Abstract
The insula region is known to be an integrating hub interacting with multiple brain networks involved in cognitive, affective, sensory, and autonomic processes. There is growing evidence suggesting that this region may have an important role in Parkinson's disease (PD). Thus, to investigate the functional organization of the insular cortex and its potential role in parkinsonian features, we used a coordinate‐based quantitative meta‐analysis approach, the activation likelihood estimation. A total of 132 insular foci were selected from 96 published experiments comprising the five functional categories: cognition, affective/behavioral symptoms, bodily awareness/autonomic function, sensorimotor function, and nonspecific resting functional changes associated with the disease. We found a significant convergence of activation maxima related to PD in different insular regions including anterior and posterior regions bilaterally. This study provides evidence of an important functional distribution of different domains within the insular cortex in PD, particularly in relation to nonmotor aspects, with an influence of medication effect. Hum Brain Mapp 37:1375‐1392, 2016. © 2016 Wiley Periodicals, Inc.
Keywords: Parkinson's disease, insula, nonmotor symptoms, dopamine, cognition, behavior
INTRODUCTION
In the past few years, the insula region has generated a great deal of interest, and while generally considered a limbic region, it is now known to be involved in numerous other functions. In fact, the insula is considered to be an integrating hub linking several functional systems, each comprised a set of anatomically and functionally different regions involved in cognitive, affective, sensory, and autonomic processes [Christopher et al., 2014a; Kurth et al., 2010]. This triangle‐shaped area located in between the frontal, parietal, and temporal lobes is divided into four functional integrative nodes [Kurth et al., 2010]. The mid‐posterior insula is implicated in somatomotor functions while the central insula participates in olfactory and gustatory tasks. The ventral anterior insula is linked to social and emotional functions, whereas the dorsal anterior insula is mainly involved in cognition.
To date, no studies have directly investigated the contribution of the insula to symptoms of PD. While PD is primarily considered a movement disorder, these patients are also afflicted by a large number of nonmotor symptoms, i.e., behavioral, cognitive, sensory, and autonomic disturbances [Chaudhuri and Schapira, 2009; Park and Stacy, 2009]. In general, imaging and neurophysiological studies investigating brain abnormalities in PD have focused on different cortical and subcortical regions but have never addressed the role of the insula, despite substantial evidence supporting its potential contribution to nonmotor symptoms in PD. While deposition of alpha‐synuclein in the insula [Braak et al., 2006] can directly impair receptor function and synaptic activity in this region, the degeneration of dopaminergic, cholinergic, and serotonergic projecting neurons to the insula [Halliday et al., 1990] may also significantly disrupt its functional integrity. The subsequent loss of neurotransmitter modulation in the insula could affect information processing through heavy interconnections between the insula and different cortical regions (i.e., frontal, temporal, parietal, cingulate cortex) [Cauda et al., 2011; Nieuwenhuys, 2012] as well the basal ganglia [Chikama et al., 1997].
In a recent review, we highlighted the possible role of the insula in nonmotor symptoms of PD [Christopher et al., 2014a]. However, in order to investigate the functional organization of the insular cortex and its potential role in parkinsonian features, we applied a quantitative meta‐analysis method to published neuroimaging studies to identify those core abnormalities consistently manifested across patient cohorts and range of tasks which are associated with the insula. We used a coordinate‐based quantitative meta‐analysis approach, the activation likelihood estimation (ALE) [Eickhoff et al., 2009; Wager et al., 2009] to overcome the classical limitations of neuroimaging studies such as heterogeneity of patient population and small sample sizes. This approach has been used consistently and successfully in a number of recent studies [Arsalidou et al., 2013; Criaud and Boulinguez, 2013; Herz et al., 2014; Kurth et al., 2010; Mutschler et al., 2009; Pan et al., 2012; Shao et al., 2014; Uddin et al., 2014].
METHODS
Literature Search
To review all imaging studies possibly involving the insula in Parkinson's disease, the Web of Science and Pubmed databases were examined from 1993 to 2015 using the following keyword combinations: “‘Parkinson's disease’ AND ‘Insula’,” “‘Parkinson's disease’ AND ‘functional magnetic resonance (fMRI)’,” “‘Parkinson's disease’ AND ‘positron emission tomography (PET)’.” This search resulted in 96 studies (Table 1). Only fMRI and PET (receptor ligands and H2O15) studies were considered. All articles were screened for eligibility with the following inclusion criteria:
Table 1.
List of studies including first author, year of publication, the contrast used, and the medication state during the task, the subcategory, the number of patients and controls, the age of each group, the UPDRS, and the state of medication during the evaluation and the modality
| First author | Year | Contrasts (ON/OFF) | Category | N PD (vs. N control group) | Age | UPDRS‐III (ON/OFF) | Modality |
|---|---|---|---|---|---|---|---|
| Ballanger | 2010 | PD with visual hallucination vs. CPD (ON) | Affective & Behavioural symptoms (Visual Hallucination) | 7 PD with visual hallucinations (7 CPD) | 69 PD (67 CPD) | 24 vs 15 (ON) | PET (18F Setoperone) |
