Exercise-induced neuroplasticity: Balance training increases cortical thickness in visual and vestibular cortical regions

Accepted Manuscript Exercise-induced neuroplasticity: Balance training increases cortical thickness in visual and vestibular cortical regions Ann-Kathrin Rogge, Brigitte Röder, Astrid Zech, Kirsten Hötting PII:

S1053-8119(18)30572-X

DOI:

10.1016/j.neuroimage.2018.06.065

Reference:

YNIMG 15071

To appear in:

NeuroImage

Received Date: 15 March 2018 Revised Date:

21 June 2018

Accepted Date: 24 June 2018

Please cite this article as: Rogge, A.-K., Röder, B., Zech, A., Hötting, K., Exercise-induced neuroplasticity: Balance training increases cortical thickness in visual and vestibular cortical regions, NeuroImage (2018), doi: 10.1016/j.neuroimage.2018.06.065. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

ACCEPTED MANUSCRIPT

RI PT

Exercise-induced neuroplasticity: Balance training increases cortical thickness in visual and vestibular cortical regions

Universität Hamburg, Biological Psychology & Neuropsychology, Germany 2 Friedrich Schiller University, Human Movement Science, Germany

M AN U

1

SC

Ann-Kathrin Rogge1*, Brigitte Röder1, Astrid Zech2 & Kirsten Hötting1

TE D

*corresponding author: Ann-Kathrin Rogge Universität Hamburg, Biological Psychology & Neuropsychology Von-Melle-Park 11, 20146 Hamburg, Germany +49 40 42838 4345 Fax: +49 40 42838 6591 [email protected]

EP

Brigitte Röder Universität Hamburg, Biological Psychology & Neuropsychology Von-Melle-Park 11, 20146 Hamburg, Germany [email protected]

AC C

Astrid Zech Friedrich Schiller University, Human Movement Science Seidelstraße 20, 07749 Jena, Germany [email protected]

Kirsten Hötting Universität Hamburg, Biological Psychology & Neuropsychology Von-Melle-Park 11, 20146 Hamburg, Germany [email protected]

2

ACCEPTED MANUSCRIPT

Abstract Physical exercise has been shown to induce structural plasticity in the human brain and to enhance cognitive functions. While previous studies focused on aerobic exercise,

RI PT

suggesting a link between increased cardiorespiratory fitness and exercise-induced neuroplasticity, recent findings have suggested that whole-body exercise with minor

metabolic demands elicit beneficial effects on brain structure as well. In the present study, we

SC

tested if balance training, challenging the sensory-motor system and vestibular self-motion perception, induces structural plasticity. Thirty-seven healthy adults aged 19-65 years were

M AN U

randomly assigned to either a balance training or a relaxation training group. All participants exercised twice a week for 12 weeks. Assessments before and after the training included a balance test and the acquisition of high-resolution T1-weighted images to analyze morphological brain changes. Only the balance group significantly improved balance

TE D

performance after training. Cortical thickness was increased in the superior temporal cortex, in visual association cortices, in the posterior cingulate cortex, in the superior frontal sulcus, and in the precentral gyrus in the balance group, compared to the relaxation group. Moreover,

EP

there was evidence that the balance training resulted in decreased putamen volume. Improved balance performance correlated with the increase of precentral cortical thickness and the

AC C

decrease in putamen volume. The results suggest that balance training elicits neuroplasticity in brain regions associated with visual and vestibular self-motion perception. As these regions are known for their role in spatial orienting and memory, stimulating visual-vestibular pathways during self-motion might mediate beneficial effects of physical exercise on cognition.

3

ACCEPTED MANUSCRIPT

1 Introduction Physical activity has been discussed as a promising means to enhance the life-long ability of the human brain to adapt to environmental demands. For instance, regular aerobic

RI PT

training has been shown to increase memory and executive functions (Cassilhas, Tufik, & Mello, 2016; Hötting & Röder, 2013; Stimpson, Davison, & Javadi, 2018). Accordingly, brain imaging studies have observed structural plasticity in the hippocampus to be associated

SC

with increased cardiorespiratory capacity (Erickson et al., 2011; Kleemeyer et al., 2015;

Maass et al., 2015). Aerobic training has been reported to increase hippocampal volume in

M AN U

young and in older adults (Erickson et al., 2011; Thomas et al., 2016). However, these results have not always been replicated (Jonasson et al., 2017; Ruscheweyh et al., 2011) and metaanalytic evidence for a direct link between aerobic training and hippocampal structural plasticity in adults is still lacking (Firth et al., 2017). Moreover, exercise studies have

TE D

observed structural changes beyond the hippocampus. For example, aerobic training has been reported to increase gray and white matter volume in prefrontal and temporal brain areas (Colcombe et al., 2006), and improved aerobic fitness was associated with increased regional

EP

white matter integrity in prefrontal and temporal cortices (Voss et al., 2013). Similar as aerobic exercise, practicing motor coordination and balance seem to elicit

AC C

structural brain changes too. For instance, motor coordination training has been related to hippocampal volume increase (Niemann, Godde, & Voelcker-Rehage, 2014). Short-term balance training resulted in gray matter increases in premotor, frontal, and parietal cortices (Taubert et al., 2010). The frontal cortex volume increase correlated with the improvement of balance performance, suggesting learning-related structural adaptations. Structural changes were not limited to cortical structures but included the putamen, for which a volume reduction was observed after the 6 weeks training period (Taubert et al., 2010). Moreover, decreased functional connectivity between the basal ganglia, in particular the caudate nucleus and the

4

ACCEPTED MANUSCRIPT putamen, and cortical areas after balance training have been observed in participants who improved in postural stability compared to a passive control group (Magon et al., 2016). (Niemann, Godde, Staudinger, & Voelcker-Rehage, 2014), in contrast, reported a volume increase of the putamen and the globus pallidus after 12 months of motor coordination

RI PT

training.

In rodents, running is known to induce hippocampal neurogenesis and to improve spatial learning (Kempermann, 2012; van Praag, Christie, Sejnowski, & Gage, 1999).

SC

Structural plasticity in the rodent hippocampus has been reported after motor skill learning on

M AN U

a rotarod, which involved balancing on a rotating cylinder (Curlik, Maeng, Agarwal, & Shors, 2013). Rotarod-training with strong balance requirements was associated with a volume increase in the prefrontal cortex, the thalamus and the amygdala, whereas volume decreases were found in the retrosplenial cortex, the cerebellum and the vestibular nuclei of the brainstem (Scholz, Niibori, Frankland, & Lerch, 2015). Moreover, behavioral improvements

TE D

after the training were associated with higher fractional anisotropy (FA) in several areas including the hippocampus and the striatum, but lower FA in the primary visual cortex and

EP

the entorhinal cortex. In mice, functional reorganization of the dorsomedial and dorsolateral striatum during motor learning and consolidation on a rotarod was observed using in vivo

AC C

striatal neural recordings (Yin et al., 2009). Thus, whole-body exercise challenging balance abilities in the absence of high

metabolic demands seems to elicit distinct structural and functional changes in the brain. It has been hypothesized that vestibular input during self-motion, inevitable during any type of physical activity, may be an essential mediator of exercise-induced neuroplasticity in general (Smith, 2017). The vestibular system assesses self-motion to quickly adjust eyes and other body parts for balance control. Balance control requires the integration of visual, proprioceptive, as well as motor-related multisensory cues, which takes place as early as in

5

ACCEPTED MANUSCRIPT

the first stage of central vestibular processing, that is, in the vestibular nuclei of the brain stem (Cullen, 2012). Pathways between the vestibular nuclei and the hippocampus, prefrontal and parietal areas provide information relevant for memory and spatial functions (Hitier, Besnard, & Smith, 2014), suggesting an essential contribution of the vestibular system to higher

RI PT

cognitive functions. For instance, vestibular deafferentiation has been reported to result in hippocampal atrophy and deficits in spatial memory (Brandt, 2005). By contrast, caloric and galvanic stimulations of the peripheral vestibular system has been found to improve verbal

SC

memory and spatial cognition (Bigelow & Agrawal, 2015). Cross-sectional studies on balance experts with several years of practice in ballet dancing or slacklining reported superior

M AN U

performance in a hippocampus-dependent learning task and a larger posterior hippocampal volume, but smaller anterior hippocampal (Hüfner et al., 2011) and putamen volumes (Hänggi, Koeneke, Bezzola, & Jäncke, 2010) compared to control participants with no experience in balance training. In addition, occipital gray matter volume was higher in the

al., 2011).

