The present pilot study has examined the transition from level walking-to-stair ascent of the elderly by comparing with young adults and investigated the effect of the VSS on the transition movement in the elderly and the changes that occur depending on the frequency of the VSS.
Participants in the present study are the same as in our previous study . Our previous study  examined the influence of local tendon vibration on postural sway during sit-to-stand movement, whereas the present study investigated the effect of the vibrotactile somatosensory stimulus on neuromotor control of the transition movement from level walking-to-stair ascent. That is, the topic, measure and analysis methods, the analyzed parameters and the motion in each study are different.
During the transition movements, in the first half of the stance, the elderly presented a higher function of the hip extensor to pull up compared to young adults. In the second half of the stance, dorsiflexion and plantar-flexor moment and power to the push-up was higher in the elderly than the young adults. When the VSS was applied, those results were further developed. The hip extensor function for the pull-up and the plantar-flexor function for the push-up increased. Consequentially, the extensor function of the lower extremity (i.e., support moment) increased during the entire stance phase. In addition, joint moment and power during the single-limb support phase were affected by the change in frequency of the VSS.
Features of the transition from level walking to stair ascent in the elderly
In the early swing phase, compared with young adults, the elderly showed higher dorsiflexion (AA_1), followed by higher hip flexion (HA_1) and dorsiflexion (AA_2). This could be the modality to pass a step of the staircase safely. In other words, it may be a strategy to ensure toe-clearance to prevent tripping over the staircase.
After that, lower dorsiflexion (AA_3) appeared in the elderly compared to young adults. It is possibly being used to clearly land the foot on the staircase through more ankle extension than the young adults. Consequently, ankle extension induces smooth foot landing, resulting in a decreasing impact force from the step during the weight acceptance.
In the stance phase, both groups used dominantly the extensor of the hip and knee joint to the lift body. However, the modality of use was opposite between the two groups. The elderly predominantly used hip extensor, whereas the young adults used the knee extensor. This can be attributed to weakness of the quadriceps femoris. It is a well-known fact that the elderly have lower muscle strength in the lower extremity compared with young adults. In particular, Hortobágyi et al.  reported that older adults had 60% lower maximal leg press moments compared with young adults. Due to this, the support moment could appear low in the elderly (SM_1). This could induce more activity of the hip extensor to compensate for the weakened knee extensor, and in the case of the plantar-flexor, it contributes to some compensation. These features could provide a clinical reference for training the quadricep muscles in the elderly.
Although joints of the hip and knee extended continuously in both groups, dorsiflexion decreased by plantar-flexor activity (AP_1) in young adults, whereas it was sustained in the elderly. This may be a strategy of the elderly to ensure stability while ascending the stairs. First, sway of the shank can be minimized by fixing the ankle joint. Second, if the ankle joint is extended, the position of the center of mass (COM) increases, which may cause postural instability due to COM acceleration. Finally, knee extension that is created by pulling the shank back, which contributes to the forward acceleration of the COM, can be prevented by restricting activity of the plantar-flexor.
In the late stance phase, dorsiflexion and knee flexion were higher in the elderly than in the young adults. This can contribute to a reduction in COM height, foot landing of the contralateral leg, and weight shift to the opposite leg. For this, the support moment (SM_3) increased more, which was attributed to the extensor moments of the hip and knee joints. This is a modality for securing energy for the subsequent push-up and compensating for the weakened quadriceps femoris in the opposite leg.
At the end of the stance phase, the plantar-flexor positive power is greater in the elderly. This was attributed to the larger dorsiflexion just earlier, and was a strategy to counter the more flexed dorsiflexion and assist the pull-up function of the proximal extensor of the opposite leg.
In summary, the elderly tended to use more hip extensors than young adults to lift the body, and to presumably ensure stability during locomotion, the extension of the ankle joint was restricted. In addition, the plantar-flexor power increased by increasing the flexion of the distal segments, possibly to assist the pull-up function of the opposite leg.
Effect of VSS on the transition from level walking to stair ascent in the elderly
In the swing phase, after toe-off, dorsiflexion (i.e., AA_1) increased, followed by a decrease in knee joint flexion. Toe clearance was ensured more by the increased dorsiflexion, and the decreased knee joint flexion could have contributed to control foot-landing by compensating for increased dorsiflexion. Consequently, the VSS induced further toe clearance, resulting in more stable stepping and more cautious foot contact.
In the stance phase, the moments and positive powers of the hip extensor increased. This could indicate that the function of the hip extensor to lift-up the body was enhanced by the VSS and that features of the elderly in the pull-up phase were also developed. In addition, the support moments increased. Hence, the results indicated that VSS increased function of the body support during the single-limb support phase and contributed to the pull-up of the body by activating the hip extensor. Furthermore, moments of the plantar-flexor increased slightly during the first half of the stance phase, which may have contributed to an increase in the support moments. Consequently, VSS contributed to an increase in the body support function while transitioning from the double limb support stance phase to the single-limb support stance.
In the second half of the stance phase, an increase in the knee joint flexion, dorsiflexion, and support moments appeared; that is, the VSS possibly further reinforced the body support (i.e., an increase in the SMs and negative plantar-flexor powers) and contributed to increasing the push-up of the plantar-flexor and assisting the pull-up of the contralateral limb.
