Mapping hippocampal-cerebellar functional connectivity across the adult lifespan

cerebellum, hippocampus, ageing, memory, functional connectivity

The ability to represent and navigate spatial environments is thought to be supported by an extended posteromedial navigation system in the brain that includes the hippocampus, as well as entorhinal, retrosplenial and parahippocampal cortices (Hodgetts et al., 2017; Murray et al., 2018; Ritchey et al., 2015, see also Ekstrom and Ranganath, 2017). Critically, this brain system is also thought to undergo pronounced structural and functional alterations across the adult lifespan (Lester et al., 2017), acting as a ‘hotspot’ for age -associated neurodegenerative processes, such as tau and amyloid -beta accumulation (Berron et al., 2021; Lace et al., 2009). In this context, it is important to better understand the organisation of the brain’s navigation system, parti cularly by considering the influence of other cortical and subcortical areas. The cerebellum, for instance, is classically associated with motor processing (Glickstein, 2007; Glickstein, 1993; Jimsheleishvili and Dididze, 2021) but has been shown to be strongly connected with higher-order cognitive areas, such as prefrontal cortex ( Kelly and Strick, 2003; Middleton and Strick, 2001; O’Reilly et al., 2009). In line with this, evidence from nonhuman species suggests that the cerebellum has functional and structural interactions with the hippocampus (Rochefort et al., 2013; Rondi-Reig et al., 2022). For example, electrophysiological studies in transgenic mice have shown that disrupting cerebellar cells alters hippocampal place cell firing and impairs navigational performance (Lefort et al., 2015; Rochefort et al., 2011). Similarly, optogenetic excitation or inhibition of midline and lateral cerebellar neurons has been shown to reduce the duration of left hippocampal seizure activity (Krook-Magnuson et al., 2014), and combined optogenetic stimulation -fMRI revealed that cerebellar stimulation (between lobule V and VI) increased dorsal hippocampal blood -oxygen level dependent .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 2 (BOLD) signal (Choe et al., 2018). Taken together, these studies establish an argument that the cerebellum modulates the physiology of the hippocampus, as well as influencing hippocampal - dependent behaviour (e.g., spatial learning and memory). While early electrophysiological evidence in primates and cats suggested a direct structural monosynaptic connection between these structures (as evidenced by short observed latencies between cerebellar fastigial nucleus stimulation and evoked hippocampal responses; Heath and Harper, 1974; Newman and Reza, 1979), recent anatomical tracing work in rodents indicates connectivity through multisynaptic pathways (Watson et al., 2019). Specifically, left hippocampus has been shown to receive input from distribute d areas of cerebellar cortex, including bilateral and vermal lobule VI, lobule VIIA (Crus I) and paraflocculus via at least two relay stations (Watson et al., 2019). This study also observed phase-locked theta coherence between cerebellar Purkinje cells in lobule VI, VIIA (Crus I) and the dorsal hippocampus during exploratory behaviour. Notably, evidence in macaque monkeys suggests that lobule VI and VIIA sends its output to the cerebellar fastigial nuclei (Coffman et al., 2011) – the same area found by the above electrophysiological studies to evoke responses in the hippocampus. This suggests that topographically distinct areas of the cerebellum send their input to the hippocampus via distinct pathways. Despite strong evidence of connectivity in nonhuman species, this is lacking in the human brain. Such evidence has the potential to extend and inform current neurobiological accounts of human memory and navigation, incorporating the cerebellum as a key str ucture, and may have implications for understanding memory decline in ageing and neurodegeneration. Studies of healthy ageing in humans, for example, indicate that the cerebellum and hippocampus are both vulnerable to age -related structural changes (Cui et al., 2020; Du et al., 2006; Ramanoël et al., 2023) and exhibit comparable grey matter loss (Bernard and Seidler, 2014; Gellersen et al., 2021; Woodruff et al., 2010). Notably, age-related reductions in cerebellar volume have been linked to poorer performance on hippocampal-dependent navigation tasks (Daugherty and Raz, 2017), suggesting that structural changes may disrupt hippocampal-cerebellar communication. Indeed, recent work in humans has found that age-related cerebellar atrophy is mainly localised to lobule HVI and lobule HVIIA (Crus I; Ramanoël et al., 2023): two regions shown to connect with the hippocampus in nonhuman animal studies (Watson et al., 2019) and co -activate with the hippocampus during navigational learning in young human adults (Iglói et al., 2015). Despite this evidence, there remains a significant gap in understanding hippocampal -cerebellar connectivity in humans, and how it may differ across the lifespan. Here, we aimed to examine these questions using resting-state fMRI data from the Cambridge Centre for Ageing and Neuroscience (CamCAN) dataset (Taylor et al., 2017). We first sought to map patterns of bilateral hippocampal connectivity (independent of age) within the cerebellum. We hypothesised that the hippocampus would show a strong functional correlation with several areas of the cerebellum, including lobule VI and HVIIA (Crus I), as predicted by prior nonhuman animal tracing work (e.g., Watson et al., 2019) and electrophysiological evidence showing that stimulating fastigial nucleus (which receives input from these areas) leads to evoked responses in the hippocampus (e.g., Coffman et al., 2011). We also contrasted the connectivity of the left hippocampus against the right, drawing on evidence that hippocampally -dependent tasks might be lateralised in the cerebellum (e.g., Riva and Giorgi, 2000). It is also possible that a distinctive pattern of connectivity exists for each hemisphere – a question that rem ains unresolved as previous anatomical studies have primarily focused on left hippocampal connectivity (e.g., Krook-Magnuson et al., 2014; Watson et al., 2019). The hippocampus is also thought to display functional gradients along its longitudinal (anterio r-posterior) axis, arising, in part, from variation in connectional anatomy (Aggleton, 2012; Poppenk et al., 2013; Strange et al., 2014). For instance, the anterior hippocampus preferentially connects to the amygdala and prefrontal cortex (Aggleton et al., 2010; Aggleton, 1986), whereas posterior hippocampus preferentially connects with the parahippocampal cortex (Aggleton, 2012). As regions within these anterior and posterior hippocampal networks might differentially connect with cerebellar cortex (see e.g ., prefrontal connections with lobule HVIIA; Ramnani, 2006), we also predicted connectivity differences when .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 3 contrasting anterior and posterior hippocampal seed regions. Finally, the large age range within the CamCAN dataset (18-87 years old) enabled us to examine how increasing age influences the degree and pattern of hippocampal -cerebellar functional connectivi ty. We predicted that the strongest age - related decreases in functional connectivity will be observed in lobules HVI and HVIIA, consistent with their vulnerability to age -related atrophy (see e.g., Gellersen et al., 2021; Ramanoël et al., 2023) and suggested involvement in spatial cognition (Iglói et al., 2015). Methodology Participants We used structural and functional MRI data from the Cambridge Centre of Ageing and Neuroscience (CamCAN) study (Taylor et al., 2017). This dataset contained 653 participants (323 males, 330 females, 18-87 years old, mean = 54.3, SD=18.6). About fifty males and females were collected from each decile (deciles: 18-27 years, 28-37 years, 38-47 years, 48-57 years, 58-67 years, 68-77 years, 78-87 years; for more information about participants see Shafto et al., 2014). Participants in the dataset were cognitively healthy, assessed via a score above 25 on the mini -mental state examination (Folstein et al.,1975), and did not have any neurological or psychiatric conditions. Written informed consent for the database was obtained in accordance with the Cambridgeshire Research Ethics Committee, and these secondary data analyses were conducted in accordance with Royal Holloway, University of London Ethics Committee processes. Neuroimaging data acquisition All CamCAN MRI data was acquired using a 3T Siemens TIM Trio Scanner at the Medical Research Council (UK) Cognition and Brain Science Unit using a 32 -channel head coil. The data used in this study forms part of a larger scanning protocol (see https://camcan-archive.mrc- cbu.cam.ac.uk/dataaccess/ or Taylor et al., 2017 for more detail). High -resolution structural images were obtained using a T1-weighted magnetisation-prepared rapid sequence (MPRAGE; TE= 2.99 ms; TR= 