**1. Introduction**

Cerebral palsy is a non-progressive disorder that a ffects the development of the brain of fetuses or infants, and presents as limited activity, movement, and postural disorders [1]. Sti ff diplegia is a disorder that shows more dysfunction in the lower extremities than the upper extremities [2], while spasticity diplegia is characterized by ununiformed abnormal movements, unstable continuous movements, and large patterns of motion [3]. Cerebral palsy also leads to limits in balance due to muscle weakness in skeletal muscles, excessive reflexes, simultaneous contraction of agonist and antagonist muscles, delayed response of the ankle muscles, and shrink posture [4].

Balance is important for most functional skill movements; this includes the integration of sensory inputs to structure the body's perception of the center of gravity, and perform appropriate musculoskeletal responses to unexpected movements or to stabilize during moments of instability [4]. However, in cerebral palsy, when balance is a ffected, it increases compensation usage of the upper extremities, which is followed by restricted movement of the upper limbs. This may cause limitation in the function of the upper extremities [5], performance, and learning activities of daily life, as well as problems in movement and a limitation of social roles and community participation [6].

Spasticity means intermittent or persistently involuntary disordered sensory motor control caused by upper motor neuron lesions [7]. Prolonged spasticity causes abnormal posture, limitation of movement, and limitation and construction of active or passive joint movement [8]. In order to improve

the spasticity and balance in children with cerebral palsy, various treatment intervention methods have been used; these include botulinum toxin injection [9,10], anticipatory postural adjustments [11], dynamic ankle-foot orthosis [12], whole-body vibration [13], and extracorporeal shock wave therapy [14]. Recently, a new method of treating upper limb motor deficits using action observation training has been proposed for patients with stroke [15,16] and cerebral palsy [17–19]. Action observation training is a cognitive intervention technique that is used to improve and learn exercise skills in sports athletes, the general public, and patients with motor impairments. This training involves using the activity of mirror neurons with excitement characteristics when actually exercising or watching others perform tasks [16]. Various studies on movement observation training have been proposed, but most relate to restoring the upper limb function of stroke and cerebral palsy patients, and studies on the spasticity, gross motor function, and balance in cerebral palsy for movement observation training are insufficient. Therefore, this study aims to contribute to the improvement of rehabilitation in children with spastic diplegia cerebral palsy by verifying the effect of action observation training on the treatment of spasticity, gross motor function, and balance.

### **2. Materials and Methods**

The participants of this study were selected from 30 children who were diagnosed with diplegia cerebral palsy and undergoing physical therapy at K-hospital and E-center in Seoul. The specific selection criteria of the study subjects were children between 5 and 11 years old diagnosed with diplegic cerebral palsy, without visual impairment and visual field defects, able to follow the researcher's instructions, GMFCS (gross motor function classification system) level I–III, and with ankle dorsal flexors and plantar flexors better than poor + in manual muscle test. The parents of the children consented to their participation in this study after the purpose of the study was explained and they were informed that they could withdraw at any time. The exclusion criteria included children with a modified assessment scale (MAS) of 2 or more, children who have not had a seizure in the last 6 months, or those who received botulinum injections 6 months prior to the study. This study was conducted with the approval of the Research Institutional Review Board of Sahmyook University. The objective and the procedures to be performed in the study were fully understood by the subjects, and all participants' parents provided informed consent for inclusion in the study. Therefore, this study was based on the ethical principles of the Declaration of Helsinki.

The past history of the 30 children who agreed to the study was examined, and other orthopedic or neurological examinations were performed by the attending doctor before treatment. Of the 30 children at K-hospital, 3 children were under GMFCS level III and 1 child had a seizure within the past 6 months. A total of 22 patients were selected, with the exception of 2 with communication disorders and 2 children who had received botulinum injection 6 months prior. The selected 22 children were divided into either the action observation training group (AOT) at K-hospital or the control group E-center, which is a cerebral palsy treatment center nearby K-hospital for the blind, and each group included 11 participants. All subjects picked a go stone with black or white stone from a box containing 22 pieces of stone. The action observation training program was conducted three times a week for 30 min, for a total of 18 times, and general physical therapy was given 5 times/week for 30 min for a total of 6 weeks. One week before training and 1 week after training proceeded the evaluation. In the AOT, two children who could not participate in the experiment due to personal reasons and seizure dropped out of the control group, and two children who could not participate in the experiment due to personal reasons dropped out. Finally, each group included 9 children, and a total of 18 children were included in the experiment.