TE D

balance experts while less gray matter was found in the insular cortex in this group (Hüfner et

EP

Hence, whole-body exercise relying strongly on the vestibular system seems to result in specific structural adaptation of this system. We have recently reported that 12 weeks of

AC C

balance training in healthy adults improved not only balance performance, but additionally associative memory and spatial cognition, compared to a control group who had participated in a relaxation training (Rogge et al., 2017). It might be hypothesized that these behavioral improvements resulted from changes in brain structures which are both part of the vestibular network and the neural network mediating memory and spatial cognition. The goal of the present longitudinal study was to assess balance training-induced structural changes underlying cognitive benefits. To this end, healthy adults were randomly assigned to either a balance training or a relaxation training. All participants were scanned

6

ACCEPTED MANUSCRIPT

with structural magnet resonance imaging (MRI) before and after the 12-week training period. The balance training was designed to cover a broad spectrum of postural control demands which involve the integration of vestibular, visual and proprioceptive information. The relaxation training served as an active control condition. In the balance group we predicted

RI PT

changes in cortical thickness and gray matter volume in brain structures of the sensory-motor

AC C

EP

TE D

M AN U

SC

and vestibular system, including the hippocampus and the basal ganglia.

7

ACCEPTED MANUSCRIPT

2 Materials and Methods 2.1 Participants A detailed description of the participant flow, behavioral assessments, and training

RI PT

protocol has been published before (Rogge et al., 2017). Briefly, participants were recruited via public advertisements in the city of Hamburg (Germany). Healthy adults between 19-65 years of age who reported no regular physical exercise (no more than five exercise sessions a

SC

month during the last five years) and no extensive experience in balance training and

relaxation techniques were eligible for the study. Exclusion criteria were untreated heart

M AN U

diseases, untreated respiratory diseases, musculoskeletal illnesses, past lower extremity trauma, past traumatic brain injury and neurological or psychiatric diseases. After a prescreening by phone, seventy participants underwent an extensive sport medical examination to ensure that they were in appropriate constitution to take part. Two participants

TE D

did not get medical approval for a physical exercise training and nine participants withdraw their willingness to take part after pretesting. Thus, n = 59 were randomized and allocated to the training groups. Forty participants successfully completed the training program including

EP

posttests. Due to scanner incompatibilities (metallic implants), n = 3 were excluded from MRI

AC C

imaging, leaving n = 37 for the final analysis (aged 19-65 years; mean ± SD, 45.00 ± 14.86, 23 females). Of the final sample, n = 2 (one in the balance group, one in the relaxation group) were on stable hypertension medication and n = 2 (one in the balance group, one in the relaxation group) on stable thyroid hypofunction medication at pre- and posttest. The study was approved by the ethical board of the German Psychological Society (DGPs) and was conducted in accordance with the principles laid down in the Declaration of Helsinki. All participants gave written informed consent.

8

ACCEPTED MANUSCRIPT Based on a meta-analysis of previous balance interventions (Lesinski, Hortobágyi, Muehlbauer, Gollhofer, & Granacher, 2015), we expected a medium effect size for balance

improvements in untrained adults. Such an effect size can be statistically detected in a Time ×

RI PT

Group design with a total sample size of 34 participants (power = 0.80, alpha = 0.05).

2.2 Experimental design

Participants were grouped into matched pairs based on age, gender, and years of

SC

education. Individuals of each pair were randomly assigned to the balance or the relaxation group, using a random number generator. Randomization occurred after the pretesting. The

M AN U

characteristics of the participants are presented in Table 1.

Individuals were assessed with a balance test and structural MRI before and after the training period. The MRI and balance assessments were scheduled on different days, the order was counterbalanced across participants. The test assessors were blinded to the participants’

TE D

group assignment.

Participants underwent 12 weeks of training in their assigned training group with two

EP

training sessions per week, each lasting 50 minutes. The training took place in groups of 1012 individuals, separately for the balance group and the relaxation group. The same coaches

AC C

supervised both groups. The primary coach is holding a PhD in sport science and licenses in movement therapy and relaxation techniques. A second coach is holding a bachelor’s degree in sport pedagogy and a trainer license supervised less than 12 % of the training sessions. Moreover, the balance training and the relaxation training were comparable with regard to group size, training and session duration, number of social contacts, time of the day and training facility, yet only the experimental group received active physical training. All participants were briefed to not change their level of leisure physical activity throughout the training period.

9

ACCEPTED MANUSCRIPT 2.2.1

Balance training The balance training was designed as a circuit training with eight different stations per

session. At different stations, each lasting five minutes, participants trained on various surfaces, such as wobble boards, wobble cushions, foam, and perturbation platforms to

RI PT

challenge functional stability limits and induce reactive stabilization. The difficulty of the exercises was varied by using the following conditions: static vs. dynamic stability, eyes open vs. closed, narrowed base of support (bipedal vs. tandem vs. one-leg stance), and stable vs.

SC

unstable undergrounds (Horak, 1997; Sibley, Beauchamp, van Ooteghem, Straus, & Jaglal, 2015). Thus, an individually adapted exercise progression was possible which is in line with

M AN U

recent recommendation for balance trainings (Lesinski et al., 2015). Upon decision of the experienced coaches, exercising difficulty was increased. No strategies were taught. Exercise stations were replaced with new ones after six weeks of training to keep the training challenging. Balance exercises were selected based on existing literature and own experience

TE D

(Hrysomallis, 2011; Lesinski et al., 2015; Zech et al., 2010). The number of training sessions and the training period of 12 weeks was chosen according to the recommendations for balance

2015).

Relaxation training

AC C

2.2.2

EP

training in healthy participants (Gebel, Lesinski, Behm, & Granacher, 2018; Lesinski et al.,

The relaxation group practiced progressive muscle relaxation (Jacobson, 1987) and

autogenic training (Stetter & Kupper, 2002), while laying or sitting on mats. During the first six weeks, the short form of progressive muscle relaxation according to the manual of (Bernstein, Borkovec, & Ullmann, 1975) was instructed. Single muscle groups (hands, arms, shoulder, back, legs, feet and face) were tensed for 5-7 sec, followed by 45 sec of relaxation, with two repetitions each per session. After the first six weeks, autogenic training was introduced, whereby participants were instructed to concentrate on the breathing rhythm as well as on heartbeat. Participants were asked to imagine body parts as warm or cold, or as

10

ACCEPTED MANUSCRIPT

heavy. Imaginations were repeated in blocks with 3-4 repetitions per session. During week 11 and 12, both relaxation methods were combined.

2.3 Balance assessment

RI PT

Dynamic balance was assessed with a stability platform with digital control (Stability Platform, Modell 16030, Lafayette Instrument, USA). The testing protocol required

participants to stand barefoot on an unstable platform with a maximum deviation of ± 15° to

SC

each side of its horizontal alignment. Participants were asked to place the hands on their hips, direct their gaze to a fixation cross straight-ahead at approx. 3 m distance, and to keep the

M AN U

platform horizontally as long as possible during a trial without leaving the testing position. A handrail was available to prevent falls. After a 60 sec practice-trial, six trials with a length of 30 sec each were run. Half of the trials were conducted with eyes open and eyes closed, respectively. Between trials, participants rested for 30 sec. The beginning and ending of the

TE D

trials were automatically signalized by a short tone. During the eyes closed conditions, participants were asked to close their eyes, but to keep the head straight-ahead. The condition to start with (eyes open / eyes closed) was counterbalanced across participants. Participants

EP

were able to rest for 2 min between conditions to prevent fatigue. The testing procedure took 15 min on average. Whenever a participant left the testing position (i.e. lifted the hands from

AC C

his/her hips, opened the eyes in eyes-closed conditions or grasped the handrail), the trial was restarted. If three unsuccessful attempts within a condition occurred, the respective test condition would have been marked as missing data. No missing data occurred in the present sample. A built-in digital encoder recorded the time (in sec) per trial the platform was in the horizontal position (±3 degrees deviation tolerance). The horizontal position was automatically referenced to 0 degrees when the platform was switched on before every testing session. For the final test score, the average over six trials (three with eyes open, three with eyes closed) was calculated.

