My initial impression while reading the paper was similar to some of the comments posted before. Mirror neurons are very interesting, and imitation does seem to play a key part of learning, but Rizzolatti doesn't do much to explain how this could lead to cognition. Perhaps a T2 that was exposed to human speech throughout the course of a lifetime to learn enough about the way humans talk and what they talk about, could one day imitate human speech to the point at which it would pass the Turing test. But I see this scenario as being analogous to Searle's Chinese room problem -imitation may constitute learning, but it does not constitute thinking.
I'm not too sure how much this has to do with mirror neurons, but one thing that the concept of mirror neurons reminds me of is phantom limbs. Often people who've had limbs amputated will still feel sensation/pain coming from their missing limb.
A way to alleviate the pain can be to put the limb in a mirror box that would make it appear to the viewer that the missing limb is in fact, right in front of them. Patients can then move their intact limb and make it appear as if their 'missing' limb was also moving. (Ramachandran, V. S., Rogers-Ramachandran, D. C., Cobb, S. (1995). "Touching the phantom". Nature 377 (6549): 489–490. doi:10.1038/377489a0. PMID 7566144.) (see also this very dramatic clip from a grey's anatomy episode,, where an amputee orders her intern to stab her in her prosthetic foot to alleviate her phantom limb pain: https://www.youtube.com/watch?v=fpjzg-JjzBc)
I wonder how much self awareness has to do with cognition ( I realise that many philosophers, including Descartes, have spoken extensively to this). I just find it interesting that mimicry -whether it's through mirror neurons, or mimicry of the self using mirror boxes -seems to play a role in cognition, or at-least self awareness.
My initial impression while reading the paper was similar to some of the comments posted before. Mirror neurons are very interesting, and imitation does seem to play a key part of learning, but Rizzolatti doesn't do much to explain how this could lead to cognition. Perhaps a T2 that was exposed to human speech throughout the course of a lifetime to learn enough about the way humans talk and what they talk about, could one day imitate human speech to the point at which it would pass the Turing test. But I see this scenario as being analogous to Searle's Chinese room problem -imitation may constitute learning, but it does not constitute thinking.
I'm not too sure how much this has to do with mirror neurons, but one thing that the concept of mirror neurons reminds me of is phantom limbs. Often people who've had limbs amputated will still feel sensation/pain coming from their missing limb.
A way to alleviate the pain can be to put the limb in a mirror box that would make it appear to the viewer that the missing limb is in fact, right in front of them. Patients can then move their intact limb and make it appear as if their 'missing' limb was also moving. (Ramachandran, V. S., Rogers-Ramachandran, D. C., Cobb, S. (1995). "Touching the phantom". Nature 377 (6549): 489–490. doi:10.1038/377489a0. PMID 7566144.) (see also this very dramatic clip from a grey's anatomy episode,, where an amputee orders her intern to stab her in her prosthetic foot to alleviate her phantom limb pain: https://www.youtube.com/watch?v=fpjzg-JjzBc)
I wonder how much self awareness has to do with cognition ( I realise that many philosophers, including Descartes, have spoken extensively to this). I just find it interesting that mimicry -whether it's through mirror neurons, or mimicry of the self using mirror boxes -seems to play a role in cognition, or at-least self awareness.
"mental functions have characteristic places in the brain, why should it matter to either side where the places are?" (Fodor, J. 1999)
Fodor tries to show us that this question is important. What is the purpose of discovering which brain areas are active during different tasks? Clinically, of course, it makes sense. For understanding human cognition however, it would be worthwhile to investigate what this sort of research can offer. If we can one day point to any section of the brain and say when it is active, would that be at all be helpful in explaining cognition? Only partly. There would have to be more to the explanation then pointing to brain areas.
Fodor's article is especially relevant considering the US Governments relatively recent BRAIN Initiative (https://en.wikipedia.org/wiki/BRAIN_Initiative) which will pour millions of dollars into brain research. I think it is impossible to know at the moment whether such extensive research can lead to significantly deeper knowledge. It is possible that we will be come greatly enlightened about the workings of the brain, with discoveries as great or greater than that of the mirror neurons. And that these discoveries lead us to have satisfactory explanations concerning human cognition. It is also possible that we continue to spin our wheels, and only collect more and more data on the brain, describing with greater detail the brains activity without having a deeper understanding of how it works.
“Jerry Fodor (LRB, 30 September) asks why learning where the carburettor is should be thought to help us understand how an engine works. By itself it doesn't, but it would be useful to discover that it's connected to the inlet port. This is perhaps an argument for closer links between functional neuro-imaging and traditional neuro-anatomy.”
The above quote comes from Graham Kemp at the University of Liverpool in response to Fodor’s musings on the usefulness of neuroimaging and function localization in the brain and, essentially, it reflects how I feel about Fodor’s article. In some ways I do agree with Fodor, especially in terms of how it seems we are “learning” more and more about increasingly specific areas (like brain functional loci), coming away from a neuroimaging experiment with lots of data but lacking particularly meaningful conclusions.
To me, the more important task should be to find out about pathways in the brain and how/where different functional areas connect: maybe the area reserved for teapots is actually linked to an automatic system responsible for thirst (that would be both interesting and useful because it tells us one way in which the brain connects an object to a physical state). Perhaps what cognitive science needs to do is find out how numerous, specific, functional localizations connect to form a larger network of thinking, feeling, or general brain-lightening-up. Thus neural imaging is not without cause, but it could be framed as in pursuit of a larger “image” instead of Fodor’s take that it is merely a quest for teacups.
