The intentional network: How the brain reads varieties of intentions

https://doi.org/10.1016/j.neuropsychologia.2007.05.011Get rights and content

Abstract

Social neuroscience provides insights into the neural correlates of the human capacity to explain and predict other people's intentions, a capacity that lies at the core of the Theory of Mind (ToM) mechanism. Results from neuroimaging research describe a widely distributed neural system underlying ToM, including the right and left temporo-parietal junctions (TPJ), the precuneus, and the medial prefrontal cortex (MPFC). Nevertheless, there is disagreement in the literature concerning the key region for the ToM network. Some authors point to the MPFC, others to the right TPJ. In the effort to make a contribution to the debate, we propose a model of a dynamic ToM network consisting of four regions. We also introduce a novel theoretical distinction among varieties of intention, which differ by the nature of an individual's pursued goal (private or social) and by the social interaction's temporal dimension (present or future). Our results confirm the crucial role of both the MPFC and the right TPJ, but show that these areas are differentially engaged depending on the nature of the intention involved. Whereas the right TPJ and the precuneus are necessary for processing all types of prior intentions, the left TPJ and the anterior paracingulate cortex are specifically involved in the understanding of social intention. More specifically, the left TPJ is activated only when a subset of social intentions are involved (communicative intentions). Taken together, these results demonstrate the progressive recruitment of the ToM network along the theoretical dimensions introduced in the present paper.

Introduction

Human beings experience other people as goal-directed and intentional agents, and researchers refer to Theory of Mind (ToM) as the ability to explain and predict the behaviour of conspecifics, based on observation of their intentional actions. In the literature, there is widespread agreement on the existence of a widely distributed neural network underpinning ToM, including the right and left temporo-parietal junctions (right TPJ and left TPJ), the precuneus, and the medial prefrontal cortex (MPFC) (Brunet, Sarfati, Hardy-Baylé, & Decety, 2000; Fletcher et al., 1995; Frith & Frith, 2003; Gallagher et al., 2000). Evidence of anatomical connections within this network comes from rhesus monkeys studies showing that these areas are connected with each other bidirectionally (Barbas & Pandya, 1989; Morecraft, Cipolloni, Stilwell-Morecraft, Gedney, & Pandya, 2004; Pandya, Van Hoesen, & Mesulam, 1981; Seltzer and Pandya, 1978, Seltzer and Pandya, 1989).

Although the prevalent view is that the MPFC is the key brain area subserving ToM (Amodio & Frith, 2006; Frith & Frith, 2006; Gallagher & Frith, 2003), it has recently been argued that the right TPJ plays a more specific role in the attribution of mental states (Saxe, 2006). We recently designed two fMRI experiments to test the specific role of the anterior Paracingulate Cortex (aPCC) – part of the MPFC – in ToM tasks (Walter et al., 2004). Results showed that the aPCC is not necessarily involved in the understanding of other people's intentions per se, but primarily, in the understanding of the intentions of people who are actually involved in social interaction, or who are preparing for future social interaction (i.e., when a given social interaction is foreseen, but has not yet occurred). The present paper is aimed at providing useful input to the scientific debate on the key ToM network regions and also introduces a novel theoretical distinction among varieties of intention, a distinction that helps to identify more clearly the role of each of the four brain regions described above. Hence, we extend our earlier results (Walter et al., 2004) by proposing a new analysis that focuses not only on the aPCC, but also on the precuneus and the TPJ bilaterally. This deeper level of analysis is based on signal time courses for the four regions of interest (ROIs) and on an explicit comparison of TPJ lateralization differences.