11

ACCEPTED MANUSCRIPT 2.4 MRI acquisition and preprocessing Scanning was performed on a 3 Tesla MR system (Magnetom Trio, Siemens, Germany) using a padded standard head coil. Participants were told to lie as motionless as possible during the scan duration of 8 min. T1-weighted images were acquired using a

RI PT

magnetization-prepared rapid gradient-echo (MPRAGE) sequence (TR = 2300 ms, TE = 2.98 ms, flip angle = 9°, FOV = 256 x 256, 240 coronal slices, voxel size = 1 mm3). Cortical thickness

SC

2.4.1

Processing of the images and surface-based morphometry were performed using the

M AN U

neuroimaging package FreeSurfer, version 6.0, http://surfer.nmr.mgh.harvard.edu/). The procedure of the automated image processing has been described elsewhere (Fischl & Dale, 2000). Briefly, the image reconstruction includes removal of non-brain tissue (Ségonne et al., 2004), Talairach transformation, volumetric segmentation of subcortical white and gray

TE D

matter structures (Fischl et al., 2002), intensity normalization, tessellation of white and gray matter boundaries, and topology correction (Fischl, Liu, & Dale, 2001). Further, surface inflation and spherical atlas registration using individual folding patterns to match cortical

EP

geometry (Fischl, Sereno, & Dale, 1999; Fischl, Sereno, Tootell, & Dale, 1999), and gyral based cortical parcellation (Desikan et al., 2006) were performed. With this method, cortical

AC C

thickness is calculated as the closest distance between the gray/white matter boundary and the gray/pial boundary at each vertex on the tessellated surface (Fischl & Dale, 2000). To reduce within-subject noise, an unbiased, robust within-subject template (Reuter &

Fischl, 2011) was created between the two time points of each participant, using the longitudinal stream of FreeSurfer (Reuter, Schmansky, Rosas, & Fischl, 2012). The preprocessing steps including motion correction, intensity normalization, Talairach registration, subcortical parcellation, surface reconstruction, cortical atlas registration and parcellations were performed based on the within-subject template information (Reuter et al.,

12

ACCEPTED MANUSCRIPT

2012; Reuter, Rosas, & Fischl, 2010). All images were visually inspected for motion blurring and correct segmentation of the white and gray matter, and manually corrected in the event of tissue segmentation errors. The cortical thickness maps were smoothed with a 10 mm full-

2.4.2

Subcortical gray matter volume

RI PT

width half-maximum Gaussian kernel.

The hippocampus and basal ganglia were selected as regions of interest (ROIs) for subcortical gray matter volume analyses as their involvement in exercise-induced plasticity,

SC

motor coordination training, and balance has been reported previously (Boisgontier et al.,

M AN U

2017; Erickson, Leckie, & Weinstein, 2014; Hänggi et al., 2010; Magon et al., 2016; Niemann, Godde, Staudinger et al., 2014; Niemann, Godde, & Voelcker-Rehage, 2014; Taubert et al., 2010; van Praag, 2009). The hippocampus was automatically parcellated into subfields by FreeSurfer, based on a statistical atlas built upon ultra-high resolution ex vivo MRI data (Iglesias et al., 2015b). The basal ganglia comprised putamen, globus pallidus and

TE D

caudate nucleus and were automatically labeled and parcellated by FreeSurfer, resulting in a mean gray matter volume for each structure, separately for the left and right hemisphere. A

EP

longitudinal processing stream based on subject-specific Bayesian atlases provided by FreeSurfer was used to increase power and robustness (Iglesias et al., 2015a; Iglesias et al.,

AC C

2016).

2.5 Statistical analyses 2.5.1

Balance performance

Data on balance performance were analyzed using R, version 3.5.0 (R Core Team, 2017). Intervention effects were compared by means of ANCOVA (Egbewale, Lewis, & Sim, 2014; Liu, Lebeau, & Tenenbaum, 2016): In the linear model, posttest scores were compared between groups by submitting pretest scores, group, and age as predictors. Age has been included as a covariate in the model as motor control and balance abilities were shown to be

13

ACCEPTED MANUSCRIPT affected by age (Seidler et al., 2010). The results are displayed as the pretest- and age-

adjusted group difference at posttest, along with the corresponding 95% confidence intervals (CI) and Cohens d as standardized effect size. The α level was set to p < 0.05, two-tailed. The

2.5.2

RI PT

test-retest correlation coefficient for the balance score was r = .62, p < .001. Brain imaging data

Cortical thickness data were analyzed in MATLAB (release 2016b, The MathWorks, Inc., USA) using linear mixed effects models (LME) for longitudinal mass-univariate data

SC

with functions provided by FreeSurfer (Bernal-Rusiel, Greve, Reuter, Fischl, & Sabuncu,

M AN U

2013). Equations included time (pretest and posttest), group, and the interaction between time and group as fixed effects, and age (at baseline, centered to the mean) as covariate of no interest, to adjust for age-related differences in cortical thickness (Barnes et al., 2010; Sowell et al., 2003). The intercept was included as random effect.

TE D

The following random intercept model was applied to the data, were Yij is the thickness value of each vertex for the ith participant on the jth time point:

EP

Yij = β0 + β1 timeij + β2 groupi + β3 (timeij × groupi) + β4 age + b0i + εij In the equation, β denotes fixed-effect estimates, b denotes subject-specific random-

AC C

effects estimates, and ε is the residual error. We chose a whole-brain approach for the cortical thickness analysis. To control for multiple comparisons, a two-stage adaptive false discovery rate (FDR) procedure with an array of q-values was used (Benjamini, Krieger, & Yekutieli, 2006). The results were thresholded at a corrected p-value of p < 0.05, two-sided. We further analyzed the data at an explorative threshold of p < .01, uncorrected, together with a minimum cluster extent threshold of k = 50 mm2 (Landin-Romero et al., 2017; Lieberman & Cunningham, 2009). Significance maps for visualization were thresholded at p < .01,

14

ACCEPTED MANUSCRIPT uncorrected, and overlaid on a Freesurfer standard brain. Coordinates are reported in MNI space. The analysis of changes in hippocampal volume focused on the hippocampus proper

RI PT

and the dentate gyrus. To increase the reliability of the measurements, mean volumes of the CA subfields (CA1-4) and the dentate gyrus were added up into one hippocampal volume of interest, separately for the left and right hemisphere. Gray matter volumes of the basal ganglia were analyzed separately for putamen, pallidum and caudate nucleus. The extracted gray

SC

matter mean volumes for each ROI were analyzed in MATLAB (release 2016b, The

M AN U

MathWorks, Inc., USA), using univariate LME functions for longitudinal data provided by FreeSurfer (Bernal-Rusiel et al., 2013). The fixed effects in the models included time, group, and the interaction between time and group. Age (at baseline, centered to the mean) and estimated total intracranial volume were submitted as covariates of no interest; the latter to adjust for head size in volume analyses (Barnes et al., 2010; Malone et al., 2015). The

TE D

intercept was included as subject specific random effect in the model. The p-values of the eight separate subcortical ROI analyses were adjusted with Bonferroni-correction for multiple

EP

comparisons using the stats package within R. Uncorrected and Bonferroni-corrected p-values are reported in the results section.

AC C

The test-retest correlation coefficients ranged from r = .82 to r = .96 for the cortical thickness clusters and from r = .96 to r = .99 for the subcortical ROIs, all ps < .001. In order to explore the relationship between cortical thickness changes and changes in

balance performance, individual mean cortical thickness values of the significant clusters at p < .01, cluster size k > 50 mm2 were extracted. The changes of cortical thickness per cluster and balance performance, respectively, were calculated by subtracting the pretest from the posttest values.

15

ACCEPTED MANUSCRIPT To assess a relationship between cortical thickness change and balance performance on a whole-brain level, a difference map between the cortical thickness at pre- and posttest per participant was calculated and used as input for multiple regression against changes in

AC C

EP

TE D

M AN U

SC

RI PT

balance performance.

16

ACCEPTED MANUSCRIPT

3 Results The groups (balance group: n = 19, relaxation group: n = 18) did not differ with respect to participants’ age, sex, number of training sessions, and balance performance before

RI PT

training (all ps > .300, see Table 1).

Balance group (n=19)

Relaxation group (n=18)

p

Age

43.9 (14.92)

46.11 (15.37)

.67a

Sex (female/male)

12/7

SC

Table 1. Characteristics of the participants (Mean, SD)

.89b

Training sessions

20.63 (2.67)

21.00 (2.74)

.68a

Balance performance at

5.29 (1.71)

5.89 (1.84)

.30a

M AN U

11/7

pretest (sec)

TE D

Note. a = independent t-test. b = Chi-square test.

EP

3.1 Balance performance

The analysis of balance performance changes yielded a significant effect of group, F (1, 33) =

AC C

7.78, p = .008. After training, the adjusted mean of the balance group was higher than the mean of the relaxation group (group difference = 1.58, 95 % CI = [0.43, 2.74], d = 0.66, see Figure 1).

17

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

3.2 Imaging results 3.2.1

TE D

Figure 1. Balance performance on the stability platform for the balance (yellow bars) and the relaxation group (blue bars) before and after the training. Error bars indicate standard deviations. Circles depict individual participant data.

