Edited by:
Reviewed by:
*Correspondence:
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Diffusion tensor imaging (DTI) is an important way to characterize white matter (WM) microstructural changes. While several cross-sectional DTI studies investigated possible links between mindfulness practices and WM, only few longitudinal investigations focused on the effects of these practices on WM architecture, behavioral change, and the relationship between them. To this aim, in the current study, we chose to conduct an unbiased tract-based spatial statistics (TBSS) analysis (
In the last two decades, white matter (WM) microstructural changes of the human brain have been widely described
Several DTI studies which have aimed at characterizing the mechanisms of FA change and at determining if these changes are the result of axon morphological modification or myelination, have also examined other diffusion parameters, such as axial and radial diffusivity (AD, RD), in the location where FA significantly changes, respectively (
Neuroimaging studies have consistently demonstrated that training and learning can modify the WM microstructure of the brain, determining related changes in behavior and/or performance (
Mindfulness practices have also been found to induce WM microstructural changes (for recent reviews see
Many cross-sectional studies have demonstrated microstructural WM differences between mindfulness practitioners and controls, as well as between novice and expert practitioners (
To our knowledge, no longitudinal investigations examined the effects of whole-body mindful movement practices on WM architecture and the possible relation between WM and behavioral changes. In addition, none of the previous studies investigated if there is a ceiling beyond which further mindful training results in no further structural changes (
Recently, a new whole-body mindful movement paradigm, the Quadrato Motor Training (QMT), was developed to enhance attention, coordination, creativity, and mindfulness (
Quadrato Motor Training incorporates all the three interdependent phases of a mindful act (
In the last years, QMT has been deeply investigated in order to highlight eventual behavioral and neurophysiological changes induced by this whole-body mindful training.
At the behavioral level, it has been demonstrated that a session or a month of daily QMT improves reaction times, ideational flexibility and spatial cognition, in contrast to several control groups, such as simple motor training and verbal training (
At the electrophysiological level, previous studies showed that a session of QMT practice significantly increases inter- and intra-hemispheric EEG alpha (α; 8–12 Hz) coherence within frontal and parietal areas in healthy adults respect to controls (
As mentioned above, no previous longitudinal study looked at the effects of whole-body mindful movement practices on WM architecture and their relationship with concomitant behavioral changes. Therefore, the first aim of the present study was to investigate the longitudinal effects of QMT on WM microstructure. In this way, we also aimed at increasing our understanding of the possible effects of this practice at a neuroanatomical level.
The second aim was to identify possible relationships between training-related longitudinal WM changes and concomitant changes in creativity, general self-efficacy and motivation. Of note, respect to previous conventional longitudinal studies in which the subjects were tested only two times, the participants in the current study were tested three times over a period of 12 weeks of daily QMT, to explore the trend of training-related microstructural WM changes over time. Furthermore, in the current work a longitudinal analysis of diffusion data was carried out following an unbiased tensor-based registration approach (
The present study is part of a larger project aimed at investigating the longitudinal effect of QMT using different brain imaging techniques. For this reason, the experimental procedure also includes electrophysiological measures, which have been analyzed and discussed elsewhere (
We recruited 50 healthy volunteers. Following the inclusion and exclusion criteria reported in
Inclusion and exclusion criteria of the present study.