At first I thought Fodor had taken a step too far in his diary, almost completely dismissing the usefulness of neuro imaging in terms of studying human cognition, but upon re-reading it, he did make some interesting points. Having worked in labs heavily dependent on fMRI studies, I sometimes do -- much like Fodor described in the diary -- get the impression that the researchers might have thrown in some seemingly random tasks and variables just to see what happens to so-and-so areas of the brain when we do a particular thing. Fodor takes this to mean that the researchers "don't much care". Even though something may appear, to a layman, like a shot in the dark, or a hunch, doesn't necessarily suggest that the researchers don't have a working hypothesis in mind. Sometimes they just don't have scientifically proven theories to back it up yet. I also found it quite interesting that Fodor, himself being a champion of the modularity view of cognition, would suggest (in his engine example) that "But why … does it matter where in the engine the carburettor is? What part of how your engine works have you failed to understand if you don’t know that? " In one of the letters responding to Fodor's original diary, the author brought up the rebuttal that although simply knowing the location of specific brain regions that light up during, say, verb comprehension, doesn't help us understand how we cognitively recognize this verb, knowing what these aforementioned regions connect to and how they interact with each other could be the key step in determining the underlying mechanism of verb comprehension. After all, the brain must have evolved to have the anatomy it has now for some reason. Unfortunately for us, however, we are have hit a wall with neuropsychology much like with the hard problem: we know (thanks to technology) exactly what the brain looks like, but we don't know how it came to be. Lastly, Fodor explained -- rather kid sibly -- the difference between empiricists and rationalists: "So empiricists, since they typically hold that all mental processes reduce to patterns of associations, would like the brain to be ‘equipotential’, whereas rationalists, since they think that there might be as many different kinds of thinking as there are different kinds of thing to think about, generally would prefer the brain to be organised on geographical principles. " If the method of studying Cognitive Science is reverse engineering, then wouldn't we most likely be doing what the rationalists are doing? Since we are trying to engineer modules of a machine that are capable of different human-like cognitive abilities: the verbal communications module of a pen-pal version TT-passing machine, for example. I wonder if there will come a day when we can piece together modules just like this one, but responsible for different cognitive capacities, and end up with a T2 or even T3 machine? Or do we have to build a equipotential, omnipotent machine, capable of (indistinguishably) performing every human cognitive function?
First off, Fodor's writing is a sight for my sore eyes, which are used to reading bone-dry research papers. His persuasive voice has a way of putting me under his spell, if only for the duration of the article, but I can't help but notice where we differ. After all, he says:
"But why (unless you’re thinking of having it taken out) does it matter where in the engine the carburettor is? What part of how your engine works have you failed to understand if you don’t know that?"
We can replace "engine" with "mind" and "carburettor" with "homunculus" to form a more familiar question: But why does it matter where in the mind the homunculus [or, cognition] is? It matters because if we figure out how the human cognitive mechanism works, then we will be much closer to solving the Easy problem. The physical trace of a cognitive action, like the BOLD signal we pick up from fMRI machines, can help us determine if this action is the emergent property of a network of activity or if it is centred around one anatomical region. It doesn't matter whether empiricists or rationalists are right - the arrival of high spatial and temporal neuroimaging techniques unlocks a door never before opened in the history of man. Why not explore the cookie crumbles (aka neural signals) left behind by cognition?
Fodor’s article argues that brain imaging is basically irrelevant in helping determine a causal mechanism behind human cognition. While he does make some good points, I can’t say that I wholly agree with what he has said…and I would like to ask why can’t it be possible that neuroimaging studies will help us think of a way to reverse engineer the mechanism behind how the mind works? Given Rizzolatti’s mirror neuron paper, we saw the importance of these neurons in imitation learning. While this doesn’t tell us how to engineer a T3 robot with these capacities, I think that it is a good start at understanding the kinds of networks and capacities the robot that we engineer has to have in order to do what we do. Before the discovery of mirror neurons, we didn’t know that it was an important part of imitation learning. Still, many people who don’t know about motor neurons don’t know this fact. However, why does it matter to them, because we will learn regardless? But for those trying to find the causal mechanism behind how the mind works, I think that it is important. If anything, in the very least to give us an idea about how to systematically think about how it can be possible.
The auto analogy that Fodor gives is misguided, but still illustrative. If a carburetor is located closer to certain components in a car, that gives us clues as to which other components it's hooked up to—car manufacturers don't place the carburetor in a random location far from its contribution to the signal path. But it's right in that localization doesn't give us all the answers—the carburetor may be hooked up to components it is relatively far away from, and may have been placed in a certain spot to save space.
I also don't buy the "know it scientifically" example. Just because some experiments are misguided and have little yield over our intuitive knowledge doesn't mean we should stop scientifically testing intuitive notions. Our intuitions are frequently shown to be false by experiments that have hypotheses based on intuition.
Why neuroscientific findings about localization are thought to matter is that seeing which parts of the brain correlate with certain mental states should give us clues about the structure of the brain and the causal mechanism that enables performance. But it seems that localization studies do not slice structural parts of the brain in a useful way: the part of the brain that lights up upon action recognition, for example, may light up in a million other categories of mental state. The organization of the brain seems to be so complex that traditional localization studies tell us virtually nothing about causal mechanisms. If we figure out how to localize in a more sophisticated way that yields specific complex networks that truly are responsible for one type of mental state, I believe there is the possibility that we will have gained actually useful clues to the structure of T3-supporting causal mechanisms, unless there really is no reason to believe brain functionality has anything to do with the interrelatedness of mental states, as Fodor claims. This claim is contentious. By my score, we have no strong empirical reasons that support or disprove it.
