Joint Action

This essay will attempt to answer the questions, when does shared agency first appear in human development? And, what role might it play in facilitating development? It will begin by discussing how the ability to engage in joint action develops over the first two years of life, and how infants are introduced to turn taking and how to interact with others.

It will then go on to discuss Bratman’s account of shared agency and consider whether this account is sufficient to describe the type of joint action that infants are likely to engage in. it will consider how joint action and theory of mind may be linked and whether evidence is now available to show that infants may develop theory of mind much earlier than previously thought.

Over the first two years of life children begin to engage in progressively more deliberate and flexible joint action with adults and peers. Infants engage in joint action with adults from very early on in life, but only begin to sustain joint action with peers towards the end of their second year, and only in a limited sense (Brownell &Brown 1992).

Joint action can be observed in infants within the first few weeks of life. Parents and caregivers interact with babies by touching, gazing, smiling and talking, and there is research that suggests that infants are prepared before birth to become engaged in these social interactions.During the early months of life infants learn to predict the effect and response in dyadic activity and learn how to coordinate their actions to suit interactions with others. A very early example of turn taking can be observed whilst mothers are breastfeeding babies. Feeding tends to happen in the form of a burst-pause pattern, sucks take place in a series of bursts, with pauses in between bursts. This leads to mothers interacting with the baby (Kaye 1977). During sucking bursts, the mother tends to be inactive and quiet, during pauses she will jiggle, stroke and talk to the infant, thus setting up a turn taking pattern, where, first one and then the other is principal actor while the other is spectator. The mother fits her actions in to the infants natural sucking rhythm, which introduces the infant to a way of interacting which is typical of many social situations.

Participating in these early forms of interaction with adults allows children to learn the basic skills required to engage in joint action and to experience the positive stimulation that helps to sustain interest and engage with others. There are a number of non-verbal ways of communication which very small infants can use to share an interest or an aspect of the environment with others.

The first of these is pointing. Pointing like movements can be seen from as early as four months, although at that age there is no indication that it is used for communication (Blake et al 1994). Pointing as a gesture emerges at around 10-12 months and is used to draw attention to objects that are interesting but too far away to reach. At this stage it is still not particularly used for communication, as the infant will not check whether an adult is following the gesture. By the second year pointing for self is replaced by pointing for others. The infant is now able to integrate both partner and object in one communicative act, and so using the gesture to influences others (Schaffer 1984).

Another non-verbal gesture used by infants is gaze direction. Following the gaze of another is the most common means of showing and sharing interest. Although, there are indications that infants do not understand the significance of this action until the end of the first year when they will begin to follow the gaze of others. This ability is initially seen under ‘easy’ conditions, when the object and person are both easily accessible (Butterworth 2004). Before this, adults take the initiative. Mothers tend to closely monitor their infants focus and almost automatically look to where they are looking (Collis and Schaffer, 1975). After establishing what the infant is looking at, the mother will often point to the object, comment on it, name it or even bring it to the child.

Most of these early interactions are asymmetrical and rely on the mothers’ sensitivity to her child’s signals. In the same way that the mother needs to fill the pauses in between sucking bouts or periods of vocalisation to embed the infants behaviour in a social situation, she must also be prepared to follow the gaze and gestures of the infant and use these cues to construct a dialogue based on the child’s interest.

These very early examples of joint action do not require the infant to share in any shared goals or desires with the caregiver, which are more important for joint action between older participants, merely the infant must be engaged in and responsive to the adult’s efforts to create a joint action. Infants often abandon these early interactions whenever something else catches their attention, adults accept this behaviour and do not expect commitment to a joint action in the way they would with another adult or older child.

When there is a sensitive responsiveness, infants can learn that their actions are of interest to the adult and that they can elicit a predictable response, thus encouraging them to eventually act to elicit that behaviour. Once an infant has learned that it has a part to play in bringing about the back and forth of social exchanges, interaction becomes a much more symmetrical affair.

After the first six moths of life, interaction between caregivers and infants begins to include objects such as toys. An adult may shake a toy to attract the infants’ attention, the child will focus attention on the toy whilst maintaining interaction with the adult. The adult joins in with the child in engaging with the toy, so that the two together are sharing the activity. These events, which are structured by the adult, scaffold the child and support them in this new form of joint activity (Carpendale & Lewis 2004).

At around one-year-old, infants can initiate joint action with adults. If a parent doesn’t take their turn during a game, such as peek a boo, the infant may attempt to re start the game by covering their face and waiting for a reaction from the adult, or vocalising to the adult. Adults support and scaffold these early forms of joint action and help the child to develop skills to be able to contribute to and control more complex interactions over the first year and into the second year of life.

