7b. Cauchoix, M., & Chaine, A. S. (2016). How can we study the evolution of animal minds? Frontiers in Psychology, 7, 358.
During the last 50 years, comparative cognition and neurosciences have improved our understanding of animal minds while evolutionary ecology has revealed how selection acts on traits through evolutionary time. We describe how cognition can be subject to natural selection like any other biological trait and how this evolutionary approach can be used to understand the evolution of animal cognition. We recount how comparative and fitness methods have been used to understand the evolution of cognition and outline how these approaches could extend our understanding of cognition. The fitness approach, in particular, offers unprecedented opportunities to study the evolutionary mechanisms responsible for variation in cognition within species and could allow us to investigate both proximate (i.e., neural and developmental) and ultimate (i.e., ecological and evolutionary) underpinnings of animal cognition together. We highlight recent studies that have successfully shown that cognitive traits can be under selection, in particular by linking individual variation in cognition to fitness. To bridge the gap between cognitive variation and fitness consequences and to better understand why and how selection can occur on cognition, we end this review by proposing a more integrative approach to study contemporary selection on cognitive traits combining socio-ecological data, minimally invasive neuroscience methods and measurement of ecologically relevant behaviors linked to fitness. Our overall goal in this review is to build a bridge between cognitive neuroscientists and evolutionary biologists, illustrate how their research could be complementary, and encourage evolutionary ecologists to include explicit attention to cognitive processes in their studies of behavior.
During the last 50 years, comparative cognition and neurosciences have improved our understanding of animal minds while evolutionary ecology has revealed how selection acts on traits through evolutionary time. We describe how cognition can be subject to natural selection like any other biological trait and how this evolutionary approach can be used to understand the evolution of animal cognition. We recount how comparative and fitness methods have been used to understand the evolution of cognition and outline how these approaches could extend our understanding of cognition. The fitness approach, in particular, offers unprecedented opportunities to study the evolutionary mechanisms responsible for variation in cognition within species and could allow us to investigate both proximate (i.e., neural and developmental) and ultimate (i.e., ecological and evolutionary) underpinnings of animal cognition together. We highlight recent studies that have successfully shown that cognitive traits can be under selection, in particular by linking individual variation in cognition to fitness. To bridge the gap between cognitive variation and fitness consequences and to better understand why and how selection can occur on cognition, we end this review by proposing a more integrative approach to study contemporary selection on cognitive traits combining socio-ecological data, minimally invasive neuroscience methods and measurement of ecologically relevant behaviors linked to fitness. Our overall goal in this review is to build a bridge between cognitive neuroscientists and evolutionary biologists, illustrate how their research could be complementary, and encourage evolutionary ecologists to include explicit attention to cognitive processes in their studies of behavior.
“Behaviors are usually what is exposed to selection from the ecological and social environment yet the relationship between cognition and complex behaviors is not always straight forward making it difficult to link fitness and cognitive behaviors is not always straight forward making it difficult to directly link fitness and cognitive performances.”
ReplyDeleteI found this article really interesting in terms of discovering and understanding the cognition of animals. What I found particularly interesting is the direct link that can be observed between the two articles that we read for this week. The human adaptations the evolutionary psychology paper looked at and the link to evolutionary psychology/fitness to understand animal behavior are very similar. This suggests a similar process for the development of cognitions of both humans and animals.
Specifically, I find that for many human adaptations, the reason, the behaviors, and the adaptations are also not always clearly linked. And, as the authors describe, the links are also difficult to make when studying animal cognition. However, due to the hard problem of cognitive science of feeling and of the uncertainty of others’ feelings, although we do not understand the links our brain makes, we take them for granted. However, for animals, we sometimes use the lack of links as justification for why their level of cognition is not as important as that of humans.
Too much doctrine and not enough explanation of the data. Anthropocentrism and the other-minds problem are handicaps. And the fact that only our species has language.
DeleteI also found this part in the article interesting, largely because it points out flaws in the logical behaviorism perspective and its ways to investigate cognition. As the authors point out, a cognitive trait can lead to any number of behaviors and a specific behavior could be produced by any number of cognitive traits. This is the challenge I believe in evolutionary psychology and cognitive science in general. Attempting to correlate observable behaviors with unobservable cognition can never be generalized. There are always multiple factors in the environment that can produce varying outcomes, and typically multiple cognitive traits are interacting with multiple behavioral manifestations. This can lead to false assumptions about cognitive evolution based by only looking at behavioral evolution.
DeleteA few pages into the first reading for this week, the concept of “reverse task-analysis” (which is essentially a version of reverse engineering) comes up. The claim is that the “bottom-up” approach to evolutionary adaptations finds behavioral phenomenon and seeks to explain it using evolutionary processes.