| Rest | |||||||
| Ballanger | 2012 | PD vs NC (OFF) | Other | 8 PD (7 NC) | 63 PD | 27 (OFF) | PET (18F MPPF) |
| Rest | |||||||
| Ballanger | 2012 | PD with depression vs NC (OFF) | Affective & Behavioural symptoms (Depression) | 4 PD with depression (7 NC) | 54 PD with depression | 26 (OFF) | PET (18F MPPF) |
| Rest | |||||||
| Ballanger | 2012 | PD with depression vs CPD (OFF) | Affective & Behavioural symptoms (Depression) | 4 PD with depression (8 CPD) | 54 with depression | 26 (OFF) | PET (18F MPPF) |
| Rest | |||||||
| Beyer | 2008 | PD with early dementia vs PD with late dementia (ON) | Cognition (Dementia) | 9 PD with early dementia (6 PD with late dementia) | 74 PD with early dementia (70 PD with late dementia) | 40 vs 40 (ON) | fMRI |
| Rest | |||||||
| Bohlhalter | 2009 | Correlation between somatosensory discrimination and working memory scale (OFF) | Cognition (Working Memory and Somatosensory Discrimination) | 12 PD (12 NC) | 59 PD (47 NC) | 17 (ON) | PET (H15O) |
| Task: somatosensory discrimination | |||||||
| Borghammer | 2012 | PD vs NC (OFF) | Other | 21 PD (11 NC) | 64 PD (60 NC) | 16 (ON) | PET (18F FDG) |
| Rest | |||||||
| Brefel‐Courbon | 2005 | Pain induced activity, PD vs NC (OFF) | Bodily Awareness (Pain) | 9 PD (9 NC) | 65 PD | 25 (OFF) | PET (H15O) |
| Task: cold water stimulation inducing painful or nonpainful sensation on the hand | |||||||
| Brefel‐Courbon | 2005 | Pain induced activity, PD vs NC (ON) | Bodily Awareness (Pain) | 9 PD (9 NC) | 65 PD | 15 (ON) | PET (H15O) |
| Task: cold water stimulation inducing painful or non‐painful sensation on the hand | |||||||
| Brefel‐Courbon | 2005 | Pain induced activity, PD OFF vs PD ON | Bodily Awareness (Pain) | 9 PD | 65 PD | 25 OFF vs 15 ON | PET (H15O) |
| Task: cold water stimulation inducing painful or non‐painful sensation on the hand | |||||||
| Brefel‐Courbon | 2013 | Pain‐induced activity, CPD vs PD with neuropathic pain (OFF) | Bodily awareness (pain) | 9 PD with neuropathic pain (9CPD) | 61 PD with neuropathic pain (65 CPD) | 28 vs 25 (OFF) | PET (H15O) |
| Task: cold water stimulation inducing painful or non‐painful sensation on the hand | |||||||
| Caproni | 2013 | 1 finger sequence, PD vs NC (OFF) | Motor | 11 PD (11 NC) | 65 PD (65 NC) | 20 (OFF) | fMRI |
| Task: finger tapping sequences executed with the right hand, conditions: 1 finger sequence, simple 5 fingers sequences, complex 5 fingers sequence. | |||||||
| Caproni | 2013 | Simple 5 fingers sequence, PD vs NC (OFF) | Motor | 11 PD (11 NC) | 65 PD (65 NC) | 20 (OFF) | fMRI |
| Task: finger tapping sequences executed with the right hand, conditions: 1 finger sequence, simple 5 fingers sequences, complex 5 fingers sequence. | |||||||
| Caproni | 2013 | Complex 5 fingers sequence, NC vs PD (OFF) | Motor | 11 PD (11 NC) | 65 PD (65 NC) | 20 (OFF) | fMRI |
| Task: finger tapping sequences executed with the right hand, conditions: 1 finger sequence, simple 5 fingers sequences, complex 5 fingers sequence. | |||||||
| Cerasa | 2006 | Synchronized tapping vs rest, PD vs NC (OFF) | Motor | 10 PD (11 NC) | 64 PD (63 NC) | 28 (OFF) | fMRI |
| Task: Synchronized tapping with right index | |||||||
| Ceravolo | 2011 | PD PPTg stimulation ON vs OFF (OFF) | Other | 6 PD | 65 PD | 74 PPTg ON vs 38 PPTg OFF (OFF) | PET (18F FDG) |
| Rest | |||||||
| Christopher | 2013 | NC vs PD with MCI (OFF) | Cognition (MCI) | 11 PD with MCI (14 NC) | 71 PD with MCI (68 NC) | 31 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Christopher | 2013 | CPD vs PD with MCI (OFF) | Cognition (MCI) | 11 PD with MCI (11 CPD) | 71 PD with MCI (69 CPD) | 31 vs 23 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Christopher | 2014 | NC vs PD with amnestic MCI (OFF) | Cognition (amnestic MCI) | 9 PD with amnestic MCI (14 NC) | 68 PD with amnestic MCI (68 NC) | 36 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Christopher | 2014 | NC vs PD with MCI (OFF) | Cognition (MCI) | 10 PD with MCI (14 NC) | 70 PD with MCI (68 NC) | 21 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Christopher | 2014 | CPD vs PD with amnestic MCI (OFF) | Cognition (amnestic MCI) | 9 PD with amnestic MCI (11 CPD) | 68 PD with amnestic MCI (69 CPD) | 36 vs 23 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Christopher | 2014 | CPD vs PD with MCI (OFF) | Cognition (MCI) | 10 PD with MCI (11 CPD) | 70 PD with MCI (69 CPD) | 21 vs 23 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Christopher | 2014 | PD with MCI vs PD with amnestic MCI (OFF) | Cognition (amnestic MCI) | 9 PD with amnestic MCI (10 PD with MCI) | 68 PD with amnestic MCI (70 PD with MCI) | 36 vs 21 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Christopher | 2014 | Correlation between brain activity and memory score (OFF) | Cognition (amnestic MCI) | 9 PD with amnestic MCI | 68 PD with amnestic MCI | 36 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Cilia | 2008 | Gambling PD vs CPD (OFF) | Affective & Behavioural symptoms (Gambling) | 11 gambling PD (40 CPD) | 57 gambling PD (55 CPD) | 18 vs 19 (ON) | SPECT (TC99m) |
| Rest | |||||||
| Cilia | 2011 | Correlation activity change and gambling score (OFF) | Affective & Behavioural symptoms (Gambling) | 15 gambling PD | 60 gambling PD | 17 (ON) | SPECT (TC99m) |
| Rest | |||||||
| Cools | 2006 | Error vs correct responses | Cognition (Learning) | 14 PD | 66 PD | 39 OFF vs 16 ON | fMRI |
| Task: probabilistic reversal learning task | |||||||
| Delaveau | 2009 | Placebo vs Levodopa during task recognition (OFF) | Affective & Behavioural symptoms (Emotion recognition) | 14 PD | 61 PD | fMRI | |
| Task: emotional face matching | |||||||
