Extended Abstract
Introduction
Vision, as a gateway for information to enter the brain, plays the most important role in facilitating motor control in the surrounding environment. 70 % of sensory receptors are associated with vision and, in the form of a dynamic system, play an important role in integrating information, shaping concepts, planning, environmental adaptation, and cognition [1, 2]. According to the International Classification of Diseases (ICD), Visual function is divided into four categories: normal vision, moderate visual impairment, severe visual impairment, and blindness. People with moderate to severe visual impairment have low vision (acuity less than 6.18 and up to 3.60), and anyone with a visual acuity of less than 6.18 after treatment has vision impairment [3]. According to the World Health Organization in 2017, there are 253 million visually-impaired people in the world [3]. 
Perceptual-motor activities through sensory integration processes improve motor and learning skills and allow a person to process sensory information with greater cognition [4]. Perceptual-motor skills play an important role in perceiving, interpreting, and responding to sensory stimuli, and allow for progress in each of the perceptual and motor components. The integration of different senses with each other in the form of perceptual-motor activities can provide the necessary motivation and ability to improve gross motor skills. Since passing the sensory-motor and perceptual stage is necessary to enter the cognitive and learning stage, if a child has appropriate perceptual-motor skills, the correct body image is formed in him/her and success in performing activities create confidence and improve cognitive activities [5]. 
In children with visual impairment, reduced vision stimulation limits the acquisition of motor experiences, including gross motor skills, which affects their perceptual-motor skills [6]. Numerous studies have examined the developmental delays of gross, fine, and perceptual-motor skills of children with visual impairment [7-10]. However, few studies have been conducted in the field of perceptual-motor intervention of these children and no specific framework for training has been provided [11, 13]. Since motor problems are very important in various aspects of development, disability is more pronounced in these aspects [5, 6, 12]. Studies have shown the effectiveness of Johnstone and Ramon Perceptual-Motor program on improving motor skills and the reading ability of elementary school children [14]. This study aimed to examine the effect of this program approved by a panel of experts, on improving the gross motor skills of children aged 7-10 years with visual impairment.
Materials and Methods
In this randomized controlled clinical trial, study population consisted of all children aged 7-10 years with visual impairment studying in three schools of Narjes (Girls’ school), Shaid Mohebbi and Shahid Khazaeli (Boys’ schools) in Tehran, Iran in 2018 (n=140). Of these, 24 were met the inclusion criteria which were: Diagnosis of vision impairment by a doctor or optometrist according to the children’s medical records (no absolute blindness or only light perception), parents’ consent, lack of physical, sensory, cognitive or motor impairment, and no participation in other clinical interventions at the same time. Due to the lack of such interventions in these children, we first examined the treatment protocol according a panel of experts [13] included four faculty members of the Department of Occupational Therapy, University of Social Welfare and Rehabilitation Sciences (expert in perceptual-motor issues), and one optometrist (a low vision specialist).
Due to the lack of time, the program was provided to the panel for 2 months, three 45-minute sessions per week (24 sessions, in total) based on the perceptual-motor model and the Johnstone & Ramon protocol [14]. The trainings were prepared for the intervention phase by applying the opinion of experts (attention to the exact number of repetitions in each set, contrast, size of the objects and their sound) and reaching an agreement above 75% [15] (Table 1). 
The examiner, who was unaware of the study groups, after receiving instructions by the researcher on how to perform the tests, measured motor skills of children using Test of Gross Motor Development-2 (TGMD-2) and Movement Assessment Battery for Children-2 (MABC-2). Then, they were divided into two groups of intervention (n=12) and control (n=12) who were matched for gender [16], age, visual acuity, and motor skills scores [7]. After 8 weeks of consecutive interventions and data collection, the second assessment phase was performed for both groups. One month after the end of the intervention, a follow-up was conducted (Figure 1).
