Elsevier

NeuroImage

Volume 54, Issue 2, 15 January 2011, Pages 1406-1414
NeuroImage

Anatomical and temporal architecture of theory of mind: A MEG insight into the early stages

https://doi.org/10.1016/j.neuroimage.2010.09.015Get rights and content

Abstract

Representation of others' actions and mental states leads to the activation of several brain networks: the mentalizing and the “mirror neuron” systems as well as a “low level” social perception component. However, respective activations of the regions belonging to these networks remain unknown with respect to chronometrical data when static drawing stimuli are presented. To determine anatomical and temporal characteristics of theory of mind processes, magnetic signals were measured in 21 subjects during a validated nonverbal attribution of intentions task. Minimum norm estimation provides chronometric and localization data showing that regions known to be involved in the mentalizing, “mirror neuron” and social perception networks have simultaneous activations between 100 and 700 ms post-stimulus, a period which may be thought as corresponding to early stages of social processes. Among some regions, different profiles as well as modulations regarding experimental conditions suggest functional distinctions between these structures, pleading for a cooperative nature of these networks. While the left temporo-parietal area and superior temporal sulcus seem more specialized in social cues coding, we demonstrate that their right homologues, as well as the right inferior parietal cortex, are preferentially recruited during attribution of intentions stimuli compared to scenarios based on physical causality from 200 to 600 ms.

Research Highlights

► Mental states attribution processes occur as soon as 200 ms post-stimulus onset. ► Right temporo-parietal areas are predominantly involved in attribution of intention. ► Left temporo-parietal areas are predominantly involved in character coding. ► All the regions were recruited during similar time-intervals. ► Attribution of intention relies on distinct but complementary networks activation.

Introduction

Social cognition encompasses the processes contributing to our ability to understand and interact with others. One of the most important and complex strategies used by people to represent and predict others' behavior consists in raising hypotheses on their putative mental states, a capability called “theory of mind” (ToM) or “mentalizing”. It is widely acknowledged that ToM relies on a brain network whose extension has increased with new findings but which is still distinct from other forms of cognitive processes such as executive functions. A large proportion of studies based on mental state attribution paradigms (i.e. based on comparison of conditions with mental states inferences with non-mentalist conditions) reports activations in temporo-parietal junction (TPJ) and the anterior part of the paracingulate cortex (aPCC, also called medial prefrontal cortex), with a notable contribution of the right hemisphere (for reviews, see Brunet-Gouet & Decety, 2006, Frith & Frith, 2006, Van Overwalle, 2009). This network is referred to as the mentalizing system.

Another set of regions is implicated in “recognizing the goal of a perceived action by matching it to a representation in our memory of our own actions” as reviewed by Van Overwalle and Baetens (2009). This network may be called the “mirror neuron” system (MNS) by reference to findings of specific neurons that responded to both the observation of simple hand movements and the initiation of a similar movement in macaque monkeys (Rizzolatti et al., 1996). The human MNS encompasses the inferior parietal sulcus (IPS), inferior parietal lobule (IPL) and inferior frontal cortex (IFC) (Rizzolatti and Craighero, 2004). The MNS has been shown to respond to human agents in action (Montgomery et al., 2007) and to the intention of an observed action (Iacoboni et al., 2005). In the present study, such an anatomical distinction of functional networks will be used and referred to as the “MNS regions”.

Finally, the posterior part of the superior temporal sulcus (pSTS) is also reported to be activated in biological motion perception (see Allison et al., 2000 for a review; Materna et al., 2008) and, especially in the right hemisphere, in intentional action understanding (Pelphrey et al., 2004, Pelphrey & Morris, 2006, Saxe et al., 2004). Interestingly, Proverbio and collaborators found significant activations in the pSTS, as well as in the MNS, even when photographs of human characters were used as stimuli (2009). From these different results, it may be hypothesized that the activation in the MNS regions reflects a low-level processing allowing a mapping of others' behavior on our own action repertoire whereas the mentalizing system involves a high-level inferential mechanism.

However, the contribution of these different systems and their interactions during social cognition remain unclear. Van Overwalle and Baetens (2009) note the lack of “evidence about precisely how these two systems [mentalizing and MNS] may cooperate and inform each other”. Methodologically speaking, the vast majority of neuroimaging studies has explored brain correlates of ToM with either positron emission tomography (PET) or functional MRI (fMRI). These techniques are powerful to localize brain responses and to bring into light the influence of conditions on the amplitude of activations using experimental designs based on cognitive subtraction. However, temporal resolution is limited to seconds and does not allow finding at which stage each brain system intervenes which is a crucial information to the current debate on the functional relations between these systems.

