Reasoning from animals to humans

A fascinating recent article on an ability usually thought to be human, but apparently shared by some birds:

von Bayern, A.M.P., & Emery, N.J. (2009). Jackdaws respond to human attentional states and communicative cues in different contexts. Current Biology, 19 (7), 602-606.

These researchers looked at the ability of jackdaws to pay attention to the eye movements of nearby humans-- a skill that human beings show from an early age. Human infants develop in the first year of life the ability to look at another person's gaze and figure out what he or she is looking at, as well as the ability to use their own eyes as "pointers" and attract another person's attention to something by looking at the object and then back at the person. However, animals like chimpanzees and dogs do not seem to be able to "read" eyes in this way, although they pay attention to the way a person's head is turned. Can birds demonstrate a skill that chimps and dogs cannot?

von Bayern and Emery put out food for their jackdaws and then timed how long it took them to approach it when an unfamiliar (so, from the jackdaws' viewpoint, possibly dangerous) person was gazing at it. The jackdaws hesitated to approach the food that a stranger was looking at, but readily approached it when a familiar person gazed at it. But the jackdaws did not seem to be able to follow a steady gaze to find food they could not see; a familiar person had to shift the eyes toward the food and point in order to get the birds to pick up on the help that was offered.

Why should jackdaws be able to use humans' eye movements or gaze to get information, when "higher" animals cannot? Does this study mean that other birds can also pick up cues from humans' eyes? One possible reason for jackdaws--but not necessarily other birds-- to have this ability is that jackdaws have dark pupils and light irises, as many humans do. This pattern makes for a complex and attention-getting visual stimulus which may
be familiar to these birds because they have looked at other jackdaws. Of course, human eyes are even more complex patterns, because they have the surrounding white area as well as the contrast between pupil and iris.

Birds that have all-dark "shoe-button" eyes may not have experienced bird eye movements as interesting or important, so they may not pay attention even to the details of human eyes. In addition, the jackdaws in this study had been reared by humans and had had a good deal of face-to-face social interaction, with a chance to learn a lot about the meaning of human eye movements. Birds who grew up in the wild might react differently.

In fact, knowing about jackdaws does not tell us directly about other kinds of birds. Only similar research with other birds can do this. In the same way, research with human-reared jackdaws does not tell us directly about wild-reared jackdaws. Jumping from bird to human, or even from bird to bird, is a dangerous form of speculation unless we have a lot more information about each species.

The ethologist and Nobel prizewinner Nikolaas Tinbergen got into some dangerous speculation years ago when he jumped from studies of ducks to consideration of autistic children. Tinbergen proposed that attracting the gaze of autistic children would help them pay more attention to human beings and lead them to behave more normally. Tinbergen even suggested the use of masks with extra-big eyes that would catch the children's attention.

Tinbergen's method proved to be of little use, but the idea that autistic children avoid attention to other people's gaze continues to be heard frequently. However, researchers like M.A. Gernsbacher have shown some evidence that autistic children are not avoiding eye contact, but instead are able to see others' eyes when they are not in the center of the visual field, a difficult task for the non-autistic to carry out.

Tempting though it may be to jump speculatively from jackdaws to humans or from ducks to autistic children, we need to study each group carefully and separately, or risk making serious mistakes. It's particularly risky to assume that more "advanced" creatures have all the characteristics of all the less advanced ones, and additional characteristics as well. Most of us would consider a chimpanzee more advanced than a jackdaw, but it appears that the jackdaw can do at least one important thing that the chimp cannot.

The animal-human analogy

Several days ago, I commented on an article by Price in which information about mouse behavior was
[over]generalized to suggest that human beings showed similar factors at work in maternal behavior. This tendency to equate behaviors of different species has become quite common, especially in textbooks for undergraduate courses, where authors hesitate to acknowledge that they are using animal studies. (This hesitation is not just because of the problems human-animal analogies create, but because of the strength of organizations like PETA that object to experimental work with animal subjects.)

The issue of human-animal comparisons is one we should think about in this Darwin anniversary year. My concern is not about rejecting all such comparisons. Some do reject the analogy on the grounds that humans are special, made in God's image, have souls, etc., and thus are not comparable to non-humans. I accept the comparison and think it can be valuable-- but only when it is used with caution.

The human-animal analogy is easily "abused" by people who consider that what is known about one must provide reliable information about the other. In fact, such reasoning only works when careful comparisons have been made of the characteristics the two groups have in common AND that are relevant to the question at hand. Mice have fur, humans have hair; this commonality is relevant to skin functions, but probably not to the study of maternal behavior. Humans mourn over dead babies; mice eat them-- does this difference mean they have nothing relevant in common? Only attention to detail can tell us where we are in any comparison of one species to another.

As J.P. Scott showed many years ago, not only species but strain differences shape social behavior in dogs. Comparing one strain of one species to members of another species may give quite different results than if another strain were used in the comparison.

Studies of animal behavior are fascinating and valuable in themselves. Sometimes it may be possible to use their data to explain human characteristics-- but sometimes, and more often, not. To find out about a species, it's best to study that species.