"It has been recognized for decades that the development of all phenotypic characteristics depends on the interaction of nature and nurture, genes and the environment, evolution and learning"
The above quote, and its different interpretations that differently value the relative contributions of biology and environment to cognition, is central to the debate re: what *and how* neuroscience can contribute to our understanding of cognition, and is also the reason why the answer is largely unclear. It’s hard to know how relevant neuroscientific findings are to explaining cognition, because we really don’t know what the various functions that we observe in the brain mean or imply. Although mirror neurons are an observable phenomena, we are unsure if their development is driven by the environment by way of associative learning, or by biology in being an adaptation. The answer, though, has implications on what the main driver of our cognitive abilities is: biological or environmental processes. Taking recent advances in epigenetics into account, like evidence that environmental effects can alter the genome during the course of development, the above quote should be modified to read “the development of all *genotypic* characteristics….” Evidence demonstrating that environment influence is fundamental to neurological development and processes, and just as “low level” as gene expression, make it nearly impossible to disentangle the causes of brain functions and draw meaning about what we observe in the brain. It is unclear whether observable neural phenomena explain a human’s behaviours or purely describe them, if the observations are a result of genetics, the environment, or interactions between the two, what the order of environmental and genetic influence is (if there even is an order), and whether they exist as some innate capacity or as a pure by-product of our particular place in time and space. It just seems like any conclusion about cognition that is based in what in what is observed in the brain is very weak, and can do little to explain what leads to and accounts for the way we think and cognize in the way we do.
It would be hard to imagine that those neurons encode the “goal” of motor behavior of another individual, when goal directed movement can be highly context-dependent and include many variables. So those same neurons are firing to different goal-directed actions, as MN corresponding to “goals” would be overly complex and our brains simply cannot store all the value and intentions at any one time. Also I agree with the author that this term of “action understanding” is problematic, and that its unclear how the genetic hypothesis aims to distinguish the term from action perception. By using the term “understanding,” the proponents of the genetic hypothesis have made it very difficult to ever definitively support this empirically as its not possible to prove that neurons can encode “understanding,” as defining this term becomes incredibly complex and subject to the other-minds problem. The associative learning hypothesis is intriguing as it leaves more room and direction for artificial intelligence, because if mirror neurons function as part of the overall associative learning system, then potentially sensorimotor grounding and unsupervised machine learning as a way to model cognition would seem more feasible.
After reading this article, I had similar thoughts about the question of defining and measuring “action understanding”. This seems to be a major flaw in the genetic account, which theorizes that mirror neurons originated specifically for this purpose. It seems to me that the overemphasis placed on the function of mirror neurons by proponents of the genetic hypothesis has elicited misinformed assumptions and media claims based on limited research. I agree with Cook et al. that the more revealing question to ask in hopes of gaining further insight into human cognitive processes primarily concerns the origins of mirror neurons. One of the strengths of the associative account—that mirror neurons develop through a process of sensorimotor learning—is that it is “functionally permissive; it does not deny that MNs make a positive—possibly even adaptive—contribution to social cognition.” Thus, evolution and learning can be seen as complementary.
RE: (1) sensorimotor learning plays a crucial, inductive role in the development of MNs, and, because of this, (2) we will get reliable information about the function of MNs only by applying an approach based on developmental history
The authors state that “Whether or not an individual has MNs, which actions are encoded by their MNs, and at what level of abstraction, will all depend on the types of sensorimotor experience received by the individual in the course of their development”
I found the importance of environmental factors (that are internalized as sensorimotor experience) on the development of MN highlighted in this article interesting. They argue that in order to understand the mirror mechanisms, we must control for the sensorimotor experiences the organism is exposed to. In attempting to reverse engineer cognitive capacity (that is partly implemented in MN), is it necessary to understand this sensorimotor experience? What implications does this have for the reverse engineering of a T2 that lacks sensorimotor experience?