ReplyDeleteThe concept comes up again, although not explicitly in this second paper. The authors mention to complex webs that cognitive functions, behavioral outcomes and genetic predispositions as a result of evolution form. One consequence of that complexity in evolutionary processes is that it renders reverse-engineering incredibly complex. Things like the lack of 1-1 links between cognition and behavioral outcomes, or the fact that some phenomena we might want to look into could actually just be an incidental by-product of evolution, like the fact that we have bellybuttons. I never thought I would say this, but there is a part of me that is sort of thinking like Fodor thinks about brain mapping: I can’t fully grasp what the purpose of all of this is beyond pure curiosity (which, yes, is reason enough to do something, but admittedly doesn’t really lead anywhere). Taking this further, I am unclear about what the functional applications of the knowledge could be. I can’t really think of a way in which evolutionary psychology might be as useful in a clinical setting as most of the other data we’re already acquiring. It’s interesting to see if we can guess what the grand design of evolution is, and I’m sure there are people out there who would think understanding past evolution might help us understand what’s going to happen in the future, but all of those outcomes feel so unimportant to me. Why are we so set on developing that field if there are practically no functional implications of our discoveries, especially knowing that we are unbelievably far from the answer?
There are some genuine insights in comparative psychology (animal cognition; evolution of cognition) but also a lot of empty nonsense. It has not contributed much to clinical psychology (but what has?).
DeleteI am currently in a biology seminar in animal communication taught by an animal biologist and a child development psychologist. Today we started the last unit of the class, in which we are taking the information we have discussed about animal communication, and attempting to understand any insights into animal cognition and human language evolution (as you can probably imagine, this exercise is proving to be complicating our ideas of explaining cognition rather than offering us any answers).
DeleteI find myself, especially in that seminar, sometimes asking myself why it is important to study these things (despite also being really curious about them, like you’ve mentioned). I sort of have my answer for you, but within it is also a criticism of the field of animal cognition. Our obsession with non-human intelligence (I will focus on animal intelligence though I’m interested in ecosystems as well) is rooted in our desire to measure other species evolutionary development in relation to ours, especially given that we visualize ourselves as the most recent (and thus most evolved) species on the tree of life (which is not true!). As a result, our intelligence is the cannon for all intelligence, so proving that monkeys and dolphins are “just like us” might make humans finally treat animals better.
There are two major problems with this mindset. Firstly, measuring other species’ cognition to our own, while fruitful given our obvious experience with human cognition, will always be a losing battle for the species we are trying to observe. We can look at our ancestor species to see our similarities, *but they will never be like us because they are not us*. This desire to prove that other animals are just like us, so that people will finally wake up and treat them better, will not work, firstly, for that reason. Other animals have a completely different perceptual set, they can see different colors, hear different sounds, and thus experience a totally different world. Of course their cognitive development is not totally insightful to human development: we humans have a completely different T3 mechanism from non-humans.
The second problem is more concise and more grim: any possibility that studying animal intelligence will make us recognize animals as intelligent, and thus treat them better, has proven in much of our society as useless. We are not able to treat our “high intelligence” conspecifics with respect (ironic that we call it “humane”) -- treating our own conspecifics with the same respect is made even more difficult for that reason.
"The link between cognition and behaviour matters because a given cognitive trait could contribute to a number of behaviors and if selection acts differently on each behaviour, you will get very different selection dynamics on the underlying cognitive trait."
ReplyDeleteMy question is how selection acts differently on each of these behaviours, and if knowing that even matters? To me, this seems like a classic case of underdetermination. I am not saying that selection has no influence whatsoever on the cognitive-input to behavioural-output system; rather, I don’t think we will ever really know what all the various acts of selection are. And if they act differently for each input-output system, there will be too many “explanations” for it. So how will we exactly know all different ways in which selection acts to get us from input to output? Moreover, what do these various acts of selection really explain?
Evolutionary theory can sometimes answer "why" questions, but in evolutionary psychology it often just offers "just-so" stories. It is better at sexual jealousy and spider phobias than at most "higher" cognition. So far it's weakest on language, which is actually evolution's most powerful product of its laziness. (Can you explain that?)
DeleteEvolution is lazy, meaning that our DNA doesn’t code for every foreseeable circumstance we might find ourselves in throughout our lifetimes. All it needs to code for is a basic system that is good enough for us to survive. The rest is offloaded to the environment: a mole’s DNA doesn’t code for it to stop growing once it reaches 35.4cm, instead the environment (i.e. a narrow tunnel) will impose constraints on the mole’s height. Learning is another example of evolutionary laziness. We aren’t born with an internal dictionary (of some language) encoded by our genes, but with the capacity to learn any language there is, provided we get enough exposure to it. Our innate ability to categorize is essential for language, given that language, in a simple sense, is just the labelling of categories. Language is such a powerful weapon (a nuclear weapon, one might say) because existing labels can be linked together to name new categories (without the need to have interacted with or even seen members of that category).
DeleteI disagree that language can be simplified as “labeling categories”. I don’t think language helps to categorize but rather than language is a way to categorize.
DeleteThe ability to learn by instructions is what distinguishes humans from current robots. We as well as robots can learn through supervised and unsupervised learning approaches but humans can learn more abstracts categories or even some precise motor skills by reading or listening to instructions— think IKEA.
Language allows us to understand these instructions; however, language itself cannot be learned by instructions because such instructions would require a prior understanding of language. Supervised and unsupervised learning are also not viable explanations of our infinite linguistic capacity, i.e. the capacity to express anything that can be expressed, because we are never exposed to counterexamples, that is things that cannot be expressed. There is something about our linguistic capacity that is innate and that allows us to use language as a form of categorization.
Hello!
