Understanding Foot Loading and Balance Behavior of Children with Motor Sensory Processing Disorder

Sensory processing disorder (SPD) could influence the neuromuscular response and adjustment to external sensory discrimination and lead to disruptions in daily locomotion. The objective of the current study was to compare plantar loadings and foot balance during walking, running and turning activities in SPD children in order to reveal the behavioral strategy of movement and balance control. Six SPD children and six age-match healthy controls participated in the test using a FootScan plantar pressure plate. The time-varying parameters of forces, center of pressure and foot balance index were analyzed using an open-source one-dimensional Statistical Parametric Mapping (SPM1d) package. No difference was found in foot balance and plantar loadings during walking, while limited supination–pronation motion was observed in the SPD children during running and turning. The plantar forces were mainly located in the midfoot region while less toe activity was found as well. Findings should be noted that SPD children had limited supination–pronation movement for shock attenuation in the foot complex and reduced ankle pronation to assist push-off and toe gripping movements. Understanding the behavior of plantar loading strategy and balance control during walking, running and turning activities may provide clinical implications for the rehabilitation and training of daily tasks.


Introduction
Sensory processing disorder (SPD) is affecting more and more school children, as documented by a ratio of around 8.3% in a recent study [1,2]. SPD often accompanies other symptoms, such as autism and attention deficit hyperactivity disorder (ADHD). SPD may result from preborn or postborn factors [3,4]. This disorder is reported as an neurological impairment that limits the brain's ability to receive, process and respond to sensory stimuli [3,5,6]. The prevalence of SPD has increased in school-aged children and children that are firstly diagnosed [7,8]. Children with SPD may have problems responding to sensory discrimination, thus resulting in disruptions such as high risks of falling during dynamic activities in daily life.
A recent study classified the SPD into three main categories, including sensory modulation disorder, sensory discrimination disorder and sensory-based motor disorder [9]. The sensory modulation disorder could be further categorized into three subtypes includes over-responsivity, under-responsivity and seeking/graving [9]. Sensory discrimination previously established Children's Sensory Integration Development Rating Scale [29][30][31]. All participants were free of lower limb injuries or surgeries within half a year prior to the test. The dominant limb of each participant is the right leg, defined as the preferred foot for ball kicking [32]. Prior to data collection, all participants were informed of the procedures about the experimental protocol and signed consent forms. This study was approved by the Human Ethics Committee in the Research Institute of Ningbo University (RAGH20201216).

Test Protocol
Foot pressures during walking, running and turning activities were measured via a FootScan plantar pressure plate (RsScan International, Olen, Belgium) with dimensions of 2 m × 0.4 m × 0.02 m with 16,384 sensors at the sample frequency set as 480 Hz, which is embedded in the 20 m runway. Before formal experimentation, the weight of each participant was measured to calibrate the pressure plate, following a protocol employed in our previous studies [19,33]. Furthermore, all participants were given five minutes to warm up and familiarize with the test environment and requirements. The walking, running and turning right (with left foot) and turning left (with right foot) tasks were included in this study ( Figure 1A,B), and all tasks were conducted under the barefoot condition.
for ball kicking [32]. Prior to data collection, all participants were informed of the procedures about the experimental protocol and signed consent forms. This study was approved by the Human Ethics Committee in the Research Institute of Ningbo University (RAGH20201216).

Test Protocol
Foot pressures during walking, running and turning activities were measured via a FootScan plantar pressure plate (RsScan International, Olen, Belgium) with dimensions of 2 m × 0.4 m × 0.02 m with 16,384 sensors at the sample frequency set as 480 Hz, which is embedded in the 20 m runway. Before formal experimentation, the weight of each participant was measured to calibrate the pressure plate, following a protocol employed in our previous studies [19,33]. Furthermore, all participants were given five minutes to warm up and familiarize with the test environment and requirements. The walking, running and turning right (with left foot) and turning left (with right foot) tasks were included in this study ( Figure 1A,B), and all tasks were conducted under the barefoot condition.
To record normal and actual locomotion performance, all participants were required to walk, run and turn at a self-selected speed. When turning right, the stance leg of left foot was analyzed. Accordingly, when turning left, the right foot was measured. Trials of incomplete step on the pressure plate would be excluded. Three successful trials of both limb from each participant were recorded during walking, running and turning tasks, respectively. Foot pressure on ten anatomic regions was further processed and analyzed, which included Toe 1 (H), Other Toes (T2-5), Metatarsals I-V (M1-M5), Midfoot (MF), Medial heel (MH) and Lateral heel (LH) ( Figure 1C). The foot pressure parameters were processed by one experienced assessor in the RsScan software system (RsScan International, Olen, Belgium).

