Introduction

The differences between human and ape communicative and social development have become the centre of a controversy that questions human uniqueness in social and communicative development. These studies have usually been carried out under the framework of Trevarthen’s theory of intersubjectivity.

Trevarthen distinguished between two stages in the development of intersubjectivity: ‘primary intersubjectivity’, which is the mutual regulation of subjectivities which occurs between mothers and two-month-old babies in protoconversations (Trevarthen, 1998), and ‘secondary intersubjectivity’, in which the mutual regulation of subjectivities occurs between mothers and babies in cooperative activities on objects (Hubley & Trevarthen, 1979).

Questioning the human uniqueness, some authors argue that human and ape social development is not as different as initially thought. They affirm that apes also have the abilities needed to engage in interactions from the very beginning of their lives (Bard, 2005) and to participate in face-to-face communication with the typical interchanges of smiles and mutual gazes that characterize human protoconversations (Maestripieri & Call, 1996).

These researchers placed the main differences between ape and human social and communicative development at the end of the first year of life. At this age, baby humans begin to perform protoimperatives, -communicative acts by which the child uses an adult to obtain a wished-for object (Bates et al., 1975), and protodeclaratives or ‘the use of non-social means to a social goal, in this case, sharing adult attention towards some referent’ (Bates, 1979, p.35).

Nowadays, whether apes use protodeclaratives is a topic of discussion. Some researchers think that they do because they use the pointing gesture with an imperative function, or to direct the other´s attention to a potential danger, or because trained-language apes name or offer the objects that their teachers ask for or respond to their questions (Lyn et al., 2011). This occurs after having obtained reinforcement for using symbols.

On the contrary, other researchers affirm that apes do not use protodeclaratives that have the sole aim of sharing attention (Gómez, 2010), even when the apes have been raised by humans (Tomasello & Camaioni, 1997), despite the fact that they can follow and direct the other´s attention (Leavens & Racine, 2009), and use pointing with an imperative function (Tomasello & Camaioni, 1997).

Many different reasons have been proposed to explain this apparent lack of protodeclaratives. The first considers that the aim of performing protodeclaratives is to affect the others´ mental states, something that requires the sender to represent the other’s mental state (Baron-Cohen, 1991), and apes can only ascribe to others intentions and desires (Call & Tomasello, 2008). The second perspective states that the aim of protodeclaratives is to elicit the receptor’s external manifestation of attention and positive affect (Moore & Corkum, 1994), and that apes are not motivated to perceive them (Gómez et al., 1993); the third assumes that the aim of protodeclaratives is to direct the other’s attention to interesting objects or events, or to provide them with information (Liszkowski, 2006), a goal that apes do not pursue if they do not have any other aim (Tomasello & Carpenter, 2005); and the last reason is that only baby humans produce protodeclaratives because only humans are motivated to express the feelings that objects, persons or events give rise to, or ‘to share with other persons their interest in or attention to some object or event’ (Tomasello & Camaioni, 1997, p.13).

Ape babies do not engage in playful triangular interactions when food or defence is not involved (Warneken et al., 2006), or in cooperative social games involving objects (Warneken et al., 2006), although apes can cooperate, patrol together to defend their territory, and take complementary roles (Boesch, 2002).

In some enriched environments, baby chimpanzees reproduce tasks such as inserting one object into other, making towers, and opening tubes, but many researchers have reported that these tasks are not usually performed cooperatively (Hayashi & Matsuzawa, 2003).

Leavens and Racine (2009) suppose that training, social experience, and different experiences could explain the differences found in human and ape communication and cooperation in playful triadic interactions. This study aims to analyse these differences. Its objective is to compare the interactive formats of three kinds of dyads: chimpanzee dyads, human dyads that have come from an occidental society, and dyads composed of a baby chimpanzee and a western person, as well as their social and communicative development. It should be noted that we are referring to three kinds of dyads that were living in Spain, and not to generic dyads.

We posed the following research questions:

  1. a.

    How did the dyadic interactions in which baby chimpanzees participated differ from those displayed by the child?

  2. b.

    Were there any differences between the triangular interactions displayed by the different dyads?

  3. c.

    Were baby chimp’s expressions in playful interactions different from the child’s expressions?

  4. d.

    Did the baby chimpanzee produce protodeclaratives?

  5. e.

    Did baby chimpanzees participate in cooperative play?

We did not have any hypotheses to pose at the beginning of the research as we were performing this as an exploratory study.

Method

Subjects

We analysed three different kinds of dyads. The first was composed of a western baby boy, Jorge, who was 2.5 months old when this study began, and either his mother, or an expert in paediatrics (H-H dyad). The second was the chimp dyad (C-C dyad) composed of a nine-month-old baby ape, Patricia, who was the only offspring that had been born in a zoo), and either her father, or her mother; they all lived at the Santillana del Mar Zoo, in Cantabria, where they were taken care of wisely and affectionately, as Pardo de Santayana (2010) reported. The third, a captive female chimpanzee, Noelia, and one of three people (H-C dyad). Noelia arrived in Spain when she was approximately 45 days old. A nurse who worked at the zoo offered to care of Noelia in her own home, for the following eight months. After this period, Noelia went to the nursery at the Madrid Zoo where the nurse continued to care for her. The research team and the people who worked at the zoo used to visit and play with Noelia frequently. They began to tape these interactions two weeks after she arrived at the zoo.