DeleteI would like to chime in because I think it is very important to understand why language is evolution’s most powerful product of its laziness. Language is an extremely powerful tool that only humans possess. Having the capacity for language means that you are able to say everything and anything using words. Other animals are not able to do this. Certain species of animals are able to respond to different threats present in their environments with specific vocalizations to warn the others, but this is not language! This is categorization!
When we say evolution is lazy, we are saying that evolution is offloading into learning rather than coding something into specific behavior patterns. Everything that evolution offloads into learning, it does not have to code. Regarding language, evolution is lazy by coding a tendency to learn quickly. As Erica stated, we were not born with a complete internal dictionary, we were born with the rapid ability to learn language.
Evolutions offloads into 3 types of learning: supervised learning (reinforcement & getting feedback from someone else), unsupervised learning (associative & seeing things over and over again to figure out patterns), and instruction (only possible if you possess language). I agree with the point that Chloe made about humans not being able to fully learn and develop language simply through these three types of learning. It is true that we cannot purely learn language by instruction because we would need a prior understanding of language. It is also true that we cannot fully learn language through supervised and unsupervised learning because we do not provide children with counterexamples of poor sentences. However, I do think it is important to say that children can learn quite a bit of language through supervised and unsupervised learning. Children can be influenced to say words and small sentences correctly by receiving positive feedback from their parents. Children also learn a lot through unsupervised learning. I took a class on child development where they talked about how children go through a type of “U shaped learning curve” regarding language. This “U shaped learning curve” explains the way children start off by correctly saying words and utterances once they first learn them and then they begin to purposefully say words and utterances incorrectly. The children observe how adults respond to their incorrect use of language before they begin to use what they have learned correctly again. I wonder if this is a way for children to explore the possibility of counterexamples in language?
I believe categorizing has a role in learning language. I do not agree with Erica, that language is simply “labeling categories”. We do not simply use language to label categories. However, categorization is necessary to understand what is being conveyed by language. In order to make and understand propositions, you need to be able to recognize that categories that are being referred to in these propositions as well. It is also important that we consider symbol grounding as a large part of language. Connecting a symbol to its referent is important to language and requires categorization and sensorimotor capacities to do so.
Erika, language is not simply “labeling categories”. Language is an instance of categorization , which is doing the right thing with the right kind of thing. When using language, so when we express any and every proposition, we categorize.
Delete"Costs of new defenses include developing the cognitive or morphological structures for new defenses as well as the added risk of expressing those defenses (e.g., rejecting own eggs), and these costs influence the evolution of recognition abilities. Plasticity in host recognition reveals the importance that making an incorrect choice can have for the evolution of egg rejection. For example, some common cuckoo hosts avoid rejecting their own eggs (recognition error) when parasites are not present by only increasing rejection rates when adult cuckoos are seen in the vicinity of the nest (Davies and Brooke, 1988)."
ReplyDeleteI found this section to be specially interesting as it speaks about the costs of incorrectly rejecting eggs for fear of being infected with parasites by potential host species. What interests me here is how the potential parasite host in question is aware of the consequences of parasites and their infestation. When a potential parasite mimics the host species, this is a clearly a categorization problem for the host with growing complexity as the "Mimicry-recognition-rejection arms race" continues to become more and more sophisticated. But on a species wide level, how was it determined what the acceptable level of mimicry risk was? Clearly there is some genetic understanding being communicated but in order for a species to succeed, the level of risk selected presumably need to be fairly optimal. I'd be interested in hearing more on this genetic evolution and how optimality is ensured.
Evolution is vegetative too, not cognitive. "Natural selection" does not cognize. And mimicry is not imitation.
DeleteI don't understand why mimicry is not imitation?
DeleteI think mimicry is when certain physical features of an organism are meant to resemble features of another organism, something like the circles on moth wings that are meant to mimic owl eyes in order to ward off predators. In this case, the eggs of the avian parasite mimic the appearance of the host eggs.
DeleteImitation refers to replications of actions (think mirror neurons). It has more to do with learning than trickery.
To add to Emmanuelle’s comment: imitation is indeed a way of learning. More specifically, it is a form of unsupervised learning. For example, an important part of infant language learning consists of listening to and imitating the speech sounds we hear in our environment.
Delete“Consequently, we know very little about what role cognition, a collection of highly plastic and flexible traits, plays in adaptation and biological evolution.”
ReplyDeleteA question that still bugs me about evolutionary psychology is the question of what is genetically encoded when we talk about cognitive traits. It seems like it puts the burden back again on studying the brain to explain cognition, and as discussed in other skywritings, the Fodor’s critique is still relevant in this case.
I get that with the mind-reading ability of ours we can assume some cognitive traits (for example we can assume that an eagle has something similar to our “visual perception”, even if we don’t and can’t know what it feels like to be an eagle) in non-human animals, and therefore we can assume an ecological role to cognition (because cognition is wildly spread in the animal kingdom). But I still don’t get how we can talk about evolution of these traits if we don’t know what is being encoded.
Re: Antoine Milette-Gagnon - "A question that still bugs me about evolutionary psychology is the question of what is genetically encoded when we talk about cognitive traits. It seems like it puts the burden back again on studying the brain to explain cognition…"
DeleteHello Antoine! My apologies about the late response.
