Task-Related Brain Connectivity Activation Functional Magnetic Resonance Imaging in Intellectual Disability Population: A Meta-Analytic Study

Abstract

Introduction:

Neuroimaging studies of intellectual disability (ID) have been published over the last three decades, but the findings are often inconsistent, and therefore, the neural correlates of ID remain elusive. This article aims to study the different publications in task-functional magnetic resonance imaging (fMRI) and different ID populations to make a qualitative and quantitative analysis on this field.

Methods:

After duplicates were removed, only 10 studies matching our inclusion criteria were incorporated. Moreover, a quality assessment of the included studies was done. Qualitative results of the different articles were analyzed, separated by type of task and type of ID. Seed-based d mapping (SDM) software was used.

Results:

The right temporal gyrus was more activated in control subjects than in ID. Concretely, the right temporal gyrus is implicated in many cognitive domains as semantic memory processing and language. Moreover, it can be highly influenced by the type of task used in every study. Heterogeneity was not detected. A jackknife sensitivity analysis was also estimated to improve the analysis reliability, and both results were confirmed.

Conclusions:

More task-fMRI studies on ID must be published to add larger samples to address the pathophysiological questions more directly.

Impact statement

In this article, the state-of-the-art in the field of functional magnetic resonance imaging (fMRI) and intellectual disability (ID) is reviewed. Moreover, we perform a meta-analysis of every article's results to summarize the principal outcomes in the field. It is very relevant because it has become the first meta-analytic study to overcome all the principal studies published in fMRI and ID to find the principal neurological substrates while the subjects are performing a task.

Introduction

Intellectual disability (ID) is a common lifetime condition that significantly impacts the individuals and their family. Important limitations in cognitive performance and adaptative behaviors characterize this population (Schalock et al., 2010). This definition is proposed by the American Association on Intellectual and Developmental Disabilities (AAIDD). It highlights the importance of the social and ecological models focused on the person and the environment. At the same time, it recognizes the importance of the individualized support application to improve the person's functioning. The conception toward people with ID has changed, and nowadays, it is not only focused on intellectual functioning (Tassé et al., 2016).

The recognition of the three intelligence components and their correlation with adaptative behavior (conceptual, social, and practical skills) has brought up a new approach toward people with ID that offers new perspectives and intervention models linked with supports. Due to the individualized support mediation, optimal environments for the development and well-being of the person with ID can be promoted (Luckasson et al., 2002). Nowadays, there is a firm conviction that with adequate support application, a person's functional abilities can be enhanced, and his or her results and quality of life can be improved.

From this perspective of ID and considering that it answers to the environmental demands thanks to the support's mediation, we ask ourselves if the brain functioning of people with ID could present some irregularities. In other words, due to the adaptative behavior and the individualized supports, people with ID can develop and have a quality of life, despite their low intellectual functioning. The brain connectivity patterns of people with ID could be different from the brain connectivity patterns of people without ID.

To study brain connectivity, our principal aim is to study the different publications in functional magnetic resonance imaging (fMRI) and people with ID.

As it is well-known, fMRI estimates functional connectivity, which describes the temporal dependence between different brain areas (Friston et al., 1993). The study of the brain as an integrative network of functional connections contributes to the human behavior knowledge and also provides helpful information for the comprehension of the neuropsychiatric disease (Greicius et al., 2007; Greicius, 2008), as well as the influence of aging in cognitive function (Damoiseaux et al., 2008). The study of specific populations, as ID population, assumes that limitations in different areas result in similar functional abnormalities in brain signal (Reiss et al., 2000).

In this sense, neuroimaging studies have been published over the last three decades (Azuma et al., 2009, 2015; Bernardino et al., 2014) in ID field. For instance, Vega and associates (2015) compared the resting-state functional connectivity in three different populations: Down syndrome (DS), Williams syndrome (WS), and healthy controls. Their findings suggest that differences in brain connectivity can be found within neurodevelopmental populations with ID compared with controls.

In general, sample sizes in these types of studies are tiny, and therefore, the effects found in the different studies are often low and inconsistent (Button et al., 2013). Consequently, researchers are driven into different methodologies as thresholding and analysis that may increase the type I errors (Eklund et al., 2016). In addition, these single imaging studies do not assemble the whole concept of ID, they are about a specific ID population, and they aim to find a specific pattern of the concrete ID population (Azuma et al., 2009, 2015; Bernardino et al., 2014).

Therefore, the neural correlates of ID persist unknown. In the field of neuroimaging, we can only find few meta-analyses that describe a single population with ID focused only on 22q11.2 deletion syndrome (22q11 DS; Scarpazza et al., 2019) or WS (Binelli et al., 2014).

From our perspective, some explanations justify why there is no meta-analysis found in fMRI with ID. First, problems that take place when using fMRI register may be exacerbated for the ID population (Carbó-Carreté et al., 2020). For instance, regarding the DS population, movements during registration can be highly increased. Consequently, experimental issues can arise (Pujol et al., 2014). In ID population, it can be a challenging task to be quiet and stay focused during the registration (Lightbody and Reiss, 2009).

Nevertheless, with considerable training and instruction for the scan (Fassbender et al., 2017), movements can be reduced. Therefore, in the last few years, some researchers have achieved the study of brain signal in the ID population. Second, it is not easy to sample this population (Carbó-Carreté et al., 2020), and people with more common disabilities will probably be more represented than people with more rare disabilities.

To date, it is therefore unclear if common aberrant connectivity patterns associated with ID can be found (Walter et al., 2009).