2250ms; TI = 900 ms; voxel size = 1x1x1 mm; field -of-view = 256x240x192 mm; flip a ngle = 9°). Resting-state fMRI data was acquired using a T2* -weighted gradient echo planar image (EPI) sequence. Rest (resting in the scanner with eyes closed) consisted of 261 volumes and lasted 8 minutes and 40 seconds (32 slices; TE = 30 ms, TR = 1970ms, voxel size: 3x3x4.44 mm; field-of-view: 192x192 mm). Data preprocessing and denoising Initially, functional and structural MRI data from a random sample of approximately 25% of cases were visually inspected for image quality and motion artefacts by KA (with support from CH). Functional MRI images were pre -processed and denoised using the standard pipeline in the CONN toolbox (RRID:SCR_009550; version 22.a; Nieto -Castanon and Whitfield -Gabrieli, 2022) and SPM12 (RRID: SCR_007037; version 12.7771; Penny et al., 2011) running on Matlab 2022b (The Mathworks Inc, Natick, Massachusetts, USA). Functional MRI data were first realigned and unwarped using SPM12 (Andersson et al., 2001). For this, all scans were co-registered to the first volume using a least-squares approach and a 6 -parameter rigid body transformation (Friston et al., 1995). These were then resampled using b -spline interpolation to correct for motion and magnetic susceptibly interactio ns. Temporal misalignment between slices was corrected using the SPM12 slice -timing correction procedure (Henson et al., 1999; Sladky et al., 2011), which involved sinc temporal interpolation to resample each BOLD timeseries slice to a common mid-acquisition time. Potential outlier scans (based on motion and global signal fluctuation) were identified using conservative (95 th percentile) outlier parameters in ART (Artifact Detection Tools; Whitfield-Gabrieli et al., 2011). Specifically, volumes with framewise displacement above 0.5 mm, or global BOLD signal changes above 3 SDs, were flagged as outliers (Nieto-Castanon, 2022; Power et al., 2014). A reference mean BOLD image was then computed for each subject by averaging all scans, excluding outliers. Followin g this, functional and structural .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 4 images were separately normalised into standard MNI space and segmented into grey matter, white matter and CSF ‘tissue’ types using the SPM12 unified segmentation and normalization algorithm (Ashburner and Friston, 2005; Ashburner, 2007). Functional and structural images were then resampled to 2 mm and 1 mm isotropic voxels, respectively, following a direct normalization procedure (Calhoun et al., 2017; Nieto -Castanon, 2022) with the default IXI -549 tissue probability map template. Finally, functional data was smoothed using a Gaussian kernel of 5 mm full-width half-maximum (FWHM; e.g., Dombrovski et al., 2020). Following preprocessing, the fMRI data were next denoised using the default pipeline in the CONN toolbox (Nieto-Castanon, 2020). This involved regressing out noise components using an anatomical component-based noise correction procedure (aCompCor), which included noise components from white matter (5 noise components), CSF (5 noise components), motion parameters (3 translational and 3 rotational and their first order derivatives; Friston et al.,1996), outliers volumes derived from scrubbing (38 factors) (Power et al., 2014), effect of rest and its first order derivatives (2 factors; default setting which removes residual trends/instabilities only at the beginning of the timeseries). These were followed by a bandpass frequency filtering of the BOLD timeseries (Hallquist et al., 2013) between 0.01 Hz and 0.09 Hz (e.g., Stefanov et al., 2020), which filters low frequencies (e.g., physiological noise). From the number of noise terms included in this denoising strategy, the effective degrees of freedom of the BOLD signal after denoising were estimated to range from 62.1 to 74.1 (average 70.5) across all subjects (Nieto-Castanon, 2022; see Morfini et al., 2023 for a full description of quality control measures, including degrees of freedom). Following quality assurance, 479 participants (242 males, 237 females, 18 -87 years old, mean= 50.7, SD= 18.2) were entered into the seed-based connectivity analysis. The remaining cases were rejected for reasons such as having more than 10% of invalid scans, motion above 0.5 mm, etc. Anatomical methods Regions of interests (ROIs) for the hippocampus were defined using probabilistic, anatomical atlases. Hippocampal seed ROIs were created by combining the hippocampal ROI from the Harvard -Oxford subcortical atlas and the subiculum ROI from the Jülich histological atlas (Amunts et al., 2005) ensuring that the hippocampal ROI was extended medially incorporating the subicular complex (see Hodgetts et al., 2017). The Harvard-Oxford atlas ROI was thresholded at 50% and the Jülich atlas ROI thresholded at 75% to ensure both ROIs were constrained to grey matter and did not extend into adjacent regions. Using this method, left and right hippocampal ROIs were defined (Figure 1a). For the long-axis analysis, the hippocampal ROIs were split into anterior and posterior zones arbitrarily at the uncal apex (Hodgetts et al., 2017; Poppenk et al., 2013), corresponding to MNI slice y = -21 (see Figure 1b for segmentation of the left hippocampus). Functional correlations within the cerebellum were visualised and interrogated using the Spatially Unbiased Infratentorial Template (SUIT) (Diedrichsen et al., 2009). As well as containing a detailed probabilistic atlas of the cerebellum, SUIT also contains a cerebellar flatmap, allowing for visualisation of functional connectivity maps within and across cerebellar lobules/subregions. The nomenclature of Larsell and Jansen (1972) was used to characterise cerebellar cortical results. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 5 Figure 1. Hippocampal regions-of-interest. (a) The left hippocampal ROI made from combining the hippocampus from the Harvard-Oxford atlas and the subicular complex from the Jüelich Histological atlas; (b) The left and right hippocampal ROI split into anterior and posterior hippocampus using the uncal apex for a landmark-based segmentation. MRI analysis Following denoising (see above), a first -level seed-based connectivity analysis was conducted using the CONN toolbox. Here, the BOLD timeseries for each hippocampal seed (left hippocampus, right hippocampus, anterior hippocampus, posterior hippocampus), as well as confound regressors (e.g., white matter and CSF) were entered as first -level covariates into general linear models (GLMs). Functional connectivity between the seeds and every other voxel in the brain was represented by the Fisher-transformed bivariate correlation coefficient ( r-to-Z) from a weighted GLM. At the first level, we conducted eight analyses whereby four of them related to single seed and four of them to between - seed contrasts. Here we considered the issue of shared variance, and this was minimised by the use of separate GLMs for each seed which avoids multicollinearity between seeds in the same model. For the single seed analyses, the left, right, anterior, and posterior hippocampus design matrices contained a single column representing all subjects, allowing the investigation of seed -to-voxel connectivity for each ROI in isolation. To examine hemispheric and long-axis differences in hippocampal connectivity with the cerebellum, we specified between-seed contrasts (referred to as between-source contrasts in CONN). Here, the design matrices included two columns representing each seed i n each contrast. Contrasts compared the connectivity of left and right hippocampus (contrast vectors: 1 -1 and -1 1). Others compared the connectivity of anterior and posterior hippocampus (contrast vector: 1 -1 and -1 1). For single seed analyses, beta images were carried to the second -level analysis for one-sample t-tests. For between- seed analyses, contrast images generated in the first -level analysis were carried over to the second - level analysis for one -sample t-test. To explore age -related variation in hippocampal -cerebellar functional correlations, we also re-specified these GLMs with age as an additional subject -level effect, firstly as a regressor-of-no-interest (contrast vector: 1 0), and then as a regressor -of-interest (contrast vector: 0 1) into a bivariate regression analysis. For these contrasts, age was mean -centred and to ensure that we captured anticorrelations, we applied a contrast vector of 0 -1 to the demeaned age data. The unthresholded group-level t-statistic maps (as well as between-source contrast maps) from CONN were interrogated in SPM. These were thresholded using a family -wise error correction p-FWE <0.05 based on Random Field Theory (Worsley et al., 1996). The thresholded images were then loaded into .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 6 SUIT to localise (and visualise) significant clusters within cerebellar subregions. This produced cerebellar flatmaps showing suprathreshold connectivity strength within and across cerebellar lobules.

