Associations and dissociations between default and self-reference networks in the human brain
Research Highlights
►Functional associations of self reference and default networks. ►Functional dissociations of self reference and default networks. ►Differential functional connectivity of self reference and default networks.
Introduction
Two independent lines of research have implicated anterior and posterior cortical midline regions, specifically medial prefrontal cortex (MPFC) and posterior cingulate extending to precuneus (PCC), in self-related cognition. In one line of research, tasks that permit or encourage explicit self-referential processing have consistently yielded activation in MPFC and PCC (i.e., a self-reference network) (d'Argembeau et al., 2005, Johnson et al., 2002, Kelley et al., 2002, Moran et al., 2006, Northoff and Bermpohl, 2004, Northoff et al., 2006). A second line of research has identified MPFC and PCC as brain regions in which activation is greater during rest than during engagement in a broad range of goal-directed tasks (e.g., Binder et al., 1999, Fox et al., 2005, Greicius et al., 2003, Greicius and Menon, 2004, Gusnard and Raichle, 2001, Mason et al., 2007, Mazoyer et al., 2001, McKiernan et al., 2003, McKiernan et al., 2006, Raichle et al., 2001, Shulman et al., 1997). This network is thought to mediate a “default mode of brain function” that may support self-reflection about internal thoughts and feelings in the absence of external stimulus processing (Gusnard and Raichle, 2001, Raichle et al., 2001). The fact that similar regions are activated during explicit self-reference and default self-reflection has often been taken as convergent evidence for the roles of MPFC and PCC in self-related thought, and there is evidence that a common ventral MPFC region may be activated in both sorts of self-related thoughts (d'Argembeau et al., 2005, Gusnard et al., 2001). d'Argembeau et al. (2005) used positron emission tomography (PET) to examine similarities and differences between unconstrained self-reference (participants were instructed to think about themselves) and rest. Surprisingly, however, the precise relations between the explicit self-reference and default self-reflection networks have not been delineated using fMRI because no study has compared both kinds of self-reference against a common control condition in a single group of participants. Here, in order to examine directly the relations between these two functional networks, we examined similarities and differences between explicit self-reference and rest conditions relative to a common baseline, within subjects, and in two independent cohorts so as to evaluate the reliability of any findings.
Identification of the neural network mediating self-reference has most often been examined via tasks that require participants to make explicit judgments about themselves. Self-reference studies contrast conditions in which participants judge whether trait adjectives (such as POLITE or TIMID) describe themselves versus conditions in which the same kinds of trait adjectives are judged as describing another person or judged semantically as being positive or negative in valence (Craik et al., 1999, Kelley et al., 2002, Kjaer et al., 2002, Heatherton et al., 2006). Both MPFC and PCC regularly exhibit greater activation during self-referential judgments relative to other-referential or valence judgments (Heatherton et al., 2006, Kelley et al., 2002, Johnson et al., 2002, Fossati et al., 2003; but see Ochsner et al., 2005, Mitchell et al., 2006). Further, MPFC is activated during passive viewing of self-relevant information (Moran et al., 2009), increases in activation linearly with the degree to which a trait is judged self-descriptive (Moran et al., 2006), and is greater at both encoding (Macrae et al., 2004) and retrieval (Fossati et al., 2004) for traits remembered versus forgotten when encoded with reference to the self. Activation in PCC is also frequently observed in self-reference tasks (Kelley et al., 2002, Kjaer et al., 2002, Lou et al., 2004, Moran et al., 2006), although it has not been consistently activated by the degree of self-relevance of items in studies that investigated this (Macrae et al., 2004, Moran et al., 2006, Moran et al., 2009, Phan et al., 2004). Additionally, transient disruption of PCC function by transcranial magnetic stimulation (TMS) reduces the recall of items that were previously encoded with reference to the self (Lou et al., 2004).
