“To increase our ability to use phylogenetic methods to study cognition, we will need to…consider how to assess cognition as a trait, representative of a species ability to solve a particular problem”
This is a very important question to address not just between species but also within, especially when it comes to cognitive states that lack an overt behavioral response. MacLean et al. suggest that “performance on a single task” may, in some cases, be able to reflect cognition as a trait. I found this claim troubling—I can’t think of an example of such a task. How can a single variable encompass the entire realm of cognition? The domain approach surely yields much more representative measures. In attempting to understand the evolution of cognition by comparative analysis, it is important that cognition is not treated as a unitary trait since different abilities evolve at different rates/times both within and across species. Appropriate tasks must be designed for the specific cognitive abilities under test, and task performance must not be interpreted as fully or faithfully reflecting an individual’s or group’s cognitive ability.
I am a bit confused about phylogenetic targeting. The authors describe how there are statistical issues with the phylogenetic signal if the sample or data set is too small. Does phylogenetic targeting avoid the problems of phylogenetic signal detection because it counts on using species that are highly related and so it has less confounding variables? Are there rules for how closely related the species have to be and how can statistical power be assured if there is no reliable error signal on the smaller data set?
Hi Julia, I have some more issues with phylogenetic targeting that I want to add to your point!
When they restrict species choices to improve the relevance of the obtained data, I think they end up producing biased data. By trying to control for a specific set of results they natural fall into confirmation bias while more accurate data could be excluded by the phylogenetic targeting algorithm constraints.
I like the approach of studying how cognition evolved in other animals might provide insight into the development of human cognition. However, I am also concerned about potential abusive experiments performed on animals which might render the whole methodology unethical.
Does it help to study cognition in other species with the assumption that there could be different types of cognition? Or should we look for general cognitive processes?
Is species difference in cognition the differences in the processes that consciousness allows the species to carry out? I.e. functional differences. Or are we looking for differences in the makeup of cognition i.e. structural differences. The latter inevitably becomes irrelevant if implementation independence is assumed. Doesn’t using a phylogenetic framework for comparison of cognition across species imply that we are examining the hardware of cognition because phylogeny is the study of the development of a feature such as an organ? My final question is: do MacLean et al. believe that we might understand HOW cognition works if we better understand why it works and how it is constructed?
The experimental methods associated with phylogenic comparative research are extremely flawed. I think until better methods are created, conclusions cannot be drawn from this type of experimental evidence. First, adapting an appropriate cognitive task for another species to be administered in an artificial environment to elicit some psychological phenomenon that is observed in humans is near impossible -- not only does the task need to produce a measurable response, but that response needs to accurately reflect whatever phenomenon is under study. It’s possible that the response tested could reflect a completely different trait. For example, in the research agendas that investigate whether dolphins are more optimistic at specific times, and if birds can synchronize to a beat like humans, the animals have been rigorously trained to perform these behaviors and engage in these tasks, and the threshold of meeting criteria is quite low. The latter example reached a positive conclusion based on the observation that a bird synchronized to a beat in bouts that were interspersed between longer episodes of unsynchronized dancing, only occurring when the beat was close to the bird’s natural rate of rhythmic motor activity. Experimental evidence can tell us what animals are capable of learning, but may just reflect artificial behavior that has no meaningful contribution to evolutionary research. Human researchers may be merely personifying behaviors seen in animals to fit research agendas.
Firstly, I found that this article focuses on the evolutionary development of the brain, but does not stress enough the social factors that play a great role in the cognition of animals. For example, when deciding which species to select for data collection in phylogenetic targeting, the researchers empathize brain size and phylogenetic independence but do not mention the effects of different social patterns in the species tested. The second comment on this article is the problematic animal testing to attempt to answer questions about human cognition. To what extent is animal intelligence and their cognitive capacities comparable to ours? How do they differ? What are the ethical repercussions in studying animals?
In the previous reading’s skywriting I mentioned how I thought that biology and genetics are components that interact with each other and other higher order components that make up human cognition. The modularity of human brain and our ability to cognize seems to have such multitudes that it seems almost impossible to distinguish. I find it harder and harder to attribute any task as a direct measure of cognition. I have found myself thinking more and more of Chomsky’s Universal Language. There are clear genetic and biological foundations to UG, but its function is both physical (the way we interact, how our mouths move, etc) but also cognitive (how it shapes how we think). This is universal across our species. Although this may seem like a clear task as a measure of cognition, we do we distinguish the physical from the mental?