| Delaveau | 2010 | NC vs PD (OFF) | Affective & Behavioural symptoms (Emotion recognition) | 14 PD (13 NC) | 61 PD (56 NC) | fMRI | |
| Task: emotional face matching | |||||||
| Delaveau | 2010 | PD placebo vs l‐dopa | Affective & Behavioural symptoms (Emotion recognition) | 14 PD (13 NC) | 61 PD (56 NC) | fMRI | |
| Task: emotional face matching | |||||||
| Dellapina | 2012 | Anti‐correlation between brain activity and pain threshold (OFF) | Bodily Awareness (Pain) | 8 PD with pain (8 CPD) | 65 PD with pain (62 CPD) | 16 vs 12 (OFF) | PET (H15O) |
| Task: cold water stimulation inducing painful or non‐painful sensation on the hand | |||||||
| Elsinger | 2003 | Synchronized tapping vs rest (OFF) | Motor | 10 PD | 67 PD | 24 (OFF) | fMRI |
| Task: Synchronized tapping with right index | |||||||
| Fregni | 2006 | NC vs PD with depression (OFF) | Affective & Behavioural symptoms (Depression) | 26 PD with depression (29 NC) | 66 PD with depression (65 NC) | 35 (ON) | SPECT (TC99m) |
| Rest | |||||||
| Gamma | 2014 | PD with visual hallucination vs PD with visual hallucination and cognitive dysfunction (ON) | Cognition (Cognitive Dysfunction in Visual Hallucinations) | 6 PD with visual hallucination and cognitive dysfunction (5 PD with visual hallucination) | 71 PD | 23 (ON) | fMRI |
| Rest | |||||||
| Goldman | 2014 | CPD vs PD with visual hallucination (ON) | Affective & Behavioural symptoms (Visual Hallucination) | 25 PD with visual hallucination (25 CPD) | 75 PD with visual hallucination (75 CPD) | 44 vs 39 (ON) | fMRI |
| Rest | |||||||
| Hanakawa | 1999 | Walk vs rest, PD vs NC (OFF) | Motor | 10 PD (10 NC) | 67 PD (67 NC) | 35 (OFF) | SPECT (TC99m) |
| Task: Walking or resting just before the scan | |||||||
| Helmich | 2010 | Connectivity from the anterior putamen, NC vs PD (OFF) | Other | 41 PD (36 NC) | 57 PD (57 NC) | 28 (OFF) | fMRI |
| Rest | |||||||
| Helmich | 2010 | Connectivity from the posterior putamen, PD vs NC (OFF) | Other | 41 PD (36 NC) | 57 PD (57 NC) | 28 (OFF) | fMRI |
| Rest | |||||||
| Hsu | 2007 | PD vs NC (OFF) | Other | 27 PD (24 NC) | 66 (62 NC) | 26 (ON) | SPECT (TC99m) |
| Rest | |||||||
| Hyoung | 2010 | CPD vs PD with MCI (OFF) | Cognition (MCI) | 18 PD with MCI (20 CPD) | 66 PD with MCI (62 CPD) | 25 vs 22 (ON) | PET (18F FDG) |
| Rest | |||||||
| Imon | 1999 | PD Hoen and Yahr stage 3 or 4 vs NC (ON) | Other | 12 PD Hoen and Yahr stage 3 or 4 (48 NC) | 63 PD Hoen and Yahr stage 3 or 4 (58 NC) | SPECT (TC99m) | |
| Rest | |||||||
| Jech | 2012 | PD before vs after STN‐DBS surgery (OFF) | Motor | 12 PD | 56 PD | 34 vs 23 (OFF) | fMRI |
| Task: Tapping with left or right hand | |||||||
| Kaasinen | 2004 | Anti‐correlation novelty seeking scale and brain activity (OFF) | Cognition (Executive) | 28 PD | 62 PD | 39 (ON) | PET (11C FLB 457) |
| Rest | |||||||
| Kahan | 2012 | Interaction movement and STN stimulation (ON/OFF), (OFF) | Motor | 10 PD | 58 PD | 21 (OFF) | fMRI |
| Task: voluntary hand movement | |||||||
| Kikuchi | 2001 | NC vs PD Hoen and Yahr stage 3/4 (ON) | Other | 11 PD Hoen and Yahr stage 3 or 4 (11 NC) | 59 PD Hoen and Yahr stage 3/4 (62 NC) | 28 (ON) | SPECT (TC99m) |
| Rest | |||||||
| Kikuchi | 2001 | PD Hoen and Yahr stage 1/2 vs PD Hoen and Yahr stage 3/4 (ON) | Other | 11 PD Hoen and Yahr stage 3 or 4 (7 PD Hoen and Yahr stage 1 or 2) | 59 PD Hoen and Yahr stage 3/4 (60 PD Hoen and Yahr stage 1/2) | 19 vs 28 (ON) | SPECT (TC99m) |
| Rest | |||||||
| Kostic | 2010 | PD vs NC (ON) | Other | 24 PD (26 NC) | 65 PD (63 NC) | 19 (ON) | fMRI |
| Rest | |||||||
| Kostic | 2010 | PD with depression vs CPD (ON) | Affective & Behavioural symptoms (Depression) | 16 PD (24 CPD) | 66 PD (65 CPD) | 23 vs 19 (ON) | fMRI |
| Rest | |||||||
| Lee | 2014a | PD with MCI vs PD with MCI and dementia (ON) | Cognition (MCI and Dementia) | 15 PD with MCI and dementia (36 PD with MCI) | 73 PD with MCI and dementia (71 PD with MCI) | 17 vs 20 (ON) | fMRI |
| Rest | |||||||
| Lee | 2014a | NC vs PD with MCI and dementia (ON) | Cognition (MCI and Dementia) | 15 PD with MCI and dementia (25 NC) | 73 PD with MCI and dementia (70 NC) | 17 (ON) | fMRI |
| Rest | |||||||
| Lee | 2014b | PD with ICD vs CPD (OFF) | Affective & Behavioural symptoms (ICD) | 11 PD with ICD (11 CPD) | 57 PD with ICD (59 CPD) | 14 vs 15 (ON) | PET (18F FP‐CIT) |
| Lee | 2014c | PD with high olfaction score vs NC (ON) | Autonomic symptoms (Olfaction) | 38 PD with high olfaction score (50 NC) | 69 PD with high olfaction score (69 NC) | 18 (ON) | fMRI |
| Rest | |||||||
| Liotti | 2003 | Phonation, before vs after voice therapy | Motor | 5 PD with hypophonia | 61 PD with hypophonia | PET (H15O) | |
| Task: speech task including phonation, reading and conversation | |||||||
| Lotze | 2009 | Correlation with error in recognition (OFF) | Affective & Behavioural symptoms (Emotion recognition) | 9 PD | 66 PD | 38 (OFF) | fMRI |
| Task: emotional and non‐emotional gesture recognition | |||||||
| Luo | 2014 | PD with depression vs NC (OFF) | Affective & Behavioural symptoms (Depression) | 29 PD with depression (30 NC) | 51 PD with depression (54 NC) | 28 (OFF) | fMRI |
| Rest | |||||||
| Maillet | 2012 | Hand movement, PD OFF vs PD ON | Motor | 12 PD | 60 PD | 40 OFF vs 10 ON | fMRI |
| Task: Hand movement and/or speech production | |||||||
| Mak | 2014 | PD with MCI vs CPD (ON) | Cognition (MCI) | 24 PD with MCI (66 CPD) | 69 PD with MCI (63 CPD) | 20 vs 17 (ON) | fMRI |
| Rest | |||||||
| Mallol | 2007 | NC vs PD (OFF) | Motor | 13 PD (11 NC) | 65 PD (62 NC) | 23 (OFF) | fMRI |
| Task: finger to thumb and hand rotation movements | |||||||
| Mattay | 2002 | PD OFF vs PD ON | Cognition (Working Memory) | 10 PD | 55 PD | 9 OFF vs 5 ON | fMRI |
| Task: N‐back | |||||||