Cortical Thickness

EP

The whole-brain analysis of the Time x Group interaction revealed significant higher

AC C

cortical thickness increases in the balance group, compared to the relaxation group, in a number of brain regions: the superior temporal gyrus extending into the circular insular sulcus, the superior transverse occipital sulcus, the superior frontal sulcus of the left hemisphere, and the posterior cingulate of the right hemisphere (p < .05, FDR-corrected, see Table 2 and Figure 2). At a more liberal threshold of p < .01, uncorrected, with a cluster extension threshold of k > 50 mm2, additional larger cortical thickness increases in the balance group compared to the relaxation group were observed in the precentral gyrus bilaterally and in the right

18

ACCEPTED MANUSCRIPT

pericalcarine gyrus (see Table 2 and Figure 2). There were not any larger cortical thickness increases in the relaxation compared to the balance group, irrespectively of the applied thresholds. There were no significant main effects of Time and Group at p < .05 (FDR-

RI PT

corrected). Table 2. Cortical thickness and subcortical volume changes for the Time x Group interaction

Superior temporal gyrus / circular insular sulcus

-40 -25 5

9.70

Superior transverse occipital sulcus Superior frontal sulcus Posterior cingulate gyrus

-19 -86 17

6.94

-20 26 42 8 -43 22

2.19 4.54

75.38 98.79 73.77

Region

LH

LH LH RH

number of vertices

β

F

p

0.09

15.43

< 0.001

11

0.09

14.83

< 0.001

4 18

0.06 0.07

14.81 16.14

< 0.001 < 0.001

SC

mm2

Hem.

22

M AN U

X, Y, Z

TE D

Cortical thickness cluster (p < .05, FDR corrected)

Cortical thickness cluster (p < .01, k > 50 mm2*) Precentral gyrus Pericalcarine gyrus Precentral gyrus

-19 -24 64 15 -77 10 33 -5 46

EP

LH RH RH

155 160 145

0.14 0.10 0.08

8.25 9.07 8.33

0.004 0.002 0.005

Subcortical volume in mm3 (p < .05)

AC C

0.019 LH Putamen -61.27 5.99 0.025 RH Putamen -48.64 5.44 Note. LH = left hemisphere, RH = right hemisphere. * = only additional significant clusters are presented. Coordinates are in MNI space.

19

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

Subcortical gray matter volume

EP

3.2.2

TE D

Figure 2. MRI results. Cortical thickness significance maps of the Time x Group interaction, depicting higher cortical thickness increases for the balance group, compared to the relaxation group. Maps are superimposed on the standard FreeSurfer brain and thresholded at p < .01, uncorrected. Color scale indicates -log 10 p-values. LH = left hemisphere, RH = right hemisphere. Top: lateral view, bottom: medial view.

Hippocampus. Changes in hippocampal gray matter volume did not differ between

AC C

groups; there was no Time x Group interaction for neither the left (F(1, 35) = .11, p = .743, d = 0.11) nor for the right hemisphere (F(1, 35) = 1.52, p = .226, d = 0.41). There were no main effects of Time (all ps > .146) and Group (all ps > .082). Basal ganglia. Analyses of Time x Group interactions yielded a significant decrease

of gray matter volume in the putamen bilaterally for the balance group compared to the relaxation group (left: F(1, 35) = 5.99, p = .019, d = 0.66, right: F(1, 35) = 5.44, p = .025, d = 0.59), but no Time x Group interaction for the pallidum and caudate nucleus (all ps > .243). The significant decrease of the putamen, however, did not survive Bonferroni-correction

20

ACCEPTED MANUSCRIPT

(correcting for the number of subcortical ROIs, all ps > .13). Analyses of the substructures of the basal ganglia revealed no significant main effects of Time (all ps >. 083). There was a significant main effect of Group in the left pallidum (F (1, 35) = 4.54, p = .040, d = 0.66) with a smaller volume in the balance group, compared to the relaxation group. This group

RI PT

difference, however, did not survive Bonferroni-correction (p = .322). There was no main effect of Group for the putamen and the caudate nucleus (all ps < .398). Correlations with balance performance

SC

3.2.3

3.2.3.1 Cortical thickness

M AN U

Improvements in balance performance significantly correlated with the cortical thickness changes in the left and right precentral gyrus across both groups (left: r = .58, p < .001; right: r = 39, p = .016, see Figure 3, left panel), indicating that a larger gain in balance performance was associated with a larger increase in precentral cortical thickness. Correlation

TE D

analyses separately run for the training groups revealed significant positive correlations for the balance group in the left precentral gyrus (left: r = .66, p = .002; right: r = .19, p = .426), but no significant associations for the relaxation group (left: r = .19. p = .437, right: r = .28, p

EP

= .270). There were no additional significant correlations between balance performance and

AC C

cortical thickness increase within the extracted clusters. Moreover, a whole-brain regression of cortical thickness changes against behavioral

change confirmed the positive correlation between the left precentral cortical thickness increase and the improved balance performance across both groups (see Table 3). The same relationship between changes in balance performance and cortical thickness was obtained for a second more inferior precentral cluster and two occipital clusters. However, none of the effects survived FDR-correction and are therefore reported at an uncorrected threshold of p < .001 with a cluster extent threshold of k > 30mm2.

21

ACCEPTED MANUSCRIPT

Table 3. Correlations between cortical thickness change and balance performance change across both groups (whole-brain analysis) Hem.

Region

X, Y, Z

mm2

Number

ß

F

p

LH

Precentral gyrus

-42 -10 58

69

167

LH

Pericalcarine sulcus

-20 -73 2

65

120

LH

Precentral gyrus

-20 -24 64

36

75

RH

Lateraloccipital sulcus

23 -92 19

35

45

RI PT

of vertices .07

16.98

< .001

.03

14.60

< .001

.05

14.57

< .001

.03

14.54

< .001

M AN U

SC

Note. LH = left hemisphere, RH = right hemisphere. Cluster at p < .001, uncorrected, and k > 30mm2. Coordinates are in MNI space.

3.2.3.2 Subcortical gray matter volume

The volume decrease of the left putamen was significantly correlated with the balance performance gain (r = - .48, p = .002, see Figure 3, right panel), that is, participants showing

TE D

a larger increase in balance performance showed a larger decrease in gray matter volume in the left putamen. The negative relationship remained significant when only the balance group

AC C

EP

was considered (r = - .47, p = .043).

22

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Figure 3. Correlation between pre-posttraining changes of the left precentral cortical thickness (left panel) and the left putamen volume (right panel) with pre-posttraining balance performance change.

23

ACCEPTED MANUSCRIPT

4 Discussion The goal of the present study was to identify the influence of balance training on cortical and subcortical gray matter structures. To this end, healthy participants were

RI PT

randomly assigned to either a balance training or a relaxation training. Both groups exercised twice a week for 12 weeks. Balance performance improved only in the balance group from pre- to posttest. At posttest, the balance group showed a larger increase in cortical thickness

SC

in the left superior temporal gyrus, superior transverse occipital, superior frontal sulci, and in the right posterior cingulate gyrus compared to the relaxation group. At a more liberal

M AN U

threshold, similar changes were observed for the left and right precentral gyrus and the right pericalcarine sulcus. By contrast, the volume of the putamen decreased in the balance group. Benefits in balance performance correlated with gray matter increases in the left precentral gyrus as well as with volume decreases in the left putamen. We did not find evidence for

TE D

specific structural changes of the hippocampus associated with balance training.

4.1 Cortical thickness

EP

Balance tasks require the integration of vestibular, visual, and proprioceptive selfmotion signals (Cullen, 2012). In line with this challenge profile, we observed training-

AC C

induced changes particularly in areas involved in the processing of self-motion. For instance, a larger cortical thickness increase in the balance group compared to the relaxation group was found in the left superior temporal gyrus extending into the circular insular sulcus. Though it is still a matter of debate whether a specific vestibular cortex exists and where it might be located (Eulenburg, Caspers, Roski, & Eickhoff, 2012), recent studies have localized the processing of vestibular signals predominantly in the parieto-insular vestibular cortex and in the posterior insular cortex in the Sylvian fissure (Frank, Wirth, & Greenlee, 2016; Lopez, Blanke, & Mast, 2012; Schlindwein et al., 2008; Wirth, Frank, Greenlee, & Beer, 2018).

