Elsevier

NeuroImage

Volume 42, Issue 1, 1 August 2008, Pages 437-449
NeuroImage

Pre-reflective and reflective self-reference: A spatiotemporal EEG analysis

https://doi.org/10.1016/j.neuroimage.2008.01.060Get rights and content

Functional imaging studies consistently support the role of the medial prefrontal cortex to be a part of a functional network of reflective self-awareness. The current study introduces a new linguistic task which (1) directly compares self-reference and other-reference, and (2) separates pre-reflective from reflective aspects of self-awareness. Twenty-six healthy volunteers evaluated trait adjectives in reference to the self or a close friend while a continuous 30-channel EEG was recorded. Low-resolution brain electromagnetic tomography (LORETA) was used for statistical brain imaging. As expected, the direct comparison of self-reference to other-reference revealed the MPFC to be significant and strongly more active during the self-reference condition. Pre-reflective aspects of self seem to be implemented to a greater degree in the ventral, reflective aspects of the self in dorsal parts of the MPFC. In the pre-reflective self condition, brain areas that receive homeostatic afferents from somatic and visceral sensation of the body such as the insula and the somatosensory cortex were strongly activated as early as 134 ms after stimulus onset. The right inferior parietal lobe seems to be involved in self-referential processing in general.

Introduction

The capacity to reflect on and evaluate oneself and others is an important component of human consciousness. Our daily experiences make us feel that we are someone who is experiencing the world. We commonly refer to this phenomenon by speaking of the ‘self’ (Ghin, 2003). The self is regarded as a multidimensional phenomenon (Gallagher, 2000, Damasio, 1998, Damasio, 2003, James, 1890, Neisser, 1988). Consequently, different perspectives and definitions on the self exist, and therefore, different tasks have been constructed to unravel the neuronal basis of the self. Despite the different material used to induce self-referential processing, many studies point consistently to cortical midline structures as brain regions essential for representing, monitoring, evaluating, and integrating stimuli that are experienced as strongly related to one's own person (Northoff et al., 2006, Gillihan and Farah, 2005). In particular, several fMRI and PET studies corroborated activation of the medial prefrontal cortex (MPFC), while subjects reflected on their own personality traits or evaluated them as opposed to doing the same for another person (Gusnard et al., 2001, Kjaer et al., 2002, Fossati et al., 2003, Johnson et al., 2002, Kelley et al., 2002, Kircher et al., 2002, Schmitz et al., 2004).

According to William James, any discussion of the self needs to include the distinction between the self as object (the “Me”) and the self as subject (the “I”) of experience (James, 1890, Northoff et al., 2006). Legrand (2003) states that there is an “obvious but crucial difference between being reflectively conscious of one's own perspective on the world and being aware of the world from one's own perspective”. The subjective first-person perspective (FPP) is thought to be needed to interact with and reflect on objects or other persons. This aspect of consciousness is thought to occur pre-reflectively and is the basis upon which any other type of self-consciousness evolves. In contrast, many imaging studies on evaluation of trait adjectives focus on the self as object, in other words, on the self evaluating itself from the third person perspective (TPP). However, it is possible that while being in the TPP, reflecting on one's own traits, processes related to FPP may co-occur. That is, the mental flow, when engaged in higher level self-consciousness may involve a rapid change from the pre-reflective FPP to the reflective TPP. This difference on the phenomenological or cognitive level must be related to specific neuronal activation.

Gallagher (2000) defines a pre-reflective aspect of the self which he calls ‘minimal self’. This minimal self enables us to perceive ourselves as the immediate subject of experience, unextended in time. In the verbal domain, this corresponds to the use of the personal pronoun “I” (Bermúdez, 1998). Gallagher, referring to Shoemaker (1984), suggests that the access to oneself in FPP is immediate and non-observational; that is, it does not involve a perceptual or reflective act of consciousness. In this sense, the immediate self that is referred to by saying or reading “I” is the pre-reflective point of origin for action, experience and thought. Besides this pre-reflective self, many authors assume a reflective self, which Gallagher denotes as ‘narrative self’. The narrative self comprises all aspects of our personality and it allows us to understand ourselves in the continuity of time. Gallagher's conception of the narrative self is akin to Damasio's (1999) ‘autobiographical self’ while his definition of the minimal self seems to be related to the “physical self” of William James (1890) or Damasio's “proto” or “core self”. A description of the different characteristics of the pre-reflective and reflective self is presented in Table 1 (for more information about self as a subject or object see also Legrand, 2007).

In short, the pre-reflective and reflective self are two aspects of the self that might interact, but still are very distinct. Both are needed to make an individual feel her- or himself as a unique personality. While the pre-reflective (minimal) self might also be present in lower species (as they might feel somehow existing), the reflective self can only exist in at least partially self-conscious species, i.e. perhaps in some primates, if not only in humans above very early childhood.