| Monchi | 2004 | Negative vs control feedback (OFF) | Cognition (Executive) | 8 PD | 57 PD | 12 (OFF) | fMRI |
| Task: Wisconsin Card Sorting Task with negative, positive and control feedback | |||||||
| Oishi | 2004 | Correlation verbal IQ score | Cognition (Executive) | 44 PD | 66 PD | SPECT (TC99m) | |
| Rest | |||||||
| Oishi | 2004 | Correlation full IQ score | Cognition (Executive) | 44 PD | 66 PD | (ON) | SPECT (TC99m) |
| Rest | |||||||
| Pavese | 2010 | CPD vs PD with Fatigue (OFF) | Affective & Behavioural symptoms (Fatigue) | 10 PD with fatigue (9 CPD) | 65 PD with fatigue (63 CPD) | 35 vs 33 (OFF) | PET (18F FDOPA) |
| Rest | |||||||
| Pavese | 2010 | CPD vs PD with Fatigue (OFF) | Affective & Behavioural symptoms (Fatigue) | 8 PD with fatigue (8 CPD) | 65 PD with fatigue (64 CPD) | 35 vs 34 (OFF) | PET (11C DASB) |
| Rest | |||||||
| Payoux | 2009 | Interaction movement and GP stimulation (OFF) | Motor | 5 PD | 58 PD | PET (H15O) | |
| Task: Opening and clenching fist | |||||||
| Peran | 2009 | Generation of action verbs (ON) | Cognition (Language) | 14 PD | 64 PD | fMRI | |
| Task: object naming and generation of action verbs related to man‐made or manipulable biological objects | |||||||
| Peran | 2009 | Generation of action verbs vs objects naming (man‐man objects) (ON) | Cognition (Language) | 14 PD | 64 PD | fMRI | |
| Task: object naming and generation of action verbs related to man‐made or manipulable biological objects | |||||||
| Peran | 2009 | Generation of action verbs with biological objects vs naming man‐man objects) (ON) | Cognition (Language) | 14 PD | 64 PD | fMRI | |
| Task: object naming and generation of action verbs related to man‐made or manipulable biological objects | |||||||
| Peran | 2009 | Naming biological objects (ON) | Cognition (Language) | 14 PD | 64 PD | fMRI | |
| Task: object naming and generation of action verbs related to man‐made or manipulable biological objects | |||||||
| Pinto | 2004 | Silent articulation, OFF STN DBS (OFF) | Motor | 10 PD with STN DBS and dysarthria | 54 PD with STN DBS and dysarthria | 46 (OFF) | PET (H15O) |
| Task: Speech, silent articulation or silence production | |||||||
| Pinto | 2011 | Hand movement, NC vs PD (OFF) | Motor | 9 PD (15 NC) | 59 PD (55 NC) | 33 (ON) | fMRI |
| Task: Hand movement and/or speech production | |||||||
| Pinto | 2011 | Hand movement and speech production NC vs PD (OFF) | Motor | 9 PD (15 NC) | 59 PD (55 NC) | 33 (ON) | fMRI |
| Task: Hand movement and/or speech production | |||||||
| Poisson | 2013 | PD with mirror movements vs CPD (OFF) | Motor | 8 PD with mirror movements (6 CPD) | 59 PD with mirror movements (65 CPD) | 18 vs 16 (OFF) | fMRI |
| Task: index to thumb opposition movement | |||||||
| Politis | 2013 | Sexual vs neutral cue, PD with hypersexuality vs CPD (OFF) | Affective & Behavioural symptoms (Hypersexuality) | 12 PD with hypersexuality (12 CPD) | 55 PD with hypersexuality (62 CPD) | 40 vs 35 (OFF) | fMRI |
| Task: Rating the follow cues: dopaminergic drugs, appetizing food, money and gambling, sexual and neutral | |||||||
| Reijnders | 2010 | Anti‐correlation apathy scale (ON) | Affective & Behavioural symptoms (Apathy) | 55 PD | 62 PD | 17 (ON) | fMRI |
| Rest | |||||||
| Robert | 2012 | Correlation apathy scale (ON) | Affective & Behavioural symptoms (Apathy) | 45 PD | 61 PD | 8 (ON) | PET (18F FDG) |
| Rest | |||||||
| Rottschy | 2013 | Memory load, PD vs NC (ON) | Motor | 23 PD (23 NC) | 67 PD (65 NC) | 24 (ON) | fMRI |
| Task: memorize and retype variably long visuo‐spatial stimulus sequences after short or long delays (immediate or delayed serial recall) | |||||||
| Sawamoto | 2007 | PD vs NC (OFF) | Cognition (Executive) | 7 PD (9 NC) | 59 PD (61 NC) | 22 (OFF) | PET (H15O) |
| Task: mental calculation of the day of the week depending on clues and instruction. | |||||||
| Schwingeschuh | 2013 | Ankle movement, PD vs NC (ON) | Other | 20 PD (10 NC) | 67 PD (35 NC) | 24 (ON) | fMRI |
| Task: Cued ankle movement | |||||||
| Sheng | 2014 | PD with depression vs NC (ON) | Affective & Behavioural symptoms (Depression) | 20 PD (25 NC) | 56 PD (57 NC) | fMRI | |
| Rest | |||||||
| Sheng | 2014 | PD with depression vs CPD (ON) | Affective & Behavioural symptoms (Depression) | 20 PD (21 CPD) | 56 PD (57 CPD) | fMRI | |
| Rest | |||||||
| Shine | 2013a | PD with visual hallucination vs CPD (ON) | Affective & Behavioural symptoms (Visual Hallucination) | 9 PD (13 CPD) | 66 PD (62 CPD) | 27 vs 21 (ON) | fMRI |
| Task: Bistable percept paradigm | |||||||
| Shine | 2013a | PD with visual hallucination vs CPD (ON) | Affective & Behavioural symptoms (Visual Hallucination) | 9 PD (13 CPD) | 66 PD (62 CPD) | 27 vs 21 (ON) | fMRI |
| Rest | |||||||
| Shine | 2013b | Complex vs simple cues, CPD vs PD with freezing gait (OFF) | Motor | 14 PD with freezing gait (15 CPD) | 63 PD with freezing gait (63 CPD) | 32 vs 29 (OFF) | fMRI |
| Task: Walk based stop signal task with simple or complex (Stoop based) cues. | |||||||
| Shine | 2013c | Walk vs freezing (OFF) | Motor | 18 PD | 67 PD | 39 (OFF) | fMRI |
| Task: Walk based stop signal task with simple or complex (Stoop based) cues. | |||||||
| Song | 2014 | NC vs PD | Other | 33 PD (33 NC) | 71 PD (67 NC) | 14 | SPECT (TC99m) |
| Rest | |||||||
| Subramanian | 2011 | Feedback vs no feedback group (ON) | Cognition (Executive) | 10 PD | 64 PD | 15 (ON) | fMRI |
| Task: Hand movement guided with or without neurofeedback | |||||||
| Tan | 2015 | NC vs PD (OFF) | Bodily Awareness (Pain) | 14 PD (17 NC) | 63 PD (61 NC) | 22 (OFF) | fMRI |
| Task: Heat‐induced pain | |||||||
| Toxopeus | 2012 | Inhibition, NC vs PD (OFF) | Motor | 13 PD (19 NC) | 59 PD | 22 (OFF) | fMRI |
| Task: wrist movement divided in initiation, inhibition and gradual movement modulation. | |||||||
