Clark, A. & Chalmers, D. (1998) The Extended Mind. Analysis. 58(1)
Where does the mind stop and the rest of the world begin? The question invites two standard replies. Some accept the demarcations of skin and skull, and say that what is outside the body is outside the mind. Others are impressed by arguments suggesting that the meaning of our words "just ain't in the head", and hold that this externalism about meaning carries over into an externalism about mind. We propose to pursue a third position. We advocate a very different sort of externalism: an active externalism, based on the active role of the environment in driving cognitive processes.
Subscribe to:
Post Comments (Atom)
-
Harnad, S. (2003b) Categorical Perception . Encyclopedia of Cognitive Science . Nature Publishing Group. Macmillan. http://www.youtube.com...
-
Harnad, S. (2003) The Symbol Grounding Problem . Encylopedia of Cognitive Science. Nature Publishing Group. Macmillan. or: Harnad, S...
-
Pinker, S. & Bloom, P. (1990). Natural language and natural selection . Behavioral and Brain Sciences 13(4): 707-784. Many people hav...
I think that the article makes a pretty clear case for the fact that the mind can extend beyond an individual’s head through various mechanisms but what I think is most interesting about this process is that we are able to actually make it happen. If we were to imagine a time before language, before any aids like calculators or notepads, when cognition was a lot more restricted to the head than it is now, it is quite exceptional that we were able to actually extend our cognition beyond the head in the first place. True it is highly adaptive to have active externalization but where did it actually come from and how could it have possibly developed? Interacting with the world around us is the most basic way for us to extend our cognition beyond the mind so maybe this is where it originally came from? This ability to interact with the environment is the method by which symbol grounding via sensorimotor categories can be achieved which later turns into our capacity for category learning via hearsay (ie: through language). Thus, active externalization is an essential element in the evolution of cognition.
ReplyDeleteChalmers touched on this a bit implicitly. It may be a possibility that the brain has evolved in such a way that factored in the "reliable presence of a manipulable external environment." We know that evolution is lazy, so it may have exploited the environment as being within the cognitive loop, instead of the tradition idea that cognition is contained in the head. The brain complements the cognitive extensions, instead of it being separate processes that merely interact. I don't know the answer to your question about where it originally come from, but I think an evolutionary exploration is a plausible avenue of research in understanding the mechanisms of active externalism.
DeleteMaria, sensorimotor organs, social communication, oral language, written language, libraries, etc. all extend our cognitive capacities. But mental states are felt states. And felt states are generated by our brains. Our brains are doing the feeling, not google.
DeleteAmber, organisms' bodies were interacting with the outside worlds before they had nervous systems or feelings. Plants and microbe interact with the world without feeling (I hope, veganly!). And so do (today's) toy robots (not Isaure!).
“many identify the cognitive with the conscious, and it seems far from plausible that consciousness extends outside the head in these cases. But not every cognitive process, at least on standard usage, is a conscious process.”
ReplyDelete“when it comes to belief, there is nothing sacred about skull and skin. What makes some information count as a belief is the role it plays, and there is no reason why the relevant role can be played only from inside the body.”
I’m not sure if this catches the main point of the article, but I really appreciated it as it sheds light on something that makes absolute sense, but which I had never thought about before. It makes perfect sense that cognition involves environment, as this is a significant part of what we cognize about. If we are exposed to zero sensory stimulation (which is mainly what the environment offers as cognitive input), we don’t have much to think about. Going further than the case of Otto, who’s cognition literally relies upon external stimuli (i.e. words in the notebook), we all necessitate some sort of input external to ourselves in order to process and make something of it. Being in a vacuum (space, or just a pitch-black sound-free room) gives us nothing to think about (unless we have been exposed to environment before). Environment allows us to form memories, relate experiences to places, and so on. It is therefore crucial that we acknowledge that cognition goes beyond our internal mental and physical selves.
"Cognitive" is a weasel word here. It need not mean "mental" (and C & C are talking about extending the mind, i.e. the mental). Mental states are felt states. Inputs to whatever mechanism generates felt states (hard problem) are not part of the felt state itself. Can the parts of the state that is being felt be wider than the head? Not the inputs to it: the state itself.
DeleteA belief has to be felt. (I am right now in the state of believing that belief has to be felt.) If, instead, a belief is just information stored somewhere inside (or outside) my brain, but I am not right now believing it, then it is misleading (or too loose) to call it a "belief." Surely a belief has to be believed (just as a feeling has to be felt). I wouldn't talk about having a feeling when I'm not having it. (I could, however, say that whenever I see a pig, it makes me feel sorry. Just as I could say that although I was not believing that 2 + 2 = 4 a moment ago, I wasn't think about it at all, but, now that you mention it, I do believe that 2 + 2 = 4. I am believing it now. I was not believing or disbelieving it a moment ago: I wasn't thinking about it at all.)
Yes, feeling organisms, with nervous systems, interact with the environment. So do non-feeling organisms, without nervous systems (like microbes or plants [I hope!]. And so do non-feeling non-organisms (like robots -- but maybe T3s like Isaure do feel too...).
Yes, organisms' environments are important, but they are not part of those of their states that are felt states (though as input they can influence those felt states).
While I understand what Chalmers is going for with the idea of extended mindedness, I think that he is trying to stretch the metaphor too far. I am willing to accept that another human who cognizes (we think) and helps you solve a problem can be incorporated into the overall thought process for problem solving. What I cannot understand is how Chalmers can include things which cannot (by nature) cognize in a thought process. In his Scrabble example, Chalmers suggests that the tiles are part of the thought process. How could a non-cognizing thing be part of a cognitive process? The way I understand it, the physical tiles create a visual percept. This percept can be changed in the head (move the "N" and the "M" to form the word "name") and that would create a new percept. But the actual cognizing and meaningful change of the tile order was in the head. The physical tiles just make it easier to visualize. Perhaps if Chalmers limited the extension of the mind to other (hopefully present) minds, the idea would be a bit less confusing.
