Cortical correlates in upright dynamic and static balance in the elderly

Falls are the second most frequent cause of injury in the elderly. Physiological processes associated with aging affect the elderly’s ability to respond to unexpected balance perturbations, leading to increased fall risk. Every year, approximately 30% of adults, 65 years and older, experiences at least one fall. Investigating the neurophysiological mechanisms underlying the control of static and dynamic balance in the elderly is an emerging research area. The study aimed to identify cortical and muscular correlates during static and dynamic balance tests in a cohort of young and old healthy adults. We recorded cortical and muscular activity in nine elderly and eight younger healthy participants during an upright stance task in static and dynamic (core board) conditions. To simulate real-life dual-task postural control conditions, the second set of experiments incorporated an oddball visual task. We observed higher electroencephalographic (EEG) delta rhythm over the anterior cortex in the elderly and more diffused fast rhythms (i.e., alpha, beta, gamma) in younger participants during the static balance tests. When adding a visual oddball, the elderly displayed an increase in theta activation over the sensorimotor and occipital cortices. During the dynamic balance tests, the elderly showed the recruitment of sensorimotor areas and increased muscle activity level, suggesting a preferential motor strategy for postural control. This strategy was even more prominent during the oddball task. Younger participants showed reduced cortical and muscular activity compared to the elderly, with the noteworthy difference of a preferential activation of occipital areas that increased during the oddball task. These results support the hypothesis that different strategies are used by the elderly compared to younger adults during postural tasks, particularly when postural and cognitive tasks are combined. The knowledge gained in this study could inform the development of age-specific rehabilitative and assistive interventions.


Introduction
Falls are usually caused by a combination of risk factors that increase with age. Indeed, adults older than 65 years of age suffer the greatest number of fatal falls each year 1 .
In the optimally functioning nervous system, the maintenance of upright stance is an automatic involuntary task, predominantly under control of cortical and spinal networks. The cortical contribution is challenged with increased cognitive load, depending on the postural task, the age of the individuals and their balance abilities 2 .
In the last decades, effective quantification of complex balance task by using different kinds of perturbations, e.g., either a platform perturbation or a cognitive task (e.g., double tasking), was mainly assessed by computerized dynamic posturography (CDP) 3 . This kind of research demonstrated that elderly -when performing a cognitive task concurrently during a challenging balance recovery -prioritize balance recovery. CDP was also quantified in virtual-reality based scene: visual interference induced a significantly larger postural sway 45 . Other studies exploring this relationship between attention, posture and gait in younger and older adults proved that postural control appears to be more cognitively demanding in older adults compared to younger adults and that the performance of a dual-task appears to have a more deleterious effect on postural control in the elderly compared to younger adults 678910111213 . In fact, it is not still clear if these differences are due to shift of attention between two different tasks, the reduction in attention capacity or the increased demand due to impairments in the postural control system in the ageing nervous system 2 .
Understanding the neurophysiological control of balance in increasingly challenging balance tasks (static, dynamic, during dual-task) may help to clarify age-relate differences. Dual-task experiments 14 are usually designed to represent real-life situations of instability and to try to introduce it in experimental settings. The quest for neural predictors of dynamic balance instability (i.e., during gait) while double-tasking (i.e., texting or conversing) in younger participants suggests a leading role of the posterior parietal cortex (PPC), involved in the processes of sensorimotor integration and gait control 14 . Indeed, experiments in younger healthy volunteers showed that, when walking naturally without engaging in any secondary tasks, PPC is active in the theta range (4)(5)(6)(7). When a double-task is added, a positive predictive relationship could be identified between PPC activity in the alpha frequency band and walking pace 141516 . The existence of neural detectors of postural instability that could trigger the compensatory adjustments to avoid falls was also supported by a burst of gamma activity that preceded the initiation of compensatory backward posture 17 .