| Turner | 2013 | Movement‐related activity, NC vs PD (OFF) | Motor | 12 PD (12 NC) | 57 PD (58 NC) | 42 (OFF) | PET (H15O) |
| Task: predictive visuomotor tracking task | |||||||
| Ulla | 2010 | Manic vs euthymic induced by STN DBS (ON) | Affective & Behavioural symptoms (Hypomania) | 5 PD with hypomania | 62 PD with hypomania | 36 (OFF) | PET (H15O) |
| Rest | |||||||
| Ventre‐Dominey | 2014 | Spatial working memory, STN‐DBS ON vs OFF (OFF) | Cognition (Working Memory) | 13 PD | 55 PD | 9 DBS ON vs 39 DSB OFF (OFF) | PET (H15O) |
| Task: spatial and non‐spatial working color and movement association | |||||||
| Weder | 2000 | NC vs PD (OFF) | Cognition (Somatosensory discrimination) | 12 PD (12 NC) | 17 (ON) | PET (H15O) | |
| Task: finger exploratory discrimination | |||||||
| Welge‐Lussen | 2009 | Olfactory stimulation (ON) | Autonomic symptoms (Olfaction) | 18 PD | 59 PD | 28 (ON) | fMRI |
| Task: Olfactory stimulation | |||||||
| Wu | 2011 | PD vs NC (OFF) | Other | 18 PD (18 NC) | 62 PD (62 NC) | 22 (OFF) | fMRI |
| Rest | |||||||
| Ye | 2012 | PD vs NC (ON) | Cognition (Language) | 16 PD (16 NC) | 62 PD (64 NC) | 16 (ON) | fMRI |
| Task: Temporal connective comprehension | |||||||
PD: Parkinson's disease patient, NC: normal control, CPD: control group of PD patients, MCI: mild cognitive impairment, STN: subthalamic nucleus, DBS: deep‐brain stimulation, PPTg: nucleus tegmenti pedunculopontini, ON: ON medication (normal intake of medication), OFF: OFF medication (withdraw of medication over night).
English articles including original data
Idiopathic PD patients
No comparison with other brain pathologies
No pharmacological trials
3D coordinates reported in stereotactic space (MNI or Talairach)
Level of significance reported (p value, cluster or voxel level, correction)
For each study included, the coordinates located in the insular cortex were collected.
Meta‐Analysis Based on Activation Likelihood Estimation
To assess the functional role of the insular cortex in parkinsonian features, all imaging studies included in the meta‐analysis were sorted into five functional categories. A total of 132 insular foci were selected from 96 published experiments (Table 1, appendix) comprising the five functional categories: cognition (30 studies), affective/behavioral symptoms (24 studies), bodily awareness/autonomic function (8 studies), sensorimotor function (21 studies), and nonspecific resting functional changes associated with disease (13 studies). All Talairach coordinates were converted to MNI space using the Lancaster transform [Lancaster et al., 2007].
The ALE is a coordinate‐based meta‐analysis method. Activation maxima reported in studies (i.e., foci) are modeled as spatial 3D Gaussian probability distributions [Laird et al., 2005]. The distribution is centered at the reported coordinates and its size is directly related to the sample size (number of participants) [Eickhoff et al., 2009]. Once all the distributions of a study have been generated, a modeled activation map is created for the study [Turkeltaub et al., 2012]. The union of the modeled activation maps describing the convergence of results across studies at each voxel represents the ALE map. To identify true congruence from noise, permutation tests were performed. The ALE map was compared to the null distribution, a randomly distributed map, and tested for significance for each voxel [Eickhoff et al., 2012]. The meta‐analysis was performed with Ginger ALE software (http://brainmap.org/ale). Statistical significance was set at a family‐wise error corrected threshold of p < 0.05 with a cluster extent of 10 voxels. The ALE value, cluster volume (in voxels), and the MNI coordinates are reported for each analysis. The maps of the ALE values were superimposed on a colin.nii atlas [Laird et al., 2005] using the Mango software (http://ric.uthscsa.edu/mango//mango.html).
A meta‐analysis was first conducted on all the imaging studies combined together to assess the contribution of all experiments to the insular activity results, and then separate meta‐analyses were conducted on specific functional subcategories (cognition, sensorimotor, affective/behavioral symptoms) to evaluate the contribution of different experiments to the insular subregional findings. An analysis was also performed focusing only on blood flow studies (78: fMRI and H2O15 PET, TC99m SPECT) representing the largest set of data in order to limit the possible confounding effect from metabolic (18F FDG) or receptor imaging and to test consistency of the results. When possible, a meta‐analysis contrasting directly patients and controls (38 studies) was also performed and another set of analyses was conducted to estimate the effect of the medication on studies in which patients were evaluated ON and OFF medication.
RESULTS
Table 1 (appendix) summarizes the demographics and experimental conditions of the 96 imaging reports included in the meta‐analysis. Briefly, these included 20 PET studies with receptor imaging, 27 PET with H2O15 or TC99m, and 51 fMRI. The total number of experiments (96) included 1852 patients (age 63 y/o ± 5 SD) and 801 controls (age 61 y/o ± 7 SD) (Table 1, appendix). Thirty studies (31%) reported experiments related to cognition (executive function, memory, language, MCI, dementia, and so on), 24 studies (25%) reported experiments evaluative affective/behavioral symptoms, 8 studies (8%) were related to bodily awareness/autonomic function, 21 studies (22%) were associated with sensorimotor function, and 13 studies (14%) reported nonspecific changes associated with disease. In total, 30 studies evaluated the patients ON medications (562 patients), while 57 evaluated them OFF medication (830 patients).
