Developmental transition to secondary intersubjectivity

The Developmental Transition to Secondary Intersubjectivity in the Second Half Year: A Microgenetic Case Study Alan Fogel Department of Psychology University of Utah alan.fogel@psych.utah.edu. Ilse DeKoeyer-Laros Department of Psychology University of Utah Abstract: To explore the relational-historical processes by which infants may develop an intersubjective sense of self in a relational context, one mother-infant dyad was observed weekly across the 9-month developmental transition, using a microgenetic research design. Qualitative research methods were used to study the developmental changes in the dynamic system of the mother-infant relationship. We discovered four developmental periods that marked qualitative differences in the patterns of communication between mother and infant and the infant’s emerging self-awareness. In the first developmental period, the mother’s attunements to the infant temporarily stabilize the infant’s attention to her own action. The ability to self-attune, crucial to emerging self and other-awareness, becomes a dynamically stable “attractor” by the second period, in which the infant attends consistently to her own and her mother’s behavior. This leads to the emergence of a stable communication frame for pounding in the third period, involving increasing temporal contiguity of self and other actions, brief glimpses of secondary intersubjectivity during the pounding frame, and changes in the mother’s behavior (such as emphasizing the word “you”). These changes lead to the consolidation and spread of secondary intersubjectivity to other frames in the mother-infant relationships (as reported by the mother’s diary) in the fourth developmental period. Introduction Some theories of the development of self-awareness in childhood have emphasized representational, reflective types of self-knowledge (Harter, 1999; Pipp, 1993). Most developmental psychologists agree that toddlers around 18 months of age show 63 64 ALAN FOGEL AND ILSE DEKOEYER-LAROS evidence of reflective self-awareness when they begin to recognize themselves in a mirror and separate the “I-self” from the “me-self” (Harter, 1999; Lewis, 1995). Research on infants prior to mirror self-recognition, however, have revealed that prerepresentational forms of self-awareness can be found in the familiar sensations and movements of the body, the experience of emotions, and the awareness of others’ emotions and intentions (Carpenter, Nagell, & Tomasello, 1998; Gibson, 1993; Neisser, 1993; Rochat, 1995; Stern, 1985; Tomasello, 1993; Trevarthen, 1993). In this paper, we are interested in examining the developmental transition in self- and other-awareness that occurs around the age of 9 months. This developmental transition has been described in many different ways in the infancy literature. Some focus on infants’ developing skills to coordinate joint attention, the onset of deictic gestures such as pointing, and the awareness that the other person has the intention to look at or refer to the same object to which their own attention is directed (Carpenter et al. 1998; Gustafson, Green, & West, 1979; Tomasello, 1993). Others focus on the change in the infant’s awareness of the feelings and intentions that are shared in the relationship between the parent and infant. Trevarthen calls this new developmental level “secondary intersubjectivity” (Trevarthen, 1993; 1998; Trevarthen & Hubley, 1978), defined as coordinating and sharing with another person one’s attention, feelings and intentions toward a third pole of an object, event, or action. Secondary intersubjectivity implies that the infant is aware of co-affectivity and co-agency with another person in relation to something else. Similarly, Stern (1985) proposes the emergence of the “intersubjective self” around 9 months of age in which the infant is thought to notice that others are affected by her agency just as she is affected by the agency of others. Secondary intersubjectivity develops after “primary intersubjectivity” (2–9 months), in which the infant is aware of moving and feeling in relation to another person, a “resonance” that is felt during interaction and that feels different to the infant than when alone (Trevarthen, 1993). A similar perspective is taken by Stern (1993) who describes the “core self” (2–9 months) in terms of emotions such that it is “not only the feeling experienced but also the experience of interpersonal evocation or regulation or sharing” (p. 205). Qualitative clinical case study observations of parent-child interactions suggest that infants in the first half year whose parents mirror the their emotions and attune to them are more likely by the end of their first year to show an awareness of and emotional responsiveness to others, and to have a secure attachment relationship with their parents (Beebe & Lachmann, 2002; Fogel, 2001; Stern, 1985; Winnicott, 1971, 1990). Quantitative research on larger samples has shown that parental responsiveness, warmth, and emotional openness are linked to more adaptive child functioning, including secure attachment, social competence, and self-esteem (Ainsworth, Behar, Waters, & Wall, 1978; Baumrind, 1989; Black & Logan, 1995; Coopersmith, 1967; De Wolff & Van IJzendoorn, 1997; Feshbach, 1987; Koren-Karie, Oppenheim, Dolev, Sher, & Etzion-Carasso, 2002; Maccoby, 1992; Oppenheim & Koren-Karie, 2002; Oppenheim, Koren-Karie, & Sagi, 2001). In addition, normally developing mothers and infants both change their behavior in relation to each other in coordinated ways over time, suggesting that the dyad develops as a shared affective and communicative system (Beebe & Lachmann, 2002; Hsu & Fogel, 2001; Lavelli & Fogel, 2005). From the late fetal period, infants are aware of at least some of their bodily states, movements, and senses (Damasio, 1999; Fifer, Monk, & Grose-Fifer, 2001). Multiple, The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 65 repeated experiences of social interaction form experience-dependent neuromotor pathways of how to do things with intimate others, called “implicit relational knowing” (Beebe, 1998; Lyons-Ruth, 1998; Panksepp, 2001; Schore, 2001; Stern, 1998). The right limbic system and pre-frontal cortex appear to be especially attuned to implicit relational knowing which is manifested in continuing, largely automatic, patterns of doing things with others (Shonkoff, & Phillips, 2000; Schore, 2001; Siegel, 2001). The neurological traces of these repeated interpersonal and emotional routines may be the foundation of “internal working models” of attachment (Bowlby, 1973; Bretherton, 1990, 1991, 1993). Intersubjectivity can also be understood from a dynamic systems perspective because the awareness of the self goes beyond the skin boundaries of the individual to include the “self-regulating other” (e.g., Beebe & Lachman, 2002; Stern, 1985). Pointing, for example, implies the awareness of another person who shares the same focus of attention. From the infant’s perspective at the end of the first year, the personal intention toward an object, the gesture (pointing), the referent (an object), and the other person appear to be part of a single system. Social referencing, to take another example of an infant’s awareness at the end of the first year, implies a personal intention to regulate emotion that is organized together with directing attention to another’s emotional expression, apparently suggesting that the infant’s emotion regulation is organized into a dyadic system. There are many experimental studies showing a developmental transition at 9 months in self-awareness, emotional communication, attachment behavior, and cognition (Behne, Carpenter, & Tomasello, 2005; Camaioni, Perucchini, Bellagamba, & Colonessi, 2004; Legerstee, 2005; Phillips, Wellman, & Spelke, 2002; Sommerville & Woodward, 2005; Striano & Rochat, 1999; Tomasello, 1993). These studies are fundamental in establishing, via experimental manipulation, that particular forms of infant behavior indeed reveal developmental changes in self- and other-awareness. On the other hand, however, they do not reveal anything about the process of change by which new levels of self-awareness emerge. Theoretical explanations refer to shifts in cognitive understanding of self and others or to neurological changes or to changes in social communication, but little is known about how these developments occur in individual children in the context of their relationships with parents. In order to study the change process from a dynamic systems perspective, two strategies are required. First, one must examine multiple elements of the system as they change together over time. As Thelen’s work on motor development from a dynamic systems perspective reveals, recurring linkages form across neurological, sensory, motor, and contextual domains. These so-called “attractors”—such as kicking, crawling, and walking—are revealed to observers as recurring sequences and cooccurrences of coordinated action in relation to particular contexts (Thelen, Fisher, & Ridley-Johnson, 1984; Thelen & Ulrich, 1991). Along the same lines, the infant’s awareness of self and others can be viewed as a dynamic psychosocial system that partakes of the relationship between the infant and significant others (Fogel, 1993, 1995, 2001; Fogel & Thelen, 1987; Garvey & Fogel, in press; Granic, 2000; Shanker & King, 2002). These attractors are called “frames” (cf. “formats,” Bruner, 1983): regularly recurring patterns of communication that involve coordinated action and shared cognition and emotion (Fogel, 1993; Fogel, Garvey, Hsu, & West-Stroming, 2006). 