ReplyDeleteI also thought that Clark & Chalmers were taking it too far, and in my opinion that's where their argument fell short. The authors seem to be claiming that anything that affords 'epistemic actions' which aid and augment cognitive processes is cognitive and therefore an extension of the mind. If I follow this criteria, I could argue that the entire world - including all of the people in this class and in my life - is an extension of my mind because it can aid my cognitive processes in some shape or form . Where does Clark and Chalmers draw the line? Consider that perhaps the authors are simply referring to general tools such as hearing aids. A hearing aid extends one of our sensorimotor capacities, but that does not mean that the aid itself is perceiving the sound. I think that epistemic actions are simply extensions of sensorimotor capacities, and not sufficient as evidence for extension of the mind.
DeleteI think the argument falls short even for other minds. In the case of two thinking individuals cognitively assisting each other, no matter how synchronized their thoughts are, they aren't part of the same thinking being. There's no "mind" that is made up of the sum of their thoughts. There's only what person A thinks, what person B thinks and what they say and do. If you and I solve a math problem together, I still don't know what's going on in your head and vice-versa.
DeleteZach, surely I am no more a part of your mind than your computer is. And when you are consciously thinking (Cogito) that state is a felt state, whereas whatever process retrieved the name of your 3rd grade school teacher, you did not do it by conscious thinking, so it is not part of your felt state, and hence it does not matter whether it happens inside your head (or outside, via google name search -- or asking a school-mate).
DeleteDevona, with hearing aids you anticipated what will be in 11b: cognitive technology. Hearing aids are not part of the felt state of hearing. They are just inputs to it.
Willem: spot-on.
It was interesting to read how Clark and Chalmers makes the comparison of Otto’s notebook and Inga’s biological memory. If these external extensions (such as Otto’s notebook) can be considered a part of our cognitive processes, I wonder where the distinction between what’s included and not included in our cognitive processes can be made. I could argue that my smartphone is part of my cognitive processes as it keeps store of my contacts and my conversations that I have with others, but there is something that doesn’t feel right when I make that statement. Also, in the future, would it be possible to offload these cognitive processes (such as having a biological memory) to save resources? How would it affect our evolution?
ReplyDelete“One reason many identify the cognitive with the conscious, and it seems plausible that consciousness extends outside the head in these cases. But not every cognitive process, at least on standard usage, is a conscious process. It is widely accepted that all sorts of processes beyond the borders of consciousness play a crucial role in cognitive processing: in memories, linguistic processes, and skill acquisition, for example.”
This passage answered the problem I had with the argument, which is how would consciousness tie in with the extended mind. I think that in this case, the extension of the mind would be limited because these extensions of our external environment cannot feel, and thus cannot be conscious. Only the cognitive processes that are not conscious would be able to have the extension.
When does a conscious state (i.e., a felt state) "extend outside the head?
DeleteI don't think that conscious states can extend outside the head because I believe that felt states are only mental states. If it were possible that conscious states could extend outside the head, then I would assume that the Hard Problem is solvable as consciousness could then be observed.
DeleteCeline, even if the physical substrate of a migraine were wider than a head, that still wouldn't explain how or why organisms feel. Whether the activities correlated with having a migraine were just in the head, or beyond the head, we're still where we were: No causal explanation of feeling; just (T4) correlates for predicting mental states (mind-reading).
Delete"It might be suggested that the cases are relevantly different in that Inga has more reliable access to the information. After all, someone might take away Otto's notebook at any time, but Inga's memory is safer. It is not implausible that constancy is relevant: indeed, the fact that Otto always uses his notebook played some role in our justifying its cognitive status. If Otto were consulting a guidebook as a one-off, we would be much less likely to ascribe him a standing belief. But in the original case, Otto's access to the notebook is very reliable - not perfectly reliable, to be sure, but then neither is Inga's access to her memory. A surgeon might tamper with her brain, or more mundanely, she might have too much to drink. The mere possibility of such tampering is not enough to deny her the belief."
ReplyDeleteTo reply to the comparison of notebook and memory- it’s not the same at all since the notebook is outside and fundamentally detachable. Yes Inga’s memory can be tampered with but alcohol and surgery are much more drastic forms of altering someone’s memory than simply taking something like a notebook away. In the case of Otto and his notebook it seems to be argued that anything he learns in a specific moment can be “integrated into his memory” simply because it’s written down in the notebook, but that’s not how a healthy person’s memory works. We hear a lot of things going through our lives that we don’t internalize and don’t become part of our memories. To me looking something up doesn’t mean that it’s in our heads. For example, if I once learned something but have since forgotten and then google it I’m relearning something that I’ve forgotten not prompting a memory (and maybe that is where I fundamentally disagree with this logic and can’t get on board with this line of thinking).
The is about the mind, hence mental states. Mental states -- the states that are felt -- can be narrower than the head. (There's plenty going on in my head that I can't feel.) But can they be wider?
DeleteThe paper argues that cognition can’t be explained solely by what goes on in the brain or the body, but that it necessarily extends to the environment. Before examining Clark and Chalmers’ reasoning behind this, it must be said that this only tries to explain doing, and not feeling. This paper attacks the easy problem (unsuccessfully in my opinion) and not the hard problem.
ReplyDeleteClark and Chalmers’ notion of an Extended Mind harks back to the Systems Reply. In essence, it argues that cognition doesn’t just go on in the head of the doer, but also in the ways the doer physically manipulates the environment, and is in turn influenced by the environment. The authors give numerous examples of ways in which we use physical tools to assist in cognitive processes (i.e. writing out the steps of a math problem as we work through it). The issues with this view boil down to one important difference between me and the paper I write out my thoughts on: I’m thinking, and the paper is not.
The authors conflate symbolically representing a belief with actually having one. What I write on the paper is just squiggles and squoggles until someone interprets them. It feels like something to believe that 2+2=4. I feel that, the paper I wrote it down on does not. The paper is not a part of me just because I wrote something I thought on it.
In the example of Otto and Inga, Clark and Chalmers argue that Inga, who consults her memory to remember where a museum is located, and Otto, who suffers from Alzheimer’s and must consult his notebook to know where the museum is, are basically equivalent. They both want to get to the same place and they both consult something to get there. It’s argued that any difference (i.e. that Otto is using a notebook) extending beyond these behaviours is pointless to talk about:
“We submit that to explain things this way is to take one step too many. It is pointlessly complex, in the same way that it would be pointlessly complex to explain Inga's actions in terms of beliefs about her memory.”