To our knowledge, there is a paucity of studies investigating the cortical activation during balance in healthy elderly, in particular using in parallel high-density electroencephalographic (EEG) systems and electromyographic (EMG) sensors 6 . The scarce number of studies addressing this issues 1819 , show that the scalp EEG power distribution increases in both elderly and younger adults control group in delta EEG frequency band during challenging postural conditions; theta rhythms, on the other hand, were more responsive to cognitive tasks in both groups, with more marked increases in younger subjects, and gamma oscillations increased in the elderly primarily over central and central-parietal cortices during challenging postural tasks.
The majority of studies analysing power spectral density changes associated with postural control are based on younger participants 4 . An increase in frontal and parietal midline theta power is observed with demanding continuous balance control 20 and fronto-central and centro-parietal theta power resulted associated with balance performance 20 . Specifically, significant increases in theta band spectral power in the left sensorimotor cortex played a larger role in sensing loss of balance during walking than the right sensorimotor cortex 21 . Cortical alpha rhythms resulted correlated with body sway during quiet open-eyes standing 22 and beta band power of the sensorimotor cortex reflected muscle contraction 23 .
Despite the wide heterogeneity in experiments, protocols, and data analyses 6 , findings on EMG data are more consistent: EMG latency, amplitude and muscular co-contraction are usually larger in the elderly compared to younger participants during walking and upright stance 242526272829 . Increased activation of antagonist muscles in old adults seems a change associated with normal healthy aging 30 .
In this study, we monitored both EEG source waveform spectral power distribution estimated by high-density (i.e., 256 electrodes) EEG recordings, and EMG amplitudes recorded by 8 wireless loggers on the main muscles of the leg and the trunk, during single postural tasks (i.e., standing on a 0%, +22%, -22% inclined planes and on a core board) and dual postural-visual tasks (adding a visual oddball -i.e., the participant is asked to count the odd stimuli presented on a screen that differ from the more frequent stimuli -to the previous conditions) in 9 healthy elderly (> 64 y) and 8 younger (< 35 y) adults. The main aim is to investigate cortical and muscular activation underlying age-related differences in postural control. Studying both the EEG and EMG signals while engaging in a secondary task in healthy subjects mimics a real-life situation of distraction that might lead to elderly instability. We chose a visual task because visual and proprioceptive inputs seem to play dominant roles in maintaining postural stability 5 .
The aim of the study is to answer to the following research questions: (Q1) Which are the age-related differences in cortical and muscular activations during static and dynamic balance and (Q2) How they are modified during dual postural-visual tasks. Toward this aim, we identified EEG power spectral content and source generators of brain regions active during different postural conditions and dual-tasking in a cohort of younger and old healthy adults and estimated EEG spectral power and EMG amplitudes differences among the two groups.

Results
Behavioral tests. To evaluate visual attention and task switching, before the dual-task experiment, participants underwent the Trail Making Test (TMT) A and B. The time in s needed for performing both TMT-A (assessing cognitive processing speed) and TMT-B (assessing executive functioning) is reported in Figure1 and in Table1 (2nd and 4th columns). Since performance on TMT-A and TMT-B is affected by both age and education 31 , we computed the cut-off for each one of our participants considering these two covariates 32 (3rd and 5th columns of Table1). No one of our participants showed an impaired performance considering her/his age and level of education comparing the time needed to perform the tasks (2nd and 4th columns of Table1) and the cut-offs (3rd and 5th columns of Table1). The time needed to perform the tests was normally distributed for both the groups, and significantly higher in the TMT-B compared to the TMT-A execution in particular for the elderly (p value < 0.001) and overall in the elderly (p value < 0.01). TMT performance declines, on average, during healthy aging 33 with performance on TMT-B declining significantly more than TMT-A in older adults 34 . Cortical and muscular activation underlying age-related differences in postural control (Q1). To emphasize the differences in cortical and muscular activation between younger adults and elderly under the different conditions, we reported the results during static and dynamic balance in Figure 2. In the baseline condition, static balance (i.e., undisturbed upright standing), elderly ( Figure 2a) exhibit a shift to lower rhythms compared to younger adults ( Figure 2c): indeed higher values in the low frequency rhythms (delta band) are present in the elderly whereas higher values in frontal and midline regions in higher frequency rhythms (i.e., beta, low and high gamma) are present in the younger adults. Significant differences (p < 0.05) between the two groups are localized over frontal regions in delta band (higher values in elderly) and over frontal and sensorimotor regions in beta and gamma bands (higher values in younger adults). During dynamic balance (i.e., on the core board), the motor/occipital theta is increasing in the elderly ( Figure 2b) compared to the younger adults ( Figure 2d) as confirmed by statistical tests (Wilcoxon test (Figure 2f)).