In this sense, as it has been mentioned above, the findings of Vega and colleagues (2015) show an altered between-network connectivity common in WS and DS. They suggest that this could be characteristic of different disorders coursing with ID. Their findings highlight the utility of using a task-based procedure, a noninvasive technique that measures brain activity while participants perform different types of tasks (e.g., executive functions, language, memory, motor, among others). This technique could help resolve the association between brain function and intellectual functioning across different disorders coursing with ID (Vega et al., 2015).

Therefore, based on previous literature, this article aims to study the different publications in task-fMRI and different ID populations to make a qualitative and quantitative analysis in this field. Previous studies sustain the reason to study task-fMRI designs such as the one of Vega and associates (2015), outlining this procedure's utility. We hope to find a pattern in different brain abnormalities with ID compared with healthy controls. Moreover, to guarantee solid and valid results, additional heterogeneity and jackknife sensitivity analysis were estimated. In this study, Seed-based d mapping (SDM) was used. It is a meta-analysis tool for neuroimaging data that has proved high levels of validity and consistency (Iwabuchi et al., 2015; Li et al., 2019; Radua and Mataix-Cols, 2009; Radua et al., 2012; Sheng et al., 2015).

Methods

Inclusion of studies

Following PRISMA guidelines, an extensive review was performed to identify studies published from January 2009 until September 3, 2020. Also, the search, the inclusion criteria, and the article selection were made by three independent investigators and there was a complete agreement between them for the study search and selection. The search was carried out in PubMed and Web of Science (WoS) databases, and the Boolean algorithm with the keywords used is presented in the Supplementary Appendix SA1. This Boolean algorithm was created based on a systematic search conducted on the six principal journals of ID. The more frequent diseases coursing with ID were included in the search.

Moreover, all the meta-analyses published in the last 5 years in those journals were revised to find the best Boolean algorithm. However, autism spectrum disorder was excluded because there is a significant variability in this population's characteristics, and ID is not the principal one. We consider that this population deserves a unique study for its characteristics. Moreover, it is still discussed the presence of ID (Thurm et al., 2019).

The inclusion criteria for the studies were as follows: (1) functional MRI scan was conducted with a specific task during the register, (2) functional activation in the brain was compared between ID persons and the control group without ID, and (3) only studies that report the specific whole-brain coordinates in Talairach or Montreal Neurological Institute (MNI) spaces were included. The selection process is shown in Figure 1.

FIG. 1.

Flowchart of the meta-analysis search conducted.

The literature review yielded 775 studies found in the WoS database and 84 studies in the PubMed database. One hundred seventy-six records were removed due to duplication in both databases. Six hundred forty-seven studies were excluded after title/abstract screening because they did not accomplish the inclusion criteria.

After this process, only 35 articles remained. However, after the full-text screening, 9 articles were excluded because they did not inform about the peak's activation coordinates, 5 articles were excluded because they did not analyze the whole brain, 6 studies did not compare groups or did not have a second-level analysis, 2 articles were excluded because the participants did not have ID, 1 article was excluded because it was not an original research,1 record was excluded because it was resting-state fMRI and no task was performed. Finally, 1 article was excluded because it had no control group. Only 10 studies matching our inclusion criteria were included and are marked with an * in the reference list.

Voxel-wise meta-analysis

SDM software was used to estimate the differences between the ID population and control people without ID. The procedure and methods of the tool have been described elsewhere (Iwabuchi et al., 2015; Li et al., 2019; Radua and Mataix-Cols, 2009; Radua et al., 2012; Sheng et al., 2015). The SDM uses the peak coordinates found in every included article to reproduce an effect size map of the differences between ID and controls.

First, the areas that were statistically significant differentiating between ID and controls were reported. All the included studies must have the same thresholding at the whole-brain level to assure possible future comparison. Second, the coordinates previously reported were established with a standard MNI map of the group difference effect size based on their peak t value through a non-normalized Gaussian kernel to the voxels near the peak. This procedure assigns larger values to the voxels nearby peaks. By trying to avoid voxels from appearing both positive and negative simultaneously, the same map was used to represent the positive and negative coordinates (Radua and Mataix-Cols, 2009). The aforementioned positive coordinates imply increased brain activation, whereas the negative coordinates estimate the decreased brain activation.

Considering the sample size, the variance of each study, and the between-study heterogeneity, the mean map was estimated. Finally, a usual p-value of 0.05 was used as a threshold with an additional peak height of z = 127.

Furthermore, we performed a descriptive analysis according to the ID sample and the fMRI tasks of the included studies. Concretely, after the calculation of the cohen's d and the confidence intervals (CIs), a descriptive analysis of every article's results was resumed in different images and tables to clarify the results obtained in every included study, separately by the type of task and by the type of ID. We pretend, with this description, to summarize and report the different results distinguishing by ID type and task type.

To validate and test the replicability of the results, a jackknife sensitivity analysis was performed. This analysis systematically excludes one study of the data set and establishes whether the results remained significant.

Quality assessment

A 10-point checklist based on previous meta-analyses (Chen et al., 2015; Du et al., 2014; Shepherd et al., 2012) was used to assess the quality of the included studies through an evaluation of the imaging-specific methodology and some demographic and clinical areas of the included studies. The quality assessment is shown in Supplementary Appendix SA2. Two authors reviewed the included articles and independently determined a complete rating. An 80% rate agreement was found between both investigators in the quality assessment. In those articles where different scores were obtained, an agreement score was obtained after discussing the two different scores.

