Vocalizations as measures of pain in monkeys

Because vocalizations have been utilized as measures of pain reactivity and are reduced in frequency by systemic morphine in rodents, vocalization patterns of monkeys were analyzed in detail to determine whether specific relationships to pain could be established. Vocalizations were recorded as monkeys performed on tasks that presented the opportunity either to acquire food reinforcement or escape electrical stimulation (ES) of one leg. Three of 5 monkeys vocalized during sessions on the escape task, and 2 of 3 monkeys vocalized while performing for food reinforcement. The amplitudes and durations of vocalizations did not vary significantly over bins of time within the intertrial intervals of 1 min, and these aspects of the monkeys' calls were not affected by the intensity of the electrocutaneous stimulation or by morphine. Although the frequency of vocalizations was greatest within a reactive period of 6-12 sec following onset of a trial, vocalization frequency was not uniquely related to the occurrence of painful stimulation. Rates of vocal calls in the reactive period were higher following presentation only of a tone (that had been paired with shock in previous sessions) than following electrocutaneous stimulation, and vocalization frequency was not directly related to the intensity of painful stimulation. Furthermore, the vocalization patterns were similar for monkeys performing aversive or appetitive responses. Finally, systemic morphine reduced the frequency of vocalizations overall, without evidence for a selective effect on vocal calls elicited by painful stimulation. This result suggests a direct effect of morphine on the neural circuitry subserving vocalizations. Spectrographs of vocalizations revealed that 2 monkeys emitted a 'broken scream' call that occurred most often as the first vocalization following strong electrocutaneous stimulation, and 0.25 mg/kg of morphine reduced the frequency with which strong shock elicited these vocalizations. Although broken screams appeared to represent vocalizations in response to aversion, they also occurred following presentation of a tone that had been paired with shock, and therefore they do not reliably indicate the presence of pain.