The role of the cerebellum in internet gaming disorder—A systematic review

Recent studies increasingly highlight involvement of the cerebellum in drug craving and addiction. However, its exact role, that is, whether the cerebellum is a critical component of a brain network underlying addictive behaviour, or whether it rather is a facilitator or mediator, is still unclear. Findings concerning the newly recognized internet gaming disorder (IGD) suggest that changes in cerebellar connectivity and functioning are associated with behavioural/non‐substance addiction. Here, we systematically review the literature on IGD and cerebellar involvement following the PRISMA guidelines. A total of 13 neuroimaging studies met the inclusion criteria. Studies utilized a broad range of diagnostic instruments and resulting cut‐off criteria, rendering it difficult to compare findings. Results on altered cerebro‐cerebellar connectivity in patients with IGD are mixed; most studies report altered or increased functional connectivity. Moreover, decreased cerebellar grey matter volume is reported. Studies have further indicated that differential activation patterns in the cerebellum may enable discrimination between healthy subjects and subjects with IGD, even allowing for prediction of treatment outcomes. Given the strong connectivity between the cerebellum and cerebral regions, the cerebellum may act as an intermediary between regions involved in craving and addiction and consequently affect symptoms of IGD. Results suggest differential involvement of the cerebellar lobes, emphasizing a need for high‐resolution parcellation of the cerebellum in future studies. However, the studies included in the present review have small sample sizes and include mostly male participants. Thus, results may have limited generalizability yet highlight a crucial role of the cerebellum in IGD that needs further investigation.

numbers of reports of co-existing motor and nonmotor cognitive impairments in patients with cerebellar lesions (e.g., 2,3 ).In recent years, many studies have corroborated the cerebellum's role in a variety of cognitive and affective functions such as working memory, [4][5][6] language and verbal fluency, 7,8 error and feedback processing, [9][10][11] performance monitoring in general, 12 emotion regulation, 13 visual attention, 14 executive control 15 and in circadian rhythmicity. 16Interestingly, findings also suggest that the cerebellum is involved in drug craving 17 and addiction. 18Neuroimaging studies have discovered cerebellar activations when participants are craving following the presentation of drug-related cues. 17Along these lines, it has been suggested that the cerebellum could be a mediator between systems involved in addiction and craving such as motor, reward, motivation and cognitive control systems. 18e form of addiction that has recently been linked to altered cerebellum function is internet gaming disorder (IGD), a newly recognized non-substance addiction.The disorder, outlined in the appendix of the DSM-5 as a condition needing further research, is marked by problematic use of computer games (whether online or offline). 19Since 2018, IGD is recognized by the ICD-11 as gaming disorder that includes online and offline gaming (https://icd.who.int/en).Currently, there is only limited evidence of effective treatment, mainly due to a limited number of studies, methodological differences, a lack of consistent definitions of the disorder and a shortage of established instruments to assess IGD. 20So far, nine criteria for a diagnosis of IGD are listed in the DSM-5, of which at least five criteria must be met in the past 12 months.Briefly summarized, these criteria are 'preoccupation', 'withdrawal', 'tolerance', 'loss of control', 'giving up other activities', 'continuation', 'deception', 'escape' and 'negative consequences'. 21wever, the criteria and set threshold may still need an adjustment. 19sides the IGD scale of the DSM-5, several instruments have emerged to assess internet gaming addiction such as the Internet Addiction Test (IAT) by Young 22 and Chen's Internet Addiction Scale (CIAS). 23uroimaging studies have attempted to identify brain alterations unique to IGD, which set this disorder apart from other addictions such as tobacco or alcohol use disorder.Several studies report increased cerebello-cerebral functional connectivity in individuals with IGD [24][25][26] as well as a hint towards decreased microstructural and macrostructural cerebellar properties. 27,28However, the results are mixed, and a systematic integration is as yet lacking.Thus, the aim of the present review was to integrate the current neuroimaging literature on IGD and the role of the cerebellum.To this end, studies including data on the cerebellum with participants diagnosed with IGD and healthy controls (HC) were reviewed.Of note, some studies also included a third group of subjects diagnosed with tobacco or alcohol use disorder.