66 ALAN FOGEL AND ILSE DEKOEYER-LAROS Particular parent-infant games, like peek-a-boo or tickle, are frames in the relationship system, as are book reading, mealtimes, bedtime rituals, and the like. The second dynamic systems research strategy is to observe change as it is occurring, not simply before and after the change. One of the central methodological principles of dynamic systems as applied to the study of developmental change is the microgenetic research design (Flynn & Siegler, 2007; Fogel, 1990; Lavelli, Pantoja, Hsu, Messinger, & Fogel, 2005; Siegler & Crowley, 1991; Thelen, 1990). Individual cases are observed frequently across a key period of developmental change. This gives a picture of the change process by observing before, during, and after with a sufficient density of observations to reveal how the different components and emergent attractors of the dynamic system become organized and re-organized as a function of time (Fogel et al., 2006). This is a very different approach to developmental research compared to cross-sectional experimental designs or to large-N longitudinal designs in which presumed factors contributing to change are assessed via patterns of shared variance across subjects. The latter uses statistical relationships between variables, while dynamic systems research focuses on real time sequential and co-occurring relationships between observed actions and on historical changes over developmental time within the same case. Microgenetic designs have been mostly applied to the study of cognitive development (Piaget, 1952; Siegler & Crowley, 1991). There are, by contrast, only a few published microgenetic studies on the development of self-awareness. Trevarthen’s work, cited above, is one example. Another study examines the developmental transition that occurs after 18 months using a discourse analysis of monthly videotapes of mealtimes with one child and her family (Forrester, 2001). Findings of this study show how the parent-child discourse shifts in a way that leads to a growing awareness of the child’s self-agency and dialogical positioning vis-à-vis the parents. It also shows clearly how the self develops in relational frames. A similar approach is taken by Garvey and Fogel (in press) on emotional development between 2 and 6 months. The purpose of the present paper is to describe the developmental changes in the communication system of one mother-infant dyad across the 9 month developmental transition, i.e., the transition between primary and secondary intersubjectivity. Our focus is on the process of change in secondary intersubjectivity, a topic which has not been previously investigated. We selected a single representative frame, a spontaneous game of pounding the table of a high chair in which the infant is sitting, and describe the ways in which actions and emotions bear on the infant’s development of secondary intersubjectivity with respect to her mother. Method Participants The dyad we selected for this case study was part of a group of 13 mother-infant dyads that were videotaped weekly over the infants’ first year of life and bi-weekly in The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 67 their second year. Between 6 and 9 months, the infants were seated in a high chair with their mothers opposite them. “Susan” is the youngest of three daughters in a Caucasian, middle-class family from a Midwestern U.S. city. So far as we could tell from our videos, Susan was developing normally. Although the transition to secondary intersubjectivity occurred in all of our dyads, we chose Susan and her mother for this investigation primarily because communication patterns specific to intersubjectivity occurred frequently on their videos, and more often than in the other dyads. In the kinds of free-play observations recorded here, there is a balance between letting each dyad develop in its own way and providing sufficient control to capture phenomena of interest. Because this one dyad’s data had a primacy of content related to the subject of this study, we cannot be certain that the change process in intersubjectivity was similar for the other dyads during times when our cameras were not on them. Procedures Video recordings. At the dyad’s first visit to the laboratory, Susan was 4 weeks old. At this time, the mother “Sheryl” was asked to play with her infant in any way she wanted. In subsequent sessions, no further instructions were given in order to leave the dyad free rein to create their own relationship in the context of the laboratory setting. At 27 weeks of age, Susan was moved into a high chair and at 42 weeks, the mother and infant started playing at a low table. With Susan in the high chair, the dyad played face-to-face without toys, although two toys were available (which could be used, for instance, if the infant became fussy). For this study, we used only the high chair sessions between 27 and 40 weeks of age. The carpeted laboratory playroom (3.81 m × 3.81 m) was equipped with two cameras, operated from the room next door. Videotapes utilized in the present study show split-screen recordings of the two cameras operated from the observation room. Mother’s and infant’s faces and upper bodies were visible as much as possible. A time code giving elapsed minutes and seconds was superimposed on the screen. Susan and Sheryl made 11 visits to the laboratory with Susan between 27 and 40 weeks of age and their interactions were video recorded for 1 hour and 55 minutes total (mean duration = 10 minutes, 30 seconds range 9:03– 11:17). Visits at 31, 38, and 39 weeks are missing because the dyad could not attend and the one at 41 weeks is missing because of technical difficulties. Journal. In addition to being videotaped every week, Sheryl was asked to make daily or weekly journal entries, noting any changes in the infant’s social, emotional, and physical development. The journal instructions did not mention intersubjectivity as a focus, again, in order to allow the mother to use her own constructions of what she considered important. For this study, entries were used that started at 26 weeks and ended at 40 weeks. Sheryl made 38 near-daily entries between 26 and 34 weeks. Starting at 34 weeks and ending at 40 weeks, she made seven weekly entries. The second author read and reread these 45 journal entries, focusing on any entries that were deemed to be observations about the infant’s emerging self-awareness and pounding behavior. All of these entries are given verbatim in this paper. 68 ALAN FOGEL AND ILSE DEKOEYER-LAROS Qualitative Analyses and Data Selection Qualitative research is particularly well-suited to the study of relationship change because communication systems are regulated by the way people make meaning of each other’s actions rather than by the physical parameters of those actions (such as velocity and force). The intersubjective meaning of communicative actions can be thought of as co-regulated within the system of communication in the same way as any other pattern formation process studied from a dynamic systems perspective. Qualitative research differs from traditional quantitative approaches by taking the observer’s perspective explicitly into account (Patton, 1990). This fits well with the dynamic systems perspective if we generalize the system under investigation to include subject, environment, and observer (Fogel, 2006). Many dynamic systems have been modeled as iterative processes in which innovation and variability transform one dynamically stable set of attractors into another (van Geert, 1994; 1998). Multiple iterative passes through videotapes and narrative transcriptions are at the heart of qualitative data analysis and interpretation. When one studies developmental change qualitatively, therefore, the iterative process of communication sessions recorded over time becomes embedded within the iterative process of data analysis and interpretation (see Figure 1). This ethnographic approach becomes scientific to the extent that we make explicit the process of change in the observer-observed system, and also provide sufficient detail from the transcriptions for the reader to follow the discourse. For that reason, we provide narrative and photographic evidence in our results. More details about qualitative dynamic systems research can be found elsewhere (Fogel, 2006; Fogel et al., 2006). The complete narrative descriptions of this study are available by contacting the first author. First Phase. Because the focus of this study was awareness of self and others related to intersubjectivity, we decided to focus our observations on moments in which the infant paid visual, auditory, or tactile attention to parts of her own body, and/or to her own actions or to her mother’s actions. Initial observations of the videos of the selected sessions in this dyad suggested that such moments most often occurred in the context of a frame that eventually developed between the mother and the infant, in which one partner would hit the tray of the high chair and the other would respond (“the pounding game”). In this phase, we as observers are letting the form of the existing data guide the selection of segments to be observed. Because of the timeintensive commitment required by qualitative analysis, observers need to be highly selective in order to focus their efforts on those segments of data most relevant to the research questions, in our case, the process of change in intersubjectivity. Second Phase. All sessions were viewed to identify (by time code) all events in which the infant hit the tray of the high chair, however slightly, or one of the toys. All instances in which the mother hit or tapped the table with her hands or fingertips were also selected because such maternal behaviors were often part of dyadic pounding events (with the mother tapping and the infant pounding). These events were then captured in digital format utilizing video-editing software (Adobe Premiere 5.1; see Secrist, de Koeyer, Bell, & Fogel, 2002). In order to capture the context of the events, video clips were captured starting 3 seconds before the onset of the pounding-related The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 69 Developmental time for participants Real time sessions for participants Initial impression Emergent Interpretive frame Developmental time for research process Real time observation and interpretation sessions FIGURE 1. The Dynamics of the Observer-World System in Qualitative Research on Developmental Change. The prospectively recorded material (narratives, videos) consists of repeated real-time observation sessions across a developmental transition (the participant’s developmental change). The timeline of this material is nested within the timeline of the research process in which the recorded material is repeatedly observed and interpreted, which leads to a developmental change in the interpretation of the observations (a developmental change in the conceptualization of the researchers about the recorded material). The results of the study are emergent from this dynamic iterative process of engagement and re-engagement with the recorded material. behavior and ending 