Table 1.
Age, sex and MABC performance for children with DCD.
Figure 1.
Experimental set-up showing a child performing the multisensory posture task.
The child stood in a modified semi-tandem stance looking at a front screen with random dots projection (not shown due to room illumination) while touching a bar lightly without triggering an auditory alarm. Both the visual display and the touch bar moved in the mediolateral direction simultaneously but with different frequencies. Markers were placed on the head and on the right side of the arm, and lower leg to track postural kinematics.
Figure 2.
Exemplar of drives and sway from a T4V2 trial of a child with DCD.
Subplot A shows the time series of the trajectories, and subplot B shows their power spectrum in the frequency domain. Note that the Tdrive and Vdrive oscillated at different frequency (0.28 and 0.2 Hz respectively).
Figure 3.
Exemplar of averaged frequency response functions (FRFs), gains and phases.
Results are averaged across three trials for each condition from all five test conditions in a 7.6-year-old child with DCD. Subplots in the first column show the FRFs in the complex plane for the Head (A) and CoM (B). The distances from the FRFs to the origin are the gains and they are plotted in the middle column subplots for Head gains (C) and CoM gains (D) to Tdrive (filled markers) and Vdrive (open face markers). Intra-modal reweighting refers to gain changes due to the change of a specific modality’s amplitude. For example, symbol such as |- INTRA (T) -| indicates that touch gain intra-modal reweighting is evaluated across T8V2 and T2V2 conditions. Similarly, |- INTRA (V) -| indicates that vision gain intra-modal reweighting is evaluated across T2V2 and T2V8 conditions. Inter-modal reweighting refers to gain changes due to the change of a coexisting modality’s amplitude. For example, symbol |- INTER (T) -| indicates that touch gain inter-modal reweighting is evaluated across T2V2 and T2V8 conditions while |- INTER (V) -| indicates that vision gain inter-modal reweighting is evaluated across T2V2 and T8V2 conditions. A significant gain difference between conditions indicates reweighting. This child demonstrates intra-modal touch reweighting. The angle between the FRFs and the positive real axis are phases and they are plotted in the right column subplots for Head phase (E) and CoM phase (F). Note that phases are all negative indicating a phase lag of postural response to drives.
Figure 4.
Linear fit of gains and phases as a function of age.
The linear fit is for each group of children for one condition. Here the gains (A) and phases (B) are from Head responses to Tdrive in a T8V2 condition. Gains (A) and phases (B) from 41 TD children (filled markers) and from 20 children with DCD (open face markers) with fitted gain lines and fitted phase lines are plotted with associated standard errors calculated from bootstrapping. The age range used for model fitting is from 4.2 to 10.8 years of age for the TD children and from 6.6 to 11.8 years for children with DCD. We evaluate the model fitting for both groups at a lower comparison age of 6.6 years and an upper comparison age of 10.8 years as shown in the following Figures 5, 6, 7 and 8. Note that children with DCD showed delayed phase of postural response to Tdrive comparing to their TD peers.
Figure 5.
Fitted CoM gains at upper (10.8 years) and lower (6.6 years) comparison ages.
Each fitted gain with its corresponding standard error was extracted from a linear model fit (exemplified in Figure 4A) for the specified condition, segment, sensory drive and comparison age. Fitted gains from 5 conditions were plotted in a subplot to summarize intra-, intermodal and total reweighting for each group (TD: triangle marker; DCD: square marker) and to contrast group difference. Subplot (A) is the CoM gain to Vdrive at upper comparison age, (B) is the CoM gain to Vdrive at lower comparison age, (C) is the CoM gain to Tdrive at upper comparison age, and (D) is the CoM gain to Tdrive at lower comparison age. Symbol such as |- INTER (V) -| indicates which two conditions are used to evaluate if there were inter-modal reweighting to Vdrive. Similar symbols are used for intra-modal reweighting and for reweighting to Tdrive. * indicates significant condition effect (***** for p<0.0001, **** for p<0.001, *** for p<0.01, ** for p<0.05, and * for marginal significance with p<0.1) for each group (labeled as TD, DCD respectively) or for group comparison (labeled as “Con by Gr”, indicating condition-by-group interaction). The solid bracket symbol indicates which two conditions are being compared. For example, the larger solid bracket is for the total reweighting between T8V2, and T2V8. The smaller bracket is for T8V2 & T2V2 condition pair or T2V2 & T2V8 condition pair.
Figure 6.
Fitted Head gains at upper (10.8 years) and lower (6.6 years) comparison ages.
Symbol notations, legends for statistical significance are the same as in Figure 5.
Figure 7.
Fitted CoM phases at upper (10.8 years) and lower (6.6 years) comparison ages.
Each fitted phase with its corresponding standard error was extracted from a linear model fit (exemplified in Figure 4B) for the specified condition, segment, sensory drive and comparison age. Symbol notations, legends for statistical significance are the same as in Figure 5.
Figure 8.
Fitted Head phases at upper (10.8 years) and lower (6.6 years) comparison ages.
Symbol notations, legends for statistical significance are the same as in Figure 5.