From my perspective, Jerry Foder is somewhat too extreme or radical when he repudiated the usefulness of brain imaging tools. Surely we have not figured out the specific mechanisms underlying cognition, it is nonetheless unreasonable to deny the importance of exploring brain structures. In fact, it is essential to at least try to examine the brain mechanisms and neural networks in order to better understand both human cognition and artificial intelligence (like the verbal, robotic and neural indistinguishability of T4).
Yeh. I agree with you and I also feel Fodor is a bit too extreme in judging neuroimaging. I do believe that, even if neuroimaging does not develop into something that enables us to figure out how we cognize and understand things, it let us know which parts of brain areas are correlated, or act together. It could (maybe) even bring us to create a successful, reverse-engineered, AI that passes T4 one day. Neuroimaging is also a good way to locate brain areas that are being activated when we do a task. Somehow it does bring us hints on how the brain works (not able to show us the exact brain pathway yet, but at least we are able to know which bunch of cells are activating so the uncertainty is somehow reduced!)
Peihong, Zhao, it's good to be optimistic and open-minded about any potential help in passing T3 or T4 (which includes T3). But the question to keep in mind is: Have data on correlations between brain activity and cognitive performance given us a clue as to how to generate (cause) the performance capacity? Because until then they're no more informative than a horoscope. (The informativeness of brain data for clinical purposes, and in the reverse-engineering of vegetative functions -- like breathing, temperature, balance, blood-sugar, heart-rate -- is not in question. But the goal in cognitive science is to reverse-engineer cognitive function.)
Thank you for your reply professor. I guess the brain imaging tools can still be useful if they can be applied to test the specific algorithms going on (thus testing out the equivalency of the candidates). Like if the same areas are activated during certain activities or certain neurotransmitters are released under certain stimulation.
As Professor Harnad pointed out, the data on correlations between brain activity and cognitive performance have thus far almost wholly failed to elucidate any causal mechanisms of cognitive function. Even just stimulating a specific brain region in a lab fails to account for all possible external stimuli that could cause the signaling cascade in question from being triggered and resulting in some cognitive function X. From previous classes that I've taken on behavioral neuroscience, we have heaps of data on the release of neurotransmitters such as oxytocin and serotonin in the context of motivation, but that information doesn't quite specify how performance capacity is exactly generated.
"It occurs to me that maybe we’re heavily invested in finding answers to which we don’t know the corresponding questions. Maybe the availability of the new technology is running the science rather than the other way round."
Are these corresponding questions about the hard problem? Is there a disregard or an ignorance about the subtleties of the hard and easy problems in the field of neuroscience?
I think a lot of neuroscience, especially the highly medicalized physiology aspects of the field, don’t necessarily have to consider these problems. I can see how it could be easy to go from studying the PNS to CNS to suddenly wind up at functions/cognition level and still be in the strictly physiology mindset – but the cognition is more than just the ‘hardware’ or the physiology. I think Fodor is right to point out that we don’t know the corresponding questions, but how could we ever know them if we are stuck with the hard and easy problems?
vɪktoɹiə, aren't questions like "How do we recognize objects? how do we learn categories? how do we learn language? how do we reason?" the right questions? What Fodor is asking here is whether correlations with neural structure and function can answer them.
I liked Fodor’s retelling of Shaw’s fable about the little girl who runs into Pavlov, who explains that the purpose of his research is to prove already evident facts “scientifically”. It’s an interesting point that I’d never really considered in the context of basic neuroscience research. I definitely agree that current researchers seem to put an inordinate amount of prestige/value/grant money into studies that correlate localized brain activity to specific actions – as Fodor points out, why does it matter where we have a “thinking about teapots” network, if we don’t know how? I agree that the how, the underlying causal mechanism of this activity, should be more of a focus than the current preoccupation with the where of this activity.
The author challenged on researchers wasting time and money to look for the neural localization of mental functions. I personally think it is just his resentment on having no clue where to start in knowing cognition/mind. At the last paragraph, “…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.” He showed a lot of disdain towards brain imaging research. Yet, we have not proved anything wrong about using neuroimaging. Cognition is related, or needed, in many tasks we can do. He might found the all-around researching in brain science through neuroimaging a waste of resources, but what if all the directions, all the research and each of the experimental tasks are needed to build up, to draw a conclusion of how our brain works? Time will prove how right or wrong we are, but right now, I don’t think he is fair enough in judging neuroimaging research in such a way though.