3 seconds after its offset. When the infant pounded repeatedly, with only seconds between the poundings, this was captured as a single episode. We included behavior indicating self-awareness related to the pounding in the captured events, such as when the infant pounded and then looked intently at her hands, lying still on the table. Also, we considered it part of the pounding event if Sheryl’s tapping evolved into a tickling game. Such events were captured until the mother stopped tickling. The captured events containing pounding or tapping and related behaviors occurred for 41% of the total elapsed time in the selected high chair sessions (total time 44 minutes, 31 seconds). The mean duration per week of the captured events was 4:03 minutes (range 1:33 minutes, occurring during week 32, to 5:54 minutes, occurring during week 34). Infant pounding events accounted for 67% of the total captured event duration (29 minutes, 54 seconds), and maternal tapping and tickling games accounted for the remaining 33% (14 minutes, 37 seconds). After all the events were captured, they were compiled and edited in sequential order to form a “developmental movie” of pounding between 27 to 40 weeks. To 70 ALAN FOGEL AND ILSE DEKOEYER-LAROS ensure that the movie contained all behaviors relevant to pounding, they were repeatedly checked against the original videotapes in an iterative process. If necessary, the movie was edited by adding or deleting captured video material. The developmental movie is a crucial step in this type of change process research. First of all, it collects everything of interest in one relatively short documentary film. Second, it makes it extremely easy to view and review the movie in an iterative way in order to assess what changes and what remains the same from one session to the next. Third Phase. After this, the developmental movie was viewed repeatedly, both within segments and for the entire movie. For a more in-depth view of what occurred, detailed descriptive narratives were also written, in an iterative process called the constant comparative method in which narratives are re-written until nothing new can be added (Patton, 1990; Strauss & Corbin, 1990). All socio-emotional events surrounding the pounding activities were described (i.e., gaze direction, emotional expressions, vocalizations, intensity, and rhythm patterns). The developmental movie and narratives were then reviewed in an iterative process to compare the weekly sessions with regard to the infant’s attention to the pounding, her emotional expressions, and dyadic interaction patterns around the pounding behaviors. Fourth Phase. After reviewing all infant pounding and maternal tapping or tickling events, a second developmental movie was created, including only those moments in which the infant focused on the pounding (i.e., by glancing or gazing at her pounding hands, repeating the pounding actions, by looking at mother after mother imitated her pounding actions, etc.). This was done because we concluded from our repeated observations that these were the most relevant to intersubjectivity. We also found that pounding events in which the infant was not focused on the pounding, or in which the infant was pounding toward a toy, did not clearly contribute to the emergence of intersubjectivity (from our observations in the earlier phases). Fifth Phase. Finally, a third developmental movie was created that contained only those events in which mother and infant paid attention to the infant’s pounding actions, attention being the key behavioral component of self-awareness. This final movie was 15 minutes and 56 seconds in length. It was reviewed repeatedly in an iterative process to clarify the processes by which intersubjectivity within this dyadic pounding frame eventually developed. Each of these research phases required months of work and the final project lasted several years. In the end, the results of this work emerged from this data analysis process (Figure 1) using the constant comparative method. Results, therefore, are not final or deterministic, but rather reflect the point at which no new interpretations emerged from our personal engagement with these data. Reliability and Validity In qualitative research, the traditional forms of reliability and validity are not useful indices of the quality of the research. Qualitative research assumes that any description of reality is colored by one’s perspective—in our case, a long-term engagement with the data—and that observers cannot be independent because of the research The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 71 process that requires personal and long-term engagement with the data. From this perspective, the results are importantly and necessarily emergent from the observer-world system. Training an independent observer is typically not desirable because of the need for long-term engagement with the data, a research process that is similar to anthropological field work. Qualitative research is assessed according to the credibility of the observers. Credibility has three criteria: prolonged engagement of the researcher with the data, the observers’ cumulative experience doing similar investigations, and making the data available for readers to inspect (Denzin & Lincoln, 1994; Rogoff, Mistry, Gioncu, & Mosier, 1993; Savage-Rumbaugh & Fields, 2000). Credibility is typically the way in which people assess the trustworthiness of professionals such as physicians, attorneys, or psychologists. It is also used, at least implicitly, in other case-based approaches, such as in the evaluation of clinical case reports, anthropological field work, and judgments in legal cases. Our research methods, described above, testify to our prolonged engagement with these data. The authors of this paper worked as a peer research team with the second author as the primary observer. Between the two authors, we have over 40 years of experience observing parent-infant communication and its development. Finally, although we cannot share the original videos because of confidentiality concerns, we do share most of our narratives and many still photographs in the results. Even researchers with fewer years of experience can be considered credible assuming that they meet the other two criteria listed here. Validity in population-based research depends in part on the sampling procedure, the size of the sample, and the confidence with which a measure is thought to reflect stable latent characteristics. Our final developmental movie, from a single dyad, is only 15 minutes and 56 seconds in duration. While this may seem scant information on which to base scientific conclusions, this needs to be placed in the context of the qualitative and microgenetic methods used here. First of all, these data are developmental: not a single observation of 15 minutes but rather a string of observations that capture a change process. Second, our lengthy sampling procedure testifies to that fact that this movie was culled from an ongoing process of communication in this dyad. The dyad, in other words, was not merely observed for a few minutes each week. The larger communicative context is essential for the emergence of the events captured in the movie and for their spontaneity. In the end, however, the final developmental movie can be considered a type of salient anecdote because of its brevity and because it is from a single dyad. How is such data credible scientifically? One must consider the problem of how an investigator might go about capturing the key events in any developmental change process. Ideally, one would live in a household and become a participant observer. Even with this amount of observation, it would still be up to the trained observer to identify key observations that illustrate most clearly a salient event or series of events. In our case, we felt fortunate that we had indeed captured a sufficient number of key events to piece together a reliable portrait of intersubjective change in at least this one dyad from our larger sample. Piaget is known to have said that “an acute observation is worth a thousand statistics” (quoted by Lock, 1992, p. 500). According to Lock (1992), 72 ALAN FOGEL AND ILSE DEKOEYER-LAROS . . . what counts as an anecdote is not merely observed, nor stumbled upon, but actively selected by an observer, filtered out of on-going reality as a result of perception intuitively pre-informed, in a way we have yet to fully elucidate. . .” (p. 501) In this paper, we have endeavored to elucidate this process, and to provide an explicit account of its potential importance to scientific discovery in the context of doing qualitative, microgenetic research. Like any research, it awaits replication and confirmation with larger samples. Results Based on the qualitative research process described above, we were able to distinguish four qualitatively different periods in the development of the infant’s attention to her pounding actions. In the first period, from 26 to 32 weeks, the pounding game was just emergent. Susan did not seem aware that she was the agent of the pounding but she would look up immediately when the mother mirrored or attuned to her actions. In the second period, 33 and 34 weeks, Susan began to attend to her own pounding and her mother again attuned to the pounding movements. During the third period, between 35 and 37 weeks, the dyad established a mutually understood game of pounding, a stable communicative frame, which Sheryl called “the slap game.” During the final period, between 38 and 40 weeks, the infant appeared to become aware of her agency in this game, and her abilities to initiate, accept, and refuse to play the game, all signs of an emergent secondary intersubjectivity. We observed action as it spontaneously unfolded over time. We rely on the support of existing experimental studies to interpret particular forms of infant behavior as indicative of self- and intersubjective-awareness. In the following descriptions, we use boldface to indicate the times when Susan appears to show intersubjective self-awareness by co-regulating her behavior with respect to her mother, as when Susan smiles and looks at her mother following her own or her mother’s pounding. Italics are used to indicate when Susan shows awareness of her own actions, typically by using some type of crossmodal self-reference, as when looking at her own hands while pounding. The First Developmental Period, The Emergence of the Pounding Behavior: 26 to 32 Weeks When Susan was 26 weeks old, Sheryl wrote in her journal about Susan’s behavior at home, “As I was holding my hand out in front of Susan, palm up, she brought her hand down on top of mine – she ‘slapped five.’ Actually, she is pretty ‘slappy’ – any hard surface in front of her feels the wrath of her hands.” Several instances of such pounding behaviors were observed in the first two videotaped high chair sessions. For example, during the first minute of the first high chair observation at 27 weeks, The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 73 Susan starts to pound energetically with her left hand, gazing to her right. She is not particularly attentive to the pounding until Sheryl starts to vocalize in the same rhythm, “Oom! Boom, boom!” Immediately, Susan looks into Sheryl’s eyes and pauses, after which Sheryl repeats “Oom!” Mother and infant smile and Susan’s jaw slightly drops. In this example, the first part of which is shown in Figure 2, Susan’s pattern of attention suggests that she does not seem aware of her abilities to intentionally initiate or modify the pounding action, or of its effects on Sheryl. Based on Susan’s gazing at her mother in response to Sheryl’s attunements (primary intersubjectivity), however, it appears as though the mother’s cross-modal attunement (vocalizing in the same rhythm and intensity as the infant’s pounding; cf. Stern, 1985) amplifies the infant’s attention to the pounding actions in line with Stern’s ideas about the way in which infants’ sense of their own actions become regulated by their mothers. FIGURE 2. The first developmental period: The emergence of pounding behavior (26 to 32 weeks). In this sequence taken from the videos, Susan looks away to her right while pounding the table of the high chair. (Note: the on-screen timer was not working for this segment; view pictures in the sequence upper left, upper right, lower left, and lower right). 74 ALAN FOGEL AND ILSE DEKOEYER-LAROS When Sheryl vocalizes along with her own tapping motions, the infant shows a different response. At 28 weeks, we observed a sequence in which both Susan’s and Sheryl’s hands are involved in pounding or tapping, while mother vocalized in tune with her own actions. Susan pounds a few times, looking at her hand. She forcefully pounds once while gazing away, then pounds again, gazing at her hand as it lands. Sheryl exclaims, “Boom!” Susan looks at her mother, drops her jaw, and pounds again. This time, as Susan’s hands hit the table tray from the top, Sheryl’s hit it from the bottom! Sheryl repeats this tap from the bottom, accompanied by “Boom!” Now, Susan’s hand wavers in the air for an instant and Sheryl repeats again. For 2 seconds, Susan stares at her mother with an open mouth, and then tries to look under the tray. Sheryl hits faster, saying “Boom! Boom!” and Susan pounds twice, as if in response to her mother’s vocalizations. Sheryl repeats Susan’s rhythm by hitting under the table twice. Susan looks up at her. They smile. Susan pounds twice, gazing at her hand, while Sheryl taps “Boom–Boom” under the table. The infant drops her jaw. A possible interpretation, following Stern (1985), is that the infant notices the auditory similarity between her own and her mother’s pounding but only feels proprioceptive sensations with her own actions (and not with her mother’s). Sheryl’s tapping, which has a similar rhythm and intensity as Susan’s pounding, is not authored by the infant and Susan seems to be curious about the ownership of these actions (pausing, experimenting with making sounds herself, and listening to her mother’s). Between 29 and 32 weeks, Susan starts to shift her gaze between Sheryl’s face, her own pounding hands, and back. At 32 weeks, for example, she watches her pounding hands with great attention before she shifts her gaze between her own hands and her mother’s face. Attention to her pounding seems related to Susan’s physiological state. Throughout all observations, Susan rarely pounds during moments of fussiness. At 30 weeks, she is particularly fussy and she rarely pounds or looks at her hands, although she is still attending to mother’s tapping hands. Susan’s attention to her pounding movements also seems to be related to her mother’s attention to the pounding. From 29 to 32 weeks, Sheryl is mainly a commentator, talking about other things and not commenting on the pounding. She does not explicitly focus on the pounding and rarely co-orients her movements or vocalizations with the infant’s pounding movements. Without Sheryl’s amplifications, Susan seems less likely to pay attention to her own pounding actions, again consistent with Stern’s hypothesis. At the beginning of the session at 29 weeks, for example, Susan pounds for a long time, occasionally glancing at her hands but mostly gazing away. When Sheryl does participate, it is more likely that Susan pays attention to her pounding actions. At 29 weeks, Sheryl vocalizes in the rhythm of the infant’s pounding movement only once, and Susan immediately looks up at her mother. At 32 weeks, Sheryl mimics the pounding movements with her voice and again Susan immediately looks up at her. When Sheryl repeats this, they smile. Similar to the earlier observations, Sheryl appears to provide a mirror for Susan’s actions, reflecting the contours of the pounding actions (up–down, up–down). One day after this observation, Sheryl writes in her journal, The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 75 While on mommy’s bed, Susan recognized her shadow on the wall. She seemed to catch something was “moving” on the wall and looked intently, probably trying to figure out what (or who) it was. She appeared to realize that the object moved with her. In summary, for this first developmental period, Susan’s ability to track her own actions is rudimentary. She seems to require her mother’s close attunements in order to begin to sense her own activity and its effects, as if the mother’s attunements amplify the feedback into her own sense of self. The example of the shadow suggests that Susan’s awareness of her own agency is becoming more stable by the end of this period. The Second Developmental Period, Attentive Pounding: 33 to 34 Weeks At 33 and 34 weeks, the various components of Susan’s pounding—mainly her hand and eye movements—appear to become more strongly coordinated as a stable system of cross-modal self-awareness. She shows more visual fascination for her pounding hands than during any of the earlier sessions. She clearly stretches out her arms and uses flat hands as she pounds, and she looks at her own hands at length and intently (see Figure 3). She also vocalizes during pounding, a possible self-attunement that may amplify her attention to the action. Finally, she varies the force and height of the movements while looking at her hands and she alternates pounding with her left and right hand. The following observation is from the outset of the session at 33 weeks: Susan starts to pound with her right hand, gazing at it attentively. Sheryl watches her, smiling and talking about other things. Susan rests her right hand upon her left and continues to look at her hands. She looks up at Sheryl and vocalizes, “Hm.” With a smile, Sheryl says, “Hi,” warmly acknowledging her daughter. They share a back-and-forth of smiles and vocalizations, while Susan continues to rest her hands on the table with her fingers touching and exploring each other (for a total of 21 seconds). Then, she watches them intently for 5 seconds, as if entranced with them. Another observation during this session was the alternation of markedly different feeling contours (Stern, 1985, 1993). Quiet moments suddenly transition into episodes of fast and intense pounding, rising with force, then suddenly dropping, and rising again. These different emotional moments are connected by a quality of visual concentration. It is likely that there is a growing awareness of self-affectivity, a possible linking for Susan of herself as the agent of these different feeling states. Although she appears to be self-focused during her moments of stillness (gazing at her own hands), Susan gazes intently at Sheryl during pounding moments. From her side of the dyad, Sheryl attunes to her daughter’s pounding as soon as she becomes aware of the intensity of the infant’s self-focus. For example, after the quiet period described above, Susan started to pound so energetically that Sheryl immediately interrupted her own talking and said, “Oooo, what a lot of noise!” Again, Susan pounded 76 ALAN FOGEL AND ILSE DEKOEYER-LAROS FIGURE 3. The second developmental period: Attentive pounding (33 to 34 weeks). In this sequence, Susan is pounding her own stationary hand while watching the stationary hand. This the first evidence of coordination between gazing, pounding, and other parts of her own body. vigorously, looking Sheryl straight into the eyes, her whole body engaged in the action. As Sheryl said, “Oooo. Bang! Bang! Bang!” moving her head up and down in the same rhythm, Susan’s jaw dropped and she faintly smiled. After this segment, the feeling contour again changed dramatically when Susan stopped pounding and watched her hands for 12 seconds with a quality of intense concentration while Sheryl was quietly observing. After this serene break, the feeling contour changed yet again when Susan suddenly started to pound vigorously, alternating left and right arms, watching her hands. Next, she started to gaze at Sheryl, who then began to vocalize, matching the rhythm and intensity of the infant’s pounding. Following this, Susan forcefully threw her body back in the chair, smiling. Seeing this, Sheryl laughed and said, in a similar outburst of energy, “Wow! You make a lot of noise.” In summary, for the second developmental period, our impression is that Susan seems to be beginning to sense her own agency, suggested by her increased attention to her own actions and her noticing (via gaze and smiling) the attunement between her pounding actions and her mother’s pounding and vocalizing. One can see the outlines The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 77 of an emerging dyadic pattern in which the possibility exists for Susan to eventually link an awareness of her own intentions with an awareness of her mother’s intentions. This is reminiscent