LETTERS TO THE EDITORS
To the Editors:
Robert Provine's article "Yawning" (November-December
2005) nicely reviewed his work on this interesting topic. I
disagree, however, that contagious yawning is restricted to humans
and chimpanzees and that it thus reflects the high-level cognitive
processes and social complexity of humans and our closest relatives.
In their 1967 paper in The Auk (84: 571-587), Edgar G. F.
Sauer and Eleonore M. Sauer presented (admittedly non-experimental)
evidence suggesting that yawning is also contagious among ostriches.
In addition, other behaviors such as preening, grooming, dustbathing
and feeding are often contagious in many species of birds and mammals.
Some confusion may result from the fact that most authors use the
term "social facilitation" rather than "contagious
behavior." While these terms are synonymous, many have
incorrectly defined social facilitation too broadly to also include
behavior that increases in the mere presence of other members of the
Staten Island, NY
Dr. Provine responds:
Sauer and Sauer observed that "a bird that yawns quite often
makes one or several neighboring birds yawn too, and yawning
… can spread out like a 'snowball effect.'" However,
unlike the fact of yawning, proof of human-like contagion in a group
requires experimental verification that observed yawns are the
vector for the synchronized response, not a commonly experienced
physiological state (e.g., waking, drowsiness) or stimulus.
This higher standard for evidence has been met only in humans and
common chimpanzees, although contagion need not be exclusive to
primates. Until the necessary research is conducted, the less
restrictive term "social facilitation" better describes
ostrich yawning and other socially synchronized acts (say, grooming
or feeding). A startled herd does not experience contagious running.
Where it is found, contagious behavior is hardly the pinnacle of
social engagement. Yawning and other contagious behavior are better
evidence for a low-level neurological substrate for sociality than a
"high-level cognitive process."