Infant joint attention, neural networks and social cognition

Abstract

Neural network models of attention can provide a unifying approach to the study of human cognitive and emotional development (Posner & Rothbart, 2007). In this paper we argue that a neural network approach to the infant development of joint attention can inform our understanding of the nature of human social learning, symbolic thought process and social cognition. At its most basic, joint attention involves the capacity to coordinate one’s own visual attention with that of another person. We propose that joint attention development involves increments in the capacity to engage in simultaneous or parallel processing of information about one’s own attention and the attention of other people. Infant practice with joint attention is both a consequence and an organizer of the development of a distributed and integrated brain network involving frontal and parietal cortical systems. This executive distributed network first serves to regulate the capacity of infants to respond to and direct the overt behavior of other people in order to share experience with others through the social coordination of visual attention. In this paper we describe this parallel and distributed neural network model of joint attention development and discuss two hypotheses that stem from this model. One is that activation of this distributed network during coordinated attention enhances the depth of information processing and encoding beginning in the first year of life. We also propose that with development, joint attention becomes internalized as the capacity to socially coordinate mental attention to internal representations. As this occurs the executive joint attention network makes vital contributions to the development of human symbolic thinking and social cognition.

Introduction

In this paper we provide an overview of theory and research on how neural networks’ outlook on joint attention assists in our understanding of the mechanisms that support human social learning, symbolic thinking and social cognition. This theory and research has stemmed in large part from the contemporary pursuit of the question; how do people become capable of sharing information with one another? One immediate answer is that it has a lot to do with the phylogeny and ontogeny of bio-behavioral systems specific to language. However, in the later part of the 20th century it became clear that vital elements of the cognitive foundations for information sharing develop before the onset of language in human infancy (Bates et al., 1979, Bruner, 1975, Werner and Kaplan, 1963). One especially pertinent observation was that by 6 to 9 months infants become increasingly capable of sharing experience about objects and events by directing or following the visual gaze of social partners (Bakeman and Adamson, 1984, Bates et al., 1979, Scaife and Bruner, 1975, see Fig. 1). Bruner (1975) referred to this type of preverbal referential behavior as joint attention. He suggested that joint attention behaviors reflect the early onset of psychological processes that are specific to social reference but are distinct from, albeit necessary to the ontogeny of language.

By and large research over the past 30 years has supported Bruner’s hypothesis about the distinct nature of communication versus language development and the primacy of infant joint attention development with regard to the former. For example, Willems et al. (2009) have recently reported evidence for two distinct but interacting neurocognitive systems in adults, one unique to communication processes and one unique to language processes. Studies and theory also suggest that joint attention is pivotal to the development of the human neurocognitive communication functions (Mundy, 1995, Mundy, 2003, Mundy et al., 2009) and that the neurocognitive joint attention functions are integral to the subsequent development of social cognition, as well as language (e.g. Baldwin, 1995, Charman et al., 2000, Hirotani et al., 2009, Kasari et al., 2008, Kwisthout et al., 2009, Meltzoff and Brooks, 2008, Mundy et al., 2003). More to the point of this article recent theory suggests that neurocognitive joint attention development is supported by a distributed neural network that synthesizes multiple sources of information in real time in social interactions and, though not necessarily species specific, contributes to a unique constellation of cognitive capacities in human beings (Mundy et al., 2009).

Bruner (1995) characterized a principal approach to research on joint attention as one that is focused on the “epistemological question”. That is to say, much of the research has been concerned with describing what the development of joint attention behaviors reveals about changes in the stages of knowledge infants possess about other people’s minds, or reveals about the nature of cognitive modules that may be dedicated to knowledge about people’s minds (e.g., Baron-Cohen, 1995, Tomasello et al., 2005). This “social cognitive” paradigm emphasizes the comparison of the types of joint attention infants are capable of at different ages. These comparisons have provided grounds for numerous experimentally controlled inferences and insights about age-related changes in infant’s knowledge about the intentions of other people (e.g., Baron-Cohen, 1995, Carpenter et al., 1998, Meltzoff and Brooks, 2008). The emphasis on knowledge has become so strong in this paradigm that in some opinions joint attention, per se, is not thought to be achievable until a critical module or stage of social cognitive development is reached. Social cognitive theory argues that this point in development is heralded by signs of “intentional” joint attention inferred from the emergence of infants’ tendency to acknowledge shared experience with the use of spontaneous gaze alternation between objects and people by about 12–15 months of age (Baron-Cohen, 1995, Tomasello et al., 2005, see Fig. 1).

A complimentary but alternative approach to the social cognitive paradigm is exemplified by the question: does joint attention lead to social knowledge acquisition (learning) and the capacity to share knowledge with others, and if so, how? This question was part of Bruner’s (1975) original motivation for the study of joint attention. He thought of joint attention as an early onset interactive key to social learning and, therefore, a potentially informative natural example to emulate in curriculum design (Bruner, 1995).

A critical difference between these two views that joint attention ostensibly follows from and is defined by social knowledge development in the social cognitive paradigm, but is conceived of as leading to knowledge development and sharing information in the alternative paradigm. Our research and theory on joint attention development has been guided by this alternative approach for quite some time. A maxim of this approach is that cognitive and psychological development is not only effectively modeled in terms of relatively discontinuous changes in knowledge or modules, but also in terms of changes and continuities in the speed, efficiency, and combinations of information processing that give rise to knowledge or psychological phenomenon (Hunt, 1999).

The approach also emphasizes a constructivist view of cognitive and neurocognitive development in which infants learn as much about the minds of other people from their own actions in episodes of joint attention, as they do from the perceptions of others behavior (Mareschal et al., 2007, Mundy, 2003, Mundy and Newell, 2007, Piaget, 1952). Indeed, this paradigm is associated with a model that describes joint attention development in terms of advances in the parallel processing of three sources of information illustrated in Fig. 2. When people interact in joint attention they actively engage in: (1) self-referenced processing of information about actions they generate (e.g. movement and orientation of one’s own eyes or control of visual attention), and about bodily state (e.g. affect and spatial position relative to a referent), (2) processing information about another person’s attention and behavior, as well as, (3) integrating those strands of input with processing information about a commonly referenced object or event (e.g., Mundy and Newell, 2007, Mundy et al., 1993, Mundy et al., 2009).

We assume that the early development of this type of parallel processing involves a transactional or dynamic interplay between neural, psychological and behavioral development. That is to say, practice with parallel self-referenced, other-referenced, and object/event-referenced information processing during joint attention in infancy is both a consequence and an epigenetic organizational impetus related to the establishment of distributed neural processing network that involves the flow of information across distal frontal, temporal and parietal cortical systems (provided below). With maturation and experience, this distributed neural network comes to serve a social executive function that enables infants to engage in increasingly effortless social coordination of attention to internal information, external social, and external objects/event information in social interactions. With development the overt operations of socially coordinating visual attention to external references become internalized as cognitive operations that allow for the joint coordination of mental attention to common representations. This internalization of joint attention process contributes to the neurocognitive foundation for social cognition, symbolic thought, and self-awareness (Mundy et al., 2009). We refer to this as the parallel and distributed processing model (PDPM) of joint attention development. The following provide more details related to this synopsis of this model and a review of research on the distributed neural networks involved in joint attention development.