Cortical activations underlying age-related differences during dual postural-visual tasks (Q2) To emphasize the differences in cortical activation in the two groups, we reported the results during the dual-task (mental counting the odd stimuli) during static and dynamic balance in Figure 4 and 5. Figure 4 shows the median values for each ROI computed among the elderly whereas Figure5 for the younger adults. In particular, in Figure 4 and in 5, we reported the difference of the power spectra between the 0.5 s-epoch post-odd stimulus and the 0.5 s-epoch pre-odd stimulus in the same postural conditions of Figure 2. The p-values of the statistical significant differences between the younger and the elderly group were reported in Figure 4c during static balance (simple standing) and in Figure 5c during dynamic balance.
During static balance (simple standing) double-tasking, the elderly ( Figure 4a) increased the values on their midline theta compared to the younger adults ( Figure 5a) and showed higher values on their frontal regions in high-gamma. During dynamic balance double-tasking, the elderly ( Figure 4b) increase their midline delta and high gamma compared to younger adults ( Figure 5b). Comparing cortical activations in the elderly during static and dynamic balance, significantly higher cortical activations are localized over sensorimotor regions in alpha and beta bands and over frontal region in high gamma during dynamic balance (Figure 4c). A different pattern is observed in younger adults, with an increase in gamma band localized over occipital regions (Figure 5c).
Interestingly, only in the younger adults delta power in the centro-parietal areas significantly increased during dynamic balance. When double-tasking, only the elderly showed a decrease in delta rhythm in the centro-parietal area parallel to an increase in theta rhythm in the occipital area during dynamic balance compared to static balance.
Comparing the performance (Figure6a and Figure6b) in dual-tasking between elderly (Figure4) and younger (Figure5) adults, we observe a significant increase in frontal and occipital theta rhythms and frontal high gamma in the elderly compared to the younger adults in dual-tasking already in static balance (Figure6a).
During dual-tasking static balance over the inclined plane (-22%), the elderly show higher cortical activations in alpha and gamma bands over occipital and frontal regions respectively. This pattern is similar to that observed in younger adults during dynamic balance, suggesting a similar cognitive load in elderly and younger adults in these two different postural conditions. Higher cortical activation in younger adults are observed, during dual-task static balance over the inclined plane (+22%) over frontal regions in theta and low gamma bands, similar to the pattern observed during static balance. During the same task, the elderly show higher activation of frontal and central regions in beta range. We refer the reader to the Supplementary material for the results of all the statistical tests for the EEG power spectra.
In Figure3, we reported the RMS values of muscular activation signal normalized by the median RMS value of the maximum voluntary contractions ( Table 2) among subjects of the same group for each assessed muscle: Erector spinae (ES); Rectus femoris (RF); Vastus lateralis (VL); Biceps femoris (BF); Tibialis anterior (TA); Peroneus longus (PL) and Gastrocnemius lateral head (GL). In particular, each barplot represents the median values computed among the elderly participants (first row, in green, of Figure3) and the younger adults (second row, in pink, of Figure 3).
