American Journal of Primatology 74:841–852 (2012)
RESEARCH ARTICLE Infant-Directed Communication in Lowland Gorillas (Gorilla gorilla): Do Older Animals Scaffold Communicative Competence in Infants? EVA MARIA LUEF1,2∗ AND KATJA LIEBAL1,2,3 Languages of Emotion, Department of Education and Psychology, Evolutionary Psychology, Freie Universitaet Berlin, Berlin, Germany 2 Department of Developmental and Comparative Psychology, Max-Planck Institute for Evolutionary Anthropology, Leipzig, Germany 3 Department of Psychology, University of Portsmouth, Portsmouth, United Kingdom
Infant-directed speech is a linguistic phenomenon in which adults adapt their language when addressing infants in order to provide them with more salient linguistic information and aid them in language acquisition. Adult-directed language differs from infant-directed language in various aspects, including speech acoustics, syntax, and semantics. The existence of a “gestural motherese” in interaction with infants, demonstrates that not only spoken language but also nonvocal modes of communication can become adapted when infants are recipients. Rhesus macaques are so far the only nonhuman primates where a similar phenomenon to “motherese” has been discovered: the acoustic spectrum of a particular vocalization of adult females may be altered when the addressees are infants. The present paper describes how gorillas adjust their communicative strategies when directing intentional, nonvocal play signals at infants in the sense of a “nonvocal motherese.” Animals of ages above infancy use a higher rate of repetitions and sequences of the tactile sensory modality when negotiating play with infants. This indicates that gorillas employ a strategy of infant-specific communication. Am. J. Primatol. 74:841–852, 2012.
C 2012 Wiley Periodicals, Inc.
Key words: gorilla; communication development; infant-directed communication; play INTRODUCTION Infant-directed speech, also called “motherese,” is a universal feature of mother–infant interaction in humans [Monnot, 1999]. When addressing infants, mothers but also other adults and older children change their language, with adaptations ranging from phonological [Kuhl et al., 1997] to syntactic and semantic changes [Monnot, 1999]. Characteristics of infant-directed speech include simplified vocabulary, a higher rate of repetitions, and a different acoustic structure of sounds, such as more extreme vowel formant frequencies, higher pitch, and slower overall tempo [see Falk, 2004, for a review]. Specific language input to infants is suggested to promote language learning by attending to the infant’s developing neural systems [Kuhl et al., 1997]. Infantdirected communication is not restricted to the vocal modality but also “gestural motherese” has been described [Bekken, 1989; Iverson et al. 1999]. The maternal gesture repertoire is modified in such a way that fewer gestures are used in general when communicating with infants and some gestures are used more repetitively, for example pointing [Bekken, 1989]. Furthermore, infant-directed gestures typically refer to the immediate context and are used
C 2012 Wiley Periodicals, Inc.
to reinforce what is expressed in spoken language [redundancy principle] [Iverson et al., 1999]. Infant-directed vocal communication was found in rhesus macaques where females use specific vocalizations to address infants of other females, however, not their own [Whitham et al., 2007]. The authors suggest that females may vocalize toward the infants to communicate a benign intent before engaging in affiliate action with them. This suggests a social function underlying the infant-specific vocalization. Older animals providing assistance in youngsters’ acquisition of communicative means has been described in vervet monkeys. Seyfarth and Cheney  found that adult vervets provide Contract grant sponsor: Alexander-von-Humboldt Foundation. ∗ Correspondence to: Eva Maria Luef, c/o Katja Liebal, Freie Un¨ Erziehungswissenschaft und Psyversitaet Berlin, Institut fur chologie, Cluster: Languages of Emotion, Habelschwerdter Allee 45, 14195 Berlin, Germany. E-mail: [email protected]
Received 18 October 2011; revised 12 April 2012; revision accepted 15 April 2012
DOI 10.1002/ajp.22039 Published online 29 May 2012 in Wiley Online Library (wileyonlinelibrary.com).