Data Processing
The time varying features of sum and regional ground reaction forces during stance are shown in Figure 2. The regional forces are selected with force values with an exclusion of zero. Then, the time series of force data from stance foot during walking, running and To record normal and actual locomotion performance, all participants were required to walk, run and turn at a self-selected speed. When turning right, the stance leg of left foot was analyzed. Accordingly, when turning left, the right foot was measured. Trials of incomplete step on the pressure plate would be excluded. Three successful trials of both limb from each participant were recorded during walking, running and turning tasks, respectively. Foot pressure on ten anatomic regions was further processed and analyzed, which included Toe 1 (H), Other Toes (T2-5), Metatarsals I-V (M1-M5), Midfoot (MF), Medial heel (MH) and Lateral heel (LH) ( Figure 1C). The foot pressure parameters were processed by one experienced assessor in the RsScan software system (RsScan International, Olen, Belgium).

Data Processing
The time varying features of sum and regional ground reaction forces during stance are shown in Figure 2. The regional forces are selected with force values with an exclusion of zero. Then, the time series of force data from stance foot during walking, running and turning tasks were processed and interpolated to a length of 51 data points via a cubic spline [34]. turning tasks were processed and interpolated to a length of 51 data points via a cubic spline [34]. The calculation of foot balance index (FBI) is based on the mean value of peak pressure in metatarsal joints and the heel regions, which is presented in Equation (1). FBI presents total foot stability with the positive value indicating foot pronation and the negative value indicating foot supination [28,33]: where Favg represents the average value of the entire plantar force. The COP trajectory is one of the key parameters to reveal gait characteristics [20,21,35,36]. In this study, the COP and FBI trajectory over the stance of walking, running and turning activities were also analyzed and interpolated to a uniform length of 51 points via a cubic spline for statistical analysis.

Statistical Analysis
Three trials of walking, running and turning each were firstly averaged as one to represent this participant, thus avoiding intertrial variations. As per the previously reported protocol [19,37,38], the time series of force is normalized by Zavg, which is calculated by using the total force divided by the sum of original data frames. For the onedimensional time-varying parameters (time-series forces, COP and FBI), the normal distribution is firstly checked and statistical analysis is then evaluated using the independent sample t-test package in statistical parameter mapping 1d (SPM1d) based on the random vector filed theory [34,39]. The time series data were assessed in MATLAB R2018a (The MathWorks, MA, USA). The variables are presented by mean and standard deviation (SD). A two-tailed confidence interval with a p value of less than 0.05 was applied to define statistically significant results. The calculation of foot balance index (FBI) is based on the mean value of peak pressure in metatarsal joints and the heel regions, which is presented in Equation (1). FBI presents total foot stability with the positive value indicating foot pronation and the negative value indicating foot supination [28,33]:

Results
where F avg represents the average value of the entire plantar force. The COP trajectory is one of the key parameters to reveal gait characteristics [20,21,35,36]. In this study, the COP and FBI trajectory over the stance of walking, running and turning activities were also analyzed and interpolated to a uniform length of 51 points via a cubic spline for statistical analysis.

Statistical Analysis
Three trials of walking, running and turning each were firstly averaged as one to represent this participant, thus avoiding intertrial variations. As per the previously reported protocol [19,37,38], the time series of force is normalized by Z avg , which is calculated by using the total force divided by the sum of original data frames. For the one-dimensional time-varying parameters (time-series forces, COP and FBI), the normal distribution is firstly checked and statistical analysis is then evaluated using the independent sample t-test package in statistical parameter mapping 1d (SPM1d) based on the random vector filed theory [34,39]. The time series data were assessed in MATLAB R2018a (The MathWorks, MA, USA). The variables are presented by mean and standard deviation (SD). A two-tailed confidence interval with a p value of less than 0.05 was applied to define statistically significant results.

