I once gave a (perfectly awful) cognitive science lecture at a major centre for brain imaging research. The main project there, as best I could tell, was to provide subjects with some or other experimental tasks to do and take pictures of their brains while they did them. The lecture was followed by the usual mildly boozy dinner, over which professional inhibitions relaxed a bit. I kept asking, as politely as I could manage, how the neuroscientists decided which experimental tasks it would be interesting to make brain maps for. I kept getting the impression that they didn’t much care. Their idea was apparently that experimental data are, ipso facto, a good thing; and that experimental data about when and where the brain lights up are, ipso facto, a better thing than most. I guess I must have been unsubtle in pressing my question because, at a pause in the conversation, one of my hosts rounded on me. ‘You think we’re wasting our time, don’t you?’ he asked. I admit, I didn’t know quite what to say. I’ve been wondering about it ever since.
See also:
ABSTRACT: Visual categorization is thought to occur in the human ventral temporal cortex (VTC), but how this categorization is achieved is still largely unknown. In this Review, we consider the computations and representations that are necessary for categorization and examine how the microanatomical and macroanatomical layout of the VTC might optimize them to achieve rapid and flexible visual categorization. We propose that efficient categorization is achieved by organizing representations in a nested spatial hierarchy in the VTC. This spatial hierarchy serves as a neural infrastructure for the representational hierarchy of visual information in the VTC and thereby enables flexible access to category information at several levels of abstraction.
Fodor’s discussion on “why the brain” is highly compelling to me. As someone who has taken many neuroscience courses, I found myself contemplating whether any of it was worth it.
ReplyDelete“Maybe the availability of the new technology is running the science rather than the other way round.”
This is an interesting point as the functional localization trend snowballed as we gained the tools to do so. After discovering different functions of the brain areas according to the outdated theory of brain modularity, it has been discovered that, in reality, the areas and their functions are deeply interconnected and much more reliant on other processes than we previously thought. So maybe if the research performed on the brain at the start was haphazard and not directed at answering specific questions, could we have come around to making it worthwhile?
But overall, I think Fodor might have a point; though neuro-imaging might give us some information, I don’t believe it has helped us understand cognition as a whole to the extent that was expected.
Fodor's point is that localizing when and where brain activity occurs when someone is doing something cognitive does not explain how it is being done by the brain. Not even the growing ConnectomeProjectdoes
DeleteBut Fodor's short article also says a lot of irrelevant things too:
The nativism/empiricism stuff is completely irrelevant. So is the utter nonsense about dualism. And he exaggerates the futility of studying the brain (though he admits it could sometimes be important for clinical research).
I was also intrigued by the above quote about “new technology running the science.” We have created the tools to study the brain and possibly we are just treading around the central question of how the brain could achieve cognition. With the localization of brain functions, we aren’t explaining the “how” directly, but I think the “what” and “where” is not to be underestimated. I don’t doubt that for the next few years there will be a plethora of new information about what goes on in the brain; even if for now we don’t have the means to dig a layer deeper into how the brain and the mind overlap, having this accumulation of knowledge about brain activity informs the technology we create (neural networks for example) in order to actually treat people. Again, from a philosophical standpoint, this information might seem redundant, but from a medical standpoint, I want to believe that we are headed in the right direction by studying these aspects. I do think that the brain is fascinating and that we’ll be able to do more in the future if we keep on with our interest in it.
DeleteRemember that this is cogsci rather than clinical neurology. And not only does when and where activity occurs in the brain not tell you how and why it produces whatever cognitive-capacity it is correlated with; it's not even clear what it means to say we know what the brain is doing. A correlation is just a correlation until you have a causal explanation (that works!)
DeleteTess:
DeleteI agree with you that "the 'what' and 'where' are not to be underestimated." However, regarding the brain's functional localization and cognitive science, Fodor's frustration seems to stem from the irrelevance of "what" and "where" to advancing our understanding of cognition. For instance, Searle could execute the right program in his CRA to "appear" to understand Chinese even if he doesn't. The fact that he doesn't FEEL like he understands Chinese is the qualitative content of a sensory state. I think what Fodor is trying to emphasize here is that the mental state of "not-understanding Chinese" has little to do with the "what" and "where." The brain's functional localization only attempts to characterize mental states' properties by juxtaposing them to correlative neuroimaging data. To illustrate, as a causal system, a refrigerator is something that keeps the food fresh. The "What" and Where" do not explain what refrigeration is because there can be different causal systems (or different implementations of a causal system) capable of achieving the same result (keeping the food fresh). Similarly, studying the brain's localized functions wouldn't bring us any closer to the answer to the easy problem, let alone the hard one.
Does the Human Connectome Project (HCP) not prove to be a worthy endeavor because it seeks to understand more than when and where brain activity occurs? Because it seeks to understand yes, the when and where, but also the complex networks which work together to produce our ability to do the things we do. It seems from Fodor's article that what he dislikes is the typical expensive fMRI experiments which seeks to show the links between menial behaviours and a specific group of neurons. I am sure the HCP uses experiments such as these but when thousands of them are combined together to further our understanding of how we do what we do is that not a worthy endeavor? Having said this I cannot help but agree to an extent with Fodor. For even if from the HCP we know one day what complex network of brain activity (down to the last neuron) produces say an empathic response, will we be on our way to explaining the easy problem? I'm not sure
DeleteIf the connectome data help to pass T3 or T4 then they will have made a contribution and refuted Fodor. If not, not.
DeleteI found the distinction between empiricism and rationalism interesting to consider in the perspective of the localization of function in the brain. Indeed, understanding whether mental processes are localized differentially in the brain prompts the question of whether our minds are homogenous or heterogeneous. Homogenous minds would imply that we can perform mental processes almost anywhere in the brain probably using similar mechanisms, whereas heterogeneous minds imply that mental functions are performed in different brain areas, thus possibly using different brain mechanisms. Empiricists argue that experience alone shapes our mental capacities, meaning that the brain would likely be equipotential since mental functions emerge purely from patterns of association. On the other hand, rationalists distinguish different types of mental processes that they suppose occur differently and in different brain areas, since according to them the mind’s structure is mostly innate. Therefore, empiricists defend a homogeneous minds theory whereas rationalists argue for a heterogeneous minds theory.
ReplyDeleteFodor’s anti-empiricism is because he is a Chomsky wannabe.
DeleteBut as we’ll find out in weeks 8 and 9, Chomsky has strong evidence and reasons for his conclusion the Universal Grammar is innate (the “Poverty of the Stimulus,” whereas Fodor’s generalization of this to all or almost all language and thinking has no evidence to support it (and hardly even makes sense). And he’s so anti-empirical that he even doubts Darwinian evolution.