I really appreciated your critique because I definitely also thought that evolutionary psychology places too much emphasis on genetics. However, Cauchoix and Chaine do, to a certain extent, also recognize the limitations of studying the brain: "the advent of powerful new neuroimaging techniques…has lead us to better understand how cognitive functions are linked to neural structures and neural activity…Despite this in depth understanding, much less progress has been made in understanding the evolutionary processes that have led to the patterns of cognition that we see." They also assert that evolutionary biologists examine behavioural evolution without really focusing on the cognitive mechanisms, but rather, the ecological and social contexts.
Nonetheless, I also recognize the role and emphasis of genetic heritability in cognitive development, exemplified by phenotypic changes resulting from sexual selection -- but I thought that Professor Harnad's point on devolution presents an interesting counter-example. Recent trends demonstrate that families of higher socioeconomic status are less likely to produce many offspring than "the less fit". This may seem like a devolution evolutionarily, but the wealthy are simply more focused on quality over quantity, so to speak. The key point here is that wealth and education are not genetic and therefore, demonstrates that behaviour extends beyond neuroanatomy and may elude evolutionary psychology.
I want to preface what I’m about to say with that I think anthropocentrism is an unfortunate ego-centric handicap, and by no means do I believe that humans have some sort of superior intelligence to other creatures – where we have developed language, I’m sure other animals have adapted other incredibly sophisticated cognitive abilities of which we can’t fathom.
ReplyDeleteThat being said, as discussed in class, language is a huge advantage for us, in communication, and improving our own cognitive capabilities (if this hasn’t already been explicitly said in class, I suspect that having language improves the sophistication with which we’re able to perform categorization).
So in light of evolutionary psychology, I wonder what specific ecological or social conditions existed for humans that allowed us to adapt language? Bear with me, as I know nothing about biology, human history, or linguistics: but could it be a combination of the fact that humans were pack-hunters (the ability to communicate is an advantage) plus we have all of the equipment for vocalizing in a sophisticated manner?
Gorillas can learn a limited amount of words in sign language, and we know they can create compound words on their own from words they’ve already learned – and of course other animals use vocal output to signal certain things - so what cognitive difference is there that was shaped by different ecological or social conditions such that we have language, and can say anything that can be said – while other animals haven’t been able to take advantage of the benefits of an improved ability to communicate vocally or gesturally?
I am in argument with you about the downfalls of anthropocentrism - is it a terrible and oftentimes debilitating view to have. I also agree that language is a huge advantage for us but I don't think I would say that having language abilities improves the sophistication with which we're able to perform categorization. I want to connect this to the Whorf Hypothesis. In trying to prove the Whorf Hypothesis some people have said that cultures with only the words 'one' 'two' and 'many' do not have as complex as number system as we do, or that languages which have more words for the variations of the colour red can see more colours than we can, or that languages of a certain peoples would have a richer understanding of a concept if they have more words to describe that concept. However, we have seen that these are all false and that in fact a 'richer' or 'more complex' language system does not give you the ability to see more things in the world. In this case, I think it would be similarly correct to say that language does not give us the ability to categorize. For example, if someone could not speak or did not use language at all (this includes non-verbal languages such as sign language) I would argue that they would still be able to categorize. They would still be able to see the differences within a pile of red, green, and blue balls and be able to sort (or categorize) these into distinct piles of red balls, green balls, and blue balls. Additionally, if someone did not have language they would be able to categorize and distinguish between the bad (poisonous) mushrooms and the good (edible) mushrooms in order to be able to survive. Therefore, I don't think it is language that gives us the ability to categorize/ability to perform categorization.
DeleteI agree with Flora regarding your anthropocentrism comment. You have said that you do not believe that humans have some sort of superior intelligence to animals, but clearly we do. We have defined intelligence generally as problem solving abilities (more or less), so obviously it would be false to say that we are not superior to animals in that respect. Although, I am not conflating this with the idea that we have superior value to animals; that I do not believe. I don't think intelligence is a sufficient quality to judge intrinsic worth on the basis of.
DeleteHowever, I disagree with Flora regarding her rejection of your statement regarding language and categorization. I think she misunderstood what you were saying. She seems to believe that you were claiming that language has given humans the ability to categorize, but it seems to me that you were stating that language has improved humans ability to categorize and I would agree with you on that.
Flora used the example of categorizing color, but that is a bad example considering that it is one of the few innate categorical perceptions we have. A good example would have involved a learned categorical perception. Moreover, the fact that we can learn categories through language (instructions) is a strong enough suggestion that language has improved our ability to categorize, and there are more reasons to believe that language has improved our ability to categorize. For instance, learning through instructions allows us to learn features of members and nonmembers of categories in a detailed and sophisticated manner. Animals may not distinguish between an object with diagonal lines going from the upper left to the lower right and the same object but with diagonal lines going from the upper right to the lower left. And we may not distinguish them as being different either, but through language, we could be instructed or informed of what distinguished these two objects and thus be able to categorize them in even more subordinate categories than before.
Super interesting discussing here. I stand with Johnny's comment on anthropocentrism. I do believe that there is something "superior" to us that than from animals. Our advantage of having a bigger prefrontal cortex allows us to better our abilities in problem solving and executive function. However, I don't believe that intelligence is necessarily a good reason for why we should exclude them from our moral arena.