Procedure

We videotaped Jorge at home interacting with his mother approximately every two weeks since he was two and a half months old (though there was no session for the 12th month, just one for the 15th and we added a final session when Jorge was 18 months old). Every second session an expert paediatrician volunteered to play with Jorge as well.

Times for all visits for Jorge were arranged to suit the time when the mother believed her baby would be alert and happy. We told her and the expert paediatrician that the aim of the study was the development of communication and we asked them to play with Jorge as naturally as possible. The length of the session depended on Jorge’s mood.

Most of the interactions took place in the sitting room where Jorge was sitting on his carriage at the beginning, though his partners could pick him up whenever they wanted to. Jorge began to stand up when he was seven months old, from then, the time spent sitting on the carriage decreased, until we removed it at the beginning of the second year. We placed a mirror beside where Jorge was to facilitate the analysis. The interactions took place in the parent’s room after a nappy change when that happened to be required.

Patricia was observed for nine hours distributed in three different sessions (of two, three and four hours) when she was nine, eleven and twelve months old. An observer described the interactions that took place between Patricia and her mother or her father, following the diary method frequently used to study a subject communicative development (Gómez, 2010). Nine observations were taken.

We assigned Noelia three humans as friends who became part of her social environment; two of them spent at least three hours per day playing with her and the third one used to go also during the weekends. Interactions took place in three different settings: the nursery, where Noelia lived for seven months being surrounded by the affectionate zoo workers who took care of her with care and love, a garden, and a 4*3-m2 cage, where she began to live with other animals when she left the nursery. When the researchers were playing with Noelia, they used to say when they had made eye contact with her. Four principles guided the selection of the videotaped interactions for this study:

  1. 1)

    Their playful character.

  2. 2)

    The absence of physical restrictions (all interactions in which there was a barrier between the partners were excluded).

  3. 3)

    Quantity of the sample (interactions held every two weeks were selected, but if the time recorded was too short additional interactions were taped on the following days).

  4. 4)

    The diversity of the sample. Any interactions in which Noelia was co-ordinating actions directed to objects with actions directed to people were included in the sample, as well as those interactions that presented peculiar characteristics. Though this principle made the sample of Noelia’s interactions atypical, it assures its diversity, which is important due to the purposes of the present research. We arranged it that some of the toys used by the human and the inter-specific dyads were the same, specifically: toy cars, a toy windmill, a toy duck that was powered by a winding mechanism, a small rubber elephant that made a sound when pressed, a ball, different types of boxes, and a photographic camera.

We recorded Jorge’s interactions with his mother (29.444 s) and with a stranger (15.172 s), and Noelia’s interactions with Partner 1 (13.001 s), Partner 2 (9862 s), and Partner 3 (15.438 s) (see Table 1). Patricia was not videotaped.

Table 1 Time of interaction analysed by Jorge’s (J) and Noelia’s (N) age in months (in seconds)
Full size table

Design

This is a qualitative descriptive and observational study. It is the kind of research that is useful when the participants are not verbal, and when researchers are interested in spontaneous behaviour (Bakeman & Quera, 2011).

Analysis of Data

Regarding Noelia and Jorge’s data, one of the female researchers incorporated a chronometer on the videotapes and followed the steps that Bakeman and Quera (2011) proposed in order to set the categories of observation: (1) she decided the unit of observation we were going to use; -it was Dore’s (1985) dialogic unit-, (2) she identified the units of observation, (3) she described them in detail, (4) she coded them (and wrote their temporal limits), (5) she watched the videotape twice or more to check if the interactive units were properly coded, and made the few changes that were necessary, and, (6) she pilot-tested the codes. The researcher then added some more abstract categories joining them with those that had some characteristics in common. Table 2 lists every category. The researchers also coded the observations of ape dyad interactions. When the observations described behaviours that had not been categorized, they added new categories in addition to the ones listed in Table 2. Detailed descriptions were also kept.

Table 2 Categories of interactions
Full size table

Dore’s (1985) ‘dialogical event’ is an interactive unit that requires at least two obligatory moves engaging between individuals. The empirical beginning of the unit of observation was the first responded or attended behaviour. The end of the unit was defined either by one of the partners’ withdrawals from the interaction or by a change in their activity. This could mark either a new unit of observation or a ‘conflict’ if a partner tried unsuccessfully to draw the other’s attention to the new activity. A partner’s withdrawal from the interaction was marked when he/she physically left the setting, when he/she threw away the object with which they were playing with, or when the young subject’s attention changed completely during more than four seconds without any adult´s intervention.