But his reservations about brain imaging and localization are worth thinking about.
In this article, Fodor makes his argument against brain imagery and the current scientific focus on figuring out the mind by studying the neural activity of the brain. He explains that we should be trying to figure out if mental functions are actually neurally encoded in the brain rather than where they are located. It makes even less sense to focus on the second question when we realize that there is no substantial evidence showing that the brain locations of certain functions will give us any information about their connection or similarity to other functions. Except for clinical purposes, which Fodor acknowledges was historically the primary goal of this type of science, it seems that the study of brain anatomy and activity misses the point (understanding the mind) completely (at least according to Fodor).
ReplyDeleteGood summary. Maybe it's a question of time. But remember the heart, which just pumps blood, so its mechanism can almost be "read off" from its structure. The brain "pumps" everything we can do (T2 and T3). That's a much taller order. I'd bet more on computational modelling (Weak AI, -- Strong CTT) than where/when localization and connectivity. (But Weak AI is not computationalism! Can you explain to kid-sib how it is not?)
DeleteSo what (if anything) does Fodor think cogsci should do instead?
DeleteWeak AI is not computationalism, since it does not hold true that cognition is computation, or, in other words, that the brain is a computer (this would be Strong AI). Rather, weak AI suggests that computation can ‘simulate’ everything in the world, including the brain, and therefore that computers are useful to explain just about anything, including cognitive science.
DeleteIn this article, it is quite straightforward to see that Fodor wants the focus in cognitive science to turn away from brain imagery research and turn it back onto other types of psychological research. He argues that we should first question whether our mental functions are neurally localised in the brain, rather than where: this is the place to start, but it seems difficult to look into this question without using the very brain imagery he dismisses. Fodor also mentions phrenology when discussing the rationalist thought on this question, but it seems unlikely to me that this would be a viable alternative to reverse-engineer our cognitive capacities.
DeleteThis piece helped me finally realize the difference between reverse-engineering vegetative and cognitive functions. To study the heart is to know how it pumps blood, which can be directly observed. But mental processes are much more intricate, and we cannot observe how the brain "pumps" behaviours and thoughts. So collecting correlative data on where the brain lights up when completing certain tasks sheds little information on how the mental process functions. This got me thinking about what we should do to reverse-engineer cognition (i.e. cognitive science).
DeleteAs Mathilda said, Fodor suggested other types of psychological research instead of brain imaging in this piece. However, considering past lectures, I believe one way to reverse-engineer cognition could be trying to build a robot by computation with sensorimotor engineering technologies. If so, could studying the neurons' underlying functionalities and how they connect with other parts of the body shed light on sensorimotor studies?
Mathilda, although I doubt it's what Fodor meant, it does make some sense to ask (and test) whether the mechanism underlying cognitive capacities in the brain is local or distributed (as in a neural net).
DeleteJenny, I'm not sure what you mean by "build a robot by computation with sensorimotor engineering technologies." If you mean a computer with optical and moving peripheral devices, that's what robotics these days is doing, but mostly just to build robots that could do useful things for people, not to reverse-engineer how people do it. Cognitive modelling (Weak AI) might be a better bet. The question is whether neuroscience discoveries would be helpful to cognitive modelling.
Hi Professor! Indeed, what I meant was robotics parts, such as optical and moving peripheral devices, would be helpful for cognitive modelling. They provide information in the world to cognitive models, similar to how humans receive information through our senses. I believe that if we were to have a working T3, the related robotics technologies (visual, sensual, vocal, etc.) must be present for T3 to function. As in Fodor's argument, localizing functions in the brain only shows us the correlation between those neurons and activities. This gives us little information on the actual mechanisms behind how we have those abilities, such as in the case of mirror neurons.
DeleteI think this is a great example of why interdisciplinary translatability is such a tricky process. The inductive manner of neuroscientific data collection and trying to derive structure from that doesn't gel easily with Fodor's philosophical deduction from the givens of experience and cognition to try and formulate structure. And Fodor has a great point that it doesn't make sense at all to shoot in the dark for structure in order to deduce mechanism.
ReplyDeleteHowever, Fodor does seem to go a bit far in claiming there is *no point* in data collection or localization per se- indeed, it seems mostly to beg for a better and more interdisciplinary research program in which the results from neuroscientific experiments are synthesized within some kind of coherent theory of cognition (which he seems to try to provide in modularity). Which leads us right back to the beginning of this comment...
Fodor was a philosopher, so he was not himself doing empirical experimentation or theoretic modelling (except briefly, in collaboration with Bever, early in his career). But he had a point about localization. (Modularity went into and out of fashion without making much progress on the easy problem, let alone the hard one.)
DeleteFodor took a very abrasive approach in this reading, which at first made me disagree with everything he wrote. He underplays the benefits of functional brain mapping and he seemed to be stating that neuroscience research is redundant. Neuroimaging is a powerful tool in clinical diagnosis and treatment. Stating that he wonders “why anybody cares” is extreme.
ReplyDeleteHowever, I now see that what Fodor was really trying to stress is that neuroimaging – although not irrelevant clinically – is irrelevant in terms of reverse-engineering T3 capacity. Finding out where something is localized in the brain when someone does something does not explain how and why someone can do it. Neuroimaging offers only correlative data, rather than a causal explanation.
Josie, your response encapsulates my thoughts as well.
DeleteThe whole time I was reading this article I found myself asking, Isn’t this just a matter of taste? Fodor says he finds astrophysics to be interesting. Well, many people would find reading about brain regions more interesting than reading about quasars! In addition, he writes one sentence about the usefulness of brain function localization. A musician being able to retain his ability to play guitar after a brain tumor surgery is a worthy endeavor and is one of the reasons “why anybody cares.”
But yes, the when/where is not the same as the how/why. Why do we feel? Why do we prefer certain foods, colors, or music genres? I’m not here to criticize Fodor’s writing style, but I do think his message is lost in his “extreme” viewpoint which is too bad. If you’re like me and this article left you feeling lacking, read Prof. Harnad’s 2019 “Can neuroimaging reveal how the brain thinks?” article which conveys the hard problem more succinctly (in my opinion), and also gives useful analogies to summarize why cognition is more than just what brain regions light up in an fMRI.
Lots of people have been frustrated by Fodor's style. But the one small point he made here (amidst a lot of stylistics and irrelevancies), about w/w vs h/w, is valid. (By the way, the "why" in "how/why" is an evolutionary "why," Week 7, not "why do I prefer this rather than that"...)