DeleteLanguage is a nuclear tool. To be able to say anything that you can possibly say and create new propositions to no end, we have a tool that allows us to have instruction and learn by instruction. We are no longer bound by solely supervised or unsupervised learning. That new avenue, in itself, I think serves to prove that there must be an advantage to language. I think language serves to help us with categorization. With instruction, you can learn much quicker and be able to identify the members from the non members in a much more precise why. Language is a very powerful tool, I highly doubt it would have no benefit when it comes to categorization.
Aligned with Erin, I do believe that anthropocentrism is an ego-centric handicap. Believing that we are superior and separate entities than other species is detrimental (especially when we deny the possibility that they may feel like we do). In agreement with Johnny however, I do believe language does make us superior to other species as it has allowed us to categorize more extensively through instruction. Like Amy, equating this superiority to being more 'intelligent' than other species is an uncomfortable assertion for me. I cannot deny and can only appreciate other communicative mechanisms species have evolved to ensure its survival as forced upon by its specific and distinct environmental pressures (e.g., bats and echolocation).
DeleteResponding back to Erin's original question, I am also fascinated by what ecological or sociocultural events in our evolutionary history resulted in the development of our ultimate nuclear weapon, language. As mentioned in class, along with being pack-hunters, it seems that superior cooperation within these groups expedited our ability to survive and flourish as a species. Extending this, it seems to me that language made it easier for us to cooperate and efficiently carry out group goals by freeing our hands and make how we share information more efficiently. This is also supported by adaptations acquired in our anatomy to form the human vocal tract at the expense of other essential survival mechanisms facilitating breathing, swallowing and chewing. Why gorillas or other chimp species who also display social behavior (e.g., grooming) have not seen the same evolution of language is also a large existential question for me however.
“The lack of exchange between cognitive scientists and behavioral ecologists has left a big gap between methods used to study contemporary selection on cognition in the wild” (11)
ReplyDeleteAt first when I read this quotation, I immediately thought that the reason why there is a lack of exchange between cognitive scientists and behavioural ecologists is because behavioural ecology may only help solve the “why” we do the things we do and “why” we feel the things we feel. It doesn’t help explain the “how”. I still think that this is true, however, evolution is inevitable and results in both positive and negative changes that have created the neural circuitry we have today, that in turn impacts cognition. Therefore, if we wanted to build a T3 robot that is completely indistinguishable from humans then it must be able to evolve too. For this reason, it may be important to work with behavioural ecologists and evolutionary psychologists in order to fully understand how evolution has impacted cognition so far, in order to design a T3 system that is also capable of evolving/changing in the future.
I am not sure but I think one important part about the evolutionary mechanism that makes the whole explanation (seem to) work is to talk about how the variations and changes would affect the reproductive and survival success of the subjects and thus influence the species across generations. This requires a long passage of time because the course of evolution that happens on one subject, one generation specifically is very minimal. It seems to me that the analysis has to be, not focus merely on a particular subject to have a basic idea of the tract of evolution. What is problematic is that when we face a T3 robot, we don't have the evolutionary history of that T3 robot, and T3 robot does not face the problem of survival either. Changes in evolution just in the particular instance of a T3 robot may lack some underlying motivation as real-life species, and it is very likely that the particular evolution of one robot only would be too trivial to discern its existence. If you are talking about plasticity, I agree with you that it is essential. If your criteria of T3 means more than that, I think that would probably bring some challenges to the underlying assumptions of evolutionary psychology.
DeleteAkhila, I find your question really interesting. If accounting for evolution is necessary to building a T3-passing robot, I agree with Mingjuan that formulizing a flexible mechanism capable of 'evolving' in response to external changes driven by the motivation to survive (just like real-like species do) is a core consideration. This is supported by Lewis et al that suggest that what drives the strength of an adaptive problem depends on the magnitude of impact on survival / reproduction and the frequency that it is faced. Specifically, they suggest that high-impact adaptive problems (regardless of whether they are low- and high-frequency) drive evolution of psychological adaptations. To formulize this, could possibly categorize certain situations as being indicative of 'threatening survival or reproduction' and others not. However, this would be made complicated since we would also have to group situations that pass the threshold and define the magnitude of something being 'so threatening' that a single occurrence by our robot would spark evolutionary change and those situations that would need to be experienced multiple times. How would we be able to define this threshold given that evolution occurs over many generations across a large span of time? How would we possibly encompass and know all the possible situations a human may have faced in its evolutionary history? Like Mingjuan suggested, this is a large hump to achieving input-output equivalence needed for a T3-passing robot.
DeleteMingjuan also brought up another interesting point: it also feels like something to sense danger and to feel that your survival is threatened. While this is outside the scope of passing-T3, if this feeling of what it is like to be alive or danger is central to evolution, this does affirm in my mind that evolutionary psychology can explain why/how we do what we do (easy problem) but in the case of the hard problem, is more limited and fails to explain why/how we feel what we feel (especially how) at all.
Hi Akhila!
