1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References

Ten dyads were observed biweekly from 10 to 24 months of infant age while playing together at home with a set of toys. The aim was to examine whether mother–infant coregulation changes over the second year of the infant’s life and whether there are individual differences in that process. Normative trends as well as variability between and within dyads were tested using a multilevel modeling technique. We found that unilateral coregulation, in which only the mother was actively involved in play, largely prevailed at the beginning of the year and then decreased linearly, while symmetrical patterns, implying that the infant was also involved, were for the most part absent at the beginning but then increased rapidly, overtaking unilateral from the middle of the year on and becoming predominant by the end. In particular, symmetrical episodes of shared affect and shared action increased first and then decreased, being replaced by shared language. Variability in data was significant between the dyads, with some dyads advancing toward symmetrical coregulation at an earlier age and more rapidly than the others. It was also significant within the dyads, as the increase in symmetrical coregulation unfolded in a quite irregular manner across the sessions, unlike the decrease in unilateral. Results are discussed with reference to a view of joint attention development as a gradual and complex process.

The ability to coordinate attention to an object and interest in a person is considered a key achievement in infant development. In the early months of life, infants are unable to attend to both of these foci at the same time (Kaye & Fogel, 1980; Trevarthen & Hubley, 1978). At around 6 months of age, however, they begin switching their gaze back and forth between the caregiver and an object (Newson & Newson, 1975), and a few months later they are also capable of clearly signaling their attempts to share with someone something outside the social interaction (Moore & Dunham, 1995). This change in attention patterns allows the mother–infant communicative system to change as well. When the mother’s face is the only object of the infant’s interest, the interaction is dyadic in nature, with the interactive process as the goal and the sharing of affect as the content (Brazelton, Koslowsky, & Main, 1974; Stern, 1974; Tronick, Als, & Adamson, 1979). When the infant’s attention to an external entity is embedded in a social exchange, the interaction becomes triadic: the infant is able to share with its partner a referent, which works as the “topic” of their joint concern (Bruner, 1983;Murphy & Messer, 1977).

Many studies have provided evidence of joint attention skills at the end of the first year of life, clearly showing the infant’s ability both to follow and to direct the other’s attention in various domains, including prelinguistic communication, imitative acts, and social referencing (for an overview, see Carpendale & Lewis, 2006; Carpenter, Nagell, & Tomasello, 1998). Nevertheless, disagreement still exists on how to interpret these skills. According to some studies, joint attention represents a unitary construct that depends on a single cognitive process—either general, such as representational capacity (Bates, Benigni, Bretherton, Camaioni, & Volterra, 1979; Leslie & Happe, 1989) and IQ (Smith & Ulvund, 2003) or specific, such as social understanding (Bretherthon, 1991; Brooks & Meltzoff, 2005; Carpenter et al., 1998; Tomasello, 1995a, 1995b, 1999; Tomasello, Carpenter, Call, Behne, & Moll, 2005). According to others, joint attention includes two distinct abilities—that of initiating an episode of joint attention and that of responding to it—which relate to different skills, follow different developmental pathways (Mundy & Sigman, 2006; Mundy et al., 2007; Slaughter & McConnell, 2003), and originate in different brain regions (Mundy, Card, & Fox, 2000). It is thus a multifaceted construct that reflects the development of multiple processes.

Although they are credited with joint attention skills, 1-year-olds prove to be quite poor at using these skills in play episodes of triadic interaction. In their pivotal study, Bakeman and Adamson (1984) observed infants from 6 to 18 months of age playing at home with their mothers and a set of appropriate toys. Only one third of 9-month-olds was found to engage in coordinated joint play. Moreover, the amount of time spent in that kind of play did not exceed 10% of the total play period until the age of 15 months, and only at 18 months were all infants observed in coordinated episodes at least once. The authors concluded that joint attention begins very early in life but develops very slowly. The same conclusion was drawn in a more recent study (Adamson, Bakeman, & Deckner, 2004) covering a subsequent age period, from 18 to 30 months, when the triadic ability is well established and becomes infused with symbols. Children were found to advance into the symbolic level of joint engagement as slowly as they had into the presymbolic level the year before. In particular, children were able to use symbols routinely only at the end of the observed period and mainly in supported episodes, where most of the responsibility for sharing fell on the mother rather than on the child. Even then, only 50% of the time spent in shared activity was symbol infused, meaning that 30-month-old children still do not use language as an integral part of an activity and need more developmental time before they are able to do so routinely (Nelson, 1996).

The gap between the first display of coordinated attention and its use in social play may be owed to the communicative demands that social play places on young children. As we found in previous studies (Aureli, 1994; Camaioni & Aureli, 2002; Camaioni, Aureli, Bellagamba & Fogel, 2003), social play in the second year of life provides mothers and infants with a privileged context for sharing meanings: first at a literal level, when the infants are at the beginning of the year and are able to act upon the toys only in a functional manner; second at a conventional level by the middle of the year, when infants begin to use toys in an adult-like manner; and finally at a symbolic level toward the end of the second year, with infants beginning to refer to the toys verbally. Therefore, social play, far from being just a playful occasion for mothers and infants to have fun together, works as a special form of triadic interaction, suited to introducing infants to the domains of cultural artifacts, such as conventional norms and symbolic language (Bruner, 1975, 1982; Tomasello, 1999). In that respect, it requires the infants to make many adjustments in order to act as full participants as they must: (1) focus on the same object as their partner, which means directing attention in a way that is different from dyadic interaction, (2) use that object together with the partner, therefore acting according to the other’s actions, and (3) say something in line with their partner’s comments and thus use language in a dialogic manner.

Joint attention skills by the end of the first year of life are too immature to allow infants to satisfy those requirements, as the infants’ poor performance in Bakeman and Adamson’s (1984) study clearly showed. Indeed, to perform better, those infants should have put their joint attention skills to work in a context of shared activity and used them as a means of collaborating—rather than simply “playing”—with another person. Becoming an effective partner in collaborative interaction is, however, a gradual process. As we know from the literature on early cooperation, infants are relatively incompetent in that domain, whatever form the cooperation may take: 12- to 18-month-olds involved in social games do not go beyond ritualized interactions (Ross & Lollis, 1987) and 14-month-old infants fail to coordinate their actions with those of another person in problem-solving tasks (Warnecken, Chen, & Tomasello, 2006; Warnecken & Tomasello, 2007). Moreover, the partner must be an adult as children can not collaborate with a peer before the age of two (e.g., Brownell & Carriger, 1990, 1991; Brownell, Ramani, & Zerwas, 2006; Eckerman, 1993; Eckerman & Peterman, 2001; Eckerman & Stein, 1990; Goldman & Ross, 1978; Hay, 1979).

Recently, the emergence and early improvement of cooperative skills has been related to the development of infants’ social understanding (Brownell et al., 2006). According to the social cognitive perspective, infants approaching their first birthday recognize other people as intentional agents like themselves and therefore come to appreciate them as potential partners in collaborative interactions. Some months later, the achievement of the so-called “we intentionality” (Tomasello et al., 2005) makes infants capable of true cooperation, with the ability of sharing goals and coordinating actions for pursuing those goals (for an overview, see Behne et al., 2008); later on, the emergence of the theory of mind provides early preschoolers with the perspective-taking ability, which allows them to collaborate systematically and explicitly with a partner, such as a peer (Ashley & Tomasello, 1998; Brownell et al., 2006; Smiley, 2001), who interacts in a more unpredictable way than an adult.

