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

https://doi.org/10.1016/j.neubiorev.2009.12.004Get rights and content

Abstract

Several studies have investigated the neural correlates of self-reflection. In the paradigm most commonly used to address this concept, a subject is presented with trait adjectives or sentences and asked whether they describe him or her. Functional neuroimaging research has revealed a set of regions known as Cortical Midline Structures (CMS) appearing to be critically involved in self-reflection processes. Furthermore, it has been shown that patients suffering damage to the CMS, have difficulties in properly evaluating the problems they encounter and often overestimate their capacities and performance. Building on previous work, a meta-analysis of published fMRI and PET studies on self-reflection was conducted. The results showed that two areas within the medial prefrontal cortex (MPFC) are important in reflective processing, namely the ventral (v) and dorsal (d) MPFC. In this paper a model is proposed in which the vMPFC is responsible for tagging information relevant for ‘self’, whereas the dMPFC is responsible for evaluation and decision-making processes in self- and other-referential processing. Finally, implications of the model for schizophrenia and lack of insight are noted.

Introduction

The aim of this paper is to review the literature on self-reflection, and in particular its differentiation from other-reflective processing, by means of a meta-analysis of the fMRI and PET studies published so far on the subject. We will present a model of self-reflection and the brain based on the results of the meta-analysis. The discussion will be expanded to encompass the literature on failure of self-reflection processes, in particular in schizophrenia patients who may be viewed as having important problems in this area (Carter et al., 2001, Baird et al., 2006, Addington et al., 2006, Sprong et al., 2007). Recent neuroimaging evidence (Taylor et al., 2007, Park et al., 2008) suggests that processes relying upon these Cortical Midline Structures (CMS), a set of regions encompassing the posterior cingulate cortex (PCC), the medial prefrontal cortex (MPFC) and the anterior cingulate cortex (Northoff and Bermpohl, 2004), may be affected in this patient group. Finally, the possible role for self-reflection abilities in illness awareness in schizophrenia patients will be discussed.

The study of self has become increasingly popular in cognitive neuroscience over the past decade (Kircher and David, 2003). Several studies have investigated the neural correlates of self-reflection or self-referential processing. In the literature these terms are used interchangably and refer to the evaluation process used to decide whether certain environmental cues apply to one's self or not. Technically, self-referential processing is a broader concept in which all information that somehow refers to oneself is processed and encompasses subconscious as well as conscious processing. Self-reflective processing on the other hand implies a conscious process in which a decision is made regarding oneself.

Having an accurate representation of one's traits, abilities and attitudes is important in evaluating one's own behavior and comparing it with the behavior of other human beings. The most commonly used paradigm to address this concept in experimental and neuroscientific research uses self-reflection in which subjects are presented with trait adjectives or sentences and are asked whether the trait or sentence applies to them. Results have consistently pointed to a role of the CMS in these self-reflection processes, but this has not allayed misgivings regarding the concept of self-reflective processing (Gillihan and Farah, 2005) and whether the processing of self-reflective information is substantially different from the processing of information concerning other people. Finally, it has also been shown that patients who have suffered damage to the CMS have difficulties in properly evaluating their problems and often overestimate their capacities as well as their performance particularly on cognitively demanding operations (Schmitz et al., 2006).

Most studies investigating self-reflection processes have found evidence for medial prefrontal cortex (MPFC), posterior cingulate cortex (PCC) and anterior cingulate cortex (ACC) involvement in distinguishing self-related information from non-self-related information. Even though these studies report a similar functional anatomy, the precise involvement of the component structures is debated. One of the first studies looking at the neural correlates of the ‘self-reference effect’, using PET was done by Craik et al. (1999). The self-reference effect refers to the finding that people tend to remember words when processed in relation to themselves better than words processed more generally (see Symons and Johnson (1997) for a meta-analysis). Craik et al. (1999) found that the retrieval of self-referential information is mediated by right prefrontal areas including the MPFC, whereas the encoding of such information is similar to the encoding of information about others and is mediated mainly by left prefrontal areas. This effect was replicated using fMRI by Kelley et al. (2002), by means of visual presentation of trait adjectives and Johnson et al. (2002) who studied self-reflection by means of a paradigm in which subjects were presented auditorily with short questions, each entailing a trait, attitude or ability (e.g. ‘I am a good friend’). As a control condition, they used simple questions entailing general semantic knowledge (e.g. ‘You need water to live’). Both studies found anterior MPFC (aMPFC) and PCC activation in the self-reflection condition. The studies by Craik et al. (1999), Kelley et al. (2002) and Johnson et al. (2002) were followed up by a number of other studies using similar paradigms and reporting similar areas of activation. Macrae et al. (2004) demonstrated that activation in MPFC regions corresponded to self-reflective judgments and memory performance related to self-descriptive trait words. Fossati et al. (2003) were specifically interested in the processing of emotionally valenced words in self-reflection. They presented positive and negative trait words and found dorsal MPFC (dMPFC) and PCC activation in a self versus baseline contrast.4 Interestingly, this dMPFC activation was not specific to either positive or negative stimuli but rather it was present regardless of valence. Gusnard et al. (2001) and Johnson et al. (2005) found only MPFC activation in a similar contrast in which subjects were asked to introspect either in response to pleasant/unpleasant visual stimuli or color preference respectively. Johnson et al. (2006) similarly used an introspection paradigm in which subjects ruminated on hopes and duties in comparison with a condition without self-reference and found MPFC and PCC activation.

