Preserved cerebellar functions despite structural degeneration in older adults
- Neuroscience of Movement and Performance Research Group, Department of Movement Sciences, KU Leuven, Belgium
- Leuven Brain Institute, KU Leuven, Belgium
- Waisman Center, The University of Wisconsin–Madison, United States
- Department of Communication Sciences and Disorders, University of Wisconsin–Madison, United States
Figures
Figure 1
Rhythmic finger tapping results.
(A) The average finger tap intervals per group, with shaded areas indicating 95% confidence intervals. The black dots represent the reference tapping interval (600 ms) from the initial auditory rhythm, which was played for the first 13 tones before becoming silent. Empty dots indicate the timing of the remaining, silent beats. (B) The clock variance, based on the last 30 unguided taps. (C) The mean tap intervals, based on the last 30 unguided taps shown with the reference interval (gray line). Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores. Analysis based on the data from 48 young adults, 77 older adults, and 29 older-old adults.
Figure 2
Grip force adjustment results.
(A) Illustration of grip force and load force peaks during three up-down arm movements. Green dots mark load force maxima, with green dashed lines indicating their timing. Orange dots represent grip force maxima, occurring just before the load force peaks. (B) Grip force anticipation is the time difference between grip and load force peaks; positive values indicate grip force peaks occurred earlier (i.e. anticipation). Analysis based on the data from 49 young adults, 77 older adults, and 30 older-old adults. (C) Average grip forces applied during the hold condition, displayed with the minimum required grip force indicated (dashed gray line). The extent to which the applied grip force exceeds this minimum reflects the grip force safety margin. Analysis based on the data from 14 young adults, 45 older adults,, and 30 older-old adults. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores.
Figure 3 with 1 supplement
Inter-joint coordination results.
(A) The timing of the activation of shoulder and elbow muscles before movement onset (at 0 s). Shoulder onset timing is indicated by the darkest color; elbow onset timing is indicated by the lighter color. (B) Shoulder anticipation timing indicates the time interval between the onset of shoulder muscle activity and the subsequent onset of elbow muscle activity. Analysis based on the data from 33 young adults, 56 older adults, and 27 older-old adults. (C) During elbow movement execution, the shoulder deviated from the fixed 60° angle (gray line) across all three age groups. (D) Total shoulder deviation during the movement. Analysis based on the data from 50 young adults, 80 older adults and 30 older-old adults. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores. See Figure 3—figure supplement 1 for the flexion condition.
Figure 3—figure supplement 1
Inter-joint coordination results from flexion condition.
(A) The timing of the activation of shoulder and elbow muscles before movement onset (at 0 s). Shoulder onset timing is indicated by the darkest color; elbow onset timing is indicated by the lighter color (B) Shoulder anticipation timing indicates the time interval between the onset of shoulder muscle activity and the subsequent onset of elbow muscle activity. (C) During elbow movement execution, the shoulder deviated from the fixed 60° angle (gray line) across all three age groups. (D) Total shoulder deviation during the movement.
Figure 4
Eye movement coordination results.
(A) The average eye movement velocity over trials for young and older adults. The dashed gray line represents the target speed, showing that the target remains stationary initially, then starts moving, and finally comes to a stop again. The gray areas indicate phases where the target was invisible, with the middle gray block representing the blanking period during which the target disappeared while moving (blank condition). Eye velocity in the blank condition (dashed colored lines) was decreased in both age groups compared to eye velocity during no-blank control trials (solid-colored lines), when the target remained visible for the whole trajectory. (B) Anticipatory eye velocity for the expected upcoming target motion. Analysis based on the data from 49 young adults, and 73 older adults. (C) Two examples of a young (left) and older participant (right) of saccadic and smooth pursuit displacements relative to total target displacement across individual trials (dots), during the blanking period. The fitted line represents the synergy between saccades and smooth pursuits, indicating mutual compensation. (D) Synergy between smooth pursuits and saccades during the blanking period. Analysis based on the data from 49 young adults, and 72 older adults. (E) Visually-guided eye velocity during eye tracking in the no-blank (control) condition. Analysis based on the data from 49 young adults, and 73 older adults. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores.
Figure 5 with 1 supplement
Results from the force matching test.
