TABLE 1.
Summary of main findings of (pharmacological) manipulation of catecholamines (NE and DA).
| Study | Modality(species) | Analysismethod | Manipulation | Main effects of manipulation on intrinsic correlations |
| Coull et al., 1999 | PET (humans) | Dynamic causal modeling | Clonidine (α2 agonist) | Rest: reduced effective connectivity from PFC to thalamus, and to and from visual cortex. Attentional task: general increase in effective connectivity, with changes centered on parietal cortex. |
| Achard and Bullmore, 2007 | fMRI (humans) | Graph theoretic analysis | Sulpride (D2 antagonist) | Reduced metrics of global and local efficiency. |
| Kelly et al., 2009 | fMRI (humans) | Seed-based correlation | L-DOPA (DA precursor) | Increased correlation between putamen and cerebellum and brainstem. Increased correlation between ventral striatum and vlPFC. Reduced correlation between DMN, and ventral striatum and caudate. |
| McCabe and Mishor, 2011 | fMRI (humans) | Seed-based correlation | Reboxetine (SNRI) | Reboxetine: reduced amygdala-OFC correlation, and reduced striatal–OFC correlation. |
| Cole et al., 2013 | fMRI (humans) | Dual regression | Haloperidol (D2 antagonist) L-DOPA (DA precursor) | Linear increase (haloperidol < placebo < L-DOPA) between BG network and sensorimotor cortex. Inerted-U between BG network and dorsal anterior-mid cingulate (placebo > haloperidol and L-DOPA). Linear decrease (haloperidol > placebo > L-DOPA) between DMN and sensorimotor cortex. Linear increase (haloperidol < placebo < L-DOPA) between DMN and SMG. |
| Akeju et al., 2016 | fMRI (humans) | Dual regression Seed-based correlation | Dexmedetomidine (α2 agonist) | Reduced correlation between DMN and bilateral thalamus and left cerebellum; increased correlation between DMN and IFG, putamen, and insula. Reduced correlation between rlFPN and cerebellum; increased correlation between rlFPN and cerebellum, precuneus, parietal operculum insula fusiform and angular gyri. Reduced correlation between llFPN and cerebellum, calcarine cortex, MFG, and SFG. |
| Metzger et al., 2015 | fMRI (humans) | Seed-based correlation | Reboxetine (SNRI) Amisulpride (D2/3 antagonist) | Reboxetine: increased correlation between brainstem, and thalamus and PCC; and thalamus and accumbens; reduced correlation between putamen-brainstem; amygdala-ACC; and between the accumens and two regions of the ACC; reduced correlation between accumbens and two regions in ACC. Amisulpride: increased correlation between: PCC-brainstem; amygdala, brainstem and thalamus; putamen-brainstem. |
| van den Brink et al., 2016 | fMRI (humans) | Graph theoretic analysis Seed-based correlation | Atomoxetine (SNRI) | Reduced metrics of global correlation strength and clustering. Reduced correlation between FPN, and DMN and visual network, and reduced correlation between the visual and sensorimotor network. Reduced correlation within a set of occipital regions. Reduced correlation between early visual cortex and the rest of the brain. |
| Guedj et al., 2017a | fMRI (rhesus macaque) | Dual regression | Atomoxetine (SNRI) | Reduced correlation within FPN, somatosensory, sensorimotor, visual, and superior temporal sulcus networks. Reduced correlation between various networks with an overall net change of reduced correlation. |
| Guedj et al., 2017b | fMRI (rhesus macaque) | Graph-theoretic analysis | Atomoxetine (SNRI) | Reduced global efficiency. Reduced global correlation strength. Increased clustering. |
| Ye et al., 2017 | fMRI (humans) | Graph theoretic analysis and seed-based correlation | Pramipexole (D2 agonist) | Reduced correlation between caudate and nodes of the sensorimotor network. No change in topological metrics. |
| Shine et al., 2018b | fMRI (humans) | Graph theoretic analysis | Atomoxetine (SNRI) | Rest: reduced metrics of integration. N-back task: increased metrics of integration. |
| van den Brink et al., 2018a | fMRI (humans) | Generalized eigenvalue decomposition | Atomoxetine (SNRI) | Increased correlations in a network that loosely resembled a rlFPN, and distribution of β (negatively) and D2 receptors. Reduced correlations in a network that loosely resembled a llFPN, sensorimotor and DMN networks, and distribution of β (positively) and α1 receptors. |
| Pfeffer et al., 2018 | MEG (humans) | DFA + computational model | Atomoxetine (SNRI) | Rest: reduced α scaling exponent due to atomoxetine. Task (bistable perception): reduced α scaling exponent due to atomoxetine. Computational model accounted for the findings as a change in excitation relative to inhibition. |
| Shafiei et al., 2019 | fMRI (humans) | Seed-based correlation Graph theoretic analysis | Phenylalanine and tyrosine depletion (DA depletion) | Reduced correlations strength in sensorimotor, salience, and temporal networks. Reduced between-module correlation of the sensorimotor and salience networks. |
| Pfeffer et al., 2019 | MEG (humans) | Graph theoretic analysis + computational model | Atomoxetine (SNRI) | Rest: no effect of atomoxetine. Task (bistable perception): Atomoxetine increased correlation strength. Computational models indicate that effects can be accounted for by an increase in gain. |
| Zerbi et al., 2019 | fMRI (mice) | Graph theoretic analysis (FCD) Dual regression | Chemogenetic LC stimulation | Increase in global correlation strength. Increased correlation in the Salience, amygdala, auditory, striato-motor, and DMN networks. |
ACC: anterior cingulate cortex; BG: basal ganglia; DA: dopamine; DFA: detrended fluctuation analysis; DMN: default mode network; vmPFC: ventromedial prefrontal cortex; LC: locus coeruleus; ll: left lateralized; MFG: middle frontal gyrus; IFG: Inferior frontal gyrus; vlPFC: ventrolateral prefrontal cortex; NE: norepinephrine; OFC: orbitofrontal cortex; SMG: supramarginal gyrus; PCC: posterior cingulate cortex; rl: right lateralized; FPN: frontoparietal network; SFG: superior frontal gyrus; SNRI: selective norepinephrine reuptake inhibitor.