Humans exploit robust locomotion by improving the stability of control signals

<strong>Background</strong> Is the control of movement less stable when we walk or run in challenging settings? Intuitively, one might answer that it is, given that adding constraints to locomotion (e.g. rough terrain, age-related impairments, etc.) makes movements less stable. Here, we investigated how young and old humans synergistically activate muscles during locomotion when different perturbation levels are introduced. Of these control signals, called muscle synergies, we analyzed the stability over time. Surprisingly, we found that perturbations and older age force the central nervous system to produce muscle activation patterns that are more stable. These outcomes show that robust locomotion in challenging settings is achieved by increasing the stability of control signals, whereas easier tasks allow for more unstable control. <strong>How to use the data set</strong> This supplementary data set contains: a) the metadata with anonymized participant information, b) the raw electromyographic (EMG) data acquired during locomotion, c) the touchdown and lift-off timings of the recorded limb, d) the filtered and time-normalized EMG, e) the muscle synergies extracted via non-negative matrix factorization and f) the code written in R (R Found. for Stat. Comp.) to process the data, including the scripts to calculate the Maximum Lyapunov Exponents of motor primitives. In total, 476 trials from 86 participants are included in the supplementary data set. The file “participant_data.dat” is available in ASCII and RData (R Found. for Stat. Comp.) format and contains: Code: the participant’s code Experiment: the experimental setup in which the participant was involved (E1 = walking and running, overground and treadmill; E2 = walking and running, even- and uneven-surface; E3 = unperturbed and perturbed walking, young and old) Group: the group to which the participant was assigned (see above for the details) Sex: the participant’s sex (M or F) Speed: the speed at which the recordings were conducted in [m/s] (two values separated by a comma mean that recordings were done at two different speeds, i.e. walking and running) Age: the participant’s age in years (participants were considered old if older than 65 years, but younger than 80) Height: the participant’s height in [cm] Mass: the participant’s body mass in [kg]. The files containing the gait cycle breakdown are available in RData (R Found. for Stat. Comp.) format, in the file named “CYCLE_TIMES.RData”. The files are structured as data frames with 30 rows (one for each gait cycle) and two columns. The first column contains the touchdown incremental times in seconds. The second column contains the duration of each stance phase in seconds. Each trial is saved as an element of a single R list. Trials are named like “CYCLE_TIMES_P0020,” where the characters “CYCLE_TIMES” indicate that the trial contains the gait cycle breakdown times and the characters “P0020” indicate the participant number (in this example the 20th). Please note that the overground trials of participants P0001 and P0009 and the second uneven-surface running trial of participant P0048 only contain 22, 27 and 23 cycles, respectively. The files containing the raw, filtered and the normalized EMG data are available in RData (R Found. for Stat. Comp.) format, in the files named “RAW_EMG.RData” and “FILT_EMG.RData”. The raw EMG files are structured as data frames with 30000 rows (one for each recorded data point) and 14 columns. The first column contains the incremental time in seconds. The remaining thirteen columns contain the raw EMG data, named with muscle abbreviations that follow those reported in the Materials and Methods section of this Supplementary Materials file. Each trial is saved as an element of a single R list. Trials are named like “RAW_EMG_P0053_OG_02”, where the characters “RAW_EMG” indicate that the trial contains raw emg data, the characters “P0053” indicate the participant number (in this example the 53rd), the characters “OW” indicate the locomotion type (E1: OW=overground walking, OR=overground running, TW=treadmill walking, TR=treadmill running; E2: EW=even-surface walking, ER=even-surface running, UW=uneven-surface walking, UR=uneven-surface running; E3: NW=normal walking, PW=perturbed walking), and the numbers “02” indicate the trial number (in this case the 2nd). The 10 trials per participant recorded for each overground session (i.e. 10 for walking and 10 for running) were concatenated into