RMS values for the Erector spinae significantly differ in all the postural conditions between the two groups: elderly show higher values compared to the younger adults and an increasing trend from 0% inclined plane to the core board (first column of Figure3). The RMS values for the RF were significantly higher in the elderly participants than in the younger adults on the -22% inclined plane and on the core board. The RMS of the GL was significantly higher on the 0% plane and on the core-board in the elderly compared to the younger adults.
The activity of the PL and GL in -22% condition revealed a plantar-flexion action, which is similar in the two groups but it is complemented by two different knee strategies between elderly and younger adults: elderly showed higher co-contraction of RF and BF, whereas younger adults prefer recruiting the knee flexors.
When looking at the results obtained for the +22% condition, results suggested a different ankle strategy to keep balance: elderly showed higher co-contraction of TA, PL and GL than younger adults, with the higher difference in the recruitment of TA; at the upper leg, extensors activity is higher than BF for both groups.
Interestingly, the task on the core board highlighted a further co-contraction of plantarflexors and dorsiflexors of the ankle joint in the elderly (i.e., higher co-contraction of TA, PL and GL) than in the younger adults, who prefer the recruitment of the PL for fine movement control. The activity of the RF, which serves both as knee extensor and hip flexor, was also found to be significantly higher in the elderly than in the younger adults on the core board.
No significant differences were found in comparing the EMG data during the postural task with and without the visual oddball.

Discussion
Our findings provide evidence of different motor control strategies to maintain static and dynamic balance in the elderly compared to younger healthy adults.
During static balance, delta power over frontal regions was higher in the elderly, whereas younger participants had higher activation of frontal and occipital regions in the fast frequencies. Both groups showed a mid-line fronto-central theta rhythms, a neural signature of the attentive process of maintaining posture. During dynamic balance, the elderly increased theta rhythms over sensorimotor and occipital areas. When adding a visual oddball task, during static balance the elderly showed stronger activations over the sensorimotor areas compared to the non-double tasking condition. Dynamic balance only magnifies these findings. In the younger group, static balance during double tasking induces an occipital theta and gamma rhythm, associated with frontal alpha increase, which again are magnified during dynamic balance.
Although the different cortical activation strategies are more evident during the dual-task, they emerged already during quite static balance. The elderly showed high delta activation over the frontal areas, a process hypothesized to contribute the maintenance of posture 1819 . The younger adults showed prevalent fast frequencies over the frontal and occipital areas, suggesting an ongoing attentive task to maintain upright posture supported by a visual fixation strategy identified by the activation over the occipital lobes.
Dual-task is a well-known strategy to increase the cognitive load in experimental conditions and mimics real-life settings in which balance maintenance is coupled with tasks such as speaking or texting. A common paradigm is the visual oddball, during which the participant is asked to count the stimuli presented on a screen that differ from the more common one. In our experiment, the double task addressed the issue of the competing cognitive resources the elderly deploy during stance. In line with previous research 234 , we expected an increase of cortical recruitment to keep up performance in the elderly. The observed activation of posterior areas in theta band supports this hypothesis; it is coupled with a recruitment of sensorimotor cortices,

5/15
confirmed by the muscle activation pattern hinting to an increased muscular firing to maintain posture.
In the younger adults, we observed during dynamic postural double tasking a strong activation of the occipital and frontal areas, congruent with the oddball task (visual stimulus and counting). While we cannot definitely disentangle the components of the visual activation, it is likely that younger participants adopt a visual fixation strategy during this more challenging task, as they did already during unchallenged static balance.
It is noteworthy that the main differences in cortical activations between elderly and younger adults appear during doubletask, suggesting that they were able to recruit adaptive reserve. This finding contrasts with previous work 35 which described increased cortical activation during double tasking static posture to be suggestive of a defective compensatory mechanism in the elderly. It is also possible that the different nature of the task -static vs dynamic -may render these findings not directly comparable.
