Skip to main content Skip to navigation

Abstracts for Perspectives on Joint Action

Joint action in one-year-old infants
Malinda Carpenter

Human cultural life has at its foundation the ability to participate in joint action and shared intentionality. I will examine the roots of this ability by presenting a series of experimental studies of one-year-old infants. First, I will provide evidence that – contrary to popular belief – infants do have the prerequisites needed to engage in joint action, even in Bratman’s sense (e.g., they have an understanding of others’ intentions and the ability and motivation to help others achieve their goals). Then I will review evidence of infants’ active participation in many different types of joint action and joint attention, touching on such topics as infants’ surprisingly complex understanding of shared experiences, evidence of cooperative communication, and infants’ collaboration with adults in more instrumental tasks. I will conclude that the uniquely human ability and motivation to participate in joint action is already seen – in complex and varied ways – in infants as young as one year of age.

Primitive Egos: Linguistic Evidence in the Ontology of Collective Intentions
Natalie Gold

In the philosophical literature there are competing claims about whether "we" is a primitive, or only "I". Research in linguistics shows that some person categories are not represented in speech, and this holds true across languages. In this case, there is reason to think that grammatically relevant concepts exhaust all cognitively relevant concepts. The project is to explore the implications of the theory and evidence from linguistics for philosophical claims and, also, more generally for the cognitive primitives involved in group reasoning.

Young Children's Understanding of Shared Goals and Joint Commitments in Collaborative Activities
Maria Gräfenhain

Young children act jointly with others in different contexts (e.g., Warneken, Chen, & Tomasello, 2006). However, in previous studies investigating young children’s collaborative abilities, children needed another person to achieve a desired effect, for example, to get access to an enclosed toy. It is thus not clear whether children merely coordinated their actions with their partner in order to achieve an individual goal or whether they truly understood joint actions as following joint goals creating joint commitments (see Bratman, 1992).

In my talk, I will present a series of studies investigating this question. We created games that children could play either alone, in parallel with another player, or jointly with another player. We found that 1- to 4-year-old children were highly motivated to play jointly with an adult partner even when they could play the games alone. The younger children, however, seemed to understand another person as acting jointly as long as she was acting in parallel to them. Only the 3- and 4-year-old children adapted their behavior to another person depending on whether they previously had formed a joint commitment to play jointly with her or not. Our findings thus suggest that children develop a relatively sophisticated understanding of joint actions between 2 and 3 years of age.

Joint action planning and anticipation of others’ actions: An EEG study
Dimitris Kourtis

Neurophysiolgical studies suggest that in joint action experiments, the action of one’s partner is represented in one’s brain structures, even prior to a prompted response (e.g. Sebanz et al., J Cogn Neurosci, 2006). A parallel line of research has demonstrated that the motor areas are activated during action observation in a qualitatively similar way as during motor execution. In addition, such activation also occurs during anticipation of an observed action (Kilner et al., Nat Neurosci, 2004; van Schie et al., Nat Neurosci, 2004).

Our experiment consisted of a choice-reaction task, where EEG was recorded simultaneously from two participants sitting opposite each other, while a third person (confederate) was sitting at right angles with them. The participants were required to prepare an action, performed either individually or jointly, or to prepare to observe their partner’s or the confederate’s action. Differences between preparing for individual vs. joint action were recorded in the amplitudes of frontocentral and parietal cognitive ERPs, largely associated with the P300 component, which is related to stimulus evaluation, working memory update and/or linking perception with action. The activation of motor areas during the preparation period was assessed by examining the amplitudes of two motor indices: the late phase of the Contingent Negative Variation (CNV) and the beta Event Related Desynchronization (ERD). No differences were found between preparation for individual action, joint action or anticipation to observe the partner’s action. Interestingly, both the late CNV as well as the beta ERD were significantly smaller when anticipating the confederate’s action, indicating that activation of motor areas during anticipation of action observation may depend on the “social” relationship between two persons.

Shared Tasks, Shared Memories
Natalie Sebanz, Terry Eskenazi, Adam Doerrfeld & Guenther Knoblich

Our actions are not carried out in isolation; we constantly and sometimes unintentionally coordinate our actions with others. When sharing a task, we plan our actions in a complementary manner to our co-actors', and in doing so we form representations of their action plans. Co-representation refers to this sharing of mental representations in social interactions.

I will present data from three new experiments that aimed at investigating whether forming shared action and task representations affects how information relevant to self or other is encoded in memory. Participants performed a categorization task alone and together. Their memory for items they had earlier responded to, items the other had responded to, and items no one had responded to was subsequently tested. We found that participants were able to not only recall items that had been relevant for their own task, but also items relevant to the other’s task. In contrast, they were much worse at recalling information that was neither relevant to self nor other. This was the case even though there was no explicit requirement to coordinate, and the memory advantage for items relevant to the other’s task remained even when participants were given financial incentives to fully focus on their own task. These findings suggest that we may be so prone to taking others’ tasks into account and that we just don’t manage to ignore each other.