DeleteYou put down my thoughts into words. While the approach was very offputting, the criticism of localization and how we only have correlative data is a valid criticism. However I do think we should consider the fact the Neuroimagery is a young practice and over time with more refined tools we could perhaps do more than imply through correlational data.
DeleteIf neuroimagery findings ever eventually help pass T2 or T3, then Fodor is refuted; if not, not. If they only contribute to T4, and T3 can be passed without them, then Fodor was still right.
DeleteIn this article, Fodor is quite dismissive of using brain imaging techniques in cognitive science to reverse engineer the mind. I think an important distinction can be made here from reverse engineering the mind and the brain. I think Fodor would agree that if the goal was to reverse engineer the brain for more vegattative functions (in the same way we would the heart), then sure we could use neuroimaging. But functional locatilastion is not enough to understand our “how the brain pumps behaviour”, especially because all the data is only correlational, neuroimaging cannot give causational data. I’m a bit of a cynic and myself largely agree with Fodor, but I'm curious as to what he proposes cognitive science focus on instead. He is a philosopher so presumably philosophy should play a role, but maybe he has ideas for more psychological explanations as well.
ReplyDeleteSophie, I'm not sure what you mean by a "psychological" explanation in cogsci, which seeks a causal explanation. But you're right that neither Fodor nor Searle, both philosophers, can tell cogsci what to do, just what not to do.
DeleteFodor's argument in the journal post is based on criticizing the popular side of recent news about neuroscientists' use of brain imaging to localize mental states in the brain. One key point he keeps emphasizing is the idea that while mental functions may have a specific location, the constant devotion to determining where they are localized in the brain is indeed irrelevant.
ReplyDeleteAs someone who has taken numerous neuroscience classes, his argument of not caring about where they are neurally localized had never occurred to me. Similarly to how he mentions the surgeon, it was always a matter of wanting to find out where things were in order to advance our clinical knowledge. And in such cases, it would make sense because neuroscience is simply a field concerned with finding scientific solutions. Specifically, it never takes into consideration the hard problem, a further step neuroscience frequently overlooks. Now, I agree that it may appear that one mental function is localized in the brain, but as Fodor emphasizes, there must be a further explanation of "how and why." Ex: How and why is it that damage to Broca’s area reduces our speech production? At the moment, we know so much about brain localization, yet we don't take into account this further step. Thus, I believe Fodor's knowledge as a philosopher is a point of view that can bring much relevance and deeper explorations than just what neuroscience has become today.
Neuroscience does not try to solve the heard problem (but no one is getting anywhere with tha). But neuroscience has not made much progress with the easy problem either, except for some simple vegetative function.
DeleteI would have to agree with you Maira. Being from a very biological background, Fodor's viewpoint of brain localization initially struck me as refutable. However, when spending more time on considering its implications, I feel I understand his take home message. While I do not believe that neuro imaging is of no use, Fodor's philosophical argument for its banality in understanding why and how things happen in the brain does hold true. Your example with Broca's area supports. this beautifully. Integration of Fodor's idea posits a new method for neuroscience research in which following localization, we work to develop new hypotheses for the ways in which this localization has arisen. It is through the insight revealed from the how and why that we will become closer to being able to reverse engineer the brain.
DeleteEmma, it is surprising how little neuroscience has so far contributed so little to reverse-engineering cognition. I'm not sure, though, that, even if his critique of localization is correct, he has provided a "new method" (just as Searle has not provided a new method).
DeleteVituperations and rants aside, Fodor's critique of neuroimaging research is an important one. In fact, I think his overall point – that we cannot explain how cognition works by simply observing the brain – is even more important than he seems to realize. It relates to a fundamental problem in the philosophy of science: that of understanding the role of evidence in science and how it relates to scientific theories. At bottom, what Fodor is rejecting (unknowingly?) is empiricism: the idea that scientific knowledge comes from observation. He is realizing, like many before him (e.g., Kant, Popper) that raw observation is quite empty, and cannot be the basis scientific theories – you can accumulate as much data as you like, and still not get a jot closer to explaining how it is the phenomenon you're observing is brought about. So, in a sense, he is right: what we need is not more observations per se, but new and better theories, which we can then test using observations – including brain imaging, but also behavioral experiments, computational models, etc. But this is what Fodor goes wrong: he seems to miss the point that, by allowing us to test our theories, observation plays a critical role in science, and therefore should not be dismissed as a something of minor importance. We absolutely need neuroimaging techniques (as well as other experimental methods) in our quest to understand the mind, just as much as physicists need particle accelerators to in their quest to understand the subatomic world. We simply must stop expecting that the paraphernalia we employ in experiments will do the explanatory work for us.
ReplyDeleteGabriel, observations (i.e., data, measurements) do not explain themselves. Turing did not expect the data (i.e., observations of what people can do) to explain themselves. Turing was proposing an empirical method for cogsci: Observe what people can do. Then go and do your creative work (maybe with the help of computational modelling (Weak AI, Strong C-TT). Then, when you think you have a computational model that looks like it works (within your computer simulation), build it in the real world (as with the test after the rocket simulations). And then Turing-test it, to see whether it really does work (as in the rocket simulations). [In other words: reverse-engineering.]
DeleteBut you seem to be suggesting that it should be tested against brain function. Well, if it’s a T4 model, yes, that too; but if it’s only a T2 or T3 model? (T3 is still a robot that has to be built; and even T2 has to be more than just a computer, as we learned from Searle’s CRA).
I like your point about how not all observation is pointless, but that it needs a directed theory and question that it is attempting to answer. Generating hypotheses and then attempting to prove or disprove them is the basis of the scientific method that has permitted great discoveries. Without the theorizing/hypothesis stage, we will get lost in never-ending “answers” to questions that were never asked and would not have been worthy of an investigation in the first place.
DeleteWhat is right, wrong, and irrelevant in Fodor's article?
DeleteHi Steven,
DeleteMy point was simply to stress that neuroimaging is not of minor importance, insofar as it is one of the types of evidence that we can use to test our hypotheses about human cognition (alongside with behavioral studies, computer simulations, robotics, etc.) But in order to do that, we first need the theories; to keep scanning the brain and looking for blunt correlations without a prior explanatory is just pointless stamp collecting.
By the way, note that I said "human cognition". I agree with you that, even if brain imaging can shed light on the specific way the human brain achieve a certain cognitive ability, it may not be needed to construct a equivalent system with the same ability (assuming there are multiple ways to achieve that functionality). To this end, other experimental tests (robotics, simulations) may suffice.