In normal aging structural and functional physiological brain changes occur. Structural changes (i.e., cortical thinning, local brain atrophy, etc.) together with functional maladaptive brain activity, namely decreased functional specificity and loss of functional lateralization, trigger the activation of compensatory mechanisms 36 . This has been referred in literature as compensatory cognitive scaffolding, whose efficacy largely depends on each one's cognitive reserve 37 . This model explains older individual's level of cognitive functioning: this relies on compensation mechanisms by which supplementary higher level cortical loops are recruited to accomplish also those tasks supposed to be highly automatized, as the control of upright stance 38 . Thus, the increased allocation of attention sources to handle a challenging postural task (as in the core board condition), together with a cognitive oddball paradigm, lead the older group to necessarily increase recruitment of cortical areas to keep balance. This is not needed by younger adults who relay on more automatized and less central processes to maintain balance, having more cognitive sources available to easily accomplish the task.
The elderly showed higher muscular activity than younger adults. The higher co-contraction of ankle plantarflexors and dorsiflexors in the elderly suggest a stiffening action at the ankle joint already during static balance and magnified in dynamic tasks. In both groups, PL shows a strong activation already during static balance, confirming the need for a greater stabilization over the mediolateral plane 39 . However, the elderly tend to complement PL action with TA and GL, whereas in challenging postural tasks (i.e., the core board) younger adults prefer to control the ankle joint mainly with the PL for fine movement control.
These mechanisms are integrated by a higher co-contraction of knee extensors (RF and VL) and flexors (BF) during dynamic balance in the elderly, most likely to add stiffness to the knee. The young adults, instead, prefer a strategy that allows lowering their center of mass to increase their stability when needed.
We can notice that the RMS values for the Erector spinae significantly differ in all the postural conditions between the two groups: elderly show higher values compared to the younger adults and an increasing trend from 0% inclined plane to the core board (first column of Figure 3). The activity of the RF, which serves both as knee extensor and hip flexor, was also significantly higher in the eldelrly during dynamic balance and complemented by a significant higher ES activity.
Overall, the elderly seem to prefer a stiffening strategy of the lower limb with a compensatory mechanism of the trunk to keep balance, whereas younger subjects work on their stabilization of the mediolateral plane, recruiting the PL for fine movement control, and lower their center of mass, i.e., a combination of ankle and knee strategy.
Our findings indicate that the elderly need to increase cortical recruitment during postural challenging tasks with additional cognitive load. Dynamic postural control during double tasking needs the involvement of the whole sensory-motor network: the correlate of this cortical finding is the preferred motor control strategy, which in the elderly sees a muscle co-contraction to stiffen lower limb and trunk. Younger subjects are apparently undisturbed by the double tasking, with a possible visual fixation strategy to maintain dynamic balance which can only be inferred from our data. Their motor strategy seems more efficient and less energetically demanding, given the overall lower contraction levels and reduced number of involved muscles.
These observations may guide future rehabilitative interventions, which need to be targeted to the compensatory mechanism in each population, as well as provide insight into the neural processes underlying postural control.
This study was carried out in accordance with the recommendations of Ethics Committee of the Teaching Hospital of Padua (n.AOP2025) with written informed consent from all subjects.

Behavioral tests
As first task, participants were asked to perform the Trail Making Test, i.e., a neuropsychological test of motor speed and visual attention composed by two parts. In part A, i.e., TMT A, the subject's task is to quickly draw lines on a page connecting 25 consecutive numbers without lifting the pen from the paper. In case errors occur, participants are alerted and correction is allowed. The performance is assessed by the time taken to complete the trial correctly. In part B, i.e., TMT B, it is tested how fast the participant can connect numbers and letters in alternating increasing sequence (i.e., 1-A-2-B, etc.). Part B is more difficult than Part A not only because it is a more difficult cognitive task, but also because of its increased demands in motor speed and visual search 40 . Thus, TMT B evaluates visual attention, motor speed, and cognitive alternation.