The interval between these units of analysis, as well as the time the partners took to respond to the other at the beginning of the unit, or to place themselves more comfortably, and the occasions in which external interruptions or internal accidents happened (i.e., a phone call, the arrival of different persons or animals, the need to change diapers, etc.) were excluded.

After having coded the units of observation, we analysed the expressions that babies (Jorge and Noelia) showed in each unit of interaction when they were performing actions on objects, and calculated an ‘index of expressivity’ according to the following formula: Eic = S Iic* 100/I, where, Eic is the index of expressivity of the expression ‘i’ in the interactive context ‘c’, Iic is the frequency of interactive units ‘c’ in which a particular expression ‘i’ appears, Iic scores double when frequent expressions appeared (as will be seen), and I is the frequency of interactive units. Table 3 shows the expressions scored.

Table 3 Babies’ emotional expressions
Full size table

To test the code system, we trained two observers. The training consisted of an explanation of the meaning of the codes, and they were provided with written definitions. They were then asked to code the ‘dialogic units’ that we had previously selected. They had to write two codes for each on a register form that they had been provided with: 1) the kind of interaction, and 2) the babies’ expressions. They coded 62.27 and 77.48 minutes, which constitute 10.22%, and 10% of Noelia and Jorge’s total time.

Four of the ten VHSs that had all the data were digitized to obtain the ‘dialogical units’. We chose one tape from the beginning and one from the end of the period of observation for Noelia and Jorge to have more different behaviours coded. We used this temporal criterion again to select the units of observations, i.e., some from the beginning, some from the middle, and some from the end of the digitized films. Finally, we began again from the beginning until we had reached 2,21 h, 10,22% of the corpus.

The analysis of the results showed that some codes had not been very well explained, so we asked a third observer to analyse and code the videos. We calculated the agreement between three different couples of observers and chose the highest.

Following the criteria provided by Landis’ & Koch’s (1977), Cohen’s Kappa coefficient was strong for: the retiring objects (1.00), substantial for: eye-contact (0.621), vocalizations (0.691) and smiles (0.607), and moderate for: body contact (0.544), and giving objects (0.502).

Cohen’s Kappa coefficient was also strong for: hide and seek (1.00) and realistic or cultural play (0.883), substantial for: protoconversations (0.710), ritual activity (0.657), vocalic imitation (0.691), information transfer (0.662), babies showing an object (0.662), forbidden actions (0.795), affective interactions (0.795), tickling and tapping (0.794), expectation game (0.692), intrinsic combination task (0.678), objects with acquired valence (0.650) and arbitrary object combination (0.740); moderate, for: body display (0.420), object observation (0.492), object exploration (0.424), distance game (0.591), intrinsic object (0.492), object with valence (0.479) and distal object (0.509) and fair for: showing off (0.324); this last result was due to the fact that the explanation of this category highlighted ‘playing funny’ and practically ignored the babies’ looking for the adult´s appreciation, consequently one of the observers coded it as a motor task, a category that obtained a Cohen’s Kappa coefficient of 0.

Results

The results are organized as a function of the research questions formulated:

  1. a.

    How did the dyadic interactions in which baby chimps participated differ from the ones displayed by the child?

As Table 4 shows all interactions in which Patricia participated involved physical contact and were PSI-encoded.

Table 4 Interactions displayed by the C-C dyad
Full size table

Table 5 shows that some of the interactions displayed by the C-C dyad were similar to the ones that the H-C dyad and the H–H dyad displayed such as ‘tickling and tapping’, and, ‘The expectation game’, but others such as ‘co-action with or without turns’ were different. Table 6 also shows that Noelia spent 3.32% of her time on DSI (Jorge, 23.66% and Patricia, 0).

Table 5 Dyads’ PSI
Full size table
Table 6 The interactions displayed by the interactive dyads
Full size table

Table 7 shows that Noelia’s single category of ‘interactions based on attention and co-action’ was ‘ritual activity’ (0.3%). During this interaction, the adult began to imitate the behaviour ‘hitting the ground’ with which Noelia used to ask the adult to tickle her again; Jorge’s ‘ritual activity’ (2.4%) consisted of the popular ‘Palmas-palmitas’, and a game which consisted of performing body actions, such as clapping hands or bending his knees, that tried to fit the rhythm of the song that her mother sang.

Noelia and Jorge’s ‘interactions based on observing the baby’ were also different. Noelia did not show ‘showing off interactions’, or ‘babies showing an object’, although Jorge did. Both displayed forbidden actions, Jorge (2.9%) by staring at his mother, while he was performing the forbidden action as if he were either challenging the adult or making fun of them; in contrast, Noelia’s ‘forbidden actions’ (1.2%) such as touching the video camera, getting bags, etc., had the aim of performing them more than anything else. (See Table 7).