Joint attention is at the core of this perspective, as the ability simultaneously to pay attention to a person and an object is considered the basic prerequisite of cooperation (Brinck & Gärdenfors, 2003; Tomasello et al., 2005). Therefore, the two abilities are supposed to be related from early on in ontogeny. In fact, Brownell et al. (2006), who directly compared joint attention and cooperative skills, provided evidence of this relationship, finding that toddlers who responded more frequently to the joint attention bids of an adult were able to coordinate better with their peer partner. On the other hand, we have seen that 1-year-olds are capable of joint attention and very poor at collaborating with another person, even when that person is a responsive adult, such as their mother. We also saw that it takes a year before they become capable of doing so routinely with an adult and even longer when collaborating with a peer. Further research is therefore needed to examine the origins of the relation between joint attention and cooperation and how it evolves over the course of development (Bronwell, Nichols, & Svetlova, 2005).

A fuller consideration of infants’ concrete experience in social interaction would contribute to that aim. We argue that the emphasis placed by joint attention research on early sociocognitive skills has largely contributed to conceiving joint attention development as an internal process, which can be properly explained only by referring to the infant’s representational capacity. Therefore, the role of social practice has largely been overlooked and early advancements in triadic interaction have not been recognized as unfolding as gradually as they appear to do. A perspective that emphasizes social understanding as an action-based process rather than a representational one may help overcome this shortcoming. According to Carpendale and Lewis (2006), joint attention behaviors are social skills that infants practice, improve, and refine by participating day after day in the real network of social interactions and that develop as the infants learn to combine these skills in increasingly complex and varied ways, with different partners, for different purposes and in different contexts (Bibok, Carpendale, & Lewis, 2008). In fact, social practice and cognitive skills are by no means independent or mutually exclusive sources of development and the two perspectives should be viewed as complementary rather than as opposite, in a closer examination of the mechanisms underlying the genesis and development of joint attention. Nevertheless, the focus on practice and learning in favor of which theorists of an action-based account of infant knowledge strongly argue (Racine & Carpendale, 2007) can be especially valuable, as it contributes to more balanced research on infant sociocognitive development than has been true historically. With respect to the current study, this focus is also beneficial, insofar as it relates the large gap between the emergence of joint attention and its efficient use in collaborative activities to the infant’s lack of specific experience.

From this perspective, we will examine social play over the second year of life with the aim of documenting the gradual development of the infant’s ability to coordinate with another person, from the time when infants are largely inattentive to their partner to when they become capable of taking into account what the partner is actually doing and saying. As our emphasis is on experience with other people as constitutive of the infant’s social development, we were interested not just in some kind of preexisting abilities supposed to act as internal forces driving the individual behavior, but in the interpersonal functioning of individuals when interacting. To analyze the developmental process in such a dynamic and situated manner, we referred to Fogel’s (1993, 2006) model of interaction as a continuous process of coregulation between the partners instead of a contiguity of discrete acts, emitted from one partner to the other. We thus observed infants’ behavior as far as it relates to their mother’s behavior, focusing not on each of the two partners separately but on their reciprocal adjustment in the ongoing interaction. As we expected to find changes in this process, we collected data in an intensive way by observing dyads bi-weekly. Moreover, as our frequent observation, multiple case, longitudinal research design provides an excellent opportunity for studying developmental trajectories (Lavelli & Fogel, 2002), we applied a multilevel modeling technique to our data in order to test normative trends and individual differences. Last, as social play occurs in an everyday context, we observed our subjects in their homes in order to strengthen the ecological validity of the study.

We examined mother–infant interaction in free play in order to observe the coregulation process as it unfolds spontaneously. In fact, although free play requires the partners to coordinate with each other triadically, as in any other collaborative activity, it does not imply a rigid set of rules, as social games do, or an explicit goal to be achieved by means of specific temporally and spatially situated actions, as problem-solving tasks do (for a similar account, see Brownell & Carriger, 1990). Instead, it gives the partners much greater freedom to choose which behaviors to adopt in order to coordinate with each other. Moreover, free play is an optimal opportunity for the infants as well, as it usually involves a parent, who typically behaves as a sensitive partner in a scaffolding-like manner, i.e., interpreting infants’ interests and goals and adapting her/his own behavior accordingly (Bornstein, 1989; Conner & Cross, 2003; Kochanska & Aksan, 2004).

Following Fogel’s theory (1993), our interest was in how mothers and children jointly contribute to the interaction. Therefore, unlike Bakeman and Adamson’s (1984), Adamson and Bakeman's (1985), and Bakeman and Gottman’s (1986) (more recently, see Bigelow, Maclean, & Proctor, 2004) studies on social play and unlike most research on social interaction (a recent example is Kochanska & Aksan, 2004), we chose the dyad as the unit of analysis rather than the individual (the infant or the mother). Accordingly, we coded that unit as a single entity, using an instrument which has been designed for the purpose of observing interaction per se, i.e., the Relational Coding System (Fogel & Lyra, 1997). Based on a corollary of Fogel’s (1993) relational theory, which posits that the organized patterns of behavior are to be found in the whole system of communication rather than in one of its components, this instrument captures the ways in which the partners adjust to each other continuously while interacting. Different patterns of coregulation are identified that correspond to the nature of this adjustment: unilateral, when only one partner is paying attention to the other while the other is engaged in something else; asymmetrical, when there is a joint focus of attention but only one partner is elaborating on the activity while the other only observes; and symmetrical, where both partners adapt to each other and together come up with innovative ways to take part in an activity.

Unlike previous studies that used the Relational Coding System to examine the first few months of life (Hsu & Fogel, 2001; Lavelli, 2005), our study focuses on a later period, from 10 to 24 months of age. It therefore contributes to extending the analysis of interpersonal coregulation from face-to-face interaction to mother–infant–object interaction. We also partly modified the original coding system according to the developmental changes in the content of interaction shown by previous studies. They found that in the first half of the second year of life infants use affective expressions (Bakeman & Adamson, 1984) or manipulative actions (Bakeman & Adamson, 1984; Camaioni et al., 2003) to interact with their mother during social play; later, with the advancement of representational skills, infants begin to produce linguistic expressions related to shared activity, such as protowords and words (Adamson et al., 2004; Camaioni et al., 2003). To account for these possible changes during the observed period, we divided symmetrical coregulation into three subcategories, which, in line with the above results, aimed at coding episodes in which affect, action or language is shared.

We expected to find a developmental sequence in the predominant patterns shared by the dyads to achieve coregulation. With respect to the main codes, the unilateral pattern would prevail at the beginning and then decrease, and the symmetrical pattern would emerge later and then increase. As the asymmetrical pattern seems to merge some features of the other two—with infants paying attention to the mother’s focus, as in symmetrical, while refraining from acting together, as in unilateral—it has been presumed to work as a transitional state between the unilateral and the symmetrical. With respect to the subcodes, we also expected symmetrical coregulation to change with advancing age, with affect sharing and action sharing occurring first and language sharing occurring later. In fact, the former patterns employ skills, like expressive and motor acts, that are already part of the infant’s repertoire at the beginning of the observational period, to communicate with others in person-focused interaction or to explore physical reality in object-focused interaction, respectively. By contrast, the latter pattern requires skills that infants still lack at the outset and that may be recruited for coregulation only in a subsequent period. Finally, as shown in previous studies on social play (Camaioni et al., 2003), we expected to see individual differences in the rate of developmental change.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References

Participants and observational procedure

Because of the focus on developmental change and individual differences, a multiple case study method (Camaioni et al., 2003; Fogel, 1990; Hsu & Fogel, 2001; Lavelli & Fogel, 2002) was used. This method implies a multiple timepoint design, providing a dual opportunity to make meaningful statements about the group and also to capture the rate and the shape of developmental trajectories for each case. Ten dyads were video-taped weekly at home, interacting with a toy tea set (dishes, forks, knives, spoons, cups, etc.) brought by the observer. Four girls and six boys were observed, with the girls belonging to dyads 1, 4, 8, 9 and the boys to dyads 2, 3, 5–7, 10. All of the infants were full term at birth; five of them were first borns, four were second borns, and one was third born. All children belonged to biparental middle-class families, living in a town of central Italy. The observations started when infants were 10-months-old (M = 10.7 months) and continued until they were 24-months-old (M = 24.9 months). Each session lasted about 5 min (M = 5 min 2 sec). Mothers were sitting with their infants at their favorite table with the toy tea set at their disposal. No other instruction was given to them than to play as usual and to ignore the observer as much as possible. All the mothers were informed about the general interest of our study and all of them agreed to participate. At the end of the study, they received an edited tape of the observational periods as a gift for their intensive participation in the project.

