**3.2.4 Task paradigm**

RW, VW and WO tasks were performed 40 s every time, with a 30 s of rest before and after each task. In order to avoid the influence by the order of the tasks, two procedures were carried. Procedure 1 was performed in the following order.

Fig. 14. Walking videos shown to the subjects during imaginary walking

Rest (30s) →VW (40s) →Rest (30s) →RW (40s) → Rest (30s) →WO (40s) →Rest (30s) →RW (40s) →Rest (30s)

Procedure 2 was performed in the following order.

Rest (30s) →RW (40s) →Rest (30s) →VW (40s)→ Rest (30s) →RW (40s)→Rest (30s) →WO (40s) →Rest (30s)

#### **3.2.5 Data analysis**

144 Infrared Spectroscopy – Life and Biomedical Sciences

CH6, 7, 10, 13, 14 covered PM, SMA and the primary motor cortex (M1) which plans and

There were three tasks, real walking (RW), virtual walking (VW), and walking observation (WO) in the experiment. During the RW task, the subjects walked on a treadmill at 1.0 km/h. For the VW task, a video clip was taken by a cameraman who was naturally walking down a corridor at 1.0km/h (Fig.14). Subjects were instructed to imagine as if they were actually walking from the first-person perspective. For the WO task, a video clip in which a person walks down the same corridor at 1.0 km/h was taken (Fig.14). Subjects were instructed to imagine that they were walking the same as the person in the video, especially

(a) Shooting location of the video

(b) Scene for walking observation. (c) Scene for virtual walking.

The subjects wore a video eyewear (Wrap920, Vuzix Corporation, NY, USA) with lightshield to block distracting outside light (Fig. 12), to display the video stimuli. This eyewear displays the video on a virtual 67-inch screen as seen from 10 feet. During the rest period between the

RW, VW and WO tasks were performed 40 s every time, with a 30 s of rest before and after each task. In order to avoid the influence by the order of the tasks, two procedures were

Fig. 14. Walking videos shown to the subjects during imaginary walking

carried. Procedure 1 was performed in the following order.

tasks, a video clip in which there is a cross mark on a white board was displayed.

executes movements related to walking.

to match their gaits to the video.

**3.2.4 Task paradigm** 

**3.2.3 Stimuli** 

Oxy-Hb was considered as an indicator of changes in blood volume. The measurements of the subjects were checked visually for artifacts due to body movements. Obtained data were analyzed by calculating the average oxy-Hb level during RW, VW and WO. In the analysis we compared the average oxy-Hb of CH4, 5, 6, 7, 8, 10, 11, 12, 13, 14 which covered PM, SMA and M1 involved in planning and execution of movements related to walking.

### **3.2.6 Results and discussions**

In Fig.15 and Fig.16, the color maps of typical activation pattern of oxy-Hb were extracted. These maps were created by interpolation to the measurement data of 24 measure points. It

Fig. 15. Color maps of typical activation pattern in procedure 1

Fig. 16. Color maps of typical activation pattern in procedure 2

Comparison of Cortical Activation During Real Walking and Mental Imagery of

the brain to the stimuli.

**5. Acknowledgment** 

WO.

23700316).

**6. References** 

pp.3770-3774, ISSN 0027-8424.

pp.1167-1177, ISSN 0031-9023.

0304-3940.

Walking – The Possibility of Quickening Walking Rehabilitation by Mental Imaginary of Walking 147

fNIRS measurement is comfortable for subjects, since it requires less constrictive circumstances of measurements and fewer movement restrictions. These advantages allow us to measure the cortical activation during real walking. The experiment results showed that the oxy-Hb during the mental imagery of walking task (VW and WO) was higher than that during the RW task. No significant difference was observed in the oxy-Hb during VW and WO, showing that they have the similar effect on the brain regions we measured. The importance of stimulus diversity was suggested in continuous stimulation of mental imagery of walking because the oxy-Hb level decreased with the adaptation of

On the other hand, the disadvantage of fNIRS is that it can only monitor limited brain areas and the spatial resolution is relatively low. fNIRS can only measure cerebral blood flow in the cortices, not in deeper structures. We need another method to find out how the other neural systems, such as cerebellum, the spinal cord and the peripheral nervous system, involve in mental imaginary of walking. Recently, a lot of studies which combined fNIRS with EEG (Wallois et al., 2011; Fazli et al., 2011) or fMRI with EEG (Horovitz et al, 2008; Mantini et al., 2010) have been reported. These combinations enable to make the best use of the advantages of each technology. The advantage of simultaneous EEG-fNIRS measurement is to provide a better temporal resolution and quantitative information about oxy-Hb and deoxy-Hb. The combination of EEG and fMRI is to make use of the high temporal resolution of EEG and the high spatial resolution of fMRI. A multimodal approach is necessary for better understanding the similarity and difference between RW, VW and

This study was supported by Grants-in-Aid for Scientific Research No. 20240058 and No. 21300212 from the Japan Society for the Promotion of Science. It was also supported by a Grant-in-Aid for Young Scientists (B) from Japan Society for the Promotion of Science (No.