Some variables at this point were coded for each chosen study to make the quality assessment and to be able to study the possible heterogeneity. Type of population, sample size in each group, % of women, and mean age are some variables codified in Table 1. Moreover, the quality assessment is also found in Table 1.

Table 1.

Demographic and Clinical Characteristics of the Participants in the 10-Neuroimaging Data Included in This Meta-analysis

Study	Imaging modality	n		ID type	% Females		Mean age		Task	Classification of task	Quality assessment
Study	Imaging modality	ID	Controls	ID type	ID	Controls	ID	Controls	Task	Classification of task	Quality assessment
Azuma et al. (2009)	1.5 T MRI	8	13	22q11DS	50	35.71	12	13	Visuospatial memory task	Visual perception and executive functions	8.5
Azuma et al. (2015)	1.5 T MRI	14	14	22q11DS	100	38.46	13	13	Facial emotion processing	Emotion recognition and visual perception	9.5
Bernardino et al. (2014)	3.0 T MRI	7	9	WS	57.14	44.40	21.57	21.22	High-level visual categorical task	Visual perception and executive functions	9
Binelli et al. (2016)	1.5 T MRI	20	20	WBS	45	40	25.42	25.65	Control and facial emotion processing task	Visual perception	9.5
Hall et al. (2009)	3.0 T MRI	10	10	FXS	100	100	18.67	14.74	Auditive perception task	Auditive perception	8.5
Montojo et al. (2013)	3.0 T MRI	16	25	22q11DS	63	60	22.5	23	Stop-signal task	Executive functions	9.5
Montojo et al. (2014)	3.0 T MRI	15	30	22q11DS	40	40	23.88	24.36	Spatial capacity working memory	Executive functions	9.5
Reynolds Losin et al. (2009)	3.0 T MRI	9	9	DS	44.4	44.4	22	17.8	Story listening task	Language, auditive and visual perception	7
Wan et al. (2017)	3.0 T MRI	38	38	PWS	28.9	28.9	13.17	13.07	T-HVOT and FPMT	Visual perception	8
Woodcock et al. (2010)	3.0 T MRI	8	8	PWS	37.5	37.5	20.7	21	Switching and location	Visual perception and executive functions	7.5

n, sample size of every group.

22q11DS, 22q11.2 deletion syndrome; DS, Down syndrome; FPMT, full picture matching test; FXS, fragile-X syndrome; ID, intellectual disability; MRI, magnetic resonance imaging; PWS, Prader–Willi syndrome; T-HVOT, two choice revised version of Hooper visual organization test; WBS, Williams–Beuren syndrome; WS, Williams syndrome.

Results

Studies included in the meta-analysis

In Table 1, the different characteristics of the studies are shown. Four studies had 22q11 DS population; two studies had Prader–Willi syndrome, one DS, one WS, one fragile-X syndrome, and finally, one Williams–Beuren syndrome. All these disorders course with ID. Regarding the tasks conducted in the resonator, these were classified in visual, audition, language perception, and executive function tasks. The specific task in every experiment is found in Table 1. As shown in the same table, some studies had different tasks that included different regions. In this case, the study was separated to analyze the coordinates separately. In Table 1, we can also find the imaging modality used in every study, and in addition, we can see the sample size of every sample. Finally, in the last column, we can see the quality assessment made by two investigators separately.

Qualitative study

As mentioned above, to our knowledge, it is the first time that a meta-analysis is done using different types of ID. Therefore, we first analyze all the articles' results and then expose the meta-analysis results.

Table 2 shows the principal results obtained in the first step of the program. In this sense, for each study, both the smallest and higher possible effect size of each voxel are estimated, fixing a lower and upper bound of the possible effect sizes (Albajes-Eizagirre et al., 2009).

Table 2.

Peak Coordinates Obtained in the Different Studies Included in the Meta-analysis

Study	Cohen's D	Coordinates	Area	Peak	Comparison
Azuma et al. (2009)	1.003	24, −78, 12	Occipital precuneus right	Max	22q11DS<C
Azuma et al. (2015)	1.246	−34, −80, −20	Left cerebellum	Max	22q11DS<C
Bernardino et al. (2014)	1.480	24, −54, 30	Caudal intraparietal sulcus right hemisphere	Max	WS < C
Bernardino et al. (2014)	−1.402	−56, −58, 4	Left middle temporal gyrus	Min	WS > C
Binelli et al. (2016)	1.106	44, −62, 8	Right lateral occipital area	Max	WBS < C
Binelli et al. (2016)	−1.024	64, −18, −16	Right middle temporal gyrus	Min	WBS > C
Hall et al. (2009)	−1.859	−4, −24, −38	Left pons	Min	FXS > C
Montojo et al. (2013)	1.576	−46, −52, 44	Left occipital cortex, angular gyrus, supramarginal gyrus, superior parietal lobule	Max	22q11DS<C
Montojo et al. (2013)	−1.219	36, 22, 32	Right inferior/middle frontal gyrus	Min	22q11DS>C
Montojo et al. (2014)	1.530	−26, −66, 54	Left superior parietal lobule, intraparietal sulcus, supramarginal gyrus, angular gyrus	Max	22q11DS<C
Reynolds Losin et al. (2009)	2.835	52, −30, 0	Right middle temporal gyrus	Max	DS < C
Reynolds Losin et al. (2009)	−1.879	16, −66, 48	Right precuneus	Min	DS > C
Wan et al. (2017)	0.697	24, −56, 42	Right precuneus	Max	PWS < C
Woodcock et al. (2010)	2.513	−26, −6, 26	Middle frontal gyrus/subgyral white matter/insula left	Max	PWS < C
Woodcock et al. (2010)	−1.476	−18, 64, −16	Inferior, anterior frontal pole right	Min	PWS > C

Max, maximum peak; Min, minimum peak.

