Elsevier

NeuroImage

Volume 124, Part A, 1 January 2016, Pages 806-812
NeuroImage

Meta-analysis of real-time fMRI neurofeedback studies using individual participant data: How is brain regulation mediated?

https://doi.org/10.1016/j.neuroimage.2015.09.042Get rights and content

Highlights

  • Previous rtfMRI studies assessed regulation success in targeted brain regions.

  • There is a need to understand networks involved in brain regulation itself.

  • IDP meta-analysis of 12 rtfMRI studies to identify networks related to neurofeedback

  • Anterior insula and basal ganglia are key components of the regulation network.

Abstract

An increasing number of studies using real-time fMRI neurofeedback have demonstrated that successful regulation of neural activity is possible in various brain regions. Since these studies focused on the regulated region(s), little is known about the target-independent mechanisms associated with neurofeedback-guided control of brain activation, i.e. the regulating network. While the specificity of the activation during self-regulation is an important factor, no study has effectively determined the network involved in self-regulation in general. In an effort to detect regions that are responsible for the act of brain regulation, we performed a post-hoc analysis of data involving different target regions based on studies from different research groups.

We included twelve suitable studies that examined nine different target regions amounting to a total of 175 subjects and 899 neurofeedback runs. Data analysis included a standard first- (single subject, extracting main paradigm) and second-level (single subject, all runs) general linear model (GLM) analysis of all participants taking into account the individual timing. Subsequently, at the third level, a random effects model GLM included all subjects of all studies, resulting in an overall mixed effects model. Since four of the twelve studies had a reduced field of view (FoV), we repeated the same analysis in a subsample of eight studies that had a well-overlapping FoV to obtain a more global picture of self-regulation.

The GLM analysis revealed that the anterior insula as well as the basal ganglia, notably the striatum, were consistently active during the regulation of brain activation across the studies. The anterior insula has been implicated in interoceptive awareness of the body and cognitive control. Basal ganglia are involved in procedural learning, visuomotor integration and other higher cognitive processes including motivation. The larger FoV analysis yielded additional activations in the anterior cingulate cortex, the dorsolateral and ventrolateral prefrontal cortex, the temporo-parietal area and the visual association areas including the temporo-occipital junction.

In conclusion, we demonstrate that several key regions, such as the anterior insula and the basal ganglia, are consistently activated during self-regulation in real-time fMRI neurofeedback independent of the targeted region-of-interest. Our results imply that if the real-time fMRI neurofeedback studies target regions of this regulation network, such as the anterior insula, care should be given whether activation changes are related to successful regulation, or related to the regulation process per se. Furthermore, future research is needed to determine how activation within this regulation network is related to neurofeedback success.

Introduction

Neurofeedback using real-time functional magnetic resonance imaging (rt-fMRI) enables participants to obtain voluntary control over multiple brain regions. Studies using this technique have demonstrated that it may be possible to successfully manipulate brain areas including the anterior cingulate cortex (ACC, Weiskopf et al., 2003, Hamilton et al., 2011), the posterior cingulate cortex (Brewer and Garrison, 2014), the anterior insular cortex (AIC, Caria et al., 2007, Caria et al., 2010, Berman et al., 2013), posterior insular cortex (PIC, Rance et al., 2014), amygdala (Posse et al., 2003, Zotev et al., 2011, Bruhl et al., 2014), primary motor and somatosensory cortex cortices (Yoo and Jolesz, 2002, Berman et al., 2012), premotor area (Johnson et al., 2012), visual cortex (Shibata et al., 2011), auditory cortex (Yoo et al., 2006, Haller et al., 2013), substantia nigra/ventral tegmental area (Sulzer et al., 2013), nucleus accumbens (Greer et al., 2014) and inferior frontal gyrus (Rota et al., 2009; for a review see Ruiz et al., 2014).

Real-time fMRI neurofeedback has also been explored as a supplementary treatment for various neurological disorders. For instance, real-time fMRI neurofeedback has shown positive benefits for diseases such as schizophrenia (Ruiz et al., 2013), depression (Linden et al., 2012, Young et al., 2014), tinnitus (Haller et al., 2010), Parkinson's disease (Subramanian et al., 2011) and nicotine addiction (Canterberry et al., 2013, Hartwell et al., 2013, Li et al., 2013). However, effect size of neurofeedback varies and in a lot of studies some participants fail to attain self-regulation. The neural mechanisms of neurofeedback as used for self-regulation of bodily functions are not well understood, which may be a roadblock to achieving consistent outcomes between studies and successful translation into clinics.

One of the most important but least understood characteristics of neurofeedback is the specificity of activation during self-regulation. Previous investigations in real-time fMRI neurofeedback have attempted to control for specificity of the self-regulation using feedback from another region (deCharms et al., 2005), subtracting the mean activity of a reference slice that does not contain involved brain regions (Caria et al., 2007, Rota et al., 2009), or using post-hoc statistical methods (Blefari et al., 2015). In contrast, we are here interested in the regions that are additionally activated during self-regulation, that is, regions that are involved in the cognitively demanding task of neurofeedback regulation.

