Elsevier

NeuroImage

Volume 183, December 2018, Pages 291-299
NeuroImage

Neural representations of the multidimensional self in the cortical midline structures

https://doi.org/10.1016/j.neuroimage.2018.08.018Get rights and content

Abstract

Self-concept consists of self-identity that distinguishes self from other people and knowledge that describes one's own attributes in different dimensions. Because self-concept plays a fundamental role in individuals' social functioning and mental health, behavioral studies have examined cognitive processes of self-identity and self-knowledge extensively. Nevertheless, how different dimensions of the self-concept are organized in multi-voxel neural patterns remains elusive. Here, we address this issue by employing representational similarity analyses of behavioral/theoretical models of multidimensional self-representation and blood oxygen level dependent responses, recorded using functional MRI, to judgments of personality traits, physical attributes and social roles of oneself, a close (one's mother) other, and a distant (celebrity) other. The multivoxel patterns of neural activities in the medial prefrontal cortex (mPFC) and posterior cingulate cortex (PCC) distinguished representations of the self from both close and distant others, suggesting a specific neural representation of the self-identity; and distinguished different dimensions of person knowledge of oneself, indicating dimension-sensitive neural representation of the self. Moreover, the pattern of PCC activity is more strongly coupled with dimensions of self-knowledge than self-identity. Our findings suggest that multivoxel neural patterns of the cortical midline structures distinguish not only self from others but also discriminate different dimensions of the self.

Introduction

Knowing oneself as a unique entity occurs early in human development and is a prerequisite for an individual's normal social functioning (Rochat, 2003). Self-concept in adults consists of self-identity that distinguishes the self from other people and self-knowledge that describes one's own attributes in different dimensions such as personality traits, physical attributes and social roles (James, 1950). The elaborated construct of self-concept is important for appreciating others' perspectives and feelings and allows successful social interactions (Decety and Sommerville, 2003).

Given the pivotal role of self-concept in human lives, the cognitive and neural mechanisms underlying self-concept have been studied extensively in psychology, neuroscience and psychiatry (Gillihan and Farah, 2005; Legrand and Ruby, 2009; Northoff et al., 2011; Ma and Han, 2010, 2011). Behavioral research has revealed faster responses to one's own face/name than those of others (Ma and Han, 2009, 2010; Tacikowski and Nowicka, 2010) and better memory of self-related items than those related to others (Klein et al., 1989; Lord, 1980; Ma and Han, 2011). Healthy adults also show more accurate and faster responses to geometric shapes associated with oneself than with familiar/unfamiliar others (Sui et al., 2012, 2013). Functional MRI studies evidenced that the cortical midline structures (CMS) including the medial prefrontal cortex (mPFC) and posterior cingulate cortex (PCC) exhibit greater blood oxygen level dependent (BOLD) responses to personality trait judgments of the self than others in the self-referential task (Hu et al., 2016; Kelley et al., 2002; Ma and Han, 2011; 2014a; 2014b; 2014c; Northoff and Bermpohl, 2004). However, other fMRI studies have reported that reflection on close others (e.g., one's mother or spouse) also induces activations in the mPFC and PCC (Denny et al., 2012; Han et al., 2016; Krienen et al., 2010; Murray et al., 2012; Romund et al., 2017; Zhu et al., 2007). In addition, previous fMRI studies employing repetition-suppression paradigm have shown that the mPFC fails to discriminate self-referential thought and mentalizing about a similar other (Jenkins et al., 2008). These findings cast doubt over the notion of unique cognitive/neural representations of the self (Gillihan and Farah, 2005; Legrand and Ruby, 2009).

Moreover, behavioral studies have suggested independent cognitive representations of the self in different aspects (Marsh and Craven, 2006; Roberts and Donahue, 1994), which make self-concept unique in terms of its multidimensional structure and elaborate contents (Symons and Johnson, 1997). Despite the behavioral evidence for multidimensional structure and contents of self-concept (Symons and Johnson, 1997), brain imaging studies showed that the mPFC activity commonly subserved different aspects of self-referential thoughts (such as personality traits and social roles) and failed to show evidence for distinct neural underpinnings of different dimensions of self-knowledge (Jenkins and Mitchell, 2011; Ma et al., 2014a; 2014c; Martinelli et al., 2013; Moran et al., 2011; Sui et al., 2012). Thus it remains unclear how different dimensions of self-knowledge (e.g., personality traits, social roles, and physical attributes) are represented in the brain.