Table 7 Time (in seconds) and percentages of time of Jorge (J)´s and Noelia´s (N) DSI
Full size table
  1. b.

    ¿Did the baby chimps produce protodeclaratives?

Whereas Jorge produced declaratives (note that protodeclaratives were ‘babies showing an object’), Noelia and Patricia did not. (See Table 7).

  1. c.

    Were there any differences between the triangular interactions displayed by the different interactive dyads?

Table 4 shows that most of C-C dyad’s interactions were dyadic. Three observations showed that the apes used objects when they were playing, but the objects did not attract their partners’ attention. Instead, Jorge and Noelia spent the greatest amount of their time on interactions based on performing actions on objects (41,55% and 47,79%). (Table 6).

Jorge spent 4.4% and 3.3% of his time on ‘observing an object’ and on ‘observing and image’ (without exploring them), and Noelia, 0.6% and 0.1%, respectively, (see Table 9). In the first case, Noelia looked at the adult’s watch which had a second hand that kept on moving, and a mirror, and in the second, she put her finger on a food poster that had images of different foods. Patricia did not participate in this kind of interaction.

Table 8 Dyads’ POI
Full size table
Table 9 Dyads’ DOI
Full size table

Table 8 shows the distribution of POI. Noelia’s favourite was the ‘distance game’ (15.1%) and the second, ‘intrinsic combination task’ (11.4%), in which Jorge spent the 7.5% and the 5.3% respectively. Noelia was very interested in opening and closing recipients such as a plastic barrel toy, a transparent plastic bottle and a metallic box, putting things in them, inserting things into holes, and pressing the button of the photographic camera. Similarly, Jorge also enjoyed inserting things into other things such as a driver’s car into the toy car, a pilot into a toy plane, pieces inside a cube and the chimney of a train, or rings on a stick. Their simplest ‘intrinsic object interactions’ were the moving parts of objects, such as the door or the wheels of a car or a rewind lever, or pressing buttons.

Jorge’s ‘object with valence’ (1.2%) and ‘objects with acquired valence’ (2,9%) included smooth dolls and objects which produced an unexpected effect such as the car which suddenly made a noise when the driver was taken out. Noelia (3.4%, 4.3%) seemed to be afraid of the little rubber objects such as the toy elephant and the ball, as well as mechanical objects such as, for example, a chicken toy even before it began to move around, and a large cushion whenever its position was changed.

There were qualitative differences in the ‘creative interactions’ displayed by Jorge (9.7%) and Noelia (3.9%). Noelia spent most of her time on ‘making sounds interactions’ (2.6%). She made those noises when she hit the suitcase where the video-camera was kept, with her hand. But she seemed to be more interested in her hitting behaviour than in the sound. Noelia’s ‘arbitrary object combination interactions’ (0.7%) consisted of destroying the towers that the adult had built. Jorge also used to destroy them, although sometimes after having tried to build them. Jorge also enjoyed arranging objects in rows, covering a car with books, putting different objects on others, or directing the movement of a ball with a stick (2.7%).

Noelia’s ‘cultural play and realistic interactions’ (.6%) were related to her physiological needs. In the first example, the adult simulated that a little animal toy was alive, and the second, the instrumental use of a spoon to drink water, and the last, an interaction in which the adult pretended to eat something that Noelia had just put into her mouth and then he put it back into Noelia’s mouth. Jorge’s interactions (4.7%) consisted of moving cars and aeroplanes as if they were real, simulating that they were taking photographs with the toy camera, talking on the toy phone, and simulating that he was going shopping, i.e., getting a basket and saying goodbye to everyone.

d. ¿Did Noelia participate in cooperative play?

Some of Noelia’s ‘intrinsic combination tasks’ were performed cooperatively. Thus, on several occasions the adult gave Noelia a barrel after showing her how to open it. She took it but, failed to open it, and so returned it to the adult who put just one half of the barrel in front of her face. Noelia took this half, and put it on the half that the person was holding on, and the barrel was formed again.

  1. d.

    Are baby chimp’s expressions in playful interactions different from the child’s ones?

Jorge´s expressions (except for ‘retires’) were equally shared among the different kind of interactions, whereas Noelia´s ones appeared linked to determined type of interactions with the occasional exception of ‘body contact’ and ‘retires’ (see Tables 10 and 11).

Table 10 Distribution of Jorge’s expressions among the interactive settings
Full size table
Table 11 Distribution of Noelia’s expressions among the interactive settings
Full size table

Thus, most of the Noelia’s ‘play-face’ expressions were displayed in ‘distance game’ interactions (index of expressivity = 71.81), where her gazes reached the highest index (55.82). She gave objects in ‘intrinsic combination task interactions’ (80.76) to ask the adult to perform an action with the object, or to help her to perform it and vocalized mainly in ‘objects with emotional valence interactions’ (64.28). Noelia’s ‘body contact’ was the expression that was more equally distributed among the different interactive settings. Jorge did not use this type of expression in these interactions (Table 10).