There have been voices challenging the research methods. For example, the use of animals. Researchers have been using rats to draw conclusions about our genes, inheritance, and pain etc. I found the use of animals to understand humans no closer than using neuroimaging to understand cognition and to try locate our mind. In pain research, fMRI is used to locate brain areas that are activated when subjects experience pain. Even though it does not give us any causal conclusion, fMRI did bring us the “pain matrix” which are the regions that activate when in pain. And what other ways researchers have been trying, tissue damage, cardiovascular measurement, stress-induction and chemicals are neither the biomarkers of pain. (And if we haven’t tried these, we will never know they aren’t working!) However, whether neuroimaging can bring us to understand how the pain works is still an open question. Pain could be barely related to cognition, but what I am trying to suggest is that I do believe we still have some hope in knowing cognition by neuroimaging. It may not be the way like what we start from the Turing Test and thinking, but it could be something that help us to merge the brain areas we found related to cognition, to the cognition we know from TT and AI(maybe). And especially there have been so much advancement in neuroimaging methods these few years, I foresee that very soon, we can have better ways to investigate what is happening inside the brain. And if unfortunately, the neuroimaging is totally a wrong way to see cognition, I guess the stupid waste of money won’t last long.
Alison, Fodor wrote his paper almost 20 years ago. Is it less true now than then? (Pain is mostly a clinical problem, and mostly a vegetative function, and no one doubts brain imagery is useful for that. And I certainly don't doubt that non-invasive brain imagery is much more welcome than hurting animals. But the question remains open as to whether correlations with brain activity help us give a causal explanation of cognitive capacity and performance.)
I have mixed feelings about Fodor…. On one hand I realize how important it is to criticize neuroscience, especially since policy makers are increasingly using its evidence to inform various public policy and health practices. Criticism is also necessary to get people to critically think about whether the treatment, advice, facts…etc they are prescribed are of value, or are complete myths that are a result of neuro-hype. At the same time, I think Fodor goes to far in his criticism, and unjustly dismisses the value of mapping the brain. I agree that brain mapping does little to *explain* cognition -- especially since the data we get from brain imaging techniques is all correlational and largely unclear as to what it even means -- but I DONT think that means this type of research is a waste of time. Knowing where mental functions happen in the brain can have great clinical value, especially because of therapies that act on particular brain areas, like non-invasive procedures. For example, TMS stimulates nerves in the brain and has been shown to have clinical use in psychiatric disorders like depression. I think Fodor makes a silly claim when he states, "And serendipity is a frail reed; 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.” This research strategy defines science, not just neuroscience. The amount of of seemingly obscure research that makes a huge impact is not a secret, and if anything this notion is not very relevant to brain localization research because of just how interconnected and ambiguous the brain is; much more investigation is necessary to understand how neural systems interact with each other. Although observing the locality of mental function in the brain might not say anything for cognition in terms of the causes, reasons, and relationships of psychological functions, this type of research can contribute to our ability to interfere with certain psychological functions, which has clinical and medical value. Instead of dismissing the research, it should be contextualized, and directed towards experiments that have potential practical treatment applications.
Fodor's article was quite polemic and interesting to read both his article and his exchange with other scholars. I do agree with Fodor to certain extent that by studying the brain excessively will not lead to an extensive understand of how the mind works. Perhaps, as we have discussed in class, it can shed light to artificial intelligent models such as the neural networks and bring us closer in reverse engineering T3 robots and study cognition in this manner. Furthermore, I want to talk about Fodor’s argument on the obsession of popular media and the general public on the findings of functional location and the implications of correlational data.
Perhaps beyond medical clinical benefits, these correlational findings in functional neuroimaging will not lead us far in studying cognition. We discussed in class and the importance of correlation and causation: it is easy to demonstrate correlation but almost seemingly impossible to prove causality. I guess from a scientific point, there are set steps to establish causation: first is the predication and hypothesis of an event under certain circumstances, then the explanation and speculation of intermediate state that can result in the outcome. Finally would be the planning for the anticipated events and observe the final outcome, if it is wrong, then reject the original causation hypothesis to find alternative ones.
