Elsevier

Cortex

Volume 123, February 2020, Pages 185-199
Cortex

Research Report
Functional connectivity of the orbitofrontal cortex, anterior cingulate cortex, and inferior frontal gyrus in humans

https://doi.org/10.1016/j.cortex.2019.10.012Get rights and content

Abstract

Parcellation of the orbitofrontal cortex, anterior cingulate cortex, and inferior frontal gyrus based on their functional connectivity with the whole brain in resting state fMRI with 654 participants was performed to investigate how these regions with different functions in reward, emotion and their disorders are functionally connected to each other and to the whole brain. The human medial and lateral orbitofrontal cortex, the ventromedial prefrontal cortex, the anterior cingulate cortex, and the right and left inferior frontal gyrus have different functional connectivity with other brain areas and with each other; and each of these regions has several parcels with different functional connectivity with other brain areas. In terms of functional connectivity, the lateral orbitofrontal cortex extends especially on the right into the orbital part of the inferior frontal gyrus and provides connectivity with premotor cortical areas. The orbitofrontal cortex, especially the lateral orbitofrontal cortex, has connectivity not only with language-related areas in the inferior frontal gyrus (Broca's area), but also with the angular and supramarginal gyri. In this context, whereas the connectivity of the orbitofrontal cortex, ventromedial prefrontal cortex, and anterior cingulate cortex is symmetrical, the connectivity of the inferior frontal gyrus triangular and opercular parts is asymmetrical for the right and the left hemispheres. These findings have implications for understanding the neural bases of human emotion and decision-making, and for their disorders including depression.

Introduction

It has been traditional, for historical reasons based on what was technically possible with the light microscope, to divide the cerebral cortex into different areas based on cytoarchitecture and myeloarchitecture (Brodmann, 1909a, Brodmann, 1909b, Henssen et al., 2016, Öngür et al., 2003, Ongür and Price, 2000, Vogt, 2009, Cajal, 1995). For example, the primary visual cortex can be identified by its prominent layer 4, with many granule cells involved in processing the massive visual sensory input from the lateral geniculate nucleus. In another example, motor cortex can be identified by its large pyramidal cells in layer 5, involved in sending motor outputs directly to the spinal cord for fine control of the distal extremities such as the fingers. However, an important way in which to define a cortical area is in terms of the functions it performs, which are related to the inputs that it receives and the regions to which it connects (Rolls, 2016a). The implication is that a different way to divide the cortex into different areas is in terms of the connectivity of each brain area with other brain areas.

In this paper we utilize a method to delineate cortical areas based on their functional connectivity with other brain areas, based on the computational concept that the functional subdivisions of the cortex are likely to be related to where they receive connections from, and where they project to (Rolls, 2016a). In essence, the concept is that the cerebral cortex, and the human brain, can be understood in term of the computations that each brain area performs, based on the inputs that it receives, and where it sends it outputs to (Rolls, 2016a). To achieve this delineation of cortical areas based in their connections with other cortical areas, we measure the functional connectivity of individual voxels in the orbitofrontal cortex and closely related areas the anterior cingulate cortex and inferior frontal gyrus with many different regions of the brain. On the basis of the functional connectivity of each voxel we divide the voxels into different groups or clusters to identify connectional subdivisions of voxels within the area being investigated. Functional connectivity refers to correlations between the fMRI BOLD signal in different brain regions, and reflects direct connections between cortical areas as shown by combined anatomical pathways tracing and functional connectivity analyses in macaques, and also some trans-synaptic effects (Van Essen et al., 2019). An advantage of functional connectivity is that it can reveal trans-synaptic effects, and is non-invasive and can be performed in humans.

In the present case, the voxels of interest are in the orbitofrontal cortex (OFC), the anterior cingulate cortex (ACC), and the inferior frontal gyrus (IFG), because all of these areas are implicated in different ways in emotion, and in emotional disorders including depression (Cheng et al., 2016, Cheng et al., 2018a, Cheng et al., 2018b, Cheng et al., 2018c, Cheng et al., 2018d, Rolls, 2014, Rolls, 2018, Rolls, 2019a, Rolls, 2019c, Rolls, 2019d, Rolls et al., 2019a, Rolls et al., 2019c). This investigation thus goes beyond a previous parcellation of the orbitofrontal cortex (Kahnt, Chang, Park, Heinzle, & Haynes, 2012) not only in terms of the robustness of the analysis (they utilized results from 13 participants, we utilize results from 654 participants for robustness and generalizability), but also because there is a whole set of connected systems involving the OFC, ACC and IFG that are important in emotion and its disorders, so that it is very important to know how all the subparts of these regions are connected. The current investigation is the first to conduct a parcellation of ACC and IFG together with OFC, in order to show how the subparts of these regions are functionally connected, given the importance of at least parts of these regions for emotion, decision-making, and their disorders. Moreover, in the approach described here, the relations and divisions between areas are based on quantitative measures of correlations between the connectivity of areas (identified with the quantitative approach of cluster analysis) with the rest of the brain, whereas anatomical investigations of subnetworks has been based on a qualitative analysis of network subdivisions (Ongür and Price, 2000, Price, 1999, Price, 2006, Price, 2007).