But explaining Inga’s behaviour in terms of her memory is exactly what we’re trying to do. You can’t claim to remember a telephone number if you can only relay it to someone after consulting a phonebook. If we want to explain Otto’s behaviour, and he uses the notebook as part of the behaviour, then we cannot conveniently omit that from the explanation. His notebook is a necessary part of the explanation, precisely because it’s NOT part of him and the cognizing HE does.
Otto’s notebook isn’t cognizing just because he writes things in it and reads from it, just as the combination of Searle and the pieces of paper he writes out Chinese computations on don’t mystically produce an understanding of Chinese.
Excellent analysis.
DeleteThis paper argues that the mind does not exist only inside of one's body but can be extended to the environment where the minds interact with external reality. In my opinion, what the author tries to argue here sounds like an overcomplicated symbol ground problem and sensorimotor categorization. Yes we need the physical world and our cognitive capacity to achieve categorization and a sense of meaning. If we live in a vacuum alone without any interaction with any of our sensors, we would not have any cognitive capacity. However, incorporate real world objects into our categorical perception and, assigning symbols that would be manipulated in the thinking process does not mean that the environment itself is part of our cognition. The thought process is in our brain, where the feelings of we are thinking is achieved. According to Decartes, the only thing we can be sure of is the fact that we feel. Therefore, in this sense, the things themselves are not important, the fact that we feel them, is crucial for cognition and the mind.
ReplyDeleteYes, the crux of it is the difference between states that are felt and states (whether internal or external) that are not.
DeleteBut symbols inside (and outside) Isaure's head even if Isaure were a T3 zombie.
("Stevan says" there are not T3 zombies. If Isaure is T3, she feels.)
I think maybe our cognitive process rely on the external world to be executed in some circumstances, like calculations, communicating, etc. however i think the root of this cognition is still internal to brain. These external “cognitions” that chalmer describes are simply external tools for internal cognitive process, however I do think most of these external process can be done without the use of the tools, the tools simply speed up the process a bit. For example, Chalmers gives an example of using scratch paper for calculations. I do agree that this is a tool for assisting the computational process, however I have a really hard time saying that the actual computation is occurring outside of the brain. I think there is a difference between assisting the cognitive process verse being the cognitive process and I view external tools as assisting.
ReplyDeleteCognitive states (as opposed to tools used by cognizers) are felt states. Can those be wider than the brain?
DeleteHardly. The only reason I think tools can assist cognitions whilst being external to the brain is precisely because these process are not felt states. Yes it feel like something to understand but most computational processes are not really felt states. However, felt states are entirely restricted to the brain in my opinion.
DeleteMy main issue with this paper is that it seems to give the same status to the interpretable as it does to the interpreter, and I'm not sure I buy it. In the MoMA example, they claim that the written address is just as much a belief as a memory of the address, however these concepts are not equivalent. Inga's memory of the location is a collection of all the information she has of it, it has already been grounded and connections have been made. Whereas Otto's notebook has no meaning until it is interpreted by Otto, it's just squiggles until someone comes along to make sense of it.
ReplyDeleteUnfelt states can be wider than the head, but unfelt states are just states, whether internal or external.
DeleteCan felt states be (all or part) external?
If we’re to replace the word “mind” with feeling (as has been the case in this course so far) this becomes a question of “the extended feeling”. For me, that makes the question easy to answer. No, there’s no extended feeling.
ReplyDeleteThe examples of how tools such as writing and google and language extend our doing capabilities are irrelevant to the question of extended feeling. They’d be relevant to the question of extended doing, but although one definitely feels what it’s like to do something, doing and feeling are two different topics.
The paper I write on is not feeling the ink I’m putting on it, google is not feeling me search on it, the acoustic sounds I’m creating when I speak aren’t feeling what the words I say mean.
We can only be sure of our own feelings (shoutout Descartes). We might have a sense of what someone else feels like based on how they describe their feelings to us through words, but it’s never possible to feel exactly what someone else feels. Thus, the idea that feeling could extend from one person to another person (or to a non- feeling item) doesn’t make sense in this framework.
I think there’s a lot of hype around an extended mind because feeling is an inherently lonely activity. No one will ever feel the same exact way I do, and likewise I’ll never feel the same exact way anyone else does. I think that’s why people bond so easily over feeling similar things - it makes them feel less alone.
spot on.
DeleteIn the beginning of the reading, the term ‘epistemic action’ is introduced. Epistemic actions are actions that are meant to alter the external world in order to increase our comprehension or knowledge. On the other hand, pragmatic actions are actions that aim to change the external world simply for the sake of changing it. For example, opening your curtains in the morning to know what the weather is like would be an epistemic action, but opening them to get some light in your room would be a pragmatic action. What I am confused about is the meaning of the “epistemic credit” and how it relates to an “extended cognition”. It seems to me like epistemic actions are not cognitive per se. The cognitive parts of the epistemic action would be the thoughts you have just before the action “Oh I wonder what the weather is like today” and the ones after you analyze the change “Great, it is sunny!”. I see how cognition is influenced by external processes, but cognition remains in the head despite the described differences in these two kinds of action.
ReplyDeleteI’m also wondering what this really means… For example, by extending our cognition is the idea that technology extends our cognitive capacities (the set of doing becomes larger)? Does it mean that it’s not just that we can do things quicker or with sustained memory by off-loading or utilizing the environment (things outside the brain).. but that we really can do more different things? So even something as simple as a pen and paper dramatically increases what cognizers can do… things that couldn’t be done without it (or similarly relevant tech). Though I’m not sure “it may be that the biological brain has in fact evolved and matured in ways which factor in the reliable presence of a manipulable external environment.” What was reliably manipulable in our environment in the evo past?
Delete"Epistemic" is a weasel word here: Is it felt states that are extended or just doing states (in which case cycling would be extended cognition...)