The hippocampus is functionally connected to widespread areas of the cerebellar cortex. Initially, we used seed-based connectivity to examine patterns of cerebellar connectivity from left and right hippocampal seed ROIs. As predicted by anatomical studies, both the left and right hippocampus showed bilateral connections to widespread areas of the cerebellum, including the vermis and lateral hemispheres. Both hemispheres showed strongest connections with the border of lobule HIV and HV, HIX, HX, and lobule HVIIA (medial parts of Crus II and laterally within the horizontal fissure at the junction of Crus I and Crus II; Figure 2a-b). Figure 2. Cerebellar regions showing significant functional connectivity with left and right hippocampus. Thresholded SPM {T} maps (p<0.05) overlaid on the cerebellar flatmap using SUIT (Diedrichsen et al., 2006) and on coronal and mid-sagittal sections of the standard MNI T1 2mm template. Contrasts for main effects of LEFT and RIGHT hippocampal seeds (a) and (b), respectively; contrasts for LEFT > RIGHT and RIGHT > LEFT hippocampal seeds, (c) and (d), respectively. To examine hemispheric differences, cerebellar functional correlations with left and right hippocampus were contrasted directly. The left hippocampus, compared to the right hippocampus, showed preferential connections with a contralateral region of lobule HVIIA (Crus I and peak in Crus II; number of voxels = 3063; p<.001), as well as lobule HIX (number of voxels = 181, p<.001; Figure 2c). Similarly, the right hippocampal seed also showed a strong preferential functional correlation with the contralateral (i.e., left) region of lobule HVIIA (peak in Crus II; number of voxels = 178, p<.001; Figure 2d). Peak cluster statistics for these hemispheric contrasts are shown in Table 1. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 7 Long-axis subdivisions of the hippocampus show both overlapping and distinct patterns of cerebellar connectivity. Next, we examined hippocampal -cerebellar functional connectivity along the longitudinal axis of the hippocampus. The anterior and posterior hippocampus showed connectivity to similar regions as those connected to the left and right hippocampus, including t he border of lobule HIV and HV, vermal parts of lobule V, lobule HVIIA (bilaterally at the fissure separating Crus I and II), lobule HIX and HX (Figure 3a-b). Figure 3. Cerebellar regions showing significant functional connectivity with anterior and posterior hippocampal seeds. Thresholded SPM {T} maps (p<0.05) overlaid on coronal, sagittal slices and cerebellar flatmaps. Contrasts for main effects of ANTERIOR and POSTERIOR hippocampal seeds (a) and (b), respectively (c) represents contrast for ANTERIOR > POSTERIOR hippocampal seed connectivity with the cerebellum. (d) represents contrast for POSTERIOR > ANTERIOR hippocampal seed connectivity with the cerebellum. However, direct contrasts revealed that the anterior hippocampus, compared to the posterior hippocampus, showed significantly greater functional correlations with lobule HI-HIV (number of voxels = 185, p<.001) and bilateral regions of lobule HVIIA, including right Crus II (number of voxels = 2778, p<.001; Figure 3c). Note, this same region of Crus II also displays significantly greater connectivity with left versus right hippocampus (Figure 2c). Additional clusters were also seen in extreme areas of right lobule HIX and HX, and bilateral HVIIB and HVIII. The posterior versus anterior hippocampus contrast showed significantly greater connectivity with widespread regions of the anterior cerebellum. Specifically, we observed a peak in the bilateral extremes of the Crus I region of lobule HVIIA (number of voxels = 4761; p<.001), which extended medially into vermal areas of lobule V (see Figure 3d). Further small clusters were found in the lateral extremes of lobule HVIIIA, bilaterally, which continued along the border of HVIIIB and HIX. Peak cluster statistics for these direct long-axis contrasts are shown in Table 1. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 8 Contrasts Cerebellar lobule Peak coordinate (mm) Probabilistic value z t Left > right hippocampus Right lobule HVIIA (Crus II) 36 -80 -42 Right Crus II (97%) Inf 10.7 28 -84 -32 Right Crus I (80%) Inf 10.14 Right Crus II (19%) 28 -76 -44 Right Crus II (76%) inf 9.23 Right lobule HIX 4 -52 -48 Right HIX (93%) Inf 8.28 Brain stem (1.2%) 6 -48 -40 Right HIX (83.2%) 7.69 7.94 Brain stem (16.8%) Right > left hippocampus Left lobule HVIIA (Crus II) -38 -64 -44 Left Crus II (87%) 6.34 6.48 Left Crus I (3%) -36 -70 -50 Left Crus II (63%) 5.96 6.07 Left VIIB (34%) -28 -74 -50 Left VIIB (63%) 5.66 5.76 Left Crus II (36%) Right lobule HI-HIV 12 -34 -20 Right HI-HIV (60%) 6.12 6.24 Brain stem (24.4%) Left lobule HVIIA (Crus II) -12 -78 -34 Left Crus II (87%) 5.64 5.74 Right lobule HVIIIA 32 -44 -56 Right HVIIIA (28%) 5.49 5.58 Right HVIIIB (6%) Left lobule VI -36 -36 -38 Left VI (62%) 5.39 5.48 Left V (4%) Left lobule VIIB -20 -76 -52 Left VIIB (88%) 5.21 5.28 Left Crus II (10%) Left lobule VIIB -40 -54 -58 Left VIIB (40%) 5.19 5.27 Left VIIIA (26%) Right lobule HVIIIB 20 -44 -56 Right HVIIIB (89%) 5.09 5.16 Right HIX (2%) Left lobule HVIIB -32 -70 -58 Left HVIIB (77%) 5.03 5.1 Left HVIIIA (6%) Anterior > posterior hippocampus Right lobule HI-HIV 12 -42 -28 Right HI-HIV (47%) Inf 13.07 Brain stem (4%) 20 -36 -30 Right HI-HIV (37%) Inf 12.11 Brain stem (15.1%) 22 -46 -30 Right HV (17%) Inf 10.28 Right VI (14%) 28 -84 -44 Right Crus II (92%) Inf 12.13 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 9 Right lobule HVIIA (Crus II) 36 -80 -40 Right Crus II (93%) Inf 11.99 Right Crus I (6%) 14 -84 -46 Right Crus II (79%) Inf 10.26 Right HVIIB (7%) Brain stem -8 -42 -28 Brain stem (69.7%) Inf 11.48 Left HI-HIV (19.0%) -18 -34 -30 Brain stem (48.6%) Inf 11.23 Left HI-HIV (20%) -26 -42 -34 Left HVI (7%) 7.68 7.93 Left HV (5%) Right lobule HIX 6 -46 -36 Right HIX (32%) Inf 8.96 Brain stem (11.3%) -2 -46 -34 Vermal X (39%) 7.37 7.59 Brain stem (2.1%) 0 -48 -44 Right HIX (42%) 6.43 6.57 Left IX (14%) Right lobule VIIIB 12 -40 -60 Right HVIIIB (12%) 5.19 5.26 Right lobule HVIIA (Crus I) 46 -58 -34 Right Crus I (93%) 5.05 5.12 Posterior > anterior hippocampus Left lobule HVIIA (Crus I) -44 -42 -34 Left Crus I (49%) Inf 16.6 Left HVI (15%) 46 -40 -34 Right Crus I (22%) Inf 15.96 Right HVI (3%) 0 -60 -6 Left HV (47%) Inf 15.8 Right HV (42%) Left lobule HVIIB -30 -68 -48 Left HVIIB (70%) 7.78 8.04 Left Crus II (17%) -40 -60 -42 Left Crus II (49%) 6.92 7.1 Left Crus I (48%) Left lobule HIX -10 -52 -50 Left HIX (71%) 6.55 6.7 Left HVIIIB (2%) Right lobule HVIIA (Crus I) 50 -62 -26 Right Crus I (64%) 6.33 6.47 Temporal Occipital Fusiform (3%) Right lobule HVIIA (Crus I) 20 -90 -26 Right Crus I (54%) 5.19 5.27 Right Crus II (23%) Table 1. Significant clusters identified within the cerebellum. Peak coordinates, z- and t-values are reported for all clusters. All statistics are reported at FWE p<0.05. NOTE: Coordinates are reported in MNI 2 mm space. Negative X-coordinates indicate the left hemisphere. Inf indicates large values that surpass threshold. Probabilistic values are derived from SPM Anatomy Toolbox (Eickhoff et al., 2005). .