glam folks and an aspiration for the plebes. Which way will you steer the ship? Too much affection for the Glam leavings as your standard and you price away the nonGlam labs. Too much meat-and-potatoes and the demiGlam labs won't even dump there anymore.

    The alternative is you take a run at Neuron and Nature Neuroscience in seriousness. First thing, publish some reviews and get that JIF headed back upward. That'll spark a virtuous cycle of "submitting their best work".

  • DWG says:

    Yes. Well, brain lesions provided insight into mechanism, but like any tool it possesses limitations. Still - even with lesion studies silver age investigators were looking for things like engrams. Even though some of it was a failure in some respects I would suggest lesion studies led to our modern notion of distributed processing and memories, and many lesion studies provided key insights into how the nervous system develops (e.g., Hubel and Wiesel and ocular dominance). So we learned something.

    More to the point of the discussion, though, I do experiments from whole cell patch clamp in brain slices to human brain imaging in my lab. Including pharmacological interventions. I have never understood what I'll call "toolism" (i.e., "...my tool or approach is best", which seems a bit intellectually stunted) since no one has a monopoly on the truth, and each tool has something to contribute.

    Philosophically, I'd like to be able to send my best work to J Neuroscience if it makes an important statement about how the brain works. I've published in it several times (though not recently) and have been rejected several times. I do agree with the basic premise that a society Journal should be receptive to the scope of work done by the membership - but that work should provide insights with impact that is similar to other work in the journal. For example, consider pharmacology that focuses on new drug but without insight into the importance of the receptor to brain function, vs. studies with the same drug but perhaps a new and different function of the receptor system is revealed. The former might be better served by a specialty pharmacology journal, of which there are several good ones - the latter in my view should be reviewed by J Neuro.

    Editors are chosen to exercise and execute a vision, and I do appreciate the comments that it's an evolving process.

  • DM says:

    Behavior is the sine qua non of "brain function".

  • Grumble says:

    "But if behavior and much of physiology is inherently a result of the function of neurons how is any of it not exploring the neuronal underpinnings?"

    Do you seriously not see a difference between a study that does nothing more than measure behavior and one that both measures behavior and either manipulates neurons or measures their activity in relation to behavior?

    "...I suspect, are all relying on the Justice Potter Stewart logic. I'd like to see us get closer to consistency."

    Nonsense. The standard is pretty clear to me, at least with respect to studies that seek to determine the neural basis of behavior (a very large portion of neuroscience). If you muck with or measure neurons, it's a study that examines mechanism. If you don't, and measure only behavior, it's not. By the way, this is not to say that non-mechanistic studies aren't valuable - and as Marina points out, they have a home at eNeuro. Sounds to me like an ideal solution to the problem of what to do with high-quality but non-mechanistic studies that people want to submit to a J Neurosci-like journal.

    "do you genuinely get the impression that the majority of the work presented there has an equal shot at publication in the Journal?"

    Why should it? You can submit the Aardvark abstract and get to present a poster on it. Why should that have an equal shot at J Neurosci?

    "What is the essential nature of the Journal of Neuroscience that must not be destroyed?"

    Easy (at least with regard to systems/behavioral/cognitive papers): It publishes rigorous studies of neural mechanisms for behavioral and cognitive phenomena; these studies tend to be more detailed and convincing with respect to mechanism than those found in lesser journals.

    "Every bloody EiC I've ever heard on the topic wants their journal to have "strong articles" and "the best work" and blah de blah. From practically the lowliest dump journal up into the high society-journal ranks (i.e., not Glam professional eds). right? How. Do. You. Do. This????"

    Marina was, I think, pretty clear about this: accept only papers that convincingly demonstrate neural mechanisms.

    I think it's that big chip on your should regarding mechanism (been burned by that one much?) that prevents you from seeing the niche that J Neurosci occupies, and why that makes it a strong journal that tends to be very well respected.

  • drugmonkey says:

    What does "manipulate neurons" mean?

  • qaz says:

    No, Grumble, I don't see a difference. There is an arrogance in neuroscience that behavioral psychology has just been pissing around for a hundred years. There have been a tremendous number of discoveries about neural function from purely behavioral studies. (How about the number of color channels in the visual system from Helmholtz's behavioral studies?)

    What is neuroscience? I don't know. I like the breadth of SFN. What should be in JNeurosci? I think anything that tells us more about how neural systems work, at whatever level. (Whether it be behavioral studies that limit the processes that can occur or subcellular protein cascades that occur within neurons or glial information processing or whatever.)

  • qaz says:

    I think a better question is What do we mean by mechanism? Especially when a lot of the most interesting things that make us human are emergent properties.

  • Grumble says:

    Emergent properties have mechanisms, too.

    For the purposes of systems neuroscience, "manipulate neurons" means artificially change their activity in a such a way that you understand (or think you understand - but that's a different issue) what the change did to the activity.