of Bruner’s (1983) idea of communication formats as “place holders” for future developmental additions and also to the dynamic systems ideas of multiple components developing at different rates as they self-organize into an attractor. Susan’s attention to both herself and to her mother is clearly a change from the first developmental period in which she primarily attuned with her mother and seemed relatively unaware of her own pounding. The mother’s attunement to Susan, however, continues to scaffold Susan’s attention. In the second developmental period, we can see how Susan’s own contrasting emotions seem to create pathways in her awareness that link different activities together. The Third Developmental Period, Emergence of a Stable Pounding Frame: 35 to 37 Weeks At 35 weeks, Sheryl indicates in her journal that the pounding (or “slapping”) has now become a game between them. She writes, “Mommy stumbled onto the ‘slap game’ with Susan and she plays with such fervor. I’ll slap the table, or my knee, with my hand. And she’ll slap the same with her hands. This will go on for 5–7 rounds, where Susan will put her whole body into the slap, staring directly at mommy the whole time.” Thus, the pounding actions seem to have evolved into a dyadically shared game or “frame” (Fogel, 1993), with each partner expecting certain reactions from the other partner and each partner implicitly aware of the other’s intentions. From our videos at 35 weeks, we can see the outlines of this frame beginning to form. Susan pounds twice, looking at her hand. Sheryl says, “Boo Boo” in the same rhythm. Susan looks at her mother immediately, with her jaw dropped, and pounds once. Sheryl taps twice, as Susan did before. Glancing at her own hand, Susan lightly pounds once, and immediately shifts her gaze to Sheryl’s hand. Sheryl taps once (as if in response), slides her hand to Susan’s hand and taps it. Susan’s gaze follows Sheryl’s hand as it covers Susan’s hand. Susan stares at the hands for 3 seconds until the toy moves and attracts her attention. The immediacy of the infant’s gaze shift and the duration of her gaze at mother’s hand after mother responded by tapping suggested to us that Susan implicitly expected an answer from her mother to her pounding movement. This expectation seems to reflect a growing awareness of her mother’s intentionality. Her pounding and Sheryl’s tapping actions follow each other sequentially in time, linked by their rhythm (note the contiguity of bold and italic fonts). It appears that this is the beginning of a sense of secondary intersubjectivity via coordinated joint attention, a shared sense of coagency, and the infant’s awareness of the link between her own and her mother’s actions (see Figure 4). At 36 weeks, it is clear that the game does not always unfold as expected. This observation not only suggests the emergence of a shared frame, but also that the infant begins to experiment with being an agent, able to refuse or to accept participation. 78 ALAN FOGEL AND ILSE DEKOEYER-LAROS FIGURE 4. Third developmental period: The emergence of a stable pounding frame (35–37 weeks). There is more coordination between self and other, as Susan watches her own pounding hand, then looks at her mother’s tapping while she continues to pound, and then looks back at her own pounding hand. Sheryl asks, “Wanna play a little slap game?” Susan looks at Sheryl’s tapping hand and then at her face. Sheryl repeats her request in various, creative ways. However, no matter how playful Sheryl is in her attempts to engage the infant in play, Susan just watches. Eventually, she pounds a few times and then smiles at her smiling mother. Shortly after, she stops and turns away. The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 79 But at 37 weeks, Susan does participate. Sheryl asks, “Show me how you pound,” tapping in a fast, staccato rhythm and gazing at her daughter. Susan looks into Sheryl’s eyes, while Sheryl repeats the request—moving her head and hand up and down in the same rhythm as her vocalization of the spoken sentence—and starts to smile. It is as if Sheryl is pounding with her head, hand, and speech. Susan now starts to pound with her right hand. She smiles—tilting her head and raising her eyebrows— and gazes intently into mother’s eyes. Sheryl exclaims, “Yeah! That’s a good girl!” Susan turns to look at her own hand as Sheryl taps twice. She alternates her gaze between Sheryl’s tapping hand and her own pounding hand, until she turns to the side, staring in the distance, forcefully pounds with both hands, looks at them, and shouts “Ah!” After this, she rests her head on her arm. For the first time, Susan alternates her gaze between her own and her mother’s hands, while continuing to pound, which we indicate with bold and italic together. This may enable Susan to notice the difference between her own and Sheryl’s intentional movements. Between 35 and 37 weeks, the mother’s behaviors also changed. First, her attunements to Susan’s actions became more accentuated. One example of this can be found in the description above. Another one was also observed at 37 weeks. Susan is pounding, while Sheryl vocalizes, “Bang! Bang!” moving her head, hand, and intonation in the same up-and-down contour. They smile at each other. Then, Susan pounds more forcefully, alternating her left and right hand. Sheryl now moves her head from left to right, continuing to vocalize rhythmically. As soon as she notices Sheryl’s left-to-right movements, Susan stops pounding and gazes into Sheryl’s eyes. They look at each other and smile, and Sheryl wrinkles her nose. Second, Sheryl also seems to accentuate the use of the word “you” more when addressing her infant. From the beginning of our observations, at 27 weeks, she often commented on Susan’s gaze direction, actions, likes and dislikes, and past and future events involving Susan, placing no special emphasis on the word “you.” For example, at 33 weeks, she says “You like to do that in the bathtub, don’t you? You like to really splash.” At 35 weeks, however, Sheryl begins to emphasize “you” to amplify the infant’s current actions. For instance, the infant makes a toy beep and mother comments enthusiastically, “Yeah! You did it! You’re making that noise now.” At 36 weeks, the same happens and mother says “You did that” and “You made it go beep.” At 37 weeks, when the infant makes the toy beep, mother repeats several times, “You made that noise!” emphasizing “you” each time. One interpretation of these changes in the mother’s behaviors toward Susan is that Sheryl is noticing Susan’s emerging awareness of being an active agent of her own pounding actions, and amplifying it with her actions and language. Another interpretation is that mother expects more awareness of self-agency to emerge at this age, and that she is providing a scaffold within which the infant can develop this awareness. 80 ALAN FOGEL AND ILSE DEKOEYER-LAROS In summary, for the third developmental period in comparison with the second, it is clear that both mother and infant change in their behavior in relation to each other and that they achieve novel types of coordinated co-actions. Of particular interest with respect to our focus on the development of intersubjectivity is the infant’s growing ability to alternate and coordinate self and other-awareness, along with the mother’s novel use of the word “you.” There is no way to tell for certain from our data whether the mother intends to highlight the infant’s growing sense of agency and intersubjectivity or whether it is an implicit action that in some way reflects the mother’s sense of her infant’s developmental changes. Nevertheless, the infant seems to have a few glimpses into a sense of secondary intersubjectivity, scaffolded by the now stable pounding frame. The Fourth Developmental Period, Emergence of Secondary Intersubjectivity: 38 to 40 Weeks Unfortunately, observations at 38 and 39 weeks are missing, because mother and infant were visiting Susan’s grandmother. Sheryl notes in her journal, “This has been a good visit for Grandma, she got a lot of firsts.” Susan gives her first kiss, claps her hands, and learns how to shake her head “no.” These are all signs of a growing referential communication, that is, communication about some shared focus of attention (Camaioni et al., 2004). At 39 weeks, Sheryl writes, “Grandma marveled at how much [Susan] had grown, changed (physically and emotionally) since we had been home [at Grandma’s] the two weeks.” At 40 weeks, mother describes a new game, “Susan likes to play a game shaking her head ‘no’ when Mommy says ‘yes.’ She responds with a big smile and starts the game all over again. The beginnings of defiance? It’s cute anyway.” This description indicates that Susan’s sense of secondary intersubjectivity has spread from a few instances within the pounding game to become a stable and more pervasive feature of different types of frames with mother and with other people. Susan seems to understand that her mother has a different intention (yes) from her own (no) and creates a teasing game around that (cf. Reddy, 2001). We observed an instance of this new game at 40 weeks, following an interaction involving pounding. Susan glances at the table and then looks at Sheryl’s face. Raising her eyebrows, she starts to pound. Pounding the table top with flat hands, she follows Sheryl’s movements with her gaze. It appears as though she wants to initiate the pounding game. But Sheryl says, “I wanna show off and show how you can do pat-a-cake.” Susan stops and reaches for Sheryl’s face. Sheryl starts to sing and clap “pat-a-cake.” Continuing to stare intently at her singing and clapping mother, Susan starts to pound the table again. Continuing to pound, she shifts her gaze to her hands and then back to Sheryl, who says, “No, show them how you clap, okay?” She takes hold of Susan’s hands and claps them until they finish the song, both smiling. Sheryl starts again, clapping her hands demonstratively in front of Susan. This time, Susan kicks the rhythm against the bottom of the high chair. Sheryl laughs, “Are you doing it with The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 81 your feet?” Susan looks at Sheryl with her jaw dropped and continues to kick. As they end the song, Susan softly laughs. Sheryl mimics this and asks, “You’re not going to do it, are ya?” while slightly moving her head side to