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 10 Figure 4. Regions in cerebellar cortex showing age -related decreases in functional connectivity with key hippocampal seeds. (a) Thresholded SPM {T} maps ( p<0.05) overlaid on cerebellar flatmaps showing ageing main effect for left (a), right (b), anterior (c) and posterior (d) hippocampus to the cerebellum. Darker colours indicate decreased functional connectivity with seed ROI with age. Hippocampal-cerebellar functional connectivity is reduced in ageing Finally, we examined the effect of age on hippocampal-cerebellar functional connectivity. We observed that several cerebellar regions showed negative correlations between age and hippocampal-cerebellar functional connectivity (i.e., lower connectivity in older participants). For the left hippocampal seed, we saw large age -related reductions in the depths of the right primary fissure that separates cerebellar lobules HV and HVI (number of voxels = 382; p<.001), which extended into adjacent parts of lobule HVIIA (Crus I; see Figure 4a). A similar spatial pattern of connectivity alterations was seen for the right hippocampus, but also revealed a strong cluster in the contralateral region of lobule HVI (number of voxels = 71; p<.001 (see Figure 4b). No differences were observed when directly contrasting left and right hippocampus. Next, we explored whether these age -related effects within the cerebellum differed across different hippocampal long -axis subdivisions. Notably, we found that the anterior hippocampus showed widespread age-related connectivity reductions within the cerebellum, and these were present primarily within the primary fissure between lobules HV and HVI (number of voxels = 837; p<.001; see Figure 4c). In contrast, the posterior hippocampus showed minimal age -related changes in functional connectivity, with very sma ll suprathreshold clusters observed in dispersed areas of the cerebellum, including lobules HV and HVI, with a peak in lobule HIX (number of voxels = 1; p = .026; see Figure .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 11 4d). However, no age-dependent connectivity differences were observed when contrasting anterior and posterior hippocampus. Cluster statistics for the ageing analyses are shown in Table 2. Seed ROI Peak cerebellar lobule Peak coordinate (mm) Probabilistic value z t Left hippocampus Right lobule HVI 32 -42 -26 Right HVI (61%) 6.7 6.87 Right HV (11%) 14 -58 -16 Right HV (55%) 6.52 6.67 Right HVI (45%) 22 -50 -22 Right HVI (68%) 6.45 6.59 Right HV (31%) Vermal lobule VI 0 -66 -16 Vermis VI (32%) 5.89 6 Right HV (26%) Left lobule HV -10 -56 -14 Left HV (95%) 5.85 5.96 Left HI-HIV (2%) Vermal lobule VIIIA 6 -64 -30 Vermis VIIIA (23%) 5.58 5.67 Right HVIIIA (8%) Left lobule HVI -10 -64 -14 Left HVI (83%) 5.52 5.62 Left HV (16%) Right lobule HVIIIA 14 -66 -44 Right HVIIIA (56%) 5.48 5.57 Right HVIIB (15%) 16 -58 -46 Right HVIIIB (46%) 5.45 5.54 Right HIX (13%) Left lobule HVI -30 -48 -26 Left HVI (94%) 5.46 5.55 Left HV (5%) -32 -38 -28 Left HVI (46%) 5.27 5.35 Left HV (23%) Left lobule HVI -22 -56 -20 Left HVI (92%) 5.44 5.52 Left HV (7%) Left lobule HVIIIA -24 -56 -48 Left HVIIIA (47%) 5.17 5.25 Left HVIIIB (39%) Right lobule HVIIIA 28 -50 -48 Right HVIIIA (62%) 5.11 5.18 Right HVIIIB (17%) Right lobule HVIIB 6 -72 -42 Right HVIIB (51%) 5.01 5.08 Right HVIIIA (19%) Right lobule HI-HIV 22 -32 -24 Right HI-HIV (30%) 4.98 5.04 Parahippocampal gyrus (23%) Left lobule HVI -32 -52 -24 Left HVI (95%) 6.21 6.34 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 12 Right hippocampus Right lobule HV 12 -60 -16 Right HV (56%) 6.12 6.24 Right HVI (42%) 14 -52 -18 Right HV (88%) 5.19 5.27 Right HVI (5%) Right lobule HVI 32 -42 -26 Right HVI (61%) 6.07 6.19 Right HV (11%) 24 -48 -22 Right HVI (68%) 6.01 6.13 Right HV (30%) 22 -30 -26 Right HI-HIV (44%) 5.92 6.04 Parahippocampal gyrus (16%) Vermal lobule VI 0 -66 -16 Vermis VI (32%) 5.59 5.69 Right HV (26%) Left lobule HVI -24 -54 -20 Left HVI (90.1%) 5.31 5.39 Left HV (6.9%) Vermal lobule VIIIA 0 -74 -40 Vermis VIIIA (56%) 5.08 5.15 Vermis VIIB (8%) Left lobule HV -10 -54 -14 Left HV (89%) 4.99 5.05 Left HI-HIV (8%) Anterior hippocampus Right lobule HV 12 -60 -16 Right HV (56%) 7.65 7.89 Right HVI (42%) 32 -42 -26 Right HVI (61%) 7.41 7.63 Right HV (11%) 0 -66 -16 Vermis VI (32%) 7.02 7.21 Right HV (26%) Left lobule HVI -22 -56 -20 Left HVI (92%) 6.06 6.19 Left HV (7%) -32 -50 -26 Left HVI (98%) 6.01 6.13 Left HV (2%) -24 -46 -22 Left HVI (49%) 5.89 6 Left HV (48%) Left lobule HVIIIA -24 -56 -48 Left HVIIIA (47%) 5.41 5.49 Left HVIIIB (39%) Left lobule HV -18 -42 -18 Left HV (71%) 5.12 5.19 Left HI-HIV (16%) Right lobule HVIIIB 14 -60 -44 Right HVIIIB (33%) 5.09 5.16 Right HIX (11%) Right lobule HVIIA (Crus I) 48 -62 -26 Right Crus I (83%) 5.07 5.14 3% Temporal Occipital Fusiform .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 13 Left lobule HV -22 -36 -22 Left HV (40%) 5.07 5.14 Parahippocampal gyrus (24%) Left lobule HVIIA (Crus I) -40 -72 -24 Left Crus I (96%) 5.03 5.1 Vermal lobule VIIIA 6 -64 -30 Vermis VIIIA (23%) 5.02 5.09 Right HVIIIA (8%) Posterior hippocampus Left lobule HIX -2 -54 -56 Left HIX (53%) 5.1 5.18 Right HIX (17%) Right lobule HVI 24 -46 -22 Right HVI (53%) 4.99 5.06 Right HV (45%) Left lobule HVI -32 -50 -26 Left HVI (98%) 4.98 5.04 Left HV (2%) Table 2. Significant clusters displaying significant age-related reductions in functional connectivity in the cerebellum. Peak coordinates, z- and t-values are shown for all clusters. All statistics are reported at FWE p<0.05. NOTE: Coordinates are reported in MNI 2 mm space. Negative X -coordinates indicate the left hemisphere. Probabilistic values are derived from SPM Anatomy Toolbox (Eickhoff et al., 2005).

