Do captive animals vocally express what they feel? If they do, a study of their vocalizations may provide an objective means for assessing animal emotionality, a behavioral dimension central to the concept of psychological wellbeing.
All else being equal, happy animals probably make healthier, more reliable research subjects than chronically distressed animals. We believe that such emotional differences in behavior are reflected in the vocalizations of animals. And, because vocalizations can be acoustically monitored non-invasively and remotely, such records can provide continuous tracks of the emotional changes contingent upon treatment, environmental, and other conditions.
Vocal records may be regarded as informationally rich, yet inexpensive, forms of documentation. Any sound occurring within the range/sensitivity of a microphone may be captured in digital or analog format. Recordings may be continuous or taken in discrete time segments spaced over long intervals. They might be the calls of an isolated individual (e.g., an animal during post-surgery recovery) or of groups of animals (e.g., a group adjusting to social changes induced by the addition of a new member). The microphone does not have to be aimed directly at the vocalizing animal to catch its sound. And, although we might acoustically identify a vocalizing individual, an ID isn't essential to decipher the emotional content of its calls. Even non-vocal sounds produced as by-products of movement (e.g., explosive bouts of cage-banging by multiple individuals) may be useful indicants of temporal sequencing in emotional expression.
Acoustical recordings can be analyzed in a variety of ways, depending on the information desired. At one end of the analysis continuum, we might simply want to know the frequencies of occurrence of vocalizations (all that are emitted or, perhaps, just one call of particular interest). At the other end of the continuum, we might want to know how the patterns of amplitude and frequency modulation in an animal's calls encode its responses to significant behavioral and environmental events (e.g., a correct response while performing in a complex cognitive discrimination task; enrichment of housing conditions). For many kinds of questions and levels of analysis, the vocal record contains in intimate detail a temporally coherent stream of acoustical behavior.
For many species, there appears to be a close tie between emotional state and vocal pattern. Adult male rats forcibly placed in confined test chambers begin emitting their intense "22 kHz cry" while dropping boli and urinating. Rhesus macaques begin moving vigorously about within their cages while emitting anticipatory "coos" in response to the sounds of footsteps from outside their living quarters, especially if a caretaker hasn't recently paid them a visit. A wild-caught mongoose "screams" and attacks the gloved hand of a researcher trying to remove it from its home cage. Fear, elation, and anger are expressed in the vocalizations of three species of captive animals.
Do the same calls occur under natural conditions in association with analogous behaviors? Rats in a burrow facing a predatory cat emit "22 kHz cries." Although this ultasonic vocalization has been referred to as a "distress call," a similar cry may be emitted by males following ejaculation. We suspect that the calls emitted in these two situations differ in their patterns of amplitude modulation. Rhesus in trees "coo" excitedly on detecting signs of food, but they also emit a subdued "coo" when potential predators end a threat and visibly withdraw. The latter appears to signal relief rather than excitement. A mongoose cornered by a predatory dog may emit a series of increasingly hostile "pants" and "chucks" so long as the possibility of escape exits, but will "scream" and attack if it can't escape.
Under natural conditions, it appears that the meaning and acoustic nature of calls co-vary with the general character of situations, as does the emotional dimension and intensity of arousal underlying specific vocal types. The implication is that animal vocalizations express the same information in captive environments as they do in nature. They are vocal components of species-specific behavior patterns, naturally linked to arousing conditions and emotional actions.
These generalizations are drawn primarily from our research on the rhesus macaque. We selected this species of primate because it is one of the most frequently used research subjects in laboratories around the world. Given recent Federal regulations that mandate provisions for, and documentation of, maintenance of behaviorally healthy animals (especially non-human primates) in captive environments, it seemed essential that new, more powerful approaches to assessment of psychological wellbeing be developed. Enrichment efforts without a capability for documenting and assessing any resultant psychological effects may not be worth the cost and contribute little to scientific knowledge.
Our research began with acoustical analyses of rhesus vocalizations recorded over several years from a large colony of free-ranging animals (n 4,000) in naturally occurring social groups. From these data, we established the vocal repertoire and associated behavior of the rhesus under natural conditions which serves as the normative base needed to interpret vocal behavior recorded under captive conditions. We then obtained vocal and video records of rhesus maintained in a variety of captive environments: out-door vs. in-door, single-caged vs. social groups, research subjects vs. breeding stock, etc. Although our research is far from complete, the data already indicate that this approach will become fruitful. Vocal patterns do appear to vary with housing and treatment conditions. The expression of emotion is there in the calls, in captivity as it is in the field. The difference is that, in captivity, an emotional state may be induced by human-imposed conditions, which may or may not be desirable. At least we now have a way of examining the question.
Animal vocalizations appear to express pretty much the same things under captive and natural conditions. To the extent that this is true, vocal records provide an objective, real-time basis for monitoring emotional variation in captive animals. Effects of environmental enrichments, new caretaking procedures, more humane research protocols, etc., can be evaluated in terms of emotional impact as expressed in the vocal record. It seems that animals can tell us how they feel if we learn to listen and understand.
(Send communications to Dr. Michael R. Murphy, AL/OER, 8308 Hawks Rd, Brooks AFB, TX 78235-5102)
Bayne, K.A.L., J.K. Hurst, and S.L. Dexter (1991). Evaluation of the preference to and behavioral effects of an enriched environment on male rhesus monkeys. Laboratory Animal Science 42:38-45.
Cheney, D.L. and R.M. Seyfarth (1992). Precis of how monkeys see the world. Behavioral and Brain Sciences 15:135-182.
Cuomo, V., R. Cagiano, M.A. DeSalvia, M.A. Maselli, G. Renna, and G. Racagni (1992). Ultrasonic vocalization as an indicator of emotional state during active avoidance learning in rats. Life Sciences 50:1049-1055.
O'Neill, P.L. and C.S. Price (1991). Customizing an enrichment program: Rhesus monkeys. Lab Animal 20:29-42.
Snowdon, C.T. (1990). Language capacities of nonhuman animals. Yearbook of Physical Anthropology 33:215-243.
Woolverton, W.L., N.A. Ator, P.M. Beardsley, and M.E. Carroll (1989). Effects of environmental conditions on the psychological wellbeing of primates: A review of the literature. Life Sciences 44:901-917.
Go to:
Contents, Animal Welfare Information Center
Newsletter
Top of Document
The Animal Welfare Information Center
U.S. Department of Agriculture
Agricultural Research Service
National Agricultural Library
10301 Baltimore Ave.
Beltsville, MD 20705-2351
Phone: (301) 504-6212
FAX: (301) 504-7125
Contact us: http://www.nal.usda.gov/awic/contact.php