“To increase our ability to use phylogenetic methods to study cognition, we will need to…consider how to assess cognition as a trait, representative of a species ability to solve a particular problem”
ReplyDeleteThis is a very important question to address not just between species but also within, especially when it comes to cognitive states that lack an overt behavioral response. MacLean et al. suggest that “performance on a single task” may, in some cases, be able to reflect cognition as a trait. I found this claim troubling—I can’t think of an example of such a task. How can a single variable encompass the entire realm of cognition? The domain approach surely yields much more representative measures. In attempting to understand the evolution of cognition by comparative analysis, it is important that cognition is not treated as a unitary trait since different abilities evolve at different rates/times both within and across species. Appropriate tasks must be designed for the specific cognitive abilities under test, and task performance must not be interpreted as fully or faithfully reflecting an individual’s or group’s cognitive ability.
I am a bit confused about phylogenetic targeting. The authors describe how there are statistical issues with the phylogenetic signal if the sample or data set is too small. Does phylogenetic targeting avoid the problems of phylogenetic signal detection because it counts on using species that are highly related and so it has less confounding variables? Are there rules for how closely related the species have to be and how can statistical power be assured if there is no reliable error signal on the smaller data set?
ReplyDeleteHi Julia, I have some more issues with phylogenetic targeting that I want to add to your point!
DeleteWhen they restrict species choices to improve the relevance of the obtained data, I think they end up producing biased data. By trying to control for a specific set of results they natural fall into confirmation bias while more accurate data could be excluded by the phylogenetic targeting algorithm constraints.
I like the approach of studying how cognition evolved in other animals might provide insight into the development of human cognition. However, I am also concerned about potential abusive experiments performed on animals which might render the whole methodology unethical.
ReplyDeleteDoes it help to study cognition in other species with the assumption that there could be different types of cognition? Or should we look for general cognitive processes?
Is species difference in cognition the differences in the processes that consciousness allows the species to carry out? I.e. functional differences. Or are we looking for differences in the makeup of cognition i.e. structural differences. The latter inevitably becomes irrelevant if implementation independence is assumed. Doesn’t using a phylogenetic framework for comparison of cognition across species imply that we are examining the hardware of cognition because phylogeny is the study of the development of a feature such as an organ? My final question is: do MacLean et al. believe that we might understand HOW cognition works if we better understand why it works and how it is constructed?
The experimental methods associated with phylogenic comparative research are extremely flawed. I think until better methods are created, conclusions cannot be drawn from this type of experimental evidence. First, adapting an appropriate cognitive task for another species to be administered in an artificial environment to elicit some psychological phenomenon that is observed in humans is near impossible -- not only does the task need to produce a measurable response, but that response needs to accurately reflect whatever phenomenon is under study. It’s possible that the response tested could reflect a completely different trait. For example, in the research agendas that investigate whether dolphins are more optimistic at specific times, and if birds can synchronize to a beat like humans, the animals have been rigorously trained to perform these behaviors and engage in these tasks, and the threshold of meeting criteria is quite low. The latter example reached a positive conclusion based on the observation that a bird synchronized to a beat in bouts that were interspersed between longer episodes of unsynchronized dancing, only occurring when the beat was close to the bird’s natural rate of rhythmic motor activity. Experimental evidence can tell us what animals are capable of learning, but may just reflect artificial behavior that has no meaningful contribution to evolutionary research. Human researchers may be merely personifying behaviors seen in animals to fit research agendas.
ReplyDeleteFirstly, I found that this article focuses on the evolutionary development of the brain, but does not stress enough the social factors that play a great role in the cognition of animals. For example, when deciding which species to select for data collection in phylogenetic targeting, the researchers empathize brain size and phylogenetic independence but do not mention the effects of different social patterns in the species tested.
ReplyDeleteThe second comment on this article is the problematic animal testing to attempt to answer questions about human cognition. To what extent is animal intelligence and their cognitive capacities comparable to ours? How do they differ? What are the ethical repercussions in studying animals?
In the previous reading’s skywriting I mentioned how I thought that biology and genetics are components that interact with each other and other higher order components that make up human cognition. The modularity of human brain and our ability to cognize seems to have such multitudes that it seems almost impossible to distinguish. I find it harder and harder to attribute any task as a direct measure of cognition. I have found myself thinking more and more of Chomsky’s Universal Language. There are clear genetic and biological foundations to UG, but its function is both physical (the way we interact, how our mouths move, etc) but also cognitive (how it shapes how we think). This is universal across our species. Although this may seem like a clear task as a measure of cognition, we do we distinguish the physical from the mental?
ReplyDelete