I’ve been meaning to start a blog for a while, but unsure where to start. However today I was struck with some inspiration courtesy of the BBC:- Can dogs look guilty? using the nifty “Science behind it” site I have been able to locate the paper on pubmed. Now the media are ridiculously prone to getting the wrong end of the stick where science is concerned. Then running around making extraordinary claims before being whacked on the nose with a rolled up blog by the likes of Ben Goldacre and the Badscience lot. Unfortunately rather then running for the hills with their tail between their leg and looking sheepish the media generally continues to publish un-scientific shaggy dog stories.
Fortunately in this case the medias reporting is reasonably good – possibly because there is little potential for generating a health scare/miracle cure story about whether or not dogs can look guilty. Although the story has generated a bit of consternation in another group of people, which may or may not include scientists and the like, pet owners. Pet owners who can tell when their dog is guilty – I mean of course it’s not displayed in their face it’s the ways the dogs behave! It’s all about body language and behaviour not puppy dog eyes!
However, despite this rather common example of cognitive dissonance (which I like to refer to the “ah, but, but…” effect or the ‘clutching at straws paradigm’ (CLASP ho ho)), the research suggests that what pet owners interpret as guilt is in fact the response of the dogs to being scolded. So rather then detecting the emotion of guilt as the abstract of the paper states:- “These results indicate that a better description of the so-called guilty look is that it is a response to owner cues, rather than that it shows an appreciation of a misdeed”. To those of you familiar with the case of Clever Hans this probably doesn’t come as much of a surprise but for some reason pet owners (and many other hooman types) are want to anthropomorphise many different things from animals to apple macs from zebras to… to… um… some other animal or inanimate object beginning with ‘z’…
From Aesop’s fables to Terry Pratchett there is a strong literary tradition of anthropomorphism and anthropologist Stewart Elliot Guthrie has suggested that all religions are anthropomorphism’s that originate due to the brain’s tendency to detect the presence or vestiges of other humans in natural phenomena in his book ‘faces in the clouds: a new theory of religion‘. I’ll admit I’ve not read the full text myself but it seems a fascinating idea. However it is worth noting that the Abrahamic faiths oppose the anthropomorphism of God; the Greek philosopher Xenophanes (570–480 BC) said that “the greatest god” resembles man “neither in form nor in mind a view shared in part by Thomas Aquinas and others. Judaism and Islam both believe that God is beyond the comprehension of us mere humans.
Yet, cause of religion or otherwise, this doesn’t explain why people are willing to believe they know what their pets are thinking, that a German horse can be a mathematical savant (well as far as equine mathematical ability goes) and perhaps the most common example of anthropomorphism that apes can master language…
The last point will perhaps be considered controversial, as it’s a well documented phenomenon and the subject of much research. However the seeming abscence of complex languages emerging among primates outwith captivity should perhaps raise alarm bells. As should the fact that many of these apes that are thought to have mastered language have been taught and researcher not by linguists and the like but by their handlers and keepers… An excellent chapter in Marc Hausers’ “Wild Minds:-What animals really think” (the title itself could perhaps be held up as an example of anthropomorphism if you wanted to go a little Descartes…) addresses the idea that the great apes (and other animals) can produce and understand complex language.
In the late twentieth century (you know that one what we just had) researchers attempted to prove this one way or the other by means of teaching dolphins and apes to understand and converse in ASL (American Sign Language) or some “artificial” language usually symbolically based. Most of the research groups posted positive results of one kind or another suggesting the animals in their charge could make requests, understand symbols for abstract concepts, objects and actions. Although it’s worth perhaps mentioning (although it isn’t made explicit in said chapter) that the spectre of Skinner (a more refined Clever Hans effect was achieved by his Skinner boxes – seek them out if you are unfamiliar from Parrots playing basketball to guided missiles guided by pigeons there is something for everyone…) seems to hang over these experiments. As in both signing and artificial language research programs a subjects success for responding to a a symbol was met with either verbal praise or a reward of food (It is not known whether the subjects in these studies drooled in their future exchanges Pavlov style).
This has been combated in much of the research by showing that some apes can apply abstract concepts, such as half and same, across contexts – although a Clever Hans effect cannot fully be discounted without proper double blind studies taking place… I’m currently unaware of any such studies having taking place, although I am fairly sure it’s unlikely no one has never done such a thing – I welcome any comments with information on this. However based on my current level of knowledge I would tend to err on the side of caution – perhaps animals can be taught to use and understand symbolic communication but the absence of such communication occurring naturally outside the lab does make me wonder if it is more related to some sort of Skinner Box or Clever Hans effect then the unmasking of an innate ability… Or rather it is our tendency to anthropomorphise, with a smidgen of confirmation bias thrown in for good measure, that perhaps leads us to conclude that, not unlike Doctor Doolittle we can talk to the animals… and that they can talk back…
If we accept that, or even if we don’t and just accept that some pet owners/computer users tend to overstate perceived human attributes in their pets/blue screen of death crash boxes, then the question arises why is this?
To be entirely honest I’m not 100% up with any research in this area although my own personal postulation would probably involve the theory of mind. The theory of mind is the ability to attribute mental states—beliefs, intents, desires, pretending, knowledge, etc.—to oneself and others and to understand that others have beliefs, desires and intentions that are different from ones own.
I suspect that as an artifact of our ability, probably key to our success as a social animal, to detect and attribute others intents and mental states we tend to attribute the same things (erroneously) to animals and inanimate objects. By analogy the process would be similar to that which makes us see faces in tacos and other simulacra, except in this case it is emotions and intent that we are attributing where they might not exist.One potential issue with this idea is that research tends to show that we, as human animals, are quite good at gauging the intent of other humans and that we rarely, no matter how many times we may beg to differ with extreme prejudice, tend to believe that our computers/vacuum cleaners/other household appliances have minds of their own. I’d also hazard a guess that we’re probably quite good at gauging the intent of certain animals in fight or flight situations (if anyone knows of any research in this area I would be fascinated to see it).
Regardless I still think that treating anthropomorphism as an artifact of our theory of mind could perhaps be an interesting avenue of research. Perhaps exploring if individuals who happen to be on the autistic spectrum use or understand anthropomorphism could be one angle to take?
Anyway I appear to have blogged (and now I feel a little bit dirty) I’d appreciate any comments/suggestions or debate about the above post.