By finding where certain brain functions are localized, we are gathering information in order to begin considering hypothesis for how these things interact. Asking whether or not "mental functions are neurally localised in the brain" needs to begin being answered somehow. At least neuroscientists are attempting to see if there are correlations between brain functions and brain areas. What else should they do - nothing until someone comes up with answer to the question of how the mind works by ways of introspection (whether your own or asking others)? Neuroscience is much more objective than psychology. Observing the brain is important if you believe it is the core and map to our every day activities, thoughts, feelings, and being.
I think my point of view is best summed up in the Letters section by Steven Rose: "A theory which integrates brain and mind processes will be a major goal for neuroscientists, psychologists and philosophers in the coming decades. It will need to understand both the particularities of the micromechanisms of nerve cells and their interactions and the dynamics of the system as a whole. The scanners have a vital part to play in providing that understanding. But the first thing we need to do is to dismiss the idea that because a particular brain region is active when, to use Fodor's example, we think about teapots, this means we have located a teapot storage site in the brain. All we have done is found part of a system enabling us to think teapots."
The bits and pieces we learn from imaging studies bring us one step closer to understanding the system. We are trying to find the players involved so that we can find a way to understand how they interact.
In Jerry Fodor's diary, he discusses an interesting phenomenon that many scientists spend a lot of time and money on brain studies, and he wants to know why.
I believe the start point of studying brains is to find out how the brain works in order to fully understand human body and human behaviors. However, human brain is so complex that even with lesion studies, neuroimaging methods and animal brains studies, researchers cannot understand how brain works exactly. Although they manage to understand which parts of brain are associated with what functions (for example, Broca's area is associated with speech production), they cannot explain the whole process (for example, why neuron fires, how it fires, what happens after it fires, and what causes speech production of different words and their combination). But during these studies, people are able to treat some neuro-diseases in spite of lack of full knowledge of the brain.
Similarly, I believe the reason why people begin to wonder how people think and how they imitate or whether a machine can imitate human behaviors is that people want to know how the human body works and why some behaviors take place. If a machine can imitate everything that a human being can do, even without full knowledge of human brain, people can still create something that has the same function (but not the exact same process) as the human brain. Then philosophers and cognitive scientists begin to wonder whether a machine can imitate all human behaviors and thoughts.
Therefore, I would say if human brains are fully studied and understood, there is no need to worry about the problem of imitation and AI because all human minds and behaviors would be able to explain. However, the technology nowadays has not granted us this ability yet, and thus philosophers and cognitive scientists still regard imitation and AI as a problem to be solved.
The notion of "mind mapping" has always been a disappointment to me - pointing to a place in the brain and saying "that is hunger" is like pointing to a football field and saying "that is football" - in of itself it is not really meaningful. However, my agreement with Fodor's critique of modern "pop" neuroscience, bears a caveat:
The localization information does not come with functional information. To continue with the football analogy, this is equivalent to pointing to the football stadium and saying "that's football and everyone goes there on Monday nights, and you can get tickets to go there at this website, and two teams of players all dressed with helmets go there and seem to have special status". Even if the information is incomplete, this research can at least provide information on further paths by which to study what is going on, or, in the better case, provide information about how this structure fits into the greater whole. In the case of the mind, the pattern of activation with additional information could provide insight into the conserved structures and functional systems used to "process information" (see my comment in the overflow of 4b here for a definition of "information processing"), as was done with the "Jeffress Model" with regards to sound localization. This sort of information is useful for questions of cognitive science as it provides a system for performing some subtask relevant to cognition, and is therefore a useful in our attempts to reverse engineer our capabilities.
I support what Fodor has to say and I agree with the fact that it is useless to get to know which part of the brain does what since we are missing the link between the brain and our minds.
However with the help of neuroscience and persons with brain damage who have volunteered to be part of research; there has been many discoveries that will lead to a positive outcome.
Indeed, it was observed through experimentation on people suffering from memory problems or from memories that caused a disregulation in brain chemistry and brain functioning that some the effect memories have can be altered through reconsolidating in a different environment. This can help people with PTSD and other mental disorders to change their perception of a certain memory or event in their lives that made them suffer. People can argue that this has been going on for a long time with psychotherapy or even induced states of trance of many sorts, but this "scientific discovery" just opened the doors for new ways of healing. On another note, research on the motor and sensory cortices might enable people with no limbs to use artificial robotic limbs which are directly connected to their brain. Or even people suffering from mental illness can find some relief for a brief period of time.(i completely disagree with the fact that they are the solution on the contrary they make the problem worse but they can provide some temporary relief) Moreover, there has also been proof that visualization and observation of certain actions can activate the motor counterparts responsible for these actions. This demonstrates that our minds are not just abstract but can also have a very concrete impact.
For what it's worth i read a quote saying “Looking for consciousness inside the brain is like looking in the radio for the announcer.” I don't think that the brain generates our consciousness but what we can't deny is that there is a strong relationship between our brain and our minds and this interaction goes both ways.
It is interesting that Fodor comments: "As far as I can see, it’s reasonable to hold that brain studies are methodologically privileged with respect to other ways of finding out about the mind only if you are likewise prepared to hold that facts about the brain are metaphysically privileged with respect to facts about the mind; and you can hold that only if you think the brain and the mind are essentially different kinds of thing"—if only because, rather than ruing research whose clinical utilities are however dubious, it neglects to pay attention to the media and, at large, society which determines said research worthy of attention.