Results
This study analyzes the center of pressure trajectory, foot balance index, ground reaction force and regional plantar forces in the left and right foot during walking, running and turning activities. In this section, the main results with significant difference are presented, while other information is included in Supplementary Materials for reference.

Center of Pressure Trajectory
The COP trajectory shows no significance between SPD children and healthy controls, apart from the push-off phase around 92-95% of stance in the right foot. Full information is illustrated in the Figure S1. It also could be noted that the left foot is in a supination position, while the right foot lands in a neutral foot position until push-off.
As shown in the Figure 3, the right foot shows significant different foot position in the COP trajectory, especially during 16-69% of stance (p < 0.001). This study analyzes the center of pressure trajectory, foot balance index, ground reaction force and regional plantar forces in the left and right foot during walking, running and turning activities. In this section, the main results with significant difference are presented, while other information is included in Supplementary Materials for reference.

Center of Pressure Trajectory
The COP trajectory shows no significance between SPD children and healthy controls, apart from the push-off phase around 92-95% of stance in the right foot. Full information is illustrated in the Figure S1. It also could be noted that the left foot is in a supination position, while the right foot lands in a neutral foot position until push-off.
As shown in the Figure 3, the right foot shows significant different foot position in the COP trajectory, especially during 16-69% of stance (p < 0.001). As shown in the Figure 4, the right foot presents significantly different foot position in the COP trajectory, particularly during 0-4% and 7-74% of stance. The left is similar with running without significance between SPD children and healthy controls.  As shown in the Figure 4, the right foot presents significantly different foot position in the COP trajectory, particularly during 0-4% and 7-74% of stance. The left is similar with running without significance between SPD children and healthy controls. This study analyzes the center of pressure trajectory, foot balance index, ground reaction force and regional plantar forces in the left and right foot during walking, running and turning activities. In this section, the main results with significant difference are presented, while other information is included in Supplementary Materials for reference.

Center of Pressure Trajectory
The COP trajectory shows no significance between SPD children and healthy controls, apart from the push-off phase around 92-95% of stance in the right foot. Full information is illustrated in the Figure S1. It also could be noted that the left foot is in a supination position, while the right foot lands in a neutral foot position until push-off.
As shown in the Figure 3, the right foot shows significant different foot position in the COP trajectory, especially during 16-69% of stance (p < 0.001). As shown in the Figure 4, the right foot presents significantly different foot position in the COP trajectory, particularly during 0-4% and 7-74% of stance. The left is similar with running without significance between SPD children and healthy controls.

Foot Balance Index
The foot balance index (FBI) of the left and right foot shows no significance between the SPD children and healthy controls (refer to the Figure S2 for full information). While the left and right foot had similar trends of foot balance index during running in SPD children with limited supination-pronation range, while during push-off, the healthy children show an obvious pronation trend in the FBI ( Figure 5), particularly in the right foot where statistical significance is found during 47-63% of stance (p < 0.001). The foot balance index (FBI) of the left and right foot shows no significance between the SPD children and healthy controls (refer to the Figure S2 for full information). While the left and right foot had similar trends of foot balance index during running in SPD children with limited supination-pronation range, while during push-off, the healthy children show an obvious pronation trend in the FBI (Figure 5), particularly in the right foot where statistical significance is found during 47-63% of stance (p < 0.001). Due to the high dynamic nature of turning motion, healthy children showed supination-pronation shifts in both feet, whereas SPD children presented restricted supinationpronation during both left turning with right foot and right turning with left foot ( Figure  6). Significance is observed in the left foot during initial landing and the push-off phase.  Due to the high dynamic nature of turning motion, healthy children showed supinationpronation shifts in both feet, whereas SPD children presented restricted supination-pronation during both left turning with right foot and right turning with left foot (Figure 6). Significance is observed in the left foot during initial landing and the push-off phase.