842 / Luef and Liebal
“reinforcement” to infants after the production of correct alarm calls by repeating the correct call again after the infant. Hauser  observed physical punishment by adults when the infants gave incorrect alarm calls. Caro and Hauser  suggest that repetitions of correct alarm calls have beneficial effects on the infants’ knowledge; in the case of inappropriate calls by infants, discouragement such as physical aggression is used as a form of “teaching method.” The data from vervet monkeys are, however, tentative and do not allow the definite conclusion that the adults’ behavior directed at infants has a direct influence on the vocal development [Caro & Hauser, 1992]. There is to date no report of apes supporting conspecifics in the acquisition of communicative skills. The present study investigates if gorillas adjust their communicative signals when the addressees are infants, in the sense of nonvocal infant-directed communication. A form of “nonvocal motherese” could be an attunement to the developmental needs of the youngsters and aid them in the learning process of nonvocal signals. Gesture Versus Signal There are certain difficulties in using the term “gesture” to describe play actions and other intentional signals in gorillas and great apes in general. Gestural communication invokes linguistic concepts from human language where gestures and speech are intertwined in communication with gestures often accompanying speech on an unconscious level [see Willems & Hagoort, 2007, for a review]. The present study will use the term “signal” to describe purposeful nonvocal communicative behavior that is directed at other individuals in order to manipulate their behavior. Signals Communication in general may be intentional and voluntary or may occur in form of involuntary produced behaviors, with the forms of the communicative signals in both instances possibly being identical. The form of the signal is therefore less important than the cognitive volition that determines the communicative instance. The deliberate, directed use of a signal for a functional end accompanied by response waiting or persistence in case there is no response of the recipient therefore characterizes instances of intentional communication in both children [Bates et al., 1975] and nonhuman primates [Leavens et al., 2005]. The focus of this study is on signals produced in the context of play initiation and termination and will consider manual gestures, body postures, and facial expressions. It is not clear if facial expressions are intentional communicative signals and if the animals possess vo-
Am. J. Primatol.
litional control over production of the expression [see Tomasello, 2008]. However, since play faces are such an integral part of play in general and serve special function in play initiations in particular [Maestripieri & Ross, 2004; Pellis & Pellis, 1996], they are considered in the current study. A metacommunicative function has been suggested for play faces where the meaning of other intentional signals becomes modified by the signal being paired with a play face [Bekoff, 1998]. Therefore, if play faces are used in a directed way, as are gestures, they will be included as “intentional communicative signals” in this study. Other facial expressions have also been included because one goal of the study was to break up the strict dichotomy between “gestures” and facial expressions [see e.g. Maestripieri, 1996a] and consider them both as interacting communicative media. Gorillas use a wide variety of intentional signals to initiate and terminate play. Studies on the gestural communication of gorillas have identified a large number of different gestures that are used rather flexibly across different contexts: Tanner  recorded 30 distinct gestures; Pika et al.  found 33 gestures, and Genty et al.  recorded 102 gestures. For the present study, 22 distinct intentional signals were described. The splitting level that we applied differed from the gesture studies and the inclusion of facial expression does not allow a direct comparison. As opposed to those existing studies, the focus of the current study is on the investigation of the recipient’s role in the shaping of communicative behavior. The study of communication from the perspective of the recipient adds an important aspect to the entirety of intentional communication [Cartmill & Byrne, 2010]. In a communicative instance, considerations of to whom the signal will be directed may influence the sender’s behavior and may be expressed in different communicative strategies employed when dealing with certain recipients. METHODS The communicative behavior of captive lowland gorillas from two zoos in Europe [Zoo Leipzig, Germany, Howletts Wild Animal Park, United Kingdom] was recorded and analyzed. Observation period was from March to June 2011, approximately 4 to 6 h a day during morning and afternoon sessions. The sampling rules were focal animal sampling [Altmann, 1974] and all-occurrence sampling [Lehner, 2002] to record instances of play. Video clips were made from the visitors’ area in the zoos with a Sony Camcorder. The videos were edited on a MacBookPro using Adobe Premier Pro. Statistical analyses were conducted with R and SPSS. The research adhered to the legal requirements of the countries in which it was conducted, to guidelines of UFAW, and to the principles of “Ethical
Infant-Directed Communication in Lowland Gorillas / 843
Treatment of Non-Human Primates,” as stated by the American Society of Primatologists.