Coding scheme

The Relational Coding Scheme developed by Alan Fogel (1993) was employed to assess mother–infant coregulation. In Fogel’s (1993) account, interpersonal relationships have regularly recurring patterns of communication, called “frames,” which segment co-actions into meaningful units, defined by the occurrence of: (1) a coherent theme, (2) a particular location in space or time, and (3) the direction of attention between participants. The units identified by the Relational Coding Scheme represent different patterns of mutual adjustment between partners and therefore the interaction corresponds to a sequence of episodes defined by an action of a partner followed by an opportunity to act for the other. To take an oral conversation as an example, one partner talks and at the same time provides the other with a variety of opportunities to reply. The partner can reply in a way that follows on from the other’s content, at the same time bringing into the conversation something new; so, their communicative episode can be considered to be coregulated in a reciprocal manner. According to the coding system, the coregulation forms we observe in a communicative process vary from unengaged to unilateral to asymmetrical to symmetrical coregulation, and breakdowns in communication can also occur (see Table 1 for the operational definitions). For the purpose of the present study, the symmetrical code was divided into three subcodes—affect, action, and language, respectively—so, the original scheme has been partly modified (see Table 1). Coding was done continuously from the video by two independent coders and the coregulation states were identified by segmenting joint activity into units, lasting at least 3 sec, corresponding to the above categories. The onset time of each code was also recorded.

Table 1.    Operational Definitions of Mother–Child Coregulation Categories and Selected Examples from Mother–Child Activity With the Toy Set
UnilateralA partner acts and elaborates upon the activity with respect to the other partner, who is engaging in something else. For example: M looks at C, who is playing with the cups and says in Italian: “[C’s name], feed the baby! She’s hungry! [C’s name], give the baby some soup!” C continues to play with the cups.
AsymmetricalA partner is engaged in an activity which the other partner too is interested in; however, the second is merely attending without commenting or acting. For example: M picks up the plastic apple and the knife, while C is puts the cup on the plate. M then says (in Italian): “Come on! Let’s cut this apple and have a snack!” B interrupts her activity and looks at M’s action.
SymmetricalBoth partners participate in a game by emoting, acting, and/or commenting together on a common focus.
 AffectBoth partners show enjoyment in the activity by exchanging expressive signals—postural, vocal, and facial—with affective content. The mother can accompany these interpersonal exchanges with language. For example: B looks at M and makes some funny noises with an amused facial expression. M smiles. C vocalizes and waves her hands.
 ActionBoth partners contribute to the elaboration upon the activity by acting in a joint manner. The mother can accompany the activity by also commenting on it. For example: M drinks from the empty glass, then approaches the glass to C’s mouth; C opens his mouth and bends his head forward.
 LanguageBoth partners act jointly and also comment on a common topic. For example: C puts the spoon in a cup while saying (in Italian): “Coffee” and M replies (in Italian): Great! I’m dying for a cup of coffee!”
UnengagedEach partner is engaged in her own activity without paying attention to the other. For example: M dresses the doll, while C is engaged in nesting cups.
DisruptionA partner interrupts the other’s action in an inappropriate manner. The interrupted partner resists the partner’s attempt. For example: C puts the spoon in her mouth. M says (in Italian): “Not in the mouth!” and attempts to take it out. C refuses to let go of the spoon.


From the coding records, durations of each category were computed and used as measures for the analysis. Because of slight variations in the session length, the raw durations in each session were transformed into proportions according to the duration of that session (proportional durations). Proportions of categories of less than .5% were excluded from the data analysis. Interobserver reliability was calculated on 30% of the entire data set. To be specific, 30% of sessions were randomly sampled for each dyad from each of the following three age periods: 44–64, 65–88, and 85–104 weeks (Bakeman & Gottman, 1986, p. 77). Kappa assessments were based on whether two independent coders agreed about the category coded in each second. Across all categories, the average kappa was not less than 80% in each of the three periods. Hierarchical random effects modeling (Goldstein, 1995, 2003; Snijders & Bosker, 1999) was used to test the advanced hypotheses. MLwiN statistical software was used to implement all the models (Goldstein et al., 1998). In the present study, data were collected on a two-level hierarchy (Rasbash, Steele, Browne, & Prosser, 2005), with the dyads at the higher level (level 2) and the set of measurement occasions (i.e., the infant’s age in weeks) for each dyad at the lower level (level 1). This conditioning of measurements over dyads allowed us to model both a growth curve common to all dyads and a growth curve specific to each dyad. To test the hypothesis, a polynomial regression function of n degree (= number of occasions minus 1) was used to model the outcome variables as a function of time for both level 2 (dyads) and level 1 (measurements; Plewis, 1996). As we were interested in linear and curvilinear (squared and cubic) trends, the average developmental curve was modeled by a third-degree polynomial function written as follows:

  • image

where Ytj is the proportional duration of variables in the jth subject at occasion t, xtj is the chronological age of the jth dyad at occasion t (the age was centered around the mean). Parameters β0, β1, β2, and β3 are estimated from data and are labeled, respectively, the intercept, linear, quadratic, and cubic coefficients. Taken together, these parameters describe the shape of the average developmental curve and are considered as fixed parameters for all dyads in the model. The age was centered around the mean in order to make the intercept parameter interpretable. To account for the variance at both dyad and age level (levels 1 and 2, respectively), the u and e parameters were considered. As the random intercept parameters (u0j and e0tj) and the random slope parameters (for linear effect: u1j and e1tj) capture the departure of each dyad from the fixed (average) parameters of intercept and developmental curve, respectively, variance/covariance parameters (σ2) were estimated for dyad level (for the random intercepts, σ2u0, slopes, σ2u1, and their covariance, σ2u01) and for age level (random intercepts, σ2e0, slopes, σ2e1, and their covariance, σ2e01). If significantly different from zero, these parameters indicated significant differences among dyads at their initial status and/or in their growth curves.

To control for the influence of background variables, the effects of infant’s gender and birth order as well as the interaction effects between each of these two variables and the infant’s age were tested. These effects were analyzed when significant.

Finally, a more elaborate regression model was explored for language coregulation patterns. To be specific, we asked whether, after controlling for the effect of the infants’ gender and the three (linear, quadratic, and cubic) effects of age, the direct effect of symmetrical coregulation as well as its interaction with the linear effect of age still predicted language proportional duration. This model is known as the Full Model to distinguish it from the Base Model that includes the same effects investigated for all the other coregulation patterns.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References

Mother–infant unilateral, asymmetrical, and symmetrical coregulation were analyzed first, according to the Fogel’s (1993) original coding system; then, symmetrical coregulation was analyzed in more detail using the subcategories created for this purpose (see the Method section).