side and clapping her hands. Susan looks at Sheryl, tilts her head, and raises her eyebrows. She smiles and then shakes her head “no.” Sheryl smiles broadly and they start a game in which Sheryl playfully repeats the words “yes” and “no,” nodding and shaking her head. Susan shakes her head a few more times. They smile and laugh together. In this observation, the latter part of which is shown in Figure 5, the dyad reveals a different pattern within the pounding frame, with new roles for mother and infant and an apparently new awareness of the infant’s unique role and ability to choose, initiate, and refuse. Consistent with experimental studies on the development of infants’ awareness of others’ intentions at this age (e.g., Behne et al., 2005), Susan seems to guess Sheryl’s intention, but chooses to act differently. Susan now initiates her own unique expression, maintaining connection to Sheryl by pounding in the same rhythm. Sheryl appears to notice the cross-modal similarity between her daughter’s pounding and kicking movements and her own clapping, suggested by her laughter and her asking, “Are you doing it with your feet?” In summary, at 40 weeks, mother and infant have a different “feel” to us as observers (cf. Stern, 1985), indicative of a developmental transition to secondary intersubjectivity. This developmental change is also marked by observable shifts in the entire system of communication and self-awareness, both for mother and infant. The infant now seems to realize that she can intentionally act differently than Sheryl wants her to: she shakes her head “no,” laughs, and her vocalizations sound more word-like (e.g., she now clearly articulates “dadada”). She is able to smoothly coordinate her gaze between Sheryl and her pounding hands. Sheryl acts differently, too. She uses her voice more matter-of-factly, speaks faster and moves more with her entire body, in comparison to all previous sessions. Such changes in maternal communication patterns have been related to infants’ ability to engage in gestural and verbal “acts of meaning” (Bruner, 1983; Stern, 1985; Trevarthen, 1998). Sheryl writes, “She likes to get my attention by wrinkling her nose into a funny face. It’s her personality starting to blossom. She really is a happy baby.” Discussion Summary of Findings The story of Susan and Sheryl provides one possible developmental pathway of the process of change in secondary intersubjectivity within the mother-infant relationship around the 9-month developmental transition. Our goal, taking a dynamic systems perspective, was to examine the historical unfolding of the change process within the dyad in order to better understand how a re-organization of relationship components may come about to create an emergent psychological milestone. 82 ALAN FOGEL AND ILSE DEKOEYER-LAROS FIGURE 5. Fourth developmental period: The emergence of secondary intersubjectivity (38–40 weeks). Susan looks at Sheryl (who wants Susan to play pat-a-cake), tilts her head, and raises her eyebrows. She smiles and then shakes her head “no.” They smile and laugh together. During the first developmental period, 26–32 weeks, the mother’s co-orientations with Susan’s pounding established an innovation of helping to focus the infant’s attention to her own movements, at least occasionally (Stern, 1985, 1993). Sheryl’s spontaneous attunements—mainly mimicking the infant’s feeling contours with her voice—served to amplify the infant’s movements and to temporarily bring them into The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 83 the infant’s awareness. They seemed to enhance the infant’s incipient awareness of herself as the agent of the pounding actions, suggested by her gazing at mother with her jaw dropped after Sheryl vocalized in tune with her pounding (see Figure 6 [a]). Stern (1985, 1993) has suggested that cross-modal attunements are particularly important to the infant’s developing sense of intersubjectivity. Attunement behaviors “. . . recast the event and shift the focus of attention to [. . .] the quality of feeling that is being shared” (Stern, 1985, p. 142). According to the intersensory redundancy hypothesis (Bahrick & Lickliter, 2000), cross-modal sensory coordination enhances attention to those sensations in both pre- and post-natal development. In our observations, cross-modal co-orientations were scaffolded by the mother and served to stabilize the infant’s hand and eye movements into a pattern of focused pounding. It is this growing stability that is the signature of an attractor or frame beginning to form in the communication system. This coordinative system was not completely stable in the first developmental period. When Sheryl co-oriented less with the pounding, between 29 and 32 weeks, Susan focused less on her pounding movements, which strengthens our impression that the mother’s actions serve to stabilize and regulate the infant’s innovative selfawareness of her own actions. Nevertheless, the infant increasingly glanced at her pounding hands and at Sheryl’s face before or after pounding, suggesting an emerging attractor for more self-sustained attention to her pounding actions and of the infant’s emerging initiative in engaging the mother in her pounding activities. A second developmental period, at 33 and 34 weeks, was marked by Susan’s increased stability of attention to her pounding and by more variations in emotional intensity than before. Susan alternated strikingly focused and deliberate pounding with prolonged pauses during which she was not pounding but quietly focused on her hands. During vigorous pounding, Susan intently gazed at Sheryl, who again attuned to the pounding vocally and through body movements. It is likely that these dyadic emotional shifts played a role in the felt experience of being with the self-regulating other, over and above the sense of motor and perceptual coordination between attention to self and to mother (see Figure 6 [b]). These types of shared dyadic emotional periods have been described as “moments of meeting” (Stern, 1998) or dyadic states of consciousness (Tronick, 1998) and are hypothesized to serve the developmental function of creating shared psychological states upon which future forms of coordination depend. These moments of meeting were the innovations that emerged in the second period. This hypothesis makes sense because what follows, between 35 and 37 weeks during the third developmental period, is the emergence of a stable pounding frame involving mutual expectations and coordinated co-actions. Susan seems to realize, “Whenever I pound, Mother will tap.” Her gaze at mother’s hands after she herself starts to pound implies an expectation of mother replying to her movements in the same rhythm. The rhythm of the pounding hands may have helped this infant and mother to further develop and stabilize shared meanings of how their mutual pounding frame unfolds, affording the infant with a sense of agency to alter the flow of events in the pounding frame. From the other side of the dyad, the mother began to call the pounding a “slap game,” and emphasized the word “you,” suggesting an innovative mutual expectation of their response to each other (see Figure 6 [c]). This 84 ALAN FOGEL AND ILSE DEKOEYER-LAROS Mother attunes to infant Gaze Mother attunes to pounding Gaze Gaze Own hand Mother (a) Co-attunement, shared emotions Own hand Mother Infant attunes Infant attunes vigorously quietly (b) Developmental Change to Emergent Secondary Intersubjectivity Gaze Gaze Temporal contiguity between self and mother reference Own hand Mother Infant attunes quietly Gaze Teasing Refusing Own hand Mother New postures, tones of voice Infant attunes vigorously (c) (d) FIGURE 6. (a) The first developmental phase (26–32 weeks). The mother’s attunement creates an emotional “container” that helps the infant self-regulate by attending to her own pounding actions and also to her mother’s face and pounding actions. When there is no maternal attunement, the infants does not attend to mother, nor to her own pounding hands. (b) The second developmental phase (33–34 weeks). Infant begins to co-attune with mother, creating two different types of contours: one more quiet and the other more vigorous and more likely to be shared with mother. (c) Third developmental phase (35–37 weeks). There is a closer contiguity between infant gazing at self and at mother, the emergence of a stable pounding frame that contains the actions and emotions, and increasingly shared intentions. The first instance of secondary intersubjectivity occurs at the end of this period. (d) Fourth developmental period (38–40 weeks). This marks the emergence of a new frame for secondary intersubjectivity in which both infant’s and mother’s behaviors have changed. New patterns emerge such as teasing, refusing, changed use of the pronoun “you” by the mother, and shifts of expression and tone of voice. illustrates a novel reorganization of many components of the pounding and the dyadic nature of the infant’s growing self-awareness. Research in music perception and mother-infant interaction have shown exquisite sensitivity to rhythm patterns as a way to establish and maintain coordinated co-action and shared emotion (Jaffe, Beebe, Feldstein, Crown, & Jasnow, 2001; Stern, 2001; Trevarthen, 1998). The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 85 The developmental transition to widespread and stable sense of secondary intersubjectivity appears to occur in the fourth developmental period, between 38 and 40 weeks. During this period, a qualitatively different communication pattern emerged and stabilized in the dyad: both mother and infant began to behave differently than before—in posture, tone of voice, and communicative expressions—suggesting true developmental change. By participating in pat-a-cake by pounding or kicking while refusing to clap, the infant shows that she understands the mother’s intention but chooses to engage in the game on her own terms. The rhythm of the game is similar between the members of the dyad but the exact form of its enactment is different, to mother’s and infant’s great enjoyment. Both appear to have developed a new understanding of themselves and how they relate to each other (see Figure 6 [d]). These observations concur with those of Trevarthen and Hubley (1978) who observed a similar novel sense of confidence and understanding of intention in the infant