I would make the argument that the reductionism which Fodor decries—and that does seem to be his focus: why examine the brain for locative proof of obscure workings of the mind, he would have us ask, when there is already ample and sufficiently differentiated evidence in our experience of daily life—is problematic in large part due to the confusion between methodology and theory between researchers and the general public (and the media in particular). Indeed, when it comes to that which arises from the mind, it seems perfectly reasonable that the search for explanations has moved from the physiological substrate of the body, in general, to that of the brain, and then to neural imaging; it is not the complete answer, perhaps, but it is among the "smallest" area of inquiry available to scientists. The ultimate folly lies in the culture, which having happily taken neuroscience as being necessary and sufficient to fully explaining the mind, have in that conflation interpreted neural imaging as a sophisticated phrenology. Whether or not The New York Times has fallen victim to a similar perspective is worrying because it may well dictate the future of research—but ultimately, not to them; their business, and goal, is selling copy, not fact. That we read Jerry Fodor's very rebuttal would seem they have succeeded.
Hi Audrey, could add a photo to your gmail account as the others have done to make it easier for me to identify and credit you and match it with our vocal performance in class? Chrs, SH
"That being so, why does it matter where in the brain their different counterparts are?"
I guess it depends who you ask.. certainly if it relates to localizing the onset of neurological diseases then I think most would agree that it's important. I think what Fodor is getting at is the more trivial stuff, research on brain functions purely for the sake of curiosity. I suppose if one's agenda is to reverse engineer the brain then a comprehensive knowledge of all the brains' counterparts is important.
"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 like that Fodor pointed this out, it seems like such an obvious truth of science that it's not often questioned. It harkens to an earlier point Fodor talks about, whether the investment society makes into collecting this data is worthwhile. I often get the feeling that much of scientific research is essentially poking in the dark in hopes of stumbling on something useful, relevant or helpful to society. Often scientific research is also motivated by many other factors as well (prestige, money, politics etc). The story of "The Little Black Girl in Search of God" I think sums up this idea well, if we intuitively know things (like the fact that dogs salivate when presented food) then why is a scientific understanding of the event necessary? How does it help society and is it worth the cost?
Economics aside, the question remains: can finding out where and when correlates occur help reverse engineer T3 capacity. It might, but are mirror neurons an example? If not, that is?
While Fodor is obviously a very intelligent person, I'm finding it difficult to understand why a philosopher is writing a critique of cognitive science methods. Not being able to come up with reasons for understanding brain networks and mirror neurons' role in imitation learning does not mean it's not worth while. It means you don't know enough about it.
As he mentioned, in the event of brain surgery (as he drew a comparison with a car), it is important to know, which structures could affect what. Does this not merit investigation? I believe it does. Aside from changes to healthcare, understanding mirror neuron functioning could be important for child development and various other uses.
Furthermore, why were we put on this earth if not to learn. Knowledge of our world and its contents has grown exponentially. Why should we stop here and just sit to be content with what we already know. Yes I do know my brain is what does the thinking but being able to figure out which structures are implicated in which mechanisms will allow for reverse engineering of the brain.
''Furthermore, why were we put on this earth if not to learn...Why whould we stop here and just sit to be content with what we already know."
I dont think Fodor is in anyway saying we should stop knowing about the brain. He is saying that this type of neuroimaging is random, and without direction and blocks other inquiries into the brain and cognition which are more pointed and hypothesis based, rather than curiosity based ("I wonder what happens when we do this!"). Neuroimaging is like creating a map where the coordinate points consist of cognitive states (eg. thinking of a teapot, playing chess) and brain states. It reminds me of the attempts to sequence the genome, except for the fact that the human genome is finite while the amount of cognitive states is infinite. I agree with Fodor when he says," its the difference between a scientist who has a hypothesis and one who only has a camera." While making a map of the brain in this way would eventually lead us to some huge insights, doing so in a scattershot, random way, is completely useless. That being said, knowing where in the brain things are happening when we cognize is still informative in general, but is very misleading in telling us how we cognize. Neuroimaging will be useful once we have an idea of why we are trying to localise these functions. But taking snapshots of random cognitive functions until a picture appears is a waste of time.
"But why (unless you’re thinking of having it taken out) does it matter where in the engine the carburettor is"
I think Fodor's point is that all that Neural Imaging does is mapping function to a location in the brain. But mapping the location of something does not in any way explain that thing. There's no doubt that Neural Imaging has helped greatly in areas such as Clinical neurology, where you need to remove damaged brain tissues precisely without affecting other mental functionalities (think about HM). However, it did not do anything to help explaining how cognition works.
I would admit that, for cognitive scientists, where a function is mapped onto the brain is irrelevant. But the study of the way how brain is implemented (hardware) is important for cognitive science. Think about artificial neural network, if not for the study of the brain, we will not be able to build machines that can learn effectively. What computation does is that it helps cognitive science to test if an assumption is correct. For example, is the parallel processing ability at the center of human reasoning? And I think there's still a lot we don't know about the implementation about the brain that is important for the constructing of computation model to test cognition hypothesis.