Most of the previous fMRI studies of self-concept depended on the mass-univariate analysis of the magnitude of BOLD responses. However, it has been recognized that the representation of stimuli or mental states can be characterized by spatially distributed patterns of neural activity that reflect neural population encoding of external stimuli or internal mental states (Georgopoulos et al., 1986; Haxby et al., 2001; Kamitani and Tong, 2005). The distributed patterns of neural activity cannot be captured by the mass-univariate analyses that rely on the average-level neural activity. Thus the current work examined spatially distributed patterns of neural activity associated with representations of multidimensional self-referential thoughts by conducting multivoxel pattern analysis (MVPA) of BOLD responses during reflection of different aspects (i.e., personality traits, social roles, and physical attributes) of oneself, one's mother (a close other), and a celebrity (a distant other). MVPA has been used to unveil spatially overlapping but distinct activity patterns for different categories of stimuli or mental processes. We adopted the representational similarity analysis (RSA, Kriegeskorte et al., 2006; Nili et al., 2014) — a novel MVPA approach that examines mental representations in terms of neural pattern similarities and assesses the geometry of representation structures of the self and others — of BOLD responses during reflection of personality traits, social roles, and physical attributes of oneself and others.

To this end, we first tested a group of healthy participants in a modified label-shape matching task where participants responded to different geometric shapes that have been associated with different dimensions of the self and others. Behavioral performances in this task provide well-validated measures of cognitive representations of the self and others (Sui et al., 2012, 2013), and allow us to build an independent behavioral dissimilarity matrix (DM) for assessing the similarity between oneself and others and between different dimensions of person knowledge of oneself/others. If different dimensions of the self are constructed more elaborated than those of other people, we would expect that the behavioral responses to shapes associated with different dimensions would be more dissimilar for oneself than others. We then sought to identify spatially distributed patterns of brain activity that differentiate between self and others and between different dimensions of self-knowledge to reveal the neural representations of multidimensional self in the brain. In light of previous findings, it was hypothesized that multi-voxel neural patterns in the CMS would not only differentiate the representation of self from close and distant others but might also distinguish the representation of different dimensions of the self, and result in discriminate neural patterns of self-reflection on different dimensions.

Section snippets

Participants

Seventy-one college students (47 males; age range, 18–27 years; mean age = 20.86 ± 2.06 years) were recruited in the fMRI experiment as paid volunteers (Supplementary Table 1). Two participants were excluded from data analysis, owing to excessive head movement during scanning, leaving 69 participants (45 males) for fMRI data analysis. An independent sample of 19 participants (9 males; age range, 18–28 years; mean age = 22.74 ± 2.66 years) was recruited for the modified label-shape matching task

Distinct neural patterns related to self-identity and multidimensional self-knowledge

We conducted whole-brain paired t-tests to assess multivoxel patterns of neural activity involved in reflection of person identity that differentiated between the self and others. The results corroborated greater neural-pattern similarity between one's mother and celebrity than between oneself and celebrity in the PCC, precuneus, mPFC and anterior cingulate cortex (ACC) (Fig. 1A; voxel-wise q (FDR) < 0.05). Furthermore, the neural-pattern similarity between one's mother and celebrity was larger

Discussion

The current work investigated neural representations of the multidimensional self by examining the spatially distributed patterns of neural activity to different dimensions of the self and others. Our results revealed multivoxel neural patterns in the CMS characterizing self-identity and multidimensional self-representations. Specifically, the self-identity (especially in the mental dimension) was evident in the multivariate patterns of activity in the mPFC, PCC, and precuneus, which

Conflicts of interest

No conflict of interest was declared.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Projects 31722026; 31771204; 91632118; 31421003; 31661143039); Open Research Fund of the State Key Laboratory of Cognitive Neuroscience, Beijing Normal University; the Fundamental Research Funds for the Central Universities (2016NT05; 2017XTCX04; 2018EYT04); Beijing Municipal Science and Technology Commission (Z151100003915122); startup funding from the State Key Laboratory of Cognitive Neuroscience and Learning,

References (54)

  • P. Rochat

    Five levels of self-awareness as they unfold early in life

    Conscious. Cognit.