While prior evidence in nonhuman animals suggest that a close functional interaction exists between the hippocampus and cerebellum, there is a limited understanding of the nature and topography of this functional connection in humans and how it be might va ry with age. In the current study, we sought to close this gap by applying seed -based functional connectivity analyses to resting -state fMRI data collected in a large -scale lifespan cohort (N = 479 from CamCAN). Our results yielded several key findings. Firstly, the shared bilateral functional connectivity between the left and right hippocampus with cerebellum involves four areas of the cerebellar cortex bilaterally: (i) almost the whole of the anterior lobe, and in addition, the anterior bank of lobule HVI ; (ii) lobule HVIIA, both the anterior and posterior bank in the mid-point of the horizontal fissure; (iii) a separate region in lobule HVIIA in Crus II, adjacent to a medial part of the ansoparamedian fissure; and (iv) lobules HIX and HX. We also consider ed laterality-related connectivity differences. When directly contrasting the left and right hippocampus, we found that the left hippocampus showed greater connectivity with contralateral lobules HVIIA (Crus I and Crus II) and HIX, whilst the right hippoca mpus showed greater connectivity with the contralateral lobule HVIIA (Crus II). Second, we found that the shared functional connectivity of the anterior and posterior hippocampus with the cerebellum encompassed similar areas as the bilateral shared functional connectivity of the left and right hippocampus. When examining differential functional connectivity between anterior and posterior hippocampus, we found that these individually correlated with distinct parts of the cerebellar cortex. The anterior compared to posterior hippocampus contrast revealed largely medial cerebellar connectivity dominated by areas posterior to the horizontal fissure (including the mid -portion of lobule HVIIA, extending into more medial parts of HVIIB and HVIII and small areas of lobule HIX and HX). The posterior compared to anterior hippocampus contrast revealed connectivity with areas of the cerebellar cortex that were largely anterior to the horizontal fissure, including the vermal portion of lobules I to VII, and hemispherical portions of lobules HI to HVIIA, including Crus I. Finally, we observed age-related decreases in hippocampal-cerebellar connectivity that were confined largely to areas of the anterior lobe. Similar patterns were observed for left and right hippocampus, whereby decreases were observed in lobule HVI, and ar ound the right primary fissure. Anterior hippocampus showed age-related decreases in connectivity with those same areas of the cerebellar .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 14 cortex, whereas the posterior hippocampus revealed no appreciable effects in any part of it. Previous electrophysiological studies have suggested a strong functional interaction exists between the hippocampus and cerebellum, based on the effects of cerebellar disruption on hippocampal place cell firing (Lefort et al., 201 5; Rochefort et al., 2011) and hippocampal -cerebellar synchronous oscillatory activity (Watson et al., 2019). The current findings support notions that these functional interactions exist in the human brain but, importantly, elucidates the topography/spatia l organisation of these functional interactions in the cerebellum. The findings of hippocampal -cerebellar connectivity within cerebellar lobule HIV, HV, HVI and bilateral regions of lobule HVIIA (Crus I and Crus II) aligns with prior electrophysiological findings (e.g., Babayan et al., 2017; Watson et al., 2019; Zeidler et al., 2020) and anatomical work in mice (e.g., Bohne et al., 2019; Watson et al., 2019). For instance, hippocampal field potentials have been shown to be modulated, most strongly, by stimulation of (vermal) lobule s IV-V (Zeidler et al.,2020), and phase -locked theta coherence has been observed between dorsal hippocampus and cerebellar lobule VI and VIIA (Crus I; Watson et al., 2019). Likewise, the combination of c-Fos imaging and graph theory suggest that two cerebe llar networks – one including lobule IV -V, VI, lobule VIIA (Crus I), and other including lobule IX, X – are differentially connected with dorsal hippocampal CA1 neurons during spatial exploration (Babayan et al., 2017). Notably, these same cerebellar lobules are also functionally correlated with left, right, anterior and posterior hippocampal seeds in this study (see also Figure 2a-b, 3a-3b). Our findings also align with previous tracing studies. For example, tracing injections into the dentate gyrus of the mouse hippocampus identified polysynaptic connectivity with lobule HIV and lobule HV (Bohne et al., 2019). Further, retrograde tracing stud ies also identified inputs to the left hippocampus from lobule HVI and lobule HVIIA (Crus I) (Watson et al., 2019) – all regions which were also found be functionally connected with the hippocampus in the current study, some of these even showing preferential connections with specific hippocampal regions (e.g., lobule HIV, HV and HVI with posterior hippocampus). Our findings also provide novel insights into hippocampal -cerebellar connectivity in humans. For example, in contrast to prior work in nonhuman species (discussed above), the strongest connectivity in lobule HVIIA was found in Crus II rather than the more commonly reported Crus I (e.g., Watson et al., 2019). This was particularly evident when contrasting anterior and posterior hippocampus (a comparison which has not been made in the animal studies). Specifically, while each long -axis subdivision showed similar patterns of cerebellar connectivity when considered in isolation (and indeed showed patterns akin to the hippocampus as a whole), when directly contrasted we found that the anterior hippocampus was strongly correlated with Crus II (extending into lobul e HVIIB and HVIII). In comparison, the posterior hippocampus showed the strong functional correlation with lobule HVIIA and extending HV (extending into lobule VI). In fMRI studies, Crus II has been shown to be involved in more abstract, non -motor function s, such as first - and second -order rule learning (Balsters et al., 2013), creative thinking (Gao et al., 2020), and mentalising (Guell and Schmahmann, 2020; Van Overwalle et al., 2020). In contrast, the regions identified in the posterior versus anterior c ontrast (notably lobule V) are implicated in sensorimotor processing (e.g., Bushara et al., 2001) and fine -grained digit representations (Grodd et al., 2001; van der Zwaag et al., 2013). The differential connectivity of anterior and posterior hippocampus to functionally distinct regions of the cerebellum is interesting as it suggests that functional subdivisions within the hippocampal formation – which are considered to emerge partially from differences in neocortical connectivity (Adnan et al., 2015; Aggle ton, 2012; Dalton et al., 2019) – are also reflected within the cerebellum. The cerebellum is thought to have motor and non-motor loops with different cortical areas (Ramnani et al., 2006). Given our data, it is possible that the anterior hippocampus participates more strongly in the cerebellar non-motor loop (consistent with its increased connectivity with prefrontal cortex), whilst the posterior hippocampus participates more strongly in the cerebellar motor loop. Such a distinction aligns with current vie ws of long -axis specialisation in the hippocampus, in which anterior hippocampus provides more abstract or coarse-grained representations in spatial and episodic memory, whereas the posterior hippocampus supports more fine-grained spatial processing (Poppenk et al., 2013; Robin and Moscovitch, 2017). .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 15 Our findings also suggest that hippocampal-cerebellar connectivity decreases with age, and this was most evident in areas of cerebellar lobules HV, HVI and their border. This finding dovetails with previous studies that have observed structural and functio nal changes in both the hippocampus and the cerebellum throughout ageing, making it plausible that this reflects, in part, age-related alterations in hippocampal-cerebellar interactions. Indeed, the current findings show that the left, right and anterior hippocampus show age-related connectivity reductions with areas of cerebellar lobules HV, HVI and their border. This evidence is in accordance with previous evidence that reported atrophy in lobule HVI during ageing (Cui et al., 2020; Ramanoël et al., 2023) . Reduced grey matter volume in lobule HVI was also negatively associated with performance on a perspective-taking task (Ramanoël et al., 2023) – a task which is thought to involve hippocampally -dependent allocentric representations (Labash et al., 2020). Further, fMRI work in young adults reported increased but differential activation for place - versus sequence -based navigation in hemispheric lobule HVI (Iglói et al., 2015). The age -related reduction in hippocampal-lobule HVI connectivity seen here, theref ore, may partly reflect age-related changes in the ability to adopt allocentric or place-based strategies during navigation, as observed in other studies (Moffat and Resnick, 2002; Rodgers et al., 2012). In this context, it is important to note that while our results strongly support a close functional interaction between hippocampus and cerebellum, they cannot speak directly to how such connectivity supports spatial/mnemonic behaviour. In nonhuman species , there is growing evidence that the cerebellum directly influences hippocampal spatial representations and navigational behaviour (e.g., Rochefort et al., 2013). One particular fMRI investigation in humans found that distinct hippocampal -cerebellar connectivity patterns might relate to the application of different strategies during virtual navigation. For instance, it was found that the right hippocampus and contralateral lobule HVIIA (Crus I) co -activated during place -based navigation whilst left hippocam pus and contralateral lobule HVIIA (Crus I) co - activated to support sequence-based navigation (Iglói et al., 2015). These same cerebellar regions were also strongly functionally correlated with the hippocampus in this study. One possible account for these interactions is that the cerebellum supports the updating and re - organisation of hippocampally-based spatial representations (e.g., cognitive maps) via its use of forward models (Ramnani, 2014; Rondi -Reig et al., 2022). This describes a model, created from an internal command, that is an internal representation of the behaviour modified by sensory feedback (Ito, 2008). Upon receiving sensory information that mismatches the internal representation, the cerebellum could support the re -computation of cognitive maps in the hippocampus when novel information is encountered. An additional finding in the current study was that several hippocampal seeds showed strong functional correlations with regions of the ‘vestibulocerebellum’ – namely, lobules IX and X. While speculative, this functional interaction is consistent with prio r work in rodents that has demonstrated the importance of vestibular information for the accuracy and stability of hippocampal representations (Sharp et al., 1995), and aligns with theoretical models that suggest that the cerebellum supplies the hippocampu s with information in an appropriate (world-centred) format to be used for allocentric navigation (Rochefort et al., 2013). Functional connectivity approaches also cannot elucidate how the hippocampus and cerebellum are structurally connected. Whilst it was previously speculated that a direct route exists between these regions (Heath and Harper, 1974), recent work points towards a more indirect route via a number of relay stations. These include thalamic regions (e.g., laterodorsal a nd ventrolateral thalamus; Bohne et al., 2019), which receive input from the cerebellum (Fujita et al., 2020) and project to areas that input to the hippocampus, such as the retrosplenial cortex (Van der Werf et al., 2002). Other relay stations include the septum and supramamillary nucleus, which are thought to be key regions involved in generating theta, and thus potentially relevant to theta couplin g observed between hippocampus and cerebellum during navigation/exploration (Watson et al., 2019) and cerebellar -dependent eye -blink conditioning (Hoffmann and Berry, 2009). It may be possible to map structural pathways between cerebellar regions/nuclei an d the human hippocampus by leveraging high -resolution diffusion MRI tractography, though the range and complexity of fiber populations (e.g., through the midbrain) would .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 16 require advanced modelling approaches and potential validation data in other species. One study, for example, used probabilistic constrained spherical deconvolution tractography (which enables complex/crossing fiber populations to be better resolved) and i dentified tractography streamlines between the hippocampus and cerebellar lobules HVIII, IX, X, Crus I, Crus II and the fastigial nucleus (Arrigo et al., 2014). Here, through the application of seed -based functional connectivity analyses within a large lifespan imaging cohort, we demonstrated that the human hippocampus is strongly functionally correlated with widespread areas of the cerebellar cortex, including lo bule I-IV, V, VI, Crus I, Crus II, IX and X. This finding provides additional important support for the idea that the two key brain structures, which are classically considered to underpin distinct memory systems, may collaborate closely in the human brain. Further, we show that hippocampal-cerebellar connectivity decreases significantly across the lifespan, particular in cerebellar regions that are vulnerable to age -related structural atrophy (lobules HV and HVI). It will be important for future studies to examine this interaction during behaviour, which will inform the development of more detailed neurobiological models of (spatial) learning and memory that incorporate these distinct regions, as well as examine the vulnerability of this functional link in neurodegenerative disorders (e.g., Alzheimer's disease).