The whole‐group ALE meta‐analysis across the 96 published experiments revealed significant convergence of activation maxima related to PD in different insular regions (Table 2 and Fig. 1). These clusters were located in the right ventral anterior insula (x = 38, y = 16, z = −2; ALE value = 0.050), left ventral anterior insula (x = −36, y = 18, z = −8; ALE value = 0.039), left dorsal posterior insula (x = −42, y = −12, z = 4; ALE value = 0.040), and right dorsal posterior insula (x = 38, y = −16, z = 4; ALE value = 0.039) (Fig. 1A).
Table 2.
Activation‐likelihood‐estimation: clusters from the whole group meta‐analysis
| Cluster | Side | k | ALE value | x | y | z |
|---|---|---|---|---|---|---|
| # 1 | R | 477 | 0.050 | 38 | 16 | −2 |
| R | 0.039 | 38 | −16 | 4 | ||
| # 2 | L | 143 | 0.039 | −36 | 18 | −8 |
| L | 0.024 | −32 | 24 | 4 | ||
| L | 0.021 | −42 | 6 | −8 | ||
| # 3 | L | 119 | 0.040 | −42 | −12 | 4 |
Location of significant convergence of activation maxima from the whole group meta‐analysis, thresholded at p < 0.05 (FWE‐corrected), the side (R: right, L: left), the number of voxels (k), the ALE value, and the MNI coordinates.
Figure 1.
ALE map of clusters showing changes in PD (p < 0.05). (A) Whole group analysis. (B) Effect of the medication.
When looking at those studies investigating only blood flow changes, there was an overlap with those clusters, in the right ventral anterior insula (x = 36, y = 16, z = 0; ALE value = 0.042), the left ventral anterior insula (x = −38, y = 18, z = −6; ALE value = 0.035), the right dorsal (x = 38, y = −16, z = 12; ALE value = 0.021), and the left dorsal insula (x = −40, y = −10, z = 8; ALE value = 0.019) (Table 3).
Table 3.
Clusters from studies investigating the blood flow changes
| Cluster | BA | Side | k | ALE value | x | y | z |
|---|---|---|---|---|---|---|---|
| # 1 | 48 | R | 337 | 0.042 | 36 | 16 | 0 |
| R | 0.034 | 44 | 12 | −6 | |||
| # 2 | 48 | L | 136 | 0.035 | −38 | 18 | −6 |
| 48 | L | 0.022 | −34 | 24 | 6 | ||
| 48 | L | 0.021 | −42 | 8 | −8 | ||
| # 3 | 48 | R | 15 | 0.021 | 38 | −16 | 12 |
| # 4 | 48 | L | 12 | 0.019 | −40 | −10 | 8 |
Location of significant convergence of activation maxima from the whole group meta‐analysis, thresholded at p < 0.05 (FWE‐corrected), the side (R: right, L: left), the number of voxels (k), the ALE value, and the MNI coordinates.
The ALE analysis performed on those studies comparing patients with healthy controls showed three significant clusters in similar regions, at the level of the right ventral anterior insula and the bilateral dorsal posterior insula (x = 40, y = −16, z = 2; ALE value = 0.022; x = −42, y = −12, z = 4; ALE value = 0.024; x = 36, y = 18, z = −4; ALE value = 0.019).
When looking at the effect of the medication, a significant convergence of activation maxima was observed for the different medication states (ON or OFF) (Fig. 1B). The ON medication studies (30 experiments) showed a significant convergence of activation maxima in the left and right ventral anterior insula (x = −38, y = 18, z = −8; ALE value = 0.022; x = 32, y = 26, z = −4; ALE value = 0.019), whereas in the OFF medication studies (57 experiments), convergence was localized more posteriorly, in the left and right dorsal posterior insula (x = −42, y = −12, z = 4; ALE value = 0.036; x = 38, y = −16, z = 4; ALE value = 0.037).
A significant convergence of activation maxima was also observed for the different functional subcategories (Table 4). In fact, combining cognitive and behavioral/affective domains (54 studies) showed a significant convergence of activation maxima in the left and right ventral anterior insula (x = −34, y = 20, z = −8; ALE value = 0.030; x = 38, y = 14, z = −4; ALE value = 0.027) and in the left and right dorsal posterior insula (x = −42, y = −12, z = 4; ALE value = 0.039; x = 38, y = −16, z = 4; ALE value = 0.037) (Fig. 2A). In contrast, the analysis limited only to the cognitive domain identified smaller overlapping clusters (Fig. 3A) in the left and right ventral anterior insula (x = −36, y = 20, z = −8; ALE value = 0.025; x = 38, y = 12, z = −4; ALE value = 0.022) and in the left and right dorsal posterior insula (x = −42, y = −12, z = 2; ALE value = 0.030; x = 36, y = −18, z = 6; ALE value = 0.025). The behavioral/affective symptoms also showed an overlapping cluster in the right ventral mid‐insula (x = 44, y = 2, z = −4; ALE value = 0.018) (Fig. 3B). When focusing on the effect of the dopaminergic medication on the behavioral/affective symptoms combined with cognition, a significant convergence of activation maxima was observed for the different medication stage (ON or OFF). The ON medication studies (21 experiments) showed a significant convergence of activation maxima in the left and right ventral anterior insula (x = −36, y = 18, z = −8; ALE value = 0.019; x = 32, y = 26, z = −4; ALE value = 0.019), whereas in the OFF medication studies (28 experiments), convergence was localized more posteriorly, in the left and right dorsal posterior insula (x = −42, y = −12, z = 4; ALE value = 0.035; x = 38, y = −16, z = 4; ALE value = 0.035) (Fig. 2B).
Table 4.