The TGMD-2 is a qualitative tool that measures gross motor skills of children under two subscales of “locomotor” and “object control”, and is used for children aged 3-11 years. The child performs each item twice and its performance is scored by 0 or 1 (a total score of 24 for each skill). The time taken to complete the test is 15-20 minutes [17]. Its psychometric characteristics for children with visual impairment have already been examined [18]. The MABC-2 is a quantities tool that measures motor proficiency of children under three subscales of manual dexterity, balance, and aiming/catching. It is used for children aged 3-16 years. Children with a total score < 5 have severe motor problems and a score of 5-15 indicates the at-risk children. The estimated time for its completion is 20-30 minutes [19]. Its psychometric characteristics for children with low vision have already been examined [20].
Results
According to the results of Shapiro-Wilk test, the distribution of data was normal (p<0.05). Hence, we used non-parametric tests including Friedman test, Mann-Whitney U test, Bonfrroni test, and Generalized Estimating Equation (GEE). The significance level was set at 0.05. The two groups of study were the same in terms of gender, visual acuity and school (p>0.05). The mean age of the children was 9.2±1.1 years in the intervention group and 8.3±1.2 years in the control group. Most of the participants in the intervention and control groups were boys (66.7%) and girls (58.3%), respectively. The visual acuity of the most of them (58.3%) was moderate, and they were from Narjes elementary school. The TGMD-2 and MABC-2 scores of participants at three phases of Pre-test, post-test, and follow-up are presented in Table 2. 
In terms of group, the mean of aiming/catching and object control was significantly different between groups at all measurement phases (p<0.05), where it was higher in the intervention group. In terms of time, there was no significant difference in aiming/catching before intervention (p>0.05), but it was significant at the post-test and follow-up phases (p<0.05). 
Regarding the balance and object control skill, there was a significant difference only in the intervention group at all measurement phases (p<0.05). In terms of time, there was no significant difference between groups before intervention (p>0.05), but it was significant at the post-test and follow-up phases (p<0.05). Regarding the locomotor skills, there was also a significant difference in the two groups at all measurement phases (p<0.05). In terms of time, there was no significant difference between groups at Pre-test and follow-up phases (p>0.05). Only at post-test phase a significant difference was observed. 
Due to the abnormality of the changes, the GEE model was used to investigate the response changes. As presented in Table 3, Bonfrroni test results showed that the mean of aiming/catching skill was significantly different between Pre-test and post-test and between Pre-test and follow-up phases in the intervention group (P<0.05), but not between post-test and follow-up phases (p>0.05). In the control group, its mean score was significantly different between Pre-test and post-test and between post-test and follow-up phases (p>0.05), but not between Pre-test and follow-up phases (p>0.05). 
The means of balance and object control skills were also significantly different between Pre-test and post-test and between Pre-test and follow-up phases in the intervention group (P<0.05), but not between post-test and follow-up phases (p>0.05). In the control group, no significant difference was found between measurement phases (p>0.05). The mean of locomotor skill was not significantly different between any measurement phases in the intervention group (p>0.05). In the control group, it was significantly different only between Pre-test and follow-up phases (p<0.05). The reason for this discrepancy may be due to the high distribution of the locomotor variable in the control group and not due to the difference in mean values. 

Discussion
Active environments, practice, and learning opportunities are among the factors influencing motor development, especially in childhood, when the nervous system undergoes changes [21]. The aim of this study was to evaluate the effectiveness of a perceptual-motor therapy program on improving visually-impaired children’s gross motor skills. Our study revealed that the most of gross motor skills in children can be improved under the influence of this perceptual-motor program. Data analysis showed that the mean score of aiming/catching in the two groups was significantly different after the intervention and this score was significantly higher in the intervention group. 
This is consistent with the results of Aki et al. and Yousefian et al. reported for the catching ball variable [22, 23]. Ashrafy et al. reported improvement in fine motor, ball skills (P=0.000), static and dynamic balance which is in agreement with our study [24]. Aiming and catching skill, which requires eye-hand coordination, is greatly affected by vision, and repeating this skill while adapting the equipment and conditions can lead to maximum use of existing vision and thus improve this skill. 