My initial impression while reading the paper was similar to some of the comments posted before. Mirror neurons are very interesting, and imitation does seem to play a key part of learning, but Rizzolatti doesn't do much to explain how this could lead to cognition. Perhaps a T2 that was exposed to human speech throughout the course of a lifetime to learn enough about the way humans talk and what they talk about, could one day imitate human speech to the point at which it would pass the Turing test. But I see this scenario as being analogous to Searle's Chinese room problem -imitation may constitute learning, but it does not constitute thinking.
ReplyDeleteI'm not too sure how much this has to do with mirror neurons, but one thing that the concept of mirror neurons reminds me of is phantom limbs. Often people who've had limbs amputated will still feel sensation/pain coming from their missing limb.
A way to alleviate the pain can be to put the limb in a mirror box that would make it appear to the viewer that the missing limb is in fact, right in front of them. Patients can then move their intact limb and make it appear as if their 'missing' limb was also moving. (Ramachandran, V. S., Rogers-Ramachandran, D. C., Cobb, S. (1995). "Touching the phantom". Nature 377 (6549): 489–490. doi:10.1038/377489a0. PMID 7566144.)
(see also this very dramatic clip from a grey's anatomy episode,, where an amputee orders her intern to stab her in her prosthetic foot to alleviate her phantom limb pain: https://www.youtube.com/watch?v=fpjzg-JjzBc)
I wonder how much self awareness has to do with cognition ( I realise that many philosophers, including Descartes, have spoken extensively to this). I just find it interesting that mimicry -whether it's through mirror neurons, or mimicry of the self using mirror boxes -seems to play a role in cognition, or at-least self awareness.
My initial impression while reading the paper was similar to some of the comments posted before. Mirror neurons are very interesting, and imitation does seem to play a key part of learning, but Rizzolatti doesn't do much to explain how this could lead to cognition. Perhaps a T2 that was exposed to human speech throughout the course of a lifetime to learn enough about the way humans talk and what they talk about, could one day imitate human speech to the point at which it would pass the Turing test. But I see this scenario as being analogous to Searle's Chinese room problem -imitation may constitute learning, but it does not constitute thinking.
ReplyDeleteI'm not too sure how much this has to do with mirror neurons, but one thing that the concept of mirror neurons reminds me of is phantom limbs. Often people who've had limbs amputated will still feel sensation/pain coming from their missing limb.
A way to alleviate the pain can be to put the limb in a mirror box that would make it appear to the viewer that the missing limb is in fact, right in front of them. Patients can then move their intact limb and make it appear as if their 'missing' limb was also moving. (Ramachandran, V. S., Rogers-Ramachandran, D. C., Cobb, S. (1995). "Touching the phantom". Nature 377 (6549): 489–490. doi:10.1038/377489a0. PMID 7566144.)
(see also this very dramatic clip from a grey's anatomy episode,, where an amputee orders her intern to stab her in her prosthetic foot to alleviate her phantom limb pain: https://www.youtube.com/watch?v=fpjzg-JjzBc)
I wonder how much self awareness has to do with cognition ( I realise that many philosophers, including Descartes, have spoken extensively to this). I just find it interesting that mimicry -whether it's through mirror neurons, or mimicry of the self using mirror boxes -seems to play a role in cognition, or at-least self awareness.
"mental functions have characteristic places in the brain, why should it matter to either side where the places are?" (Fodor, J. 1999)
ReplyDeleteFodor tries to show us that this question is important. What is the purpose of discovering which brain areas are active during different tasks? Clinically, of course, it makes sense. For understanding human cognition however, it would be worthwhile to investigate what this sort of research can offer. If we can one day point to any section of the brain and say when it is active, would that be at all be helpful in explaining cognition? Only partly. There would have to be more to the explanation then pointing to brain areas.
Fodor's article is especially relevant considering the US Governments relatively recent BRAIN Initiative (https://en.wikipedia.org/wiki/BRAIN_Initiative) which will pour millions of dollars into brain research. I think it is impossible to know at the moment whether such extensive research can lead to significantly deeper knowledge. It is possible that we will be come greatly enlightened about the workings of the brain, with discoveries as great or greater than that of the mirror neurons. And that these discoveries lead us to have satisfactory explanations concerning human cognition. It is also possible that we continue to spin our wheels, and only collect more and more data on the brain, describing with greater detail the brains activity without having a deeper understanding of how it works.
“Jerry Fodor (LRB, 30 September) asks why learning where the carburettor is should be thought to help us understand how an engine works. By itself it doesn't, but it would be useful to discover that it's connected to the inlet port. This is perhaps an argument for closer links between functional neuro-imaging and traditional neuro-anatomy.”
ReplyDeleteThe above quote comes from Graham Kemp at the University of Liverpool in response to Fodor’s musings on the usefulness of neuroimaging and function localization in the brain and, essentially, it reflects how I feel about Fodor’s article. In some ways I do agree with Fodor, especially in terms of how it seems we are “learning” more and more about increasingly specific areas (like brain functional loci), coming away from a neuroimaging experiment with lots of data but lacking particularly meaningful conclusions.
To me, the more important task should be to find out about pathways in the brain and how/where different functional areas connect: maybe the area reserved for teapots is actually linked to an automatic system responsible for thirst (that would be both interesting and useful because it tells us one way in which the brain connects an object to a physical state). Perhaps what cognitive science needs to do is find out how numerous, specific, functional localizations connect to form a larger network of thinking, feeling, or general brain-lightening-up. Thus neural imaging is not without cause, but it could be framed as in pursuit of a larger “image” instead of Fodor’s take that it is merely a quest for teacups.