Data acquisition
High-density EEG recordings were acquired at Padova Neuroscience Center inside a dimly lit sound-attenuated and electrically shielded room with the Geodesic Sensor Net with 256 electrodes (Electrical Geodesic Inc., Eugene, OR, USA). Electrode-skin impedances were maintained < 40 kΩ. The recordings were sampled at 500 Hz, referenced to Cz. In parallel, synchronized EMG recordings were acquired from the following muscles in the right leg (Figure 3a)): Rectus femoris (RF), Vastus lateralis (VL), Tibialis anterior (TA), Peroneus longus (PL), Gastrocnemius lateral head (GL), Biceps femoris (BF); and from Erector spinae (ES) with Cometa MiniWave Waterproof EMG sensors (Cometa srl, Milan, Italy). An expert researcher individualized the minimal crosstalk areas for the EMG electrode placement 41 as suggested by 42,43 . Skin preparation was performed removing dead cells and humidifying the areas if participants had a dry skin, or rubbing it with alcoholic wipes when they had a oily skin 43 . Electrode-skin impedances were maintained < 74 kΩ. The recordings were sampled at 2 kHz.
Baseline. Before performing the dual-task experiment, 3 minutes of quiet upright and barefoot standing in front of a black screen were acquired for each participant at the beginning and at the ending of the experiment.
The dual-task experiment consisted in a static or dynamic postural task while performing a visual task. Both the static and dynamic postural tasks were tested with and without the cognitive perturbation, i.e., a visual oddball. In particular, 30 s of standing before and after the visual oddball were recorded.
Static postural conditions. Participants were standing upright and barefoot on a plane -feet position was maintained consistent from trial to trial and at shoulder distance -inclined at different angles: 0% +22% (i.e., 12 • ) and -22%.
Dynamic postural condition. Participants were keeping balance on a 50cm 2 wooden core board with a 12 • front-to-back tilt.
Visual oddball. Participants were asked to focus on a series of coherent pictures (i.e., red squares) presented on a screen in front of them with about 0.5 Hz rate. One incoherent picture (i.e., a yellow square) appeared at a random time during each trial. Participants were instructed to keep their gaze in front of them. Each block contained circa 80 standard (frequent) stimuli, and circa 20 incoherent (odd) stimuli. The total experiment included four blocks (3 for the static postural conditions, i.e., 0% +22% and -22%, and one for the dynamic postural condition) of 100 stimuli presented in a pseudo-randomized order. Each block lasted about 3 minutes. Stimulus duration was 500 ms and was presented centrally on a black background. Inter-stimuli interval was varying between 500 ms and 1 s. Participants were asked to count the incoherent pictures. A short pause between each block was offered to participants. Data processing EEG processing. EEG data were zero-phase-filtered in the interval  Hz through a 4 order Butterworth filter avoiding phase distortion. Channels in the cheeks and in the neck were discarded (204 channels left). EEG epochs were then extracted from the continuous dataset and time-locked from -500 to 500 ms relative to the onset of each image whereas the 30-s recorded before and after the dual-task were divided in non-overlapping epochs of 1 s. Noisy channels were identified by visual inspection and were interpolated using the nearest-neighbor spline method (average percentage of channels interpolated: 1.5%). Individual epochs containing non-stereotyped artifacts, peri-stimulus eye blinks and eye movements (occurring within ± 500 ms from stimulus onset) were also identified by visual inspection and removed from further analysis (average percentage of epochs removed: 10%). Data were cleaned from remaining physiological artifacts (eye blinks, horizontal and vertical eye movements, muscle potentials and other artifacts) through a Principal Component Analysis (PCA)-informed Independent Component Analysis (ICA) algorithm implemented in EEGLAB (average percentage of components removed: 9.1%). We applied the LAURA algorithm implemented in Cartool 44 to compute the source reconstruction taking into account the patient's age to calibrate the skull conductivity 454647 . The method restricts the solution space to the gray matter of the brain. Then, the cortex was parcellated into 45 brain regions of interest (ROIs) 48  Lastly, we performed two-sided Wilcoxon rank sum tests on the relative power spectra values for each ROI under the same postural/dual-task condition between the two groups (i.e., elderly and younger adults) to find the significant different activation between elderly and younger adults. Moreover, we performed two sided paired samples Wilcoxon signed rank tests inside the same group to compare relative power spectra values for each ROI between the different postural/dual-task condition to emphasize the seeds of each task. In particular, as input of the statistical tests, we used the relative power spectral values during the baseline condition (i.e., simple standing in front of a black screen) and we computed the difference between the relative power spectral value post-stimulus (i.e., 500 ms after the stimulus onset) and pre-stimulus (i.e., 500 ms before the stimulus onset) for the odd and frequent stimuli.