DeleteHugo, What was reliably manipulable in our environment ? Wood and stone to make tools for various use (hunting, skinning, cloth making etc..), fire, water, food. I do not think technology extends our cognitive capacities per say. With the example of the pen and paper, we can still write down words, or make mental calculations in our heads. What the pen and paper do allow us to do is to put it our there in the environment, it affords the possibility to write faster, to make it available to others for them to recognize and interpret what is written. Therefore, it extends our DOING capacities, but since cognition is more than doing, i'm not sure if we can say that it "extends cognition". This would imply that our felt, intentional states would be extended to the piece of paper. Yet, we know the symbols remain squiggles and sqaggles on paper until someone interprets them. How we felt while writing does not extend onto the paper.
ReplyDeleteTechnology and external tools make the cognitive processes faster, easier, or more efficient, maybe. However, while these tools are part of the cognitive process in that they assist and provide possibilities in the production of an output, they are in no way part of, containing, or causing felt states.
Cognition is what generates both our capacity to do and to feel. A crane or remote control warfare extend my capacity to do (easy problem), but they do not extend my "mind," because mental states are felt states, and all my feeling is done by my brain (somehow: hard problem). A telescope or hearing aid extends my senses, but again, they are not part of my "mind," because the feeling of what reaches my senses is again done by my brain (somehow), not the telescope or the hearing aid (nor even by a cochlear transplant); they are just inputs to the felt state, not parts of it.
DeleteBut be careful, symbol grounding, too, is just a doing capacity. T3 robots like Isaure are grounded whether or not they feel.
The example of Otto using his notebook to augment his memory reminded me of Thad Starner, a professor at Georgia Institute of Technology who has been wearing a computer as a memory tool since 1993. He has a keyboard in his hand and a small LED screen clipped onto one of the lenses of his glasses. Starner first started doing this as an undergraduate student at MIT because he found that it was difficult to both listen well in lectures and take good notes. His tool helped him take notes all the while looking at his professors and focussing better. Ever since, he has been taking notes on interesting things people tell him. When he is in conversation, he can read information from the Internet or past conversations, and he says that this allows him to have much deeper and smarter interactions. In fact, Starner and a colleague designed a program called the Remembrance Agent that allows his portable computer to provide relevant information to each conversation in real time. In other words, as Starner types onto his keyboard, he sees words pop up onto his screen about past events or conversations that could be brought up in the current context. Starner says that this tool increases his memory immensely and helps him feel more confident in social situations, because he can remember more details about people’s personal lives and communicate with them in a more meaningful way.
ReplyDeleteI think that Clark and Chalmers would agree that Starner’s tool is an example of extended cognition. Like Otto, Starner uses his tool to increase his cognitive capacities. Starner’s device is attached to him most of the time (he takes it off to sleep and shower), therefore it might even come closer to the brain implant example than to the notebook example: Starner’s tool, although it can be removed, is attached to his body and is thus less easy to “decouple”. If such a device could be attached to someone’s body in a more permanent way, would there be fewer objections to saying that a tool like Starner’s can be part of the cognitive process?
That being said, upon further reflection, I disagree with Clark and Chalmers’ idea of extended cognition. While Otto’s notebook and Starner’s portable computer are tools that help them do the things they do, they have nothing to do with feeling. As we have seen in lectures, cognition is concerned with both doing and feeling, and “extended cognition” completely leaves out feeling. Therefore, even the most sophisticated brain implant would only be a tool and not part of the cognitive process, since it can never be a part of the mind.
Delete"The human organism is linked with an external entity in a two-way interaction, creating c oupled system that can be seen as cognitive system in its own right".
ReplyDeleteIt is clear that all living organisms evolve, act in and upon a certain environment. The environment affects us and that in turn we alter the environment, all the time (e.g. ecological niches). Thus, we can say that everything we can do is dependent upon our environment. We cannot function without oxygen, water, food, people and things to learn to categorize and make sense of the world. So yes, it true that we cannot be decoupled from our environment. Is there a scenario where our brains function completely autonomously from the environment ?
"If we remove the external component the system's behavioral competence will drop, just as it would if we remove a part of its brain".
Their claim implies that there are two fixed entities. The human organism and the environment. We are the locus of the cognizing that extends on to the environment. They ask,how far does it extend and/or what if the extension goes the other way around ?
I think we all have an intuitive idea that whatever is inside my skin is my system, and that my clothes, even my hair (although we get very attached to its sense of identity) are not me. My computer, my phone, my car, all these devices that have become so entrenched in our lives, that we *think* and maybe feel that we cannot live without them. However, they are only external systems that allows and affords the possibility to enhance or make easier certain processes and functions. Our felt states do not extend to the phone, the car or the computer. If we lose our phones, we might cry, we might rejoice, but ultimately we have not literally lost a part of our mind.
In the case of Otto's notebook, it is the same! He cannot, in effect, function without it because he has Alzheimer's, but his notebook is in no way feeling what Otto is feeling. It does not possess any intentional felt states even though Otto is offloading everything that he can in there.
"The biological brain is in similar danger, and occasionally loses capacities temporarily in episodes of sleep, intoxication and emotion" (p4). Is there decoupling within the brain ? A decoupling with the environment ? I find it so interesting that they throw emotions in the same lot as sleep and intoxication. Does this imply that emotions are thus a pure internal state? Doesn't this go to refute their claim that "cognition is extended" ?
In this piece Clark and Chalmers articulate their belief in active externalism: the active role of the environment in driving cognitive processes. Quite simply, I don’t quite understand why this belief needs to be fleshed out. It seems so obvious as to be ridiculous to even devote an entire essay to. All things we do involve an interaction with the world around us. Moreover, this interaction is inherently active. How could we conceivably interact with the world around us passively? Our vision is guided by selective and active attention, our memory by active mechanisms of recall. Some may argue for dual process theory: one active mechanism and one passive mechanism of cognition. I will agree that passive/habitual cognition occurs. But even in classic habitual cognition examples (e.g. highway trance) then we are habitually interacting with our environment in an active manner, i.e by responding to our environment or manipulating it. Therefore, the notion of passive externalism makes no sense to me. Though Clark and Chalmers go into very little detail, it seems as though Putnam and Barge make a case about historically interacting with the environment in the same way. This would fall into the category of habitual/passive interactions, which I have already addressed. But even so they make some claims about distal interactions and twins that are wholly unclear. How could you distally interact with the environment? How could the features of the present not be relevant? Your environment always has a part in influencing your actions, this is as obvious as saying that a fish actively interacts with water.