This work was supported a Sarah Parker Remond PhD studentship at Royal Holloway, University of London, and the Biotechnology and Biological Sciences Research Council (BBSRC) [BB/V010549/1]. For the purpose of open access, the authors have applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising. We would also like to thank Samuel Berry, Marie-Lucie Read, Eleanor Alderman and Kieran Allen for their valuable advice during this project. CRediT author statement Kavishini Apasamy: Conceptualization, Formal Analysis, Data Curation, Visualization, Writing – Original Draft, Writing – Reviewing and Editing. Narender Ramnani: Conceptualization, Data Curation, Visualization, Supervision, Writing – Reviewing and Editing. Carl J. Hodgetts: Conceptualization, Visualization, Supervision, Writing – Reviewing and Editing, Funding acquisition.

Adnan, A., Barnett, A., Moayedi, M., McCormick, C., Cohn, M., & McAndrews, M. P. (2015). Distinct hippocampal functional networks revealed by tractography-based Parcellation. Brain Structure and Function, 221(6), 2999–3012. https://doi.org/10.1007/s00429-015-1084-x Aggleton, J. P. (2012). Multiple anatomical systems embedded within the primate medial temporal lobe: implications for hippocampal function. Neuroscience & Biobehavioral Reviews, 36(7), 1579-1596. Aggleton, J.P., Hunt, P. R., & Rawlins, J. N. P. (1986). The effects of hippocampal lesions upon spatial and non-spatial tests of working memory. Behavioural Brain Research, 19(2), 133– 146. https://doi.org/10.1016/0166-4328(86)90011-2 Aggleton, J. P., O’Mara, S. M., Vann, S. D., Wright, N. F., Tsanov, M., & Erichsen, J. T. (2010). Hippocampal–anterior thalamic pathways for memory: Uncovering a network of direct and indirect actions. European Journal of Neuroscience, 31(12), 2292–2307. https://doi.org/10.1111/j.1460-9568.2010.07251.x .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 17 Amunts, K., Kedo, O., Kindler, M., Pieperhoff, P., Mohlberg, H., Shah, N. J., Habel, U., Schneider, F., & Zilles, K. (2005). Cytoarchitectonic mapping of the human amygdala, hippocampal region and entorhinal cortex: Intersubject variability and probability maps. Anatomy and Embryology, 210(5–6), 343–352. https://doi.org/10.1007/s00429-005-0025-5 Andersson, J. L. R., Hutton, C., Ashburner, J., Turner, R., & Friston, K. (2001). Modeling Geometric Deformations in EPI Time Series. NeuroImage, 13(5), 903–919. https://doi.org/10.1006/nimg.2001.0746 Arrigo, A., Mormina, E., Anastasi, G. P., Gaeta, M., Calamuneri, A., Quartarone, A., De Salvo, S., Bruschetta, D., Rizzo, G., Trimarchi, F., & Milardi, D. (2014). Constrained spherical deconvolution analysis of the limbic network in human, with emphasis on a direct cerebello- limbic pathway. Frontiers in Human Neuroscience, 8. https://www.frontiersin.org/articles/10.3389/fnhum.2014.00987 Ashburner, J. (2007). A fast diffeomorphic image registration algorithm. NeuroImage, 38(1), 95–113. https://doi.org/10.1016/j.neuroimage.2007.07.007 Ashburner, J., & Friston, K. J. (2005). Unified segmentation. Neuroimage, 26(3), 839-851. Babayan, B. M., Watilliaux, A., Viejo, G., Paradis, A.-L., Girard, B., & Rondi-Reig, L. (2017). A hippocampo-cerebellar centred network for the learning and execution of sequence-based navigation. Scientific Reports, 7(1), Article 1. https://doi.org/10.1038/s41598-017-18004-7 Balsters, J. H., Whelan, C. D., Robertson, I. H., & Ramnani, N. (2013). Cerebellum and Cognition: Evidence for the Encoding of Higher Order Rules. Cerebral Cortex, 23(6), 1433–1443. https://doi.org/10.1093/cercor/bhs127 Bernard, J. A., & Seidler, R. D. (2014). Moving forward: age effects on the cerebellum underlie cognitive and motor declines. Neuroscience & Biobehavioral Reviews, 42, 193-207. Berron, D., Vogel, J. W., Insel, P. S., Pereira, J. B., Xie, L., Wisse, L. E., Yushkevich, P. A., Palmqvist, S., Mattsson-Carlgren, N., Stomrud, E., Smith, R., Strandberg, O., & Hansson, O. (2021). Early stages of tau pathology and its associations with functional connectivity, atrophy and memory. Brain, 144(9), 2771–2783. https://doi.org/10.1093/brain/awab114 Bohne, P., Schwarz, M. K., Herlitze, S., & Mark, M. D. (2019). A new projection from the deep cerebellar nuclei to the hippocampus via the ventrolateral and laterodorsal thalamus in mice. Frontiers in Neural Circuits, 13. https://doi.org/10.3389/fncir.2019.00051 Bushara, K. O., Wheat, J. M., Khan, A., Mock, B. J., Turski, P. A., Sorenson, J., & Brooks, B. R. (2001). Multiple tactile maps in the human cerebellum. Neuroreport, 12(11), 2483–2486. https://doi.org/10.1097/00001756-200108080-00039 Calhoun, V. D., Wager, T. D., Krishnan, A., Rosch, K. S., Seymour, K. E., Nebel, M. B., Mostofsky, S. H., Nyalakanai, P., & Kiehl, K. (2017). The impact of T1 versus EPI spatial normalization templates for fMRI data analyses. Human Brain Mapping, 38(11), 5331–5342. https://doi.org/10.1002/hbm.23737 Choe, K. Y., Sanchez, C. F., Harris, N. G., Otis, T. S., & Mathews, P. J. (2018). Optogenetic fmri and electrophysiological identification of region-specific connectivity between the cerebellar cortex and forebrain. NeuroImage, 173, 370–383. https://doi.org/10.1016/j.neuroimage.2018.02.047 Coffman, K. A., Dum, R. P., & Strick, P. L. (2011a). Cerebellar vermis is a target of projections from the motor areas in the cerebral cortex. Proceedings of the National Academy of Sciences, 108(38), 16068–16073. https://doi.org/10.1073/pnas.1107904108 Cui, D., Zhang, L., Zheng, F., Wang, H., Meng, Q., Lu, W., Liu, Z., Yin, T., & Qiu, J. (2020). Volumetric reduction of cerebellar lobules associated with memory decline across the adult lifespan. Quantitative Imaging in Medicine and Surgery, 10(1), 148–159. https://doi.org/10.21037/qims.2019.10.19 Dalton, M. A., McCormick, C., & Maguire, E. A. (2019). Differences in functional connectivity along the anterior-posterior axis of human hippocampal subfields. NeuroImage, 192, 38–51. https://doi.org/10.1016/j.neuroimage.2019.02.066 Daugherty, A. M., & Raz, N. (2017). A virtual water maze revisited: Two-year changes in navigation performance and their neural correlates in healthy adults. NeuroImage, 146, 492-506. Diedrichsen, J., Balsters, J., Flavell, J., Wams (néé Cussans), E., & Ramnani, N. (2009). A probabilistic MR Atlas of the human cerebellum. NeuroImage, 46, 39–46. https://doi.org/10.1016/j.neuroimage.2009.01.045 Dombrovski, A. Y., Luna, B., & Hallquist, M. N. (2020). Differential reinforcement encoding along the hippocampal long axis helps resolve the explore–exploit dilemma. Nature Communications, 11(1). https://doi.org/10.1038/s41467-020-18864-0 Du, A.-T., Schuff, N., Chao, L. L., Kornak, J., Jagust, W. J., Kramer, J. H., Reed, B. R., Miller, B. L., Norman, D., Chui, H. C., & Weiner, M. W. (2006). Age effects on atrophy rates of entorhinal .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 18 cortex and hippocampus. Neurobiology of Aging, 27(5), 733–740. https://doi.org/10.1016/j.neurobiolaging.2005.03.021 Eickhoff, S. B., Stephan, K. E., Mohlberg, H., Grefkes, C., Fink, G. R., Amunts, K., & Zilles, K. (2005). A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data. NeuroImage, 25(4), 1325–1335. https://doi.org/10.1016/j.neuroimage.2004.12.034 Ekstrom, A. D., & Ranganath, C. (2017). Space, time, and episodic memory: The hippocampus is all over the cognitive map. Hippocampus, 28(9), 680–687. https://doi.org/10.1002/hipo.22750 Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “mini-mental state.” Journal of Psychiatric Research, 12(3), 189–198. https://doi.org/10.1016/0022-3956(75)90026-6 Friston, Karl. J., Ashburner, J., Frith, C. D., Poline, J.