24

ACCEPTED MANUSCRIPT

Electrical stimulation of the superior temporal gyri in epilepsy patients evoked illusory body motion including perceived head tilts (Kahane, Hoffmann, Minotti, & Berthoz, 2003), indicating a contribution of the superior temporal gyrus to self-motion perception, a function

RI PT

specifically targeted with balance tasks. Larger gray matter changes for the balance group were additionally observed in visual association areas in the superior transverse occipital sulcus. This region overlaps with area V3A, a motion-selective region involved in the processing of optic flow. It has been

SC

demonstrated that area V3A reliably responses to objective planar motion, suggesting an

M AN U

essential contribution to visual self-motion perception (Fischer, Bülthoff, Logothetis, & Bartels, 2012; Strong, Silson, Gouws, Morland, & McKeefry, 2017). Additionally, area V3A receives vestibular input, indicating an early visual-vestibular interaction (Roberts et al., 2017). The present findings are in line with results in professional ballet dancers and slackliners showing larger gray matter volumes in visual association areas compared to non-

TE D

balance experts (Hüfner et al., 2011). While it was not possible to unequivocally associate balance experience and cortical gray matter changes due to the cross-sectional nature of this

EP

study, the present experiment provides strong evidence for a causal relation between balance exercise and changes in visual cortex. Since the implemented balance exercise tasks required

AC C

planar motion detection to maintain an upright posture, the thickness increase of area V3A is likely related to visual-vestibular motion processing during the balance training. Similar group differences were observed for the right posterior cingulate gyrus, a brain

region proposed to be an anatomical interposition of the parieto-medial temporal pathway to the hippocampus (Kravitz, Saleem, Baker, & Mishkin, 2011). While receiving input from almost all sensory modalities, including the vestibular system (Hitier et al., 2014; Wirth et al., 2018), the posterior cingulate has been reported to be involved in the processing of depth motion cues (Kovács, Cziraki, & Greenlee, 2010) and the perception of self-location

25

ACCEPTED MANUSCRIPT (Guterstam, Björnsdotter, Gentile, & Ehrsson, 2015). Noteworthy, the observed cluster

coordinates are located at the borders of the retrosplenial cortex, being reciprocally connected with the subiculum, the parahippocampus and the enthorhinal cortex (Vann, Aggleton, & Maguire, 2009). The retrosplenial cortex has been found to be functionally connected with

RI PT

area V3A during path integration (Sherrill et al., 2015). The posterior cingulate gyrus seems to contribute to the integration of egocentric and allocentric spatial representations and spatial navigation (Kravitz et al., 2011; Maguire, 2001). Additionally, the retrosplenial cortex has

SC

been found to be involved in verbal and visual memory (Maguire, 2001; Vann et al., 2009). As memory and spatial cognition have been shown to benefit from physical exercise, in

M AN U

particular from balance trainings (Dordevic, Hokelmann, Muller, Rehfeld, & Muller, 2017; Hüfner et al., 2011; Rogge et al., 2017), it might be speculated that the posterior cingulate gyrus contributes to the beneficial effects of exercise trainings on these cognitive processes. The balance training induced a greater cortical thickness increase in the left superior

TE D

frontal sulcus. This area overlaps with those reported to have lower gray matter volumes in professional ballet dancers as compared to a control group with no balance training

EP

experience (Hänggi et al., 2010). The opposite effects might be related to the different study designs: while we investigated gray matter changes following short-term balance training,

AC C

Hänggi et al. (2010) analyzed long-term effects of ballet expertise in a cross-sectional design. It is known that morphological adaptations after motor trainings depend on the stage of practicing (Wenger, Brozzoli, Lindenberger, & Lövdén, 2017). For instance, it has recently been demonstrated that tissue changes followed an inverse-quadratic shape during motor learning: an initial expansion of precentral gray matter partly renormalized to its baseline value despite continued practice and further increase in task proficiency (Taubert et al., 2010; Wenger, Kühn et al., 2017). These authors interpreted this pattern as an initial overproduction of synapses followed by a selective stabilization of relevant new connections. In the present study, exercising difficulty was progressively increased in order to promote continuous motor

26

ACCEPTED MANUSCRIPT learning. Thus, we speculate that the phase of selective stabilization was not yet reached at posttest. However, balance training studies with repeated measurements are necessary to prove this hypothesis.

RI PT

Cortical thickness increased in the precentral gyri bilaterally after the balance training. However, the effects were significant only at an explorative threshold. The left precentral thickness change correlated significantly with the improved balance performance, providing evidence for an association with the balance training. This result is in line with previous

SC

exercise studies reporting balance training-related gray matter increase in the primary motor

M AN U

cortex (Taubert et al., 2010; Taubert, Mehnert, Pleger, & Villringer, 2016). Changes in cortical thickness after the balance training were primarily left lateralized. A left hemispheric dominance for motor and coordination skills has been reported in recent training studies using balance (Taubert et al., 2010; Taubert et al., 2016) or dancing exercises

TE D

(Müller et al., 2017) and might be related to hemispheric specialization and handedness (Serrien, Ivry, & Swinnen, 2006).

EP

4.2 Subcortical gray matter volume

In contrast to the cortical results, the balance training induced a bilateral gray matter

AC C

decrease in the putamen. The volume reduction of the left putamen significantly correlated with the benefit in balance performance, suggesting a link between these structural changes and balance performance. However, the change in putamen volume did not survive Bonferroni-correction for multiple comparisons. Nevertheless, this result is in line with reports of a previous balance training study (Taubert et al., 2010). Moreover, cross-sectional findings in balance experts revealed smaller putamen volumes compared to controls (Hänggi et al., 2010), and lower volume of basal ganglia predicted better postural performance in young and older adults (Boisgontier et al., 2017).

27

ACCEPTED MANUSCRIPT In contrast to previous exercise studies in humans and rodents, we did not find a significant group difference for gray matter volume changes of the hippocampus. Such

hippocampal structural increases have, however, mostly been reported after aerobic exercise training (Erickson et al., 2011; Thomas et al., 2016) and related increases of cardiorespiratory

RI PT

fitness (Erickson et al., 2011; Kleemeyer et al., 2015; Maass et al., 2015). Motor coordination and dance interventions reporting changes in hippocampal volume lasted for 12 to 18 months (Niemann, Godde, & Voelcker-Rehage, 2014; Rehfeld et al., 2017). It might be speculated

SC

that hippocampal volume changes follow balance training too, provided they last for longer durations than implemented in the present study. Finally, we might speculate that dancing and

M AN U

coordination training involves higher memory demands than the balance tasks used in the present study, such as choreographies participants had to memorize accurately. This difference might have contributed to the beneficial effects of motor coordiantion and dancing on the hippocampus reported in previous studies (Niemann, Godde, & Voelcker-Rehage,

4.3 Limitations

TE D

2014; Rehfeld et al., 2017).

EP

Some limitations of the study need to be mentioned. The effects in the precentral gyrus and the putamen did not survive correction for multiple comparisons and should be interpreted

AC C

with caution. Physical exercise intervention studies are time consuming; the present findings should nevertheless be replicated with larger samples. Moreover, to trace the time course of structural changes, multiple assessments during the intervention and a follow-up after several months would be necessary. In addition, the MRI scanning was performed with a standard T1 sequence resulting in a voxel size of 1mm3. A higher resolution and complementing T2-weighted images would allow a more precise segmentation of hippocampal subfields (Iglesias et al., 2015a; Wisse et al., 2017) which in turn would have increased the sensitivity for detecting hippocampal

28

ACCEPTED MANUSCRIPT

volume changes. Rodent studies have reported that stimulating the vestibular system seems to modulate hippocampal theta activity and the firing pattern of hippocampal place cells (Aitken, Zheng, & Smith, 2018). Thus, functional brainimaging studies might have a higher sensitivity

RI PT

to detect first changes in the medial temporal lobe than structural imaging studies.

4.4 Conclusion

The present results suggest that 12 weeks of balance training is capable of inducing

SC

structural plasticity in the superior temporal cortex, in the visual association cortex, in the posterior cingulate cortex and in the superior frontal sulcus. These regions are known to be

M AN U

involved in visual-vestibular self-motion processing and in the integration of spatial information from different sensory signals. Moreover, these areas contribute to higher cognitive functions such as memory and spatial cognition. Thus, physical exercise-induced neuroplasticity in these regions might mediate not only the increase in balance performance,

AC C

EP

TE D

but also the beneficial effects of physical exercise on cognition.

29

ACCEPTED MANUSCRIPT

Acknowledgements This research was supported by a grant from the European Comission [ABBI, 611452, FP7-ICT-2013-10] to Brigitte Röder, and a grant from the German Research Foundation

RI PT

[DFG Ro 2625/10-1] to Brigitte Röder. We thank Volker Nagel for contributing to the training conception, Gudrun Nagel for carrying out the training programs, and Klaus-Michael Braumann and Karsten Hollander of the Institute for Sports Medicine at the University of

M AN U

Conflict of Interest Statement

SC

Hamburg for conducting the medical examinations.

