3 ms·

The article mentions physical changes in the mice as a result of the smell (more specific receptors and neurons) which are explained within the current model by

The article mentions physical changes in the mice as a result of the smell (more specific receptors and neurons) which are explained within the current model by epigenetics: expression of "dormant genes" for those physical changes. So, if those physical changes explain the mouse behavior response, then this has nothing to do with memories[1], and is not an unexpected outcome of the study within the current model. This article is much too sensationalist, in my opinion. [1] for reasonable definitions of a memory, of course.

Sniffnoy 13y ago
I think you're possibly being overly narrow with your notion of "memory". Such a thing would not be a memory in the sense of something stored in the nervous system, but it would still be a memory in the sense of stored information based on past occurrences that influence future behavior. (And all memories are physical changes, after all!) And while that's certainly nowhere near as noteworthy as if memories from the nervous system were somehow being directly passed down, it does not seem to be a correct statement that this "is not an unexpected outcome of the study within the current model". I mean, sure, the model doesn't exclude this outcome, but it certainly seems that nobody expected it to happen in a case like this.
- computer 13y ago
  Related-- Would you call instinct a memory? Instinct does not have physical attributes that completely and directly explain it, so the described behavior here is less of a memory than much of (the rest of) instict is.
  - Sniffnoy 13y ago
    No, I wouldn't call instinct a memory, but I really think you're focusing on the wrong thing here. The relevant distinction here is not how "completely and directly" something is "explained by physical attributes"; to be honest, I'm not even really sure what that means. All behavior comes down to physical phenomena, after all; whether or not we can currently make that reduction explicitly is more a property of our current state of knowledge, than a property of the physical system itself. The relevant distinction here is the fact that this behavior, like a memory proper, was acquired during the mouse's lifetime, not inherited, and yet passed on to offspring anyway. And yes, as you've noted, instances of that had already been observed -- this wasn't something revolutionary -- but still this wasn't a case where people expected to see that.
    - hcarvalhoalves 13y ago
      Maybe the term "imprint" could serve well. I wouldn't be surprised if more instances of that experiment get discovered and reproduced, after all this is describing a feedback loop leaving patterns on the next iteration (parent -> genetics -> expression on descendant), and feedback loops are to be expected in all natural systems.
    - computer 13y ago
      I don't have a problem with the research itself, I love that. But the presentation in this article is extremely sensationalist, using the word "memory" to make it seem like memories (in the human interpretation) are actually being transfered. I have a big issue with using sensationalism in such a way that the average reader will read it in a way that's just completely wrong.

Sniffnoy 13y ago

I think you're possibly being overly narrow with your notion of "memory". Such a thing would not be a memory in the sense of something stored in the nervous system, but it would still be a memory in the sense of stored information based on past occurrences that influence future behavior. (And all memories are physical changes, after all!) And while that's certainly nowhere near as noteworthy as if memories from the nervous system were somehow being directly passed down, it does not seem to be a correct statement that this "is not an unexpected outcome of the study within the current model". I mean, sure, the model doesn't exclude this outcome, but it certainly seems that nobody expected it to happen in a case like this.

computer 13y ago
Related-- Would you call instinct a memory? Instinct does not have physical attributes that completely and directly explain it, so the described behavior here is less of a memory than much of (the rest of) instict is.
- Sniffnoy 13y ago
  No, I wouldn't call instinct a memory, but I really think you're focusing on the wrong thing here. The relevant distinction here is not how "completely and directly" something is "explained by physical attributes"; to be honest, I'm not even really sure what that means. All behavior comes down to physical phenomena, after all; whether or not we can currently make that reduction explicitly is more a property of our current state of knowledge, than a property of the physical system itself. The relevant distinction here is the fact that this behavior, like a memory proper, was acquired during the mouse's lifetime, not inherited, and yet passed on to offspring anyway. And yes, as you've noted, instances of that had already been observed -- this wasn't something revolutionary -- but still this wasn't a case where people expected to see that.
  - hcarvalhoalves 13y ago
    Maybe the term "imprint" could serve well. I wouldn't be surprised if more instances of that experiment get discovered and reproduced, after all this is describing a feedback loop leaving patterns on the next iteration (parent -> genetics -> expression on descendant), and feedback loops are to be expected in all natural systems.
  - computer 13y ago
    I don't have a problem with the research itself, I love that. But the presentation in this article is extremely sensationalist, using the word "memory" to make it seem like memories (in the human interpretation) are actually being transfered. I have a big issue with using sensationalism in such a way that the average reader will read it in a way that's just completely wrong.