| METHOD
This systematic review was performed according to the updated Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 29Consequently, the systematic literature search was conducted by two independent researchers.Relevant studies were identified using the database PubMed exclusively.The following search terms were used: 'cerebellum' AND 'gaming'.Research articles published in PubMed until December 2022 were included if they met the following inclusion criteria: (i) published original research article, (ii) English language, (iii) peer-reviewed, (iv) participants included were individuals diagnosed with IGD and HC, (v) methods used for data collection where either magnetic resonance imaging (MRI) or functional MRI (fMRI), (vi) for fMRI studies, activation was measured either with resting state or task-based paradigms and (vii) participants had no history of substance-dependence disorders and did not take any medication affecting the central nervous system.Publications that matched all of the above-listed criteria were included in an Excel file.The main findings of each publication were recorded, especially those that contained information on the cerebellum.In case of any discrepancies regarding inclusion/exclusion, a third independent reviewer was consulted to discuss the final decision.If, in case of cerebellar activations, peak coordinates were provided but cerebellar subregions were not further specified, we identified the subregion using the label4MRI package in R and the AAL atlas taxonomy.If no coordinates were provided, cerebellar regions were approximated by visual inspection of the results figures in the publications.

| Study selection
The initial literature search identified 53 publications of which 10 matched all the inclusion criteria listed above.Three additional studies were retrieved from a list of references in another study that was among the results of the initial literature search.However, there were a number of publications that were excluded, resulting in a total of 13 publications included in the systematic review (Figure 1).Most of the excluded publications were either focused on other disorders that were irrelevant to this review or the research was conducted only in healthy participants playing games.Two studies that had been found during the systematic literature search did not report any activations in the cerebellum.They were among the search results due to their list of references that contained one of the search terms.As the review is focused on cerebellar involvement in IGD, these two studies were excluded.In the following sections, we will focus solely on study results concerning the cerebellum, even though most studies reveal alterations in several brain regions.We will cluster studies by methodology, reporting findings of studies that used fMRI first, here distinguishing between resting-state and task-based fMRI.In the following sections, we will review studies that used structural MRI, covering macrostructural findings first, followed by microstructural findings.

| Risk of biases
Because only 5 out of the included 13 studies investigated both male and female subjects, results might not be representative of the general population due to selection bias. 30Possibly, this is because men are more likely to exhibit IGD than women ( g = 0.479) as confirmed by a recent meta-analysis. 31However, gender-specific changes in functional connectivity have been found in brain regions implicated in executive control and reward processing. 32For a more comprehensive understanding of the underlying neurobiological changes in IGD, it is thus important to include females as well.In most of the reviewed studies, relatively young subjects were recruited, which could add to the selection bias as well.Furthermore, most of the included studies tested rather small samples, which might limit the generalizability of the findings.To identify subjects with IGD, different diagnostic questionnaires were used which might result in measurement bias.Publication bias frequently affects systematic reviews because a study is more likely to be published if findings support the research hypothesis while studies finding no effects are neglected. 30This is particularly problematic when the existing body of research is still quite small, as is the case for IGD and cerebellar involvement.
For the sake of brevity, we focus the reporting on the neuroimaging results.Further information regarding sample composition and study design (e.g., questionnaires and cut-off score used to define IGD) and further questionnaires on behavioural measures are listed in Tables 1-3.

| fMRI studies
A total of 10 studies analysed functional changes in the cerebellum in IGD, 7 of which analysed alterations in resting-state fMRI and 3 investigated the role of the cerebellum in cue reactivity with task-based fMRI (Tables 1 and 2).