Developmental trajectories for unilateral, asymmetrical, and symmetrical coregulation patterns

Our first hypothesis was that there are age effects on dyadic coregulation in mother–infant joint activity during the second year of life. In particular, we expected unilateral patterns to prevail at an earlier period and symmetrical to prevail later. Asymmetrical patterns were supposed to be a transient frame between the two, emerging first, then peaking, and then declining.

With respect to group data (fixed effects; Table 2), the intercept parameters were significant for unilateral, asymmetrical, and symmetrical patterns (χ2[1] = 79.17, p < .001; χ2[1] = 87.64, p < .001; χ2[1] = 60.44, p < .001, respectively); the linear effect of age (β1) was significant for each pattern (χ2[1] = 7.79, p < .01; χ2[1] = 7.06, p < .01; χ2[1] = 12.20, p < .01, respectively); and a quadratic effect (β2) was significant for asymmetrical and symmetrical patterns (χ2[1] = 16.81, p < .01; χ2[1] = 7.21, p < .01, respectively).

Table 2.    Fixed and Random Coefficients of the Modeled Developmental Trajectories for Coregulation Categories
 UnilateralAsymmetricalSymmetricalAffectActionLanguage (Base Model)Language (Full Model)
  1. Note. β0 = constant effect; β1 = linear effect of age; β2 = squared effect of age; β3 = cubic effect of age; σ2u0 = interdyadic variance for intercept; σ2u01 = interdyadic covariance between intercept and age linear trend; σ2u1 = interdyadic variance in age linear trend; σ2u02 = interdyadic covariance between intercept and age quadratic trend; σ2u2 = interdyadic variance for age quadratic trend; σ2e0 = intradyadic variance for intercept; σ2e01 = intradyadic covariance between intercept and age linear trend; σ2e1 = intradyadic variance for age linear trend; σ2e02 = intradyadic covariance between intercept and age quadratic trend; σ2e2 = intradyadic variance for age quadratic trend.

  2. Standard errors in parentheses.

  3. * p  <  .05. ** p  <  .01.

Fixed Part
β0: Constant .44559** (.05007).08707** (.00930).45086** (.05799).06928** (.02653).25701** (.02988).05485 (.03572).05446* (.02249)
Direct effects:
Gender.09831 (.06380)−.02290* (.01141).08237 (.07415).06063 (.03359)−.11026** (.03851).07639** (.02125).06383** (.01783)
Symmetrical      .38671** (.03217)
β1: Age linear trend −.00691** (.00248)−.00155** .00058).00880** (.00252).00019 (.00108).00050 (.00077).00645** (.00150).00278** (0.00103)
β2: Age squared trend .00002 (.00003).00006** (.00001)−.00007** (.00003)−.00007** (.00002)-.00018** (.00007).00021** (.00003).00010** (.00003)
β3: Age cubic trend .00000 (.00000).00000 (.00000).00000 (.00000).00000 (.00000).00000 (.00000).00000 (.00000).00000 (.00000)
Interaction effects:
Gender × Age linear trend.00060 (.00298).00016 (.00055).00098 (.00303).00182 (.00115)   
Gender × Age squared trend    .00009 (.00010)−.00012** (.00004)−.00010** (.00003)
Symmetrical × Age linear trend      −.01486** (.00151)
Random Level: Dyad
σ2u0: Constant .00930* (.00436).00022 (.00014).01271* (.00589).00250* (.00121).00282 (.00158).01035* (.00480).00331* (.00160)
σ2u01: Age linear trend/constant .00029 (.00017).00000 (.00000).00036 (.00020)−.00008* (.00004) .00043* (.00020).00014* (.0007)
σ2u1: Age linear trend .00002* (.00001).00000 (.00000).00002* (.00001).00000 (.00000) .00002* (.00001).00001* (.00000)
σ2u02: Age squared trend/constant     .00000 (.00000)  
σ2u2: Age squared trend     .00000 (.00000)  
Random Level: Age
σ2e0: Constant .01881** (.00204).00227** (.00031).01871** (.00202).00785** (.00088).01406** (.00181).01285** (.00100).00972** (.00074)
σ2e01: Age linear trend /constant .00001 (.00004).00000 (.00001).00013** (.00004).00004* (.00002) .00040** (.00004).00027** (.00003)
σ2e1: Age linear trend −.00000 (.00001).00001* (.00000).00000 (.00001).00000 (.00000) .00001** (.00000).00001** (.00000)
σ2e02: Age squared trend/constant     .00001 (.00001)  
σ2e2: Age squared trend     .00000 (.00000)  
FIT Statistics
-2*log likelihood: 419.851089.54415.60759.33628.50723.63840.04
Units: dyad10101010101010
Units: age410410410410410410410

As in Figure 1, unilateral and asymmetrical patterns decreased during the second year of life, whereas symmetrical increased. In particular, unilateral prevailed at the beginnings of the year and decreased gradually and linearly, whereas symmetrical increased rapidly (but nonlinearly) and crossed over unilateral at around the 20th month. Asymmetrical patterns were a little more frequent than symmetrical at the beginning, they then decreased rapidly and remained very low until the end.


Figure 1.  Observed versus predicted average developmental trajectories for unilateral, asymmetrical, and symmetrical coregulation patterns (grouped data).

Download figure to PowerPoint

According to our first hypothesis, mother–infant coregulation in the second year of life followed a developmental sequence, with unilateral occurring first as the predominant form of communication and symmetrical occurring later. Contrary to our hypothesis, asymmetrical decreased gradually instead of showing an inverted U-shaped trajectory, thus revealing that it did not play a bridging role in the transition between the other two frames.

Only asymmetrical patterns were influenced by the fixed effect of infant’s gender (χ2[1] = 4.02, p < .05), with girls showing greater proportional durations of this pattern than boys.

With respect to interindividual variability (random effect at two-level variance, Table 2), dyads differed in unilateral and symmetrical patterns, both with respect to the initial status (random intercept effects [σ2u0], χ2[1] = 4.54, p < .05; χ2[1] = 4.66, p < .05, respectively) and the growth rate (random slopes for linear effects of age [σ2u1]; χ2[1] = 4.28, p < .05; χ2[1] = 4.32, p < .05, respectively). As in Figure 2, unilateral decreased very rapidly for half of the dyads (dyads 2, 7–10) and remained high and practically unaltered for the other half. Dyads also differed with respect to symmetrical trend as shown in Figure 3; all of them were quite low at the beginning, but at around 15 months half of them (dyads 2, 7–10) increased much steeper than the other half. In both cases, the initial differences became greater as a function of time.


Figure 2.  Individual developmental trajectories for unilateral patterns.

Download figure to PowerPoint


Figure 3.  Individual developmental trajectories for symmetrical patterns.

Download figure to PowerPoint

Finally, with respect to intraindividual variance—i.e., variability owing to differences within each dyad across observations (random level 1 variance)—two significant effects were found: the linear effect of age for asymmetrical patterns (σ2e1 =0.00001, χ2[1] = 23.90, p < .01) and the covariance effect between the intercept and the linear effect of age (σ2e01 =0.00013, χ2[1] = 8.79, p < .01) for symmetrical. Therefore, the variability of the proportional duration of these two frames within dyads was a function of time. To be more precise, asymmetrical intradyadic variability showed a U-shaped relationship, indicating a maximum of variability both at the beginning (11th month) and at the end (24th month) with a minimum variability around the 18th month; symmetrical intradyadic variability increased with time so that the proportional durations of symmetrical patterns differed more in the latter part of the year than in the former. This greater variability between sessions at the end compared with the beginning could signal a certain degree of systematic fluctuation for symmetrical patterns. It was not found for either unilateral or asymmetrical.

Developmental trajectories for affect, action, and language frames

The second hypothesis of the study was about the age effects on each of the three different types of symmetrical coregulation. We expected that affect and action patterns would be prevalent at an earlier age and verbal exchanges would be prevalent at the end.