At first I thought Fodor had taken a step too far in his diary, almost completely dismissing the usefulness of neuro imaging in terms of studying human cognition, but upon re-reading it, he did make some interesting points.
ReplyDeleteHaving worked in labs heavily dependent on fMRI studies, I sometimes do -- much like Fodor described in the diary -- get the impression that the researchers might have thrown in some seemingly random tasks and variables just to see what happens to so-and-so areas of the brain when we do a particular thing. Fodor takes this to mean that the researchers "don't much care". Even though something may appear, to a layman, like a shot in the dark, or a hunch, doesn't necessarily suggest that the researchers don't have a working hypothesis in mind. Sometimes they just don't have scientifically proven theories to back it up yet.
I also found it quite interesting that Fodor, himself being a champion of the modularity view of cognition, would suggest (in his engine example) that "But why … does it matter where in the engine the carburettor is? What part of how your engine works have you failed to understand if you don’t know that? " In one of the letters responding to Fodor's original diary, the author brought up the rebuttal that although simply knowing the location of specific brain regions that light up during, say, verb comprehension, doesn't help us understand how we cognitively recognize this verb, knowing what these aforementioned regions connect to and how they interact with each other could be the key step in determining the underlying mechanism of verb comprehension. After all, the brain must have evolved to have the anatomy it has now for some reason. Unfortunately for us, however, we are have hit a wall with neuropsychology much like with the hard problem: we know (thanks to technology) exactly what the brain looks like, but we don't know how it came to be.
Lastly, Fodor explained -- rather kid sibly -- the difference between empiricists and rationalists: "So empiricists, since they typically hold that all mental processes reduce to patterns of associations, would like the brain to be ‘equipotential’, whereas rationalists, since they think that there might be as many different kinds of thinking as there are different kinds of thing to think about, generally would prefer the brain to be organised on geographical principles. " If the method of studying Cognitive Science is reverse engineering, then wouldn't we most likely be doing what the rationalists are doing? Since we are trying to engineer modules of a machine that are capable of different human-like cognitive abilities: the verbal communications module of a pen-pal version TT-passing machine, for example. I wonder if there will come a day when we can piece together modules just like this one, but responsible for different cognitive capacities, and end up with a T2 or even T3 machine? Or do we have to build a equipotential, omnipotent machine, capable of (indistinguishably) performing every human cognitive function?
First off, Fodor's writing is a sight for my sore eyes, which are used to reading bone-dry research papers. His persuasive voice has a way of putting me under his spell, if only for the duration of the article, but I can't help but notice where we differ. After all, he says:
ReplyDelete"But why (unless you’re thinking of having it taken out) does it matter where in the engine the carburettor is? What part of how your engine works have you failed to understand if you don’t know that?"
We can replace "engine" with "mind" and "carburettor" with "homunculus" to form a more familiar question: But why does it matter where in the mind the homunculus [or, cognition] is? It matters because if we figure out how the human cognitive mechanism works, then we will be much closer to solving the Easy problem. The physical trace of a cognitive action, like the BOLD signal we pick up from fMRI machines, can help us determine if this action is the emergent property of a network of activity or if it is centred around one anatomical region. It doesn't matter whether empiricists or rationalists are right - the arrival of high spatial and temporal neuroimaging techniques unlocks a door never before opened in the history of man. Why not explore the cookie crumbles (aka neural signals) left behind by cognition?
Fodor’s article argues that brain imaging is basically irrelevant in helping determine a causal mechanism behind human cognition. While he does make some good points, I can’t say that I wholly agree with what he has said…and I would like to ask why can’t it be possible that neuroimaging studies will help us think of a way to reverse engineer the mechanism behind how the mind works?
ReplyDeleteGiven Rizzolatti’s mirror neuron paper, we saw the importance of these neurons in imitation learning. While this doesn’t tell us how to engineer a T3 robot with these capacities, I think that it is a good start at understanding the kinds of networks and capacities the robot that we engineer has to have in order to do what we do. Before the discovery of mirror neurons, we didn’t know that it was an important part of imitation learning. Still, many people who don’t know about motor neurons don’t know this fact. However, why does it matter to them, because we will learn regardless? But for those trying to find the causal mechanism behind how the mind works, I think that it is important. If anything, in the very least to give us an idea about how to systematically think about how it can be possible.
The auto analogy that Fodor gives is misguided, but still illustrative. If a carburetor is located closer to certain components in a car, that gives us clues as to which other components it's hooked up to—car manufacturers don't place the carburetor in a random location far from its contribution to the signal path. But it's right in that localization doesn't give us all the answers—the carburetor may be hooked up to components it is relatively far away from, and may have been placed in a certain spot to save space.
ReplyDeleteI also don't buy the "know it scientifically" example. Just because some experiments are misguided and have little yield over our intuitive knowledge doesn't mean we should stop scientifically testing intuitive notions. Our intuitions are frequently shown to be false by experiments that have hypotheses based on intuition.
Why neuroscientific findings about localization are thought to matter is that seeing which parts of the brain correlate with certain mental states should give us clues about the structure of the brain and the causal mechanism that enables performance. But it seems that localization studies do not slice structural parts of the brain in a useful way: the part of the brain that lights up upon action recognition, for example, may light up in a million other categories of mental state. The organization of the brain seems to be so complex that traditional localization studies tell us virtually nothing about causal mechanisms. If we figure out how to localize in a more sophisticated way that yields specific complex networks that truly are responsible for one type of mental state, I believe there is the possibility that we will have gained actually useful clues to the structure of T3-supporting causal mechanisms, unless there really is no reason to believe brain functionality has anything to do with the interrelatedness of mental states, as Fodor claims. This claim is contentious. By my score, we have no strong empirical reasons that support or disprove it.