| Resting-state fMRI
Alterations in regional homogeneity of resting-state brain activity in subjects with IGD were explored by Dong et al. 25 This study included only male participants (15 IGD and 14 HC).The left inferior parietal lobule was selected as the seed region; functional connectivity was analysed between the seed and other brain regions; and regional homogeneity measures were applied to detect abnormal functional integrations.Results showed that IGD was associated with increased regional homogeneity in the left cerebellar lobule VIIB, brainstem, inferior parietal lobule and left middle frontal gyrus.However, the correlation between IAT scores and the averaged peak values in all clusters did not reach significance after correcting for multiple testing.Also, functional connectivity did not differ significantly between IGD and HC. 25 Ding et al. scanned a total of 17 adolescent IGD subjects aged 14-17 years and 24 gender-and age-matched HC. 24 29 superior temporal gyrus in IGD individuals.Region of interest-based correlation analyses revealed a significant negative correlation between IGD severity (IAT score) and fALFF in the cerebellum.Moreover, there were decreased amplitude values in the higher frequency band (e.g., slow-2) and increased amplitude values in the lower frequency band (slow-6) in the cerebellum in the IGD group compared with HC, indicating a dynamic alteration of fALFF in the cerebellum. 33n et al. compared functional connectivity in male IGD participants (n = 15) and males diagnosed with alcohol dependence (n = 16), without including an HC group.In both groups, subjects showed positive functional connectivity between the dorsolateral prefrontal cortex (selected seed region) and cingulate gyrus and bilateral cerebellum (Crus I), but negative functional connectivity between the dorsolateral prefrontal cortex and the orbitofrontal cortex.Furthermore, while individuals with alcohol dependence demonstrated positive functional connectivity between the dorsolateral prefrontal cortex, temporal lobe and striatal areas, individuals with IGD revealed negative functional connectivity between the dorsolateral prefrontal cortex, temporal lobe and striatal areas. 34Of note, the authors defined a positive connection as positive correlations between T A B L E 2 Summary of task-based fMRI studies included and respective results for the cerebellum in IGD. in recreational game use. 37A total of 148 IGD subjects and 226 participants who play games recreationally were included.MVPA was applied to classify individuals with IGD from recreational game users based on regional brain features, respectively, regional homogeneity and functional connectivity.MVPA could not distinguish individuals with IGD from recreational game users based on regional homogeneity and functional connectivity features when using DSM-5 scores ≥5.

Study
However, the classification was successful when stricter DSM-5 scores of ≥6 were used.Here, bilateral inferior cerebellum, right lobule VIIB and right Crus I, orbital frontal cortex, cuneus, inferior temporal gyrus, middle frontal gyrus and parahippocampal gyrus displayed strong discriminative power.Furthermore, functional connectivity between the left inferior cerebellum and the right middle temporal gyrus as well as the left inferior temporal gyrus and right cerebellar lobule VIIB were among the most important brain regions that contribute to classifying IGD.Unfortunately, the authors did not investigate potential sex differences even though gender-specific changes in functional connectivity have been reported in brain regions implicated in executive control and reward processing. 32eng et al. investigated similarities and differences between male and female subjects with IGD (n = 92) or tobacco use disorder (n = 96) and controls (n = 107). 38The authors performed independent component analysis and defined several resting state networks of brain regions that are anatomically separated but functionally linked and show high continuous functional connectivity during resting state.Nine networks were derived: subcortical, auditory, sensorimotor, cerebellar, visual, salience, default mode, attention and executive network.Intra-network analysis revealed significant differences in the subcortical network, with higher connectivity in the orbital superior frontal gyrus, caudate and thalamus in tobacco use disorder patients compared with HC, and greater connectivity in the orbital superior frontal gyrus, caudate, and putamen compared with IGD patients.There were no differences between IGD and HC in the subcortical network.Furthermore, in both addiction disorders relative to HC, higher functional connectivity was found in the left (likely anterior) cerebellum.Correlation analysis between the cerebellar network and craving scores yielded a negative correlation between craving and the cerebellum, indicating that high functional connectivity of the cerebellum was associated with lower craving.
Furthermore, an inter-network analysis was performed, revealing high connectivity between the salience network and the cerebellar network in all groups.There was no significant difference between IGD and HC.
Taken together, findings regarding altered cerebello-cerebral connectivity in IGD patients are mixed.Prior studies used different diagnostic criteria to define IGD and rather small sample sizes.However, most studies did report altered, and often increased, functional connectivity of the cerebellum with neocortical regions such as the dorsolateral prefrontal cortex, the middle temporal gyrus and the posterior T A B L E 3 Summary of structural studies included and respective results for the cerebellum in IGD.cingulate cortex in IGD subjects.Moreover, the bilateral inferior cerebellum showed strong discriminative power for identifying IGD relative to recreational game use when applying MVPA.