Northoff and Bermpohl (2004) reviewed the literature on self-processing and neuroimaging and discussed the role of the CMS in self-processing. They discussed each area separately and came to the conclusion that different areas within the CMS represent different functions, such as representation, monitoring, evaluation and integration. However, their review focused on the processing of self only. Many other recent studies have included an ‘other’ condition in which the subject is asked to reflect upon another person, either an unfamiliar person, a relative, close friend or someone famous, while presented with trait words (Kelley et al., 2002, Macrae et al., 2004, Schmitz et al., 2004, Ochsner et al., 2005, Heatherton et al., 2006, Schmitz and Johnson, 2006, D’Argembeau et al., 2007, Gutchess et al., 2007, Zhu et al., 2007), trait sentences (Pfeifer et al., 2007, Modinos et al., 2009), or when instructed to introspect upon emotional pictures (Ochsner et al., 2004, Jenkins et al., 2008) or food preference (Seger et al., 2004). The involvement of the MPFC (Macrae et al., 2004, Ochsner et al., 2004, Schmitz et al., 2004, Johnson et al., 2005, Schmitz and Johnson, 2006, Zhu et al., 2007, Modinos et al., 2009) and PCC, precuneus (Ochsner et al., 2004, Seger et al., 2004, Heatherton et al., 2006, Johnson et al., 2006, Schmitz and Johnson, 2006, Modinos et al., 2009) in self-reflective processing is broadly confirmed. However, regarding ‘other’-reflective processing and the extent to which this differs from self-reflective processing, the literature does not yield a consensus.

Even though most studies report differences in self-processing versus other-processing, the brain region that is mostly reported to be functionally associated specifically with self-processing is referred to as MPFC (Kelley et al., 2002, Ochsner et al., 2004, Heatherton et al., 2006, D’Argembeau et al., 2007, Gutchess et al., 2007, Pfeifer et al., 2007, Zhu et al., 2007, Jenkins et al., 2008, Modinos et al., 2009), while the same studies report this area to be involved in other-processing as well. Additional areas that are reported to be involved specifically in self-processing are the anterior cingulate cortex (Ochsner et al., 2005, Heatherton et al., 2006, D’Argembeau et al., 2007, Zhu et al., 2007, Modinos et al., 2009), dorsolateral PFC (DLPFC) (Schmitz et al., 2004), superior parietal regions (Seger et al., 2004) and lateral temporal regions (Kjaer et al., 2002, Lou et al., 2004). The observation that the involvement of the MPFC is reported for self as well as other-processing indicates that there is a need to be more specific regarding the areas within the MPFC that are involved in either self- or other-processing or both.

Some studies suggest that a functional distinction should be made between the ventral MPFC (vMPFC) and the dMPFC (Frith and Frith, 1999, Gusnard et al., 2001, Northoff and Bermpohl, 2004, Northoff et al., 2006, Schmitz and Johnson, 2007) in which the dMPFC might process the non-emotional, and the vMPFC the more emotional content of the information. However, most studies mentioned above do not make this explicit distinction. Hence we still do not have a definitive answer to the question formulated by Gillihan and Farah (2005), namely, does the processing of self-referential information in the brain indeed differ as compared to the processing of other-referential processing? And if so, which brain areas are functionally involved in these processes?