So from Fodor’s paper, it seems that to go from the what/where to how/why is seemingly difficult: such that correlation could only provide a speculation for a hypothesis that is needed to be further decided if it is a causation. So in the end, to find causality, we would need to pursue to determine the intermediate state of cognition, in this case, reverse engineer to pass the TT.
“Very well, then: just what is the question about the mind-brain relation in general, or about language in particular, that turns on where the brain’s linguistic capacities are? And if, as I suspect, none does, why are we spending so much time and money trying to find them?”
I am in full agreement with Fodor on the authority of neuroscience and therefore the validity of its findings. This article is reminiscent of Husserl’s book, The Crisis of European Sciences and Transcendental Phenomenology. Fodor points out the very problem of the computational model of the mind; the very phenomenon of settling on the computational model of the mind. It still is not true that this is the model to pursue, yet our public funding is squandered on the blind race for publication. The cause for the lack of the birth of Giants in neuroscience may be attributable to the logical fallacies with which the current neuroscientists are displaying. I dare say that this article should be a required reading for all Cognitive Science students at McGill.
Neuroimaging is a powerful tool that allows the mapping out of the human cortex, finally giving us a way of connecting years of molecular and functional organization research of animal brains and now being able to finally see how these translate or correspond to human brains. By doing extensive research into the mapping of brain regions, we can now look to see what are the effects when one part of this highly organized mapping is disrupted and how this might be giving rise to cognitive disorders. The potential of the tool is much more powerful than “you never can tell, it might pan out.” For example, brain topographical organization is found consistently throughout the brain, and there is much to suggest that the specific organization of the brain is not just randomly associating or localizing things together such as “taking naps” alongside “thinking of teapots,” but playing a foundational role in cognitive processing. These neural maps also have huge implications and may reveal a neural substrate for computing one-to-one mappings, which would give great insight into these types of computations and how they can be modeled and altered. I can understand Fodor’s frustration in researchers just producing data for the sake of generating data that is more likely to be published because of the popularity of the technique, but to just rebuke the immense tool of brain imaging and cognitive mapping altogether is an over-exaggeration.
In regards to cognitive research, I agree that brain localization research alone will not tell us how cognitive capacity is generated. However, I believe that in his argument, Fodor seriously overreaches and fails to provide a reasonable alternative. By no means is neuroimaging an end-all solution to the study of cognition, but I would argue that it certainly has the potential to lead us to a closer approximation of the mind; that is, a means by which we may come closer to explaining questions of “how/why”. Also, I think that Fodor’s brain/engine analogy completely weakens his argument by trivializing the broader implications of cognitive research. Is it really fair to compare the mechanics of a motor vehicle, which has been forward-engineered, to an incredibly complex and poorly understood biological phenomenon that we are trying to reverse-engineer?
Teresa, I am in complete agreement in your comments as to Fodor's weakness in the motor vehicle and brain comparison. In addition to your comments, I think he fails to see the broader reaching MEDICAL implications of cognitive research. Not only are we attempting to create a map of the brain, but were hoping along the way that this map reveals something about brain disorders and diseases. There is no doubt that neuroimaging has made profound contributions to the fields of neuroscience, and in no way is its sole mission to understand cognition.
All these observations and experiments on the brain help drawing a functional map of the brain. I understand that imagery has been overused sometimes without a concrete purpose, but in the end what we have, while it may not be an explanation for any of the problems about the mind, is a powerful tool for these aims. A map of the brain is simply a map: looking at a map won’t tell you what you really want about what it maps, but it will definitely guide you through it, and provide landmarks to find your way. If we shift the understanding of brain processes from being the end to bein the mean, then there is no problem in trying to make it more and more accurate and rich.
I guess the brain imaging tools are not only making a map? At least we have the brain divided into different areas with different functions thanks to the invention of brain imaging methods. So we are one step closer to testing out the equivalence of processes of generating outputs from inputs?
I think it would be premature to dismiss understanding HOW mental functions happen by first understanding WHERE they happen because we cannot say with certainty that understanding the WHERE will shed no light on the HOW. Nevertheless, I agree with Fodor in that investigating neural imagery should have more direction instead of using it like a new toy. Nevertheless, further exploration into neural imagery might at least tell us that the WHERE is not important to the HOW with more certainty, which brings us closer to understanding the HOW. At best it can narrow down where we should look for the HOW. Furthermore, neural imagery still retains its medical usefulness. However, I still appreciate Jerry Fodor's entry because it makes us question our purposes and methods in neural imagery a bit more diligently.