| Task-based fMRI studies
A total of three studies examined changes in the cerebellum by applying task-based fMRI (Table 2).Sun et al. performed fMRI in 10 male addicts of the game World of Warcraft (WoW) and 10 male HC using WoW-associated cue pictures and neutral pictures, the latter of which were scrambled versions of the original pictures. 39Subjects reported their craving scores after viewing WoW-related cue pictures.They

| Structural MRI studies
A total of four studies analysed structural changes in the cerebellum, with three studies demonstrating changes on the macrostructural level and one study investigating changes on the microstructural level via MRI (Table 3).3).
To summarize, studies on macrostructural and microstructural alterations in IGD highlight decreased grey matter volume in anterior and posterior cerebellar regions.As only few studies have focused on structural alterations in IGD, more research that investigates potential region-specific effects in the cerebellum is needed.

| DISCUSSION
Currently, findings on altered cerebro-cerebellar connectivity associated with IGD are mixed.However, several studies have reported altered or increased functional connectivity of the cerebellum in IGD, especially with frontal areas such as the dorsolateral prefrontal cortex, the posterior cingulate cortex and the middle temporal gyrus (see Table 1).Studies also indicate that IGD might be associated with a decrease in cerebellar microstructure as reflected in lower mean kurtosis metrics and axial kurtosis obtained by diffusion kurtosis imaging, 27 and with grey matter volume in the bilateral posterior (possibly lobule VII, VIII and IX) cerebellum. 26,28,42However, very few studies have investigated structural cerebellar changes in IGD (see Table 3).Interestingly, some convergent findings derived from MVPA indicate that cerebellar activation patterns either during resting-state or during cue-related task performance could help discriminate between HC and subjects with IGD and aid in predicting treatment outcomes. 37,41en integrating the studies included in this review, the broad range of diagnostic instruments and resulting cut-off criteria is obvious and certainly limits the comparability of their findings.As evidenced by the study by Dong et al., 37 the application of different criteria or thresholds to diagnose IGD (in this case DSM-5 scores ≥5 or ≥6) can impact whether cerebellar alterations are associated with discrimination of IGD and recreational players.This illustrates that standardized, unified diagnostic instruments and scores are urgently needed.
Given the important role of the cerebellum in emotional and cognitive processing, and particularly in cognitive processes such as reward or feedback processing, 10 it has been hypothesized that the cerebellum acts as an intermediary between regions involved in craving and addiction, that is, regions involved in reward processing, motivation and cognitive control. 13,18Consequently, altered connectivity with any such region or altered activation in response to cravingrelated cues could contribute to symptoms associated with IGD.
Findings of previous studies appear to hint at differential involvement of several cerebellar subregions in IGD.For example, microstructural changes in IGD were only present in the right anterior lobe. 27Increased activation in a response inhibition task in IGD compared with HC was found in the left posterior cerebellum. 40Interestingly, posterior cerebellar regions, albeit in the right hemisphere, were key regions for discriminating between HC and IGD based on activity patterns associated with processing of addiction-related cues. 41These findings are particularly interesting in light of a functional cerebellar topography in which lateral and posterior regions are recruited for complex motor and cognitive/nonmotor functions. 43 Functional gradients (e.g., from more concrete to more abstract) appear to be preserved even within these representations.