### *2.1. Action Observation Training*

In this study, action observation training (AOT) focused on spasticity of lower extremities, contracture, gross motor function measurement (GMFM), and balance ability. Children with cerebral palsy watched a video on a 42-inch screen, installed 1 m in front of their chairs, while sitting comfortably with their arms resting, but they were not allowed to physically follow the video or move. The model of the video's motion observation exercise was performed by a therapist who treated the child, and the training video consists of 4 stages that varied by di fficulty, and the video of each step was watched for the entire week. The participants watched a video of a task presented by a therapist, and after completing the assignment, they performed the steps, if a step was too di fficult to perform, retraining was conducted. The first stage consisted of movements to improve balance in the sitting position, the second stage consisted of sit-to-stand movements, the third stage consisted of standing movements to improve balance, and the fourth stage consisted of walking sideways (Table 1). The viewing time was 15 min, and 5 min of physical training was conducted with the therapist based on the content of the video, after 5 min of watching. In order to enhance the e ffectiveness of the action observation training, the participants watched the video at a designated time in a quiet place without noise. The children were instructed to concentrate on the video for 1 min intervals to allow for the attention span of children. Entire experiments were conducted by the same investigator from the beginning to the end of the experiment.

### *2.2. General Physical Therapy*

Neurodevelopment treatment is a 1:1 treatment between a patient and therapist. The participants received 6 weeks of general physical therapy, 5 times a week, for 30 min each session, according to the treatment schedule of the hospital. The exercise program included lying to sitting position, moving in the sitting posture, sitting and standing up, posture training for learning a normal gait pattern, weight bearing and weight movement training in the straight posture, walking training on the flat floor, and stair walking.

### *2.3. Outcome Measurements*

### 2.3.1. Spasticity of Ankle Joint

In this study, changes in the spasticity was used; ankle stiffness and Modified Tardieu Scale. An electronic joint goniometer (Gemred, China, 2014) was used to measure the ankle stiffness. In the supine position, the examiner extended the knee joint and examined the ankle stiffness in a relaxed state without muscle contraction. After the subject's heel fixed and dorsi-flexion with manual force as much as possible, it was maintained for 4 s in the end range. A Modified Tardieu Scale (MTS) was used to measure the muscle spasticity; the reliability for children with cerebral palsy was ICC = 0.54–0.95, which is defined as a high reliability [17]. The Tardieu scale can measure muscle spasticity by testing the response of the muscle to stretch at three types of velocity (i.e., slow as possible, speed of the limb segmen<sup>t</sup> when falling, and as fast as possible).

### 2.3.2. Gross Motor Function

The gross motor function measure-88(GMFM) is a tool for measuring and recording changes in an exercise level over time or as a result of treatment; the scores are ordinal on a 4-point grade scale after observing the movements of children with cerebral palsy. The evaluation items were composed of the following: A scale, lying and rolling position with 17 items; B scale, 20 items in the sitting position; C scale, 14 items in the standing position of the instrument and knees; D scale, 13 items in the standing position; and E scale, 24 items, including walking, running, and jumping activities. In cerebral palsy children, the inter-evaluator reliability was ICC = 0.929 and the test-retest reliability was ICC = 0.92–0.99 [18]. The results were compared using the sitting posture (B scale), device and knee standing posture (C scale), standing posture (D scale), and walking, running, and leap activity (E scale).

### 2.3.3. Balance Function Measurement

The pediatric arm stretch test is a modified version of the functional reach test (FRT) with forward stretch and side stretch in a sitting position. The test-retest reliability of cerebral palsy children was *r* = 0.54–0.88 and the inter-tester reliability *r* = 0.50–0.93 [19]. The side and forward distances of children from the sitting position were measured before and after intervention.

### 2.3.4. Data Analyses

SPSS ver. 21.0 was used to calculate the mean and standard deviation. The normality of variables was tested using the Shapiro–Wilks test, and all variables were normally distributed. General characteristics of the subjects were provided as descriptive statistics, and independent *t*-tests were conducted to identify the differences among the groups. Paired *t*-tests were conducted before and after the action observation training. Statistical significance was assumed when *p* < 0.05.
