Experience-based human perception of facial expressions in Barbary macaques (Macaca sylvanus)

Abstract

Background: Facial expressions convey key cues of human emotions, and may also be important for interspecies interactions. The universality hypothesis suggests that six basic emotions (anger, disgust, fear, happiness, sadness, and surprise) should be expressed by similar facial expressions in close phylogenetic species such as humans and nonhuman primates. However, some facial expressions have been shown to differ in meaning between humans and nonhuman primates like macaques. This ambiguity in signalling emotion can lead to an increased risk of aggression and injuries for both humans and animals. This raises serious concerns for activities such as wildlife tourism where humans closely interact with wild animals. Understanding what factors (i.e., experience and type of emotion) affect ability to recognise emotional state of nonhuman primates, based on their facial expressions, can enable us to test the validity of the universality hypothesis, as well as reduce the risk of aggression and potential injuries in wildlife tourism.

Methods: The present study investigated whether different levels of experience of Barbary macaques, Macaca sylvanus, affect the ability to correctly assess different facial expressions related to aggressive, distressed, friendly or neutral states, using an online questionnaire. Participants' level of experience was defined as either: (1) naïve: never worked with nonhuman primates and never or rarely encountered live Barbary macaques; (2) exposed: shown pictures of the different Barbary macaques' facial expressions along with the description and the corresponding emotion prior to undertaking the questionnaire; (3) expert: worked with Barbary macaques for at least two months.

Results: Experience with Barbary macaques was associated with better performance in judging their emotional state. Simple exposure to pictures of macaques' facial expressions improved the ability of inexperienced participants to better discriminate neutral and distressed faces, and a trend was found for aggressive faces. However, these participants, even when previously exposed to pictures, had difficulties in recognising aggressive, distressed and friendly faces above chance level.

Discussion: These results do not support the universality hypothesis as exposed and naïve participants had difficulties in correctly identifying aggressive, distressed and friendly faces. Exposure to facial expressions improved their correct recognition. In addition, the findings suggest that providing simple exposure to 2D pictures (for example, information signs explaining animals' facial signalling in zoos or animal parks) is not a sufficient educational tool to reduce tourists' misinterpretations of macaque emotion. Additional measures, such as keeping a safe distance between tourists and wild animals, as well as reinforcing learning via videos or supervised visits led by expert guides, could reduce such issues and improve both animal welfare and tourist experience.

Keywords: Ethnoprimatology; Facial expressions; Human-macaque interactions; Primates; Tourism; Universal hypothesis.

Figure 1. The six pictures presenting the six Barbary macaques’ facial expressions related to the four emotional stated tested.

(A) and (B) ‘aggressive’ or ‘threat’ face; (C) and (D) ‘distressed’ or ‘submissive’ face; (E) ‘friendly’ or ‘affiliative’ face; (F) ‘neutral’ face. The description of each facial expression is detailed in the text. Photo credit: Laëtitia Maréchal (A, C), Julia Fischer (B, D, E), and Andrew Forsyth (F).

Figure 2. Bar graph representing the percentage of correct identification scores of Barbary macaques’ emotional state by experience and emotional states.

*** p = 0.01, ** P < 0.01, * P < 0.05. Error bars represent ±1 standard error of the mean.