DeleteI think that it is not important to work with behavioral ecologists and evolutionary psychologists in order to design a t3 system that is capable of changing in the future. Firstly, the test for t3 is lifelong performance capacity, I think that it is irrelevant to try to create the t3 so that it changes overtime. If it is a t3 according to any of our life spans, then it is a t3. I think creating a t3 concisely answers the easy problem, and clearly the easy problem is the only problem for creating the t3. Secondly, studying the evolutionary psychology is clearly not necessary, even if we want to create a t3 that is supposed to evolve by time. It is due to underdetermination. And again, in a world where we solved the easy problem (either by describing it on a piece of paper, or by creating a t3 level robot), the only condition required for t3 is that system to be able to show lifelong performance. From where we (as friends of that t3 level robot) stand, it doesn’t matter if that t3’s cognitive abilities were changed by the knowledge that we gained from evolutionary psychology, or if the creator of that t3 just tweaked some settings so that that t3 still shows lifelong performance x years after it was created.
"If two cognitive traits act in concert to produce a specific behavior, you can get correlated evolution and even genetic coupling of those cognitive traits (i.e., linkage disequilibrium between genes coding different cognitive functions). This problem is exacerbated for neural structures since a given neural structure likely underlies a number of cognitive abilities and each cognitive ability relies on a number of brain regions."
ReplyDeleteThis sounds like another scenario where underdetermination appears to be a huge problem.
As the only thing we can observe is our current behavior which is gradually shaped by evolution (selection) and fundamentally ordered by our cognition, cognitive science and evolutionary psychology can only start from our overtly exhibited behaviors and try to reverse-engineer the reason and the mechanism underlying them. Therefore, the explanation chain we are attempting to build is: in evolutionary psychology, from behavior to adaptive advantages during selection; in cogsci, from behavior to cognitive capacities, then towards underlying mechanisms (neural structures).
However, as pointed out above, if one behavior generates multiple adaptive advantages which thus subject to multiple evolutionary theories, and at the same time that behavior may originates from multiple cognitive traits that are further supported by multiple neural structures, and each of the cognitive traits can also have different explanations given that multiple mechanisms could account for every single one of them, then the link between what evolutionary psychology can infer from one behavior and the real cognitive process that gives rise to that behavior just seems to be outrageously mysterious and underdetermined. Then what can we get from studying evolutionary psychology bearing a cognitive science intention?
I had a similar reaction to this passage. It felt like this statement rendered their proposed strategy ineffective; the entanglement of behavior, cognitive ability and neural structures makes it extremely difficult to attribute causality to existing cognitive processes through evolutionary hypotheses.
DeleteThey address this: “Without clear mapping of different agents of selection on multiple behaviors as well as mapping the links between multiple behaviors and multiple cognitive abilities, we will have only a poor understanding of the evolutionary dynamics underlying cognition.” Yet, I can't see how the first condition can be accurately achieved.
Additionally, they state the baseline conditions for applying Darwin's evolutionary theory: "(1) there is variability in cognition between individuals, (2) that this variability in cognitive performances is heritable, and (3) that this variation is related to variance in fitness (survival, reproductive success) under specific environmental conditions." But, in addressing each of these points they admit to difficulty measuring and little direct evidence for each condition. While their logical justification is clear, it was hard to be fully on board with a method that off bat is lacking solid evidence.
And as stated in many other skywritings--what will evolutionary histories really even tell us moving forward?
I am highly skeptical of the research referenced regarding guppies demonstrating the heritability of brain size, especially given that there was a “numerical learning test” that the guppies supposedly performed. I hold animals in very high regard, but I highly doubt our ability to create a research experiment that specifically tests the numerical learning of fish. I can imagine that counting would be useful for guppies and am not saying it’s impossible that they can learn in this way, but by what measure are we judging them? I doubt our ability to conclude intent and learning in species with such little resemblance to us (literally: we can’t judge their facial expressions). But this is likely besides the larger point of the article and I may be underestimating cognitive ecologists!
ReplyDeleteI am glad that they made the clarification that brain size is not necessarily a good representation of cognition. Elephants and whales have larger brains than humans and yet humans continue to assume that we’re at the top of the food chain, so why tout the significance of general brain size? I agree that, instead, “analyses should begin to compare specific regions of the brain, neural structure, or brain function rather than coarse measures of brain size” (p.6).
In this paper, they say the field of evolution of cognition has some problems that need to be addressed and list four. Then, in the very next paragraph, the authors propose expanding the scope of the field. Why? Will these new studies be the first to address the three issues previously mentioned? If so, that makes sense. But it seems like they are broadening the scope of the field without first addressing the current issues in setting a proper methodology to answer questions within the current scope of the field. It reminds of a Gazzaniga paper, where he says at this point in time it's better to just test a bunch of things and hypothesize later in the field of neuroscience. Many would frown upon that, as it could bias hypotheses to fit pre-existing data. I feel what this paper proposes is jumping the gun in a similar sense, moving forward into new questions without first having a framework to answer the original question.
ReplyDelete“Hippocampus size, however, varies predictably with the need for excellent spatial memory in brood parasites. Brood parasites have an enlarged hippocampus in the breeding season (Clayton et al., 1997), the sex that searches for nests tends to have a larger hippocampus than the other sex (Sherry et al., 1993; Reboreda et al., 1996), and brood parasites have a relatively larger hippocampus than closely related non-parasites (Reboreda et al., 1996; Corfield et al., 2013).”