With respect to group data (fixed effects), Table 2 shows a quadratic fixed effect (β2) of age for affect, action, and language (χ2[1] = 15.68, p < .001; χ2[1] = 5.97, p < .05; χ2[1] = 10.51, p < .001, respectively) and an additional linear effect (β1) for language (χ2[1] = 7.25, p < .001). As shown in Figure 4, proportional durations for both affect and action were very low at the beginning, then increased to a peak at 18 months (70 weeks), and finally decreased slowly, in both cases showing an inverted U-shaped trajectory. Language was almost absent in the first weeks and accelerated nonlinearly between the 14th and 18th months, crossing the affect curve after the 18th month and the action curve at the 21st month; then it continued to increase very steeply until the end. Therefore, mother–infant symmetrical coregulation advanced in the second year of life from nonverbal to verbal form, as hypothesized. An unexpected result, however, was that affect and action exchanges evolved through inverted U-shaped trajectories, suggesting a transitional role played by both frames from unilateral to language.


Figure 4.  Observed versus predicted average developmental trajectories for affect, action, and language patterns (grouped data).

Download figure to PowerPoint

Both action and language patterns were significantly affected by gender (χ2[1] = 8.20, p < .01; χ2[1] = 12.92, p < .01, respectively), with boys being lower in action and higher in language than girls. For language patterns, however, this effect was qualified by the interaction effect between gender and time (χ2[1] = 11.80, p < .01). As revealed by the simple slopes analysis (Bauer & Curran, 2005; Cohen, Cohen, West, & Aiken, 2003), girls exhibited a significant increase in language proportional durations as a function of time (β = .0002 [.0000], z = 6.33, p < .001), whereas this relation was not significant for boys (β = .0001 [.0001]; z = 1.27; NS). Actually, we found that at the beginning of the observational period boys were higher in proportional durations of language, but toward the end they were outperformed by girls.

Group data analysis showed that symmetrical and language coregulation followed a similar trend, although displaced over time with the first increasing earlier and the second later in development. Therefore, to shed light on the possible relationship between the two coregulation forms, the interaction effect of linear age by symmetrical pattern was added to the language model (Table 2). The inclusion of this term significantly improved the statistical fit (χ2[2] = 116.41, p < .01) of the new model with respect to the previous one. Moreover, the interaction effect was significant (χ2[1] = 144.46, p < .001; χ2[1] = 97.07, p < .001, respectively). Therefore, an increment in proportional durations of symmetrical predicted significantly an increment in proportional duration of language. As in Figure 5, the simple slopes analysis (Bauer & Curran, 2005; Cohen et al., 2003) showed that dyads who were faster (i.e., 1 SD above the average) in symmetrical proportional durations exhibited higher language proportional durations as a function of time (β = .0058 [.0011], z = 5.06, p < .01), whereas this relation was nonsignificant for the slower dyads (i.e., 1 SD below the mean) in symmetrical proportional durations (β = −.0008 [.0011], z = −.71, p = .4761).


Figure 5.  Simple slopes for the interaction effect of time (linear) and symmetrical patterns on language proportional durations.

Download figure to PowerPoint

Turning to interdyadic differences (random effects, Table 2), affect and language patterns showed significant values. With respect to affect, the covariance between intercept and linear effect of age was significant (χ2[1] = 4.51, p < .05), with the variability decreasing nonlinearly toward the end of the second year of life. As regards language, significant differences between dyads were found for the intercept (σ2u0), the slopes (σ2u1), and the covariance between intercept and slopes (σ2u01) for the linear trend (respectively, χ2[1] = 4.27, p < .05; χ2[1] = 4.13, p < .05; χ2[1] = 4.21, p < .05). As shown in Figure 6, three of 10 dyads (dyads 8–10) started to increase the proportional duration of language patterns from about 14 months (65 weeks), whereas the others remained quite low until 18 months (80 weeks). Only at that age did these latter dyads begin to accelerate, although at a slower rate than the former. Finally, the covariance effect signals that differences among dyads in the use of language become more and more evident over time.


Figure 6.  Individual developmental trajectories for language patterns.

Download figure to PowerPoint

Finally, intradyadic variance for the affect and language patterns showed a systematic time-dependent pattern. As to affect, the covariance between the intercept and the linear effect of age was significant (σ2e01 =.00004, χ2[1] = 3.73, p < .05), meaning that variability among sessions increased at the end of the observational period. As to language, the difference in proportional duration of these frames among sessions was time dependent (σ2e1 = .00001, χ2[1] = 22. 56, p < .00), meaning that this difference increased rapidly and in a nonlinear way with advancing infant age. As the covariance between the intercept and the linear component (σ2e01 =.00027, χ2[1] = 79.77, p < .00) was also significant, the sessions differed more at the end of the second year than at the beginning. Therefore, as for symmetrical patterns, language patterns also increased with a certain degree of fluctuation.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References

This study aimed to examine mother–infant social play in the second year of life. With reference to Fogel’s (1993) model of interaction as a continuous adjustment between partners instead of a sequence of discrete acts, we focused on mother–infant interpersonal functioning during play rather than on individual behaviors. Communicative patterns were identified (Fogel, 1993) to distinguish different forms of coregulation, an intensive longitudinal design was adopted to match the developmental process as closely as possible, a multiple case study was used to make claims about the group as well as the individuals and, finally, a hierarchical linear analysis was performed to model the trajectories of different coregulation forms. We expected to find developmental transitions and individual differences.

Developmental changes

As a first result, we found that unilateral patterns of interaction predominated in the first half of the observational period and decreased gradually, being replaced by symmetrical patterns, which were emergent in the former part of the period, increased sharply, and became prevalent in the latter part. We conclude that social play occurring in the second year of life evolves from episodes when only the mother is sensitive to the infant, by directing her attention to and acting on the object of the infant’s focus, to episodes where both partners are mutually involved and influence each other continuously. This finding parallels Bakeman and Adamson’s (1984) results, thus confirming that infants enter their second year as quite independent agents in social interaction and end that year as effective partners, who appreciate the other’s contributions and are capable of coordinated joint engagement. In our terms, mother–infant dyads playing together around a common set of objects become more coregulated in the course of the second year of infant life by increasing the time devoted to interacting symmetrically.

A closer look at the kind of patterns used by the dyads for achieving symmetry showed that advancement toward a good coregulated interaction was very gradual. Patterns of shared affect and shared actions emerged early, peaked soon after, and then decreased, to be replaced by shared language, which emerged later and then increased. This is an expected finding. Expressive and motor behaviors are commonly used by infants to interact in dyadic contexts—with people or with objects, respectively—and are therefore at their disposal at the outset of social play. Later, such behaviors wane as soon as linguistic skills, which are specifically designated for interacting triadically, become available. From this perspective, shared affect and shared actions are primarily transient forms in coregulation development, used to achieve symmetry in a period when no other content can be shared by mother and infant, and are destined to disappear because of the appearance and strengthening of more advanced skills. A comprehensive view of the whole process suggests, however, a more substantial role played by these two patterns. As we have seen, both shared affect and shared action evolved with an inverted U-shaped trajectory. According to Fogel’s (1993) model of frame transition, such a trajectory signals so-called “bridging frames,” i.e., patterns that mediate the passage from a historically predominant form to an emerging form. As shared affect and shared action occur between an old form—the unilateral—which is decreasing, and a new form—shared language—which is increasing, they resemble such a frame, which mediates the transition from a pattern in which no common focus is shared by the partners to a pattern in which language is also shared.