Subjects Twenty-four individuals [n = 24] in six different groups [Howletts: 5, Leipzig: 1] were recorded for the study and divided into four age groups [see Genty et al., 2009]: infant [0–3 years], juvenile [4–5 years], subadult [6–8 years], adult [females and blackback males over 8 years, and silverbacks, the dominant males of groups] See Table I. The mean frequency of signal types for age groups was: infants: mean = 10, ±SD = 2; juveniles: mean = 11.8, ±SD = 3.7; subadults: mean = 8.2, ±SD = 3.7; adults: mean = 8.4, ±SD = 3.6. The mean frequency for signal use in general was the following: infants: mean = 78.4 [±31.1]; juveniles: mean = 87.2 [±56.1]; subadults: mean = 40.7 [±36.4]; adults: mean = 40.5 [±28.9]. The focus of the analysis was on communication in the context “play,” in particular the initiation and the termination of play bouts through single signals or the use of signal combinations. Communication was defined as an exchange of expressive and non-instrumental signals that involved the head, limbs, and body posture in order to manipulate another individual’s behavior [also see Pika
et al., 2003; Tomasello et al., 1985]. The intentional use of signals was assumed when the sender established eye contact [either before, during, or shortly after the placement of the signal] and waited for a response from the recipient. For a signal to be included in the analysis, it was imperative that the recipient responded positively and that play between the two partners was commenced or stopped. Play was defined as motor patterns directed at another individual that occur to engage the other individual in playful activities, such as wrestling, mock-biting, hitting, and facial expressions associated with playful functions [e.g. Bateson, 1956; Flack et al., 2004; Pellis & Pellis, 1996]. Two independent raters agreed on the occurrence of signals with a Kappa value of 0.79 for 15% of the data, which is rated as “good agreement” [Altmann, 1991]. All statistical tests were conducted as two-tailed and a null hypothesis was rejected at an alpha level of 0.05. Types of Signals Signals may be produced as single signals or as combinations of signals [compounded signals]. When signals are compounded, they may be produced simultaneously [termed “multimodal”], in sequences [“sequential”], or as a mixed compound of multimodal and sequential signals [termed “mixed”].
TABLE I. Twenty-four Subjects and Observed Play Time; Gesture Types and Tokens per Individual
Subjects Affy Nkoumou Boula Mumbou Zola Imbizo Kisane Masindi Oundi Louna Baloo MahMah Bitanu Kibara Popa Ebeki Kwimba Kimba2 Sanki Otana Boma Djanghou Juju Viringika
Signals per minute of play time
Signals per subject
0,8 1,0 3,0 0,3 1,11 4,10 4,4 3,10 3,11 3,9 5,2 7,4 6,5 5,9 5,4 6,10 13 12 12 9 14 17 48 15
Infant Infant Infant Infant Infant Juvenile Juvenile Juvenile Juvenile Juvenile Subadult Subadult Subadult Subadult Subadult Subadult Adult Adult Adult Adult Adult Adult Adult Adult
Howletts 1 Howletts 1 Howletts 3 Howletts 2 Leipzig Howletts 2 Howletts 1 Howletts 1 Howletts 3 Leipzig Howletts 2 Howlette 2 Howletts 5 Leipzig Howletts 4 Howletts 3 Howletts 1 Howletts 1 Howletts 1 Howletts 2 Howletts 1 Howletts 1 Howletts 5 Leipzig
3.7 2.3 3.7 3.4 3.7 3.1 3.2 4.0 2.1 4.4 2.6 2.6 3.7 3.7 1.3 1.6 2.7 3.1 2.0 3.9 3.8 3.6 4.3 4.3
10 12 12 8 8 8 17 14 11 9 15 9 5 8 4 8 12 14 5 10 3 9 7 7
85 121 52 44 90 40 170 95 30 101 102 66 11 33 10 22 68 85 14 70 19 25 13 30
Am. J. Primatol.
844 / Luef and Liebal
TABLE II. Signal Repertoire Sensory modality Visual
Intentional signal Play face Shake head Bared teeth display Pouted lips Reach arm Present rump
Pirouette Somersault Roll over
Peer Straw wave Jump Grab Slap Touch Mouth stroke Hand on Beat object Chest beat Clap Body beat
Description Relaxed open mouth, upper teeth ridge may be showing, chuckling sounds may be produced. Fast movement of the head from left to right, similar to human head shaking. Mouth wide open, lips retracted, upper and lower teeth ridges showing; often ended play bouts. Lips are protruded to form a pout. Extending the arm toward another individual, often the wrist was presented. Upon approaching others, juveniles may turn around, bend over, and bring their rearside close to the other’s face. The other individual grabs the rearside and starts with play biting, play commences; only observed in one group between two juveniles and one adult female. Upon approaching, the head is lowered to the ground, possibly touching the play partner and engaging him/her in play. This signal is probably derived from mothers’ luring behavior to engage newborns in crawling [see Whiten, 1999]. Whole body spins around the horizontal axis. Making a forward flip. From an upright sitting or standing posture, the animal rolls down on its back and looks at its partner. Staring at another individual’s face from close distance (