Tracey around 9 months of age. They are also reminiscent of Reddy’s (2001; Reddy, Hay, Murray, & Trevarthen, 1993) observations of teasing games, during which infants deliberately provoke their partners to mischievously disrupt the others’ ongoing actions, not comply with a previously accepted prohibition, or disrupt normally accepted gestures and routines. Teasing presumes secondary intersubjectivity. During such provocations, infants have also been observed to immensely enjoy their partners’ exaggerated protests, as during Susan’s “shaking the head ‘no’” game. Inferences about the Process of Developmental Change From these observations, we can infer systematic regularities in the process of change. In each developmental period, there appears to have been some kind of dyadic innovation that emerged via the coordinated and mutually amplifying action of the mother and infant. In the first developmental period, for example, Sheryl’s attunements seem to temporarily stabilize Susan’s attention to her own action. The ability to self-attune, crucial to emerging self and other-awareness, becomes a dynamically stable attractor by the second period, in which Susan attends consistently to her own and her mother’s behavior. This leads, in the second developmental period, to an emerging attractor of coordinated action and attention across multiple modalities of emotion, action, and gazing, an innovation that becomes the stable pounding frame in the third period. Innovation in the third period involved the increasing contiguity of self and other actions, brief glimpses of secondary intersubjectivity during the pounding frame, and changes in the mother’s behavior (such as emphasizing the word “you”). These innovations seem to lead to the consolidation and spread of secondary intersubjectivity to other frames in the mother-infant relationships (and in other family relationships) in the fourth period. These observations fit well with our previous research on the developmental change process. In a larger sample (n = 13) of change in infants between 2 and 6 months (Fogel et al., 2006), we found that innovations in one developmental period always preceded and foreshadowed the emergence of new frames in the next period. This seems to be the chief mechanism by which the dyad conserves its own history 86 ALAN FOGEL AND ILSE DEKOEYER-LAROS while at the same time gradually transforming its relationship. Second, we found that major developmental change, the fourth developmental period in this study, always had the characteristic of “permeability.” This means, as we observed with the infant Susan as well, that multiple frames in the relationship system began to blend as the history of innovations spreads across the entire system. We found, for all of our 13 dyads, that permeability, and hence developmental change and the emergence of new frames, occurred relatively rapidly, within one or two sessions. This rapid reorganization of the entire system of frames, the hallmark of a developmental change, is not, however, discontinuous or non-linear. One can observe the threads of change occurring systematically during each developmental period, with new attractors that are built from historically familiar and readily accessible components. It is as if the dyad can change quickly when it is ready, but readiness implies carrying the familiar existing patterns of co-action with them (in such a way as to avoid catastrophic non-linear, unprecedented) transition. Research on clinical cases and on developmental psychopathology suggests that non-linear change in human development is most often experienced by participants as traumatic with deleterious psychological consequences. Put another way, normally developing social relationships create bridges between the past and future that buffer participants from precipitous changes (cf. Fogel et al., 2006). Conclusions One advantage of our qualitative microgenetic research approach is the opportunity to trace developmental change. Within a single dyad, we can see how new forms of dyadic communication emerge in each different developmental period. More crucially, however, we can see how each new developmental period depends upon the specifics of what occurred earlier. This relational-historical perspective is missing in much of developmental science but it is central to understanding the process of how development unfolds over time. Siegler’s microgenetic studies (cited earlier), to take one example, have contributed much to the understanding of how children learn novel cognitive strategies (e.g., math). We cannot generalize these findings to larger populations. Focusing on a small number of observations from a single dyad from a dynamic systems perspective, however, is not a limitation but rather a strength because it allows for an in-depth examination of the process of change in the whole system of communication. The ability to observe a system before, during, and after a change using a microgenetic design provides insights about change that are generative if not generalizable. They are generative in the sense that our findings of particular processes of mutual attunement— rhythm and affect matching—as amplifiers for infant attention, the emergence of stable attractors for shared awareness as a foundation for later forms of coordination, and our particular description of the four developmental periods could be applied to study other dyads across the same transition. According to dynamic systems principles, generalities about the population must emerge by finding out what is common across cases, and not by statistical averaging and group-based correlational models (Bergman, Cairns, & Nilsson, 2000; Fogel, 1990; Thelen, 1990). Future work using The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 87 this approach can also facilitate understanding of the specific historical processes in cases where development is derailed or delayed. Acknowledgments This study was supported by grants to Alan Fogel by the National Institutes of Health (R01 HD21036) and the National Institute of Mental Health (R01 MH48680 and MH 57669). We thank Susan and her mother for their participation in this study. The authors thank Manuela Lavelli for her helpful comments on this paper. Correspondence concerning this article should be addressed to Alan Fogel, Department of Psychology, University of Utah, 380 South 1530 East, Salt Lake City, UT, 84112-0251, USA. References Ainsworth, M., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment. Hillsdale, NJ: Lawrence Erlbaum Associates Publishers. Bahrick, L. E., & Lickliter, R. (2000). Intersensory redundancy guides attentional selectivity and perceptual learning in infancy. Developmental Psychology, 36, 190–201. Baumrind, D. (1989). Rearing competent children. In W. Damon (Ed.), Child development today and tomorrow (pp. 349–378). New York: Jossey Bass. Beebe, B. (1998). A procedural theory of therapeutic action: Commentary of the symposium, “Interventions that effect change in psychotherapy.” Infant Mental Health Journal, 19, (3), 333–340. Beebe, B., & Lachmann, F. M. (2002). Infant research and adult treatment: Co-constructing interactions. Hillsdale, NJ: The Analytic Press. Behne, T., Carpenter, M., & Tomasello, M. (2005). One-year-olds comprehend the communicative intentions behind gestures in a hiding game. Developmental Science, 8(6), 492–499. Bergman, L. R., Cairns, R. B., & Nilsson, L. (2000). Developmental science and the holistic approach. Mahwah, NJ: Lawrence Erlbaum Associates Publishers. Black, B., & Logan, A. (1995). Links between communication patterns in mother-child, father-child, and child-peer interactions and children’s social status. Child Development, 66, 255–271. Bowlby, J. (1973/1991). Attachment and Loss: Vol. 2. Separation, anxiety and anger. London: Penguin Books. Bretherton, I. (1990). Open communication and internal working models: Their role in the development of attachment relationships. In R. Dienstbier & R. A. Thompson (Eds.), Nebraska symposium on motivation 1988: Socioemotional development (pp. 57–113). Lincoln: University of Nebraska Press. Bretherton, I. (1991). Pouring new wine into old bottles: The social self as internal working model. In M. R. Gunnar & L. A. Sroufe (Eds.), Self processes and development (pp. 1–41), Vol. 23. Hillsdale, NJ: Lawrence Erlbaum Associates Publishers. Bretherton, I. (1993). From dialogue to internal working models: The co-construction of self in relationships. In C. A. Nelson (Ed.), Memory and affect in development: The Minnesota symposia on child psychology, 26, 237–263. Hillsdale, NJ: Lawrence Erlbaum Associates Publishers. Bruner, J. (1983). Child’s talk: Learning to use language. New York: Norton. Camaioni, L., Perucchini, P., Bellagamba, F., & Colonnesi, C. (2004). The role of declarative pointing in developing a theory of mind. Infancy, 5(3), 291–308. 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, no. 176. 