Philosophy of mind describes a conceptual difference between prior intention and intention in action. Searle (1983) defines prior intention as an initial representation of the goal of an action prior to initiation of the action itself; this kind of intention is formed in advance. By contrast, an intention in action is the proximal cause of the physiological chain leading to overt behaviour. Typical ToM tasks employed a third person perspective, namely prior intentions are comprehended after an action has been observed (e.g., by observing an agent looking for a book in a bookcase, one can comprehend that the agent intends to read a book). On the contrary, in first person perspective a prior intention logically and temporally precedes the intention in action (e.g., an agent intends to read a book hence looks for it in a bookcase). The purpose of the present paper is to examine how human beings represent other people's prior intentions from the observation of their actions. In particular, we claim that, starting from the specific observation of an action, an individual can represent two types of prior intention: private intentions and social intentions, and that the two differ in terms of the nature of their inherent goals (see Fig. 1).

Private intentions (PInt) involve the representation of a private goal. We define a private goal as one involving only the actor satisfying that particular goal. Conversely, a social intention involves the representation of a social goal. We define a social goal as the goal of an actor (A) that implies at least one other person (B), who is a necessary element for satisfying that goal. Furthermore, in social intention, we can distinguish between present interaction and prospective (future) interaction. When A and B interact, the social intention is shared at that moment, i.e., in the present. The prototypical example of a social intention shared in the present is a communicative intention (CInt), i.e., the intention to communicate a given meaning to someone else, plus the intention that this intention should be recognized by the addressee (Bara, 2007, Grice, 1975). However, there are social intentions in which the social goals lie in the future. This kind of social intention involves the representation of a social goal when A and B are not actually interacting but B is part of A's goal, i.e., when a given social interaction is not present at the moment but the social intention is potentially shared in the future. We define this type of social intention as prospective social intention (PSInt). In order to emphasize the temporal dimension of this kind of social intention, in our experimental protocol we employed scenarios where only one agent (A) was present, but she was preparing to interact with B, who was never present in the scenario (see Appendix A).

Note that these two types of social intentions (CInt and PSInt) are not mutually exclusive. There might be actual social interactions without communication, e.g., two people dancing together, as well as communicative intentions directed at the future, e.g., writing a letter. In literature, rich descriptions of different kinds of intentions were proposed (for a deeper analysis see Becchio, Adenzato, & Bara, 2006; Jeannerod & Pacherie, 2004; Pacherie, 2000, Pacherie, 2006). However, here we have investigated a single category of social intention shared in the present: communicative intention. The other kinds of social intentions (such as, for example, joint intentions) are not the focus of the present study.

Based on these conceptual distinctions, we implemented an experimental protocol that was designed to investigate the role of each area of the ToM network in understanding other people's prior intentions. In our experiment, participants were asked to read short comic strips and then choose a picture that showed the only logical ending to the story, a procedure that induced participants to take the third-person perspective.

Comic strips pertained to the following experimental categories:

  • (i)

    PInt: Private intention. Participants in this condition represented another person's intention, based on observation of that person's isolated action, e.g., observing a single person (A) changing a broken bulb in order to read a book;

  • (ii)

    PSInt: Prospective social intention (potentially shared in the future). Participants in this condition represented another person's intention to socially interact with someone else in the future, based on the observation of that person's isolated action, e.g., observing a single person (A) preparing a romantic dinner for another person (B), who is not yet present in the scenario;

  • (iii)

    CInt: Communicative intention (shared in the present). Participants in this condition represented the intentions to communicate based on the observation of two people interacting, e.g., observing a person (A) obtaining a glass of water by asking another person (B) to get it for her.

The control condition was physical causality (Ph-C), in which participants represented non-intentional causal links among objects, e.g., a ball blown by a gust of wind knocking over and breaking a glass of water. For a complete description of the experimental conditions see Appendix A.

We hypothesised that the apparently contradictory findings concerning the priority of a key region (i.e., the MPFC or the TPJ) in ToM tasks could be resolved by taking the conceptual differences between different types of prior intention (private and social) into account (see Fig. 1). Therefore, we propose a model of a dynamic network consisting of four brain regions, each of which has its own specific function, depending on what type of prior intention is represented from an observed action. The entire network encompasses the standard ToM regions, i.e., the right and left TPJs, the precuneus and the aPCC. In particular, we proposed that the ToM network becomes more extensively activated whenever people are trying to comprehend social intentions, i.e., in PSInt and CInt conditions. Finally, we aimed at investigating lateralization differences in TPJ activation, in relation to the conceptual categories outlined above.