(A) Mean linear regression fits of the matched forces against each of the target forces for each age group. The dashed black line is a reference to indicate perfect correspondence between matched and target forces. The dashed colored line with corresponding circles show the results in the slider condition, which was used as a reference. The solid-colored line and corresponding squares show the button condition results. (B) Average overcompensation in the button condition. (C) Average slopes of linear relationship between target forces and matched forces during the slider condition, indicating sensory sensitivity. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores. Analysis based on the data from 48 young adults, 80 older adults, and 29 older-old adults. See Figure 5—figure supplement 1 for the lever condition.
Figure 5—figure supplement 1
Force matching results from the lever condition.
(A) Mean linear regression fits of the matched forces against each of the target forces for each age group. The dashed black line is a reference to indicate perfect correspondence between matched and target forces. The dashed colored line with corresponding circles show the results in the slider condition, which was used as a reference. The solid-colored line and corresponding triangles show the lever condition results. (B) Average overcompensation per age group in the lever condition.
Figure 6
Results from the reaching adaptation task.
(A) Trajectory of average baseline-corrected hand adaptation across all trials, per age group. Baseline correction used the final 20 trials of the baseline phase as a reference (shown by the first gray area). Perturbation began after trial 40 (vertical solid line). Breaks were provided at 80 and 160 trials after the onset of perturbation (dashed vertical lines at trials 120 and 200, respectively). The average late hand adaptation was calculated from the final 40 perturbation trials, referred to as the late adaptation period (second gray area). (B) Final hand adaptation. (C) Reach accuracy during the baseline phase. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores. Analysis based on the data from 50 young adults, 80 older adults, and 30 older-old adults.
Figure 7
Results from the speech adaptation task.
(A) Trajectory of the baseline-corrected F1 trajectories. Baseline correction used the final 30 trials of the baseline phase as a reference (first gray area). Perturbation was applied after trial 30 and removed after trial 120 (vertical lines). After every 25 trials, a break was provided (dashed vertical lines). (B) Final speech adaptation. (C) Baseline trial-to-trial variance in F1 frequency. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores. Analysis based on the data from 50 young adults, 79 older adults, and 28 older-old adults.
Figure 8
Results from the mental rotation task.
(A) Reaction times at different rotation angles (symbols) with corresponding regression lines. The black dashed lines represent the regression line of the young adults, for visual comparison with those of older and older-old adults. (B) Slopes of the reaction time regression lines, representing mental rotation speed. Steeper slopes indicate slower mental rotation performance. (C) Choice reaction time during baseline trials, when no mental rotation was required. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant scores. Analysis based on the data from 50 young adults, 80 older adults, and 30 older-old adults.
Figure 9
Cerebellar structural measure.
The left side of the figure represents our findings in gray matter volumetric results and the right side the white matter results, illustrated with the gray and white matter segments of a young participant. (A, B) Cerebellar gray and white matter as percentages of the TIV. (C, D) Our results (colored squares) compared to the Cam-CAN dataset. The gray circles indicate the individual data points from the Cam-CAN participants, while the black lines illustrate the quadratic age-related declines in both gray and white matter. (E, F) Cerebellar gray and white matter volumes as a percentage of the total gray matter and total white matter, respectively. Bar heights and error bars represent robust means and standard deviations, respectively. Dots show individual participant volumes.
Figure 10
Overview of effect sizes and Bayesian Factors across behavioral and structural measures.
Left panel: Each dot represents the effect size from statistical comparisons between age groups: young vs. older adults (orange) and young vs. older-old adults (pink). Horizontal error bars indicate the 95% confidence intervals. Positive values reflect better performance or larger brain volumes in the older groups relative to young adults, while negative values indicate poorer performance or smaller volumes. Cohen’s D effect sizes are interpreted as follows: no effect (<0.2, white), small effect (0.2–0.5, red), medium effect (0.5–0.8, yellow), and large effect (>0.8, green). Right panel: each dot represents the Bayesian factor from statistical comparisons between age groups: young vs. older adults (orange) and young vs. older-old adults (pink). Left from the gray area corresponds to the presence of evidence that the data from the young participants are not stronger (for cerebellar specific and general sensorimotor measures) or larger (for cerebellar volumes) than those of the older participants. On the right side, there is evidence that the performance of the young participants is stronger or that the cerebellar volumes of the young participants are larger than those from the older groups. The gray area represents the area of anecdotal evidence where no conclusion about an age-related effect can be drawn.
Figure 11
Grip force adjustment task.