We are aware of the fact that our considerations about the self and its different aspects are somewhat “philosophical”. Indeed, there is a long-standing discussion within the philosophical but also psychological literature whether a difference between these distinct view-points (early emphasized by William James) really exists. Yet, the literature strongly distinguishes the psychological processes associated with the perception of stimuli representing the self as subject or as object as it can also be experienced by ourselves. Each reflection and cognition related to the perception of the self as subject changes the perspective of the self immediately to the one of an object: as soon as the “I” reflects about itself it changes to “me”. In other words, we cannot contemplate our “self“ as subject but only as an object, since the perspectives of subject and object exclude each other. This is the theoretical assumption on which this study is based.

We are aware of the fact that there are currently no specific data available which are exactly supporting our idea. However, our intention was to design a new experimental approach which would allow us to study the old questions about different aspects of self awareness in the context of modern cognitive neuroscience. We intended to exclusively focus on the true separation between the perception of the self as subject or object.

In this study we will concentrate on these two different aspects of the self, the pre-reflective and the reflective self, which are both thought to operate at different time points and to be controlled by different neural networks. Our subjects viewed sequentially presented personal pronouns and trait adjectives in three-word sentences under self-reference and other-reference conditions. When reading the personal pronouns, subjects performed a pre-reflective, minimal reference to the self, and when reading the trait adjectives, they performed a reflective, narrative self-reference.

Using conventional brain imaging methods such as fMRI or PET, it would be nearly impossible to disentangle these different aspects, especially the different time course, because of the rather coarse time resolution of these methods. Therefore, we designed a high time resolution EEG-experiment in combination with functional electric brain tomography, allowing the calculation of the intracerebral sources of EEG activity in the millisecond range. Most fMRI experiments published so far trying to disentangle cortical activations to the perception of the self have not distinguished between the two perspectives (the self as object or the self as subject). There are only few fMRI studies (e.g., Ochsner et al., 2005) taking the same approach. However, the new aspect of our study is to continue the research endeavor to distinguish these two perspectives by use of electrical tomography, in order to enhance the time resolution of cortical mapping. Our hypothesis is that by use of this technique, neural activations associated with the processing of the self as subject or object will occur at different time epochs and brain areas.

Based on previous brain imaging experiments, we hypothesize that the MPFC will be the core region for processes associated with the control and generation of the reflective self. With respect to the pre-reflective self, we hypothesize that the insula and somatosensory areas will be involved, reminiscent to brain areas which are engaged in the control of the proto-self as proposed by Damasio (Damasio, 2003).

The aim of this paper was to study the dynamic activation of brain areas involved in pre-reflective and reflective self-reference. When talking about self-reference, we are assuming a self capable of linguistic communication—i.e., the person is capable of using the first-person pronoun (Gallagher, 2000). But because the mechanisms of the first-person perspective and the reference to a minimal self remain largely pre-reflective and not accessible to language, the concept seems to lie between what Gillihan and Farah (2005) call “physical self” and “psychological self”. Therefore, we hypothesized that the MPFC would be activated in both self-reference induced by personal pronouns and trait adjectives. In particular, we suggested that dorsal and lateral parts of this brain area, which constitute the more cognitive part of the prefrontal cortex, would be activated in the trait-adjective condition and more ventral parts in the personal pronoun condition (Damasio, 2003, Damasio, 1999, Ochsner et al., 2005).

Since we are using a brain imaging method, which allows cortical mapping of psychological functions in the milliseconds range, we are able to examine the dynamic evolvement of cortical activations in the context of self-consciousness. Based on Damasio's theory of what he calls a proto-self, we hypothesize that the brain activations during the generation of the pre-reflective self will antecede those activations accompanying the generation of the reflective self.

Section snippets

Subjects

Twenty-six healthy volunteers (13 females, 13 males, mean age 25, SD 2.8), took part in the study and gave their written informed consent. All subjects were consistently right-handed (Annett handedness questionnaire, Annett, 1970) and native speakers of Swiss German. None of the subjects reported any neurological or psychiatric disease, prior head trauma, sensory impairment or subjective cognitive impairment. Besides these exclusion criteria, subjects were required not to consume alcohol, drugs

Results

Fig. 1, Fig. 2 show the basic ERP data used in the analyses.

  • (1) Time-average LORETA analysis: In both the pre-reflective and the reflective self-conditions, self-reference induced consistently stronger activation than other-reference. Fig. 3, Fig. 4 reveal the qualitative similarity of the activation in the two self-referential tasks: Fig. 3 illustrates the LORETA image for the pre-reflective self condition, and Fig. 4 for the reflective self condition.