I found this article made very good points about the flaws of a lot of neuroimaging studies, however, i have to admit that it does not seem to contain any concrete solution. It is true that there seems to be an obsession with the localization of mental functions that is reminiscent of pseudosciences like phrenology. The frenetic search for differences can also be seen in comparisons between different brain locations (I'm thinking of all of the exaggerated claims about the right/left hemisphere), different populations (sexual dimorphisms are frequently reported, but it turns out that in many cases this is due to a flawed design, etc), all of this partly attributable to the fact that only hypothesis-driven studies that can be easily vulgarized/publicized, and in which "significance" is found receives attention and money. Fodor seems almost angry at scientists in neuroimaging, but to be fair this concerns all branches of research, and it just so happens that in this discipline it is easier to get the public's attention. Maybe Fodor meant to criticize the non-scientific community when he called the Times a "closet dualist".
In response to the idea that people who have high general intelligence should be good at multiple things simultaneously and that learning Latin can give rise to other capacities, I’m not too sure that I agree exactly with what Fodor is saying. Perhaps I don’t clearly understand what he is suggesting, but in my opinion, someone with general intelligence is someone who can easily understand concepts and maybe connect ideas better than the norm. I don’t think that it means that they have to be an expert in all fields. And in response to the learning different disciplines and being able to apply that to other fields, I think that is feasible. You might be able to find that your lesson in historical linguistics could help you understand why certain buildings may be built a certain way (far-fetched, but it might be possible).
Fodor terribly underestimates the power and implications of findings out what part of our brain is responsible for what. It is the first step in finding out how we do things and maybe why. What else can we possibly do to come closer to the question of how of the brain and cognition, then to start from the building blocks. Those parts of the brain that light up when we do a task can tell us what actions it is responsible for, although it might have been found accidently. It is the first step in finding out a part of who we are and how we do something that that makes us so distinct and enigmatic. Apart from understanding our kind, and especially how cognition works better, the results from these researches have a lot of clinical implications as well. It helps with many treatments. I feel like Fodor tried to make something look very trivial without even researching and backing his theory up with enough arguments.
Fodor makes a good point in arguing that data showing patterns of when and where things happen in the brain is not sufficient to explain how or why we are able have cognitive capacities. That being said, I also think that he oversimplifies the research that is being done with neuroimaging to make his point. Neural activity recorded when thinking of teapots is induced is not only about finding “where [or when] in that brain its thinking about teapots happens”, but about trying to understand how different brain regions interact (and communicate) with one another when presented with a sensorimotor experience. I agree that the technology we have available today is not powerful enough to explain the function of certain brain regions (I have hopes that some day it will), but maybe investigating these regions as part of neural networks can shed light on how cognition is implemented in the brain.
RE: It occurs to me that maybe we’re heavily invested in finding answers to which we don’t know the corresponding questions. Maybe the availability of the new technology is running the science rather than the other way round.
Although I disagree with Fodor’s dramatic conclusion in dismissing brain imaging altogether, he brings up a good point about the motivations that drive science today. I appreciate his discussion on the competitive nature of research funding, and suggests that money is a powerful motivator for science. In order to get funding, researchers must be able to produce not only credible but interesting findings about their topic of research. Today, the sexy topic of cognition is often explained in neuroscientific terms allowing for a moment of neuro-enchantment where the public stops thinking critically about the information that is being presented due to an awe produced by the fMRI picture. Of course Times Magazine is interested in presenting these discoveries to their audience, just as public funding is generally given to researchers who can produce these findings. The problem with this cycle, as Fodor points out, is that “the new technology is running the science”, suggesting that the findings are not answering the scientific questions but the public’s demand for charming phenomena about the human brain.
From my perspective, Jerry Foder is somewhat too extreme or radical when he repudiated the usefulness of brain imaging tools. Surely we have not figured out the specific mechanisms underlying cognition, it is nonetheless unreasonable to deny the importance of exploring brain structures. In fact, it is essential to at least try to examine the brain mechanisms and neural networks in order to better understand both human cognition and artificial intelligence (like the verbal, robotic and neural indistinguishability of T4).