Gabriel, good points.
DeleteFrom what I understood after reading the article, Fodor seems to have a virulent opinion about the neuro-imaging research. According to him, except for medical reasons, knowing what area of the brain is activated when one is engaging in a particular task is not relevant. I can see that knowing the location of the activation in the brain doesn’t answer the question “why does that area get activated?”, but I think that Fodor is a bit too extreme. Everyone in the comments already talked about the debate between rationalists thinking that the mind is heterogenous and the empiricists thinking that it is homogenous, so I will mention that one thing that I found interesting in the article is the how money is used in research. He indeed said “science is expensive, and it’s largely publicly funded, and there’s never enough money to do all the research that might be worth doing. In particular, brain imaging is expensive compared to other ways of trying to find out about the mind.” That quote made me think about whether or not everything that is done in science is absolutely necessary. As professor Harnad repeated many times in class, for example, are experiments on animals really necessary when it comes to cognitive science?
ReplyDeleteCharlene the question is not “Why does that area get activated?” but “How and why does the brain activity cause the capacity with which it is correlated?”
DeleteBrain imagery experiments are non-intrusive and harmless, though expensive, so I don’t like to compare them with animal experiments, which are much, much worse (especially in cogsci, which is not even clinical). Other sources of data are behavioral testing, and other ways to test causal hunches are through computer modelling (Weak AI).
Like those who responded to Fodor’s article and some students in this comment section, I found his criticism of brain networks and modules frustrating because it doesn’t advance how mental states can be connected to the brain. As Prof Harnad suggests in “Can neuroimaging reveal how the brain thinks?”, we need a new way of thinking creatively, which is likely true but even more obscure. Sure, explaining where signals occur in the brain during a certain cognitive process doesn’t solve the hard problem, but what options do we have other than mapping things out? If it is possible to elucidate the cognition problem, it seems plausible that discovering the physical properties of the brain could help in some sort of way, even if it’s by refuting once and for all the possibility of a causal relationship.
ReplyDeleteRosalie, no one has an algorithm for creativity, but Turing’s method is to try to computer-model human cognitive capacities, and then Turing-test whether your model works through T2, T3 or T4. That’s “reverse-engineering” (and empiricism). Only T4 requires the model to explain the activity going on inside our brains, rather than “explaining” everything (cognitive) that our bodies can do in the world. (See other replies on computer modelling (“Weak AI”) compared to computationalism (“Strong AI”).)
DeleteThe main point of Fodor’s article is that he thinks neuroimaging is an excessively expensive and ineffective method of solving the problem of how cognition works. By using it to discover physical properties of the brain, it does contribute to discoveries by eliminating incorrect hypotheses, but is that really the most effective way of solving problems? That is alike randomly testing every possible option until we find the answer, which may be acceptable if it wasn’t so expensive to perform each of theses tests. We are unfortunately bounded by resources (time & money) so it makes sense to prioritize using these tools for problems that absolutely require it. Or at the minimum request that these neuroscientists have a specific objective when conducting these experiments rather than “stamp collecting”.
DeleteFodor’s idea in his paper “Why the brain?” is that neuroscience is focused on discovering which parts of the brain pertain to which actions. A lot of funding, research and development is put into answering questions such as which part of the brain is activated when I think of a teapot. However, he feels that money could be put elsewhere, as having this information does not unveil how the brain works in its entirety. Indeed, I believe the brain is to be understood as a complete entity and we must understand how different parts of the brain communicate and share information to come up with the concept of a teapot.
ReplyDeleteIn order to solve the intelligence problem, explaining all our cognitive functions and how we do things should come from a cooperation between subjects such as neuroscience and philosophy, we should study this through the lens of cognitive sciences. Fodor’s point of view should not be to disregard the work of neuroscientists but rather to help them and interact with them. Maybe neuroscientists have not discovered how we produce thought but with the help of other core subjects, we might.
Étienne, yes, there’s always a tomorrow, but mirror-neurons are an example – and not an atypical one – of how localization and correlation can fail to arrive at a causal explanation of cognitive capacity. Roboticists, for example, do not ignore neuroscience: they just have not learned much from it to help yet (see the robotic optional reading on mirror neurons). And don’t forget Turing’s reverse-engineering and T-testing.
DeleteI agree that Fodor's critique could be a bit more constructive that questioning the utility of 'which part of the brain lights up when I think of a teapot'. On the other hand, I think that those type of 'smaller' neuroimaging experiments shouldn't be discredited - they serve to understand mental function, whether involving more trivial things or not.
DeleteI usually I enjoy critical pieces more so than this one, however I appreciate Fodor's requestioning on a lot of popular psychological debates or experiments, especially Pavlov's one being undoubtebly unethical in order to make a not-very-out-of-the-box discovery. I also have to add that it was a pleasantly humorous reading.
Ariane, Fodor was a gadfly, and sometimes unwarrantedly dismissive, but he was clever and stimulating.
DeleteAdding on to what I previously said about Fodor, I find his comparison of the argument between homogeneous minds and heterogenous minds to the argument between empiricists and rationalists to be an interesting observation. Bringing in a philosophical approach to neuroscience/cog sci is definitely interesting and always an important point of view. However, I noticed that within his text there were quite a lot of “I think” ’s, which of course being common in philosophy, might be a little too contemplative for commentary on a scientific subject.
Delete“But why (unless you’re thinking of having it taken out) does it matter where in the engine the carburetor is?” Reminded me of the example about the difference between reverse-engineering the heart and the brain. In the case of the hearts the organ wears it’s function on it’s sleeve - what it does outside resembles what it does on the inside (pump blood) but you can’t “see” how the brain does what it does - “pump” behavior.
ReplyDeleteAs far as I understand, according to Fodor, neuroimaging is probably not getting us anywhere on the how question. It seems like this is because the “let’s see what it looks like” approach in neuroimaging is not giving us nearly a detailed enough map for there to be any insight gained from it.
But I don't think mapping out the brain would work to understand cognitive functions. The reason for this is that despite both of them being physical objects, brains use their physical components to represent things symbolically whilst car engines use their physical components only to execute mechanical actions, (with no symbolic representations involved).
Melis, what is a “symbol,” a “representation” and a “symbolic representation”?
DeleteWhile yes a car engine uses physical components to make the vehicle move forwards, but does the brain not do the same? Are the symbols you speak of not just the result of the physical matter of the brain, the firing of neurons and their interaction with the environment?