  • In the pre-reflective self condition, SR

Discussion

The results of the present study shed new light on the time course and the different neuronal networks involved in the implementation of the pre-reflective and the reflective self reference. Both are implemented at least partially in distinct brain areas. During the entire first 700 ms after stimulus presentation, activation in the pre-reflective self condition was located in the left insula, the dorsal and ventral MPFC, and left precentral gyrus. During the reflective self condition,

Conclusion

This study replicates and extends, in an EEG-LORETA design, previous findings of the MPFC as being crucially involved in self-referential processing. Additionally, we show that pre-reflective and reflective aspects of the self recruit different neuronal substrates, and we therefore suggest that in forthcoming studies a careful analysis of the phenomenal requirements of self-referential tasks be performed. Furthermore, our data strongly supports the involvement of structures that do not lie on

Acknowledgment

This work is partly funded under the European Union FET project PRESENCCIA (Contract Number 27731).

References (60)

  • LegrandD.

    Subjectivity and the body: Introducing basic forms of self-consciousness

    Conscious. Cogn.

    (2007)
  • LegrandD.

    How not to find the neural signature of self-consciousness

    Conscious. Cogn.

    (2003)
  • NorthoffG. et al.

    Cortical midline structures and the self

    Trends Cogn. Sci.

    (2004)
  • NorthoffG. et al.

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

    NeuroImage

    (2006)
  • OchsnerK.N. et al.

    The neural correlates of direct and reflected self-knowledge

    NeuroImage

    (2005)
  • OfekE. et al.

    Neurophysiological correlates of subjective significance

    Clin. Neurophysiol.

    (2005)
  • PankseppJ.

    Affective consciousness: Core emotional feelings in animals and humans

    Conscious. Cogn.

    (2005)
  • Pascual-MarquiR.D. et al.

    Low resolution electromagnetic tomography: a new method for localizing electrical activity in the brain

    Int. J. Psychophysiol.

    (1994)
  • PhanK.L. et al.

    Neural correlates of individual ratings of emotional salience: a trial-related fMRI study

    NeuroImage

    (2004)
  • PhanK.L. et al.

    Functional neuroanatomy of emotion: a meta-analysis of emotion activation studies in PET and fMRI

    NeuroImage

    (2002)
  • PhillipsC. et al.

    Anatomically informed basis functions for EEG source localization: combining functional and anatomical constraints

    NeuroImage

    (2002)
  • SchmitzT.W. et al.

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

    NeuroImage

    (2004)
  • StrikW.K. et al.

    Three-dimensional tomography of event-related potentials during response inhibition: evidence for phasic frontal lobe activation

    Electroencephalogr. Clin. Neurophysiol.

    (1998)
  • TowleV.L. et al.

    The spatial location of EEG electrodes: locating the best-fitting sphere relative to cortical anatomy

    Electroencephalogr. Clin. Neurophysiol.

    (1993)
  • VogeleyK. et al.

    Mind reading: neural mechanisms of theory of mind and self-perspective

    NeuroImage

    (2001)
  • AnnettM.

    A classification of hand preference by association analysis

    Br. J. Psychol.

    (1970)
  • BermúdezJ.L.

    The Paradox of Self-Consciousness

    (1998)
  • BlankeO. et al.

    The out-of-body experience: disturbed self-processing at the temporo-parietal junction

    Neurosci.

    (2005)
  • BlankeO. et al.

    Stimulating illusory own-body perceptions

    Nature

    (2002)
  • CraigA.D.

    How do you feel? Interoception: the sense of the physiological condition of the body

    Nat. Rev., Neurosci.

    (2002)
  • Cited by (72)

    • Cortical activity during social acceptance and rejection task in social anxiety disorder: A controlled functional near infrared spectroscopy study

      2021, Progress in Neuro-Psychopharmacology and Biological Psychiatry
      Citation Excerpt :

      Frontopolar activity was associated with pessimistic future thinking (Katayama et al. 2019) and may also play a role in the integration of one's preferences with those of others (Raposo et al. 2011). Additionally, in one study activity contrasts in the frontal pole were found during self-referential compared to other-referential thinking (Esslen et al. 2008), a bias which is included in cognitive models of SAD particularly during social threat (Hirsch et al. 2003). Our findings are also in line with some cognitive models in SAD implicating the role of attentional biases.

    • Immediate effects of visual–motor illusion on resting-state functional connectivity

      2020, Brain and Cognition
      Citation Excerpt :

      In our study, the RSFC increased between the left DLPFC and the left PMC, between the right PMC and the right Pa, between the left PMC and left Pa, and between the left Pa. Our results was in agreement with previous studies reporting on networks related to a sense of body ownership (Arizono et al., 2016; Ehrsson et al., 2004; Ehrsson et al., 2005; Esslen et al., 2008; Gentile et al., 2011, Lenggenhager et al., 2011; Kanayama et al., 2009; Zeller et al., 2011). Based on these, our results suggest that RSFC reflects change in the sense of body ownership by the VMI.

    View all citing articles on Scopus
    1

    The first and second authors have contributed equally to this publication.

    View full text