ReplyDeleteChachoix and Chaine use increased hippocampal volume in brood parasites to “illustrate the power of the fitness approach in linking ecological context to the evolution of cognitive abilities and neural structure”. I immediately thought of the classic “London Taxi Drivers” experiment where it was shown that the taxi drivers had larger hippocampuses due to their extensive spatial mapping of London streets. Couldn’t this be the case for brood parasites?
Earlier, they address this concern in stating “While the appearance of adaptation is clear, such differences could reflect either local adaptation shaped by natural selection or result from plasticity in brain structure and behavior generated from the local environment”, and they go own to demonstrate support for cases of genetically inherited differences in spatial memory (i.e. genetically determined mating strategies used by side-blotched lizards that require more spatial memory correlate with hippocampus and dorsal cortex differentiation). However, it wasn’t clear to me that the hippocampal differences in the brood parasite case study was heritable/genetic rather than acquired through learning and neural plasticity.
I am glad you critiqued their analysis on hippocampus size because they put forth incomplete information. The ‘London Taxi Drivers’ study shows the increased neural plasticity in the hippocampus of taxi drivers. They were not born with bigger hippocampi – it was learned. I think Cauchoix & Chaine could have explained the findings in a more convincing light. After looking into the studies cited in-text, I found the sex of brood parasitic birds which search for nests have larger hippocampi compared to the other sex even before breeding occurs. Brood parasitic birds aren’t picked up by headhunters to lay eggs in hosts’ nests. This is genetically engrained in their breeding process. Larger hippocampi are found without any learning. However, I do not think heritability is the only factor causing more hippocampus size. I see neural plasticity also playing a role in parasitic brood species – similar to the London Taxi Drivers or the homunculus.
DeleteThe largest issue I take with all of these studies is that they do not appear to give us any insight into making a T3.
ReplyDeleteUnderstanding that guppies perform better on an arbitrary cognitive task, a "numerical learning" task, if they have larger brains tells us very little about *how* or *why* they do so.
I might mention that, while this task may be significant for human cognition, it appears rather unclear that this task is essential for guppy cognition.
Frustratingly, even *human* intelligence tasks do not tell us about the underlying human cognitive mechanism. Thus, interspecies studies appear to be a place that cognitive scientists can start. Personally, I think this is unfounded, and believe that it could leads to claims about cognition that we could simply not know about because of the OMP.
Studying another species' cognitive capacities may be able to tell us *something*, but (at least for now) comparative studies appear to not be applicable to the kinds of questions - the hows and whys - that cognitive science needs to answer.
I agree with you that these studies don't directly tell us anything to do with the questions cognitive science is concerned with. In the paper they say that, "Neuroimaging techniques have led us to better understand how cognitive functions are linked to neural structures and neural activity in several species including humans." I think it's good that there are people doing studies like the ones mentioned throughout the paper because it gives us a place to start when thinking about the hows and whys of cognitive science, but it's frustrating because we don't really know how to interpret any of it. It seems like there's a lot of missing information that could make sense of why those studies are relevant.
DeleteBeing able to directly link a cognitive function to a neural structure can help us create that specific cog. function in a robot for example, but only if we know everything else that the function is linked to and every other function that it influences and gets influenced by. Just finding the location of one function isn't going to help us with cognitive science for a very long time, so for now it seems like it's not useful to use the findings from studies like these when trying to answer the hows and whys of cognitive science.
"Interestingly, comparison of ecological and social factors in ungulates, showed that relative brain size is influenced by both social and ecological factors while relative neocortex size is only influenced by sociality (Shultz and Dunbar, 2006). Finally, species with larger brains have been shown to survive better in novel environments (Sol et al., 2005, 2007, 2008) in support to the cognitive buffer hypothesis (Sol, 2009)."
ReplyDeleteThese evolutionary psychology hypothesis are way too generalized in my opinion to provide a satisfying explanation of humans cognition and abilities. I find the statement that the size of the neocortex is only being influenced by sociability compared to the relative size of the brain absurd. First how can one environmental factor can be said to affect the overall size of a species brain? Then, what about all of the other factors like plasticity, genome, individual characteristics ? This is the danger and limitation of evolutionary psychology: verging to the ridiculous and overgeneralizing cognition which is way more complex.
Completely agree with you on the tendency to over generalize in evolutionary psychology research. Correlations are treated as causations. And like you mentioned, other factors that influence the same trait are not considered - the takeaways are under-deterministic.
DeleteTo be fair to the authors, I think we have to give them the fact that they have not blurred the lines between correlation and causation in the same way that other authors we read in this course have (e.g. Cook et al.). In fact, I found these authors to be very careful about not falling in the trap of reductionism (i.e. neuroscience-will-explain-it-all kind of claims). They always said things like (I’m paraphrasing): “the two fields have evolved in parallels and should work more collaboratively”, “their findings could complement each other and improve knowledge”, etc. It is not false that we can learn a lot about evolution using neural imaging results, for example. However, it will not help us understand the "how" question, and maybe, but probably not the "why" question either.
DeleteTo be fair to your own comment, though, it is true that a lot of evolutionary psychologists have made this type of claims in the past and continue to do it. And it is not clear if, although not stating this overtly, the authors of this paper do think that the findings that could be made from collaborations between the fields of research would answer this question, which we know from class and from reflecting on Fodor’s criticism that they could not. You are also right about the disregard of other important factors.