Section snippets

Participants

Twelve right-handed volunteers (six females; age range = 19–27; M = 24.75; S.D. = 2.63) with no history of medical or neurological illness were recruited. All participants gave written informed consent. The study was approved by the local ethics committee.

Experimental design

A detailed description of our experimental design is available in Walter et al. (2004) and a similar experimental design was utilized by Brunet et al. (2000) and Langdon et al. (1997). Briefly, we presented comic strips consisting of a sequence of

Behavioural results

For each of the four conditions, response accuracies (number of correct answers, maximum score = 11) and reaction times in ms (for correct answers only) with their standard deviations, were as follows: Ph-C 10.58 ms (±0.669) and 2863 ms (±600), PInt 9.83 ms (±0.577) and 3164 ms (±543), PSInt 10 ms (±1.045) and 3392 ms (±425), and CInt 9.83 ms (±0.937) and 2908 ms (±403). There was no significant condition effect on response accuracy, F(3, 44) = 2.212, p = 0.100, and although a condition effect on reaction

Discussion

The present work examined the relative contribution of four key ToM network regions (the aPCC, precuneus, right TPJ, and left TPJ) in the comprehension of different types of prior intention. Based on our results, we propose a neurocognitive framework for the human ability to read other peoples intentions.

The most important result was that the ToM network showed different activation patterns in relation to the nature of the intentions participants were dealing with. Only the comprehension of a

Conclusions

Human beings conceive conspecifics as social agents. This attitude implies the explanation and the prediction of intentions underlying the actions of others. Without this competence, other people's behaviour would be meaningless from a third person perspective: behaviour would be observed, but the meaning of actions would not be understood. Researchers have proposed that two different key brain regions underpin this competence: the aPCC as a part of the MPFC (Amodio & Frith, 2006) and the right

Acknowledgments

We would like to thank Cristina Becchio, Claudia Chiavarino, Ana Solodokin, and two anonymous reviewers for valuable comments to an early version of the manuscript. This work was supported by MIUR of Italy (cofin 2005, protocol no. 2005119758_004) and by Regione Piemonte (Bando regionale per la ricerca scientifica 2004, cod. A239).

References (40)

  • M. MacSweeney et al.

    Dissociating linguistic and nonlinguistic gestural communication in the brain

    NeuroImage

    (2004)
  • R. Saxe et al.

    People thinking about thinking people. The role of the temporo-parietal junction in “theory of mind”

    Neuroimage

    (2003)
  • R. Saxe et al.

    A region of right posterior superior temporal sulcus responds to observed intentional actions

    Neuropsychologia

    (2004)
  • R. Saxe et al.

    Making sense of another mind: the role of the right temporoparietal junction

    Neuropsychologia

    (2005)
  • R. Saxe

    Uniquely human social cognition

    Current Opinion of Neurobiology

    (2006)
  • B. Seltzer et al.

    Afferent cortical connections and architectonics of the superior temporal sulcus and surrounding cortex in the rhesus monkey

    Brain Research

    (1978)
  • L. Wang et al.

    Pantomine, praxis, and aphasia

    Brain and Language

    (1992)
  • D.M. Amodio et al.

    Meeting of minds: The medial frontal cortex and social cognition

    Nature Review Neuroscience

    (2006)
  • B.G. Bara

    Cognitive pragmatics

    (2007)
  • H. Barbas et al.

    Architecture and intrinsic connections of the prefrontal cortex in the rhesus monkey

    Journal of Comparative Neurolology

    (1989)
  • Cited by (0)

    View full text