ReplyDeleteSo to add a bit to my comment, though I believe that cognition necessarily involves input from our environment I do not believe that this input is part of our mind. E.g. if we take away this input our mind still exists.
DeleteThis comment has been removed by the author.
ReplyDeleteI am struggling to accept Clark and Chalmers’ contention that cognition is extended in the dynamic interactions between our minds and the environment (AKA ‘active externalism’). While we may engage with tools that assist with our cognitive processes, the tools themselves are not cognizing. Let’s take the example of using one’s fingers to calculate 9+18=27. This calculation is carried out through the combination of: 1) internal cognitive processes (i.e., knowledge of an addition algorithm) and 2) features of the external environment (i.e., one’s fingers). The fingers (2), however, are not cognizing, but rather being used by the cognizer/cognitive processes (1) to reduce the cognitive load and compensate for our limited memory capacity. Thus, it is our internal cognitive processes that ENABLE us to use and interact with these external features in such a way as to provide this compensation/support. Therefore, while cognitive processes may at times be carried out in conjunction with the external environment, the environment merely serves as an aid to cognition – and that is only because our cognitive processes manipulate it to do so. The external environment itself is not cognizing.
ReplyDeleteMoreover, Clark and Chalmers appear to be focussed on the ‘doing’ aspect of cognition (the easy problem), but as Steven says, ‘cognitive states are felt states’, so what about ‘feeling’ (the hard problem)? As Willem mentioned above, it feels like something to know and believe a calculation (ex. that 9 plus 18 is equal to 27). Thus, after DOING this calculation (even with the assistance of paper or fingers) I have a FEELING; my fingers that were used, however, did not DO the calculation and surely do not FEEL it.
I think that this paper is interesting but I'm not sure if they clarify the functions of the mind and how this delineation applies to the problem of extended minds. As has been discussed, there are felt and unfelt states. What I believe Chalmers and Clark were referring to about the the extended minds problem, is extended computation. The fact that we use tools outside of our brains to aid us in computational and cognitive functions. I don't believe that the extension applies to felt states, when they cite beliefs in their argument they're just referring to the information that may generate that belief, but not the feeling of belief. Therefore I believe the term of extended minds to be incorrect, extended computation would be a better fit (in my view)
ReplyDeleteI find the thought experiment proposed by Clark and Chalmers to be an interesting argument, but am altogether unconvinced. The thing that makes the mind what it is in actuality is the plastic nature of it. Simply having access to a set of directions in a notebook does not extend the mind to those written directions. Furthermore, once Otto sets down his notebook and no longer physically possesses his set of directions to the museum, the extension of the mind would appear to be retracted from the notebook because the notebook does not have a mind of its own. While there are many tools that humans have created which help in everyday life, like calculators and GPS systems, these tools do not cognize, they just compute (which we already established after the Chinese Room Argument as proposed by Searle) so how could we argue that cognition is extended to these objects?
ReplyDeleteThe question regarding how much cognition is present in the three problem-solving situations is intriguing. If you imagine yourself in the three situations it is clear that (1) would take the most time. The second scenario would seemingly reduce the amount of time and thinking/cognition required. (3) seems like it would be the same speed as the second situation, but it is hard to imagine how the implant would work and how much time it would take for the agent to decide which method to use. This could potentially add time and required cognition.
ReplyDeleteNext I wonder if epistemic actions alter the world “so as to augment cognitive processes such as recognition and search”, demanding epistemic credit, and these functions are processes which done in the head would be considered cognition, are then labeled as cognition, are the blocks in the game the part of the environment that is cognition, or is it the implant? I am confused on this point.
The idea of the Naked Mind was interesting to me because it makes sense that we have our own cognitions that are internal regardless of environment, and these operations and resources are always available to us. But, this decouples the mind and the external world in a way that I find inconsistent with the author’s main points. If the Naked Mind exists, then some aspects of our interactions would be solely governed by these cognitions, which would require the environment interaction and extension of these cognitive resources. I am not sure if I am clear on how this fits into the idea of the Extended Mind as a whole.
ReplyDeleteThis is an interesting article because a lot of people would tend to agree with Chalmers that the mind can be extended with the help of the environment. However, although external aids such as my phone are helpful in making decisions and guiding me places, my mind is still the one feeling and deciding.
ReplyDeleteSo, in the case of Otto who uses his notebook as a memory storage, his mind is still the one feeling whether the information is right. Without his notebook he would feel lost and will feel like he needs to ask for directions but I wouldn't say his notebook is an extension of his mind, it is just a helpful tool that allows him to go through the day without being completely lost.
"Could mental states be partly constituted by the states of other thinkers? We see no reason why not, in principle"
ReplyDeleteI feel very conflicted by the arguments in this paper. While I think Chalmers and Clark make interesting points about the technology we use in our cognizing processes today, I found it hard to accept some of their ideas based solely on my gut intuition.
I appreciated their argument regarding google and the notebook scenario, in the sense that there are many things we know/believe that we do not have constant access to. It would be impossible for us to maintain all of the information at once, and the time and energy consuming nature of these impossibility would make it an adaptive disadvantage. This reminds me of our conversation regarding categorizing, which can help us organize things efficiently and process our environment more efficiently then Funes.
But these speculations went a bit far for me in considering our extension to the world wide web, which would essentially reduce the world into one giant cognizing system. The internet is such a vast resource, it would be impossible for one person to even dream of seeing every piece of content that has been and is currently being uploaded. Despite the capacity for google or the internet to be a reliably used system for solving all of one's problems, I do not feel convinced that the results of an online search could count as one's own beliefs and mind.
In addition to the connections I made with categorization and mental efficiency, I also found myself questioning the potential of their point of view, given that I can't feel the minds of others, nor the information that is on the world wide web (outside of a sensorimotor context). I am not sure I could ever consider something a part of my mind that I could not myself feel.