Another interesting finding is that alterations of cerebellar functional connectivity do not seem to involve neocortical or striatal motor regions.One might expect computer gaming addicts to develop enhanced motor coordination, which does not seem to be reflected in the findings on functional connectivity.Alternatively, gaming might require the development of timing mechanisms, and given the hypothesized involvement of the cerebellum in timing, 44 some of the cerebro-cerebellar changes in functional connectivity could reflect differences in timing proficiency.
A critical aspect to consider is that different genres of games may need to be distinguished because of diverging demands.Along these lines, games that involve complex sensorimotor integration (e.g., firstperson shooter games or games involving multiplayer online battle arenas) are likely to induce neural plasticity processes in sensorimotor networks, which might dilute the addiction-related changes of IGD.In contrast, social simulation games or certain sports games likely share some of the addictive features but differ in sensorimotor demands.
Unfortunately, the studies included in this review did not differentiate between or report the type of games played by the participants.To truly disentangle whether the reported changes in fronto-cerebellar connections found in IGD reflect gaming-induced neural plasticity or whether they relate to addictive behaviour, future studies will need to carefully track and take into account the types of games played.Furthermore, they should include the number of hours played and, if possible, illness duration to disentangle these effects.With regard to time spent gaming, it has to be noted that some of the reviewed studies 34,39,41 used weekly gaming time as an additional criterion for IGD but did not quantify gaming in controls.This should be addressed in future work.
Previous research emphasizes cerebellar involvement in visual attention 14 and executive control, 15 which are both needed when gaming.Consequently, intensive gaming as seen in IGD might also alter networks related to visual attention and executive control that involve both cerebral and cerebellar regions.Importantly, cerebrocerebellar interaction is mediated by a closed-loop architecture of the cerebellum as defined by Kelly and Strick. 45The major neocortical input sources to a given circuit are also its output targets.Along these lines, alterations of lateral and posterior cerebellar regions in IGD likely affect cerebro-cerebellar cognitive loops that are linked to executive functions and cognitive demand. 46terestingly, studies on substance use or addiction also point to a critical role of the posterior cerebellum.For example, long-term use of heroin, cocaine, alcohol or nicotine is associated with reduced grey matter in the posterior cerebellum. 18Moreover, when studies on substance use or addiction include coordinates on cerebellar responses to craving or in response to acute drug use, increased activation is found in the posterior cerebellum. 18But it is yet to be determined whether chronic drug use or addiction results in reduced cerebellar volume, or whether a small cerebellum increases the risk to develop addiction. 18e role of the cerebellum was further corroborated in a metaanalysis that highlights the cerebellum among overlapping regions of reactivity to reward across drug, gambling, food and sexual cues. 47udies in mice have shown that the cerebellar deep nuclei not only project directly to the ventral tegmental area, one of the brain regions crucial for processing and encoding reward, but optogenetic stimulation of the cerebellar afferents to the ventral tegmental area are rewarding for the mice and support conditioned place preference as well as normal social behaviour. 48th regard to reward processing, recent findings also point to coding of reward prediction errors in cerebellar activity (for a review, see Hull   10 which requires behaviour to be adapted to changing task rules.These regions and processes may also play a crucial role in the development and maintenance of addictive behaviour. 18Unfortunately, most of the reviewed studies on IGD did not specify cerebellar subregions in their data.More research that focuses particularly on cerebellar subdivision and that incorporates high-resolution imaging and parcellation of the cerebellum is clearly needed to further elucidate specific associations between cerebellar subregions and IGD.