DeleteIf we imagine a reverse engineered brain, does this not imply that the symbols the brain compute are formed from the physical reverse engineered 'computer' that hosts them?
Emma, I could not quite follow you. You seem to be ignoring Searle's refutation of computationalism. What is the difference between a computer and a "computer"?
DeleteI think that a symbol is just an arbitrary shape manipulated by its shape and not its meaning. As said in a previous reading, “Symbols in a physical symbol system are physical objects that are part of the real world, even though they may be internal to computers and brains”. So the representation seems to be how we interpret that arbitrary shape (the symbol) in our minds, in the abstract world. And symbolic representation might be the result of that procedure of representing symbols in our minds.
DeleteFodor's commentary reminded me of the "stamp collecting" critique I learned about in a past psychology class. This critique specifically applies to neurophysiology and critiques the practice of studying neural mechanisms without first knowing their purpose and how they may contribute to advancing science. Without a clear goal or understanding of what a study may be contributing to the field, researchers can practice science for the sake of science even when a study may be unethical (see: Hubel & Wiesel cat visual perception experiment). Fodor's question at the end of the paper to the neuroscientists at the science convention almost applies this same critique- instead of just searching for the underlying neural mechanisms of behavior, we must first understand if (and if so, what) these experiments actually contribute to the field before their (and subsequent studies on more and more granular levels) execution. Parallels of this critique can be found in Fodor’s critique of cognitive science; besides the clinical capacities that these neuroimaging techniques may offer, a lot of neuroscience studies end up as practice of “stamp collecting” than actually contributing to understanding how and why cognitive capacities arise from the cortical behavior they are examining.
ReplyDeleteDarcy, true, but not just in neurosci!
DeleteYour comment made me think about the benefits of achieving the goal of cognitive science. Yes, we know the goal is to reverse engineer US, or our cognition, but what are the benefits to society at large. I am curious to know what the leaders in cognitive science are trying to achieve, though it is a long ways away. Of course, it would be informative for understanding ourselves further. But, similarly to Fodors opinion, why do we gain from this knowledge
DeleteLaura, if you were saying that researchers should not do anything that harms society (which they do, a lot), I would agree. But I'm not sure that if research doesn't produce benefits to society then it does not provide knowledge...
DeleteI feel like Fodor's missing the point in that finding where certain functions are localized in the brain can be useful to us in many ways. For one, localizing function can help us make clinical improvements from scientific findings. I believe that this clinical significance carries a greater weight than anything else, as helping people should be our primary goal. I understand that Fodor is arguing that we should be looking at whether mental functions are neurally encoded in the brain, but won’t we answer that question if we find where they are located anyway? Understanding the mind has many components to it and we don't yet understand all of how the brain gives rise to the mind, but I don't think building a map of the brain in evolving detail is a bad step to take in order to achieve that goal. If we one day figure out how the brain represents the mind in every bit of detail there is, many possibilities can arise. Scientists have already mapped all 302 neurons in the nematode C. elegans, forming a Connectome, helping them better understand the structural connectivity of the organism's nervous system. Although we have 86 billion neurons, achieving the same sort of mapping within ourselves can help us achieve the goal of better understanding the mind.
ReplyDeleteAlexander, as I agree with some of your points, I believe Fodor is talking about cases where scientists are looking for answers to questions that they are not even sure about. From what I understood, his main point is that even if we know what part of a brain is activated while thinking about a certain thing, what use is it since we know it's already somewhere in the brain? So in what sense do you think knowing which part of the brain is activated during a certain task would be an advantage for us? I would really like to know your thoughts!
DeleteHi Alexander,
DeleteI think the main critic of Fodor is from the point of view of cognitive science. Knowing the structures of the brain is what the brain scientists do not cognitive science. For cognitive scientists, knowing the structure of the brain does not tell us about the "how" and "why" about cognition (from what I read from the replies above). Reverse engineering cognition does (from professor Harnad). I am not sure what Fodor would have said.
In other words, studying the brain does not suffice the purpose. Cognitive science aims to understand human's cognitive capacities. Is it true that these capacities arose from the brain itself alone? Maybe but very unlikely...
DeleteCorrection: "cognitive capacities" not "human's cognitive capacities".
DeleteAlexander, see other replies about connectome.
DeleteAlara & Yumeng, good replies to Alexander, but where else do you think cognitive capacities arise from in organisms if not from their brains, Yumeng . Or are you referring to T2 and T3?
I think there is a clear winner between the two competing theories of empiricism and rationalism in that we already know our brains are not homogenous. In line with the rationalist view, different types of mental processes do happen in different brain areas, but that doesn’t mean that empiricism is completely wrong; because of neural plasticity. Experience does shape our mental capacities in some way, but not all. For example, London Taxi drivers exhibit an increased volume of their hippocampi from memorizing the complex streets of London over the years. The research shows that the hippocampi grow in the process of learning the streets, rather than the drivers starting out with an increased memory capacity. Given this knowledge, we still know that there is localization of function within the brain, where different areas are devoted to different tasks. So it’s a bit of both, but empiricism is too extreme.
ReplyDeleteOf course organisms can learn, and that's already in the TT. But, as explained in class, Fodor's empiricism/rationalism is a red herring. (How?)
DeleteI was very amused reading Fodor’s article about the pointlessness of neuroimaging, especially coming off of Searle’s adamance that neuroscience is the best and only way to study cognitive science. Searle’s staunch opposition to dualism fueled his insistence that neuroscience is the only avenue for cognitive science. And here, Fodor insists that one must believe in dualism if one is “prepared to hold that facts about the brain are metaphysically privileged with respect to facts about the mind.”
ReplyDeleteI think this fundamental disagreement between the two really drives the point that the field of cognitive science is nowhere near having an agreed-upon direction in terms of answering the major questions.
Searle and Fodor are philosophers. Dualism is philosophy, and irrelevant to the cogsci's (and Turing's) program to reverse-engineer cognitive capacities.
DeleteMany of Fodor’s points made sense to me in this reading. As I have read the text about mirror neurons before this one, I automatically started to think how much, if at any level, the discovery of mirror neurons was useful. Fodor draws attention to a point about Pavlov’s dogs: “ ‘But we already knew that,’ she says, in some perplexity. ‘Now we know it scientifically’.” When I think about this quote with the mirror neurons reading in mind, I arrive at a similar conclusion: we already knew that we could imitate or observe, far before mirror neurons were discovered. Then what use is it to us to know that mirror neurons are the reason we perform these tasks, unless it explains the causal mechanism behind it?