AC C

EP

TE D

The authors declare no competing financial interests.

30

ACCEPTED MANUSCRIPT

References Aitken, P., Zheng, Y., & Smith, P. F. (2018). The modulation of hippocampal theta rhythm by the vestibular system. Journal of Neurophysiology, 119(2), 548–562. https://doi.org/10.1152/jn.00548.2017.

RI PT

Barnes, J., Ridgway, G. R., Bartlett, J., Henley, S. M. D., Lehmann, M., Hobbs, N.,. . . Fox, N. C. (2010). Head size, age and gender adjustment in MRI studies: A necessary nuisance? NeuroImage, 53(4), 1244–1255. https://doi.org/10.1016/j.neuroimage.2010.06.025

SC

Benjamini, Y., Krieger, A. M., & Yekutieli, D. (2006). Adaptive linear step-up procedures that control the false discovery rate. Biometrika, 93(3), 491–507. https://doi.org/10.1093/biomet/93.3.491

M AN U

Bernal-Rusiel, J. L., Greve, D. N., Reuter, M., Fischl, B., & Sabuncu, M. R. (2013). Statistical analysis of longitudinal neuroimage data with Linear Mixed Effects models. NeuroImage, 66, 249–260. https://doi.org/10.1016/j.neuroimage.2012.10.065 Bernstein, D.A., Borkovec, T.D., & Ullmann, L.P. (1975). Progressive relaxation training: A manual for the helping professions (3rd pr). Champaign: Research Press. Bigelow, R. T., & Agrawal, Y. (2015). Vestibular involvement in cognition: Visuospatial ability, attention, executive function, and memory. Journal of vestibular research: equilibrium & orientation, 25(2), 73–89. https://doi.org/10.3233/VES-150544

TE D

Boisgontier, M. P., Cheval, B., Chalavi, S., van Ruitenbeek, P., Leunissen, I., Levin, O.,. . . Swinnen, S. P. (2017). Individual differences in brainstem and basal ganglia structure predict postural control and balance loss in young and older adults. Neurobiology of Aging, 50, 47–59. https://doi.org/10.1016/j.neurobiolaging.2016.10.024

EP

Cassilhas, R. C., Tufik, S., & Mello, M. T. de. (2016). Physical exercise, neuroplasticity, spatial learning and memory. Cellular and molecular life sciences : CMLS, 73(5), 975– 983. https://doi.org/10.1007/s00018-015-2102-0

AC C

Colcombe, S. J., Erickson, K. I., Scalf, P. E., Kim, J. S., Prakash, R., McAuley, E.,. . . Kramer, A. F. (2006). Aerobic exercise training increases brain volume in aging humans. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 61(11), 1166–1170. Cullen, K. E. (2012). The vestibular system: Multimodal integration and encoding of selfmotion for motor control. Trends in Neurosciences, 35(3), 185–196. https://doi.org/10.1016/j.tins.2011.12.001 Curlik, D. M., Maeng, L. Y., Agarwal, P. R., & Shors, T. J. (2013). Physical skill training increases the number of surviving new cells in the adult hippocampus. PLoS ONE, 8(2), e55850. https://doi.org/10.1371/journal.pone.0055850 Desikan, R. S., Ségonne, F., Fischl, B., Quinn, B. T., Dickerson, B. C., Blacker, D.,. . . Killiany, R. J. (2006). An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage, 31(3), 968–980. https://doi.org/10.1016/j.neuroimage.2006.01.021

31

ACCEPTED MANUSCRIPT

Dordevic, M., Hokelmann, A., Muller, P., Rehfeld, K., & Muller, N. G. (2017). Improvements in Orientation and Balancing Abilities in Response to One Month of Intensive SlacklineTraining. A Randomized Controlled Feasibility Study. Fronties in Human Neuroscience, 11, 55. https://doi.org/10.3389/fnhum.2017.00055

RI PT

Egbewale, B. E., Lewis, M., & Sim, J. (2014). Bias, precision and statistical power of analysis of covariance in the analysis of randomized trials with baseline imbalance: A simulation study. BMC medical research methodology, 14, 49. https://doi.org/10.1186/1471-2288-1449 Erickson, K. I., Voss, M. W., Prakash, R. S., Basak, C., Szabo, A., Chaddock, L.,. . . Kramer, A. F. (2011). Exercise training increases size of hippocampus and improves memory. Proceedings of the National Academy of Sciences, 108(7), 3017–3022. https://doi.org/10.1073/pnas.1015950108

SC

Erickson, K. I., Leckie, R. L., & Weinstein, A. M. (2014). Physical activity, fitness, and gray matter volume. Neurobiology of Aging, 35 Suppl 2, S20-8. https://doi.org/10.1016/j.neurobiolaging.2014.03.034

M AN U

Eulenburg, P. zu, Caspers, S., Roski, C., & Eickhoff, S. B. (2012). Meta-analytical definition and functional connectivity of the human vestibular cortex. NeuroImage, 60(1), 162–169. https://doi.org/10.1016/j.neuroimage.2011.12.032 Firth, J., Stubbs, B., Vancampfort, D., Schuch, F., Lagopoulos, J., Rosenbaum, S., & Ward, P. B. (2017). Effect of aerobic exercise on hippocampal volume in humans: A systematic review and meta-analysis. NeuroImage, 166, 230–238. https://doi.org/10.1016/j.neuroimage.2017.11.007

TE D

Fischer, E., Bülthoff, H. H., Logothetis, N. K., & Bartels, A. (2012). Human areas V3A and V6 compensate for self-induced planar visual motion. Neuron, 73(6), 1228–1240. https://doi.org/10.1016/j.neuron.2012.01.022

EP

Fischl, B., & Dale, A. M. (2000). Measuring the thickness of the human cerebral cortex from magnetic resonance images. Proceedings of the National Academy of Sciences, 97(20), 11050–11055. https://doi.org/10.1073/pnas.200033797

AC C

Fischl, B., Liu, A., & Dale, A. M. (2001). Automated manifold surgery: Constructing geometrically accurate and topologically correct models of the human cerebral cortex. IEEE transactions on medical imaging, 20(1), 70–80. https://doi.org/10.1109/42.906426 Fischl, B., Sereno, M. I., & Dale, A. M. (1999). Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system. NeuroImage, 9(2), 195–207. https://doi.org/10.1006/nimg.1998.0396 Fischl, B., Salat, D. H., Busa, E., Albert, M., Dieterich, M., Haselgrove, C.,. . . Dale, A. M. (2002). Whole Brain Segmentation. Neuron, 33(3), 341–355. https://doi.org/10.1016/S0896-6273(02)00569-X Fischl, B., Sereno, M. I., Tootell, R. B.H., & Dale, A. M. (1999). High-resolution intersubject averaging and a coordinate system for the cortical surface. Human brain mapping, 8(4), 272–284. https://doi.org/10.1002/(SICI)1097-0193(1999)8:4<272::AIDHBM10>3.0.CO;2-4

32

ACCEPTED MANUSCRIPT Frank, S. M., Wirth, A. M., & Greenlee, M. W. (2016). Visual-vestibular processing in the human Sylvian fissure. Journal of Neurophysiology, 116(2), 263–271. https://doi.org/10.1152/jn.00009.2016

Gebel, A., Lesinski, M., Behm, D. G., & Granacher, U. (2018). Effects and Dose-Response Relationship of Balance Training on Balance Performance in Youth: A Systematic Review and Meta-Analysis. Sports medicine (Auckland, N.Z.). Advance online publication. https://doi.org/10.1007/s40279-018-0926-0

RI PT

Guterstam, A., Björnsdotter, M., Gentile, G., & Ehrsson, H. H. (2015). Posterior cingulate cortex integrates the senses of self-location and body ownership. Current biology : CB, 25(11), 1416–1425. https://doi.org/10.1016/j.cub.2015.03.059

SC

Hänggi, J., Koeneke, S., Bezzola, L., & Jäncke, L. (2010). Structural neuroplasticity in the sensorimotor network of professional female ballet dancers. Human brain mapping, 31(8), 1196–1206. https://doi.org/10.1002/hbm.20928 Hitier, M., Besnard, S., & Smith, P. F. (2014). Vestibular pathways involved in cognition. Frontiers in integrative neuroscience, 8, 59. https://doi.org/10.3389/fnint.2014.00059

M AN U

Horak, F. B. (1997). Clinical assessment of balance disorders. Gait & posture, 6(1), 76–84. https://doi.org/10.1016/S0966-6362(97)00018-0 Hötting, K., & Röder, B. (2013). Beneficial effects of physical exercise on neuroplasticity and cognition. Neuroscience and biobehavioral reviews. Advance online publication. https://doi.org/10.1016/j.neubiorev.2013.04.005