ReplyDeleteYeh. I agree with you and I also feel Fodor is a bit too extreme in judging neuroimaging. I do believe that, even if neuroimaging does not develop into something that enables us to figure out how we cognize and understand things, it let us know which parts of brain areas are correlated, or act together. It could (maybe) even bring us to create a successful, reverse-engineered, AI that passes T4 one day. Neuroimaging is also a good way to locate brain areas that are being activated when we do a task. Somehow it does bring us hints on how the brain works (not able to show us the exact brain pathway yet, but at least we are able to know which bunch of cells are activating so the uncertainty is somehow reduced!)
DeletePeihong, Zhao, it's good to be optimistic and open-minded about any potential help in passing T3 or T4 (which includes T3). But the question to keep in mind is: Have data on correlations between brain activity and cognitive performance given us a clue as to how to generate (cause) the performance capacity? Because until then they're no more informative than a horoscope. (The informativeness of brain data for clinical purposes, and in the reverse-engineering of vegetative functions -- like breathing, temperature, balance, blood-sugar, heart-rate -- is not in question. But the goal in cognitive science is to reverse-engineer cognitive function.)
DeleteThank you for your reply professor. I guess the brain imaging tools can still be useful if they can be applied to test the specific algorithms going on (thus testing out the equivalency of the candidates). Like if the same areas are activated during certain activities or certain neurotransmitters are released under certain stimulation.
DeleteHi Alison and Zhao,
DeleteAs Professor Harnad pointed out, the data on correlations between brain activity and cognitive performance have thus far almost wholly failed to elucidate any causal mechanisms of cognitive function. Even just stimulating a specific brain region in a lab fails to account for all possible external stimuli that could cause the signaling cascade in question from being triggered and resulting in some cognitive function X. From previous classes that I've taken on behavioral neuroscience, we have heaps of data on the release of neurotransmitters such as oxytocin and serotonin in the context of motivation, but that information doesn't quite specify how performance capacity is exactly generated.
"It occurs to me that maybe we’re heavily invested in finding answers to which we don’t know the corresponding questions. Maybe the availability of the new technology is running the science rather than the other way round."
ReplyDeleteAre these corresponding questions about the hard problem? Is there a disregard or an ignorance about the subtleties of the hard and easy problems in the field of neuroscience?
I think a lot of neuroscience, especially the highly medicalized physiology aspects of the field, don’t necessarily have to consider these problems. I can see how it could be easy to go from studying the PNS to CNS to suddenly wind up at functions/cognition level and still be in the strictly physiology mindset – but the cognition is more than just the ‘hardware’ or the physiology. I think Fodor is right to point out that we don’t know the corresponding questions, but how could we ever know them if we are stuck with the hard and easy problems?
vɪktoɹiə, aren't questions like "How do we recognize objects? how do we learn categories? how do we learn language? how do we reason?" the right questions? What Fodor is asking here is whether correlations with neural structure and function can answer them.
DeleteI liked Fodor’s retelling of Shaw’s fable about the little girl who runs into Pavlov, who explains that the purpose of his research is to prove already evident facts “scientifically”. It’s an interesting point that I’d never really considered in the context of basic neuroscience research. I definitely agree that current researchers seem to put an inordinate amount of prestige/value/grant money into studies that correlate localized brain activity to specific actions – as Fodor points out, why does it matter where we have a “thinking about teapots” network, if we don’t know how? I agree that the how, the underlying causal mechanism of this activity, should be more of a focus than the current preoccupation with the where of this activity.
ReplyDeleteOlivia, so the open question remains whether we can get to from when and where to how and why.
DeleteThe author challenged on researchers wasting time and money to look for the neural localization of mental functions. I personally think it is just his resentment on having no clue where to start in knowing cognition/mind. At the last paragraph, “…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.” He showed a lot of disdain towards brain imaging research. Yet, we have not proved anything wrong about using neuroimaging. Cognition is related, or needed, in many tasks we can do. He might found the all-around researching in brain science through neuroimaging a waste of resources, but what if all the directions, all the research and each of the experimental tasks are needed to build up, to draw a conclusion of how our brain works? Time will prove how right or wrong we are, but right now, I don’t think he is fair enough in judging neuroimaging research in such a way though.