"It has been recognized for decades that the development of all phenotypic characteristics depends on the interaction of nature and nurture, genes and the environment, evolution and learning"
ReplyDeleteThe above quote, and its different interpretations that differently value the relative contributions of biology and environment to cognition, is central to the debate re: what *and how* neuroscience can contribute to our understanding of cognition, and is also the reason why the answer is largely unclear. It’s hard to know how relevant neuroscientific findings are to explaining cognition, because we really don’t know what the various functions that we observe in the brain mean or imply. Although mirror neurons are an observable phenomena, we are unsure if their development is driven by the environment by way of associative learning, or by biology in being an adaptation. The answer, though, has implications on what the main driver of our cognitive abilities is: biological or environmental processes. Taking recent advances in epigenetics into account, like evidence that environmental effects can alter the genome during the course of development, the above quote should be modified to read “the development of all *genotypic* characteristics….” Evidence demonstrating that environment influence is fundamental to neurological development and processes, and just as “low level” as gene expression, make it nearly impossible to disentangle the causes of brain functions and draw meaning about what we observe in the brain. It is unclear whether observable neural phenomena explain a human’s behaviours or purely describe them, if the observations are a result of genetics, the environment, or interactions between the two, what the order of environmental and genetic influence is (if there even is an order), and whether they exist as some innate capacity or as a pure by-product of our particular place in time and space. It just seems like any conclusion about cognition that is based in what in what is observed in the brain is very weak, and can do little to explain what leads to and accounts for the way we think and cognize in the way we do.
It would be hard to imagine that those neurons encode the “goal” of motor behavior of another individual, when goal directed movement can be highly context-dependent and include many variables. So those same neurons are firing to different goal-directed actions, as MN corresponding to “goals” would be overly complex and our brains simply cannot store all the value and intentions at any one time. Also I agree with the author that this term of “action understanding” is problematic, and that its unclear how the genetic hypothesis aims to distinguish the term from action perception. By using the term “understanding,” the proponents of the genetic hypothesis have made it very difficult to ever definitively support this empirically as its not possible to prove that neurons can encode “understanding,” as defining this term becomes incredibly complex and subject to the other-minds problem. The associative learning hypothesis is intriguing as it leaves more room and direction for artificial intelligence, because if mirror neurons function as part of the overall associative learning system, then potentially sensorimotor grounding and unsupervised machine learning as a way to model cognition would seem more feasible.
ReplyDeleteHi Julia,
DeleteAfter reading this article, I had similar thoughts about the question of defining and measuring “action understanding”. This seems to be a major flaw in the genetic account, which theorizes that mirror neurons originated specifically for this purpose. It seems to me that the overemphasis placed on the function of mirror neurons by proponents of the genetic hypothesis has elicited misinformed assumptions and media claims based on limited research. I agree with Cook et al. that the more revealing question to ask in hopes of gaining further insight into human cognitive processes primarily concerns the origins of mirror neurons. One of the strengths of the associative account—that mirror neurons develop through a process of sensorimotor learning—is that it is “functionally permissive; it does not deny that MNs make a positive—possibly even adaptive—contribution to social cognition.” Thus, evolution and learning can be seen as complementary.
RE: (1) sensorimotor learning plays a crucial, inductive role in the development of MNs, and, because of this, (2) we will get reliable information about the function of MNs only by applying an approach based on developmental history
ReplyDeleteThe authors state that “Whether or not an individual has MNs, which actions are encoded by their MNs, and at what level of abstraction, will all depend on the types of sensorimotor experience received by the individual in the course of their development”
I found the importance of environmental factors (that are internalized as sensorimotor experience) on the development of MN highlighted in this article interesting. They argue that in order to understand the mirror mechanisms, we must control for the sensorimotor experiences the organism is exposed to. In attempting to reverse engineer cognitive capacity (that is partly implemented in MN), is it necessary to understand this sensorimotor experience? What implications does this have for the reverse engineering of a T2 that lacks sensorimotor experience?