– Age between 25 and 45 year |
– Right-handedness |
– No history of current or past drug addiction/abuse or antidepressant use |
– No motor, emotional, cognitive or developmental coordination disorders |
– No previous practice of the QMT or other motor activation programs |
– History of traumatic injury, previous neurosurgery, stroke, inflammatory/infective brain disease |
– Co-morbidity of congenital metabolic diseases or malformations |
– Diagnosis of one histologically proven primary cancer (< 1 years) |
– Vitamin B12 deficiency, positive serology for secondary dementia (RPR/ VDRL, HIV, anti-Borrelia), abnormal thyroid function |
– Clinical evidence of depression or other psychiatric conditions, epilepsy, drugs or alcohol addiction (according to DSM IV-TR) |
– Severe cognitive impairment (Mini Mental State Examination ≤ 24) |
– Diagnosis of malnutrition |
– Chronic or acute inflammatory disease |
– Hearing or visual impairment or motor deficits incompatible with the workout |
– Hormone replacement therapy |
– Current or recent history of smoking (i.e., not smoking during the last year) |
All procedures were explained to participants, verifying sufficient understanding and written informed consent was obtained in accordance with the declaration of Helsinki. The ethical committee of the Università Campus Bio-Medico di Roma, Rome, Italy, approved the experimental phase I study entitled “Effect of quadrato MOtor Training On the BRAIN of healthy volunteers” (MOTO-BRAIN, 09/14 PAR ComEt CBM. Compliance with GCP (Good Clinical Practice) was warranted and data were collected following the ALCOA (Attributable, Legible, Contemporaneous, Original and Accurate) algorithm. The TREND checklist was also accomplished (S1 TREND Checklist). Participants were free to interrupt the QMT and drop-out from the study at any time for any reason, without any prejudice.
Volunteers were asked to consent to a longitudinal evaluation at our institution as a pre-requisite for recruitment. The longitudinal protocol consisted of three time points: (i) baseline – the day of recruitment (T0), (ii) 6 weeks after daily QMT (T1), and (iii) 12 weeks after daily QMT (T2) (see
To check for compliance to the exercise, subjects were asked to fill up a personal diary on a daily basis and collect information about their practice and habits during the period of exercise. At each time point, the diary needed to be accurate and complete as a pre-requisite for proceeding to the next time point measurements.
At each time point, participants underwent magnetic resonance imaging (MRI) scanning brain and electroencephalography (EEG). Clinical interview and cognitive examination were performed in a dedicated room beside the MRI magnet site. Handedness was assessed by the Edinburgh Handedness Inventory (
The QMT, created by Patrizio Paoletti, requires standing at one corner of 0.5 m × 0.5 m square and making movements to different corners of the square in response to verbal instructions given by an audio tape recording indicating the next corner to which the participant should move (see
Graphical illustration of the Quadrato Motor Training. The participants stood in a quiet room at one corner of a 0.5 m × 0.5 m square and made movements to the different corners of the square in response to verbal instructions given by an audio tape recording, indicating the next corner to which the participant should move (for example, “one four” means move from corner 1 to corner 4). Participants were instructed to keep their eyes focused straight ahead, their hands loose at the side of the body and to begin all movements with the leg closest to the center of the square. In this longitudinal experimental protocol, the daily training consisted of a sequence of 69 commands lasting 7 min.
The AUs Task is an established psychometric test to assess divergent creative thinking (
The GSE (
The Mot scale was inspired by the Motivated Strategies for Learning Questionnaire (MSLQ), (
In this task, the participant was asked to state the degree to which he or she agrees with each of the statements in the questionnaire on a scale ranging from “1” (strongly disagree) to “7” (strongly agree) related to the level of motivation and enjoyment prior, during and following the training. The total scores on the questionnaire range from 10 to 70, with the highest score reflecting higher motivation.
Imaging data were acquired using a Siemens 1.5-T MAGNETOM Avanto (Siemens, Erlangen, Germany) whole body scanner equipped with a 12-element designed Head Matrix coil, as part of the standard system configuration. Diffusion weighted images (DWIs) were acquired using an axial pulsed-gradient spin-echo echo-planar sequence (7600/103; 38 sections; section thickness, 3.0 mm with no intersection gap), with diffusion-encoding gradients applied in 12 non-collinear directions (b factor 0 and 1000 s/mm2; number of acquired signals, four). A 2D fluid attenuated inversion recovery (FLAIR) T2 weighted scan was also used to exclude the presence of small vessel ischemic disease and other supra- or infra-tentorial brain lesions (TR = 11460 ms, TE = 102 ms, TI = 2360 ms, FOV = 280 mm × 330 mm, NEX = 2, matrix = 248 × 320, 1.00 × 1.00 mm2 in-plane resolution, horizontal slices with slice thickness of 3.0 mm and no gap). Structural images were collected using a sagittal magnetization-prepared rapid acquisition gradient echo (MPRAGE) T1-weighted sequence (TR = 2400 ms, TE = 3.61 ms, TI = 1000 ms, flip angle = 15°, FOV = 240 mm × 280 mm, NEX = 1, matrix = 192 × 192, 1.00 × 1.00 mm2 in-plane resolution, horizontal slices with slice thickness of 1.2 mm and no gap). Whole brain functional scans were also acquired in 25 contiguous axial slices approximately parallel to the anterior-posterior commissure plane with interleaved multi-slice T2 echo-planar imaging (TR = 3560 ms, TE = 50 ms, field of view = 22 cm, flip angle = 90°, voxel size = 3.4 × 3.4 × 3 mm, slice thickness = 3 mm, no inter-slice gap, 135 volumes). Since the present paper focused on WM microstructural QMT- related changes, resting-state data will not be discussed further.