Chance, B., Zhuang, Z., UnAh, C., Alter, C., & Lipton, L. (1993). Cognition-Activated

Dickstein, R., Dunsky, A., & Marcovitz, E. (2004). Motor Imagery for Gait Rehabilitation in

Fukuyama, H., Ouchi, Y., Matsuzaki, S. Nagahama, Y., Yamauchi, H., Ogawa, M., Kimura,

Fazli, S., Mehnert, J., Steinbrink, J., Curio, G., Villringer, A., Müller, K. R., & Blankertz B.

*NeuroImage*, (August 2011), doi:10.1016, ISSN 1053-8119.

Lowfrequency Modulation of Light Absorption in Human Brain. *Proceedings of the National Academy of Sciences of the United States of America*, Vol.90, No.8, (April 1993),

Post-Stroke Hemiparesis, *Physical Therapy*, Vol.84, No.12, (December 2004),

J., & Shibasaki, H. (1997) Brain Functional Activity during Gait in Normal Subjects: a SPECT study. *Neuroscience Letters*, Vol.228, No.3, (June 1997), pp.183–186, ISSN

(2011). Enhanced Performance by a Hybrid NIRS–EEG Brain Computer Interface,

can be observed from the experiment results that both in procedure 1 and procedure 2, oxy-Hb increased significantly no matter the subject really walked or just imagined walking in first-person perspective and third-person perspective. The experiment results suggest that the cortical areas related to walking was activated by both real walking and imaginary walking. The oxy-Hb in procedure 2 was generally lower that in procedure 1. The same video clips were used in procedures 1 and 2. The adaptation of the subject to the tasks might be the reason for the decrease in oxy-Hb.

The average oxy-Hb of the 4 subjects during the tasks in procedures 1 and 2 are listed in Table 2. In procedure 1, for all the subjects, oxy-Hb during VW and RW were higher that that during the RW before them. In procedure 2, although the oxy-Hb was generally lower than procedure 1, the same pattern (oxy-Hb were higher during VW and RW than the RW after them.) was observed except one subject QQ with oxy-Hb during WO of -0.054 and oxy-Hb during RW of -0.031. We can conclude from Table 2 that although individual difference was significant, on the average, the oxy-Hb levels during VW and WO were higher than the RW, regardless of the order of the tasks. There was no significant difference observed in the oxy-Hb during VW and WO. Whether the subjects imagined from first-person perspective or from third-person perspective did not lead to significant difference in cortical activation of the brain regions we measured.


\*\*oxy-Hbs are given in mM·mm

Table 2. Average oxy-Hb during RW, VW and WO tasks

### **4. Conclusion**

In our study, in order to verify the possibility of conducting neurological rehabilitation by mental imagery of walking and to find an effective way to activate the motor area in mental imagery, we compared the activation in motor areas during RW, VW, and WO, making use of the advantages of fNIRS. fNIRS measures cerebral hemodynamics noninvasively by monitoring the attenuation of near infrared light passing through tissue. fNIRS measurement is comfortable for subjects, since it requires less constrictive circumstances of measurements and fewer movement restrictions. These advantages allow us to measure the cortical activation during real walking. The experiment results showed that the oxy-Hb during the mental imagery of walking task (VW and WO) was higher than that during the RW task. No significant difference was observed in the oxy-Hb during VW and WO, showing that they have the similar effect on the brain regions we measured. The importance of stimulus diversity was suggested in continuous stimulation of mental imagery of walking because the oxy-Hb level decreased with the adaptation of the brain to the stimuli.

On the other hand, the disadvantage of fNIRS is that it can only monitor limited brain areas and the spatial resolution is relatively low. fNIRS can only measure cerebral blood flow in the cortices, not in deeper structures. We need another method to find out how the other neural systems, such as cerebellum, the spinal cord and the peripheral nervous system, involve in mental imaginary of walking. Recently, a lot of studies which combined fNIRS with EEG (Wallois et al., 2011; Fazli et al., 2011) or fMRI with EEG (Horovitz et al, 2008; Mantini et al., 2010) have been reported. These combinations enable to make the best use of the advantages of each technology. The advantage of simultaneous EEG-fNIRS measurement is to provide a better temporal resolution and quantitative information about oxy-Hb and deoxy-Hb. The combination of EEG and fMRI is to make use of the high temporal resolution of EEG and the high spatial resolution of fMRI. A multimodal approach is necessary for better understanding the similarity and difference between RW, VW and WO.
