Robert M. Seyfarth - US grants

The overall aim of research is to increase understanding of the vocal communication of non-human primates. This is important both for a general understanding of animal communication and because research on non-human primates can clarify ways in which these species may potentially serve as animal models for the study of human speech. Our specific focus is vocal development. Subjects are captive rhesus and Japanese macaques, housed in six social groups that accurately reflect the two species' social organization under natural condition. Over the next three years we will use observational sampling, tape- recording, and computer-based acoustic analysis to document the normal course of vocal development in each species. Simultaneously, we will cross-foster infants between species, thereby exposing developing individuals to an auditory environment that is critically different from the one they would normally experience. Two infants - one from each species - have already been successfully cross-fostered, indicating that our methods are feasible and likely to provide important new data on non-human primate vocal development. If cross-fostered infants conform to their adopted species' pattern of vocalizations, this would provide the first strong evidence of environmentally- induced vocal modifiability in primates. Conversely, evidence that infants follow their own species' pattern would indicate a fundamental difference in vocal production between human and non-human primates.

A major objective of the proposed research is to increase understanding of vocal communication systems, by investigating the vocalizations of free-ranging East African vervet monkeys, and comparing these with human communication. Our previous work has revealed striking similarities between the use of vocalizations by vervets and the simplest use of words by humans. One study will examine how vervets preceive vocal signals, and if they use complex sequences of calls under natural conditions. A second study will examine the acoustic properties of vervet vocalizations, allowing a comparison of vocal production in human and non-human primates. A third study will examine the ontogeny of vervet vocalizations, focusing especially on those areas in which the vervet's vocal development parallels the development of human speech. In each case, methods will include sound spectography and computer-aided acoustic analysis, plus observations, tape-recordings, and filmed playback experiments in the monkey's natural habitat. Comparison between human and non-human primate vocalizations, in both physical aspects of the acoustic signals, their perception, and the contexts of their production, can provide important information concerning human linguistic and 'paralinguistic' vocalizations. Comparative research on non-human primates cn lead to new diagnostic tests and theraputic techniques for humans, and is essential if we are to understand the neural control of different human communication systems.

Seyfarth 9520566 Abstract Alarm calls are vocal signals given by animals when they detect predators. Alarm calls are potentially costly, because they may attract predators. The questions of how such signals function, and how they evolve, are intriguing. Alarm calls given by nonhuman primates are of particular interest because they provide clear evidence of semantic communication outside of human language. This research has three objectives. The first is to examine the current function of alarm calls, and to elucidate their evolution. The second is to examine the semantic content of alarm calls, and the third is to determine what callers know about the predator-prey relations that exist among sympatric species. Studies will focus on red colobus monkeys, whose vocal and behavioral responses to tape playback of noises from potential predators and other prey species will be monitored in their natural environment. The results will increase our understanding of animal communication and insight into the evolution of semantic communication.