To this end, a meta-analysis was conducted to integrate and extend the published findings. The aim was to get a more objective and quantitative picture of which regions are involved in self-referential processing specifically and those which are implicated in other-referential processing. The neuroimaging meta-analysis software algorithm that was applied allows visualization and precise localization of foci of activation that are consistent across the majority of studies.

Section snippets

Method

Articles included in the meta-analysis were identified through a literature search using the search terms ‘self AND (reflect* OR referen*) AND (fMRI OR “functional magnetic resonance imaging” OR PET OR “positron emission tomography”)’ in PubMed until November 2008. Furthermore, the references of the selected papers were searched for additional papers on self-reflection and neuroimaging that did not appear in the PubMed search. A total of 20 out of 29 PET or fMRI studies were selected based on

Results

The results of this meta-analysis are presented in Fig. 1 (self > baseline) and 2 (self > other). Table 3 (self > baseline) and 4 (self > other) summarize brain regions that were revealed by this meta-analysis. The self > baseline contrast yielded four clusters (see Table 3). The first large cluster with 58% and 55% of the studies reporting activation in this area included the posterior cingulate/precuneus (BA23/30) and the cuneus (BA 18), respectively. A second large cluster, with a similar 58% and 51%

Cortical midline structures

The areas that were most consistently activated across studies in this meta-analysis largely converge with what one would expect from individual studies that investigated self-reflection. Regarding self-reflection > baseline the integration of the published evidence leaves it beyond doubt that the Cortical Midline Structures (CMS) are associated with self-processing. However, this meta-analysis also pointed to other areas that have not been particularly highlighted before in relation to

Acknowledgements

This work was supported by the European Science Foundation EURYI grant (NWO no. 044035001) to Professor A. Aleman. The authors would like to thank Marieke Pijnenborg for her useful comments on the manuscript. Sergi Costafreda would like to gratefully acknowledge support from the National Institute for Health Research (NIHR) Specialist Biomedical Research Centre for Mental Health award to the South London and Maudsley NHS Foundation Trust and the Institute of Psychiatry, King's College London.

References (107)

  • H.T. Ghashghaei et al.

    Pathways for emotion: interactions of prefrontal and anterior temporal pathways in the amygdala of the rhesus monkey

    Neuroscience

    (2002)
  • B. Goldstein et al.

    The impact of frontal and non-frontal brain tumor lesions on Wisconsin Card Sorting Test performance

    Brain Cogn.

    (2004)
  • B.J. Harrison et al.

    Task-induced deactivation of midline cortical regions in schizophrenia assessed with fMRI

    Schizophr. Res.

    (2007)
  • M.W. Haut et al.

    On the nature of Wisconsin Card Sorting Test performance in schizophrenia

    Psychiatry Res.

    (1996)
  • T.T. Kircher et al.

    Recognizing one's own face

    Cognition

    (2001)
  • T.W. Kjaer et al.

    Reflective self-awareness and conscious states: PET evidence for a common midline parietofrontal core

    NeuroImage

    (2002)
  • F. Krueger et al.

    The medial prefrontal cortex mediates social event knowledge

    Trends Cogn. Sci.

    (2009)
  • R.J. Maddock et al.

    Remembering familiar people: the posterior cingulate cortex and autobiographical memory retrieval

    Neuroscience

    (2001)
  • J.P. Mitchell et al.

    Dissociable medial prefrontal contributions to judgments of similar and dissimilar others

    Neuron

    (2006)
  • G. Northoff et al.

    Cortical midline structures and the self

    Trends Cogn. Sci.

    (2004)
  • G. Northoff et al.

    Self-referential processing in our brain—a meta-analysis of imaging studies on the self

    NeuroImage

    (2006)
  • K.N. Ochsner et al.

    The neural correlates of direct and reflected self-knowledge

    NeuroImage

    (2005)
  • I.H. Park et al.

    Dysfunctional modulation of emotional interference in the medial prefrontal cortex in patients with schizophrenia

    Neurosci. Lett.

    (2008)
  • A. Sapara et al.

    Prefrontal cortex and insight in schizophrenia: a volumetric MRI study

    Schizophr. Res.

    (2007)
  • T.W. Schmitz et al.

    Self-appraisal decisions evoke dissociated dorsal-ventral aMPFC networks

    Neuroimage

    (2006)
  • T.W. Schmitz et al.

    Relevance to self: a brief review and framework of neural systems underlying appraisal

    Neurosci. Biobehav. Rev.

    (2007)
  • T.W. Schmitz et al.