To avoid a type I error induced by the effect of WM hyperintensities on brain connectivity results, two expert radiologists (CCQ, YE) examined all MRIs. Subjects were excluded when more than 3 lesions with a maximum diameter of 5 mm were detected in the subcortical or periventricular WM on axial FLAIR images (
All DWIs were visually inspected for artifacts and preprocessed using different tools from FDT (FMRIB Diffusion Toolbox, part of FSL (FMRIB’s Software Library v.5.0.8,
FA, RD, and AD data from each participant were furtherly analyzed using the Tract-Based Spatial Statistics (TBSS;
To investigate on the effects of QMT on creativity, perceived self-efficacy and motivation, repeated measure analyses of variance (ANOVA) over the three time points (T0, T1, and T2) were performed, separately for ideational fluency, flexibility and originality AUs’ subscales scores, GSE and Mot measures. The differences between time points were finally computed for the all the AUs’ subscales scores, the GSE scores and the Mot scores. Statistical analyses on behavioral data were performed using Statistica v.7 software (StatSoft Inc., United States).
White matter microstructural changes were also investigated performing three separate one-sample
Voxelwise statistical analyses were carried out using permutation-based non-parametric statistics using the FSL Randomize permutation-based program (
Randomize tool (5,000 permutations) was also used to examine the statistical correlation between significant longitudinal changes of diffusion parameters and longitudinal behavioral changes. Resulting statistical maps were thresholded at pFWE < 0.05.
All the results were anatomically localized using the JHU ICBM-DTI-81 White-Matter Labels and the JHU White-Matter Tractography atlases included in the FSL distribution
As shown by repeated measures ANOVAs, QMT significantly increased the originality subscale score of the AU task over time (
At T1 respect to T0, TBSS analysis revealed a significant (pFWE < 0.05) FA increase of several WM tracts (
Significant increases in FA and decreases in RD after 6 weeks of daily QMT (T1) respect to T0 baseline (pFWE < 0.05, TFCE corrected). RD changes were investigated in the locations where FA significantly changed. The study-specific FA skeleton, representing the centers of principal WM tracts, is displayed in green, overlaid on the mean FA map. The vertical lines on the coronal view indicate the sagittal slices displayed. The horizontal lines on the sagittal view indicate the axial slices displayed. The red–yellow and blue–light blue color bars represent level of significance for FA increase and RD decrease, respectively.
Significant increases in FA at T1 respect to T0 (pFWE < 0.05 TFCE-corrected).