Regarding the balance skill, our results can be in agreement with the results of Aki et al. and Sajedi and Barati in terms of balance maintenance and heel-to-toe walking [22, 25], and Yousefian et al. in terms of balance maintenance [23]. Our results in this area are also consistent with the results of Ashrafy et al., and Mavrovouniotis et al. [24, 26, 27]. According to Ayres [28], by training and repeating the activities that stimulate the utricle and saccule receptors of the ear, deep sensory receptors can be stimulated, which in turn enhances the balance skills.
The mean score of locomotor skill in the two groups did not differ significantly which is contrary to the findings of Aki et al., Yousefian et al., and Houwen et al. [22, 23, 29]. A possible reason for the discrepancy between our results and those of Yousefian et al. can be the high initial scores of locomotor skill in the current study and the consequent lack of significant change after the intervention, the use of different training periods, the difference in exercise type, age range and sample size. Houwen et al. reported that children who participated in sports programs apart from school programs earned higher locomotor skill scores than children with visual impairments who did not attend these programs. 
The children participated in a sports program 60-180 minutes a week in school. As a result, they received more exercises than the children in our study, which may be a reason for the difference in the results. The reason for the difference between the results of Aki et al. and this study can be the greater number of participants or its longer intervention period. It is important to note that locomotor activities requires less visual acuity than other gross skills and is less affected in children with visual impairment [29], so performing this intervention has not had a significant effect on improving this skill.
Another result of our study was that the mean score of the object control skill was significantly different after the intervention in the two groups, and this score was significantly higher in the intervention group. This is consistent with the findings of Aki et al. , Yousefian et al., and Houwen et al. [22, 23, 29]. According to studies, the development of fine and gross motor skills development in children with visual impairment is delayed due to limitations in receiving stimuli compared to normal peers [8]. The provision of perceptual-motor exercises leads to the growth and development of body image, spatial awareness and coordination of organs. As a result, the promotion of this skill leads to the improvement of motor quality. The integration of different senses with each other in the form of perceptual-motor activities can provide the necessary motivation and ability to improve gross motor skills [30].
Conclusion
The perceptual-motor exercises for children with visual impairments can improve their gross motor skills such as object control, aiming/catching, and balance. Improving these skills can facilitate their community participation and independence in daily activities, especially school activities that are essential at school age. This study faced limitations and problems such as limited number of samples, limited treatment protocols in the field of perceptual-motor for such children, limited time in holding treatment sessions and limited valid and reliable tools for these children. It is recommended that other treatment programs be designed and implemented to enhance the gross motor skills of children with different levels of visual impairment and age. Examining the psychometric properties of the instruments used for measuring these children in Iran is another recommendation of the present study.
Ethical Considerations
Compliance with ethical guidelines
This article approved in University of Social Welfare and Rehabilitation Sciences Ethics Committee (Code: IR.USWR.REC.1397.058)
Funding
The present paper extracted from the MSc. thesis of the fourth author Department of Occupational Therapy, University of Social Welfare and Rehabilitation Sciences.
Authors' contributions
Conceptualization: Hooshang Mirzaie, Seyed Ali Hosseini, Abbas Riazi, Fatemeh Ghasemi Fard; Methodology, validation, analysis: Hooshang Mirzaie, Samaneh Hossein Zadeh, Fatemeh Ghasemi Fard, Mehdi Jafari Oori, Narges Hooshmand Zadeh; Editing, drafting the final manuscript: Hooshang Mirzaie, Fatemeh Ghasemi Fard, Mehdi Jafari Oori.
Conflicts of interest
The authors declared no conflict of interests.
Acknowledgements
The authors would like to thank to Department of Occupational Therapy, University of Social Welfare and Rehabilitation Sciences. Also the authors would like to thank the members of the panel of experts, Dr. Abbas Riazi, Dr. Ebrahim Pishiareh, Dr. Nazila Akbar Fahimi, Dr. Saeed Fattahchi, Ms. Fatemeh Behnia and Dr. Behzad Amini Kamal and thank the principals, the school staff and all the students who participated in this study.