Ground Reaction Force and Regional Plantar Forces
The foot balance index (FBI) of the left and right foot shows no significance between the SPD children and healthy controls (refer to the Figure S2 for full information). While the left and right foot had similar trends of foot balance index during running in SPD children with limited supination-pronation range, while during push-off, the healthy children show an obvious pronation trend in the FBI (Figure 5), particularly in the right foot where statistical significance is found during 47-63% of stance (p < 0.001). Due to the high dynamic nature of turning motion, healthy children showed supination-pronation shifts in both feet, whereas SPD children presented restricted supinationpronation during both left turning with right foot and right turning with left foot ( Figure  6). Significance is observed in the left foot during initial landing and the push-off phase.

Ground Reaction Force and Regional Plantar Forces
The ground reaction force in the left and right foot of SPD children and healthy controls showed no significance during walking (refer to the Figure S3 for full information), and regional plantar force presented no significance during walking (refer to the Figure S4 for information).
The magnitudes of running GRF are higher in healthy controls than SPD children, especially in the left foot where healthy controls are significantly larger than SPD children from 22-63% during stance. Similar magnitudes were observed in the right foot, although no significance is found (Figure 7). trols showed no significance during walking (refer to the Figure S3 for full information), and regional plantar force presented no significance during walking (refer to the Figure  S4 for information).
The magnitudes of running GRF are higher in healthy controls than SPD children, especially in the left foot where healthy controls are significantly larger than SPD children from 22-63% during stance. Similar magnitudes were observed in the right foot, although no significance is found (Figure 7). Contrary results are observed during turning (or cutting) motions with the left and right foot. The left foot of SPD children shows greater GRF than the healthy control during initial landing (1-2% and 11-15%) phase while turning right. The right foot of SPD children shows larger GRF around the mid stance (35-52%) and afterwards, although without significance ( Figure 8).  Contrary results are observed during turning (or cutting) motions with the left and right foot. The left foot of SPD children shows greater GRF than the healthy control during initial landing (1-2% and 11-15%) phase while turning right. The right foot of SPD children shows larger GRF around the mid stance (35-52%) and afterwards, although without significance ( Figure 8).
trols showed no significance during walking (refer to the Figure S3 for full information), and regional plantar force presented no significance during walking (refer to the Figure  S4 for information).
The magnitudes of running GRF are higher in healthy controls than SPD children, especially in the left foot where healthy controls are significantly larger than SPD children from 22-63% during stance. Similar magnitudes were observed in the right foot, although no significance is found (Figure 7). Contrary results are observed during turning (or cutting) motions with the left and right foot. The left foot of SPD children shows greater GRF than the healthy control during initial landing (1-2% and 11-15%) phase while turning right. The right foot of SPD children shows larger GRF around the mid stance (35-52%) and afterwards, although without significance (Figure 8).  In terms of the regional plantar force during running, the differences between SPD children and healthy controls were mainly found in the toe (Toe 1 and Other Toes), medial forefoot (M1, M2 and M3) and midfoot regions (Figures 9 and 10) during certain phases (in percentage) of the stance. Specifically, healthy controls had greater force in Toe 1 (during 7-62%, p < 0.001), Other Toes (during 98-100%, p = 0.049), M1 (during 1-4%, p = 0.038), M2 (during 33-69%, p < 0.001), M3 (during 14-62%, p < 0.001) and MedHeel (during 56-83%, p < 0.001) regions than SPD children in the left foot during running, while the force in the Midfoot of SPD children is significantly larger than the healthy controls (during 29-57%, p < 0.001). As for the right foot, the healthy controls had larger forces in Toe 1 (during 8-26%, p < 0.001), Other Toes (during 19-22%, p = 0.045), M1 (during 23-26%, p < 0.01; and 63-72%, p = 0.001) and LatHeel (during 14-24%, p = 0.002), whilst the force in the Midfoot of SPD children is far greater than the healthy controls (during 6-100%, p < 0.001).   The main difference between SPD children and healthy controls was found in the toe, forefoot and midfoot regions (Figures 11 and 12) during certain phases (in percentage) of the stance. Specifically, in the left foot while turning right, the difference was located at the starting and end phase of Toe 1. The healthy controls had larger force than the SPD children in Other Toes (during 56-66%, p = 0.001), while SPD children had greater force in Midfoot (during 1-70%, p < 0.001) than the healthy controls. As for the right foot while turning left, the healthy controls had larger forces in Toe 1 (during 16-24%, p < 0.001) and Other Toes (during 23-27%, p = 0.029; and 45-47%, p = 0.001), while SPD children had larger forces in M1 (during 57-97%, p < 0.001), M2 (during 94-100%, p = 0.028), M3 (during 1-3%, p = 0.046; and 30-94%, p < 0.001), M4 (during 17-56%, p < 0.001) and Midfoot (during 23-68%, p < 0.001; and 88-97%, p < 0.001). Children 2022, 9, x FOR PEER REVIEW 11 of 17 Figure 11. The regional plantar forces in the left foot during turning in the SPD children and healthy controls with highlighted statistics. Figure 11. The regional plantar forces in the left foot during turning in the SPD children and healthy controls with highlighted statistics. Children 2022, 9, x FOR PEER REVIEW 12 of 17 Figure 12. The regional plantar forces in the right foot during turning in the SPD children and healthy controls with highlighted statistics. Figure 12. The regional plantar forces in the right foot during turning in the SPD children and healthy controls with highlighted statistics.