88 ALAN FOGEL AND ILSE DEKOEYER-LAROS Coopersmith, S. (1967). The antecedents of self-esteem. San Francisco, CA: W.H. Friedman and Company. Damasio, A. (1999). The feeling of what happens: Body and emotion in the making of consciousness. San Diego, CA: Harcourt, Inc. De Wolff, M. S., & Van IJzendoorn, M. H. (1997). Sensitivity and attachment: A meta-analysis on parental antecedents of infant attachment. Child Development, 68, 571–591. Denzin, N. K., & Lincoln, Y. S. (1994). Handbook of qualitative research. Thousand Oaks, CA: Sage Publications, Inc. Feshbach, N. D. (1987). Parental empathy and child adjustment/maladjustment. In N. Eisenberg & J. Strayer (Eds.), Empathy and its development (pp. 271–291). New York: Cambridge University Press. Fifer, W., Monk, C., & Grose-Fifer, J. (2001). Prenatal development and risk. In G. Bremmer and A. Fogel (Eds.), Blackwell handbook of infant development (pp. 505–542). United Kingdom: Blackwell Publishers. Flynn, E. & Siegler, R. (2007). Measuring change: Current trends and future directions in microgenetic research. Infant and Child Development, 16, 135–149. Fogel, A. (1990). The process of developmental change in infant communicative action: Using dynamic systems theory to study individual ontogenies. In J. Colombo & J. Fagen (Eds.), Individual differences in infancy: Reliability, stability and prediction. Hillsdale, NJ: Lawrence Erlbaum Associates Publishers. Fogel, A. (1993). Developing through relationships: Origins of communication, self, and culture. Chicago: University of Chicago Press. Fogel, A. (1995). Relational narratives of the prelinguistic self. In P. Rochat (Ed.), The self in infancy: Theory and research. Amsterdam: Elsevier. Fogel, A. (2001). A relational perspective on the self and emotions. In H. A. Bosma & E. S. Kunnen (Eds.), Identity and emotion: Development through self-organization (pp. 93–114). Cambridge, UK: Cambridge University Press. Fogel, A. (2006). Dynamic systems research on interindividual communication: The transformation of meaning making. Journal of Developmental Processes, 1, 7–30. Fogel, A., Garvey, A., Hsu, H., & West-Stroming, D. (2006). Change processes in relationships: Relational–historical research on a dynamic system of communication. Cambridge, UK: Cambridge University Press. Forrester, M. A. (2001). The embedding of the self in early interaction. Infant and Child Development, 10, 189–202. Garvey, A. & Fogel, A. (in press). Dialogical change processes, emotions, and the early emergence of self, International Journal of Dialogical Sciences. Gibson, E. J. (1993). Ontogenesis of the perceived self. In U. Neisser (Ed.), The perceived self (pp. 25–42). New York: Cambridge University Press. Granic, I. (2000). The self-organization of parent-child relations: Beyond bidirectional models. In M. Lewis & I. Granic (Eds.), Emotion, Development and Self-Organization: Dynamic Systems Approaches to Emotional Development (pp. 267–348). New York: Cambridge University Press. Gustafson, G. E., Green, J. A., West, M. J. (1979). The infant’s changing role in mother-infant games: The growth of social skills. Infant Behavior and Development, 2, 301–308. Harter, S. (1999). The construction of the self: A developmental perspective. New York: Guilford Press. Hsu, H. C., & Fogel, A. (2001). Infant vocal development in a dynamic mother-infant communication system. Infancy, 2, 87–109. Jaffe, J., Beebe, B., Feldstein, S., Crown, C. L., & Jasnow, M. D. (2001). Rhythms of dialogue in infancy: Coordinated timing in development. Monographs of the Society for Research in Child Development, 66. Koren-Karie, N., Oppenheim, D., Dolev, S., Sher, E., & Etzion-Carasso, A. (2002). Mothers’ insightfulness regarding their infants’ internal experience: Relations with maternal sensitivity and infant attachment. Developmental Psychology, 38, 534–542. The Developmental Transition to Secondary Intersubjectivity in the Second Half Year 89 Lavelli, M. & Fogel, A. (2005). Developmental changes in the relationship between infant’s attention and emotion during early face-to-face communication: The 2-month transition. Developmental Psychology, 41, 265–280. Lavelli, M., Pantoja, A. P. F., Hsu, H., Messinger, D., & Fogel, A. (2005) Using microgenetic designs to study change processes. In D. G. Teti (Ed.), Handbook of research methods in developmental psychology. (pp. 1–50) Oxford, UK: Blackwell Publishers. Legerstee, M. (2005). Infants’ sense of people: Precursors to a theory of mind. New York: Cambridge University Press. Lewis, M. (1995). Embarrassment: The emotion of exposure and evaluation. In J. P. Tangney & K. W. Fisher (Eds.), Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride (pp. 198–218). New York: The Guildford Press. Lock, A. (1992). Anecdotal events may be really important events. Human Movement Science, 11, 497–504. Lyons-Ruth, K. (1998). Implicit relational knowing: Its role in development and psychoanalytic treatment. Infant Mental Health Journal, 19, 282–289. Maccoby, E. E. (1992). The role of parents in the socialization of children: An historical overview. Developmental Psychology, 28, 1006–1017. Neisser, U. (1993). The perceived self: Ecological and interpersonal sources of self-knowledge. Emory symposia in cognition, 5. New York: Cambridge University Press. Oppenheim, D., & Koren-Karie, N. (2002). Mothers’ insightfulness regarding their children’s internal worlds: The capacity underlying secure child-mother relationships. Infant Mental Health Journal, 23, 593–605. Oppenheim, D., Koren-Karie, N, & Sagi, A. (2001). Mothers’ empathic understanding of their preschoolers’ internal experience: Relations with early attachment. International Journal of Behavioral Development, 25, 16–26. Panksepp, J. (2001). The long-term psychobiological consequences of infant emotions: Prescriptions for the twenty-first century. Infant Mental Health Journal, 22, (1–2), 132–173. Piaget, J. (1952). The origins of intelligence in children. New York: International Universities Press. Patton, M. Q. (1990). Qualitative evaluation and research methods (2nd ed.). Thousand Oaks, CA: Sage Publications. Phillips, A., Wellman, H., & Spelke, E. (2002). Infants’ ability to connect gaze and emotional expression to intentional action. Cognition, 85(1), 53–78. Pipp, S. (1993). Infants’ knowledge of self, other, and relationship. In U. Neisser (Ed.), The perceived self: Ecological and interpersonal sources of self-knowledge (pp. 185–204). New York: Cambridge University Press. Reddy, V. (2001). Mind knowledge in the first year: Understanding attention and intention. In G. Bremner & A. Fogel (Eds.), Blackwell handbook of infant development (pp. 241–264). Oxford: Blackwell Publishers. Reddy, V., Hay, D., Murray, L., & Trevarthen, C. (1993). Communication in infancy: Mutual regulation of affect and attention. In G. Bremner, A. Slater, & G. Butterworth (Eds.), Infant development: Recent Advances, pp. 247 – 274. Hove, UK: Psychology Press. Rochat, P. (1995). Early objectification of the self. In P. Rochat (Ed.), The self in infancy: Theory and research (pp. 53–71). Amsterdam: Elsevier. Rogoff, B., Mistry, J., Goncu, A., & Mosier, C. (1993). Guided participation in cultural activity by toddlers and caregivers. Monographs of the Society for Research in Child Development, 236, 58(8). Savage-Rumbaugh, E. S., & Fields, W. M. (2000). Linguistic, cultural and cognitive capacities of bonobos. Culture & Psychology, 6, 131–153. Schore, A. (2001). Effects of a secure attachment on right brain development, affect regulation, and infant mental health. Infant Mental Health Journal, 22, (1–2), 7–66. Secrist, C., de Koeyer, I., Bell, H., & Fogel, A. (2002). Combining digital video technology and narrative methods for understanding infant development. Forum for Qualitative Sociology Research, 3, (2). Retrieved from http://www.qualitative-research.net/fqs-texte/2-02/2-02 secristetal-e.htm. 90 ALAN FOGEL AND ILSE DEKOEYER-LAROS Shanker, S. & King, B. (2002). The emergence of a new paradigm in ape language research, Behavioral & Brain Sciences, 25, 605–626. Shonkoff, J. P., & Phillips, D. A. (2000). From neurons to neighborhoods: The science of early childhood development. Washington, DC, US: National Academy Press. Siegel, D. (2001). Toward an interpersonal neurobiology of the developing mind: Attachment relationship, “mindsight,” and neural integration. Infant Mental Health Journal, 22, (1–2), 67–94. Siegler, R. S., & Crowley, K. (1991). The microgenetic method: A direct means for studying cognitive development. American Psychologist, 46(6), 606–620. Sommerville, J., & Woodward, A. (2005). Pulling out the intentional structure of action: the relation between action processing and action production in infancy. Cognition, 95(1), 1–30. Stern, D. N. (1985). The interpersonal world of the infant. A view from psychoanalysis and developmental psychology. New York: Basic Books. Stern, D. N. (1993). The role of feelings for an interpersonal self. In U. Neisser (Ed.), The perceived self: Ecological and interpersonal sources of self-knowledge (pp. 205–215). New York: Cambridge University Press. Stern, D. N. (1998). The process of therapeutic change involving implicit knowledge: Some implications of developmental observations for adult psychotherapy. Infant Mental Health Journal, 19, 300–309. Stern, D. N. (2001). Putting time back into our considerations of infant experience: A michrodiachronic view. Infant Mental Health Journal, 21, 21–28. Strauss, A. & Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park, CA: Sage Publications. Striano, T. & Rochat, P. (1999). Developmental links between dyadic and triadic social competence in infancy. British Journal of Developmental Psychology, 17, 551–562. Thelen, E. (1990). Dynamical systems and the generation of individual differences. In J. Colombo & J. Fagen (Eds.), Individual differences in infancy: Reliability, stability, prediction. (pp. 19– 43). Hillsdale, NJ: Lawrence Erlbaum Associates. Thelen, E., Fisher, D. M., & Ridley-Johnson, R. (1984). The relationship between physical growth and a newborn reflex. Infant Behavior and Development, 7, 479–493. Thelen, E., & Fogel, A. (1989). Toward an action-based theory of infant development. In J. Lockman & N. Hazen (Eds.), Action in social context: Perspectives on early development (pp. 23– 64). New York: Plenum Press. Thelen, E., & Ulrich, B. D. (1991). Hidden skills. Monographs of the Society for Research, Child Development, 56, 6–97. Tomasello, M. (1993). On the interpersonal origins of self-concept. In U. Neisser (Ed), The perceived self: Ecological and interpersonal sources of self-knowledge. Emory symposia in cognition, 5 (pp. 174–184). New York: Cambridge University Press. Trevarthen, C. (1993). The self born in intersubjectivity: An infant communicating. In U. Neisser (Ed.), The Perceived Self: Ecological and Interpersonal Sources of Self-Knowledge, (pp. 121–173). New York: Cambridge University Press. Trevarthen, C. (1998). The concept and foundations of infant intersubjectivity. In S. Bråten (Ed.), Intersubjective Communication and Emotion in Early Ontogeny, (pp. 15–46). Cambridge: Cambridge University Press. 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. 183–227). New York: Cambridge University Press. Tronick, E. Z. (1998). Dyadically expanded states of consciousness and the process of therapeutic change. Infant Mental Health Journal, 19, 290–299. van Geert, P. (1994). Dynamic systems of development: Change between complexity and chaos. Hertfordshire, UK: Harvester Wheatsheaf. van Geert, P. (1998). We almost had a great future behind us: The contribution of non-linear dynamics to developmental-science-in-the-making. Developmental Science, 1(1), 143–159. Winnicott, D. W. (1971). Playing and reality. New York: Basic Books. Winnicott, D. W. (1990). Playing and reality. London/New York: Routledge. (Originally published in 1971).