Discussion

This exploratory study examines the dyadic and triadic interactions in some western human-human dyads, chimpanzee-chimpanzee dyads, and human-chimpanzee dyads to address the issues posed in this section.

How did the Dyadic Interactions in which the Baby Chimps Participated Differ from the Ones Displayed by the Child?

The dyadic interactions in which Noelia and Patricia participated were different from those in which Jorge did. One of the main differences was that Noelia and Patricia’s interactions always involved actions, expectations, or body contact. These types of interactions have been observed in zoos and in nature. Our observations are like the researchers who analysed mother baby interactions in free-living apes. Plooij (1979) described the evolution of an interactive pattern of tickling and biting in which mother and young made their moves in succession which is like the interactions that Patricia and their parents displayed. Mothers stroke their babies’ faces and placed their index fingers against or in their mouths making the baby display play-faces and biting. Gradually, the young chimpanzee became more likely to reach for and to bite his mother’s fingers after she had stopped stroking or had removed them from his face. Mother’s tickling often resulted in babies’ laughing, which acted as feedback for the mother through which she could know if the baby was receiving the right amount of stimulation. Therefore, it is neither that they do not look at each other, nor that they do not display facial expressions. They show these expressions in zoos and in nature (Amici et al., 2024; Maestripieri & Call, 1996). It is that these expressions inform the adult whether the baby is enjoying the game or not. Perceiving these expressions may regulate the parent´s actions, as Plooij (1979) explained.

The chimpanzees that took part in our study did not display protoconversations (Trevarthen, 1974), although it should be noted that we did not have the opportunity to check if they had displayed them in their first months of life. On the contrary, Jorge did like many babies, at least those who lived in an occidental society.

Along the sixth month, as Rivière (1997) noted, the previous emotional interchange (that can be seen in protoconversations) is nearly replaced by interactions in which both partners try to fit their activity to a shared rhythm or to a behavioural pattern (our ‘ritual activity interactions’); thus Jorge let his mother guide his behaviour, either its shape as in clapping hands (“palmas palmitas”) or its rhythm as occurs in interactions where the mother sang and Jorge joined her rhythm clapping his hand on a chair or bending his knees.

There were ritual interactions in the H-C dyad; but in this case, the person imitated the movement that Noelia used to perform after being tickled (i.e., hitting the floor with the hands), unless the action imitated had to do with food.

Patricia and her parents also engaged in interactions based on co-action since they enjoyed biting each other; these interactions were again based on physical contact, such as those displayed by the captive chimpanzees observed by Bard et al. (2005).

The theoretical framework outlined by Nuttin and MacMurray may lead to a peculiar interpretation of the differences among the interactions observed. Nuttin (1984) defined motivation as a mark imprinted in the mental apparatus of organisms which determines the range of relations they can establish with the world. Because Noelia and Patricia’s interactions differed from Jorge’s, it may be that they had different motivational systems.

To clarify the differences between them, we borrow MacMurray’s (1956) distinction between the psychological relationships that can be established with the world through actions of the ‘subject as agent’: (a) ‘The means relationship’ and (b) ‘The end relationship’. In the first one, subjects perceive the world as a mean to fulfil physiological and psychological needs. On the contrary, the second one is that in which the individual perceived the world as an end by itself with immediate subjective ‘qualia’. In this case, the organism does not try to change the object, nor use it as a guide of its own movement, nor as a potential mean for future actions.

Werner and Kaplan (1963) also described this different way of perceiving reality. They distinguished ‘things of action’ from ‘things of contemplation’, that is, ‘objects that one regards out there…. from things upon which one merely acts in the service of immediate biological needs-satisfaction’ (Werner & Kaplan, 1963, p. 67).

It might have been that Patricia and Noelia were only able to establish with the world the first kind of relationship of action with experience, whereas Jorge’s motivational system contained both, ‘the mean relations with the world’, and another in which the world is perceived as an end in itself. If this were true, we could say that Noelia and Patricia’s minds were composed of a ‘sensorimotor unit’ regulated by emotional states, in which the mechanisms of attention picked up from the environment only that information which was useful to guide actions. It always generated intentional states since it made them perceive the world as ‘objects of action’ (Werner & Kaplan, 1963). Whereas Jorge’s mind contained, an additional ‘attributive unit’, as it was referred to by Rivière (personal communication). It would be composed of the same systems as the sensorimotor unit but with a different order of functional priority. In this case, action served perception, creating a sort of ‘contemplative states’ in which the world is perceived as an ‘object of contemplation’ (Werner & Kaplan, 1963) without the aim to complete actions and without any cognitive interest.

We would say, drawing on Searle’s (1983) terms, that the direction of fit between our babies’ mind and the environment takes opposite ways depending on the unit responsible of the relation established between them. When they relate to the world through the ‘sensorimotor unit’, there is always an intention in their mind to which they try to fit the action to the object. The direction of fit goes from the world to the mind, whereas the direction of causality goes from the mind to the world.