Discussion
The purpose of this study was to compare differences in plantar loading and foot balance during the walking, running and turning activities in children with SPD via a comparative analysis with healthy controls in both left (non-dominant) and right (dominant) feet. The walking tasks had no significant difference between SPD children and healthy controls in both feet. During running, SPD children showed a supinated foot position during midstance with reduced plantar forces in the toes and metatarsals but focused forces in the midfoot of both feet. Similar loading patterns and foot COPs were found during turning where children with SPD had supinated foot posture during left turning and supination-pronation movement was restricted in both feet during turning. While compared to the healthy controls, the left foot had the most plantar force during right turning and the right foot had the most loading in the metatarsals and midfoot but less in toes during left turning among SPD children.
The current study found that the maximum force loading of the entire foot was around 1.5 BW during walking and 2-2.5 BW during running and turning activities, which were in accordance with the findings of previous studies [33,38,40]. A slightly higher sum force in the right limb might be due to limb dominance, as the non-dominant limb was responsible for support and stability, while the dominant limb played the role of propulsion [41]. The maximum force in the Midfoot region of SPD children in this study was around 0.7 BW, which was greater than the healthy controls in the range of previous studies [33,38]. This may be a key observation that should be noted, where children with SPD had reduced foot control during dynamic running and turning activities, while the healthy control group results were similar with the findings in the previous study, where age may affect the larger values of plantar force loadings [42]. Another information was that the children's feet may be different to adult ones because of the plastic feature of the foot-ankle complex, especially in SPD children with the motor disorder of this study. The foot-ankle complex was often characterized by altered foot shape and function [43][44][45][46]. Because of the plasticity in this complex, the loading in the MF region increased greatly during running and turning, suggesting less stability or reduced control of the longitudinal arch [43] and reduced force in the distal segment of toes [47,48]. These could explain the larger MF (midfoot) loading but lesser toes loadings compared to the healthy controls in this study.
Assessments of the plantar loading and dynamic foot balance were employed to understand the foot structure and mechanics during dynamic conditions [49,50]. During walking, COP and FBI showed no significance between SPD children and healthy controls in both dominant and non-dominant feet. This may be because the SPD children of motor disorder cohort had stable behavioral automation [51] or less sensory hypersensitivity [6] in the neuromuscular system during preferred walking in a less dynamic condition. However, the foot had restricted supination-pronation motions during running, showing a supinated COP trajectory and limited FBI range over the stance. This may be due to hypersensitivity from the neuroanatomical structure [6] in order to control motions of the foot-ankle complex with overprotection, thus restricting supination-pronation, which s reported as a key function for impact absorption [52]. This change was also linked with less plantar force in the medial and lateral heel regions of both feet, whilst the midfoot showed far greater force in both feet of SPD children compared to healthy controls. This difference may explain with the higher plasticity in the medial foot (arch region) during midstance [43]. Following the midstance was the push-off phase during running [43,44], which manifested with greater medial forefoot (M1, M2 and M3) force loadings in the healthy controls compared with the SPD children. The findings from this study may be useful, and it should be noted that SPD children had limited supination-pronation movement for shock attenuation in the foot complex and reduced ankle pronation to assist push-off and toe gripping movements.
During cutting, a more apparently limited motion in the foot-ankle complex was found, as shown in the COP trajectory and FBI of both feet compared with healthy controls. During the stance phase, most plantar loadings were exerted in the medial side of the stance foot with the foot in a pronated position among the healthy controls [53]. By nature, the COP trajectory transferred medially compared to the running in the healthy control, which also proved increasing foot pronation during the stance phase of turning. On the contrary, this transference was not observed in SPD children, and it still showed restricted supinationpronation motion. The finding suggested that sensory hypersensitivity may hinder the normal function in the foot-ankle complex, thus affecting balance and postural control during dynamic movement [6], such as the turning task of direction changes (or cutting maneuver). As documented in the literature, the youth population tended to employ ankle strategies to stay balance compared to the hip strategy of the aged population [21,54].
In this study, higher force loadings under M1 and M2 regions were found in both feet during turning, which agreed with a previous study [53]. The pressure load on the first metatarsal was positively associated with gait speed [55], suggesting that both SPD children and healthy controls had a similar range of run-turning gait speed. During turning activity (or known as cutting maneuvers), participants may decelerate in the original direction and redirect to the intended direction during turning, requiring acceleration [56]. Accordingly, the direction changing mechanism would account for the increased force loading under the M1 and M2 regions, and it is facilitated with active toe gripping motion in healthy controls (but not observed in SPD children) [46,48,57].
However, the significantly higher force profiles in the lateral forefoot of SPD children deserve special attention where active propulsive push-off occurred in these anatomical regions, which may result from reduced pronation motions. This alteration, thus, exposed SPD children into risks of falling and postural instability or ankle sprain during turning activities. In terms of force loadings in the midfoot regions, the reason was similar with running where there is high plasticity in the arch region. Turning is higher in terms of dynamics compared with running, which was also analyzed as a cutting maneuver in several specific sports such as soccer and basketball with recruited participants of college students or athletes. In this study, children with SPD were involved in order to be compared with healthy controls, which might be responsible for the different results. In this study, children with SPD more specifically had dyspraxia by conducting Sensory Integration and Praxis Tests [51]. The disorder in sensory processing might produce certain influences on motor behavior and disrupt the quality of movement [58].
Several limitations of current study should be considered. Firstly, this study only compared the difference in plantar loading, COP and FBI between SPD children and healthy control. The kinematic parameters and the muscle activities of the lower limb shall be analyzed to comprehensively investigate the movement characteristics of children with sensory processing disorder in future studies. Secondly, smaller numbers of participants were recruited in this study, which lacked generality to other age and gender cohorts, as this study was a pilot to reveal the movement patterns of this special SPD group. It was difficult to recruit more SPD children of different ages and genders, which shall be a primary aim in future study.