-B., Heather, J. D., & Frackowiak, R. S. J. (1995). Spatial registration and normalization of images. Human Brain Mapping, 3(3), 165–189. https://doi.org/10.1002/hbm.460030303 Friston, K. J., Williams, S., Howard, R., Frackowiak, R. S. J., & Turner, R. (1996). Movement-Related effects in fMRI time-series. Magnetic Resonance in Medicine, 35(3), 346–355. https://doi.org/10.1002/mrm.1910350312 Fujita, H., Kodama, T., & du Lac, S. (2020). Modular output circuits of the FASTIGIAL nucleus for diverse motor and nonmotor functions of the cerebellar vermis. eLife, 9. https://doi.org/10.7554/elife.58613 Gao, Z., Liu, X., Zhang, D., Liu, M., & Hao, N. (2020). The indispensable role of the cerebellum in visual divergent thinking. Scientific Reports, 10(1), Article 1. https://doi.org/10.1038/s41598- 020-73679-9 Gellersen, H. M., Guell, X., & Sami, S. (2021). Differential vulnerability of the cerebellum in healthy ageing and alzheimer’s disease. NeuroImage: Clinical, 30, 102605. https://doi.org/10.1016/j.nicl.2021.102605 Glickstein, M. (1993). Motor skills but not cognitive tasks. Trends in Neurosciences, 16, 450–451. https://doi.org/10.1016/0166-2236(93)90074-V Glickstein, M. (2007). What does the cerebellum really do? Current Biology, 17(19). https://doi.org/10.1016/j.cub.2007.08.009 Grodd, W., Hülsmann, E., Lotze, M., Wildgruber, D., & Erb, M. (2001). Sensorimotor mapping of the human cerebellum: FMRI evidence of somatotopic organization. Human Brain Mapping, 13(2), 55–73. https://doi.org/10.1002/hbm.1025 Guell, X., & Schmahmann, J. (2020). Cerebellar Functional Anatomy: A Didactic Summary Based on Human fMRI Evidence. The Cerebellum, 19(1), 1–5. https://doi.org/10.1007/s12311-019- 01083-9 Hallquist, M. N., Hwang, K., & Luna, B. (2013). The nuisance of nuisance regression: Spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity. NeuroImage, 82, 208–225. https://doi.org/10.1016/j.neuroimage.2013.05.116 Heath, R. G., & Harper, J. W. (1974). Ascending projections of the cerebellar fastigial nucleus to the hippocampus, amygdala, and other temporal lobe sites: Evoked potential and histological studies in monkeys and cats. Experimental Neurology, 45(2), 268–287. https://doi.org/10.1016/0014-4886(74)90118-6 Henson, R. N. A., Buechel, C., Josephs, O., & Friston, K. J. (1999). The slice-timing problem in event- related fMRI. NeuroImage, 9, 125. Hodgetts, C. J., Voets, N. L., Thomas, A. G., Clare, S., Lawrence, A. D., & Graham, K. S. (2017). Ultra-High-Field fMRI Reveals a Role for the Subiculum in Scene Perceptual Discrimination. Journal of Neuroscience, 37(12), 3150–3159. https://doi.org/10.1523/JNEUROSCI.3225- 16.2017 Hoffmann, L. C., & Berry, S. D. (2009a). Cerebellar theta oscillations are synchronized during hippocampal theta-contingent trace conditioning. Proceedings of the National Academy of Sciences, 106(50), 21371–21376. https://doi.org/10.1073/pnas.0908403106 Iglói, K., Doeller, C. F., Paradis, A.-L., Benchenane, K., Berthoz, A., Burgess, N., & Rondi-Reig, L. (2015). Interaction Between Hippocampus and Cerebellum Crus I in Sequence-Based but not Place-Based Navigation. Cerebral Cortex, 25(11), 4146–4154. https://doi.org/10.1093/cercor/bhu132 Ito, M. (2008). Control of mental activities by internal models in the cerebellum. Nature Reviews Neuroscience, 9(4), 304–313. https://doi.org/10.1038/nrn2332 Jimsheleishvili, S., & Dididze, M. (2021). Neuroanatomy, Cerebellum. In StatPearls. StatPearls Publishing. http://www.ncbi.nlm.nih.gov/books/NBK538167/ Kelly, R. M., & Strick, P. L. (2003). Cerebellar loops with motor cortex and prefrontal cortex of a .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 19 nonhuman primate. The Journal of Neuroscience, 23(23), 8432–8444. https://doi.org/10.1523/jneurosci.23-23-08432.2003 Krook-Magnuson, E., Szabo, G. G., Armstrong, C., Oijala, M., & Soltesz, I. (2014). Cerebellar Directed Optogenetic Intervention Inhibits Spontaneous Hippocampal Seizures in a Mouse Model of Temporal Lobe Epilepsy. ENeuro, 1(1), ENEURO.0005-14.2014. https://doi.org/10.1523/ENEURO.0005-14.2014 Labash, A., Aru, J., Matiisen, T., Tampuu, A., & Vicente, R. (2020). Perspective Taking in Deep Reinforcement Learning Agents. Frontiers in Computational Neuroscience, 14, 69. https://doi.org/10.3389/fncom.2020.00069 Lace, G., Savva, G. M., Forster, G., de Silva, R., Brayne, C., Matthews, F. E., Barclay, J. J., Dakin, L., Ince, P. G., & Wharton, S. B. (2009). Hippocampal tau pathology is related to neuroanatomical connections: An ageing population-based study. Brain, 132(5), 1324–1334. https://doi.org/10.1093/brain/awp059 Larsell, O., & Jansen, J. (1972). The comparative anatomy and histology of the cerebellum: the human cerebellum, cerebellar connections, and cerebellar cortex Lefort, J. M., Rochefort, C., & Rondi-Reig, L. (2015). Cerebellar contribution to spatial navigation: New insights into potential mechanisms. The Cerebellum, 14(1), 59–62. https://doi.org/10.1007/s12311-015-0653-0 Lester, A. W., Moffat, S. D., Wiener, J. M., Barnes, C. A., & Wolbers, T. (2017). The Aging Navigational System. Neuron, 95(5), 1019–1035. https://doi.org/10.1016/j.neuron.2017.06.037 Middleton, F. A., & Strick, P. L. (2001). Cerebellar projections to the prefrontal cortex of the primate. The Journal of Neuroscience, 21(2), 700–712. https://doi.org/10.1523/jneurosci.21-02- 00700.2001 Moffat, S. D., & Resnick, S. M. (2002). Effects of age on virtual environment place navigation and allocentric cognitive mapping. Behavioral Neuroscience, 116(5), 851–859. https://doi.org/10.1037//0735-7044.116.5.85 Morfini, F., Whitfield-Gabrieli, S., & Nieto-Castañón, A. (2023). Functional connectivity MRI quality control procedures in Conn. Frontiers in Neuroscience, 17. https://doi.org/10.3389/fnins.2023.1092125 Murray, E. A., Wise, S. P., & Graham, K. S. (2018). Representational specializations of the hippocampus in phylogenetic perspective. Neuroscience Letters, 680, 4–12. https://doi.org/10.1016/j.neulet.2017.04.065 Newman, P. P., & Reza, H. (1979). Functional relationships between the hippocampus and the cerebellum: An electrophysiological study of the cat. The Journal of Physiology, 287, 405– 426. Nieto-Castanon, A. (2020). FMRI denoising pipeline. Handbook of Functional Connectivity Magnetic Resonance Imaging Methods in CONN, 17–25. https://doi.org/10.56441/hilbertpress.2207.6600 Nieto-Castanon, A. (2020). Handbook of functional connectivity Magnetic Resonance Imaging

in CONN. Hilbert Press. Nieto-Castanon, A. (2022). Preparing fMRI Data for Statistical Analysis (arXiv:2210.13564). arXiv. https://doi.org/10.48550/arXiv.2210.13564 Nieto-Castanon, A., & Whitfield-Gabrieli, S. (2022). Conn Functional Connectivity Toolbox: RRID SCR_009550, Release 22. https://doi.org/10.56441/hilbertpress.2246.5840 O’Reilly, J. X., Beckmann, C. F., Tomassini, V., Ramnani, N., & Johansen-Berg, H. (2009). Distinct and overlapping functional zones in the cerebellum defined by resting state functional connectivity. Cerebral Cortex, 20(4), 953–965. https://doi.org/10.1093/cercor/bhp157 Penny, W. D., Friston, K. J., Ashburner, J. T., Kiebel, S. J., & Nichols, T. E. (2011). Statistical Parametric Mapping: The Analysis of Functional Brain Images. Elsevier. Poppenk, J., Evensmoen, H. R., Moscovitch, M., & Nadel, L. (2013). Long-axis specialization of the human hippocampus. Trends in Cognitive Sciences, 17(5), 230–240. https://doi.org/10.1016/j.tics.2013.03.005 Power, J. D., Mitra, A., Laumann, T. O., Snyder, A. Z., Schlaggar, B. L., & Petersen, S. E. (2014).