On the contrary, when the mind tries to fit the world and the direction of causality goes from the world to the mind, those relationships are established through the ‘attributive unit’ of imagination. This direction of fit of movement with the physical world will be evident in dancing to music, given that any kind of dance supposes an attempt to fit the grace and energy of movements with the rhythm of the music listened to. This kind of dance has never been seen in apes as far as we know. Though Francis (1991) reported the different kinds of dancelike behaviours in apes described in the literature, that show that apes can perform rhythmic movements, but the rhythm does not fit an external pattern of artful composition.

In the social context of a child’s early experience this predisposition to adapt oneself to the world leads mother and infants to that kind of ‘co-action’ (Nuttin, 1984) which exemplify the tendency of an organism to transfer the locus of control of its own activity. We have observed this kind of co-action in Jorge’s ‘protoconversations’. In these interactions, the other becomes an object of contemplation and oneself is experienced as object of contemplation too. Both partners’ goals are reduced to experience the feeling of being engaged in an interchange in which both subjectivities control are trying to fit to each other (Trevarthen, 1998).

Patricia and her parents engaged in interactions based on co-action and enjoyed themselves performing the same action (biting to each other). We would say that in this kind of ‘ ape protoconversations’, but while they are at the same time actors and perceivers as in Jorge’s ones, each one does not perceive itself as the other’s object of contemplation, so there is no transference of the locus of control of the own activity. These ‘ape-protoconversations’ do not evolve to a ritual activity in which the imitation of the other´s action does not include physical contact. It does not become a ‘symbol’.

Were There any Differences between the Triangular Interactions Displayed by the Different Dyads?

Our data show that Noelia’s and Jorge’s triangular interactions were also different. Noelia’s DOI seemed to be led by a cognitive interest, because the object observed had a part which kept on moving (the second hand of a watch) or an image in a mirror.

The fact that perceiving a visual or auditory stimulation is not an end by itself for Noelia seemed also the case in the special characteristics of the interactions she made with sounds. This category included the typical Piaget’s ‘secondary circular reactions’ (SCR) in which the baby acts to reproduce visual and/or auditory effects. Noelia displayed an interaction of this kind when she hit the video camera case. But she was hitting the object each time with stronger force, rather than tapping it to listen the sound effects. It seemed that Noelia and Jorge’s SCR were led by different motivation.

Chevalier-Skolnikoff (1977) reported the existence of SCR in anthropoids. She gave an example, which is especially useful to clear up one of the differences we find between the SCR of humans and apes. Her subjects were pealing a bell. The photographs shown in her study are very meaningful. The child, sitting on the floor, holds a bell with his hand and smiles. The chimpanzee, in a woman’s arms, has a bell hung with a string in front of him. The bell sounded probably as the consequence of the chimpanzee’s attempts to get it.

Gómez (1989) concluded that SCR were apparent in the behaviours of young gorillas. The examples of SCR given such as hitting a wall, a ball, a glass, a can, rubbing the spiral of a notebook, splashing about the water are also characterised by the simultaneity of the motor actions performed and the visual and/or audible effects produced.

Are Chimp Baby’s Expressions in Playful Interactions Different from the Child’s Ones?

It is necessary to observe the primitive communicative expressions to understand the communicative development of any subject/agent in relationships. Most of Noelia’s expressions are linked to interactive contexts as if their expressions were elicited by the context or by the objects used instead of being elicited by their social partners, whereas Jorge’s ones, excepting actions to ‘retire’ from engagement, appear distributed among the different interactive settings. Noelia’s highest index of expressivity of ‘play faces’ and ‘eye to eye contact’ appears in the teasing with distance game. In this game, adults have an interesting object and keep on approaching it to the baby and putting it out of reach when the babies try to get it. Babies may look at their partners’ eyes attempting to anticipate their movements and thus being successful in getting the objects desired or in keeping the game going.

Most of Noelia’s vocalizations appeared in reference to ‘objects with valence’. Some objects made her vocalize and look for the partner’s protection, since most of her ‘body contact’ appeared in this context.

It has been proposed that the property of context-independence is the mark which distinguishes human from animal communication, since that property is the consequence of having behaviours patterns which are involuntarily triggered by certain types of contexts (see Brinck, 1999). Our study demonstrates that Noelia’s communicative expressions were context-dependent, whereas Jorge’s were context-independent.

Did the Chimp Babies Produce Protodeclaratives?

As with prelinguistic and linguistic communication (Bruner, 1978), there might be a continuous line between pre-intentional and intentional communicative acts. The role played by the social partner in triadic interactions does not change when babies can formulate communicative actions on purpose. Babies’ intentional communicative acts are just a regulative element of the relations that they establish with the physical world when their previous social relations are subordinated to the relations that these babies establish with objects in social-object-games. Their intentional communicative acts have just an imperative function.

