Mobbing vocalizations as a coping response in the common marmoset

Abstract

Using a non-invasive method of sampling saliva followed by assay for cortisol levels, we found that common marmosets (Callithrix jacchus) show a decrease in cortisol levels after seeing a snake-model stimulus that reliably elicits mobbing (tsik) calls. In fact, there was a significant positive correlation between the number of tsik vocalizations made and the magnitude of the decrease in the cortisol concentrations. Furthermore, marmosets with higher levels of cortisol prior to being exposed to the stimulus produce more tsik calls than those with lower levels of cortisol. Subsequent experiments showed that, in response to 15 min of isolation with no visual or auditory contact with conspecifics (a traditional stressor), cortisol levels increased significantly. However, playback of the mobbing calls of a familiar conspecific to individual isolated marmosets not only prevented the rise in cortisol, but also actually caused a decrease in the levels of this hormone. This suggests that the mobbing calls serve to calm the marmoset after experiencing a stressful situation. This finding results in a greater understanding as to the role of physiological responses during communication in this species and could have implications for the welfare of marmosets in captivity.

Introduction

As in other species, when common marmosets are exposed to a stressful situation, they exhibit behavioral indicators of fear, including distress vocalizations (Norcross and Newman, 1999) and an increase in activity (Mendoza and Mason, 1986) as well as elevated levels of cortisol (Cross et al., 2004). Methods of eliciting stress responses in marmosets include housing in an unstable peer group (Johnson et al., 1996) or novel environment (Smith et al., 1998), social isolation (Jones et al., 2004) and presentation of a stress-inducing stimulus, such as a snake-model (Hook-Costigan and Rogers, 1998). How an individual reacts to a disturbing situation will differ between animals of the same species as extrinsic factors (such as presence of familiar peers and prior experience with the challenge), as well as intrinsic factors (such as an animals cognitive interpretation of the situation and its temperament) and will determine each individuals' physiological and behavioral response to stress (Boccia et al., 1995).

An animal's behavioral response to stress, however, is not always a reliable indicator of the physiological response and vice versa (Hennessy, 1985). For example, in inadequate captive housing conditions, animals may display activities thought to indicate stress (e.g. performance of stereotypic behaviors), but this may not always be consistent with their physiological state (Geverink et al., 2003). Therefore, we were interested in investigating both the behavioral and physiological responses of marmosets when they were likely to be stressed by exposure to a stimulus that induces mobbing.

In their natural habitat, marmosets emit mobbing “tsik” calls in response to seeing snakes and other predators. Mobbing occurs when social groups cooperatively attack, harass or pursue a potential predator (Wittenberger, 1981). These mobbing calls are used to alert other members of the group to the presence of the predator and act to drive it away (Epple, 1968). This has been shown to be a behavior that is learned from other group members (Mineka and Cook, 1988). Alternatively, as in the case of the rapid approach of a raptor (one of their major predators), they may emit a brief alarm call, take cover and freeze (Ferrari and Lopes-Ferrari, 1990). Both types of anti-predator response are thought to involve heightened fear and stress. Mobbing or freezing responses can be elicited by the visual, olfactory and auditory characteristics of predators (Searcy and Caine, 2003). The continuous issuing of alarm vocalizations during mobbing may serve to discourage or drive away predators that rely on surprise attacks (Zuberbühler et al., 1999), and the presence and proximity of conspecifics can influence the rate of vocalizations emitted (Coe et al., 1982). Previous studies here have shown that the marmosets in our colony emit bouts of tsik calls and perform mobbing responses to specific visual stimuli, such as a snake-model stimulus (Hook-Costigan and Rogers, 1998). We performed this study to determine how seeing an arousing stimulus and hearing mobbing (tsik) calls performed by other members of the group affect the marmosets physiologically and to assess if the hearing of the tsik calls would cause the marmoset to become more stressed in the presence of a predator or if hearing the calls would in fact decrease stress levels and calm the marmoset as each individual realizes the presence of other group members who will group together to drive away the predator. We did this by measuring cortisol levels, using an established salivary sampling method (Cross et al., 2004).