By finding where certain brain functions are localized, we are gathering information in order to begin considering hypothesis for how these things interact. Asking whether or not "mental functions are neurally localised in the brain" needs to begin being answered somehow. At least neuroscientists are attempting to see if there are correlations between brain functions and brain areas. What else should they do - nothing until someone comes up with answer to the question of how the mind works by ways of introspection (whether your own or asking others)?
ReplyDeleteNeuroscience is much more objective than psychology. Observing the brain is important if you believe it is the core and map to our every day activities, thoughts, feelings, and being.
I think my point of view is best summed up in the Letters section by Steven Rose: "A theory which integrates brain and mind processes will be a major goal for neuroscientists, psychologists and philosophers in the coming decades. It will need to understand both the particularities of the micromechanisms of nerve cells and their interactions and the dynamics of the system as a whole. The scanners have a vital part to play in providing that understanding. But the first thing we need to do is to dismiss the idea that because a particular brain region is active when, to use Fodor's example, we think about teapots, this means we have located a teapot storage site in the brain. All we have done is found part of a system enabling us to think teapots."
The bits and pieces we learn from imaging studies bring us one step closer to understanding the system. We are trying to find the players involved so that we can find a way to understand how they interact.
In Jerry Fodor's diary, he discusses an interesting phenomenon that many scientists spend a lot of time and money on brain studies, and he wants to know why.
ReplyDeleteI believe the start point of studying brains is to find out how the brain works in order to fully understand human body and human behaviors. However, human brain is so complex that even with lesion studies, neuroimaging methods and animal brains studies, researchers cannot understand how brain works exactly. Although they manage to understand which parts of brain are associated with what functions (for example, Broca's area is associated with speech production), they cannot explain the whole process (for example, why neuron fires, how it fires, what happens after it fires, and what causes speech production of different words and their combination). But during these studies, people are able to treat some neuro-diseases in spite of lack of full knowledge of the brain.
Similarly, I believe the reason why people begin to wonder how people think and how they imitate or whether a machine can imitate human behaviors is that people want to know how the human body works and why some behaviors take place. If a machine can imitate everything that a human being can do, even without full knowledge of human brain, people can still create something that has the same function (but not the exact same process) as the human brain. Then philosophers and cognitive scientists begin to wonder whether a machine can imitate all human behaviors and thoughts.
Therefore, I would say if human brains are fully studied and understood, there is no need to worry about the problem of imitation and AI because all human minds and behaviors would be able to explain. However, the technology nowadays has not granted us this ability yet, and thus philosophers and cognitive scientists still regard imitation and AI as a problem to be solved.
The notion of "mind mapping" has always been a disappointment to me - pointing to a place in the brain and saying "that is hunger" is like pointing to a football field and saying "that is football" - in of itself it is not really meaningful. However, my agreement with Fodor's critique of modern "pop" neuroscience, bears a caveat:
ReplyDeleteThe localization information does not come with functional information. To continue with the football analogy, this is equivalent to pointing to the football stadium and saying "that's football and everyone goes there on Monday nights, and you can get tickets to go there at this website, and two teams of players all dressed with helmets go there and seem to have special status". Even if the information is incomplete, this research can at least provide information on further paths by which to study what is going on, or, in the better case, provide information about how this structure fits into the greater whole. In the case of the mind, the pattern of activation with additional information could provide insight into the conserved structures and functional systems used to "process information" (see my comment in the overflow of 4b here for a definition of "information processing"), as was done with the "Jeffress Model" with regards to sound localization. This sort of information is useful for questions of cognitive science as it provides a system for performing some subtask relevant to cognition, and is therefore a useful in our attempts to reverse engineer our capabilities.
I support what Fodor has to say and I agree with the fact that it is useless to get to know which part of the brain does what since we are missing the link between the brain and our minds.
ReplyDeleteHowever with the help of neuroscience and persons with brain damage who have volunteered to be part of research; there has been many discoveries that will lead to a positive outcome.
Indeed, it was observed through experimentation on people suffering from memory problems or from memories that caused a disregulation in brain chemistry and brain functioning that some the effect memories have can be altered through reconsolidating in a different environment. This can help people with PTSD and other mental disorders to change their perception of a certain memory or event in their lives that made them suffer. People can argue that this has been going on for a long time with psychotherapy or even induced states of trance of many sorts, but this "scientific discovery" just opened the doors for new ways of healing.
On another note, research on the motor and sensory cortices might enable people with no limbs to use artificial robotic limbs which are directly connected to their brain. Or even people suffering from mental illness can find some relief for a brief period of time.(i completely disagree with the fact that they are the solution on the contrary they make the problem worse but they can provide some temporary relief) Moreover, there has also been proof that visualization and observation of certain actions can activate the motor counterparts responsible for these actions. This demonstrates that our minds are not just abstract but can also have a very concrete impact.
For what it's worth i read a quote saying “Looking for consciousness inside the brain is like looking in the radio for the announcer.” I don't think that the brain generates our consciousness but what we can't deny is that there is a strong relationship between our brain and our minds and this interaction goes both ways.