In the present investigation, the connectivity between each voxel in the areas OFC/ACC/IFG and every AAL3 brain area (Rolls, Huang, Lin, Feng, & Joliot, 2019) was measured by the Pearson correlation between their BOLD signals using resting state functional magnetic resonance imaging (fMRI). The concept here is that in a system in which a task is not being performed, the noise-produced perturbations in the system will influence other nodes in the system according to the strength of the connections between any two nodes in the system. The noise in the system is produced by the almost random (Poisson) times of firing of the neurons in the system for a given mean firing rate (Cabral et al., 2014, Deco et al., 2013, Rolls and Deco, 2010), which in turn can be related to factors such as noise in ion channels (Faisal et al., 2008, Rolls and Deco, 2010). A list of abbreviations of AAL3 areas is provided in Table S1 (see Table 1).

Section snippets

Participants

There were 254 healthy participants subjects (age: 39.7 ± 15.8, Male/Female: 166/88) from Xinan (First Affiliated Hospital of Chongqing Medical School in Chongqing, China); and there were 400 healthy participants (age: 40.6 ± 21.4, Male/Female: 147/253) from the NKI cohort Nathan Kline Institute–Rockland Sample (NKI-RS) dataset (Nooner et al., 2012). All the functional connectivity-driven parcellations were based on the resting-state fMRI data of the 254 subjects in the Xinan cohort and 400

Results

The parcellation based on the resting state fMRI of 654 participants from the NKI and Xinan datasets is shown in Fig. 1, with the average functional connectivity (FC) of the voxels of each of the 24 parcels to the AAL3 brain areas shown in Fig. 2. The parcels and their connectivity are as follows. Each of the parcels had connectivity with at least one of the AAL3 areas shown in Fig. 2 that was significantly different at p = 1.1 × 10−46 (Bonferroni corrected). Fig. 3 illustrates the connections

The connectivity and some of the implications

The findings of this parcellation of the orbitofrontal cortex, anterior cingulate cortex, and inferior frontal gyrus based on the functional connectivity with the whole brain include the following.

First, three parcels (2–4) located mainly in areas 13 and 11 in the medial/mid orbitofrontal cortex had strong connectivity with each other, and moderate connectivity with posterior to mid-temporal cortical areas and insula (which are likely to provide visual, auditory and taste inputs), with the

Funding

J.F. is partially supported by the key project of Shanghai Science & Technology Innovation Plan (No. 15JC1400101 and No. 16JC1420402) and the National Natural Science Foundation of China (Grant No. 71661167002 and No. 91630314). The research was also partially supported by the Shanghai AI Platform for Diagnosis and Treatment of Brain Diseases (No. 2016–17). The research was also partially supported by Base for Introducing Talents of Discipline to Universities No. B18015. W.C. is supported by

CRediT authorship contribution statement

Jingnan Du: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing - review & editing. Edmund T. Rolls: Conceptualization, Methodology, Software, Formal analysis, Investigation, Writing - original draft, Writing - review & editing, Visualization. Wei Cheng: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation,

Declaration of Competing Interest

None of the authors declares a conflict of interest.

Acknowledgements

The use of resting state fMRI data from the Nathan Kline Institute–Rockland Sample (NKI-RS) dataset (Nooner et al., 2012) is gratefully acknowledged.

References (55)

  • M.L. Kringelbach et al.

    Neural correlates of rapid reversal learning in a simple model of human social interaction

    NeuroImage

    (2003)
  • J.D. Power et al.

    Methods to detect, characterize, and remove motion artifact in resting state fMRI

    NeuroImage

    (2014)
  • E.T. Rolls

    Functions of the anterior insula in taste, autonomic, and related functions

    Brain and Cognition

    (2016)
  • E.T. Rolls

    A non-reward attractor theory of depression

    Neuroscience and Biobehavioral Reviews

    (2016)
  • E.T. Rolls

    The orbitofrontal cortex and emotion in health and disease, including depression

    Neuropsychologia

    (2019)
  • E.T. Rolls et al.

    Implementation of a new parcellation of the orbitofrontal cortex in the automated anatomical labeling atlas

    Neuroimage

    (2015)
  • M.F. Rushworth et al.

    Valuation and decision-making in frontal cortex: One or many serial or parallel systems?

    Current Opinion in Neurobiology

    (2012)
  • K. Brodmann

    Localisation in the cerebral cortex (L. J. Garey, trans.)

    (1909)
  • K. Brodmann

    Vergleichende Lokalisationslehre der Grosshirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues

    (1909)
  • S.T. Carmichael et al.

    Connectional networks within the orbital and medial prefrontal cortex of macaque monkeys

    Journal of Comparative Neurology

    (1996)
  • Y. Chao-Gan et al.

    DPARSF: A MATLAB toolbox for “pipeline” data analysis of resting-state fMRI

    Frontiers in Systems Neuroscience

    (2010)
  • W. Cheng et al.

    Medial reward and lateral non-reward orbitofrontal cortex circuits change in opposite directions in depression

    Brain

    (2016)
  • W. Cheng et al.

    Functional connectivity of the human amygdala in health and in depression

    Social Cognitive and Affective Neuroscience

    (2018)
  • W. Cheng et al.

    Increased functional connectivity of the posterior cingulate cortex with the lateral orbitofrontal cortex in depression

    Translational Psychiatry

    (2018)
  • W. Cheng et al.

    Functional connectivities in the brain that mediate the association between depressive problems and sleep quality

    JAMA Psychiatry

    (2018)
  • W.L. Deng et al.

    Separate neural systems for behavioral change and for emotional responses to failure during behavioral inhibition

    Human Brain Mapping

    (2017)
  • A.A. Faisal et al.

    Noise in the nervous system

    Nature Reviews Neuroscience

    (2008)
  • Cited by (0)

    1

    Co-first author.

    View full text