(A) Participants held the force sensor between the thumb and index finger. Grip force and load force were measured in X and Z directions, respectively. (B) In the up-down condition, the manipulandum was lifted 10 cm from the table (dashed line). From there it was moved up and down over a distance of 30 cm (black curves). (C) In the hold condition, participants held the manipulandum stationary at 10 cm above the table.
Figure 12
Inter-joint coordination task.
The participant’s hand starts at the home position (white dot), with the elbow positioned at 75°. In this example, the task requires an elbow extension movement towards the goal target on the right (yellow dot), reaching an elbow angle of 45°. Throughout the movement, the shoulder angle should remain fixed at 60°.
Figure 13
Eye movement coordination task.
(1) The task starts with target fixation for 1 s. (2) followed by a brief disappearance of the target for 0.3 s. (3) When the target reappears it starts moving constantly from the left to the right side of the screen for 0.8 s. (4) In the blanking trials, the target is invisible for a part of the trajectory for 1 s. (5) For the last part of the moving trajectory the target reappears 0.6 s. (6) At the end of the trajectory the target disappears briefly before the next trial starts 0.3 s.
Figure 14
Force matching task.
(A) In the slider condition (control), participants used the slider located on top of the device to manually control the force applied to the finger via the torque motor. (B) In the button condition, a button was used which directly applied the same amount of force to the finger via the torque motor as was applied to press the button. (C) In the lever condition, participants pressed the lever on top of the finger to apply the force.
Figure 15
Reach adaptation task.
(A) In the baseline condition, participants moved the cursor, representing their hand location, from the home position to the goal target. (B) In the perturbation condition, the cursor was artificially offset from the actual hand position, with a clamped deviation of 30° counterclockwise from the goal target. The dashed arrows indicate the adjustment of the hand in the opposite direction of the perturbation, demonstrating adaptive behavior to the perturbation over trials.
Figure 16
Speech adaptation task.
During the baseline trials, participants pronounced the word ‘bed’ into a microphone and heard their own voice played back through headphones exactly as they had spoken it. During the perturbation trials, the first formant (F1) frequency of the pronounced word was altered, with a –125 Mel shift in the auditory feedback. As a result, when participants pronounced ‘bed,’ they heard it as sounding more like ‘bid’.
Figure 17
Mental rotation task.
A rotated letter was presented on the screen that had to be judged whether it was presented normally or mirrored by pressing the left or right arrow button, respectively. Feedback about the correctness of the response was given afterwards, followed by a short fixation period before the next trial started.
Tables
Table 1
Overview of behavioral test battery.
| Task name | Task protocol references | Cerebellar references | Cerebellar-specific measures | General sensorimotor measures |
|---|---|---|---|---|
| Rhythmic finger tapping | Wing and Kristofferson, 1973 | Franz et al., 1996 | Clock estimate extracted from tapping variance. Reflecting timekeeping abilities. | Total rhythm deviation. Reflecting rhythm tempo drift. |
| Grip force adjustment | Flanagan et al., 1993 | Rost et al., 2005 | Grip-load force peak timing difference. Reflecting grip force anticipation abilities. | Grip force surplus. Reflecting use of a safety margin. |
| Inter-joint coordination | Maeda et al., 2017 | Bastian et al., 1996 | Shoulder muscle onset timing. Reflecting shoulder anticipation for upcoming torque. | Shoulder angle deviation from fixed position. Reflecting inefficient motor coordination or reduced shoulder stability. |
| Eye movement coordination | Orban de Xivry et al., 2006 | Lekwuwa, 1995 | Anticipatory eye velocity. | Visually-guided eye velocity during target tracking. |
| - | Smooth pursuit vs. saccadic displacement. Reflecting eye movement coordination. | |||
| Force matching | Wolpe et al., 2016 | - | Force overcompensation during self-applied touch. Reflecting sensory attenuation. | Force matching discrimination between different target forces. Reflecting sensory sensitivity. |
| Reach adaptation | Morehead et al., 2017 | Morehead et al., 2017 | Final hand adaptation. Reflecting implicit motor adaptation. | Baseline reach error. Reflecting reach accuracy. |
| Speech adaptation | Parrell et al., 2017 | Parrell et al., 2017 | Final F1 component adaptation. Reflecting implicit speech adaptation. | Baseline F1 frequency. Reflecting voice frequency. |
| Mental rotation | McDougle et al., 2022 | McDougle et al., 2022 | Slope of reaction times with increased rotation. Reflecting mental rotation pace. | Choice reaction time at baseline. |
Key resources table
| Reagent type (species) or resource | Designation | Source or reference | Identifiers | Additional information |
|---|---|---|---|---|
| Software, algorithm | Matlab | Matlab | RRID:SCR_001622 | Multi-paradigm numerical computing environment and fourth-generation programming language developed by MathWorks. Allows matrix manipulations, plotting of functions and data, implementation of algorithms, creation of user interfaces, and interfacing with programs written in other languages; |
| Software, algorithm | SPM | SPM | RRID:SCR_007037 | Software package for analysis of brain imaging data sequences. Sequences can be a series of images from different cohorts, or time-series from the same subject. Current release is designed for analysis of fMRI, PET, SPECT, EEG, and MEG. |
| Software, algorithm | Computational Anatomy Toolbox for SPM (CAT) | SPM | RRID:SCR_019184 | Software toolbox as extension to SPM12 to provide computational anatomy. This covers diverse morphometric methods, such as voxel-based morphometry, surface-based morphometry, deformation-based morphometry, and region- or label-based morphometry. |
Appendix 1—table 1
Split-half reliability per task measure.