The occurrence of transitional patterns gives coregulation development the quality of a process that unfolds in a very smooth way. As this quality also characterizes the infant’s advancement in joint engagement, as in Bakeman and Adamson’s (1984) study, the infant’s entry into the domain of triadic interaction seems to be a continuous course of development. This claim is far from being uncontroversial. According to the social cognitive perspective, the ability to be jointly engaged with a partner is brought about by a strong reorganization of infant mind—the so-called 9-month cognitive revolution (Tomasello, 1995a, 1995b, 1999)—occurring at around the end of the first year of life, owing to the emergence of the infant’s understanding of other persons as intentional agents. Therefore, that ability is viewed as a sudden achievement that appears in quite an abrupt way and pushes infants from the dyadic to the triadic period. Recent research has challenged this view. Infants younger than 9 months of age actively coordinate their attention between people and objects (Flom & Pick, 2005; Striano & Bertin, 2005; Striano, Stahl, & Cleveland, 2009) and even 3-month-olds can appreciate the triadic situation if they are provided with a facilitated condition, such as when the adult’s gaze on an object is coordinated with the infant’s gaze (Striano & Stahl, 2005). The few neurophysiological data available so far are also consistent with the above findings, as 5-month-olds’ attention to an object, measured as activation of neural correlates, was higher in joint attention condition, where the experimenter alternated her gaze from the object to the infant’s eyes, than in nonjoint attention condition (Parise, Reid, Stets, & Striano, 2007), and 4-month-olds exhibited enhanced neural processing when looking at an object at which the adult did not look compared with the object the adult looked at, suggesting that the cued object is perceived as more familiar than the uncued one (Reid, Striano, Kaufman, & Johnson, 2004).

Overall, infants appear to be sensitive to key components of triadic interaction very early in development. It is thus hard to argue for a sharp discontinuity between the dyadic and triadic period owing to the alleged sociocognitive shift. Instead, infants’ earlier appreciation of rudimentary aspects of triadic interactions in the dyadic period could represent the first step in joint attention development (Moore, 1996; Striano & Rochat, 1999; Striano & Stahl, 2005), giving it the nature of a process that is “nurtured during the early period of face-to-face play and expands during the emergence of the triadic interactive system” (Bakeman & Adamson, 1984, p. 1288). Indeed, recent literature supports the continuity perspective (Müller, Carpendale, Budwig, & Sokol, 2008). In particular, relational theories, by grounding later intentionality sharing in the emotionally charged episodes occurring in early face-to-face interaction, claim a full continuity from the dyadic to the triadic period (Hobson, 2002; Hobson & Hobson, 2008; Reddy, 2008); conversely, social cognitive theories, by recognizing that a combination of representational and motivational components underlie shared intentionality, limit continuity to the development of the latter, whose origins are traced back to mother–infant affective engagement in proto-conversations occurring in an earlier period of life (Behne et al., 2008; Tomasello et al., 2005). In both cases, a smooth stream of experience seems to accompany infants’ advancement in their attunement to other persons from the dyadic to triadic period (Striano & Stahl, 2005). Our modeled trajectories showing such smoothness even later, in coregulation development over the triadic period, add to this hypothesis.

Individual differences

Looking at the individual trends, we see that all dyads advanced in coregulation according to the same developmental pattern of age-related changes, but differed with respect to the rate of their advance. Half of the dyads were both later and slower in passing from unilateral to symmetrical than the other half, with the latter group departing from the former very early on. Interdyadic differences were even greater in shared language, with three dyads being much earlier and much faster in adopting such an advanced pattern. Moreover, the difference increased in a nonlinear way, meaning that the dyads entered the year provided with quite a similar ability to coregulate and became progressively more different during the year.

To identify some factors responsible for differentiating the dyads with respect to the speed of development, infants’ gender was included in the modeling of language trajectories, and an interaction effect was found: dyads with girls were much lower than dyads with boys at the beginning of the year, but increased later at a faster rate, so that at the end the former outperformed the latter. Interestingly, the age point of this overtaking is around 20 months, virtually coinciding with the so-called vocabulary explosion. Previous studies have already found that girls are more proficient than boys in several measures of linguistic skills (Bornstein & Haynes, 1998) and have also found an interaction effect on early vocabulary growth, with girls being significantly better than boys until 20–24 months but not after (Huttenlocher, Haight, Bryk, Seltzer, & Lyons, 1991). Our data found that dyads with girls performed better than dyads with boys from the age of 20 months. It could be that the greater proficiency of girls at an earlier age, shown by previous studies, is put to work in verbal exchanges later, as our study showed. In other words, girls are more likely than boys to share language in social play as their language is rich enough to infuse joint activity.

Another factor that helps to explain individual differences pertains to the relationship between earlier and later forms of symmetrical coregulation. We found that the rate of increase in proportional duration of shared affect and shared action predicted the rate of shared language. This result, while confirming the “bridging” role played by shared affect and shared action in coregulation development which we saw above, also shows that the more capable the dyads were of achieving symmetry at an earlier age, the more capable they were of it later, although the skills needed for coregulating symmetrically were different, expressive/motor or linguistic, respectively. We also found that the three dyads that showed longer language patterns were also those more capable of symmetry. The role of language in the development of joint engagement has also been underlined by Brinck and Gärdenfors (2003). According to them, earlier forms of joint attention are based on information present in the actual context, but later forms imply communication about absent goals and therefore require agents to use symbolic means of sharing. In their words, “a major reason for evolution of language is that it enhances co-operation” (p. 492). Our study, by showing that symmetrical exchanges are longer in more dialogical dyads, adds to this claim.

Finally, the variance in the observational sessions differed between coregulation patterns, increasing significantly in symmetrical and language patterns and remaining stable in unilateral. In other words, the unilateral form of coregulation decreased over time without fluctuating from one session to the next, whereas symmetrical and language forms increased with an increasing local fluctuation. With reference to the dynamic system perspective, which claims a greater instability of a phenomenon when emerging (Thelen & Smith, 1994), we could trace this difference to the timing of the developmental appearance of the two forms. As symmetrical patterns are emergent in the second year of life, the dyads advance toward the symmetry with a certain degree of uncertainty, so the duration of these patterns increases in an irregular manner. On the other hand, the unilateral form is more familiar in that period and on the wane; so, it decreases in a much more controlled manner.

To conclude, we identified a normative trend in interpersonal coregulation between mother and infant when they interact in social play during the second year of infant life. As we found, coregulation changes from unilateral to symmetrical mode and this change occurs around the middle of the year. We also verified that this trend is not completely predictable but is accompanied by a great deal of individual variability which affects the rate of the transition. As regards the factors which possibly account for this effect, preliminary analyses showed that they differ in relation to different processes. The increase in language exchanges, for example, varied between the dyads owing to some constitutive aspects, such as infant gender moderated by infant age; conversely, differences in symmetrical trends are influenced by earlier modes of interaction, so depending on more particular aspects, such as each dyad’s unique history. Finally, we found that the above trend occurred with some degree of uncertainty, as shown by the significant increase in variability across sessions with respect to symmetrical and language frames. As this phenomenon could reflect the emergent status of those coregulation forms in that period of infant life, it involves another source of variance, which is related to the timing of developmental events.