Figures 2–5 were visualized with the BrainNet Viewer (Xia et al., 2013). Figure 2 represents the different positive and negative coordinates of the results in every article after the first analysis made with SDM. Moreover, the size of every node is proportioned to Cohen's d calculated through the meta-analysis, and so, the bigger the node is, the more total the effect. More concretely, Table 3 shows the same results classified by type of ID. Figure 3 represents the different peak coordinates separating by type of ID. Another time, the node's size is proportioned to Cohen's d.

FIG. 2.

Representation of the positive and negative peak coordinates obtained in the different studies included in the meta-analysis. The size of every node is proportioned to the Cohen's d calculated through the meta-analysis. Blue color represents the positive peaks and green color represents the negative peaks. 1: Azuma et al. (2009); 2: Azuma et al. (2015); 3: Bernardino et al. (2014); 4: Binelli et al. (2016); 5: Hall et al. (2009); 6: Montojo et al. (2013); 7: Montojo et al. (2014); 8: Reynolds Losin et al. (2009); 9: Wan et al. (2017);10: Woodcock et al. (2010).

FIG. 3.

Representation of the coordinates obtained in the different studies included in the meta-analysis classified by the different types of ID. The size of every node is proportioned to the Cohen's d calculated through the meta-analysis. Blue color represents the positive peaks and green color represents the negative peaks. 1: 22q11DS; 2: Williams syndrome; 3: Williams–Beuren syndrome; 4: fragile-X-syndrome; 5: Down syndrome; 6: Prader–Willis syndrome. 22q11DS, 22q11.2 deletion syndrome; ID, intellectual disability.

FIG. 5.

Representation of the differences in brain activation between ID patients and controls in the meta-analysis.

Table 3.

Peak Coordinates Classified by Intellectual Disability Type

ID type	Cohen's D	Coordinates	Area	Peak	Comparison
22q11DS	1.003	24, −78, 12	Occipital precuneus right	Max	22q11DS<C
	1.246	−34, −80, −20	Left cerebellum	Max	22q11DS<C
	1.576	−46, −52, 44	Left occipital cortex, angular gyrus, supramarginal gyrus, superior parietal lobule	Max	22q11DS<C
	−1.219	36, 22, 32	Right inferior/middle frontal gyrus	Min	22q11DS>C
	1.530	−26, −66, 54	Left superior parietal lobule, intraparietal sulcus, supramarginal gyrus, angular gyrus	Max	22q11DS<C
WS	1.480	24, −54, 30	Caudal intraparietal sulcus right hemisphere	Max	WS < C
WS	−1.402	−56, −58, 4	Left middle temporal gyrus	Min	WS > C
Williams–Beuren syndrome	1.106	44, −62, 8	Right lateral occipital area	Max	WBS < C
Williams–Beuren syndrome	−1.024	64, −18, −16	Right middle temporal gyrus	Min	WBS > C
Fragile-X-syndrome	−1.859	−4, −24, −38	Left pons	Min	FXS > C
DS	2.835	52, −30, 0	Right middle temporal gyrus	Max	DS < C
DS	−1.879	16, −66, 48	Right precuneus	Min	DS > C
Prader–Willis syndrome	0.697	24, −56, 42	Right precuneus	Max	PWS < C
	2.513	−26, −6, 26	Middle frontal gyrus/subgyral white matter/insula left	Max	PWS < C
	−1.476	−18, 64, −16	Inferior, anterior frontal pole right	Min	PWS > C

We can see that the activated areas in 22q11 DS syndrome in front of controls were both the right inferior and middle frontal gyrus. The deactivation areas of 22q11 DS syndrome in front of controls were right occipital precuneus, left cerebellum, left occipital cortex, angular gyrus, supramarginal gyrus, superior parietal lobe, and intraparietal sulcus. In the case of Prader–Willi, the more activated areas were inferior and the anterior frontal pole right compared with controls, and the more deactivated areas were right precuneus and middle frontal gyrus. Finally, in the same way, Table 4 shows the peak coordinates classified by type of tasks. Figure 4 shows the results regarding the type of tasks used following the classification in Table 1.

FIG. 4.

Representation of the coordinates obtained in the different studies included in the meta-analysis classified by the different tasks used in the different studies. The size of every node is proportioned to the Cohen's d calculated through the meta-analysis. Blue color represents the positive peaks and green color represents the negative peaks. 1: Visual perception and executive functions; 2: emotion recognition and visual perception; 3: visual perception; 4: auditive perception; 5: executive functions; 6: language, auditive, and visual perception.

Table 4.