I really had an issue with the memory example Clark and Chalmers used to defend their "extended mind" argument.
ReplyDeleteIn my opinion, it is not possible that Otto's brain and Otto's notebook are on the same mental states, because notebooks are zombies (whereas we are assuming here for the sake of the argument that Otto feels and thus is not a zombie); therefore, Otto's notebook could not be considered an extension of his mind. In contrast, Inga's memories are contained within her brain, which (again, for the sake of argument) contains her mind (feeling state) as well. When Inga recalls a memory, it feels like something to do so, whereas because Otto has Alzheimer's, it would not feel like something to consult the notebook and find the answer written there.
The retrieval process in Inga's mind, which includes feelings, is not the same as the retrieval process that Otto utilizes, which is akin to him looking up a word in the dictionary or typing a calculation on a calculator. Otto may be doing something that involves cognizing, but it does not necessarily involve feeling.
I would be interested to entertain an argument for extended cognition where both people in the situation are like Inga; they both have intact memory, except Inga #2 uses a notebook instead of natural memory recall to have her questions answered, in the same way that Otto would use the notebook. I believe the answer above is still relevant in this example, and even more obvious: since both Ingas can use their memories and it FEELS like something to do so, the fact that they can also refer to a notebook is simply an example of using cognitive technology, but it is NOT an example of an extended mind.
The idea of an extended mind is quite compelling as the objects used would be extending the capacity of what your mind is able to do such as finding the fastest route to your next destination or using paper and pencil to write down important things for a later date. The term extended mind also seems very feasible as all these examples are things that are used as an extension. However, these things help us DO stuff and still avoids the feeling issue that our mind generates and therefore doesn't help us solve anything or should'nt be considered as an extension as those things do not feel
ReplyDeleteI agree that the paper, pencil and notebook idea is not really an extension of the mind but rather just an extra piece that helps us navigate through the environment that we are integrated in. It is an extension in the sense that it is there as an aid but it is not really part of the processes within the mind and cognition.
DeleteIn the example given about mental rotation as an example of cognition, I see serious flaws in their reasoning. I think it's very obvious that physically rotating an image is different from mentally rotating. Cognition is present in all three cases, but the mental rotation requires "more" of it. Not only does the person sitting at the computer screen have to do more things to accomplish the same goal as the person who is just sitting and pushing a button to rotate it physically, their mental rotation is also prone to mistakes and misinformation. With physical rotation, you would never doubt your perception of the rotated image, but with mental rotation you have to rely on your own spatial abilities and acknowledge that your rotation may be tainted by poor human perception. Because of this opportunity for mistakes, it makes sense to me that there is a human tendency to lean on environmental supports. So while both may be an example of cognition, with one within the head and one not, I don’t believe the two processes are at all equivalent.
ReplyDeleteAdditionally, their explanations of behavioral equivalence remind me of Searle's Chinese room argument. Just because the output is physically indistinguishable, does not mean they are the same. Namely, the problem of feeling vs not feeling is key.
I would argue that even if the person is physically rotating the object, they are still using the visuo-spatial areas of their brain; that is, the are still mentally rotating the object but in parallel with physical rotation. However, I do agree that there can be differences in judgement when physically having an object in your hand and only using your perceptual knowledge from looking at a picture with no immediate feedback.
DeleteI found Chalmers' TedTalk very interesting and engaging; the extended mind is a fun subject to toy with. I find the idea of, for example, the iPhone as an extension of the mind very curious, as the points Chalmers was making about the use of a phone for memory, planning, and navigating were valid. These are tasks we are essentially giving our phone to do instead of our minds, which can free up room to focus on more important and complex priorities. This speaks to “the general tendency of human reasoners to lean heavily on environmental support,” pointing to the idea that cognitive processes aren’t all in the head, but part of an interaction with the environment. This exemplifies the concept of active externalism; whereby, external features are actively involved in the way we cognize. As it seems, the mind is not just limited to inside our skulls. That being said, the mind is still what is the core driving force of any action that is carried out – even if I use my phone for navigation, it is still my mind that has to make sense of the directions. Even if I store reminders in my calendar, it is still my mind that is making a point to record important information. So, instead of an extended mind, I would view the environment as providing tools to enhance the mind.
ReplyDelete
ReplyDeleteThis concept of extended cognition is tough to digest, but once the point that cognition is not equivalent to consciousness was made, it became easier. The fact that not everything cognitive has to be felt, means that cognition could be unrestricted by the brain, and could include entities outside the human body. Humans rely heavily on their environment to do many things, for example pen and paper or a calculator for long multiplication. You could argue that without the brain, these operations could not be performed with pen and paper. But the opposite is true as well - without these external environment is required for brain processes to be fulfilled. Like Chalmers and Clark argue, I agree that a dynamic exchange system is a more accurate description of cognition. However, this only furthers our knowledge about the easy problem, and the hard problem is not explored at all. Clark and Chalmers are providing an alternative explanation for the doing but are leaving consciousness (aka feeling) out of the equation. They equate believing, desiring, and experiencing to feeling, which I think is stretching the meaning. It feels like something to believe, desire, and experience, therefore those states are mental when they are being felt, but what happens when they are not being felt? This brings into question whether feeling can be extended as well as the computational processes in the brain. Could feeling actually be generated outside of the brain? Could dualism be a potential path for further exploration?
Also, I am still confused on what is meant by passive vs. active externalism. From what I understand, passive externalism is when the outside forces (outside meaning not in the head) do not play a role in altering or do not contribute to the processes occurring in the brain as they occur. Active externalism, on the other hand, has a direct impact on the organism's cognitive processes and its behaviour.
The main argument made by Chalmers and Clark is that cognition is not restricted to the mental processes in the head, but also the physical manipulations, and aides (like notebooks) that are available in the environment, and that these become an extension of mind.