TE D

Hrysomallis, C. (2011). Balance ability and athletic performance. Sports medicine (Auckland, N.Z.), 41(3), 221–232. https://doi.org/10.2165/11538560-000000000-00000 Hüfner, K., Binetti, C., Hamilton, D. A., Stephan, T., Flanagin, V. L., Linn, J.,. . . Brandt, T. (2011). Structural and functional plasticity of the hippocampal formation in professional dancers and slackliners. Hippocampus, 21(8), 855–865. https://doi.org/10.1002/hipo.20801

EP

Iglesias, J. E., Augustinack, J. C., Nguyen, K., Player, C. M., Player, A., Wright, M.,. . . van Leemput, K. (2015a). A computational atlas of the hippocampal formation using ex vivo, ultra-high resolution MRI: Application to adaptive segmentation of in vivo MRI. NeuroImage, 115, 117–137. https://doi.org/10.1016/j.neuroimage.2015.04.042

AC C

Iglesias, J. E., Augustinack, J. C., Nguyen, K., Player, C. M., Player, A., Wright, M.,. . . van Leemput, K. (2015b). A computational atlas of the hippocampal formation using ex vivo, ultra-high resolution MRI: Application to adaptive segmentation of in vivo MRI. NeuroImage, 115, 117–137. https://doi.org/10.1016/j.neuroimage.2015.04.042 Iglesias, J. E., van Leemput, K., Augustinack, J., Insausti, R., Fischl, B., & Reuter, M. (2016). Bayesian longitudinal segmentation of hippocampal substructures in brain MRI using subject-specific atlases. NeuroImage, 141, 542–555. https://doi.org/10.1016/j.neuroimage.2016.07.020 Jacobson, E. (1987). Progressive Relaxation. The American Journal of Psychology, 100(3/4), 522. https://doi.org/10.2307/1422693 Jonasson, L. S., Nyberg, L., Kramer, A. F., Lundquist, A., Riklund, K., & Boraxbekk, C.-J. (2017). Aerobic Exercise Intervention, Cognitive Performance, and Brain Structure:

33

ACCEPTED MANUSCRIPT Results from the Physical Influences on Brain in Aging (PHIBRA) Study. Frontiers in Aging Neuroscience, 8, CD005381. https://doi.org/10.3389/fnagi.2016.00336

Kahane, P., Hoffmann, D., Minotti, L., & Berthoz, A. (2003). Reappraisal of the human vestibular cortex by cortical electrical stimulation study. Annals of neurology, 54(5), 615– 624. https://doi.org/10.1002/ana.10726

RI PT

Kempermann, G. (2012). New neurons for ’survival of the fittest’. Nature reviews. Neuroscience, 13(10), 727–736. https://doi.org/10.1038/nrn3319 Kleemeyer, M. M., Kühn, S., Prindle, J., Bodammer, N. C., Brechtel, L., Garthe, A.,. . . Lindenberger, U. (2015). Changes in fitness are associated with changes in hippocampal microstructure and hippocampal volume among older adults. NeuroImage. Advance online publication. https://doi.org/10.1016/j.neuroimage.2015.11.026

SC

Kovács, G., Cziraki, C., & Greenlee, M. W. (2010). Neural correlates of stimulus-invariant decisions about motion in depth. NeuroImage, 51(1), 329–335. https://doi.org/10.1016/j.neuroimage.2010.02.011

M AN U

Kravitz, D. J., Saleem, K. S., Baker, C. I., & Mishkin, M. (2011). A new neural framework for visuospatial processing. Nature reviews. Neuroscience, 12(4), 217–230. https://doi.org/10.1038/nrn3008 Landin-Romero, R., Kumfor, F., Leyton, C. E., Irish, M., Hodges, J. R., & Piguet, O. (2017). Disease-specific patterns of cortical and subcortical degeneration in a longitudinal study of Alzheimer's disease and behavioural-variant frontotemporal dementia. NeuroImage, 151, 72–80. https://doi.org/10.1016/j.neuroimage.2016.03.032

TE D

Lesinski, M., Hortobágyi, T., Muehlbauer, T., Gollhofer, A., & Granacher, U. (2015). Doseresponse relationships of balance training in healthy young adults: A systematic review and meta-analysis. Sports medicine (Auckland, N.Z.), 45(4), 557–576. https://doi.org/10.1007/s40279-014-0284-5.

EP

Lieberman, M. D., & Cunningham, W. A. (2009). Type I and Type II error concerns in fMRI research: Re-balancing the scale. Social cognitive and affective neuroscience, 4(4), 423– 428. https://doi.org/10.1093/scan/nsp052

AC C

Liu, S., Lebeau, J.-C., & Tenenbaum, G. (2016). Does Exercise Improve Cognitive Performance? A Conservative Message from Lord's Paradox. Frontiers in psychology, 7, 1092. https://doi.org/10.3389/fpsyg.2016.01092 Lopez, C., Blanke, O., & Mast, F. W. (2012). The human vestibular cortex revealed by coordinate-based activation likelihood estimation meta-analysis. Neuroscience, 212, 159– 179. https://doi.org/10.1016/j.neuroscience.2012.03.028 Maass, A., Duzel, S., Goerke, M., Becke, A., Sobieray, U., Neumann, K.,. . . Duzel, E. (2015). Vascular hippocampal plasticity after aerobic exercise in older adults. Mol Psychiatry, 20(5), 585–593. Retrieved from http://dx.doi.org/10.1038/mp.2014.114 Magon, S., Donath, L., Gaetano, L., Thoeni, A., Radue, E.-W., Faude, O., & Sprenger, T. (2016). Striatal functional connectivity changes following specific balance training in elderly people: MRI results of a randomized controlled pilot study. Gait & posture, 49, 334–339. https://doi.org/10.1016/j.gaitpost.2016.07.016

34

ACCEPTED MANUSCRIPT

Maguire, E. A. (2001). The retrosplenial contribution to human navigation: A review of lesion and neuroimaging findings. Scandinavian Journal of Psychology, 42(3), 225–238. Malone, I. B., Leung, K. K., Clegg, S., Barnes, J., Whitwell, J. L., Ashburner, J.,. . . Ridgway, G. R. (2015). Accurate automatic estimation of total intracranial volume: A nuisance variable with less nuisance. NeuroImage, 104, 366–372. https://doi.org/10.1016/j.neuroimage.2014.09.034

RI PT

Müller, P., Rehfeld, K., Schmicker, M., Hökelmann, A., Dordevic, M., Lessmann, V.,. . . Müller, N. G. (2017). Evolution of Neuroplasticity in Response to Physical Activity in Old Age: The Case for Dancing. Frontiers in Aging Neuroscience, 9, 56. https://doi.org/10.3389/fnagi.2017.00056

SC

Niemann, C., Godde, B., Staudinger, U. M., & Voelcker-Rehage, C. (2014). Exercise-induced changes in basal ganglia volume and cognition in older adults. Neuroscience, 281C, 147– 163. https://doi.org/10.1016/j.neuroscience.2014.09.033

M AN U

Niemann, C., Godde, B., & Voelcker-Rehage, C. (2014). Not only cardiovascular, but also coordinative exercise increases hippocampal volume in older adults. Frontiers in Aging Neuroscience, 6, 170. https://doi.org/10.3389/fnagi.2014.00170 R Core Team. (2017). R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. Retrieved from URL https://www.Rproject.org/

TE D

Rehfeld, K., Müller, P., Aye, N., Schmicker, M., Dordevic, M., Kaufmann, J.,. . . Müller, N. G. (2017). Dancing or Fitness Sport? The Effects of Two Training Programs on Hippocampal Plasticity and Balance Abilities in Healthy Seniors. Fronties in Human Neuroscience, 11, 305. https://doi.org/10.3389/fnhum.2017.00305 Reuter, M., & Fischl, B. (2011). Avoiding asymmetry-induced bias in longitudinal image processing. NeuroImage, 57(1), 19–21. https://doi.org/10.1016/j.neuroimage.2011.02.076

EP

Reuter, M., Rosas, H. D., & Fischl, B. (2010). Highly accurate inverse consistent registration: A robust approach. NeuroImage, 53(4), 1181–1196. https://doi.org/10.1016/j.neuroimage.2010.07.020