Jorge’s communicative interchanges do not limit themselves to regulate the way he relates with objects, as when he requested the adult to perform a desired action on the object; he took the other into account in all the interactive settings, so at the intentional communicative stage, with his communicative acts he expressed the feelings that the objects arise, as he has been expressing himself before while experiencing himself as the other’s object of contemplation.

Experiencing being the other’s object of contemplation leads the actor to show their own mental states (his or her satisfaction for the acquisition of new motor abilities, the feelings that the objects arise, etc.), as Reddy (2003) stated. We observed how Jorge performed showing off interactions and forbidden actions with the purpose of letting the mother know his determination of performing them despite the prohibition. And he also produced protodeclaratives. On the contrary, Noelia did not display any showing off interactions, her prohibited actions had the sole aim of performing them, and she did not show any object without asking for something.

Reddy (2003) stated that babies become aware that what they perceive might also be the other’s object of attention and they make protodeclarations. Jorge showed these protodeclaratives (‘showing an object interaction’), but Noelia did not. Reddy (2003) argue that ‘perceiving attention in others could emerge from the experience of being an object of attention’ (p.399). But this is not enough. Noelia did experience herself as being the other’s focus of attention, but she displayed neither ‘showing off’, nor ‘showing an object interaction’.

It is necessary to distinguish between being the object of attention as a subject of action and being the object of attention as a subject of contemplation. Only in the latter case it is possible to develop a sense of ‘me’ whom others enjoy seeing and hearing or attending in any other way without any other end, and a special sensitivity to the other as spectator (Trevarthen, 1989). Perceiving the other as a subject of contemplation means that one has been perceiving oneself as the other’s object of contemplation, and this might lead to the communication of whatever one might feel, see, experience, or think.

Baron-Cohen (1991) suggested that the aim of protodeclaratives is to affect others’ minds. Though, at the beginning, protodeclaratives might just be a product of this tendency to communicate whatever is in one’s mind. It might also be a process through which children realize that protodeclaratives can also affect the other’s mind, a kind of representational redescription (Karmiloff-Smith, 1995), and affecting the other’s mind might eventually become the aim of at least some protodeclaratives.

Apes performing protodeclaratives have been described when researchers asked the apes to name some objects after having shown them a communicative system, or to inform (Lyn et al., 2011), or after defining protodeclaratives as a communicative act aimed at sharing attention to a distant object (Leavens, 2004), without indicating the communicative function of that act. However, Wilke et al. (2022) recently observed in Kibale Forest, in Uganda, a chimpanzee called Fiona showing her mother a leaf when she was grooming her. The authors stated that the ape was sharing with her mother her interest in the object for ‘sharing’s sake’, though since the mother was using leaves to groom the ape, the possibility that Fiona was asking her to use that particular leaf to groom her remains open, because, as the research team assert, other apes in the Kibale Forest ask their mates to groom a particular part of their bodies (Pika y Mitani, 2006).

This study took place in a western country where babies experience distal caregiving practices; the manifestation of being an object and a subject of contemplation might be different in cultures that employ proximal caregiving practices, that is, cultures in which infants are nearly always in physical contact with their caretakers (Bard et al., 2021). In those societies, babies might also feel the focus of a subject of contemplation’s attention and making tactile contact might be the way in which these children express the feelings that objects and events arise, (i.e., protodeclaratives) as Sorenson (1979) suggested, when he observed the Fore People in New Guinea.

Did Chimp Babies Participate in Cooperative Play?

In this study, we observed two kinds of playful triangular interactions that had a cooperative nature. In the first, the cooperative game was the result of both partners having the same goal as in the ‘distance game’. They shared the goal of trying to get an object. Both Noelia and Jorge enjoyed it. In a sense, the game was cooperative since, as Bekoff (2001) explained, no one hurt or struggled with the other to get the object wanted. What was funny was ‘trying’ to get it and failing. But this kind of cooperative game between intentional beings does not require fitting the own subjectivity to the subjectivity of others. The partners have the same goal, but no one’s intentions must fit the other’s ones.

The second kind of cooperative game had the three features that characterized cooperative interactions (Moll & Tomasello, 2007): (1) a shared joint goal (i.e., closing a green barrel), (2) the existence of complementary roles (to close the barrel, placing half of the barrel on top of the half that the researcher had on his hand) and, (3) the willingness to help one another in order to accomplish their aim.

Perceiving the other just as an intentional agent may lead to attribute the other determined interests and intentions (i.e., an interest in getting objects). However, Noelia gave the barrel to the adult to have it opened. Gómez (2010) also observed how a baby gorilla progressively acquired the skill to engage in cooperative behaviours with a person although Warneken et al. (2006) did not see chimpanzees involved in these kinds of cooperative games. Perinat and Dalmau (1989) could not observe either of these kinds of cooperation in manipulating objects among the games played by human and gorilla babies.