ReplyDeleteThere have been voices challenging the research methods. For example, the use of animals. Researchers have been using rats to draw conclusions about our genes, inheritance, and pain etc. I found the use of animals to understand humans no closer than using neuroimaging to understand cognition and to try locate our mind. In pain research, fMRI is used to locate brain areas that are activated when subjects experience pain. Even though it does not give us any causal conclusion, fMRI did bring us the “pain matrix” which are the regions that activate when in pain. And what other ways researchers have been trying, tissue damage, cardiovascular measurement, stress-induction and chemicals are neither the biomarkers of pain. (And if we haven’t tried these, we will never know they aren’t working!) However, whether neuroimaging can bring us to understand how the pain works is still an open question. Pain could be barely related to cognition, but what I am trying to suggest is that I do believe we still have some hope in knowing cognition by neuroimaging. It may not be the way like what we start from the Turing Test and thinking, but it could be something that help us to merge the brain areas we found related to cognition, to the cognition we know from TT and AI(maybe). And especially there have been so much advancement in neuroimaging methods these few years, I foresee that very soon, we can have better ways to investigate what is happening inside the brain. And if unfortunately, the neuroimaging is totally a wrong way to see cognition, I guess the stupid waste of money won’t last long.
Alison, Fodor wrote his paper almost 20 years ago. Is it less true now than then? (Pain is mostly a clinical problem, and mostly a vegetative function, and no one doubts brain imagery is useful for that. And I certainly don't doubt that non-invasive brain imagery is much more welcome than hurting animals. But the question remains open as to whether correlations with brain activity help us give a causal explanation of cognitive capacity and performance.)
DeleteI have mixed feelings about Fodor…. On one hand I realize how important it is to criticize neuroscience, especially since policy makers are increasingly using its evidence to inform various public policy and health practices. Criticism is also necessary to get people to critically think about whether the treatment, advice, facts…etc they are prescribed are of value, or are complete myths that are a result of neuro-hype. At the same time, I think Fodor goes to far in his criticism, and unjustly dismisses the value of mapping the brain. I agree that brain mapping does little to *explain* cognition -- especially since the data we get from brain imaging techniques is all correlational and largely unclear as to what it even means -- but I DONT think that means this type of research is a waste of time. Knowing where mental functions happen in the brain can have great clinical value, especially because of therapies that act on particular brain areas, like non-invasive procedures. For example, TMS stimulates nerves in the brain and has been shown to have clinical use in psychiatric disorders like depression. I think Fodor makes a silly claim when he states, "And serendipity is a frail reed; 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.” This research strategy defines science, not just neuroscience. The amount of of seemingly obscure research that makes a huge impact is not a secret, and if anything this notion is not very relevant to brain localization research because of just how interconnected and ambiguous the brain is; much more investigation is necessary to understand how neural systems interact with each other. Although observing the locality of mental function in the brain might not say anything for cognition in terms of the causes, reasons, and relationships of psychological functions, this type of research can contribute to our ability to interfere with certain psychological functions, which has clinical and medical value. Instead of dismissing the research, it should be contextualized, and directed towards experiments that have potential practical treatment applications.
ReplyDeleteNancy, clinical usefulness is uncontested, Fodor is asking about explaining cognition.
DeleteFodor's article was quite polemic and interesting to read both his article and his exchange with other scholars. I do agree with Fodor to certain extent that by studying the brain excessively will not lead to an extensive understand of how the mind works. Perhaps, as we have discussed in class, it can shed light to artificial intelligent models such as the neural networks and bring us closer in reverse engineering T3 robots and study cognition in this manner. Furthermore, I want to talk about Fodor’s argument on the obsession of popular media and the general public on the findings of functional location and the implications of correlational data.
ReplyDeletePerhaps beyond medical clinical benefits, these correlational findings in functional neuroimaging will not lead us far in studying cognition. We discussed in class and the importance of correlation and causation: it is easy to demonstrate correlation but almost seemingly impossible to prove causality. I guess from a scientific point, there are set steps to establish causation: first is the predication and hypothesis of an event under certain circumstances, then the explanation and speculation of intermediate state that can result in the outcome. Finally would be the planning for the anticipated events and observe the final outcome, if it is wrong, then reject the original causation hypothesis to find alternative ones.