Joint action is not only studied from a psychological point of view, it is also something that many philosophers are interested in. There is a lot of debate over what joint action is, and what is needed for joint action to take place. Michael Bratman claims that for joint action to take place there needs to be shared intentions.

In Bratman’s account, he claims that each agent not only intends that the group perform a joint action, but also that they have ‘meshing subplans’, that they have a shared intention. Bratman’s claim is that for us to have a shared intention that we J, it is sufficient that:

  1. ‘a) I intend that we J and, b) you intend that we J;’

  2. ‘I intend that we J in accordance with and because of 1a, 1b and meshing subplans of 1a and 1b; you intend that we J in accordance with and because of 1a, 1b and meshing subplans of 1a and 1b;’

  3. ‘1 and 2 are common knowledge between us’. (Bratman 1993)

But is Bratman’s claim too complex to account for instances of joint action involving infants?For there to be common knowledge and meshing subplans the infants would need to have the ability to form beliefs, and to form beliefs about what others believe. To be able to do those things requires a developed theory of mind.

Theory of mind involves the ability to understand why someone may act in a certain way or to predict how another person will act. It requires having the knowledge that other people have beliefs, knowledge, emotions and intentions and using that understanding to navigate social situations. It is thought that children begin to show understanding of others’ beliefs at around 4 years old. At this point they can successfully take part in false belief tasks which require them to answer questions based on beliefs or knowledge which other individuals may have, with few errors.

False belief understanding has been linked to various aspects of social functioning including the ability to maintain friendships, and the ability to engage in conversation, understanding the beliefs and emotions of others plays a role in the development of social competency.

So, it would seem that there is a relationship between joint action and theory of mind, as the ability to co-ordinate actions with others requires understanding of the beliefs and desires of others. Or, maybe it is the case that the co-ordination of actions allows the understanding that others have desires and beliefs which may be different to yours?

Joint action skills emerging in infants may later lead to the development of theory of mind. Joint action skills have been observed in several non-human animals, such as birds in flight, herding cattle or primates engaging in play, with no compelling evidence for theory of mind in non-humans. This may show that joint action is a more primitive ability, which is likely to develop first, without the need for theory of mind.

The alternative is that theory of mind develops first, which then assists in the ability to take part in joint action. (Tomasello 1995) suggests that theory of mind, or theory of mind like behaviours are necessary to drive the development of joint action.

There has been some recent research which may support this claim by showing that infants can show evidence of theory of mind much earlier than was previously thought.

Studies have shown that infants younger that one year old appear to understand the reasons why others may be acting the way that they do and use this understanding to behave appropriately towards others in social situations. (Behne et al 2005) found that 9-month olds behave differently towards adults who are unable to give them a toy than they do to adults who are unwilling to give them a toy, they are more patient with the adults who are unable to (due to failed attempts for example).

There is some evidence, although not as much due to there being less research done, that by 1-year old infants may understand not only the goals that others have, but also the intentions of others. (Gergely et al 2002) studied 14-month olds, the study showed that the infants would imitate the actions of an adult, when shown a new way to execute a task, only if the infants considered the new action a rational alternative. The results were interpreted by the testers to show that preverbal infants use a selective process when deciding which action to take, rather than simply re-enacting what they have been shown.

If these findings are correct, it would appear that infants may have a far more robust theory of mindbefore their first year of life, and are capable of understanding what their partners goals are and so, have the ability to mesh subplans. Meshing subplans is not necessarily the same as having common knowledge, however.

Common knowledge is difficult as it requires, firstly, understanding something about the knowledge that others’ have, and secondly, being able to make sense in some way of the regressive nature of common knowledge (I know that you know that I know…). To make Bratman’s proposal more suitable for infants it has been suggested that joint attention could be substituted for common knowledge (Tollefsen 2005) This could work when discussing infants and could also be used for simple joint action in adults as it allows both partner to have mutual awareness that they are focussed on the same task.

However, there is also evidence that one-year olds do have some basic understanding of common knowledge. (Moll et al 2008) conducted a study of 14-month olds, who showed knowledge of objects being referred to based solely on experiences the adult and child had shared previously. The infants responded not based on what either they nor the adult knew individually, but what they knew together. This study showed that by 14 months old infants are aware of what we know together.

These studies onto theory of mind in infants appear to show that the idea of joint action may not need to be simplified to accommodate infants as much as previously thought, and that in terms of their understanding, by their first birthday infants may have some idea of others’ goals and intentions and that they share knowledge with others. It seems that in the same way that infants develop other skills, such as language and co-ordination, their ability to understand of what is required to work successfully with others is always developing and improving as they grow.