| Task | Cerebellar- specific | General sensorimotor | ||||||
|---|---|---|---|---|---|---|---|---|
| Measure | Young | Older | Older-Old | Measure | Young | Older | Older-Old | |
| Rhythmic finger tapping | Clock variance | 0.66 | 0.79 | 0.89 | Rhythm deviation | 0.98 | 0.98 | 0.99 |
| Grip force adjustment | Grip force anticipation | 0.94 | 0.92 | 0.84 | Grip safety margin | - | - | - |
| Inter-joint coordination | Shoulder anticipation | 0.87 | 0.83 | 0.89 | Shoulder deviation | 0.98 | 0.99 | 0.98 |
| Eye movement coordination | Eye velocity anticipation | 0.55 | 0.61 | - | Eye tracking accuracy | 0.97 | 0.97 | - |
| Saccade-pursuit coordination | 0.97 | 0.93 | - | |||||
| Force matching | Sensory attenuation | 0.94 | 0.94 | 0.96 | Sensory accuracy | 0.73 | 0.84 | 0.86 |
| Reach adaptation | Reach adaptation | 0.96 | 0.98 | 0.98 | Reach accuracy | 0.41 | 0.43 | 0.60 |
| Speech adaptation | Speech adaptation | 0.99 | 0.99 | 0.98 | Speech variance | 0.90 | 0.79 | 084 |
| Mental rotation | Mental rotation pace | 0.73 | 0.86 | 0.71 | Choice reaction time | 0.91 | 0.92 | 0.94 |
Author response table 1
Split-half reliabilities.
| Task | Cerebellar- specific | General sensorimotor | ||||||
|---|---|---|---|---|---|---|---|---|
| Measure | Young | Older | Older-Old | Measure | Young | Older | Older-Old | |
| Rhythmic finger tapping | Clock variance | 0.66 | 0.79 | 0.89 | Rhythm deviation | 0.98 | 0.98 | 0.99 |
| Grip force adjustment | Grip force anticipation | 0.94 | 0.92 | 0.84 | Grip safety margin | - | - | - |
| Inter-joint coordination | Shoulder anticipation | 0.87 | 0.83 | 0.89 | Shoulder deviation | 0.98 | 0.99 | 0.98 |
| Eye movement coordination | Eye velocity anticipation | 0.55 | 0.61 | - | Eye tracking accuracy | 0.97 | 0.97 | - |
| Force matching | Sensory attenuation | 0.94 | 0.94 | 0.96 | Sensory accuracy | 0.73 | 0.84 | 0.86 |
| Reach adaptation | Reach adaptation | 0.96 | 0.98 | 0.98 | Reach accuracy | 0.41 | 0.43 | 0.60 |
| Speech adaptation | Speech adaptation | 0.99 | 0.99 | 0.98 | Speech variance | 0.90 | 0.79 | 084 |
| Mental rotation | Mental rotation pace | 0.73 | 0.86 | 0.71 | Choice reaction time | 0.91 | 0.92 | 0.94 |
Additional files
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Preserved cerebellar functions despite structural degeneration in older adults
eLife 15:RP109440.
https://doi.org/10.7554/eLife.109440.3