In their landmark study, deCharms et al. reported that reduced pain perception via ACC regulation may have resulted from the contribution of a higher order region despite rigorous controls (deCharms et al., 2005). If so, exactly which regions would be responsible for effects of self-regulation?

To answer this question, it is important to consider the cognitive processes involved during neurofeedback and the corresponding networks. One of these networks is the central executive network (CEN) that is active in most cognitively demanding task, likely reflecting working-memory involvement and decision-making (Koechlin and Summerfield, 2007, Miller and Cohen, 2001). It includes the dorsolateral prefrontal cortex (dlPFC) and the posterior parietal cortex (Sridharan et al., 2008). In addition, the saliency network that is comprised of the AIC and the ACC as main components will be involved in neurofeedback relevant tasks including attentional control and monitoring. Menon and Uddin (2010) suggest that this network coordinates task-related information processing by recruiting various other, more specialized networks. For neurofeedback, these might include reward-learning areas, recruiting the striatum (Hollerman et al., 1998, Samejima et al., 2005, Daniel and Pollmann, 2014), the frontal cortex (Watanabe, 1996, O'doherty et al., 2003) and areas responsible for interoception (Craig, 2002, Lerner et al., 2009) such as parts of the AIC. Neurofeedback will likely use subnetworks cutting through all the above-mentioned networks.

Indeed, studies using a single region of interest suggest involvement of the dorsolateral prefrontal cortex (dlPFC), the dorsomedial prefrontal cortex (dmPFC, Zotev et al., 2013), the ventromedial prefrontal cortex (vmPFC, Haller et al., 2010) and the anterior mid-cingulate cortex (Lee et al., 2012) to anterior cingulate cortex (Lawrence et al., 2013, Zotev et al., 2013) in the regulation process. A number of feedback studies show activation of the posterior ACC (pACC,), although this area was not targeted (e.g. Caria et al., 2007, Rota et al., 2009, Lee et al., 2012, Veit et al., 2012, Lawrence et al., 2013). Similarly, several studies reported activation of the insula during neurofeedback runs (e.g. Rota et al., 2009, Haller et al., 2010, Lee et al., 2012, Paret et al., 2014).

In the current investigation, we assess the brain network mediating regulation in real-time fMRI neurofeedback. We hypothesized that regardless of the target region used, a common brain network is involved in the regulation process itself. Consequently, we performed a meta-analysis using individual participant data (IPD meta-analysis) across multiple previously reported rt-fMRI neurofeedback studies with different target regions in order to cancel out target region-specific effects and identify those activations commonly related to the regulation process. It should be noted that, at the current stage, we cannot distinguish between self-regulation processes and other processes involved in neurofeedback including feedback processing and learning as the current study does not include control runs without feedback (“transfer runs”). Our results suggest the existence of a neurofeedback network consisting of the anterior insula, basal ganglia, dorsal parts of the parietal lobe extending to the temporo-parietal junction, ACC, dlPFC, ventrolateral prefrontal cortex (vlPFC) and visual association areas including the temporo-occipital junction.

Section snippets

Study selection

Studies were selected based on a Web of Knowledge (https://apps.webofknowledge.com) search for the keywords: “real time fMRI”, “real time functional” or “rtfMRI” (in January 2014) as well as studies indicated in the real-time community ([email protected], updated in August 2015 to [email protected]) literature updates. This search provided us with a total of 316 publications. Next, we used the following selection criteria, 1) rt-fMRI neurofeedback, 2) 1.5 or 3.0 T static field

Results

The third level mixed effects analysis of all 12 studies yielded two main regions that are consistently activated during neurofeedback: the bilateral anterior insula and the basal ganglia. Considering the subsample analysis with a larger field of view (n = 8 studies) additional significant areas include the posterior ACC (pACC), the bilateral ventrolateral prefrontal cortex (vlPFC) and an area in the bilateral dorsolateral prefrontal cortex (dlPFC) extending to the premotor cortex (PMC), a large

Discussion

The IPD meta-analysis of rt-fMRI neurofeedback studies with a variety of target regions identified a regulation network that includes notably the anterior insula, the basal ganglia, the temporo-parietal area, the ACC, the dlPFC, the vlPFC and the visual association area including the temporo-occipital junction (see Fig. 2).

Anterior insula activation is known to occur during interoceptive cognition and self-awareness processes (Craig, 2002, Critchley et al., 2004). Additionally, specifically the

Conclusion

Brain self-regulation during rt-fMRI neurofeedback involves a complex regulation network, including notably AIC, BG and the ACC. Taking into account the limitation that the current investigation is a retrospective IPD meta-analysis of rt-fMRI studies, which were not specifically designed for this purpose, our results suggest that some target regions of rt-fMRI neurofeedback studies (notably insula and ACC) are also implicated in the process of regulation per se. This may therefore represent a

Support

This work was supported by the Swiss National Science Foundation (project 320030_147126/1, 320030_127079/1 and the Marie Heim-Vögtlin grants PMCDP2_145442 and PMCDP2_162223) and the Center for Biomedical Imaging (CIBM, Geneva, Switzerland).