It is interesting that Fodor comments: "As far as I can see, it’s reasonable to hold that brain studies are methodologically privileged with respect to other ways of finding out about the mind only if you are likewise prepared to hold that facts about the brain are metaphysically privileged with respect to facts about the mind; and you can hold that only if you think the brain and the mind are essentially different kinds of thing"—if only because, rather than ruing research whose clinical utilities are however dubious, it neglects to pay attention to the media and, at large, society which determines said research worthy of attention.
ReplyDeleteI would make the argument that the reductionism which Fodor decries—and that does seem to be his focus: why examine the brain for locative proof of obscure workings of the mind, he would have us ask, when there is already ample and sufficiently differentiated evidence in our experience of daily life—is problematic in large part due to the confusion between methodology and theory between researchers and the general public (and the media in particular). Indeed, when it comes to that which arises from the mind, it seems perfectly reasonable that the search for explanations has moved from the physiological substrate of the body, in general, to that of the brain, and then to neural imaging; it is not the complete answer, perhaps, but it is among the "smallest" area of inquiry available to scientists. The ultimate folly lies in the culture, which having happily taken neuroscience as being necessary and sufficient to fully explaining the mind, have in that conflation interpreted neural imaging as a sophisticated phrenology. Whether or not The New York Times has fallen victim to a similar perspective is worrying because it may well dictate the future of research—but ultimately, not to them; their business, and goal, is selling copy, not fact. That we read Jerry Fodor's very rebuttal would seem they have succeeded.
Hi Audrey, could add a photo to your gmail account as the others have done to make it easier for me to identify and credit you and match it with our vocal performance in class? Chrs, SH
Delete"That being so, why does it matter where in the brain their different counterparts are?"
ReplyDeleteI guess it depends who you ask.. certainly if it relates to localizing the onset of neurological diseases then I think most would agree that it's important. I think what Fodor is getting at is the more trivial stuff, research on brain functions purely for the sake of curiosity. I suppose if one's agenda is to reverse engineer the brain then a comprehensive knowledge of all the brains' counterparts is important.
"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 like that Fodor pointed this out, it seems like such an obvious truth of science that it's not often questioned. It harkens to an earlier point Fodor talks about, whether the investment society makes into collecting this data is worthwhile. I often get the feeling that much of scientific research is essentially poking in the dark in hopes of stumbling on something useful, relevant or helpful to society. Often scientific research is also motivated by many other factors as well (prestige, money, politics etc). The story of "The Little Black Girl in Search of God" I think sums up this idea well, if we intuitively know things (like the fact that dogs salivate when presented food) then why is a scientific understanding of the event necessary? How does it help society and is it worth the cost?
Economics aside, the question remains: can finding out where and when correlates occur help reverse engineer T3 capacity. It might, but are mirror neurons an example? If not, that is?
ReplyDeleteWhile Fodor is obviously a very intelligent person, I'm finding it difficult to understand why a philosopher is writing a critique of cognitive science methods. Not being able to come up with reasons for understanding brain networks and mirror neurons' role in imitation learning does not mean it's not worth while. It means you don't know enough about it.
ReplyDeleteAs he mentioned, in the event of brain surgery (as he drew a comparison with a car), it is important to know, which structures could affect what. Does this not merit investigation? I believe it does. Aside from changes to healthcare, understanding mirror neuron functioning could be important for child development and various other uses.
Furthermore, why were we put on this earth if not to learn. Knowledge of our world and its contents has grown exponentially. Why should we stop here and just sit to be content with what we already know. Yes I do know my brain is what does the thinking but being able to figure out which structures are implicated in which mechanisms will allow for reverse engineering of the brain.
''Furthermore, why were we put on this earth if not to learn...Why whould we stop here and just sit to be content with what we already know."
DeleteI dont think Fodor is in anyway saying we should stop knowing about the brain. He is saying that this type of neuroimaging is random, and without direction and blocks other inquiries into the brain and cognition which are more pointed and hypothesis based, rather than curiosity based ("I wonder what happens when we do this!"). Neuroimaging is like creating a map where the coordinate points consist of cognitive states (eg. thinking of a teapot, playing chess) and brain states. It reminds me of the attempts to sequence the genome, except for the fact that the human genome is finite while the amount of cognitive states is infinite. I agree with Fodor when he says," its the difference between a scientist who has a hypothesis and one who only has a camera." While making a map of the brain in this way would eventually lead us to some huge insights, doing so in a scattershot, random way, is completely useless. That being said, knowing where in the brain things are happening when we cognize is still informative in general, but is very misleading in telling us how we cognize. Neuroimaging will be useful once we have an idea of why we are trying to localise these functions. But taking snapshots of random cognitive functions until a picture appears is a waste of time.
"But why (unless you’re thinking of having it taken out) does it matter where in the engine the carburettor is"
ReplyDeleteI think Fodor's point is that all that Neural Imaging does is mapping function to a location in the brain. But mapping the location of something does not in any way explain that thing. There's no doubt that Neural Imaging has helped greatly in areas such as Clinical neurology, where you need to remove damaged brain tissues precisely without affecting other mental functionalities (think about HM). However, it did not do anything to help explaining how cognition works.
I would admit that, for cognitive scientists, where a function is mapped onto the brain is irrelevant. But the study of the way how brain is implemented (hardware) is important for cognitive science. Think about artificial neural network, if not for the study of the brain, we will not be able to build machines that can learn effectively. What computation does is that it helps cognitive science to test if an assumption is correct. For example, is the parallel processing ability at the center of human reasoning? And I think there's still a lot we don't know about the implementation about the brain that is important for the constructing of computation model to test cognition hypothesis.