ReplyDeleteI believe the author contradicts himself when he says "case (2) with the rotation button displays the same sort of computational structure as case (3), although it is distributed across agent and computer instead of internalized within the agent. If the rotation in case (3) is cognitive, by what right do we count case (2) as fundamentally different? " The author started off with properly stating that case 2 and 3 were "computational" which is exactly what they are input/output manipulations, but then a sentence later he stats that case 3 is cognitive, which tells me he's conflating cognition with computation. Furthermore, there seems to be a focus on physical manipulations which are "doings" not feelings, and that is another steadfast error in his argument. By his externalism argument, he equates a notebook of a person with severe Alzheimer's to the memory of a person not afflicted with the disease, a false equivalence. A notebook can be a representation of something you know or understand, say for example writing down your name in English, and then being handed a notebook with your name in Arabic, you would not be able to understand that what you are reading, nor categorize it as a name in the first place. Thus the notebook maximally serves the same function as your sensorimotor interactions, they can provide input by which you can use to cognize (computing +feeling), but a pair of eyeballs can not cognize by themselves and neither can a piece of paper, or any other technological aide (be it an implant or a calculator).
Further he writes, "Of course, one could always try to explain my action in terms of internal processes and a long series of "inputs" and "actions", but this explanation would be needlessly complex. If an isomorphic process were going on in the head, we would feel no urge to characterize it in this cumbersome way." In this quote, I can dissect that firstly he realizes he is dealing with action (a matter of the easy problem, and he fails to realize that cognition is the easy and the hard problem), secondly he talks of inputs and outputs again which is computation and lastly he states that this explanation can not be possible because it is "needlessly complex", which is an outright error, and can not be a viable argument for dismissing arguments against externalism.
The author could begin to have validness to his externalism idea only under the parameters that cognition is simply computation.
ReplyDeleteIt seems to me that much like with mirror neurons, Clark & Chalmers are just describing things we already knew we could do/happen in the brain. We knew we were good at mirroring others’ actions, and we already knew we outsource performance capacity to technology; they’re not really explaining anything revolutionary about feelings (“the mind”). From the examples they provide, it seems like they attribute “the mind” to more than just feeling. They address that cognitive processing (like memory storage) can take place in the environment (which I don’t dispute), but they go on to say that although “mental states” such as “beliefs” “desires” and “emotions – i.e. feelings – are determined internally, they can be impacted significantly by external factors that drive the cognitive process (and so those external factors are part of the cognitive system/mind). But when I outsource memory storage to my notebook, my notebook isn’t part of the thing that’s FEELING and hence not part of my MIND (the feeling part of brain processes). The interdependent couple for whom one partner plays the same role as Otto’s notebook are still feeling their own SEPARATE feelings, even when they’re relying on each other.
ReplyDelete“The moral is that when it comes to belief, there is nothing sacred about skull and skin.”
I would argue that when it comes to FEELINGS (including beliefs), everything about skull and skin is sacred. Where can the feeling part of cognizing come from BUT inside that body? I really liked Willem’s connection to the Systems Reply to support this objection; how could the system of notebook and person be somehow have feeling that’s unified and more than its parts?
To me, Clark and Chalmers mostly make a good case for how our minds can extend to the external environment. I agree that things like spoken and written language serve to extend our cognitive functions, but I get a little sceptical when they bring up beliefs.
ReplyDeleteThey say, “While some mental states, such as experiences, may be determined internally, there are other cases in which external factors make a significant contribution. In particular, we will argue that beliefs can be constituted partly by features of the environment, when those features play the right sort of role in driving cognitive processes. If so, the mind extends into the world.”
In this case, they use the address of a museum to serve as an external factor, or belief, that play a role in driving cognitive processes. While I understand and agree that the external world can influence our cognitive processes, I don’t agree that the mind “extends into the world” in this sense. It seems to me like anything happening inside of our heads can be called a mental state and any mental state is something that we are feeling. According to Descartes, one of the only things we can know for sure is that we are feeling when we think we are feeling (i.e. we are cognizing when we are cognizing) and that we are ALLOWED to doubt what this cognition might be telling us about the world. Therefore, it seems kind of illogical to me that a belief could be an external factor that serves as some type of mental state.
Where does the mind stop and the rest of the world begin? This question is addressed well in discussions of Deep Brain Stimulation and Transcranial Magnetic Stimulation. These two techniques are very similar to each other in that magnetic fields are used to stimulate neurons in the brain. Once we move from TMS to DBS, what used to be external moves to the internal and connects with neurons (internal hardware). Our feelings and moods are highly driven by internal states, which are seemingly untouchable with and unreachable with such hardware. This could be a potential next step in the field. With the prevalence of mood disorders and other affective disorders, this technology could prove extensively useful. Few treatments for these disorders keep their efficacy in the long-term; dosage is always increasing, psychotherapy remains incredibly expensive, and reoccurrence is all too common. If we can find a way to connect outer hardware to our innermost states in the way that DBS does, we may be able to treat these conditions more effectively and increase the quality of life for many patients.
ReplyDeleteThe concept of the extended mind seems plausible enough when explained by Clark and Chalmers, especially in the Otto and Inga example. You don’t want to deny Otto the belief that the museum is on 53rd street simply because his belief does not come from his memory. But I believe what is the reason it seems to plausible when you first read through it, is that the definition of belief they employ allows them to say that it can come from different places. If you restrain the definition of belief as something that the person thinks is true in the moment that they are thinking about it, then both Otto and Inga are only believing the museum is on 53rd street when they are both thinking about it, no matter where they retrieve that belief from.
ReplyDeleteThe concept of active externalism sounds convincing with regards to the requirement in symbol grounding, which essentially involves sensorimotor interactions with the external environment. In these cases, it is clear that features of the external environment are just as relevant as the brain’s processes that link symbols with their referents, since they must fundamentally be grounded in the physical objects they represent. In this case, interactions with the external environment in the form of the physical object in question are necessary and part of the cognitive process. This is equally relevant to the process of identifying categories based on affordances. In learning these categories, interactions with the object itself are important and thus form a critical component of the cognitive process of categorization. Importantly, if learning categories through a method of hearsay, the symbols used in the learning process must themselves be grounded, thus requiring interactions with the external environment. The relevance to category learning becomes ever more clear when Clark and Chalmers state the fact that if the internal structure is constant but the external environment changes, so does the behaviour. Intuitively this seems to counter the notion that active externalism is relevant to categorization since every instance is infinitely different yet our ability to identify features of that instance is constant. However, this statement can also be taken to capture the fact that when the features of category members that are critical for identifying them as members change, we must refine our categories.