ReplyDeleteThis is a compelling point. My only thought is that perhaps the causal mechanism behind mirror neurons is their evolutionary role in language? In other words, perhaps our need for language as a species caused the mirror mechanism to evolve, yet this still doesn't explain how/why the mirror mechanism lights up and affects mental states when we observe others' (non-)intentional actions/emotions...
DeleteThe differentiation between common sense and knowing something “scientifically” is also what stood out most to be in this reading. Fodor emphasizes the point that it appears certain questions are being asked in science just so modern technologies can be put to use, and so scientists can claim they’ve contributed to the field. The only sense of purpose I’ve gathered from this type of experimentation is through what Fodor has mentioned in this writing which was that if the brain is doing the tasks in different areas (ie localizations), it may do these tasks in different ways as opposed to using the same neural mechanism for different kinds of thinking if there isn’t localization of tasks in the brain.
DeletePolly, we’ll be discussing evolution in Week 7. (1) It’s not just humans who have mirror-capacities. (2) Evolution does not work by anticipating future needs.
DeleteThis is an interesting point, because we've been talking a fair bit about inutility of searching for neural correlates of certain behavioural tasks, etc. ("stamp collecting") without having an underlying theory. However, the discovery of mirror neurons (for example) has had a metascientific effect on neuroscience where it has opened up a whole new research program searching for correlations between behavioural and neural phenomena as relate to mirror neurons in something of a frenzy; the same goes for many similar discoveries. This critique has a similar shape as Fodor's of what happens to a field like neuroscience upon the popularization of a new technology. It may just be that this itself is the result of not having a well-defined or articulated theory, and so independent discoveries like that of mirror neurons are propped up as the theory, or part of what is meant to be one. But it seems tough to be able to tell, then, what is the emergence of a potentially directed theory with testable questions (which isn't *not* the case for some of these research programs as with mirror neurons) and what is more aimless red-herring stamp-collecting.
DeleteFodor once wrote “what our cognitive science has done so far is mostly to throw some light on how much dark there is.” I felt as though this quote sums up his doubts surrounding brain imaging research. His hesitations around brain localization experiments demonstrate the limitations of narrowing our focus, since – from his point of view – they still do not help us answer the question of how the brain does what it does. But Fodor posed a very curious thought: why is it that we are fascinated by neuroscientific explanations about our brain? Why do we continue to invest so much of our time and energy in neuroscientific research to understand the brain? We have seen studies that make it clear that both the use of vivid neurological images, and the complexity of neuroscientific literature alone don’t drive our tendency to prefer for brain explanations of psychological phenomena, and further the exploration into why the brain does what it does. But perhaps it is because reducing scientific phenomena at one level by appealing to a more basic level is a concrete way of bringing us one step closer to answering this question. But, as Fodor suggested in his article, sure advancing brain-based research tells us what the brain is doing when we perform certain cognitive function, but fails to address why and how it does this.
ReplyDeleteHere is a paper that I came across entitled "People’s explanatory preferences for scientific phenomena" that inspired this skywriting https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6249345/
DeleteFodor is a bit of a computationalist, which is another reason he rejects brainware explanations. He thinks there is a “language of thought” (“mentalese”) that is like language, verbal reasoning, but it is also like computation in that it too is hardware-independent. Hence the brain is irrelevant.
DeleteFodor is wrong that thinking is only verbal (propositional). Nonverbal animals and preverbal children can think too, nonverbally. But he is right that language is extremely important in human cognition. More about this in Weeks 6 and 8 (unsupervised learning, supervised learning, verbal learning).
This isn't as closely linked to our specific course material, but the quote you mention reminds me of Stuart Firestein's book on Ignorance and How it Drives Science and some other exposures to the field of "ignorance studies," which speak about how vital research is that sheds light on how much dark there is. It seems Fodor underappreciates this aspect of the work! Though this diary shows he has more specific critiques that would underline/underlie his dissatisfaction with this particular avenue of prodding in the dark.
DeleteThe how/why question, for the brain producing cognitive capacity, is not explained by localizing its correlated activity. That’s what I understood from Fodor’s article, and further crystalized from peer discussion. It seems to me that the question posed to neuroimaging here is, how instructive is it to the goal of cognitive science. That goal being to answer the (not so) “easy problem” of how organisms can do everything they can do, and the looming “hard problem” of feeling/consciousness where I think neuroimaging is trying to play a role. Cog sci’s best bet right now for the “hard problem” is to hope that the answer lies within the answer to the “easy problem”. But neuroimaging falls under a hardware/software issue, furthermore an argument of the details of the hardware- superfluous in our endeavour towards even understanding how we can do what we do. The localization of hardware informs us on how to fix the hardware (clinical relevance), but not on understanding the “how” of the software. I think neuroimaging, in Fodor’s article, is being evaluated on its ability to instruct cognitive capacity; but I do not believe that is what it was set out (or should be set out) to accomplish. Rather clinical applications are the areas in which it should be focussed and evaluated.
ReplyDeleteGood points, but don't forget T4. It is possible that T3 can only be passed if it includes some of the properties of T4 (e.g., biochemistry).
DeleteI actually think that Fodor's reply to his Rutgers colleague helped to add some nuance to the original article. In it, we can see that rather than seeing all neuroimaging as useless to cognitive science, he has an issue with the popularity of neuroimaging studies that serve no wider purpose or theory. However, even in this I disagree with him a little bit, as although it is more informational for studies that support a wider thesis, we can also gain important information from neuroimaging studies that simply show us where certain faculties are held in the brain. This can help break down complex processes into simpler ones and tell us which capabilities are utilized when completing certain tasks, an analysis that can't always be done accurately by self-reflection.
ReplyDeleteI agree with you, Elena. Perhaps some of these neuroimaging papers are produced for a specific target audience, however that does not discredit the practice as a whole. We have learned so much about ambiguous traits, like empathy, for example, from neuroimaging. Although, I can agree that knowing exactly WHERE this activity lies does not help us understand HOW and WHY we do it. Ironically, I felt I resonated with Fodor's frustration towards neuroscience in the same way that I get frustrated with the more philosophical and computer science related topics we discuss in class. I often find class discussions quite philosophical (which is a subject I avoid at all costs, despite my cognitive science major). At the end of the day, we know that our brains are essential for cognition so it follows that we would study neuroscience in trying to understand it.
Deletehey Laura, I read over your reply and I do agree with you on the point of not discrediting neuroimaging. Maybe WHERE is important in terms of mapping some function-structure in close proximity if we thinking about the wiring economy from the neurosci class? However the HOW and WHY are more important when we reverse engineer
DeleteElena, yes, but can localizing function help reverse-engineer it?