Further research is encouraged to confirm these findings. Although we examined an average of 41 observations for each dyad and the study lasted 14 months, the low number (10) of dyads involved in the study, owing to the difficulties of collecting data so frequently and over a long period of time, unfortunately reduced the power of the statistical test we used. Moreover, limiting data collection to only one instrument did not permit us to explore possible relationships between coregulation development and other measures, and confined the analysis of individual differences to the information provided by our coding system. A research design which includes other individual and environmental variables is needed to identify those factors which may account for variability in development. Despite these limitations, our findings are fine grained and quite consistent; so, they can be reliably taken as further evidence of the infant’s entry into the realm of secondary intersubjectivity as a gradual and multidetermined process.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References
  • Adamson, L. B., & Bakeman, R. (1985). Affect and attention: Infants observed with mothers and peers. Child Development, 56, 582593.
  • Adamson, L. B., Bakeman, R., & Deckner, D. F. (2004). The development of symbol-infused joint engagement. Child Development, 75, 11711187.
  • Ashley, J., & Tomasello, M. (1998). Cooperative problem-solving and teaching in preschoolers. Social Development, 7, 143163.
  • Aureli, T. (1994). Shared focus in mother–child joint activity in the second year of life. Early Development and Parenting, 3, 145152.
  • Bakeman, R., & Adamson, L. (1984). Coordinating attention to people and objects in mother–infant and peer–infant interaction. Child Development, 55, 12781289.
  • Bakeman, R., & Gottman, J. M. (1986). Observing interaction: An introduction to sequential analysis. New York: Cambridge University Press.
  • Bates, E., Benigni, K., Bretherton, I., Camaioni, L., & Volterra, V. (1979). The emergence of symbols: Cognition and communication in infancy. New York: Academic Press.
  • Bauer, D. J., & Curran, P. J. (2005). Probing interactions in fixed and multilevel regression: Inferential and graphical techniques. Multivariate Behavioral Research, 40, 373400.
  • Behne, T., Carpenter, M., Gräfenhain, M., Liebal, K., Liszkowsky, U., Moll, H., . . . Wyman, E. (2008). Cultural learning and cultural creation. In U.Müller, J.Carpendale, N.Budwig, & B.Sokol (Eds.), Social life and social knowledge (pp. 65102). New York: Erlbaum.
  • Bibok, M. B., Carpendale, J. I. M., & Lewis, C. (2008). Social knowledge as social skill. In U.Müller, J.Carpendale, N.Budwig, & B.Sokol (Eds.), Social life and social knowledge (pp. 145169). New York: Erlbaum.
  • Bigelow, A. E., Maclean, K., & Proctor, J. (2004). The role of joint attention in the development of infant’s play with object. Developmental Science, 7, 518526.
  • Bornstein, M. H. (1989). Maternal responsiveness: Characteristics and consequences. San Francisco, CA: Jossey-Bass.
  • Bornstein, M. H., & Haynes, O. M. (1998). Vocabulary competence in early childhood: Measurement, latent structures and predictive validity. Child Development, 69(3), 654671.
  • Brazelton, T. B., Koslowsky, B., & Main, M. (1974). The origins of reciprocity. In M.Lewis & R. L.Rosemblum (Eds.), The effect of the infant on the caregiver (pp. 4976). New York: Wiley.
  • Bretherthon, I. (1991). Intentional communication and the development of an understanding of mind. In D.Frye & C.Moore (Eds.), Children’s theories of mind: Mental states and social understanding (pp. 4975). Hillsdale, NJ: Erlbaum.
  • Brinck, I., & Gärdenfors, P. (2003). Co-operation and communication in apes and humans. Mind & Language, 18, 484501.
  • Bronwell, C. A., Nichols, S., & Svetlova, M. (2005). Early development of shared intentionality with peers. Behavioral and Brain Sciences, 28, 693694.
  • Brooks, R., & Meltzoff, A. (2005). The development of gaze following and its relations to language. Developmental Science, 8, 535543.
  • Brownell, C. A., & Carriger, M. S. (1990). Changes in cooperation and self-other differentiation during the second year. Child Development, 61, 11641174.
  • Brownell, C. A., & Carriger, M. S. (1991). Collaboration among toddler peers: Individual contribution to social contexts. In L. B.Resnick, J. M.Levine, & S. D.Teasley, (Eds.), Perspective on socially shared cognition (pp. 365383). Washington, DC: American Psychological Association.
  • Brownell, C. A., Ramani, G. B., & Zerwas, S. (2006). Becoming a social partner with peers: Cooperation and social understanding in one-and two-year-olds. Child Development, 77, 803821.
  • Bruner, J. (1982). The organization of action and the nature of adult–infant transactions. In E.Tronick (Ed.), Social interchange in infancy: Affect, cognition and communication (pp. 2336). Baltimore, MD: University Park Press.
  • Bruner, J. (1983). Child’s talk: Learning to use language. New York: Norton.
  • Bruner, J. S. (1975). Child’s talk: Learning to use language. New York: Norton.
  • Camaioni, L., & Aureli, T. (2002). Trajectories developpementales et individuelle e la transition vers la communciation symbolique. Enfance, 3, 259275.
  • Camaioni, L., Aureli, T., Bellagamba, F., & Fogel, A. (2003). A longitudinal examination of the transition to symbolic communication in the second year of life. Infant and Child Development, 12, 126.
  • Carpendale, J., & Lewis, C. (2006). How children develop social understanding. Oxford: Blackwell.
  • Carpenter, M., Nagell, K., & Tomasello, M. (1998). Social cognition, joint attention, and communicative competence from 9 to 15 months of age. Monographs of the Society for Research in Child Development, 63(4), Serial No. 255.
  • Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analyses for the behavioral sciences, 3rd ed. Mahwah, NJ: Lawerence Erlbaum Associates, Inc.
  • Conner, D., & Cross, D. (2003). Longitudinal analysis of the presence, efficacy and stability of maternal scaffolding during informal problem solving interactions. British Journal of Developmental Psychology, 21, 315334.
  • Eckerman, C. O. (1993). Toddlers’ achievement of coordination actions with conspecifics: A dynamic systems perspective. In L. B.Smith & E.Thelen (Eds.), A dynamic systems approach to development: Application MIT Press/Bradford books series in cognitive psychology (pp. 333357). Cambridge, MA: MIT press.
  • Eckerman, C. O., & Peterman, K. (2001). Peers and infant social/communicative development. In G.Bremner & A.Fogel (Eds.), Blackwell handbook of infant development. Handbooks of developmental psychology (pp. 326350). Malden, MA: Blackwell Publishers.
  • Eckerman, C. O., & Stein, M. R. (1990). How imitation begets imitation and toddler’s generation of games. Developmental Psychology, 26, 370378.
  • Flom, R., & Pick, A. D. (2005). Experimenter affective expressions and gaze following in 7-month-olds. Infancy, 7, 207218.
  • Fogel, A. (1990). The process of developmental change in infant communicative action: Using dynamic system theory to study individual ontogenies. In J.Colombo & J.Fagen (Eds.), Individual differences in infancy: Reliability, stability and prediction (pp. 341358). Hillsdale, NJ: Erlbaum.
  • Fogel, A. (1993). Developing through relationships. Chicago: University of Chicago Press.
  • Fogel, A. (2006). Dynamic systems research on interindividual communication: The transformation of meaning-making. The Journal of Developmental Processes, 1, 730.
  • Fogel, A., & Lyra, M. C. D. P. (1997). Dynamics of development in relationships. In F.Masterpasqu & P. A.Perna (Eds.), The psychological meaning of chaos: Translating theory into practice (pp. 7594). Washington, DC: American Psychological Association.