Activation‐likelihood‐estimation from different functional subcategories
| Cluster | Side | k | ALE value | x | y | z |
|---|---|---|---|---|---|---|
| Cognition and affective/behavioral symptoms | ||||||
| # 1 | R | 115 | 0.027 | 38 | 14 | −4 |
| R | 0.024 | 42 | 4 | −2 | ||
| R | 0.022 | 32 | 26 | −4 | ||
| # 2 | L | 115 | 0.039 | −42 | −12 | 4 |
| # 3 | R | 104 | 0.037 | 38 | −16 | 4 |
| # 4 | L | 54 | 0.030 | −34 | 20 | −8 |
| Cognition | ||||||
| # 1 | L | 85 | 0.030 | −42 | −12 | 2 |
| # 2 | R | 62 | 0.025 | 36 | −18 | 6 |
| # 3 | R | 52 | 0.022 | 38 | 12 | −4 |
| R | 0.022 | 30 | 26 | −4 | ||
| # 4 | L | 37 | 0.025 | −36 | 20 | −8 |
| Affective/behavioral symptoms | ||||||
| # 1 | R | 15 | 0.018 | 44 | 2 | −4 |
| Motor | ||||||
| # 1 | L | 32 | 0.018 | −42 | 6 | −8 |
| # 2 | R | 11 | 0.014 | 44 | 18 | −2 |
Location of significant convergence of activation maxima from the subcategory meta‐analyses, thresholded at p < 0.05 (FWE‐corrected), the side (R: right, L: left), the number of voxels (k), the ALE value, and the MNI coordinates.
Figure 2.
ALE map of clusters showing changes in PD (p < 0.05). (A) Cognitive and affective/behavioral symptoms analysis. (B) Effect of the medication on cognitive and affective/behavioral symptoms.
Figure 3.
ALE map of clusters showing changes in PD (p < 0.05). (A) Cognitive analysis. (B) Affective/behavioral symptoms.
A significant convergence of activation maxima related to the sensorimotor function (21 studies) was seen instead in the left mid‐insula (x = −42, y = 6, z = −8; ALE value = 0.018) and the right anterior insula (x = 44, y = 18, z = −2; ALE value = 0.014), which did not overlap with other domain‐related clusters (Table 4).
DISCUSSION
Surprisingly, no neuroimaging studies have directly addressed the contribution of the insula as a critical region in PD pathology. This study confirmed the importance of the insula in PD in acting as a hub for processing critical information related to the body state and for integrating cognitive–affective, sensorimotor, and autonomic information (Fig. 1). This report provides evidence of an important functional distribution of different domains within the insular cortex in PD, particularly in relation to nonmotor aspects, with changes related to the effect of disease and medication state.
When focusing on the cognitive and behavioral/affective domains of the disease, the insula show a bilateral involvement of both anterior and posterior regions (Figs. 2 and 3). The anterior involvement is quite consistent with accumulating research in healthy subjects showing that this region plays a central role in directing cognitive processes and implementing/maintaining task set [Dosenbach et al., 2006, 2007]. In conjunction with the anterior cingulate cortex (i.e., salience network), the anterior insula allows switching between neural networks required for executive functions [Menon and Uddin, 2010; Seeley et al., 2007; Swick et al., 2011]. It also has a well‐established contribution to processing affect and emotion and it is critically involved in social interactions requiring self‐awareness, interoception, and integration of both affective/emotional and environmental stimuli [Craig, 2009]. In other words, it is very likely that neurodegenerative processes affecting this region could disrupt both cognitive and socio/affective functions in PD [Christopher et al., 2014a; Nieuwenhuys, 2012].
An interesting finding was the observation in these PD patients (performing cognitive and behavioral/affective tasks) of a significant convergence of activation maxima in the posterior regions of the insula (Fig. 2B). While the biological explanation of this finding is not entirely clear, this may be the result of the dopaminergic changes described previously [Christopher et al., 2014b]. Indeed, when focusing on the effect of medication, we found some evidence supporting the role of dopamine depletion: PD patients OFF medication presented with a significant convergence of activation maxima mainly in dorsal posterior regions of the insula, while PD patients ON medication showed, in contrast, a more physiological involvement of bilateral anterior insula (Fig. 2B).
In patients studied OFF medication, the bodily sensations generated by the motor symptoms (i.e., bradykinesia and rigidity), it is possible that led to an abnormal activation of the posterior insula, an area generally implicated in the processing of position, movement, and sensation of the body [Chang et al., 2013; Cerasa et al,. 2006]. This abnormal interoceptive information of how the body “feels” could lead to an abnormal salience processing in the anterior insula affecting how affective/emotional sensations are perceived in patients with PD, which is partially restored by dopaminergic medication. Alternatively, the more posterior activation in the OFF medication state (during cognitive and behavioral/affective tasks) may simply be related to a compensatory activation [Daselaar et al., 2015] due to inadequate recruitment of the anterior insula. In contrast, when ON dopaminergic medication, patients might be better able to recruit the anterior insula during cognitive processing.
A significant convergence of activation maxima related specifically to sensorimotor tasks was seen in the mid‐insula confirming the role of this region in processing bodily awareness in relation to somatosensory information and coordination of movement.
Although the ALE technique overcomes the classical limitations of neuroimaging studies, it only reveals convergences of activity from different studies, not the actual activations. To reduce potential confounding factors, we conducted as well a number of different analysis limited to only blood flow studies and comparing patients to healthy controls. In any case, interpretations of the results should be made taking into consideration the limitation inherent to this technique. Despite this, the technique offers a valuable approach to investigate an under‐recognized region involved in the pathogenesis of Parkinson's disease. The insula is substantially affected by alpha‐synuclein deposition in PD and the insular abnormalities found in neuroimaging studies strongly point toward its contribution to a wide range of nonmotor symptoms, including somatosensory disturbances. Thus, as an important hub involved in integrating diverse information, the insula should be considered a region of interest when investigating cognitive and behavioral changes, as well as disruptions in viscerosensory or somatosensory processes in Parkinson's disease.