Within the MPFC there is evidence for a functional dissociation between dorsal and ventral regions. A factor analysis of 27 imaging studies of self-reference concluded that there were three statistically independent clusters of activations across studies in ventral MPFC, dorsal MPFC, and PCC (Northoff et al., 2006). It has been proposed that dorsal MPFC is involved in the process of evaluating whether stimuli are self-referential, and ventral MPFC is involved in representing self-referential stimuli (Northoff and Bermpohl, 2004). Two studies have further elucidated the nature of the dissociation between ventral and dorsal MPFC. In one study, judgments about people who held views similar to participants yielded ventral MPFC activation, whereas judgments about people who held views dissimilar to participants yielded dorsal MPFC activation (Mitchell et al., 2006). Ventral MPFC is connected anatomically to striatal, limbic, and midbrain regions related to emotions (Northoff et al., 2006), and people may feel more emotional sympathy with a similar individual. Dorsal MPFC is connected to lateral prefrontal cortex (Northoff et al., 2006), and people may analyze dissimilar individuals in a more cognitive or analytic fashion (Mitchell et al., 2006). In a second study, dorsal MPFC was engaged when participants thought about whether trait characteristics described a person, both themselves and others, whereas ventral MPFC was engaged when participants encountered information that was particularly self-relevant (Moran et al., 2010). Thus, it appears that dorsal MPFC subserves the analysis of character in people in general, whereas the ventral MPFC subserves the analysis of personally relevant information.
The most consistent activations associated with the default mode of brain function also occur in MPFC and PCC/precuneus midline regions during rest relative to tasks, but interpretation of the psychological processes signified by these activations is difficult because participants are not performing any task at rest. Gusnard and Raichle (2001) investigated default-mode processes by comparing rest with tasks in which participants viewed scenes and made either external (non self-relevant) indoor/outdoor judgments or internal (self-relevant) pleasant/unpleasant judgments. Activation in MPFC was greater during both rest and internal judgments relative to external judgments. This suggests that MPFC activation during rest reflects self-relevant thoughts. Further evidence along these lines comes from a study that compared activation during well-practiced working memory tasks relative to novel tasks (Mason et al., 2007). This contrast revealed greater activation in both MPFC and PCC, which further linearly increased with an increasing tendency to mind-wander or daydream. This suggests that performing practice tasks allows the mind to wander towards processing information that is internal in nature and hence self-referential. In addition, the amount of self-referential thoughts reported by subjects during unrelated cognitive tasks correlated with activation in MPFC and PCC (McKiernan et al., 2003, McKiernan et al., 2006).
Although there are clear-cut similarities between the brain regions activated by explicit self-reference tasks and by the default mode of brain function during rest, it is unknown whether these activations identify a unitary self-referential neural system, or whether there are brain regions that are differentially engaged by explicit self-reference versus rest. We aimed to address this open question by comparing activations associated with explicit self-reference and rest in a single group of participants (Experiment 1), and then to assess the reliability of any findings in a second group of participants (Experiment 2). In both experiments there were three conditions: (1) explicit self-reference during a task in which participants decided whether or not trait adjectives applied to them, (2) rest, and (3) a control condition in which participants decided whether trait adjectives were positive or negative in valence. The critical question was to what extent the self-reference and rest conditions, relative to the valence condition, yielded similar and dissimilar activations.
In order to further dissociate midline brain regions, we investigated their differential resting state networks using seed-driven resting state functional connectivity analyses (rs-fcMRI). Spontaneous, very low-frequency fluctuations (< 0.1 Hz) in fMRI BOLD signal reveal temporal correlations between brain regions that appear to define functional networks of the human brain (Biswal et al., 1995, Fox et al., 2005). Here we asked whether functional dissociations between self-reference and default-mode activations would be further supported by convergent rs-fcMRI dissociations.