Peak Coordinates Classified by Type of Task Used in the Different Studies

Type of task	Cohen's D	Coordinates	Area	Peak	Comparison
Visual perception and executive functions	1.003	24, −78, 12	Occipital precuneus right	Max	22q11DS<C
	1.480	24, −54, 30	Caudal intraparietal sulcus right hemisphere	Max	WS < C
	−1.402	−56, −58, 4	Left middle temporal gyrus	Min	WS > C
	2.513	−26, −6, 26	Middle frontal gyrus/subgyral white matter/insula left	Max	PWS < C
	−1.476	−18, 64, −16	Inferior, anterior frontal pole right	Min	PWS > C
Emotion recognition and visual perception	1.246	−34, −80, −20	Left cerebellum	Max	22q11DS<C
Visual perception	1.106	44, −62, 8	Right lateral occipital area	Max	WBS < C
	−1.024	64, −18, −16	Right middle temporal gyrus	Min	WBS > C
	0.697	24, −56, 42	Right precuneus	Max	PWS < C
Auditive perception	−1.859	−4, −24, −38	Left pons	Min	FXS > C
Executive functions	1.576	−46, −52, 44	Left occipital cortex, angular gyrus, supramarginal gyrus, superior parietal lobule	Max	22q11DS<C
	−1.219	36, 22, 32	Right inferior/middle frontal gyrus	Min	22q11DS>C
	1.530	−26, −66, 54	Left superior parietal lobule, intraparietal sulcus, supramarginal gyrus, angular gyrus	Max	22q11DS<C
Language, auditive, and visual perception	2.835	52, −30, 0	Right middle temporal gyrus	Max	DS < C
Language, auditive, and visual perception	−1.879	16, −66, 48	Right precuneus	Min	DS > C

Compared with controls, people with ID in visual perception and executive function tasks demonstrated increased brain activation in left middle temporal gyrus and inferior anterior frontal pole right. However, they showed decreased activation compared with controls in occipital precuneus right, caudal intraparietal sulcus of the right hemisphere, and middle frontal gyrus. In visual perception tasks, people with ID appeared to have higher activation in the right middle temporal gyrus and lower activation in the right lateral occipital area and right precuneus. Concerning executive functions tasks, people with ID displayed a higher activation in the right inferior/middle frontal gyrus and decreased activation in left occipital cortex, angular gyrus, supramarginal gyrus, superior parietal lobe, and intraparietal sulcus.

Meta-analysis result

In Supplementary Appendix SA3, the forest plot of the studies with their effect size, low CI, and high CI is shown. As we can see, the more significant effects are found in the study of Woodcock and colleagues (2010), and as in all articles CI 0 is not included, there are significant effects in all the studies.

In the meta-analysis, people with ID did not show any increased brain activation compared with controls. However, people with ID showed decreased activation in right temporal gyrus, concretely, in the right middle temporal gyrus, in Brodmann area (BA) 37, compared with controls. Two clusters were found in this same area, one of two voxels and one of four voxels. The results are shown in Table 5. Moreover, in Figure 5, we can see the represented areas of the study.

Table 5.

Differences of Brain Activation Between Intellectual Disability Patients and Controls in the Main Meta-analysis

Region	MNI coordinates			SDM-Z score	p	Voxels	Cluster breakdown (voxels)
Region	x	y	z	SDM-Z score	p	Voxels	Cluster breakdown (voxels)
ID<HC (A) Right temporal gyrus	48	−62	6	−1.688	0.045716286	4	Right middle temporal gyrus, BA 37
(B) Right temporal gyrus	46	−66	6	−1.675	0.047921253	2	Right middle temporal gyrus, BA 37

BA, Brodmann area; MNI coordinates, Montreal Neurological Institut; SDM, Seed-based d mapping.

To determine the possible heterogeneity across studies, Q and I ² statistics were calculated. The results are shown in Supplementary Appendix SA4. In positive and negative peaks, nonsignificant Q statistics were found, and so, the results demonstrate that the variability is due to random variation rather than study heterogeneity. Due to the nonexistence of heterogeneity in the effect sizes, it is unnecessary to study the possible moderate variable effects. Therefore, it was decided not to conduct a meta-regression with external variables.

Reliability analysis

The jackknife sensitivity analysis (Supplementary Appendix SA5) reported that the right middle temporal gyrus (with coordinates in 48, −62, 69, and BA 37) was replicable in all 16 data sets. However, the same right middle temporal gyrus in different coordinates (46, −66, 6, and BA 37) was replicable in 14 data sets. This reveals a significant reliability of the results reported in this study.

Discussion

This study aimed to identify brain abnormalities mainly associated with ID. To the best of our knowledge, it is the first meta-analysis exploring differences in brain activation between people with ID and control subjects. In this sense, it is important to highlight that although our aim and our search were centered on all types of ID, our search concludes that only studies with fMRI in different genetical syndromes associated with ID exist. The inclusion of genetic syndromes only could be because of the facility identifying the genetic syndromes and increasing the possibilities to control the different variables. The studies in fMRI with this population are very recent, and we hope that more studies with different types of ID are published in some years.

Regarding the qualitative part of the study, there are common patterns regarding the type of ID and the type of task that are important to highlight.

About the type of ID, in 22q11 DS, we can find an overall deactivation in different areas of the left occipital cortex and lobule, as well as of the superior parietal lobule. This is congruent with some abnormalities in the thickness of the occipital cortex found in this population in a group with psychosis (Sun et al., 2020). In the case of Prader–Willi syndrome, we can find a deactivation in comparison with the control population near the insula in both hemispheres.

Regarding the type of task used in the study, in visual perception and executive functions, deactivation in the occipital and parietal areas, left and right, was found. These areas have been identified as necessary in visual perception (Kaas et al., 2010). However, there is an activation in ID compared with controls in the middle temporal gyrus and in the frontal pole right. The most interesting result is in executive functions, where we can find a clear common pattern. We can see that there is deactivation in the left occipital cortex, supramarginal gyrus, superior parietal gyrus, and angular gyrus, also congruent with certain studies about executive functions (Fassbender et al., 2017).