MNI coordinates |
||||||
---|---|---|---|---|---|---|
Cluster size | WM structures | |||||
7409 | 5.93 | <0.001 | –16 | –7 | 3 | |
5.79 | <0.001 | –20 | –18 | 42 | Left corticospinal tract | |
4.32 | 0.007 | –24 | 25 | 6 | Left inferior fronto-occipital fasciculus, Left Uncinate fasciculus | |
4.01 | 0.007 | –34 | 4 | 2 | Left superior longitudinal fasciculus | |
3.57 | 0.008 | –20 | 18 | 30 | Left superior longitudinal fasciculus (temporal part) | |
3.47 | 0.018 | –3 | 21 | –2 | Forceps minor/ |
|
2266 | 4.93 | 0.003 | –10 | –17 | 16 | Left anterior thalamic radiation |
3.69 | 0.019 | 9 | –28 | –11 | Right anterior thalamic radiation | |
1163 | 4.44 | 0.021 | 10 | –1 | –4 | Right anterior thalamic radiation |
3.89 | 0.023 | 28 | –19 | 19 | Right corticospinal tract | |
2.77 | 0.038 | 18 | –20 | –7 | Right corticospinal tract/ |
|
1159 | 4.86 | 0.019 | –53 | –37 | –19 | Left superior longitudinal fasciculus (temporal part) |
4.67 | 0.015 | –37 | –28 | –2 | Left inferior fronto-occipital fasciculus/ |
|
4.23 | 0.019 | –42 | –4 | –26 | Left inferior longitudinal fasciculus | |
316 | 3.74 | 0.021 | –21 | –53 | 60 | Left superior longitudinal fasciculus |
153 | 3.42 | 0.029 | –46 | –14 | –11 | Left inferior longitudinal fasciculus |
128 | 4.05 | 0.043 | 17 | 7 | 8 | Right anterior thalamic radiation/ |
106 | 3.56 | 0.044 | 24 | –18 | 35 | Right corticospinal tract |
75 | 2.79 | 0.046 | 33 | 5 | –11 | Right uncinate fasciculus/ |
45 | 3.13 | 0.047 | –6 | –39 | –21 | Left anterior thalamic radiation/ |
44 | 2.55 | 0.047 | 17 | 19 | –11 | Right uncinate fasciculus/ |
28 | 2.23 | 0.047 | –15 | 11 | 29 | |
28 | 2.44 | 0.033 | –33 | –35 | 8 | Left inferior longitudinal fasciculus/ |
The increase of FA at T1 respect to T0 was accompanied by a significant bilateral decrease of RD in corticospinal tracts and anterior thalamic radiations, also including the posterior limbs of internal capsule. RD decrements were also found in the left uncinate, inferior fronto-occipital, and superior longitudinal fasciculi, as well as in the right anterior limb of the internal capsule and cerebral peduncle (
Significant decreases in RD at T1 respect to T0 (pFWE < 0.05 TFCE-corrected).
MNI coordinates |
||||||
---|---|---|---|---|---|---|
Cluster size | WM structures | |||||
4615 | 4.82 | <0.001 | –16 | –7 | 4 | Left anterior thalamic radiation/ |
4.80 | 0.002 | –12 | –15 | 67 | Left superior longitudinal fasciculus | |
3.87 | <0.001 | –26 | –20 | 22 | Left corticospinal tract | |
2.24 | 0.009 | –23 | 21 | 5 | Left inferior fronto-occipital fasciculus, Left uncinate fasciculus | |
906 | 5.27 | 0.007 | –10 | –16 | 16 | Left anterior thalamic radiation |
3.68 | 0.011 | 10 | –15 | 15 | Right anterior thalamic radiation | |
468 | 4.42 | 0.012 | 12 | –3 | –4 | Right anterior thalamic radiation |
2.31 | 0.035 | 24 | –6 | 18 | Right corticospinal tract/ |
|
388 | 3.73 | 0.014 | 28 | –19 | 18 | Right corticospinal tract/ |
122 | 3.47 | 0.032 | 18 | –20 | –7 | Right corticospinal tract/ |
74 | 4.93 | 0.028 | 17 | 7 | 8 | Right anterior thalamic radiation/ |
At T2 respect to T0, a significant (pFWE < 0.05) increase of FA was still present (
Significant increases in FA and decreases in RD after 12 weeks of daily QMT (T2) respect to T0 baseline (pFWE < 0.05, TFCE corrected). See
Significant increases in FA at T2 respect to T0 (pFWE < 0.05 TFCE-corrected).