Section snippets
Participants
Participants were naïve, right-handed young adults (Experiment 1, N = 10, 4 women; Experiment 2, N = 25, 10 women). Participants were native English speakers, had normal or corrected-to-normal vision, and gave written informed consent in accordance with the requirements of internal review boards at MIT. They were healthy undergraduate students who self-reported that they were not contra-indicated for metal implants, prior psychiatric illness, or any neurological impairment. Participants were
Self reference
There was greater activation during the self-reference task than during the valence task in regions typically associated with self reference, including MPFC extending into ACC and PCC/Precuneus (Fig. 1) (Table 1).
Default
There was greater activation during rest than during the valence task in regions typically associated with the default network, including MPFC extending into ACC as well as a large posterior region of activation, peaking in the PCC/Precuneus and extending into lateral parietal cortex,
Discussion
This study aimed to identify associations and dissociations in midline cortical structures between the self-reference network and the default network. The two networks were associated by their common inclusion of ventral MPFC (BA 10) and PCC (BA 31). The two networks were doubly dissociated by preferential engagement of dorsal MPFC (BA 9) for self-reference and of precuneus (BA 7) for rest. The general patterns of functional associations and dissociations of the three conjunctions were similar
Conflict of interest
The authors of the study have no conflict of interest to declare.
Acknowledgments
Supported by a grant from Simons Foundation Autism Research Initiative and the Poitras Center for Affective Disorders Research.
References (43)
- et al.
A component based noise correction method (CompCor) for BOLD and perfusion based fMRI
Neuroimage
(2007) - et al.
Self-referential reflective activity and its relationship with rest: a PET study
Neuroimage
(2005) - et al.
Distributed self in episodic memory: neural correlates of successful retrieval of self-encoded positive and negative personality traits
Neuroimage
(2004) - et al.
Characterizing evoked hemodynamics with fMRI
Neuroimage
(1995) - et al.
Reflective self-awareness and conscious states: PET evidence for a common midline parietofrontal core
Neuroimage
(2002) - et al.
Cortical networks for working memory and executive functions sustain the conscious resting state in man
Brain Res. Bull.
(2001) - et al.
Interrupting the “stream of consciousness”: an fMRI investigation
Neuroimage
(2006) - et al.
Dissociable medial prefrontal contributions to judgments of similar and dissimilar others
Neuron
(2006) - et al.
Cortical midline structures and the self
Trends Cogn. Sci.
(2004) - et al.
Self-referential processing in our brain
Neuroimage
(2006)
The neural correlates of direct and reflected self-knowledge
Neuroimage
Neural correlates of individual ratings of emotional salience: a trial-related fMRI study
Neuroimage
Delineating self-referential processing from episodic memory retrieval: common and dissociable networks
Neuroimage
Self-appraisal decisions evoke dissociated dorsal — ventral aMPFC networks
Neuroimage
Event-related fMRI studies of episodic encoding and retrieval: meta-analyses using activation likelihood estimation
Neuropsychologia
Likeableness ratings of 555 personality-trait words
J. Pers. Soc. Psychol.
Conceptual processing during the conscious resting state: a functional MRI study
J. Cogn. Neurosci.
Functional connectivity in the motor cortex of resting human brain using echo-planar MRI
Magn. Reson. Med.
The brain's default network anatomy, function, and relevance to disease
Ann. N. Y. Acad. Sci.
Cortical hubs hevealed by intrinsic functional connectivity: mapping, assessment of stability, and relation to Alzheimer's disease
J. Neurosci.
The precuneus: a review of its functional anatomy and behavioural correlates
Brain
Cited by (307)
The neural bases of familiar music listening in healthy individuals: An activation likelihood estimation meta-analysis
2023, Neuroscience and Biobehavioral ReviewsDefault mode network connectivity and social dysfunction in children with Attention Deficit/Hyperactivity Disorder
2023, International Journal of Clinical and Health PsychologyStratifying ASD and characterizing the functional connectivity of subtypes in resting-state fMRI
2023, Behavioural Brain ResearchSleep and dreaming in the light of reactive and predictive homeostasis
2023, Neuroscience and Biobehavioral Reviews
- 1
Whitfield-Gabrieli and Moran contributed equally to this work.