Conclusions
The current study aimed to compare plantar loading and foot balance during walking, running and turning activities in SPD children. No difference was found in foot balance and plantar loading during less dynamic walking, while limited supination-pronation motion was observed in SPD children during high dynamic running and turning. The plantar forces were mainly located in the midfoot region while less toes activity was found as well. It should be noted that SPD children had limited supination-pronation movement for shock attenuation in the foot complex and reduced ankle pronation to assist push-off and toe gripping movements. Understanding the behavior of plantar loading strategy and balance control during walking, running and turning activities may provide clinical implications for the rehabilitation and training of daily tasks.

Supplementary Materials:
The following supporting information can be downloaded at the following website: https://www.mdpi.com/article/10.3390/children9030379/s1, Figure S1: The COP of trajectory in the left foot and right foot during walking with highlighted direction of pronation (Blue arrow) and supination (Red arrow); Figure S2: The foot balance index (FBI) in the left foot and right foot during walking with highlighted direction of pronation (blue arrow) and supination (red arrow); Figure S3: The vertical ground reaction force (GRF) in the left foot and right foot during walking in the SPD children and healthy controls; Figure S4: The regional plantar force in the left foot and right foot during walking in the SPD children and healthy controls. Informed Consent Statement: Informed consent was obtained from all participants involved in the study.

Data Availability Statement:
The data may be available upon reasonable request from the corresponding authors.