MNI coordinates |
||||||
---|---|---|---|---|---|---|
Cluster size | WM structures | |||||
4337 | 4.89 | 0.012 | –20 | 51 | –2 | Left anterior thalamic radiation |
4.79 | 0.016 | –23 | 27 | –2 | Left uncinate fasciculus, Left inferior fronto-occipital fasciculus | |
4.25 | 0.021 | –16 | 53 | 18 | Forceps minor | |
4.05 | 0.014 | –35 | –3 | –14 | Left uncinate fasciculus | |
3.90 | 0.023 | –28 | –28 | 7 | Left inferior fronto-occipital fasciculus/ |
|
3.38 | 0.022 | –22 | –5 | 15 | Left anterior thalamic radiation/ |
|
1.95 | 0.031 | –43 | –28 | –7 | Left inferior longitudinal fasciculus | |
653 | 4.57 | 0.025 | –21 | –24 | 40 | Left corticospinal tract |
Significant decreases in RD at T2 respect to T0 (pFWE < 0.05 TFCE-corrected), investigated in the locations where FA significantly changed.
MNI coordinates |
||||||
---|---|---|---|---|---|---|
Cluster size | WM structures | |||||
139 | 4.11 | 0.008 | –10 | –4 | –6 | Left anterior thalamic radiation |
40 | 3.30 | 0.033 | –27 | 10 | –13 | Left uncinate fasciculus |
The repeated measures ANOVA performed on the mean FA values at T0, T1, and T2, showed a significant effect for time (
ANOVA repeated measures - Time (3) factor for mean FA values. Data show a significant principal effect for the Time factor. Bonferroni
The correlation analyses between significant longitudinal changes of diffusion parameters and behavioral changes yielded no FWE-corrected results. However, since previous studies already reported significant correlations between QMT-related changes of psychological well-being measures and electrophysiological indices (
Results of voxelwise correlation analyses between longitudinal changes in FA and RD maps and concomitant changes in behavioral tests (
MNI coordinates |
||||||
---|---|---|---|---|---|---|
Cluster size | WM structures | |||||
34 | 2.65 | <0.001 | 15 | –2 | 7 | Right anterior thalamic radiation/ |
23 | 3.31 | <0.001 | –34 | –18 | 35 | Left superior longitudinal fasciculus |
56 | 3.44 | <0.001 | –17 | –14 | 55 | Left superior longitudinal fasciculus |
38 | 4.26 | <0.001 | 16 | –2 | 6 | |
31 | 3.31 | 0.001 | –29 | –3 | 27 | Left superior longitudinal fasciculus (temporal part) |
82 | 3.33 | <0.001 | –23 | 18 | 13 | Left anterior thalamic radiation |
30 | 4.31 | <0.001 | –22 | –3 | 15 | Left anterior thalamic radiation / |
29 | 3.88 | 0.001 | 9 | –7 | 11 | Right anterior thalamic radiation |
24 | 4.79 | <0.001 | –37 | –24 | 31 | Left superior longitudinal fasciculus, Left Superior longitudinal fasciculus (temporal part) |
29 | 3.51 | 0.001 | –21 | –3 | 16 | Left anterior thalamic radiation/ |
22 | 3.93 | <0.001 | –22 | 15 | 12 | Left anterior thalamic radiation/Left |
In this work, we investigated for the first time the longitudinal effects of daily QMT on WM microstructure in a healthy group of subjects. Of note, respect to conventional pre–post training longitudinal studies, our subjects were tested three times over a period of 12 weeks of QMT, to examine the trend of training-related microstructural WM changes over time. Furthermore, we used an unbiased DTI analysis pipeline for tracking longitudinal WM changes, following recent advances in tensor-based image registration (
Our results revealed that QMT daily practice significantly affected WM microstructural architecture over time. Respect to the baseline (T0), FA values increased after 6 weeks of training (T1) in different bilateral tracts and in major associative tracts of the left hemisphere. Significant training-induced FA increments were still present after 12 weeks of QMT (T2) respect to T0, although less widespread and only localized in the left-hemisphere. No significant FA changes were found at T2 respect to T1.