    Metacognitive evaluation, self-relevance, and the right prefrontal cortex

    Neuroimage

    (2004)
  • T.W. Schmitz et al.

    Neural correlates of self-evaluative accuracy after traumatic brain injury

    Neuropsychologia

    (2006)
  • C.A. Seger et al.

    Cortical activations during judgments about the self and an other person

    Neuropsychologia

    (2004)
  • M.U. Shad et al.

    Prefrontal subregions and dimensions of insight in first-episode schizophrenia—a pilot study

    Psychiatry Res.

    (2006)
  • M.U. Shad et al.

    Insight and prefrontal cortex in first-episode Schizophrenia

    Neuroimage

    (2004)
  • M. Startup

    Awareness of own and others’ schizophrenic illness

    Schizophr. Res.

    (1997)
  • S.F. Taylor et al.

    Medial frontal hyperactivity in reality distortion

    Biol. Psychiatry

    (2007)
  • A. Aleman et al.

    Insight in psychosis and neuropsychological function: meta-analysis

    Br. J. Psychiatry

    (2006)
  • X.F. Amador et al.

    Insight and Psychosis: Awareness of Illness in Schizophrenia and Related Disorders, 2

    (2004)
  • N.C. Andreasen et al.

    Remembering the past: two facets of episodic memory explored with positron emission tomography

    Am. J. Psychiatry

    (1995)
  • R.L. Atkinson et al.

    Paired-associate learning by schizophrenic and normal subjects under conditions of personal and impersonal reward and punishment

    J. Abnorm. Soc. Psychol.

    (1961)
  • A. Baird et al.

    Social and emotional functions in three patients with medial frontal lobe damage including the anterior cingulate cortex

    Cogn. Neuropsychiatry

    (2006)
  • A. Bechara et al.

    Deciding advantageously before knowing the advantageous strategy

    Science

    (1997)
  • Y. Benjamini et al.

    Controlling the false discovery rate: a practical and powerful approach to multiple testing

    J. R. Stat. Soc. B

    (1995)
  • B.R. Buchsbaum et al.

    Meta-analysis of neuroimaging studies of the Wisconsin card-sorting task and component processes

    Hum. Brain Mapp.

    (2005)
  • R.L. Buckner et al.

    Frontal cortex contributes to human memory formation

    Nat. Neurosci.

    (1999)
  • S.T. Carmichael et al.

    Sensory and premotor connections of the orbital and medial prefrontal cortex of macaque monkeys

    J. Comp. Neurol.

    (1995)
  • P. Carruthers

    How we know our own minds: the relationship between mindreading and metacognition

    Behav. Brain. Sci.

    (2009)
  • C.S. Carter et al.

    Anterior cingulate cortex activity and impaired self-monitoring of performance in patients with schizophrenia: an event-related fMRI study

    Am. J. Psychiatry

    (2001)
  • R. Corcoran

    Theory of mind in schizophrenia

  • R. Corcoran et al.

    Autobiographical memory and theory of mind: evidence of a relationship in schizophrenia

    Psychol. Med.

    (2003)
  • R. Corcoran et al.

    Thematic reasoning and theory of mind: accounting for social inference difficulties in schizophrenia

    J. Evol. Psychol.

    (2005)
  • S.G. Costafreda

    Pooling fMRI data: meta-analysis, mega-analysis and multi-center studies

    Front. Neuroinform.

    (2009)
  • A.D. Craig

    How do you feel—now? The anterior insula and human awareness

    Nat. Rev. Neurosci.

    (2009)
  • Cited by (433)

    • Autobiographical memory and health in childhood and adolescence

      2023, Encyclopedia of Child and Adolescent Health, First Edition
    View all citing articles on Scopus
    1

    Centre for Neuroimaging Sciences, Institute of Psychiatry, Box P089, De Crespigny Park, London SE5 8AF, United Kingdom. Tel.: +44 0 203 228 3052; fax: +44 0 203 228 2116.

    2

    University Medical Center Groningen, Department of Neurosciences, Cognitive Neuropsychiatry Group, P.O. Box 196, 9700 AD Groningen, The Netherlands. Tel.: +31 0 50 363 5111; fax: +31 0 50 363 8875.

    3

    Section of Cognitive Neuropsychiatry, P.O. Box 68, Institute of Psychiatry, King's College, Denmark Hill, London SE5 8AF, United Kingdom. Tel.: +44 0 20 7 848 0138; fax: +44 0 20 7 848 0572.

    View full text