(A sort-of followup to Cats, Dogs, Strings, and Causality and A Small, Informal Cat Cognition Experiment)
Today I tried looking up studies and/or scholarly articles on the phenomenon of Cats Knocking Things Over, but didn't come up with any interesting results. The majority of writing on this subject seems to be in the context of advice on "cat behavior problems", e.g., Dealing With Cats That Knock Things Down, How can I get my cat to stop knocking stuff over?, etc.
(Of course, if anyone has any links to papers on this subject please feel free to share!)
But anyway. It occurred to me, upon seeing Shadow (one of three ex-feral littermates sharing my home) push a container of cat treats off my desk for the nth time earlier today, that (if it was indeed being done deliberately) such an action might represent a fairly well-developed understanding of certain physical principles. I.e., the fact that if one cannot readily access the contents of a treat-containing object, one might be able to gain access via utilizing the tendency of objects to fall over when pushed or similarly manipulated.
(Full disclosure: as a person on the autistic spectrum who is also interested in neuroscience and relevant cognitive research, I must admit it tends to catch my attention whenever I encounter what to me looks like an interesting ability or trait, in any species, being largely written off as a "behavior problem". Often it seems to me something is being missed when this occurs, so I'm driven to investigate in situations like this!)
Certainly the standard "I am not a professional researcher, this was all done completely informally, my home is not a laboratory, etc." disclaimer must be applied to these results. Moreover, I am well aware that interpretation is not data (and vice versa), and in truth the only thing that can be said for sure is that at least one of my cats pretty consistently knocks over objects when is is conceivable that he has reason to believe these objects could contain treats.
The following three videos appear in chronological order. All were filmed on 5 September 2010, in the afternoon, within the space of maybe twenty minutes.
In this first video ("Cats and Gravity I"), Shadow is shown pushing a sealed container of treats off my desk.
He had precedent for doing this, as this particular container is the usual one I put the kitties' daily allotment of treats in. Once, about a week ago, I left the lid part-way off, to see what he would do (seeing as he'd definitely be able to smell the treats within). And in that case he proceeded to nudge the container around with his nose until it fell off the table, scattering treats hither and thither, much to his and his siblings' delight.
Now, I am fairly certain that this first knockdown was an accident. However, since then, Shadow has pushed the treat container off multiple other surfaces (besides the coffee table), on multiple separate occasions. "Cats and Gravity I", then, seems like it could very well represent Shadow's having learned that "if I push this container, sometimes treats fall out!" I don't know that anyone in the cognitive research field actually believes this level of reasoning is beyond the domestic feline (I suspect not) but in any event, it makes for a good "baseline" data point in terms of the variables I am interested in observing.
In this next video ("Cats and Gravity II"), we have an interesting situation involving three cats. You may remember Coraline and Brodie from all those string and zip-tie trials I ran as a rough check of what experimental design conditions might be improved so that cats could better demonstrate their actual cognitive capacities in string-pulling tasks. In the case of that set of puzzles, Cora and Brodie were the only participating felines; Shadow preferred to simply watch.
However, in "Cats and Gravity II", you will observe that Brodie "experiments" with the treat-containing bottle but doesn't succeed in getting anything out of it, whereas Shadow makes one single decisive swipe and sends the thing crashing down. Note that this bottle is different from the treat container in the first video, but it is one that I've put treats in several times prior to this, so the cats would definitely be familiar with it.
(Coraline, meanwhile, is ignoring the whole business and seems more excited about the fact that I've gotten up from my computer chair, giving her an opening to steal it. To me this mainly suggests that at different points in time, different cats may have very different priorities!)
And then we come to "Cats and Gravity III". The outcome of this scenario completely caught me off guard (you may even be able to hear me exclaim "HOLY CRAP!" at one point). In this case I took a treat container that would be new to all the cats (another empty vitamin bottle, this time a dark purple one slightly larger than the white one used in "Cats and Gravity II"). I let them watch me putting treats into it, and then placed this bottle on top of a small end table in the living room (a different surface in a different part of the house than my desk, to control for position habit).
In this case, the video shows that initially none of the cats really showed interest in the new treat bottle when I first placed it on the end table. However, after I went over and shook it a bit, Shadow went over to the table, put his front paws on the top surface, and then proceeded to lift the bottle up with his mouth and throw it down onto the floor. (That's where I exclaimed "HOLY CRAP!", by the way.)
So...what to make of this?
Subjectively speaking (yes, I'm about to offer an interpretation), it looked to me like Shadow spontaneously came up with a really creative way of getting the treats he knew were inside the purple bottle. Which would suggest that he's learned to generalize beyond "if I paw at this maybe it will fall and treats will come out" and now understands that it is not the mechanical motion of pawing or nosing that's important, but rather, the falling of the bottle itself, if one's goal is to get the treats out of the bottle. This, to me, seems pretty significant, and again I'm curious to know if there's any literature out there saying one thing or another about this type of cognition in felines.
However, this was of course a tiny sample set. And I did not do this series of "mini-trials" in response to another study I'd be able to cite and/or comment on -- like I said at the beginning, I couldn't find any studies about cats knocking stuff over. No peer-reviewed references = not "ResearchBlogging". Plus, for all I know, cats' understanding of gravitational cause-and-effect is already well documented and known and I just fail at searching for this documentation.
That said, at the very least, I think anyone who really wants to study cats' understanding of cause-and-effect as it pertains to objects in a broader sense would do well to try out a variety of different scenarios involving different types of objects, and requiring different types of attentiveness and planning on the cats' part.
I find it terribly problematic (and this goes back to my discussion of the string experiments again) when a single particular test is taken (whether by the study authors, the media, or both) as meaning something globally significant about a given population's abilities or lack thereof. In the absence of a single task (or task type) with huge amounts of existing data backing up its ability to test "general" cognitive ability in a given domain, multiple tasks of varying attributes would seem to me required for appropriate levels of rigor.
Also, I have to say that another reason I wanted to post these videos is because now more than ever I am beginning to think it is very important to have as much of an experiment on record (for multiple parties to view and evaluate) as possible. Even though I (hopefully) disclaimered the heck out of my string experiments, I still would rather do things as close to "right" as possible for a layperson -- just because I'm not a real researcher doesn't mean I can't practice holding my informal stuff to higher standards.
Finally, I would just like to say that I would be extremely interested to get people's comments on what it looks like is actually happening in the videos above. As in, if you think Shadow is doing what he's doing deliberately, what aspects of his actions lead you to think that? I'm curious about this because I see my cats doing all sorts of things all the time, some of which (to me) look "deliberate", whereas other things they do look thoroughly "accidental". Only I haven't come up with a good way to describe what "deliberateness" looks like in quantitative terms.
I suspect that in general this sort of issue comes up a lot in animal cognition research, which has me curious as to whether there even exists any kind of objective way to measure something so "internal". Behaviorism (in my opinion) fails miserably to account for everything that could potentially be important (for one thing it often seems to completely fail to account for, say, different sensory and perceptual modalities on the part of the researcher vs. subject), and much of what I hear from "evolutionary psychology" sounds like it's been pulled straight from someone's nether orifice, to put it politely. So I'd be really intrigued to know what other tools or paradigms may currently be out there that might be more promising.