We also examined the pattern of AD and RD changes in tracts where FA significantly increased and found a significant decrease of RD both at T1 and at T2, supporting the relevance of myelination processes in training-related FA changes. Behavioral analyses showed that our subjects remained motivated during the entire course of training, confirmed and deepened the knowledge of the longitudinal effect of QMT on creativity and revealed a training-related effect on self-efficacy. Finally, we found significant correlations between individual WM microstructural changes and individual improvements of self-efficacy and originality. These findings support the effectiveness of QMT in improving WM integrity and suggest the relevance of these microstructural changes for psychological well-being.
The unique combination of motor and cognitive components, which distinguishes the QMT from other mindfulness practices, could explain the present pattern of results, which comprises WM tracts related to sensorimotor functions as well as critical tracts for high-level cognitive operations.
Respect to T0, significant FA increments were found bilaterally in the corticospinal tract at T1. These changes are probably related to the sensorimotor effect of the QMT, in accordance with previous studies (
Significant FA increments were also found at T1 in the anterior thalamic radiations, which are generally related to executive function, memory encoding and planning of complex behaviors (
Other significant microstructural FA increments were found at T1 in both left and right uncinate fasciculi, which persisted in the left-hemisphere at T2. These fasciculi play a role in emotion regulation, emotional learning, memory, and language functions (
Fractional anisotropy increments at T1 were also localized in the genu of the corpus callosum and forceps minor, as well as in the superior cerebellar peduncles. Longitudinal increases of FA in the genu and forceps minor have been already reported after mindfulness (
At T1, an increase of FA was also detected in the left superior longitudinal fasciculus, which connects parietal to frontal ipsilateral regions, as well as in its temporal part, which instead connects temporal with ipsilateral frontal areas, also including fibers belonging to the arcuate fasciculus (
Robust long lasting (at both T1 and T2 respect to T0) left-lateralized FA increases were present in two long association fiber tracts, the inferior longitudinal and inferior fronto-occipital fasciculi. While the inferior longitudinal fasciculus is thought to mediate fast transfer of visual signal to temporal regions (
In the present work, no significant FA changes were instead found at T2 respect to T1. However, the repeated measures ANOVA showed that mean FA values at T2 were lower than at T1. This leads to the suggestion that the training-related increase of FA could have already met at T2 a descending phase. Notably, the same trend has been previously reported in several domains of expertise, such as motor sequence learning (
Although previous studies reported left-lateralized changes following different types of training like working memory (
Likewise the present results, several works reported the same pattern of FA increase and RD decrease after training (
In the present work, training-induced improvements in originality and general self-efficacy (GSE) were associated with increased FA/decreased RD in the right anterior thalamic radiation, and left superior longitudinal fasciculus. Creativity measures have been previously related to the anterior thalamic radiation, a fiber tract associated to creative cognition (
Similar to the originality results, we found that higher GSE scores were associated with increased FA/decreased RD in the anterior thalamic radiations and left superior longitudinal fasciculus. This is the first study that investigated on possible correlations between training-related WM changes and measures of self-efficacy. GSE has been positively related to optimism, self-respect and internal control (
There are a few limitations to this study, which should be noted. The first is the lack of a control group with no training or a control group with the same type of motor activity (but with reduced cognitive demands) or cognitive effort (but reduced motor load). However, several studies have already demonstrated the longitudinal reliability of DTI measures, including previous learning studies where control groups did not show FA changes (
The effectiveness of the Quadrato Motor Training, together with the ease of learning and the minimal time and space it requires, make this training a very promising and feasible paradigm for children, adolescents and elderly people, contributing to well-being in the healthy but also useful for neurorehabilitation. Future research should examine QMT efficacy on different populations suffering from altered WM microstructural connectivity and impaired cognitive performance, such as mild cognitive impairment patients, or with decreased motor and/or cognitive functions, such as learning disabilities, language disorder and Parkinson disease. Furthermore, exploring how WM microstructural changes are related to measures of well-being and health may have relevant implications for cognitive and educational neuroscience as well as psychotherapeutic programs.
TB-S, FC, and CQ designed the research. CQ, CM, and YE performed the research. CP, FC, and FM analyzed the data. CP and FC wrote the paper. TB-S, FM, and CQ contributed to the writing process.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.