**4. Discussion**

Spasticity is the most common feature of cerebral palsy and occurs in approximately 85% of children with cerebral palsy, and correct assessment and treatment of spasticity are considered important. [20]. Most children with cerebral palsy spasticity have an asymmetrical body and disordered balance, leading to difficulties in daily activities [21]. In this study, changes in spasticity before and after training were confirmed by an ankle joint test and knee joint MTS. Curtis et al. (2014) [22] used an interactive dynamical stander to study low lateral flexor muscle spasticity of the ankle in children with cerebral palsy. Six children with cerebral palsy aged 4–10 years who were GMFCS level I–III were trained to activate ankle dorsi flexion for 10 weeks through ankle movements of the interactive

dynamic stander of computer games. As a result, it was noted that the median active and passive dorsi flexion of the ankle increased by 5◦ and 10◦, respectively; therefore, this training could function as a new clinical conservative treatment of ankle flexion in cerebral palsy children. In the right ankle joint contraction test in spasticity of ankle joint, both groups had a significant increase from before training to after training (*p* < 0.05), and in the left ankle joint contraction test, the AOT and control group increased from before training to after training (*p* < 0.05). Both groups' range of motion of ankle was significantly increased because of symmetrical weight-bearing during the intervention and its stimulation of proprioceptive sensibility. Improvement in the range of motion of the ankle, which is a functional ankle movement, is possibly due to the voluntary forward, backward, left, and right weight shift movements of the lower extremity during the stage 1 to 3 in AOT. Increase in the range of motion of the ankle in the control group and the exercise program included sitting and standing up, weight-bearing, and weight movement training in the straight posture.

In the MTS used to examine the degree of spasticity, the right knee joint of the action observation training participants was 4.78◦ before training and 2.91◦ after training. The knee joint decreased from 3.77◦ before training to 1.91◦ after training, which is consistent with the results of previous studies and suggests that there is a connection between construction and rigidity. These results indicate that the range of motion of the ankle joint is increased with regards to ankle construction, and the action observation training program images provide the children to simultaneously watch the full-motion images and the enlarged images of the movements, which are an important part of the joint's movement, were shown enlarged of children with diplegic cerebral palsy.

GMFM-88 is used as a most common tool to evaluate the function of children with cerebral palsy and down's syndrome [23] by measuring and recording changes in exercise levels over time or over treatment outcomes [24]. In the current study, the B scale of the AOT was 93.33% before training and 98.45% after training, the C scale was 88.36% before training and 94.92% after training, the D scale was 60.68% before training and 77.48% after training, and the E scale was 46.26% before training and 52.77% after training. There was a significant di fference before and after training (*p* < 0.05), and also in the B scale, C scale, D scale, and E scale in the control group (*p* < 0.05). This was supported by the results of the AOT group in this study since a more significant change in the GMFM-E results was observed compared to the control group. Although conducted in a relatively short amount of time, the repeated viewing of the videos (including shifting the body weight forward and backward), standing movements to improve balance in the third stage, and walking sideways in the fourth stage allowed for an easier understanding of the specific positions for each action and their order. This increased the exercise's learning e ffects.

Park et al. reported [25] the e ffects of horseback riding treatment on gross motor function and functional performance in children with sti ff cerebral palsy, and demonstrated a significant di fference between the experimental group and the control group. Mahasup et al. studied [26] the e ffects of motor observation on 30 children with sti ff diplegia cerebral palsy by applying action observation training for 2 months. In total, 15 control groups received general physical therapy that included the Bobath concept once a week, and 15 experimental groups performed action observation training three times a day; they demonstrated a significant di fference in running and running and jumping between the two groups. The results of the current study are in agreemen<sup>t</sup> with previous studies and confirm that action observation training shows positive e ffects in improving GMFM. Therefore, action observation training improves the function of participants according to exercise level, and the action observation training was considered to contribute to the improvement of daily life skills and mobility of spastic bilateral lower limbs in cerebral palsy.

Children with cerebral palsy often have di fficulty with sitting posture balance and have unstable postures, such as asymmetrical trunks and bending [27]. In addition, children with cerebral palsy have reduced movements in the trunk, pelvis, and lower extremities, so they stand and walk in unprepared condition, raising the upper extremities or excessively extending the upper bodies to compensate for insu fficient antigravity activity [28]. Auld and Johnston (2014) [29] investigated the e ffects of an 8-week local community-based strengthening and balance exercise group on exercise in children with cerebral palsy. Five children with spastic diplegia cerebral palsy and five children with hemiplegia cerebral palsy participated in the study, and the results demonstrated that the participants' balance ability increased significantly in Movement Assessment Battery for Children and anterior and lateral extension (*p* < 0.05). In the current study, the right side stretch of the AOT increased from 15.52 cm before training to 18.8 cm after training, the left side of the side stretch increased from 14.34 cm before training to 18.42 cm after training, the right side of the forward stretch increased from 22.78 cm before training to 27.38 cm after training, and the left side of the forward stretch increased from 22.6 cm before training to 26.85 cm after the training. The improvement of the arm stretch test indicates an improvement in balance, and it is thought that the movement observation training contributed to uniform weight bearing, postural alignment, and the ability to change direction by improving the muscles. There was also a significant difference before and after training in the two groups (*p* < 0.05); this may be due to the effect of action observation training on brain activity in the primary motor area, and the activation of cognitive activities related to motor memory formation and understanding of other people's behavior through imitation [30].

This study has the following limitations: this study has a short 6-week intervention period, and this study comprises a small sample size. This makes it difficult to generalize the findings to all children with CP. It is also difficult to control all of the factors that might affect the child's hormones affect because of puberty, scoliosis or hip problems, any influence of bracing on knee or ankle range of motion and spasticity, and activities of daily living. Furthermore, several of the participants had a short attention span while concentrating on the action observation, making it difficult for the treatment to last long as planned. This could be explained by the fact that children with spastic CP are not only impaired by the regulation ability of the muscular system and sensory deprivation, but are also deprived of cognitive function [31]. This study confirmed that action observation training has shown positive effects in improving the spasticity of ankle joint, gross motor function, and balance in children with cerebral palsy.

**Author Contributions:** Conceptualization, Y.-a.J., and B.-H.L.; Data curation, Y.-a.J.; Methodology, Y.-a.J., and B.-H.L.; Project administration, B.-H.L.; Supervision, B.-H.L.; Visualization, Y.-a.J.; Writing–original draft, Y.-a.J.; Writing–review & editing, B.-H.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This study was supported by a gran<sup>t</sup> from the NRF (NRF-2018R1D1A1B07045746), which is funded by the Korean government.

**Conflicts of Interest:** The authors declare no conflict of interest.