Compared with controls, people with ID did not show any increased activation in the quantitative analysis of every study. However, people with ID showed decreased activation in two areas located in the right middle gyrus compared with controls. This area comprised the BAs 21, 22, 37, and 39. In particular, in our case, only the BA number 37 was deactivated in the case of ID population compared with controls. Early functional neuroimaging articles have demonstrated that the middle temporal gyrus is involved in specific cognitive domains (Cabeza and Nyberg, 2000; Laufer et al., 2011; Raposo et al., 2006; Sass et al., 2009), for instance, semantic memory processing or language (Cabeza and Nyberg, 2000; Chao et al., 1999; Tranel et al., 1997). These domains can be associated with 70% of the tasks used in the studies included in the meta-analysis.

In schizophrenia, functional deficits in these cognitive domains, that is, language (Kuperberg et al., 1998), semantic memory (Nestor et al., 1998), and complex visual perception (Tek et al., 2002), have been demonstrated. In the same line, functional deficits, concretely, in the middle temporal gyrus (Chang et al., 2016; Fusar-Poli et al., 2011) have been found in current studies with samples at high risk of psychosis.

In a recent meta-analysis where the aim was to study the differences between 22q11 DS and controls, some regional differences were identified. Specifically, the results reported that the 22q11 DS population has structural and functional abnormalities. In particular, these abnormalities were mainly found within the right precuneus and superior temporal gyrus, bilateral inferior parietal lobe, and posterior cingulate cortex (Scarpazza et al., 2019). This could be congruent with our findings because 40% of our sample has 22q11 DS.

Moreover, two articles were shared between this paper and the meta-analyisis from Scarpazza et al. (2019). In the same line, a recent study (Marshall et al., 2017) conducted with 321 participants identified the 22q11 DS gene as one of the most linked to psychosis (Marshall et al., 2017). Consequently, the middle temporal gyrus and inferior temporal gyrus could be related to the appearance of schizophrenia and therefore may have a relationship with 22q11 DS.

Limitations

This study has several limitations. First, there exists a scarcity of studies in task-fMRI with the ID population. This is probably attributable to the relatively low commonness of this population and, in addition, the difficulties of using a resonator in this particular disease. Therefore, all our studies analyzed consisted of a tiny sample, and also, there are few studies in these meta-analyses. It is important to remark that the available studies do not allow us to draw clear conclusions about brain functional irregularities in ID. We should interpret our results with caution. Therefore, functional connectivity in the ID population remains unexplored.

Second, and due to the limited data, the effects of some variables such as type of ID, type of task used, % of women, or mean age of the participants have not been analyzed due to the low heterogeneity found. Our heterogeneity indexes were very low, and this can be due to the limited sample. Finally, the studies included used different tasks. In general, it is essential to highlight the low heterogeneity in all the included articles of this study. Therefore, we are not able, with these preliminary results, to generalize to the ID population. To address the functional brain abnormalities in this population, more task-fMRI studies are needed, and a task-specific meta-analysis should address this topic more directly.

Conclusions

Previous studies in ID and neuroimaging suggest an altered between-network connectivity that could be common of different disorders involving ID, including DS and WS. Considering the perspective of ID based on the individualized supports and the adaptative behavior, this study aimed to examine if differences in the connectivity pattern exist in this population compared with healthy controls. However, this population faces different problems that complicate the register. This article aims to integrate the last years of knowledge in this field in a meta-analytic review.

Only 10 articles met the inclusion criteria, and all the disabilities were associated with genetical syndromes. Regarding the qualitative part of the study, common patterns of dysconnectivity and connectivity versus controls are found in specific populations such as 22q11 DS and Prader–Willi syndrome. Also, in the type of task, and considering that all the different syndromes are mixed, common patterns were found in executive functions and visual perception.

In conclusion, this meta-analysis revealed changes in brain activation in the right middle temporal gyrus regarding the quantitative part of the study. This area is implicated in cognitive areas, including semantic memory processing and language. These results can be highly influenced by the type of task that has been made in all the studies included in the meta-analysis, including 70% of tasks that would include the domains mentioned before. However, other issues can influence.

For instance, the high percentage of the 22q11 DS population that we have included in the studies (a 40%) could mediate the functional disconnection seen in the right middle temporal gyrus (Scarpazza et al., 2019). This region is linked with schizophrenia and psychosis, and 22q11 DS is at high risk of presenting this pathology (Nestor et al., 1998). Despite the efforts of the present study to provide state-of-the-art on this issue, it was not possible to demonstrate the differences in brain irregularities of the ID population and controls. Indeed, the current results are based on the available literature and do not allow the assumption of differences between the two populations. Clarifying if a pattern underlying the different types of ID exists compared with healthy controls would lead to a better understanding of the ID population.

Data Availability

The data sets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Footnotes

Authors' Contributions

All authors have made substantial contributions to the study, have approved the submitted version, and have agreed both to be personally accountable for the author's own contribution and to ensure that questions related to the accuracy or integrity of any part of the work are appropriately investigated, resolved, and the resolution documented in the literature. Concretely, C.C.-M. and J.G.-O. conceived the study as well as acquired, analyzed, and interpreted the data. M.C.-C. and M.P.-C. substantively revised the work and also interpreted the results.

Authors Disclosure Statement

No competing financial interests exist.

Funding Information

Funders: Ministerio De Ciencia, Innovación Y Universidades, Agencia Estatal De Investigación, Grant No. PGC2018-095829-B-I00.

