The old saying tells us that "three's a crowd", but when we watch babies developing their social skills, we see that they sometimes need practice with more than just two people. Playing with one grown-up can be fun for both parties and is an important beginning to social life, but it doesn't develop all the abilities we use in our social interactions.
Babies a month or so old will look at and "talk to" a familiar caregiver, and pay attention for quite a while as long as the adult uses high-pitched, emphatic "infant-directed talk". In one nice experiment, mothers were videotaped talking to their babies while the two looked at each other through a sheet of Plexiglass. Then the lighting was changed, so that although the babies could still see their mothers, the mothers saw the reflection of another adult and began to talk to that person in "adult-directed talk." The babies soon stopped paying attention to the mothers who were not talking to them. Thus, the presence of another adult presented a new social task to the babies-- a situation in which the mother could be seen and heard to be talking, but she was behaving in quite a different way from her manner when she could see her baby. The presence of a third person meant the baby could no longer interpret the mother's actions according to the rules that had worked before. The baby would need to experience and learn about this in order for behaviors of people in small groups to seem at all reasonable or predictable.
One necessary social skill is the ability to figure out whether another person is addressing oneself or another member of a group. There are some clear cues, like calling the person by name. If you say "Matilda, what kind of sandwich do you want?", Ethelred and Fridewide will probably not answer-- although it's possible that one of them might, if Matilda looks distracted and the other person knows what she prefers. But not all comments have equally clear targets. Sometimes we look at one person while speaking to another, especially if we are speaking about the first person. Ethelred might say, while looking at Matilda, "I think she wants a chicken Caesar wrap." He is not talking TO Matilda, but looking at her for correction in case she's going for the turkey Reuben. Adults can usually figure out the interplay of speakers, listeners, and topics by paying attention to context, the recent history of the conversation, facial expression and gaze direction, tone of voice, and so on. (But even with all our skills we sometimes have to ask, "You talking to me?")
Obviously no baby could learn these communicative skills just by interactions with one other person. They usually get a lot of practice with triangular communications in their family or other care settings. A baby may experience multiple interactions while with mother and father, with one parent and another child, with grandparents, or with non-parent caregivers inside or outside the home. By 2 months of age, babies already can co-ordinate some turn-taking in interactions with two other people (see Fivaz-Depeursinge, E., & Corboz-Warnery, A. [1999]. "The Primary Triangle." New York: Basic Books). They shift attention rapidly from one familiar adult to another, sometimes showing pleasure and interest, sometimes protesting if an interaction becomes too intense. The ability to interact in this triangular way is not just evidence of the baby's developing social skill, but has been considered as a good index of family functioning that indicates how well the parents work together and how they communicate to each other and to the baby.
Experience with adults who can communicate well would seem to be an important foundation for babies learning how to operate within a social group. Cues like direction of gaze, body posture, or voice tone should be easier to learn in a simple pair interaction than they will be in a larger group. Experience in switching back and forth from one person to another should also be easier in a triangle than in a crowd.
Work by Elisabeth Fivaz-Depeursinge and others has suggested that mothers who are depressed or otherwise emotionally disturbed may give paradoxical cues to their infants. For example, they may talk to the baby, but at the same time bring their bodies either too near or too far for good communication. The babies may respond by averting their eyes or by staring at the mother's forehead. Even when the father is able to communicate better, these families have unusual patterns of triangular interaction. It's plausible that babies of depressed mothers may fail to learn good social skills as a result of these difficult patterns of interaction. However, we need to keep in mind that there can be genetic or other factors that help determine the babies' development. The babies' signals to their mothers may be unusual in some way we have yet to understand.
Although it's not completely clear that triangular interactions cause the development of good social skills, they do seem associated with important later abilities. This raises questions about how infant day care should best be organized--- and suggest that two may not be enough company.