  • Goldman, B., & Ross, H. (1978). Social skills in actions: Analysis of early peer games. In J.Glick & K. A.Clarke-Stewart (Eds.), The development of social understanding (pp. 177212). New York: Gardner.
  • Goldstein, H. (1995). Multilevel statistical models. London: Edward Arnold. New York: Wiley.
  • Goldstein, H. (2003). Multilevel statistical models. Kendall’s Library of Statistics 3. London: Arnold.
  • Goldstein, H., Rasbash, J., Plewis, I., Draper, D., Browne, W. S., Yang, M., . . . Healy, M. (1998). A user’s guide to MLwiN. London: Institute of Education.
  • Hay, D. (1979). Cooperative interactions and sharing between very young children and their parents. Developmental Psychology, 15, 647653.
  • Hobson, P. (2002). The cradle of thought. Exploring the origins of thinking. London: Macmillan.
  • Hobson, P., & Hobson, J. (2008). In the beginning is relation…and then what? In U.Müller, J.Carpendale, N.Budwig, & B.Sokol (Eds.), Social life and social knowledge (pp. 103122). Hillsdale, NJ: Erlbaum.
  • Hsu, H.-C., & Fogel, A. (2001). Infant vocal development in a dynamic mother–infant communication system. Infancy, 2, 87109.
  • Huttenlocher, J., Haight, W., Bryk, A., Seltzer, M., & Lyons, T. (1991). Early vocabulary growth: Relation to language input and gender. Developmental Psychology, 27, 236248.
  • Kaye, K., & Fogel, A. (1980). The temporal structure of face-to-face communication between mothers and infants. Developmental Psychology, 16, 454464.
  • Kochanska, G., & Aksan, N. (2004). Development of mutual responsiveness between parents and their young children. Child Development, 75, 16571676.
  • Lavelli, M. (2005). Sviluppo dei primi processi di co-regolazione madre-lattante: la transizione evolutiva del secondo mese [The development of early mother–infant coregulation processes: The second month transition]. Età Evolutiva, 81, 5063.
  • Lavelli, M., & Fogel, A. (2002). Developmental changes in mother–infant face-to-face communication: Birth to 3 months. Developmental Psychology, 38, 288305.
  • Leslie, A., & Happe, F. (1989). Autism and ostensive communication: The relevance of meta-representation. Development and Psychopathology, 1, 205212.
  • Moore, C. (1996). Theories of mind in infancy. British Journal of Developmental Psychology, 14, 1940.
  • Moore, C., & Dunham, P. J. (1995). Joint attention: Its origins and development. Hillsdale, NJ: Erlbaum.
  • Müller, U., Carpendale, J., Budwig, N., & Sokol, B. (Eds.). (2008). Social life and social knowledge. New York: Erlbaum.
  • Mundy, P., Block, J., Delgado, C., Pomares, Y., Vaughan Van Hecke, A., & Parlade, M. V. (2007). Individual differences and the development of joint attention in infancy. Child Development, 78, 938954.
  • Mundy, P., Card, J., & Fox, N. (2000). EEG correlates of the development of infant joint attention skills. Developmental Psychobiology, 36, 325338.
  • Mundy, P., & Sigman, M. (2006). Joint attention, social competence and developmental psychopathology. In D.Cicchetti & D.Cohen (Eds.), Developmental Psychopathology, 2nd ed., Vol. 1: Theory and methods (pp. 293332). Hoboken, NJ: Wiley.
  • Murphy, C. M, & Messer, D. J. (1977). Mothers, infants and pointing: A study of a gesture. In 0H. R.Schaffer (Ed.): Studies in mother-infant interaction (pp. 386419). London: Academic Press.
  • Nelson, K. (1996). Language in cognitive development: Emergence of the mediated mind. Cambridge: Cambridge University Press.
  • Newson, J., & Newson, E. (1975). Intersubjectivity and the transmission of culture. Bulletin of the British Psychological Society, 28, 437445.
  • Parise, E., Reid, V., Stets, V., & Striano, T. (2007). Eye gaze direction influences object processing in 5-month-old infants. 13th European Conference on Developmental Psychology, August, Jena.
  • Plewis, I. (1996). Statistical methods for understanding cognitive growth: A review, a synthesis and an application. British Journal of Mathematical and Statistical Psychology, 49, 2542.
  • Racine, T. P., & Carpendale, J. (2007). The role of shared practice in joint attention. British Journal of Developmental Psychology, 25, 325.
  • Rasbash, J., Steele, F., Browne, W., & Prosser, B. (2005). A User’s Guide to MLwiN: Version 2.0. Bristol: Centre for Multilevel Modelling, University of Bristol.
  • Reddy, V. (2008). Experiencing others: A second person approach to other-awareness. In U.Müller, J.Carpendale, N.Budwig, & B.Sokol (Eds.), Social life and social knowledge (pp. 123144). New York: Erlbaum.
  • Reid, V., Striano, T., Kaufman, J., & Johnson, M. H. (2004). Eye gaze cueing facilitates neural processing of objects in 4-month-old infants. Neuroreport, 15, 25532555.
  • Ross, H., & Lollis, S. P. (1987). Communication within infant social games. Developmental Psychology, 23, 241248.
  • Slaughter, V., & McConnell, D. (2003). Emergence of joint attention: Relationships between gaze following, social referencing, imitation and naming in infancy. Journal of Genetic Psychology, 164, 5471.
  • Smiley, P. (2001). Intention understanding and partner-sensitive behaviors in young children’s peer interactions. Social Development, 10, 330354.
  • Smith, L., & Ulvund, L. (2003). The role of joint attention in later development among preterm children: Linkages between early and middle childhood. Social Development, 12, 222234.
  • Snijders, T., & Bosker, R. (1999). Multilevel analysis. An introduction to basic and advanced multilevel modeling. London: SAGE.
  • Stern, D. (1974). The goal and the structure of mother–infant play. Journal of the American Academy of Child Psychiatry, 13, 402421.
  • Striano, T., & Bertin, E. (2005). Social-cognitive skills between 5 and 10 months of age. British Journal of Developmental Psychology, 23, 559568.
  • Striano, T., & Rochat, P. (1999). Developmental link between the dyadic and triadic social competence in infancy. British Journal of Developmental Psychology, 17, 551562.
  • Striano, T., & Stahl, D. (2005). Sensitivity to triadic interaction in early infancy. Developmental Science, 8, 333343.
  • Striano, T., Stahl, D., & Cleveland, A. (2009). Taking a closer look at social and cognitive skills: A weekly longitudinal assessment between 7 and 10 months of age. European Journal of Developmental Psychology, 6(5), 567591.
  • Thelen, E., & Smith, L. B. (1994). A dynamic systems approach to the development of cognition and actions. Cambridge, MA: MIT Press.
  • Tomasello, M. (1995a). Joint attention as social cognition. In C.Moore & P.Dunham (Eds.), Joint attention: Its origins and role in development (pp. 103130). Hillsdale, NJ: Erlbaum.
  • Tomasello, M. (1995b). The cultural origins of human cognition. Cambridge, MA: Harvard University Press.
  • Tomasello, M. (1999). The cultural origins of human cognition. Cambridge: Harvard University Press.
  • Tomasello, M., Carpenter, M., Call, J., Behne, T., & Moll, H. (2005). Understanding and sharing intentions: The origins of cultural cognition. Behavioral and Brain Sciences, 28, 675691.
  • Trevarthen, C., & Hubley, P. (1978). Secondary intersubjectivity: Confidence, confiding and acts of meaning in the first year. In A.Lock (Ed.), Action, gesture, and symbol: The emergence of language (pp. 4975). New York: Academic.
  • Tronick, E. Z., Als, H., & Adamson, L. B. (1979). Structure of early face-to-face communicative interactions. In M.Bullowa (Ed.), Before speech: The beginning of interpersonal communication (pp. 349372). Cambridge: Cambridge University Press.
  • Warnecken, F., Chen, F., & Tomasello, M. (2006). Cooperative activities in young children and chimpanzees. Child Development, 77, 640663.
  • Warnecken, F., & Tomasello, M. (2007). Helping and cooperation at 14 months of age. Infancy, 11, 271294.