ReplyDeleteAdditionally, Clark & Chalmers bring up the example of evolution as countering in reliable features of the external environment into the evolution of cognitive processes. This is related to the fact that evolutionarily based mechanisms work based on predictability in the environment. Take the example of hedonic foods, which humans are intrinsically motivated to seek precisely because of their high sugar and fat content, which made sense in the adaptive environment. Though the unpredictability of food sources that characterized the adaptive environment has changed, this evolutionary mechanism remains and has caused some to drastically change their consuming behaviours. They have changed their behaviours despite the fact that internal structure or internal cognitive processes have remained constant and precisely because the external environment has changed and they no longer have to rely on consuming other living organisms or foods with high sugar content for survival.
Chalmers and Clark make a really convincing argument for active externalism. However, it reminds me of mirror neurons. We already knew it was possible to mirror others because we discovered the neural correlates. Likewise, we already knew that we use the external environment to help us cognize, but can we necessarily say that the environment is cognizing? Chalmers and Clark try to make this point with the coupled system. We use calculators, but we cant say that the calculator is cognizing because then we would submit to computationalism. In the context of the hard problem, we know that feeling is pertinent in cognition because it feels like something to remember, to understand, etc. Can the calculator feel? No. Chalmers and Clark state a common objection that active externalism is identifying cognition with consciousness, and reply that not every cognitive process is a conscious one, which is true. But every felt state is a conscious one because theres no such thing as an unfelt feeling. My remaining question is if everyone conscious state is felt? Because if its not, then maybe there is some argument to be made for active externalism.
ReplyDelete" the human organism is linked with an external entity in a two-way interaction, creating a coupled system that can be seen as a cognitive system in its own right. All the components in the system play an active causal role, and they jointly govern behavior in the same sort of way that cognition usually does. If we remove the external component the system's behavioral competence will drop, just as it would if we removed part of its brain. "
ReplyDeleteI found this really interesting because I like how it describes the environments role within cognition. Although cognition occurs within our mind, it goes hand in hand with the environment to conduct behaviour. Without the environmental aspect, the behaviour would not be conducted in the same manner as it should be. I think that this is important when considering the mind. It is not just an internal mechanism but rather one that works with external factors to help elicit the appropriate behaviour.
The discussion about epistemic actions makes a really crucial point about cognition: that cognition doesn’t take place solely in the mind. Epistemic actions alter the world to “aid and augment cognitive processes.” Clark and Chalmers use the examples of rotating a block to see if it fits in a slot (like Tetris), and using a paper and pen to assist with doing math. Feedback from performing epistemic actions and their active features creates a cognitive loop with the mind. During this course, we have sought many different mechanisms for cognition, but have not yet considered that cognition could be something (at least partly) external to the mind. However, this idea creates a problem with the distinction between the environment and external features of cognition. I am inclined to say that the whole environment accessible to us via our senses is an active feature (or collection of features) included in a feedback loop. External features of cognition are the environment - they are one in the same. If I am incorrect to say that the environment as the whole is an active feature, or a collection of active features, how does the mind seek out relevant features in the external environment?
ReplyDelete"We cannot simply point to the skin/skull boundary as justification, since the legitimacy of the boundary is precisely what is at issue. But nothing else seems different."
ReplyDeleteI don’t this is a very strong argument in favour of extended cognition. Firstly, it seems that the authors have invented the position of "active externalism" which prior to this article didn't seem to exist in the cognitive science literature, and then created a problem around the position. They follow to explain what seems to me to be three extremely different situations which employ extremely different processes, both internally and externally. Finally, they claim that the processes are identical, and toss out the "skin/skull boundary" counterargument on the basis that its legitimacy is already in question and it is therefore not a valid counterargument. However, the authors themselves are the people who brought the skin/skull boundary into question in the first place! They seem to be structuring the paper so that there is only one logical conclusion, and hoping that we don't notice that the entire argument is invalidated from the very start.
"So the mere fact that external processes are external where consciousness is internal is no reason to deny that those processes are cognitive."
I feel like this argument is a huge stretch of reasoning and I also think a school of thought like active externalism will bring about more questions than it will successfully answer. I completely agree that many cognitive processes take place outside the consciousness, including processes such as memory retrieval, motor skill acquisition, etc. But it is another thing all together to claim that physical boundaries such as the skull are as meaningless here as the boundaries between the conscious and the unconscious mind. It is almost unbelievable to assert that whatever participates in a cognitive action, be it the mind, the limbs that the mind controls, or the tool that the limbs control, are each equally a part of the process of cognition. It seems to me that this extension could be argued to never really end, but go on infinitely until everything in the universe would be included in what the authors consider to be "cognition". Once again, the authors are structuring an argument where their conclusions hold, but one which also have very faulty premises.
"For Otto, his notebook plays the role usually played by a biological memory."
Does this not go back to simulation versus reality and the Church-Turing thesis? Although Otto's notebook "plays the role usually played by a biological memory" (is simulating biological memory), can it be said to be a part of cognition because of its successful simulation (formal modelling)? I personally think no. Just as the simulation of a vacuum cannot perform as a standard vacuum can, neither can a simulated biological memory perform in the same way a true memory can. Even likening the two to me seems like a gross oversimplification of the intricacies of a human's memory capacities.
I thought it was interesting that Clark & Chalmers point to the “demarcations of skin and skull” as being the core disputes for their arguments of an extended mind. To me it is not so much that something is outside of my body that makes me skeptical about its ability to be part of my mind, but more the fact that it is not something I feel. We talk a lot about conscious states being felt states, so can something that is unfelt be part of my mind? Well, unconscious, unfelt states are still parts of our mind aren’t they? Well, I guess if we argue that consciousness is inherently felt then no – those processes are in our brains but not our minds (as we said in class). Clark & Chalmers at no point address this – so besides feeling as if they have taken the whole extended minds thing too far (is everyone in my life an extension of my mind? It would seem so in accordance with their definition), but they do not address feeling.
ReplyDelete