Species Categories and Psychophysics

• Welcome to the BOL Forum on Evolutionary Psychology by John Grohol, 3/19/97
  • (...)
    
    • Species: Got Any New Ones Lately? by James Brody, 6/14/97
      
      

Extracted from the Paleo_Psych ListServe:

"One argument that has been made for the ``reality'' of species is that they're often classified the same way by different cultures. For example, Murray Gell-Mann cites Ernst Mayr's account of finding 127 bird species in New Guinea when he was there as a young researcher, while the local tribe counted 126, disagreeing with him only about two species of gerygone that they called one. (Gell-Mann, The Quark and the Jaguar, Freeman, 1994, p. 13) However all this establishes, if it is universally true, is that species break down the same to all people. Not to all the universe.

David Berreby, May 12, 1998

"Ed Wilson reports in one of his books that when he tried the same test, also in PNG, about ants, he got two classes (big ones and little ones) for several scores of species. Autochthonous classification is based around the sorts of discriminations that culture makes for cultural reasons, but when they make them, the classifications tend to be natural.

John Wilkins, May 12, 1998

"Behaviorally, we have hybrids. This happens even in my lowly spiders - populations of the same species can have very different courtship behaviors. In some cases, these differences are so different to cause one member to not recognize the other, even though they are the same species! And as my mentor Dan Papaj mentioned in his book "Insect learning" sometimes these behaviors are learned, not genetically wired.

Kelly C. Kissane to IPP 5/2/98

"My own work involves differences in courtship behaviors between populations of the fishing spider Dolomedes triton, which has led to some reproductive isolation....These animals can produce viable, fertile offspring when different populations are "force-mated" - a process that involves putting the female in a sleepy mode so she doesn't kill her suitor - and these offspring often are "hybrids" behaviorally, having characteristics of both parents. This is also seen in two sister species of Schizocosa, work by Gail Stratton found that hybrids were often rejected by both parental species, but the hybrids were responsive to each other.

Kelly Kissane to IPP 5/03/98"

---------------

Donald Blough once derived spectral sensitivity curves for pigeons and found they matched human performance. (Scientific American, July 1961; J. Optical Society of America, (1957), 47(9), 827-833) Thus, it is possible to test animal perception for its agreement with human perception.

Another worker -- I apologize for not recalling his name -- trained pigeons to peck photographic images of people (the positive stimulus) but not those of non-people. He then tested them on two new images, one of a house and the second of a house with smoke coming from the chimney. The birds responded positively to the "smoke" on the first trial. One possibility is that pigeons form concepts of "people."

Cross species agreement be assessed for species identification. Similar techniques to Blough's could test agreement between humans and other primates or humans and various birds. "Agreement" might be a function of "survival relevance" (Birds track species of foxes? Of other birds? "Alpha" status of members of their own species?), phylogeny (Birds agree with other birds? With humans?), or ontogeny ("Agreement" occurs at some developmental ages but not others).

There are a couple of reasons:

Categorical sorting of species (a la Thurstone) produced impressive interobserver agreement (usually) with humans. Studies of the evolution of cognition might well gain from assessing systematic shifts in category formation in relation to species, survival relevance, and mating preferences.

The phenomenon of "categorization" appears relatively simple to describe if we borrow one of Stu Kauffman's models. "Phase transitions" appear to supply an on/off switch for clusters of decision units, whether neurons or groups of people. Category judgments usually entail a binary decision rather than analog. It's then a bit easier to imagine language coming a bit later to match "one word with one referent" in a useful but sometimes misleading way.

Besides, it would be fun to know if birds sort other birds (or people) the same way that we do ... if their mating will match their visual decisions? If they distinguish between predators such as different kinds of hawks?

Phylogenetic history (our closer relatives make the same judgments that we do) could be a variable as well as ontogenetic (agreement with us as a function of age and experience).

• Sign Stimuli and Human Taxonomies by John Fentress, 5/18/98