To overcome these limitations, in the present study we propose the use of magnetoencephalography (MEG) which benefits from an excellent .8 ms resolution while maintaining a centimetric precision. It is necessary to note that in ToM domain, no data on the characteristics of MEG signals and the efficient way to adapt paradigms are available yet. In a step by step approach, we chose to use a task previously validated (Brunet et al., 2000) and replicated in neuroimaging conditions (Ciaramidaro et al., 2007, Völlm et al., 2006, Walter et al., 2004) and pathology assessment (Benedetti et al., 2009, Brunet et al., 2003, Walter et al., 2009). These earlier studies showed that this task based on comic-strips, requiring attribution of intentions to human figures, elicited robust hemodynamic and metabolic responses in several regions commonly reported in ToM literature such as the right aPCC, pSTS, temporal pole (TP), orbito-frontal cortex (OFC), bilateral TPJ and right IFC. Furthermore, the procedure used allowed varying the nature of the logical inferences by contrasting sequences with intentions (AI), sequences with human characters involved in physical events (PCCH), and sequences only depicting objects (PCOB). Here, MEG signals were recorded while subjects were presented with a similar material. Minimum norm algorithm was used in order to provide localization information (Baillet et al., 2001).

Having laid theses methodological bases which improve the level at which we can examine brain activity, several questions and hypotheses were raised and implemented experimentally:

  • #1:

    Attribution of intentions (AI) to others, as a core component of ToM, is based on inferences about mental states leading to measurable activations in the mentalizing network, including the TPJ and the aPCC, preferentially in the right hemisphere. AI will elicit stronger activations in these regions compared to stimuli depicting physical causality with human characters (PCCH).

  • #2:

    Does the observation of static and symbolic series of drawings involving attribution of intentions to characters in action elicit activations within all or parts of the MNS regions? As participating to a low-level non-inferential mechanism to represent and understand others, MNS regions, if activated, should precede that of the mentalizing system as defined in #1.

  • #3:

    Mentalization from observation of visual sequences requires fast detection and extraction of social cues related to the presence of human characters. Activity should be found in bilateral pSTS with a cumulative effect of characters and intentions, especially in the right hemisphere. From other authors (Hirai et al., 2003, Jokisch et al., 2005), we expect that the pSTS activation occurs before 300 ms post-stimulus.

Section snippets

Subjects

Twenty-one healthy volunteers (14 males), aged from 21 to 40 years old (mean age: 28; sd: 5.74), right-handed, native French speakers with normal or corrected to normal vision were included. All of them gave their informed consent in agreement with the French Ethical Committee (Comité de Protection des Personnes).

Task

The material consisted of 74 comic-strips each made of four black and white pictures. The images were extracted from previous works (Brunet et al., 2000), cropped and simplified (some

Results

Participants completed the task with a mean accuracy of 91.5%. The accuracy for the AI condition reached 90.2%, while the PCCH and PCOB conditions reached 90.4% and 94.1% respectively. A one-way ANOVA showed an absence of condition effect (F(2;54) = 2.03, p = .14). Mean reaction time for the average of congruous and incongruous trials was 1640 ms in AI, 1523 ms in PCCH and 1464 ms in PCOB condition. Here again, a one-way ANOVA revealed no effect of experimental conditions (AI vs. PCCH vs. PCOB) on

Discussion

Inferring mental states such as desires and intentions to other people is a core ability required for appropriate human communication and fluent navigation into the social world. Several studies have elucidated the anatomical bases of this capacity with either hemodynamic, metabolic measures or neuropsychological investigations. In the present study, we add to this knowledge base by providing insights into the integration between the temporal dynamics and the spatial localization of ToM. MEG

Funding

This work was supported by a grant of Neuropôle (DV), a grant of Fonds d'Etudes et de Recherche du Corps Médical (EBP), and an Institut National de la Santé Et de la Recherche Médicale interface contract (EBG). EA4047, CHV is member of Fondation FondaMental.

Acknowledgments

We are thankful to Bernard Renault, Antoine Ducorps, Denis Schwartz, Lydia Yahia-Cherif, Jean-Didier Lemarechal, Nathalie Georges (MEG Center, Hopital La Pitié Salpêtrière), Amélia Lemoalle, Audrey Angelard (Versailles), Philip Jackson (CIRRIS and CRULRG, Université Laval, Québec) and Pierre-Emmanuel Michon (CIRRIS, Université Laval, Québec), for their help at each stage of this experimental work. The EA 4047 (Université Versailles Saint-Quentin) — Service de Psychiatrie Adulte (Centre

References (47)

Cited by (0)

1

These authors have equally contributed to this work.

View full text