Supplementary Material

Supplementary Appendix SA1

Supplementary Appendix SA2

Supplementary Appendix SA3

Supplementary Appendix SA4

Supplementary Appendix SA5

References

Albajes-Eizagirre

, Solanes

, Vieta

, et al. 2009. Voxel-based meta-analysis via permutation of subject images (PSI): theory and implementation for SDM. Neuroimage, 186:174–184.

* Azuma

, Daly

, Campbell

, et al. 2009. Visuospatial working memory in children and adolescents with 22q11. 2 deletion syndrome; an fMRI study. J Neurodev Disord, 1:46–60.

* Azuma

, Deeley

, Campbell

, et al. 2015. An fMRI study of facial emotion processing in children and adolescents with 22q11. 2 deletion syndrome. J Neurodev Disord, 7:1.

* Bernardino

, Rebola

, Farivar

, et al. 2014. Functional reorganization of the visual dorsal stream as probed by 3-D visual coherence in Williams syndrome. J Cogn Neurosci, 26:2624–2636.

* Binelli

, Muñiz

, Subira

, et al. 2016. Facial emotion processing in patients with social anxiety disorder and Williams–Beuren syndrome: an fMRI study. J Psychiatry Neurosci, 41:182–191.

Binelli

, Subirà

, Batalla

, et al. 2014. Common and distinct neural correlates of facial emotion processing in social anxiety disorder and Williams syndrome: a systematic review and voxel-based meta-analysis of functional resonance imaging studies. Neuropsychologia, 64:205–217.

Button

, Ioannidis

, Mokrysz

, et al. 2013. Power failure: why small sample size undermines the reliability of neuroscience. Nat Rev Neurosci, 14:365–376.

Cabeza

, Nyberg

. 2000. Imaging cognition II: an empirical review of 275 PET and fMRI studies. J Cogn Neurosci, 12:1–47.

Carbó-Carreté

, Cañete-Massé

, Peró-Cebollero

, et al. 2020. Using fMRI to assess brain activity in people with Down syndrome: a systematic review. Front Hum Neurosci, 14:147.

10.

Chang

, Womer

, Bai

, et al. 2016. Voxel-based morphometry in individuals at genetic high risk for schizophrenia and patients with schizophrenia during their first episode of psychosis. PLoS One, 11:e0163749.

11.

Chao

, Haxby

, Martin

. 1999. Attribute-based neural substrates in temporal cortex for perceiving and knowing about objects. Nat Neurosci, 2:913–919.

12.

Chen

, Du

, Zhao

, et al. 2015. Voxel-wise meta-analyses of brain blood flow and local synchrony abnormalities in medication-free patients with major de- pressive disorder. J Psychiatry Neurosci, 40:401–411.

13.

Damoiseaux

, Beckmann

, Sanz Arigita

, et al. 2008. Reduced resting-state brain activity in the “default network” in normal aging. Cereb Cortex, 18:1856–1864.

14.

, Liu

, Chen

, et al. 2014. Brain grey matter volume alterations in late-life depression. J Psychiatry Neurosci, 39:397–406.

15.

Eklund

, Nichols

, Knutsson

. 2016. Cluster failure: why fMRI inferences for spatial extent have inflated false-positive rates. Proc Natl Acad Sci U S A, 113:7900–7905.

16.

Fassbender

, Mukherjee

, Schweitzer

. 2017. Reprint of: minimizing noise in pediatric task-based functional MRI; Adolescents with developmental disabilities and typical development. Neuroimage, 154:230–239.

17.

Friston

, Frith

, Liddle

, et al. 1993. Functional connectivity: the principal-component analysis of large (PET) data sets. J Cereb Blood Flow Metab, 13:5–14.

18.

Fusar-Poli

, Borgwardt

, Crescini

, et al. 2011. Neuroanatomy of vulnerability to psychosis: a voxel-based meta-analysis. Neurosci Biobehav Rev, 35:1175–1185.

19.

Greicius

. 2008. Resting-state functional connectivity in neuropsychiatric disorders. Curr Opin Neurol, 21:424–430.

20.

Greicius

, Flores

, Menon

, et al. 2007. Resting-state functional connectivity in major depression: abnormally increased contributions from subgenual cingulate cortex and thalamus. Biol Psychiatry, 62:429–437.

21.

* Hall

, Walter

, Sherman

, et al. 2009. The neural basis of auditory temporal discrimination in girls with fragile X syndrome. J Neurodev Disord, 1:91–99.

22.

Iwabuchi

, Krishnadas

, Li

, et al. 2015. Localized connectivity in depression: a meta-analysis of resting state functional imaging studies. Neurosci Biobehav Rev, 51:77–86.

23.

Kaas

, Weigelt

, Roebroeck

, et al. 2010. Imagery of a moving object: the role of occipital cortex and human MT/V5+. Neuroimage, 49:794–804.

24.

Kuperberg

, McGuire

, David

. 1998. Reduced sensitivity to linguistic context in schizophrenic thought disorder: evidence from on-line monitoring for words in linguistically anomalous sentences. J Abnorm Psychol, 107:423–434.

25.

Laufer

, Negishi

, Lacadi

, et al. 2011. Dissociation between the activity of the right middle frontal gyrus and the middle temporal gyrus in processing semantic priming. PLoS One, 6:e22368.

26.

, Chn

, Gong

, et al. 2019. Voxel-wise meta-analysis of task-related brain activation abnormalities in major depressive disorder with suicide behavior. Brain Imaging Behav, 14:1298–1308.

27.

