released October 21, 2003
Mark Feinstein, a Hampshire College cognitive science professor interested in bioacoustics, has in the course of his research on animal communication found a method of measuring stress in sheep through recording and analyzing their vocal behavior. His research may have positive economic and humane consequences as well as shedding new light on the vocal signals of sheep.
To some, stress in sheep may sound like a joke, but to farmers, agronomists and animal welfare experts it’s a serious subject, with concerns ranging from humane treatment to the economic impacts of reduced reproductive rates.
Dr. Feinstein, whose past research, working with Hampshire colleague Ray Coppinger, has included the vocalizations of sheep-herding and sheep-guarding dogs as well as New Guinea Singing Dogs, spent academic year 2002-2003 in Ireland as a visiting scientist at Teagasc, a government agricultural research organization.
While there, he conducted a series of experiments in different stress-inducing circumstances, including isolation of individual sheep and comparisons of vocal behavior of just-weaned sheep with their vocal behavior after a period of adjustment to separation from the mother, and found a pattern of well defined acoustic markers of stress.
Feinstein is continuing his research on vocalizations of stress by sheep at the Hampshire College Farm Center in Amherst, with Hampshire College students joining him as research assistants.
Feinstein’s findings—presented to the Irish Agricultural Research Forum 2003 and published in the proceedings—are significant for a number of reasons:
They run counter to the belief that sheep, unlike some other animals including cows, do not vocalize stress. What differed in Feinstein’s work from prior studies of sheep is that he investigated the full acoustic spectrum of each bleat, not just the basic pitch of a signal or the number of vocalizations an animal made. Looking at the fine-grained acoustic characteristics of individual signals, he found distinctive and consistent markers within waveforms that showed a statistical correlation with stressful conditions.
They suggest that sheep reflect stress in their vocalizations (and possibly communicate it to other animals) by altering the timbre of their vocalizations, or the overall quality of sounds, rather than by changes in pitch or loudness over time. This ability is characteristic of human speech, but most other animals are unable to change the timbre of a signal as they have less ability to manipulate their vocal tracts.
Past methods of measuring stress in sheep have involved invasive procedures, drawing blood to measure corticosteroids or hooking an animal up to a heartbeat monitor. Feinstein’s method is non-invasive, using an external device that records the sounds and measures waveforms. When he presented his findings at a national agricultural conference in Ireland, officials developing European guidelines for humane treatment of animals were interested in the possibilities presented by Feinstein’s system.
Monitoring stress within a flock has economic implications for farmers concerned about reproductive levels and quality of food.
Elaine Thomas, Director of Communications, 413.559.5482,
Dr. Mark H. Feinstein, 413.559.5551, firstname.lastname@example.org