to detect, characterize, and remove motion artifact in resting state fMRI. NeuroImage, 84, 320–341. https://doi.org/10.1016/j.neuroimage.2013.08.048 Ramanoël, S., Durteste, M., Perot, V., Habas, C., & Arleo, A. (2023). An Appraisal of the Role of the Neocerebellum for Spatial Navigation in Healthy Aging. The Cerebellum, 22(2), 235–239. https://doi.org/10.1007/s12311-022-01389-1 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 20 Ramnani, N. (2006). The primate cortico-cerebellar system: Anatomy and function. Nature Reviews Neuroscience, 7(7), 511–522. https://doi.org/10.1038/nrn1953 Ramnani, N. (2014). Automatic and controlled processing in the Corticocerebellar System. Progress in Brain Research, 255–285. https://doi.org/10.1016/b978-0-444-63356-9.00010-8 Ritchey, M., Libby, L. A., & Ranganath, C. (2015). Cortico-hippocampal systems involved in memory and cognition: the PMAT framework. Progress in Brain Research, 45–64. https://doi.org/10.1016/bs.pbr.2015.04.001 Riva, D., & Giorgi, C. (2000). The cerebellum contributes to higher functions during development: Evidence from a series of children surgically treated for posterior fossa tumours. Brain, 123(5), 1051–1061. https://doi.org/10.1093/brain/123.5.1051 Robin, J., & Moscovitch, M. (2017). Details, Gist and schema: Hippocampal–neocortical interactions underlying recent and remote episodic and Spatial Memory. Current Opinion in Behavioral Sciences, 17, 114–123. https://doi.org/10.1016/j.cobeha.2017.07.016 Rochefort, C., Arabo, A., André, M., Poucet, B., Save, E., & Rondi-Reig, L. (2011). Cerebellum Shapes Hippocampal Spatial Code. Science, 334(6054), 385–389. https://doi.org/10.1126/science.1207403 Rochefort, C., Lefort, J. M., & Rondi-Reig, L. (2013). The cerebellum: A new key structure in the navigation system. Frontiers in Neural Circuits, 7, 35. https://doi.org/10.3389/fncir.2013.00035 Rodgers, M. K., Sindone, J. A., & Moffat, S. D. (2012). Effects of age on navigation strategy. Neurobiology of Aging, 33(1), 202.e15-202.e22. https://doi.org/10.1016/j.neurobiolaging.2010.07.021 Rondi-Reig, L., Paradis, A.-L., & Fallahnezhad, M. (2022). A Liaison Brought to Light: Cerebellum- Hippocampus, Partners for Spatial Cognition. The Cerebellum, 21(5), 826–837. https://doi.org/10.1007/s12311-022-01422-3 Sharp, P., Blair, H., Etkin, D., & Tzanetos, D. (1995). Influences of vestibular and visual motion information on the spatial firing patterns of hippocampal place cells. The Journal of Neuroscience, 15(1), 173–189. https://doi.org/10.1523/jneurosci.15-01-00173.1995 Sladky, R., Friston, K. J., Tröstl, J., Cunnington, R., Moser, E., & Windischberger, C. (2011). Slice- timing effects and their correction in functional MRI. NeuroImage, 58(2), 588–594. https://doi.org/10.1016/j.neuroimage.2011.06.078 Stefanov, K., McLean, J., Allan, B., Cavanagh, J., & Krishnadas, R. (2020). Mild inflammation causes a reduction in resting-state amplitude of low-frequency fluctuation in healthy adult males. Brain and Neuroscience Advances, 4, 239821282094935. https://doi.org/10.1177/2398212820949353 Strange, B. A., Witter, M. P., Lein, E. S., & Moser, E. I. (2014). Functional organization of the Hippocampal Longitudinal Axis. Nature Reviews Neuroscience, 15(10), 655–669. https://doi.org/10.1038/nrn3785 Taylor, J. R., Williams, N., Cusack, R., Auer, T., Shafto, M. A., Dixon, M., Tyler, L. K., Cam-CAN, & Henson, R. N. (2017). The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) data repository: Structural and functional MRI, MEG, and cognitive data from a cross-sectional adult lifespan sample. NeuroImage, 144, 262–269. https://doi.org/10.1016/j.neuroimage.2015.09.018 Van der Werf, Y. D., Witter, M. P., & Groenewegen, H. J. (2002). The intralaminar and midline nuclei of the thalamus. Anatomical and functional evidence for participation in processes of arousal and awareness. Brain Research Reviews, 39(2), 107–140. https://doi.org/10.1016/S0165- 0173(02)00181-9 van der Zwaag, W., Kusters, R., Magill, A., Gruetter, R., Martuzzi, R., Blanke, O., & Marques, J. P. (2013). Digit somatotopy in the human cerebellum: A 7T fmri study. NeuroImage, 67, 354– 362. https://doi.org/10.1016/j.neuroimage.2012.11.041 Van Overwalle, F., Ma, Q., & Heleven, E. (2020). The posterior crus II cerebellum is specialized for social mentalizing and emotional self-experiences: A meta-analysis. Social Cognitive and Affective Neuroscience, 15(9), 905–928. https://doi.org/10.1093/scan/nsaa12 Watson, T. C., Obiang, P., Torres-Herraez, A., Watilliaux, A., Coulon, P., Rochefort, C., & Rondi-Reig, L. (2019). Anatomical and physiological foundations of cerebello-hippocampal interaction. ELife, 8, e41896. https://doi.org/10.7554/eLife.41896 Whitfield-Gabrieli, S., Nieto-Castanon, A., & Ghosh, S. (2011). Artifact detection tools (ART). Cambridge, MA. Release Version, 7(19), 11 Woodruff-Pak, D. S., Foy, M. R., Akopian, G. G., Lee, K. H., Zach, J., Nguyen, K. P. T., Comalli, D. M., Kennard, J. A., Agelan, A., & Thompson, R. F. (2010). Differential effects and rates of normal aging in cerebellum and hippocampus. Proceedings of the National Academy of .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint 21 Sciences of the United States of America, 107(4), 1624–1629. https://doi.org/10.1073/pnas.0914207107 Worsley, K. J., Marrett, S., Neelin, P., Vandal, A. C., Friston, K. J., & Evans, A. C. (1996). A unified statistical approach for determining significant signals in images of cerebral activation. Human Brain Mapping, 4(1), 58–73. https://doi.org/10.1002/(sici)1097-0193(1996)4:13.3.co;2-l Zeidler, Z., Hoffmann, K., & Krook-Magnuson, E. (2020). HippoBellum: Acute Cerebellar Modulation Alters Hippocampal Dynamics and Function. The Journal of Neuroscience, 40(36), 6910– 6926. https://doi.org/10.1523/JNEUROSCI.0763-20.2020 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted October 28, 2024. ; https://doi.org/10.1101/2024.10.28.620678doi: bioRxiv preprint