In this paper, Cauchoix and Chaine aim to describe how and why certain cognitive processes evolved. They describe the advantageous behavior certain species of birds display and link these behaviors with brain structure. In the beginning, they describe cognition as the sum total of all brain activity involved in transforming information. They do not explicitly mention whether feeling is considered a part of cognition, or the topics they discuss.
ReplyDeleteThe 3 three main approaches adopted by the paper to link evolution and cognition show shortcomings:
Comparative approach: The studies reviewed in the paper connecting brain size with cognition can only point to correlations, not causality, as the authors themselves point as well. The studies in birds relating brain structure to cognitive capacity demonstrated in natural habits and in laboratory studies are both cross-sectional, and hence, do not address the questions of whether the adaptive traits were the result of evolution or how it evolved over time. It addresses why some traits are adaptive, and then the assumption seems to be that evolution led to these traits being selected, without showing a heritable basis to these traits. The latter was stated in the paper itself as a necessary condition to show that a certain trait evolved under direct natural selection.
Fitness approach and case studies again provide a cross-sectional explanation, rather than a longitudinal one.
Overall, the paper might be able to address some factors influencing why certain traits evolved, but not how they evolved. They also do not mention the hard problem of why we feel.
Re: Anna Jos
DeleteI totally agree with you! Going one step further down that path, the authors failed to convince me that I should care. I am quite skeptical of the field of evolutionary psychology and its methods to start with (see my skywriting for 7.A), which might explain why I remain unconvinced, but I’m also not convinced by the overall desire to explain why things evolved? Sure understanding how certain cognitive abilities came to be fulfills our curiosity, but why does it matter? Shouldn’t we focus on understanding how it functions today? And why sometimes it doesn’t? Wouldn’t that be a better use of our time than generating untestable speculations for a time we no longer have access to, probably dismissing important factors that are just no longer relevant to our modern lives? (I use "us" and "our" very broadly to refer to both human and non-human animals)
The comparative approach lacks the advantage of fully understanding the natural context in which cognitive functions evolve by its very nature as it looks at evolutionary history. If the information is taken from previously conducted detailed research on specific organisms (without evolutionary context) then: Assuming that we believe previous researchers have accurately documented their findings on the natural context (and without bias), there are necessarily things that were missed. Apart from simple human error, we must also consider that the context in which animals were being studies was not always in line with the purpose for which that historical evolutionary research is now being used. If you use a magnifying glass on area A of a puzzle, while it may overlap with area B, it does not give an accurate picture of area B. Studying contemporary selection offers insights that evolutionary history simply cannot.
ReplyDeleteCognition is said to “present all the characteristics of traits under selection (variation, heritability and fitness benefits)” which is another strong point endorsing the use of the fitness approach. This provides insight into the natural selection of animals, providing details on the already established evolutionary knowledge. This allows for understanding of proximate and ultimate factors, but also the connection between the two, the connection between neural and developmental as well as ecological and evolutionary origins respectively. I personally think this integrative approach using methods across domains is the most compelling part of this paper. Much is lost when different domains do not communicate. Overlap provides opportunity for new perspectives on findings and new insightful questions all together.
ReplyDelete“Attention in such [behavioral or evolutionary] studies is placed on the quality of information and the outcome of a decision, but there is little understanding of how information is processed and how cognitive abilities enhance or constrain decisions based on the available information” (3).
ReplyDeleteThe Cauchoix and Chaine (2016) suffers from the prevalence of underdetermination in evolutionary and behavioral study. That is, that in the study of evolution of animal psychological processes, common approaches generally emphasize the linkages between the interpretation of outside information and the behavioral result. As such, they inevitably work under weak-equivalence, as they neglect to examine the exact mental processes undergoing in the mind as it encounters outside stimulus. For example when trying to demonstrate comparative studies of brain and cognition evolution, the authors run into issues of attributing evolution of brain size to broad environmental factors (5) which results in overgeneralizing (or leaving out all together) the complex persistence of cognition. Without understanding the actual process of cognition itself, we are left only with correlations and no causal linkage, which, in my opinion, is problematic, as we’ve likely reduced animals, at least in our scientific perception to robots which interpret information and spit out behaviour. In this case, this perception could possibly be due to the hard problem, and the other minds problem, in that we cannot deduce what is really going on inside the animal’s head, and neither can we feel it or explain what it feels like to cognize ourselves, never mind for animals.
The goal of this article was to emphasize how the research in cognitive neuroscience and behavioral evolution/ecology could benefit from being integrated when trying to explain animal cognition. The two methods explained for integrating these fields were the comparative approach and the fitness approach.
ReplyDeleteThe authors argue that the fitness approach is a better method to take on the road to integrating the two fields, and they give two reasons for that. One is that the fitness approach provides better linkage between proximal and distal causes of explaining animal cognition. The other advantage is that the fitness approach can help identify the agent(s) that selected for a cognitive trait, or give clue to what ecological environment selected for that specific trait.
I think this article is very well intended. However, as mentioned by many others, a lot of the supporting evidence seems like it's a reach at some points, and there's a lot of inferring causal relationships based on correlational observations. In an ideal world, it would be interesting to know the distal causes for animal (and our) cognitive traits. But we don't necessarily need to know the distal causes to solve the easy problem.