Lightbody

, Reiss

. 2009. Gene, brain, and behavior relationships in fragile X syndrome: evidence from neuroimaging studies. Dev Disabil Res Rev, 15:343–352.

28.

Luckasson

, Borthwick-Duffy

, Buntinx

, et al. 2002. Mental Retardation: Definition, Classification, and Systems of Supports. Washington, DC: American Association on Mental. Retardation.

29.

Marshall

, Howrigan

, Merico

, et al. 2017. Contribution of copy number variants to schizophrenia from a genome-wide study of 41,321 subjects. Nat Genet, 49:27–35.

30.

* Montojo

, Ibrahim

, Karlsgodt

, et al. 2014. Disrupted working memory circuitry and psychotic symptoms in 22q11. 2 deletion syndrome. Neuroimage Clin, 4:392–402.

31.

* Montojo

, Jalbrzikowski

, Congdon

, et al. 2013. Neural substrates of inhibitory control deficits in 22q11. 2 deletion syndrome. Cereb Cortex, 25:1069–1079.

32.

Nestor

, Akdag

, O'Donnell

, et al. 1998. Word recall in schizophrenia: a connectionist model. Am J Psychiatry, 155:1685–1690.

33.

Pujol

, Macià

, Blanco-Hinojo

, et al. 2014. Does motion-related brain functional connectivity reflect both artifacts and genuine neural activity?. Neuroimage, 101:87–95.

34.

Radua

, Mataix-Cols

. 2009. Voxel-wise meta-analysis of grey matter changes in obsessive–compulsive disorder. Br J Psychiatry, 195:393–402.

35.

Radua

, Mataix-Cols

, Phillips

, et al. 2012. A new meta-analytic method for neuroimaging studies that combines reported peak coordinates and statistical parametric maps. Eur Psychiatry, 27:605–611.

36.

Raposo

, Moss

, Stamatakis

, Tyler

. 2006. Repetition suppression and semantic enhancement: an investigation of the neural correlates of priming. Neuropsychologia, 44:2284–2295.

37.

Reiss

, Moss

, Stamatakis

, et al. 2000. Brain imaging in neurogenetic conditions: realizing the potential of behavioral neurogenetics research. Ment Retard Dev Disabil Res Rev, 6:186–197.

38.

* Reynolds Losin

, Rivera

, O'Hare

, et al. 2009. Abnormal fMRI activation pattern during story listening in individuals with Down syndrome. Am J Intellect Dev Disabil, 114:369–380.

39.

Sass

, Sachs

, Krach

, et al. 2009. Taxonomic and thematic categories: neural correlates of categorization in an auditory-to-visual priming task using fMRI. Brain Res, 1270:78–87.

40.

Scarpazza

, Lattanzi

, Antoniades

, et al. Systematic review and multi-modal meta-analysis of magnetic resonance imaging findings in 22q11. 2 deletion syndrome: is more evidence needed? Neurosci Biobehav Rev 2019;107:143–153.

41.

Schalock

, Borthwick-Duffy

, Bradley

, et al. 2010. Intellectual Disability: Definition, Classification, and Systems of Supports. Washington DC: American Association on Intellectual and Developmental. Disabilities.

42.

Sheng

, Ma

, Zhong

, et al. 2015. Motor and extra-motor gray matter atrophy in amyotrophic lateral sclerosis: quantitative meta-analyses of voxel-based morphometry studies. Neurobiol Aging, 36:3288–3299.

43.

Shepherd

, Matheson

, Laurens

, et al. 2012. Systematic meta-analysis of insula volume in schizophrenia. Biol Psychiatry, 72:775–784.

44.

Sun

, Ching

, Lin

, et al. 2020, Large-scale mapping of cortical alterations in 22q11. 2 deletion syndrome: convergence with idiopathic psychosis and effects of deletion size. Mol Psychiatry 25:1822–1834.

45.

Tassé

, Luckansson

, Schalock

. 2016. The relation between intellectual functioning and adaptative behaviour in the diagnosis of intellectual disability. Intellect Dev Disabil, 54:381–390.

46.

Tek

, Gold

, Blaxton

, et al. 2002. Visual perceptual and working memory impairments in schizophrenia. Arch Gen Psychiatry, 59:146–153.

47.

Thurm

, Farmer

, Salzman

, et al. 2019. State of the field: differentiating intellectual disability from autism spectrum disorder. Front Psychiatry, 10:526.

48.

Tranel

, Damasio

. 1997. A neural basis for the retrieval of conceptual knowledge. Neuropsychologia, 35:1319–1327.

49.

Vega

, Hohman

, Pryweller

, et al. 2015. Resting-state functional connectivity in individuals with Down syndrome and Williams syndrome compared with typically developing controls. Brain Connect, 5:461–475.

50.

Walter

, Mazaika

, Reiss

. 2009. Insights into brain development from neurogenetic syndromes: evidence from fragile X syndrome, Williams syndrome, Turner syndrome and velocardiofacial syndrome. Neuroscience, 164:257–271.

51.

* Wan

, Chiang

, Chen

SCJ

, et al. 2017. The effectiveness of the computerized visual perceptual training program on individuals with Down syndrome: an fMRI study. Res Dev Disabil, 66:1–5.

52.

* Woodcock

, Humphreys

, Oliver

, et al. 2010. Neural correlates of task switching in paternal 15q11–q13 deletion Prader–Willi syndrome. Brain Res, 1363:128–142.

53.

Xia

, Wang

, He

. 2013. BrainNet Viewer: a network visualization tool for human brain connectomics. PLoS One, 8:e68910.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.25 MB