So from Fodor’s paper, it seems that to go from the what/where to how/why is seemingly difficult: such that correlation could only provide a speculation for a hypothesis that is needed to be further decided if it is a causation. So in the end, to find causality, we would need to pursue to determine the intermediate state of cognition, in this case, reverse engineer to pass the TT.
“Very well, then: just what is the question about the mind-brain relation in general, or about language in particular, that turns on where the brain’s linguistic capacities are? And if, as I suspect, none does, why are we spending so much time and money trying to find them?”
ReplyDeleteI am in full agreement with Fodor on the authority of neuroscience and therefore the validity of its findings. This article is reminiscent of Husserl’s book, The Crisis of European Sciences and Transcendental Phenomenology. Fodor points out the very problem of the computational model of the mind; the very phenomenon of settling on the computational model of the mind. It still is not true that this is the model to pursue, yet our public funding is squandered on the blind race for publication. The cause for the lack of the birth of Giants in neuroscience may be attributable to the logical fallacies with which the current neuroscientists are displaying. I dare say that this article should be a required reading for all Cognitive Science students at McGill.
Neuroimaging is a powerful tool that allows the mapping out of the human cortex, finally giving us a way of connecting years of molecular and functional organization research of animal brains and now being able to finally see how these translate or correspond to human brains. By doing extensive research into the mapping of brain regions, we can now look to see what are the effects when one part of this highly organized mapping is disrupted and how this might be giving rise to cognitive disorders. The potential of the tool is much more powerful than “you never can tell, it might pan out.” For example, brain topographical organization is found consistently throughout the brain, and there is much to suggest that the specific organization of the brain is not just randomly associating or localizing things together such as “taking naps” alongside “thinking of teapots,” but playing a foundational role in cognitive processing. These neural maps also have huge implications and may reveal a neural substrate for computing one-to-one mappings, which would give great insight into these types of computations and how they can be modeled and altered. I can understand Fodor’s frustration in researchers just producing data for the sake of generating data that is more likely to be published because of the popularity of the technique, but to just rebuke the immense tool of brain imaging and cognitive mapping altogether is an over-exaggeration.
ReplyDeleteIn regards to cognitive research, I agree that brain localization research alone will not tell us how cognitive capacity is generated. However, I believe that in his argument, Fodor seriously overreaches and fails to provide a reasonable alternative. By no means is neuroimaging an end-all solution to the study of cognition, but I would argue that it certainly has the potential to lead us to a closer approximation of the mind; that is, a means by which we may come closer to explaining questions of “how/why”. Also, I think that Fodor’s brain/engine analogy completely weakens his argument by trivializing the broader implications of cognitive research. Is it really fair to compare the mechanics of a motor vehicle, which has been forward-engineered, to an incredibly complex and poorly understood biological phenomenon that we are trying to reverse-engineer?
ReplyDeleteTeresa, I am in complete agreement in your comments as to Fodor's weakness in the motor vehicle and brain comparison. In addition to your comments, I think he fails to see the broader reaching MEDICAL implications of cognitive research. Not only are we attempting to create a map of the brain, but were hoping along the way that this map reveals something about brain disorders and diseases. There is no doubt that neuroimaging has made profound contributions to the fields of neuroscience, and in no way is its sole mission to understand cognition.
DeleteAll these observations and experiments on the brain help drawing a functional map of the brain. I understand that imagery has been overused sometimes without a concrete purpose, but in the end what we have, while it may not be an explanation for any of the problems about the mind, is a powerful tool for these aims. A map of the brain is simply a map: looking at a map won’t tell you what you really want about what it maps, but it will definitely guide you through it, and provide landmarks to find your way. If we shift the understanding of brain processes from being the end to bein the mean, then there is no problem in trying to make it more and more accurate and rich.
ReplyDeleteI guess the brain imaging tools are not only making a map? At least we have the brain divided into different areas with different functions thanks to the invention of brain imaging methods. So we are one step closer to testing out the equivalence of processes of generating outputs from inputs?