In order to develop their coding system, Bard et al. (2021) made narratives of observations and described the triadic interactions that dyads of humans and chimpanzees, that lived in different cultures and contexts, displayed. Unfortunately, these descriptions were not included in their paper. It would have been interesting to check if any of them participated in this kind of cooperative play.

Noelia’s protoimperatives show that she trusted her human partners enough to give them the object she was interested in, and that she expected them to perform just the action she was requiring of them, instead of taking the object and running with it away. This kind of trust is something that could be added to the social cognitive or behavioural skills that chimpanzees may acquire because of human influence (for further discussion see Tomasello & Call, 2004).

We propose that in these interactions the adult’s intentions fitted Noelia’s ones, which followed the motor affordances of the object. Through the period of observation, Noelia learnt that persons may fulfil her intentions and that she could rely on them and to take a complementary role. But her human partners were the ones who fit Noelia’s intentions; it was, as Gómez (1998) explained, a one-sided-intersubjectivity.

Noelia´s mind was not prepared to fit the other´s mental states. She did not perceive individuals as objects of contemplation and consequently she could neither fit her intentions to the partner’s intention nor could she perceive reality as her partners did. She did not perform on objects any actions that were not dictated by the objects themselves, that is she just perceived in the objects the traits that showed their utility for fulfilling her needs (Gibson, 1982); (see Tomasello, 1999, for further discussion).

On the contrary, Jorge showed that he had begun to grasp the social meaning of objects because he started to use the toy phone to talk and to follow the adult’s goal of making a tower.

The PPD expanded these triangular interactions to those which required performing actions that were not suggested by objects, but by one of the partner’s mental image or imagination. These interactions required the partners to perceive the intention of the other as an object of contemplation and, accordingly, to fit to it their own intentions.

Jorge was interested not only in the object but also in what the other wanted to do with that object, and therefore he was able to make his psychological states try to fit those of others. This capacity which rests on an ‘attributive unit’ (Rivière personal communication) makes Jorge able to admit that a person mediates his or her relationship with the world, to perceive reality as each other does it, from their point of view, and to perform actions to fulfil the partner’s goals. These abilities are the core of that secondary intersubjectivity which opens the door of the cultural world (Trevarthen, 1989).

Limitation of the Study

This study has some limitations. The most important one is the adoption of the perspective that Henrich et al. (2010) called `WEIRD’ (Western, Educated, Industrialized, Rich, and Democratic) (Bard et al., 2021). It is a perspective that emphasizes objects and distal interactions, and it only applies to certain groups of human infants. We adopted this approach because in this study we were observing western dyads. However, it is important to note that we believe that having the ‘attributive unit’ makes distal interactions possible, and also that those distal interactions could adopt a different form in different cultures. The ‘attributive unit’ may lead the distal interactions displayed by the babies who come from non-western rural societies and their mothers to take a different form; those babies might not need to call their mothers’ attention on themselves (in a ‘western way’), because they might feel tactilely that they are frequently the focus of their mothers’ attention. Bard et al. (2021) showed that they get involved in ‘ritual activity’ and employ a tactile way to share the emotions they feel when they see objects and persons, a communicative act that they named: ‘tactile engagement with objects’ and that Sorenson (1979) has also described; and obviously, they might also enjoy ‘contemplating’ objects and sounds. Unfortunately, we did not include one non-western human dyad in our study. It would be really interesting to replicate this study in one of those societies, and to be able to analyse how their dyadic and triadic interactions developed over a year and a half. The only chimpanzee family that could participate in this study lived in a zoo. This is a ‘Rare Rearing Environment’ (Leavens, Bard, & Hopkins, 2010), and means that the results obtained could not be generalized to the whole population of chimpanzees. It would be interesting to analyse the development of the dyadic interactions with and without objects that mother and baby chimpanzees that live in different contexts establish.

Another important limitation is that the babies who participated in this research could not be matched to their pre-observational experience, age, and sex. The contexts in which the dyads were observed were also different. So, obviously, the results of this study cannot be generalized to the whole respective populations. Furthermore, this research cannot prove if the differences found were due to the babies’ earlier experience, or any of the other variables stated above. This study, which was of an exploratory nature, identifies the similarities and the differences between the interactions displayed by the different dyads, and proposes some hypothetical explanations. Further research should also be carried out to examine the correspondence between the behavioural patterns observed and the psychological processes that were proposed.

Conclusions

This exploratory research has lead us to propose some hypotheses: (1) that the differences between Jorge and the baby chimpanzees’ interactions had a motivational nature, (2) that, probably, this motivational difference made Jorge’s mind have an ‘attributive unit’ which made it able to perceive people and objects as objects of contemplations, (3) that this kind of perception made it possible to establish, with the physical and the social world, interactions in which the mind tried to fit those worlds (cooperative interactions), and (4) that the experience of having been perceived as an object of contemplation led Jorge to express the own mental states, i.e., the ones which led him to perform protodeclaratives among them. All of these hypotheses assume that there are intrinsic and evolutionarily derived differences between Jorge and the two chimpanzees.