DeleteLaura, you’re right about getting too philosophical, and we know computation alone is not enough.
Monica, yes, the geography of the different mirror-capacity correlates could eventually give some clues as to how to reverse-engineer it – but so far it hasn’t given many…
Dr. Harnad, while knowing the location of a function is not very helpful in understanding its function, knowing that, for example, some of the same areas of the brain are used in imagining music and processing musical input, could be quite helpful in reverse-engineering both processes, as this indicates a shared faculty. The example may seem unnecessary, but there are many other processes that may be less obvious.
DeleteElena, yes, the hope is that correlations of multiple cognitive functions with activity in the same brain area will give a clue as to how the area is producing the function. Do you know examples where it has done so?
DeleteAlthough it does not give exact answers, one example in which imaging helped was in understanding whether facial recognition is different from recognition of other objects. It seems highly unlikely that faces would be processed in a specialized area if the process is exactly the same. Furthermore, it was hypothesized that the FFA also played a role in the recognition of items with which a person has expertise. However, a study showed that it was active only when chess pieces were placed in relation to each other, suggesting that it, rather than recognizing individual objects or specific scenes, recognizes variations and relationships between familiar objects.
Deletechess study: https://pubmed.ncbi.nlm.nih.gov/27082047/
It's interesting to think about how most of what we have now are correlations, e.g., X is highly correlated (almost certain) to Y, or X, Y, X are correlated to Q. I couldn't agree more with Fodor's point that perhaps "we’re heavily invested in finding answers to which we don’t know the corresponding questions". There are indeed limited resources and it can take some years, if not decades for preclinical findings be put to use in clinical trials. This long experimental time costs many lives of animals, facilities, and it may in the end the information we gain lead to minimal clarity to what this line of research is for. (the above it's just some personal opinions on certain animal research)
ReplyDeleteNow back to the reading:
Fodor expressed how as a philosopher he couldn't quite understand why neuroscientists are doing and or choosing how to research. To me, it's probably not because there's absolutely no point in studying the brain, it's about how Fodor and neuroscientists have different views on how to go about things, like like empiricism and rationalism, if the fundamental ground is different, then the branches derived from is would not have much overlap after all.
I think Fodor less proposed a way in which to go about solving cogscis problems and more so just opposed what was being done at the time. His explanation of empricism and rationalism wasn't him necessarily taking a stance. It reads as though he took the extreme position of stating that studying the brain is irrelevant when there is still value to be found in correlating brain imaging to what humans do. Perhaps in a less modular way than described for each thought but a general study of brain activity. Fodor still appears to head in the right direction of looking at the bigger picture of cognition, very different than what computation chose to focus on (the meaningless algorithms)
DeleteIn my opinion, Fodor wanted to disrupt the field of neuroscience with his provocative piece. He wanted us to rethink why money and time is being allocated to functional localization brain-imaging studies that fall under the umbrella of "cognitive science" but do little to advance the field. I found this piece very thought-provoking because it encouraged me to scrutinize neuroscientific research being done in the name of "cognitive science" with a critical lens that I hadn't before applied to it.
ReplyDeleteThe paradox raised about whether science drives technology or the other way round made me consider how the modernity of brain-imaging makes it a desirable means to study cognition, yet that doesn't mean that it's necessarily the most appropriate method. Perhaps it is better to fund research that spends more time thinking about the bigger picture, namely 'how the brain pumps behaviour', than flashy neuroscience if we're truly going to make strides in this interdisciplinary discipline.
This article prompted me to wonder: contrary to functional localization findings which suggest that our brain is organized on 'geographical principles', the equipotential model of the brain seems evolutionarily likely since evolution tends to be an energy-conserving process, and having all our cognitive capacities reliant on heavily overlapping circuits would be a logical way to conserve energy. Taking a step back to make a realization like this does seem to undermine the necessity of pinpointing particular hubs of cognition, cementing Fodor's argument.
There is localized function, distributed function, redundant function, and plasticity of function in the brain. So far, neither that, nor single cell function, has led to much success in reverse-engineering the cognitive capacities with which they are correlated,
DeleteAlthough I agree with Fodor's sentiment that some research studies may seem frivolous and not worthy of getting any public funding, we only come to this conclusion in retrospect after the fact. Likewise, public financing is not actively throwing out the money in the air for researchers to grab. They actively seek original ideas that might advance our knowledge on a particular topic. In this case, it's about neuroimaging.
ReplyDeleteMaybe over 20 years ago, it was the case that neuroimaging may not have been interesting as exciting and seemed pointless. But today and future advancements would not be possible for the earlier pioneering work. Even if we may never find whether our mental states or functions are localized within the brain, it is still not a waste of time and money.
DeleteAs said above, Fodor motivates his argument by saying that since funding is a limited resource it should be allotted to what is most useful. He attacks in particular the utility of neural imaging/ emphasis on localizing brain functions. He brings up the analogy of an engine carburetor, arguing it is important to know its function but its specific location in the engine is irrelevant. For the brain, though the absolute location of language processing areas may seem only relevant to clinical research, but it doesn’t seem utterly foolish that mapping relative locations of brain functions may shed light onto the steps involved in more complex integrative functions. He acknowledges that this might “pan out” as he says, and argues that such a wanton research strategy of waiting for things to pan out doesn’t amount to much of a strategy at all; I think that is most relevant point he makes. It seems very true that “we’re heavily invested in finding answers to which we don’t know the corresponding questions,” and that relying on serendipity isn’t a very mature way to conduct science, but it doesn’t make the questions we don’t know how to pose any less interesting, at least in my view.
Searle concludes that computation is irrelevant, Fodor concludes that brain function is irrelevant.
DeleteBoth are partly right. But progress is more likely to come from those who show what IS relevant.
From what I understand, Fodor does not just reject the idea of theorizing on observation but also regard some of the theories as insignificant or unnecessary. He thinks that what we know, we know; e.g., we know the difference of nouns and verbs intuitively, and we do think about teapots. Even if we are able to explain why we can think about teapots, it is insignificant as we have already known that we can think about teapots. It is only that we can now explain it in 'scientific terms', and that's all. I think that this is incorrect, but it does point out something important. Furthering knowledge of the brain and computational and clinical applications can be the most prominent reasons of why investigating the brain is significant. However, from my own experience, I think when the research aim is on details of a very small, limited area, researchers are often so focused on the small area that they actually fail to recognize the overall significance of how it can contribute to the knowledge of a certain function or pathway and how it can be applied. Studies like this certainly lead to the so-called 'stamp collecting' and yield less interesting conclusions. I think that, when doing research, stopping for a minute and looking at how this current study can fit into the global picture is important and is what we should do.