    "behavioral psychology has just been pissing around for a hundred years"

    Maybe. I've never said that behavioral psychology isn't valuable. I'd say it can be extremely valuable. And yes, those sorts of studies constrain what the neural mechanism must be. But I also don't see a problem with drawing a line, for the purposes of what's acceptable to a given journal, around what constitutes a "mechanism" study.

  • qaz says:

    So, Grumble, no electrophysiology in J Neuroscience? Given the tremendous gains in knowledge that correlational studies have given us over the last hundred years, and the fact that correlational ephys are still telling us very important things about mechanism, I think we shouldn't get too wrapped up in the causal-neural-manipulation fad.

    I liked it better when we were arguing what defines mechanism rather than trying to limit the definition to manipulation.

    On the question of whether behavior that informs mechanism is neuroscience, we'll just have to agree to disagree. If I were editor (and I am not!), I would be more generous as to what defines as neuroscience. Especially since the mechanisms of emergent properties may well arise from interactions and complex properties of many small factors. If you are going to argue that looking at high level emergent properties is only enough if you've identified all the small factors, then I can argue that looking at small factors without understanding how they interact to reach the high level emergent properties is not enough. Personally, I think we can do more by being inclusive. That's part of what makes SFN so wonderful - it has everything from high level psychological constructs (it's a forest!) to the subcellular properties of single channels (call 'em bark people - forget the forest, they can't even see the trees!).

  • drugmonkey says:

    Do Ramachandran's phantom limb experiments "artificially change activity"? Yes.

  • DJMH says:

    I had never noticed that J Neuro didn't publish some segment of neuroscience. Can someone be more explicit about what J Neuro doesn't touch? From my standpoint there's mol/cell, dev, systems/circuits, clinical...what are they missing?

    I can't think why a pure behavior study would go to a journal for neuroscience. If you take a Skinnerian black-box approach to the brain, why would a journal that publishes brain research be interested?

  • qaz says:

    By that definition, behavioral experiments artificially change activity (especially if they are well controlled) and then you can look at the consequences of that change in activity with a behavioral measure. So we've danced ourselves back around to purely behavioral experiments as being important, especially if they tell us about mechanism! Sounds good to me.

    In fact, Ramachandran's phantom limb experiments tell us a lot about how cortical representations are structured, especially in the light of the rest of the neuroscience literature.

  • The Other Dave says:

    J. Neuroscience is from back when 'neuroscience' didn't really exist so much. It was psychology, physiology, or maybe cell biology. The journal was an important part of making the relatively new field a recognizable entity.

    Does the journal need a new purpose now? What is its purpose?

  • drugmonkey says:

    It'll come to you in a minute DJMH.

  • The Other Dave says:

    To be more explicit: J Neuroscience was created to fill a gap. Behavioral stuff, psychopharmacological stuff, membrane biophysics stuff... there wasn't so much a gap there.

  • Grumble says:

    "So, Grumble, no electrophysiology in J Neuroscience? "

    That's not what I said. Above, I defined a study of mechanism as "one that both measures behavior and either manipulates neurons or measures their activity in relation to behavior."

    "So we've danced ourselves back around to purely behavioral experiments as being important"

    Of course they are important. The question is whether they should be published in J Neurosci.

    "Do Ramachandran's phantom limb experiments "artificially change activity"? Yes."

    Those were behavioral experiments that were designed to answer a specific question about neural mechanism: "Is phantom limb pain due to the activity of cortical or peripheral neurons?" This was answered by the observation that touching the body elicits the pain, which suggests a specific mechanism (sensory information from the periphery activates cortical areas that received input from the phantom limb). Most behavior-only studies are not designed to address a question of specific neural mechanism. I'd say that studies that use behavioral experiments explicitly to determine how a neural circuit or set of circuits produces behavior, sensation, cognition, etc fall under the category of mechanistic.

  • drugmonkey says:

    Now we are getting somewhere

  • The Other Dave says:

    ha ha. I want to see you guys argue about the value of cell culture experiments.

  • drugmonkey says:

    it has a place. In the cellular biology disciplines

  • The Other Dave says:

    Is neuroscience a cellular biology discipline?

  • drugmonkey says:

    Hell no

  • The Other Dave says:

    Aren't we talking about neuroscience?

    Neurons aren't cells?

    What is the brain made of?

  • drugmonkey says:

    Is transcription "neuroscience"?

  • The Other Dave says:

    Transcription is a biological phenomenon, much like action potentials and axon growth and sleeping and drug addiction.

    I don't think that transcription is neuroscience. But it might be relevant.

  • qaz says:

    Neuroscience is not cell biology, but some cell biology is neuroscience. (And some neuroscience is cell biology.) But neuroscience is more than cell biology, just as neuroscience is more than psychology. (But some psychology is neuroscience.)

  • The Other Dave says:

    If only these comments allowed Venn diagrams.

Leave a Reply