    (2003)
  • R. Saxe et al.

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

    Neuroimage

    (2003)
  • E. Svoboda et al.

    The functional neuroanatomy of autobiographical memory: a meta-analysis

    Neuropsychologia

    (2006)
  • P. Tacikowski et al.

    Allocation of attention to self-name and self-face: an ERP study

    Biol. Psychol.

    (2010)
  • N. Tzourio-Mazoyer et al.

    Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain

    Neuroimage

    (2002)
  • L. van der Meer et al.

    Self-reflection and the brain: a theoretical review and meta-analysis of neuroimaging studies with implications for schizophrenia

    Neurosci. Biobehav. Rev.

    (2010)
  • Y. Zhu et al.

    Neural basis of cultural influence on self-representation

    Neuroimage

    (2007)
  • B.T. Denny et al.

    A meta-analysis of functional neuroimaging studies of self-and other judgments reveals a spatial gradient for mentalizing in medial prefrontal cortex

    J. Cognit. Neurosci.

    (2012)
  • A.P. Georgopoulos et al.

    Neuronal population coding of movement direction

    Science

    (1986)
  • S.J. Gillihan et al.

    Is self special? A critical review of evidence from experimental psychology and cognitive neuroscience

    Psychol. Bull.

    (2005)
  • S. Han et al.

    Shared neural representations of self and conjugal family members in Chinese brain

    Culture and Brain

    (2016)
  • J.V. Haxby et al.

    Distributed and overlapping representations of faces and objects in ventral temporal cortex

    Science

    (2001)
  • D. Herold et al.

    Common and distinct networks for self-referential and social stimulus processing in the human brain

    Brain Struct. Funct.

    (2016)
  • W. James

    The Principles of Psychology: in 2 Volumes

    (1950)
  • A.C. Jenkins et al.

    Repetition suppression of ventromedial prefrontal activity during judgments of self and others

    Proc. Natl. Acad. Sci. Unit. States Am.

    (2008)
  • A.C. Jenkins et al.

    Medial prefrontal cortex subserves diverse forms of self-reflection

    Soc. Neurosci.

    (2011)
  • Y. Kamitani et al.

    Decoding the visual and subjective contents of the human brain

    Nat. Neurosci.

    (2005)
  • Cited by (34)

    • Disrupted network switching in euthymic bipolar disorder: Working memory and self-referential paradigms

      2023, Journal of Affective Disorders
      Citation Excerpt :

      Previous research using the same self-processing paradigm have already highlighted that the dmPFC plays a relevant role moderating the posterior DMN functioning (Davey et al., 2016). This interaction may bring PCC self-representations into consciousness depending on the dmPFC interpretation of the task demands (Feng et al., 2018; Leech and Smallwood, 2019). By itself, the dmPFC has been closely related to abstract cognitive processes, including social cognition tasks (Zamani et al., 2022) as well as autobiographical abstract reasoning (D’Argembeau et al., 2014), self-relevant judgments and rumination (Denny et al., 2012; Zhou et al., 2020).

    • Disrupted fronto-parietal network and default-mode network gamma interactions distinguishing suicidal ideation and suicide attempt in depression

      2022, Progress in Neuro-Psychopharmacology and Biological Psychiatry
      Citation Excerpt :

      Our exploratory analysis was consistent with previous studies that abnormal affective network during processing of negative emotional stimuli could cause the impairment to cognitive and emotion processing which may lead to incremental suicide risk (Jeurissen et al., 2014; Liang et al., 2012; Miller et al., 2018). Similarly, DMN have an important role in psychological processes related to suicide behaviors such as controlling the vividness of negative mental imagery, improving self-referential processing (Feng et al., 2018). Decreased DMN activity has been observed in suicide attempts with mood disorders (Minzenberg et al., 2015).

    View all citing articles on Scopus
    1

    C.F. and X.Y. contributed equally to this work.

    View full text