| Recommendation for future studies
Future studies need to address whether changes in fronto-cerebellar connections found in IGD result from gaming-induced neural plasticity or whether they relate to addictive behaviour.To this end, detailed analysis of time spent playing and type of games played will be critical.
Moreover, in taking into account the high comorbidity between IGD and ADHD, 50 it is important to note that ADHD and autism spectrum disorder have comorbidity rates of 50% to 70%. 51We therefore recommend that future studies carefully screen their participants for both neurodevelopmental disorders, as both are associated with a reduced cerebellar volume. 52Consequently, the question arises whether the differences in cerebellar functions seen in IGD might simply be reflective of changes in executive control and response inhibition that have resulted from such neurodevelopmental disorders.
Although most studies did screen participants for other psychiatric diagnoses, such as depression or substance use disorder, and applied respective exclusion criteria, only one study included IGD individuals with and without ADHD and reported differential effects. 26To clearly distinguish cerebellar changes in IGD from changes influenced by comorbid ADHD or autism spectrum disorder, future studies need to be more nuanced with regard to their experimental groups.

| CONCLUSION OR SUMMARY
There is a growing body of evidence supporting cerebellar involvement in IGD.Most of the reviewed studies reported increased functional cerebro-cerebellar connectivity in IGD compared with HC.Furthermore, some findings indicate that IGD might be associated with a decrease in cerebellar anterior lobe microstructure and posterior lobe grey matter volume.Interestingly, several studies show that differential cerebellar activation patterns in response to addictionrelated cues allow for discrimination between healthy participants and subjects with IGD and may even allow a prediction of treatment outcomes.However, results also indicate that functional and structural alterations associated with IGD typically do not occur cerebellumwide but are rather limited to specific subregions.Moreover, it must be stressed that studies included in the present review used a range of diagnostic criteria and diagnostic tools to define IGD and were often conducted in samples of mostly male participants.Consequently, the results may have limited generalizability.Nevertheless, it can be concluded that the existing empirical findings highlight a key role of the cerebellum, particularly posterior/posterolateral and anterior regions, in IGD that needs further investigation.
All participants were nonsmokers and right-handed.Compared with HC, IGD subjects showed increased functional connectivity between the bilateral posterior lobe (Crus II) of the cerebellum and the middle temporal gyrus.Connectivity between the posterior cingulate cortex and lobule VI of the right cerebellar hemisphere and the left superior parietal lobule was negatively correlated with CIAS scores, indicating that reduced connectivity between posterior cingulate cortex and lobule VI was associated with higher levels of addiction.Lin et al. focused on exploring spontaneous brain activity in IGD subjects.To this end, the resting-state fractional amplitude of lowfrequency fluctuation (fALFF) was recorded in male HC (n = 26) and male participants diagnosed with IGD (n = 26).When comparing HC with IGD individuals, decreased fALFF was found in the cerebellum's posterior lobe in IGD while increased fALFF values were found in F I G U R E 1 PRISMA flow diagram depicting the process of identifying, screening and inclusion of the literature.Adapted from Page et al.
were then asked to remember game experiences and count their scores.Comparison between HC and IGD (presumably for the contrast WoW-associated pictures minus neutral images; unfortunately, the methods description is somewhat unclear here) showed increased activation in several brain regions in IGD subjects, among those left Crus II and left cerebellar lobules VIIB and VIII.Liu et al. investigated brain activations associated with responseinhibition under gaming cue distraction in 11 IGD men and 11 male controls.During the fMRI scan, subjects performed a go/no-go task with or without distraction.In the undistracted condition, participants were shown white polygons (except pentagons) on black background.They had to press the button as quickly as possible when seeing a polygon, but not a pentagon.In the game-distraction session, different gaming pictures were shown 0.5 s prior to the presentation of the polygons.Analyses revealed increased activation in the left posterior cerebellum (Crus I) and several frontal regions for response inhibition in the IGD group.Interestingly, when analysing differences in brain activation for response inhibition with/without gamingdistraction between groups, HC subjects showed greater activation under gaming distraction than IGD subjects in the left cerebellar posterior lobe, right superior parietal lobe and right dorsolateral prefrontal cortex.40Wang et al. explored neural mechanisms underlying IGD using a cue reactivity task in males.41All subjects (40 IGD and 19 HC) had to watch three different video clips (gaming video clip, control video clip and fixation) and rate their craving afterwards.Twenty-three of the included IGD subjects participated in a craving behavioural intervention.The authors performed MVPA to detect a brain signature for cue reactivity to classify IGD and found that the right posterior lobule IX of the cerebellum, the precuneus and the bilateral middle frontal gyrus were the most informative regions to discriminate IGD from HC subjects.For predicting intervention outcome via MVPA, the right Crus I and the right lobule IX of the cerebellum, the right middle frontal gyrus, superior frontal gyrus, supramarginal gyrus and left postcentral gyrus were the most informative brain regions.Furthermore, negative correlations were found in the intervention group between weekly gaming time change and beta values in the anterior lobe of the cerebellum, right middle frontal gyrus, middle/superior frontal gyrus, superior frontal gyrus, supramarginal gyrus and left postcentral gyrus.To summarize, task-based studies point towards increased activation in different (anterior and posterior) cerebellar regions.Additionally, differential activation patterns in the cerebellum enable discrimination between HC and IGD and allow prediction of treatment outcome.But, given the small study number, more research on cue reactivity and cerebellum activation in IGD is needed.

3. 4 . 1 |
figure in the article).Yoon et al. compared cortical thickness and volume in male individuals with IGD (n = 19) with those suffering from alcohol use disorder (n = 20) and HC (n = 25) using voxel-based morphometry.42The authors report that significantly smaller volume in bilateral cerebellar lobule VIII was found in alcohol use disorder compared with HC, but not in IGD.Furthermore, left cerebellar volume was negatively correlated with illness duration and bilateral cerebellar volume correlated negatively with total errors in alcohol use disorder participants, but again not in IGD subjects.The study by Lee et al. described in Section 3.3.1 analysed grey matter volume in regions that demonstrated differences in functional connectivity and found no differences between groups in bilateral Crus II.26 More recent evidence points to multiple motor and nonmotor cerebellar representations: a first sensorimotor representation located in anterior regions (lobules I-VI), two nonmotor representations in reverse orientation that converge around the Crus I/II border, a second sensorimotor representation in lobule VIII and a third nonmotor representation in the inferior cerebellum (lobules IX-X) (for a review, see Stoodley et al. 43 ). 49