AC C

Reuter, M., Schmansky, N. J., Rosas, H. D., & Fischl, B. (2012). Within-subject template estimation for unbiased longitudinal image analysis. NeuroImage, 61(4), 1402–1418. https://doi.org/10.1016/j.neuroimage.2012.02.084 Roberts, R. E., Ahmad, H., Arshad, Q., Patel, M., Dima, D., Leech, R.,. . . Bronstein, A. M. (2017). Functional neuroimaging of visuo-vestibular interaction. Brain structure & function, 222(5), 2329–2343. https://doi.org/10.1007/s00429-016-1344-4 Rogge, A.-K., Röder, B., Zech, A., Nagel, V., Hollander, K., Braumann, K.-M., & Hötting, K. (2017). Balance training improves memory and spatial cognition in healthy adults. Scientific Reports, 7, 572. https://doi.org/10.1038/s41598-017-06071-9 Ruscheweyh, R., Willemer, C., Krüger, K., Duning, T., Warnecke, T., Sommer, J.,. . . Flöel, A. (2011). Physical activity and memory functions: An interventional study. Neurobiology of Aging, 32(7), 1304–1319. https://doi.org/10.1016/j.neurobiolaging.2009.08.001

35

ACCEPTED MANUSCRIPT

Schlindwein, P., Mueller, M., Bauermann, T., Brandt, T., Stoeter, P., & Dieterich, M. (2008). Cortical representation of saccular vestibular stimulation: VEMPs in fMRI. NeuroImage, 39(1), 19–31. https://doi.org/10.1016/j.neuroimage.2007.08.016 Scholz, J., Niibori, Y., Frankland, P. W., & Lerch, J. P. (2015). Rotarod training in mice is associated with changes in brain structure observable with multimodal MRI. NeuroImage, 107, 182–189. https://doi.org/10.1016/j.neuroimage.2014.12.003

RI PT

Ségonne, F., Dale, A. M., Busa, E., Glessner, M., Salat, D., Hahn, H. K., & Fischl, B. (2004). A hybrid approach to the skull stripping problem in MRI. NeuroImage, 22(3), 1060–1075. https://doi.org/10.1016/j.neuroimage.2004.03.032

SC

Seidler, R. D., Bernard, J. A., Burutolu, T. B., Fling, B. W., Gordon, M. T., Gwin, J. T.,. . . Lipps, D. B. (2010). Motor control and aging: Links to age-related brain structural, functional, and biochemical effects. Neuroscience and biobehavioral reviews, 34(5), 721– 733. https://doi.org/10.1016/j.neubiorev.2009.10.005

M AN U

Serrien, D. J., Ivry, R. B., & Swinnen, S. P. (2006). Dynamics of hemispheric specialization and integration in the context of motor control. Nature reviews. Neuroscience, 7(2), 160– 166. https://doi.org/10.1038/nrn1849. Sherrill, K. R., Chrastil, E. R., Ross, R. S., Erdem, U. M., Hasselmo, M. E., & Stern, C. E. (2015). Functional connections between optic flow areas and navigationally responsive brain regions during goal-directed navigation. NeuroImage, 118, 386–396. https://doi.org/10.1016/j.neuroimage.2015.06.009

TE D

Sibley, K. M., Beauchamp, M. K., van Ooteghem, K., Straus, S. E., & Jaglal, S. B. (2015). Using the systems framework for postural control to analyze the components of balance evaluated in standardized balance measures: A scoping review. Archives of Physical Medicine and Rehabilitation, 96(1), 122-132.e29. https://doi.org/10.1016/j.apmr.2014.06.021

EP

Smith, P. F. (2017). Is hippocampal neurogenesis modulated by the sensation of self-motion encoded by the vestibular system? Neuroscience and biobehavioral reviews. (83), 489– 495. https://doi.org/10.1016/j.neubiorev.2017.09.013

AC C

Sowell, E. R., Peterson, B. S., Thompson, P. M., Welcome, S. E., Henkenius, A. L., & Toga, A. W. (2003). Mapping cortical change across the human life span. Nature Neuroscience, 6(3), 309–315. https://doi.org/10.1038/nn1008 Stetter, F., & Kupper, S. (2002). Autogenic training: a meta-analysis of clinical outcome studies. Applied psychophysiology and biofeedback, 27(1), 45–98. Stimpson, N. J., Davison, G., & Javadi, A.-H. (2018). Joggin' the Noggin: Towards a Physiological Understanding of Exercise-Induced Cognitive Benefits. Neuroscience and biobehavioral reviews, 88, 177–186. https://doi.org/10.1016/j.neubiorev.2018.03.018 Strong, S. L., Silson, E. H., Gouws, A. D., Morland, A. B., & McKeefry, D. J. (2017). Differential processing of the direction and focus of expansion of optic flow stimuli in areas MST and V3A of the human visual cortex. Journal of Neurophysiology, 117(6), 2209–2217. https://doi.org/10.1152/jn.00031.2017 Taubert, M., Draganski, B., Anwander, A., Muller, K., Horstmann, A., Villringer, A., & Ragert, P. (2010). Dynamic Properties of Human Brain Structure: Learning-Related

36

ACCEPTED MANUSCRIPT Changes in Cortical Areas and Associated Fiber Connections. The Journal of neuroscience, 30(35), 11670–11677. https://doi.org/10.1523/JNEUROSCI.2567-10.2010

Taubert, M., Mehnert, J., Pleger, B., & Villringer, A. (2016). Rapid and specific gray matter changes in M1 induced by balance training. NeuroImage, 133, 399–407. https://doi.org/10.1016/j.neuroimage.2016.03.017

RI PT

Thomas, A. G., Dennis, A., Rawlings, N. B., Stagg, C. J., Matthews, L., Morris, M.,. . . Johansen-Berg, H. (2016). Multi-modal characterization of rapid anterior hippocampal volume increase associated with aerobic exercise. NeuroImage, 131, 162–170. https://doi.org/10.1016/j.neuroimage.2015.10.090 Van Praag, H., Christie, B. R., Sejnowski, T. J., & Gage, F. H. (1999). Running enhances neurogenesis, learning, and long-term potentiation in mice. Proceedings of the National Academy of Sciences, 96(23), 13427–13431.

SC

Van Praag, H. (2009). Exercise and the brain: something to chew on. Trends in Neurosciences, 32(5), 283–290. https://doi.org/10.1016/j.tins.2008.12.007

M AN U

Vann, S. D., Aggleton, J. P., & Maguire, E. A. (2009). What does the retrosplenial cortex do? Nature reviews. Neuroscience, 10(11), 792–802. https://doi.org/10.1038/nrn2733 Voss, M. W., Heo, S., Prakash, R. S., Erickson, K. I., Alves, H., Chaddock, L.,. . . Kramer, A. F. (2013). The influence of aerobic fitness on cerebral white matter integrity and cognitive function in older adults: Results of a one-year exercise intervention. Human brain mapping, 34(11), 2972–2985. https://doi.org/10.1002/hbm.22119

TE D

Wenger, E., Brozzoli, C., Lindenberger, U., & Lövdén, M. (2017). Expansion and Renormalization of Human Brain Structure During Skill Acquisition. Trends in Cognitive Sciences, 21(12), 930–939. https://doi.org/10.1016/j.tics.2017.09.008

EP

Wenger, E., Kühn, S., Verrel, J., Martensson, J., Bodammer, N. C., Lindenberger, U., & Lovden, M. (2017). Repeated Structural Imaging Reveals Nonlinear Progression of Experience-Dependent Volume Changes in Human Motor Cortex. Cerebral cortex (New York, N.Y. : 1991), 1;27(5), 2911–2925. https://doi.org/10.1093/cercor/bhw141

AC C

Wirth, A. M., Frank, S. M., Greenlee, M. W., & Beer, A. L. (2018). White Matter Connectivity of the Visual-Vestibular Cortex Examined by Diffusion-Weighted Imaging. Brain connectivity, 8(4), 235–244. https://doi.org/10.1089/brain.2017.0544 Wisse, L. E. M., Daugherty, A. M., Olsen, R. K., Berron, D., Carr, V. A., Stark, C. E. L.,. . . La Joie, R. (2017). A harmonized segmentation protocol for hippocampal and parahippocampal subregions: Why do we need one and what are the key goals? Hippocampus, 27(1), 3–11. https://doi.org/10.1002/hipo.22671 Yin, H. H., Mulcare, S. P., Hilário, M. R. F., Clouse, E., Holloway, T., Davis, M. I.,. . . Costa, R. M. (2009). Dynamic reorganization of striatal circuits during the acquisition and consolidation of a skill. Nature Neuroscience, 12(3), 333–341. https://doi.org/10.1038/nn.2261 Zech, A., Hübscher, M., Vogt, L., Banzer, W., Hänsel, F., & Pfeifer, K. (2010). Balance Training for Neuromuscular Control and Performance Enhancement: A Systematic Review. Journal of Athletic Training, 45(4), 392–403. https://doi.org/10.4085/1062-605045.4.392

37

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT