corvus
Posted : 1/4/2011 5:52:53 PM
Kevin Behan
But at any rate I'm not saying that dogs don't have a memory, but rather that they are informed by a "physical memory" so that the present moment's intensity/resistance value triggers these older memories and thus the dog relives the past by way of the present.
That seems to be more or less what researchers are saying about episodic memory in animals, except that it appears some animals have some sense of time in that they can make decisions based on how long it has been since they last had this same decision before them (see the chickadees paper, for example). In a controlled environment it's possible to take away all other possible cues and rely entirely on the animal's internal sense of time, whether that be an internal clock or a sense of time relative to sequences of events or "long" vs "short" periods. I suspect the latter. I think it's useful to note that some species have skills that other species do not. Animal species that cache are well known for spatial and temporal memory, and I think it's ground squirrels that have such an extensive and detailed suite of vocalisations that it has been bandied about that they have language. They aren't especially brilliant; they just live in an environment and have a social structure where auditory communication is especially useful.
Kevin Behan
I believe that a dog senses the affective change from compression (right before the explosion) to release (this can be a coherent kind of behavior rather than a simple overload explosion) and that this is a basis for responding coherently to the way things change, but not in the linear way of understanding that A caused B and this led to C. Rather, my premise is that from the dog's point of view, whatever it is feeling it then associates with the process of change, in other words, just as if its feelings are what caused things to change.
I think we can all agree that dogs (and other animals, including humans) tend to be superstitious about what they can affect and what they can't. We all know about the confirmation bias. I think the main problem with this idea of feelings causing changes for me is simply clicker training. My dogs have a definite "training mode" that I (and Panksepp, as it happens) equate to SEEKING. When they are in this mode they are all about working out how to get a mark. There is no discernable change in their demeanour when they get a click, generally speaking. We know they get a little surge of dopamine, which is thought to mediate learning. They learn to do stuff without changing emotional states as far as I'm aware. So their feelings can't be what caused things to change for them because they can learn several different behaviours in one training session, all without a change in feelings. I think I must be missing something again.
Mind you, I don't disagree that it can occur the way you suggest. Aggression would be a good example. Dog feels pressured, strikes out, and the person/dog backs away and the dog relaxes. I (and Steven Lindsay, it seems), think this is an opponent-process, where over time the pressure before the dog strikes lessens, but the feeling of relief after the dog strikes becomes stronger. You must love the opponent-process theory.