DeleteI think it would be premature to dismiss understanding HOW mental functions happen by first understanding WHERE they happen because we cannot say with certainty that understanding the WHERE will shed no light on the HOW. Nevertheless, I agree with Fodor in that investigating neural imagery should have more direction instead of using it like a new toy. Nevertheless, further exploration into neural imagery might at least tell us that the WHERE is not important to the HOW with more certainty, which brings us closer to understanding the HOW. At best it can narrow down where we should look for the HOW. Furthermore, neural imagery still retains its medical usefulness. However, I still appreciate Jerry Fodor's entry because it makes us question our purposes and methods in neural imagery a bit more diligently.
ReplyDeleteI found this article made very good points about the flaws of a lot of neuroimaging studies, however, i have to admit that it does not seem to contain any concrete solution. It is true that there seems to be an obsession with the localization of mental functions that is reminiscent of pseudosciences like phrenology. The frenetic search for differences can also be seen in comparisons between different brain locations (I'm thinking of all of the exaggerated claims about the right/left hemisphere), different populations (sexual dimorphisms are frequently reported, but it turns out that in many cases this is due to a flawed design, etc), all of this partly attributable to the fact that only hypothesis-driven studies that can be easily vulgarized/publicized, and in which "significance" is found receives attention and money. Fodor seems almost angry at scientists in neuroimaging, but to be fair this concerns all branches of research, and it just so happens that in this discipline it is easier to get the public's attention. Maybe Fodor meant to criticize the non-scientific community when he called the Times a "closet dualist".
ReplyDeleteIn response to the idea that people who have high general intelligence should be good at multiple things simultaneously and that learning Latin can give rise to other capacities, I’m not too sure that I agree exactly with what Fodor is saying. Perhaps I don’t clearly understand what he is suggesting, but in my opinion, someone with general intelligence is someone who can easily understand concepts and maybe connect ideas better than the norm. I don’t think that it means that they have to be an expert in all fields. And in response to the learning different disciplines and being able to apply that to other fields, I think that is feasible. You might be able to find that your lesson in historical linguistics could help you understand why certain buildings may be built a certain way (far-fetched, but it might be possible).
ReplyDeleteFodor terribly underestimates the power and implications of findings out what part of our brain is responsible for what. It is the first step in finding out how we do things and maybe why. What else can we possibly do to come closer to the question of how of the brain and cognition, then to start from the building blocks. Those parts of the brain that light up when we do a task can tell us what actions it is responsible for, although it might have been found accidently. It is the first step in finding out a part of who we are and how we do something that that makes us so distinct and enigmatic. Apart from understanding our kind, and especially how cognition works better, the results from these researches have a lot of clinical implications as well. It helps with many treatments. I feel like Fodor tried to make something look very trivial without even researching and backing his theory up with enough arguments.
ReplyDeleteFodor makes a good point in arguing that data showing patterns of when and where things happen in the brain is not sufficient to explain how or why we are able have cognitive capacities. That being said, I also think that he oversimplifies the research that is being done with neuroimaging to make his point. Neural activity recorded when thinking of teapots is induced is not only about finding “where [or when] in that brain its thinking about teapots happens”, but about trying to understand how different brain regions interact (and communicate) with one another when presented with a sensorimotor experience. I agree that the technology we have available today is not powerful enough to explain the function of certain brain regions (I have hopes that some day it will), but maybe investigating these regions as part of neural networks can shed light on how cognition is implemented in the brain.
ReplyDeleteRE: It occurs to me that maybe we’re heavily invested in finding answers to which we don’t know the corresponding questions. Maybe the availability of the new technology is running the science rather than the other way round.
Although I disagree with Fodor’s dramatic conclusion in dismissing brain imaging altogether, he brings up a good point about the motivations that drive science today. I appreciate his discussion on the competitive nature of research funding, and suggests that money is a powerful motivator for science. In order to get funding, researchers must be able to produce not only credible but interesting findings about their topic of research.
Today, the sexy topic of cognition is often explained in neuroscientific terms allowing for a moment of neuro-enchantment where the public stops thinking critically about the information that is being presented due to an awe produced by the fMRI picture. Of course Times Magazine is interested in presenting these discoveries to their audience, just as public funding is generally given to researchers who can produce these findings. The problem with this cycle, as Fodor points out, is that “the new technology is running the science”, suggesting that the findings are not answering the scientific questions but the public’s demand for charming phenomena about the human brain.