REFERENCES
- Arsalidou M, Duerden EG, Taylor MJ (2013): The centre of the brain: Topographical model of motor, cognitive, affective, and somatosensory functions of the basal ganglia. Hum Brain Mapp 34:3031–3054. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Braak H, Bohl JR, Müller CM, Rüb U, de Vos RA, Del Tredici K (2006): Stanley Fahn Lecture 2005: The staging procedure for the inclusion body pathology associated with sporadic Parkinson's disease reconsidered. Movement Disord 21:2042–2051. [DOI] [PubMed] [Google Scholar]
- Cauda F, D'Agata F, Sacco K, Duca S, Geminiani G, Vercelli A (2011): Functional connectivity of the insula in the resting brain. Neuroimage 55:8–23. [DOI] [PubMed] [Google Scholar]
- Cerasa A, Hagberg GE, Peppe A, Bianciardi M, Gioia MC, Costa A, Castriota‐Scanderbeg A, Caltagirone C, Sabatini U. (2006): Functional changes in the activity of cerebellum and frontostriatal regions during externally and internally timed movement in Parkinson's disease. Brain Res Bull 71:259–269. [DOI] [PubMed] [Google Scholar]
- Chang LJ, Yarkoni T, Khaw MW, Sanfey AG (2013): Decoding the role of the insula in human cognition: Functional parcellation and large‐scale reverse inference. Cereb Cortex 23:739–749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chaudhuri KR, Schapira AH (2009): Non‐motor symptoms of Parkinson's disease: Dopaminergic pathophysiology and treatment. Lancet Neurol 8:464–474. [DOI] [PubMed] [Google Scholar]
- Chikama M, McFarland NR, Amaral DG, Haber SN (1997): Insular cortical projections to functional regions of the striatum correlate with cortical cytoarchitectonic organization in the primate. J Neurosci 17:9686–9705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Christopher L, Koshimori Y, Lang AE, Criaud M, Strafella AP (2014a): Uncovering the role of the insula in non‐motor symptoms of Parkinson's disease. Brain 137:2143‐2143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Christopher L, Marras C, Duff‐Canning S, Koshimori Y, Chen R, Boileau I, Segura B, Monchi O, Lang AE, Rusjan P, Houle S, Strafella AP (2014b): Combined insular and striatal dopamine dysfunction are associated with executive deficits in Parkinson's disease with mild cognitive impairment. Brain 137:565–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Craig AD (2009): How do you feel—now? The anterior insula and human awareness. Nat Rev Neurosci 10:59–70. [DOI] [PubMed] [Google Scholar]
- Criaud M, Boulinguez P (2013): Have we been asking the right questions when assessing response inhibition in go/no‐go tasks with fMRI? A meta‐analysis and critical review. Neurosci Biobehav Rev 37:11–23. [DOI] [PubMed] [Google Scholar]
- Daselaar SM, Iyengar V, Davis SW, Eklund K, Hayes SM, Cabeza RE (2015): Less wiring, more firing: Low‐performing older adults compensate for impaired white matter with greater neural activity. Cereb Cortex 25:983–990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dosenbach NU, Visscher KM, Palmer ED, Miezin FM, Wenger KK, Kang HC, Burgund ED, Grimes AL, Schlaggar BL, Petersen SE (2006): A core system for the implementation of task sets. Neuron 50:799–812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dosenbach NU, Fair DA, Miezin FM, Cohen AL, Wenger KK, Dosenbach RA, Fox MD, Snyder AZ, Vincent JL, Raichle ME, Schlaggar BL, Petersen SE (2007): Distinct brain networks for adaptive and stable task control in humans. Proc Natl Acad Sci USA 104:11073–11078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eickhoff SB, Laird AR, Grefkes C, Wang LE, Zilles K, Fox PT (2009): Coordinate‐based activation likelihood estimation meta‐analysis of neuroimaging data: A random‐effects approach based on empirical estimates of spatial uncertainty. Hum Brain Mapp 30:2907–2926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Eickhoff SB, Bzdok D, Laird AR, Kurth F, Fox PT (2012): Activation likelihood estimation revisited. Neuroimage 59:2349–2361. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Halliday GM, Li YW, Blumbergs PC, Joh TH, Cotton RG, Howe PR, Blessing WW, Geffen LB (1990): Neuropathology of immunohistochemically identified brainstem neurons in Parkinson's disease. Ann Neurol 27:373–385. [DOI] [PubMed] [Google Scholar]
- Herz DM, Eickhoff SB, Løkkegaard A, Siebner HR (2014): Functional neuroimaging of motor control in parkinson's disease: A meta‐analysis. Hum Brain Mapp 35:3227–3237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kurth F, Zilles K, Fox PT, Laird AR, Eickhoff SB (2010): A link between the systems: Functional differentiation and integration within the human insula revealed by meta‐analysis. Brain Struct Funct 214:519–534. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Laird AR, Fox M, Price CJ, Glahn DC, Uecker AM, Lancaster JL, Turkeltaub PE, Kochunov P, Fox PT (2005): ALE meta‐analysis: Controlling the false discovery rate and performing statistical contrasts. Hum Brain Mapp 25:155–164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lancaster JL, Tordesillas‐Gutierrez D, Martinez M, Salinas F, Evans A, Zilles K, Mazziotta JC, Fox PT (2007): Bias between MNI and Talairach coordinates analyzed using the ICBM‐152 brain template. Hum Brain Mapp 28:1194–1205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Menon V, Uddin LQ (2010): Saliency, switching, attention and control: A network model of insula function. Brain Struct Funct 214:655–667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mutschler I, Wieckhorst B, Kowalevski S, Derix J, Wentlandt J, Schulze‐Bonhage A, Ball T (2009): Functional organization of the human anterior insular cortex. Neurosci Lett 457:66–70. [DOI] [PubMed] [Google Scholar]
- Nieuwenhuys R (2012): The insular cortex: A review. Progr Brain Res 195:123–163. [DOI] [PubMed] [Google Scholar]
- Pan PL, Song W, Shang HF (2012): Voxel‐wise meta‐analysis of gray matter abnormalities in idiopathic Parkinson's disease. Eur J Neurol 19:199–206. [DOI] [PubMed] [Google Scholar]
- Park A, Stacy M (2009): Non‐motor symptoms in Parkinson's disease. J Neurol 256:293–298. [DOI] [PubMed] [Google Scholar]
- Shao N, Yang J, Li J, Shang HF (2014): Voxelwise meta‐analysis of gray matter anomalies in progressive supranuclear palsy and Parkinson's disease using anatomic likelihood estimation. Front Hum Neurosci 8:63. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Seeley WW, Menon V, Schatzberg AF, Keller J, Glover GH, Kenna H, Reiss AL, Greicius MD (2007): Dissociable intrinsic connectivity networks for salience processing and executive control. J Neurosci 27:2349–2356. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Swick D, Ashley V, Turken U (2011): Are the neural correlates of stopping and not going identical? Quantitative meta‐analysis of two response inhibition tasks. Neuroimage 56:1655–1665. [DOI] [PubMed] [Google Scholar]
- Turkeltaub PE, Eickhoff SB, Laird AR, Fox M, Wiener M, Fox PT (2012): Minimizing within‐experiment and within‐group effects in activation likelihood estimation meta‐analyses. Hum Brain Mapp 33:1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Uddin LQ, Kinnison J, Pessoa L, Anderson ML (2014): Beyond the tripartite cognition‐emotion‐interoception model of the human insular cortex. J Cogn Neurosci 26:16–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wager TD, Lindquist MA, Nichols TE, Kober H, Van Snellenberg JX (2009): Evaluating the consistency and specificity of neuroimaging data using meta‐analysis. Neuroimage 45:S210–S221. [DOI] [PMC free article] [PubMed] [Google Scholar]