Abstract 9514001 Cheney The goal of this research is a better understanding of the vocal communication and social behavior of freeranging nonhuman primates. Focus will be on vocal communication for a variety of reasons: first, because of its potential relation to studies of language and cognition in humans; second, because vocalizations lend themselves to quantitative signal analysis and to experimental tests of specific hypotheses. Finally, prior research has shown that playback experiments can be used to test hypotheses not only about vocal communication per se but also about the cognitive mechanisms that underlie complex social interactions. Subjects are free-ranging baboons (Papio cynocephalus ursinus) living in the Okavango Delta, Botswana. Animals in the study population are fully habituated to human observers, are recognized as individuals, and have known reproductive histories. The proposal has five inter-related components. The first component of this proposal focuses on the use of vocalizations in reconciliation. After aggression, female baboons sometimes 'reconcile' by grunting to their lower-ranking opponents. Previous experiments have shown that these grunts function at least in part to reduce former victims' anxiety. A second component focuses on the acoustic structure and possible semantic content of grunts. Baboons appear to use acoustically different grunts in different social situations. The PIs plan a series of observations and experiments are planned to determine whether acoustically different grunt types designate specifically different information to baboons. A third component focuses on contact barks that appear to be exchanged between mothers and infants when they become separated in wooded areas. If these experiments indicate that mothers give contact barks primarily with respect to their own (rather than their infants') location, then they will add to a growing body of evidence suggesting that nonhuman primates do not call with the intent of providing inform ation to others. A fourth component focuses on the view that pair bonds are formed in response to the threat of infanticide by recent immigrant males. These experiments, however, are incomplete, largely because they must be conducted opportunistically during the weeks after a new, potentially infanticidal male has joined the group. A fourth goal of the research will be to complete these experiments. The final component focuses on the perception of cause-effect relations by nonhuman primates. Humans readily distinguish between two events that occur together because one caused the other and two events that occur together purely by association. Most research conducted to date suggests that monkeys do not apply causal reasoning when manipulating tools, though language-trained apes perform markedly better on similar experiments. Previous experiments conducted with this group of baboons, however, suggest that monkeys' ability to perceive cause-effect relations in the domain of social interactions may be quite well developed. Female baboons respond significantly more strongly to playbacks of vocal sequences that are causally anomalous than to those that are not. To investigate this question further, a series of experiments is proposed that make explicit predictions about both the strength and the timing of subjects' responses to particular call sequences. ***

Many animals give distinct vocalizations when feeding or upon discovering food. A number of hypotheses attempt to explain the evolution of these calls. One hypothesis argues that food calls are simply the involuntary consequence of excitement when the caller finds food. Alternatively, food calls may inform others about the type, location, or amount of food that has been discovered. The meaning of food calls to others, and hence their presumed function, is often inferred by observing the responses of nearby listeners. However, calls that elicit the same response may be caused by different underlying mechanisms. Call meaning, in this sense, may be different from the signaler's and the recipient's perspective. The researchers propose field observations and experiments to study the causes of food-calling in wild white-faced capuchin monkeys living in Costa Rica. They will investigate how calling is influenced by (1) the presence or absence of an audience, (2) food quantity and subdivisibility, and (3) foraging context. Prior experience at the study site assures that the work can be completed in 12 months and that the experiments are feasible. Results will make a significant contribution to our understanding of the mechanisms that underlie vocal production in nonhuman primates. These data, in turn, will help to formulate theories about the evolution of human language.

Group living confers a variety of benefits to individuals within the group, including predator detection, defense and high rates of food acquisition. Animals that form groups with close genetic relatives may also benefit from opportunities to help kin at minimal cost, thus increasing the helper's indirect fitness. Individuals in kin groups also gain direct fitness advantages through decreased within group aggression, lowered risk of infanticide, and coalitionary support. However, much of what we know about the effects of kinship comes from relatively simple, single level social systems. We know very little about how kinship works in complex, multi-level societies like those found in modern humans. Moreover, until recently only long-term studies have had the pedigree data to map maternal relatives. Now, genetic techniques that combine non-invasive sampling with PCR-based genotyping allow researchers to evaluate the interaction between relatedness and social systems in species where relationship data are lacking. This project applies these new genetic techniques and detailed behavioral observation to the unusual, highly complex social system of gelada monkeys (Theropithecus gelada). Geladas live in large, multi-level social systems resembling those of many human hunter-gatherer societies. Gelada social structure appears to be considerably more complex than that of their close phylogenetic relatives, the well-studied baboons. Therefore this research provides important new data on kinship-behavior interactions and the evolution of complex societies in mammals.

This project, in collaboration with the University of Michigan Gelada Research Project (UMGRP), offers both genetic and behavioral information on geladas, which are listed as "rare" by the World Conservation Union. The project employs Ethiopian scientists and rangers, and established both multi-intuitional and interdisciplinary collaborations. The co-PI participates in community outreach lectures to Ethiopian guides, tourists, and local K-12 students, as well as giving presentations to elementary school students in the Northeast U.S.