ReplyDeleteSee above reply.
DeleteFodor does make a point, and I agree with him that brain imaging and localization is not the way to go if we want to address the “how” and “why” question. However, he did not say what we should do instead. (if we want to know the how and why part of cognition). Also, I was wondering whether it means he disfavours using neuroscience, in general, to try to solve cognition problems, or just specifically brain imaging and localization. I feel like we should not try to decide in advance which problems can/cannot, will/will not be able to be solved by science. If (assume) Fodor thinks philosophy might be the way to go to solve the easy problem (since he is a philosopher), how much weight does he think we should give to “thought experiments”?
ReplyDeleteI actually disagree in a way with Fodor’s opinion that brain imaging and localization is not how we are going to solve the questions of “how” and why” in terms of consciousness and cognition; I think that at this point in our research into the hard problem, we have exhausted the philosophical perspective, and must rely on new technological advancements in things like neural imaging and fMRI to further study the biology of the brain (more accurately, the chemistry and physics that make up the biology of the brain). Only this way are we ever going to get to the true root of the mechanism responsible for consciousness and our more complex cognitive processes.
DeleteHow about the easy problem?
DeleteBrandon, I understand where you are coming from. It's true that imaging and localization are supporting us to figure out the mechanisms that evolution has to offer, but that's probably still on the level of the easy problem (i.e., how and why can organisms do what they can do) and not the hard problem which cogsci also tries to understand (eventually).
DeleteFodor’s writing reminded me of this idea of ‘material collectives’ that I learned about in my BASC 201 class. Essentially, material collectives come to exist when groups of people mutually exchange the same representations for abstract entities. A simple example of this idea is the concept of time. Time is this abstract entity that physicists and scientists have standardized and made measurable in such a way that is universally accepted.
ReplyDeleteI’d argue that neural imaging aims to do the exact same, and Fodor is trying to make us aware of this in his article. It is true that neuroscience answers the ‘where’ and ‘what’ of brain activity, but this doesn’t really help us understand ‘WHY’ and ‘HOW’ localized brain activity causes a specific cognitive behaviour. To cognitive science, neural imaging is merely a ‘material collective’ that makes brain activity measurable to answer the questions of ‘what’ and ‘where.’
I found this distinction very interesting. Some people think that we are born with the thinking processes that we have as adults. Thus, brain activity mapping would result in the same mapping for a child or for an adult. Other people think that thinking and certain mental processes are acquired through experience. Thus, through plasticity, the brain gains more knowledge and capacities throughout aging. Here, equipotent means that every region in an infant’s brain would have the same potential for localized mental functions. Thus, I can imagine that scientists that believe in one theory or the other would not approach the easy problem and the hard problem the same way.
ReplyDelete“Likewise for anyone who cares about how much of the mind’s structure is innate (whatever, exactly, that means). If you think a lot of it is, you presumably expect a lot of localisation of function, not just in the adult’s brain but also in the infant’s. Whereas, if you think a lot of mental structure comes from experience (whatever, exactly, that means), you probably expect the infant’s brain to be mostly equipotential even if the adult’s brain turns out not to be.”
All very speculative
DeleteTo answer Fodor’s question which he starts his article with (“why does everyone go on so about the brain”), I think that it is for the same reason that the New York Times publishes articles on quasars and blackholes: there is so much about the brain that we don’t know, and of course, scientists and laymen alike are fascinated by the unknown. I disagree with Fodor on the front that it does not matter where in the brain specific capabilities of cognition lie, since in my opinion, if we want to one day understand the unknown, we need to examine the organ that is responsible for this cognition with increasing detail, and part of this is not necessarily labeling regions, but using technology to just generally learn more about how it works and how synaptic connections impact cognition.
ReplyDeleteI agree with your view and believe that, as you have eluded to, if one looks at the big picture (say something like the Connectome Project) then all of the smaller experiments matter. I do think there is substantive value in Fodor's critique however and think reverse engineering the brain is one tool among others we should be using to reverse engineer cognition. "if the best you can say for your research strategy is ‘you can never tell, it might pan out,’ you probably ought to have your research strategy looked at" is a valid claim and maybe there ought to be more money going into computational modelling, but I think both avenues should be pursued.
DeleteOn connectome see other replies
DeleteTo be honest, I've never really taken the time to consider such a fundamental question. The way Fodor describes the influence of neural imaging in research is similar to a child discovering a new toy. He writes "Maybe the availability of the new technology is running the science rather than the other way round." and although I think part of that is largely true, I also think we have an inherent interest in trying to understand what makes us "human" and any tool that'll allow us to come up with explanations will most definitely be exploited as much as possible.
ReplyDeleteInjy: I think that what Foodor tried saying with that phrase is that we are using the technology that is already available to us in order to create new hypotheses or perhaps ‘make new science’ as he says. Foodor expresses many times that he thinks the use of fMRI as an attempt to locate specific neurons responsible for doing a certain thing is rather pointless or at least not of so much usage compared to the amount of time and resources needed for that. After reading the comments, I started to see his point and agree that the amount of resources put into this will not help us to understand how and why we produce a certain cognitive capacity. With that being said, I agree with you that we do have an enormous interest in trying to understand what makes us human but this is a very big and complex question that has been broken down into many others such as solving the easy and hard problems… And again, neuroscience or even more specifically these neuro imaging techniques that Foodor describes has done very little contribution to solving such issues.
DeleteThe distinction Fodor makes between knowing something versus knowing something scientifically pin points exactly where cognition and neuroscience diverge. There are many aspects of cognition that we know yet are not (at least for the time being) able to explain scientifically (and by scientifically I mean with empirical data that assigns some sort of objectivity). On the other hand, neuroscience explains the brain scientifically. Generally speaking, something is not neuroscience if it isn't explaining something about the brain empirically.
ReplyDeleteP.s. as a phantom of the opera fan, I don't think the hit on the music of Andrew Lloyd Webber was necessary >:(
It was interesting to read about the distinction Fodor makes between empirical understanding and general understanding because it diverges how we understand concepts in certain fields. As previous replies have mentioned, in neuroscience empirical data is required compared to cognitive science where empirical data isn’t always accessible and not everything can be explained scientifically. Hence neuroscientists are empicists and cognitive scientists are rationalists. Similarly to others in the earlier comments, I did find his comments on brain networks and modules to dismissive and frustrating.
ReplyDelete