We are very grateful to all of the researchers who kindly supplied us with data from their studies. This work was supported by data from Brian Berman & Silvina Horovitz, Markus Breimhorst, Annette Brühl, Andrea Caria, Sabine Frank, Steve Johnston & David

References (77)

  • J.H. Lee et al.

    Real-time fMRI-based neurofeedback reinforces causality of attention networks

    Neurosci. Res.

    (2012)
  • M. Lotze et al.

    Motor imagery

    J. Physiol. Paris

    (2006)
  • J.P. O'doherty et al.

    Temporal difference models and reward-related learning in the human brain

    Neuron

    (2003)
  • S. Posse et al.

    Real-time fMRI of temporolimbic regions detects amygdala activation during single-trial self-induced sadness

    Neuroimage

    (2003)
  • M.E. Raichle et al.

    A default mode of brain function: a brief history of an evolving idea

    Neuroimage

    (2007)
  • F. Robineau et al.

    Self-regulation of inter-hemispheric visual cortex balance through real-time fMRI neurofeedback training

    Neuroimage

    (2014)
  • S. Ruiz et al.

    Real-time fMRI brain computer interfaces: self-regulation of single brain regions to networks

    Biol. Psychol.

    (2014)
  • F. Schneider et al.

    The resting brain and our self: self-relatedness modulates resting state neural activity in cortical midline structures

    Neuroscience

    (2008)
  • S.M. Smith et al.

    Advances in functional and structural MR image analysis and implementation as FSL

    Neuroimage

    (2004)
  • J. Sulzer et al.

    Neurofeedback-mediated self-regulation of the dopaminergic midbrain

    Neuroimage

    (2013)
  • N. Weiskopf et al.

    Physiological self-regulation of regional brain activity using real-time functional magnetic resonance imaging (fMRI): methodology and exemplary data

    Neuroimage

    (2003)
  • M.W. Woolrich et al.

    Multilevel linear modelling for FMRI group analysis using Bayesian inference

    Neuroimage

    (2004)
  • H.D. Zimmer

    Visual and spatial working memory: from boxes to networks

    Neurosci. Biobehav. Rev.

    (2008)
  • C. Amiez et al.

    Anterior cingulate error-related activity is modulated by predicted reward

    Eur. J. Neurosci.

    (2005)
  • M. Arsalidou et al.

    The centre of the brain: topographical model of motor, cognitive, affective, and somatosensory functions of the basal ganglia

    Hum. Brain Mapp.

    (2013)
  • S. Arzy et al.

    Neural basis of embodiment: distinct contributions of temporoparietal junction and extrastriate body area

    J. Neurosci.

    (2006)
  • B.W. Balleine et al.

    Human and rodent homologies in action control: corticostriatal determinants of goal-directed and habitual action

    Neuropsychopharmacology

    (2010)
  • B.D. Berman et al.

    Modulation of functionally localized right insular cortex activity using real-time fMRI-based neurofeedback

    Front. Hum. Neurosci.

    (2013)
  • M.L. Blefari et al.

    Improvement in precision grip force control with self-modulation of primary motor cortex during motor imagery

    Front. Behav. Neurosci.

    (2015)
  • J.A. Brewer et al.

    The posterior cingulate cortex as a plausible mechanistic target of meditation: findings from neuroimaging

    Ann. N. Y. Acad. Sci.

    (2014)
  • A.B. Bruhl et al.

    Real-time neurofeedback using functional MRI could improve down-regulation of amygdala activity during emotional stimulation: a proof-of-concept study

    Brain Topogr.

    (2014)
  • M. Canterberry et al.

    Sustained reduction of nicotine craving with real-time neurofeedback: exploring the role of severity of dependence

    Nicotine Tob. Res.

    (2013)
  • A.D. Craig

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

    Nat. Rev. Neurosci.

    (2002)
  • H.D. Critchley et al.

    Neural systems supporting interoceptive awareness

    Nat. Neurosci.

    (2004)
  • R.C. Decharms et al.

    Control over brain activation and pain learned by using real-time functional MRI

    Proc. Natl. Acad. Sci. U. S. A.

    (2005)
  • K. Emmert et al.

    Comparison of anterior cingulate vs. insular cortex as targets for real-time fMRI regulation during pain stimulation

    Front. Behav. Neurosci.

    (2014)
  • N.A. Farb et al.

    Attending to the present: mindfulness meditation reveals distinct neural modes of self-reference

    Soc. Cogn. Affect. Neurosci.

    (2007)
  • S. Frank et al.

    The obese brain athlete: self-regulation of the anterior insula in adiposity

    PLoS One

    (2012)
  • Cited by (0)

    1

    Both authors contributed equally.

    View full text