EMG processing. EMG data (time-locked with the EEG recordings) were zero-phase-filtered in the interval  Hz through a 4th-order Butterworth filter. EMG segments containing artifacts were identified by a threshold (|EMG| > 5 * standard deviation) and removed from further analyses (percentage of data-points removed: < 3.5 % for the elderly; < 3% for the younger adults). We computed the Root Mean Square (RMS) value for each EMG signal. In order to compare the muscular activity between the two groups (i.e., elderly and younger adults), the RMS of each muscle activation during each task was divided by the within-group median RMS value of the same muscle while performing a Maximum Voluntary Contraction (MVC). Due to constraints of the experimental setup, MVCs were performed with the participants standing in the shielded room and equipped with the EEG cap. Not all the participants were able to perform a true MVC and normalizing the RMS within-subjects would lead to unreliable results. However, RMSs obtained from each muscle and within groups showed low variability and we can trust median values to be representative of the muscular activity for elderly and younger groups. We then compared the normalized RMS value of each muscle under the same postural condition (i.e., 0%, ±22%, core board) between the two groups through two-sided Wilcoxon rank sum tests and among all the four different postural conditions within each muscle and group through paired samples two sided Wilcoxon signed rank tests.  . EEG power spectra during dual-tasking in static (0%) and dynamic (core board) balance in the elderly. Median values computed among elderly for each ROI represented by a sphere centered on the cortical region, whose radius is linearly related to the magnitude. Such information is also coded through a color scale. We reported the difference (∆) of the power spectra between the 0.5 s-epoch post-odd stimulus and the 0.5 s-epoch pre-odd stimulus in the same postural conditions of Figure 2. The p-values of the statistical significant differences between a) and b) condition are reported in c). Each p-value < 0.05 is represented by a sphere centered on the cortical region, whose radius is linearly related to the magnitude, coded also through a color scale. 14/15 Figure 5. EEG power spectra during dual-tasking in static (0%) and dynamic (core board) balance in the younger adults. Median values computed among younger adults for each ROI represented by a sphere centered on the cortical region, whose radius is linearly related to the magnitude. Such information is also coded through a color scale. We reported the difference (∆) of the power spectra between the 0.5 s-epoch post-odd stimulus and the 0.5 s-epoch pre-odd stimulus in the same postural conditions of Figure 2. The p-values of the statistical significant differences between a) and b) condition are reported in c). Each p-value < 0.05 is represented by a sphere centered on the cortical region, whose radius is linearly related to the magnitude, coded also through a color scale. Figure 6. Statistics in elderly vs younger adults during odd stimuli in static (0%) and dynamic (core board) balance. The p-values of the statistical significant differences between the younger and the elderly group during a) visual odd stimuli and static balance (0%) and b) during visual odd stimuli and dynamic balance (core board). Each p-value < 0.05 is represented by a sphere centered on the cortical region, whose radius is linearly related to the magnitude, coded also through a color scale.