These results indicated that the VSS could further develop characteristics of the transition movement in the elderly. In summary as follows:
The VSS possibly ensures safer toe clearance by increasing dorsiflexion, enhancing lift-up function of the hip extensor, increasing body support during the single-limb support phase, and assisting in the push-up of the plantar-flexor and the pull-up of the contralateral limb.
Despite the listed effects of VSS, the duration of the transition can be lengthened due to the increase in joint flexion and support moment. With increasing joint flexion, more extensor activity and antagonist muscle activation are required to prevent lower-limb collapse, and to restrict joint movement, respectively. These are factors that increase the joint stiffness. Thus, future studies should include spatiotemporal variables and electromyography (EMG).
Parameters depending on the frequencies of the VSS
For the hip joint angle, although there were no significant differences under non-stimulus and stimulus conditions, HA_2 had a tendency to increase. For the knee joint angle, there was a significant difference between non-stimulus and stimulus in KA_1 and KA_3, despite no difference between frequencies. For the ankle joint angle, there was a statistically significant difference between non-stimulus and stimulus in the AA_1 and AA_4. In particular, statistical significance was found between 180 Hz and 250 Hz for AA_4.
Overall, the VSS can be considered to cause a change in kinematics, and the trend seems to be common at all frequencies. This common trend may have resulted from the fixed dimensions of the staircase. However, differences depending on frequencies may occur in kinetic parameters considering that the joint angles of each stimulus condition were different.
As shown in Table 2, statistically significant differences between the non-stimulus and stimuli appeared between each frequency pair. Furthermore, significant frequency-dependent differences were found in the support moment, a sum of extensor moments of the lower limb joints [36, 38], as shown in Table 4. Steyvers et al.  measured motor evoked potential (MEP) depending on vibration frequency and revealed that frequency-dependent effect appeared in the MEPs. Therefore, it is indicated that the VSS affects joint kinetics while transitioning from level walking to stair ascent, and thus it can be presumed that joint kinetics can be regulated depending on the VSS frequency; that is, the VSS could influence neuromotor control (NMC). As a result, performance degree in sub-biomechanical tasks (i.e., weight acceptance, pull-up, body support, and push-up) during transitioning from flat ground walking to stair climbing will be affected.
There were significant differences between each frequency pair in the KNP for foot landing, the HPP_1 for the pull-up, the SM for preventing lower-limb collapse, and the ANP for ensuring stability, supporting the lower-limb and assisting the opposite leg. Although not all frequencies, this indicated that the same function is performed during the transition, but the degree varies with the vibration frequency; that is, it suggests that the NMC of the transition is frequency dependent.
In conclusion, the VSS has the possibility to control the degree of performance of sub-biomechanical tasks by affecting the NMC. It can further develop the characteristics of the transition movement in the elderly. For more insight, future investigations on EMG and muscle synergy are required.
Role of VSS in assisting level walking-to-stair ascent and its application
In this study, it was found that the elders’ characteristics of the transition motion differed from those of young adults. The predominant characteristics were larger hip joint flexion (HA_1), sustained dorsiflexion and larger dorsiflexion (AA_4), higher moment (HM_2) and power (HP_2) of the hip extensor, higher power of the plantar-flexor (AP_3) and the lower support moment (SM_1). These characteristics of the elderly might be an adaptation of motor control strategy according to neuro-physiological changes due to aging.
When the VSS was applied, some characteristics (i.e. AA_4, HM_2, HP_2, and SM_1) and unexpected parameters (i.e. AM_2, AM_3, AP_2, SM_2, and SM_3) were affected; that is, the function of the hip extensor in the first half of the stance phase, function of the plantar-flexor in the second half of the stance phase, and support moment in the entire stance phase. Consequently, the VSS enhanced motor control of the transition movement in the elderly, and motor control can be regulated depending on the frequency of the VSS as shown in Tables 1, 2, 3 and 4.
These findings suggest that the VSS can be utilized as a means of assisting the transition movement in the elderly. To apply the VSS, various commercial linear actuators can be used. The linear actuator used in the present study was small. Its diameter and thickness were 9 mm and 3.4 mm, respectively. In addition, to detect the transition movement or any motions, an inertial measurement unit (IMU) device and a device combined with an accelerometer and gyroscope can be used.
There is already a variety of sensors and methodology to detect and recognize human movement. In other words, if the VSS, IMU, power supply, communication module, and regulator for voltage and frequency are combined and miniaturized, the assistance device can be realized. Park et al.  and So et al.  used inertial sensors to detect gait events and small vibrators to apply somatosensory stimulation, and suggested an algorithm for detecting and stimulating. In addition, portable devices that detect motion and apply stimulation have been used in various clinical studies [42,43,44]. Therefore, the results of this study can be sufficiently utilized as clinical reference to assist the elderly in their movements, and support the design of assistive devices to enhance motor control by combining the above-mentioned devices and methodologies.
This pilot study has some limitations. 1) the number of participants in this study was small and the sample size between both groups did not the same. 2) the steady-state stair ascent was not investigated in this study. That is, further insight into elders’ stair walking, it is required that more sample size and comparing the transition from level walking-to-stair ascending with the steady-state stair ascent.