one. The filtered and time-normalized emg data is named, following the same rules, like “FILT_EMG_P0053_OG_02”. The files containing the muscle synergies extracted from the filtered and normalized EMG data are available in RData (R Found. for Stat. Comp.) format, in the files named “SYNS_H.RData” and “SYNS_W.RData”. The muscle synergies files are divided in motor primitives and motor modules and are presented as direct output of the factorization and not in any functional order. Motor primitives are data frames with 6000 rows and a number of columns equal to the number of synergies (which might differ from trial to trial) plus one. The rows contain the time-dependent coefficients (motor primitives), one column for each synergy plus the time points (columns are named e.g. “Time, Syn1, Syn2, Syn3”, where “Syn” is the abbreviation for “synergy”). Each gait cycle contains 200 data points, 100 for the stance and 100 for the swing phase which, multiplied by the 30 recorded cycles, result in 6000 data points distributed in as many rows. This output is transposed as compared to the one discussed above to improve user readability. Each set of motor primitives is saved as an element of a single R list. Trials are named like “SYNS_H_P0012_PW_02”, where the characters “SYNS_H” indicate that the trial contains motor primitive data, the characters “P0012” indicate the participant number (in this example the 12th), ), the characters “PW” indicate the locomotion type (see above), and the numbers “02” indicate the trial number (in this case the 2nd). Motor modules are data frames with 13 rows (number of recorded muscles) and a number of columns equal to the number of synergies (which might differ from trial to trial). The rows, named with muscle abbreviations that follow those reported in the Materials and Methods section of this Supplementary Materials file, contain the time-independent coefficients (motor modules), one for each synergy and for each muscle. Each set of motor modules relative to one synergy is saved as an element of a single R list. Trials are named like “SYNS_W_P0082_PW_02”, where the characters “SYNS_W” indicate that the trial contains motor module data, the characters “P0082” indicate the participant number (in this example the 82nd) ), the characters “PW” indicate the locomotion type (see above), and the numbers “02” indicate the trial number (in this case the 2nd). Given the nature of the NMF algorithm for the extraction of muscle synergies, the supplementary data set might show non-significant differences as compared to the one used for obtaining the results of this paper. The files containing the MLE calculated from motor primitives are available in RData (R Found. for Stat. Comp.) format, in the file named “MLE.RData”. MLE results are presented in a list of lists containing, for each trial, 1) the divergences, 2) the MLE, and 3) the value of the R² between the divergence curve and its linear interpolation made using the specified amount of points. The divergences are presented as a one-dimensional vector. MLE are one number like the R² value. Trials are named like “MLE_P0081_EW_01”, where the characters “MLE” indicate that the trial contains MLE data, the characters “P0081” indicate the participant number (in this example the 81st) ), the characters “EW” indicate the locomotion type (see above), and the numbers “01” indicate the trial number (in this case the 1st). All the code used for the preprocessing of EMG data, the extraction of muscle synergies and the calculation of MLE is available in R (R Found. for Stat. Comp.) format. Explanatory comments are profusely present throughout the scripts (“SYNS.R”, which is the script to extract synergies, “fun_NMF.R”, which contains the NMF function, “MLE.R”, which is the script to calculate the MLE of motor primitives and “fun_MLE.R”, which contains the MLE function).<strong>研究背景</strong> 在具有挑战性的环境中行走或跑步时，运动控制是否会更不稳定？直觉上，人们可能会给出肯定的答案，因为在运动中增加约束条件（例如崎岖地形、年龄相关的运动功能损伤等）会使运动稳定性下降。本研究旨在探究年轻与老年人类在运动过程中，当施加不同程度扰动时，肌肉的协同激活模式。我们将这类控制信号称为肌肉协同（muscle synergies），并分析其随时间的稳定性。令人意外的是，我们发现扰动与衰老会促使中枢神经系统生成更稳定的肌肉激活模式。该结果表明，在挑战性环境中实现稳健的运动，需通过提高控制信号的稳定性来达成；而相对简单的任务则允许更不稳定的控制策略。 <strong>数据集使用说明</strong> 本补充数据集包含以下内容：a) 包含匿名受试者信息的元数据；b) 运动过程中采集的原始肌电图（electromyographic, EMG）数据；c) 记录肢体的触地与离地时间；d) 经过滤波与时间归一化的EMG数据；e) 通过非负矩阵分解（non-negative matrix factorization）提取的肌肉协同数据；f) 用于处理数据的R语言（R Found. for Stat. Comp.）代码，其中包含计算运动基元最大李雅普诺夫指数（Maximum Lyapunov Exponents）的脚本。本补充数据集共包含来自86名受试者的476组试验数据。 文件"participant_data.dat"支持ASCII与RData（R Found. for Stat. Comp.）两种格式，其内容包括： 受试者编号：受试者的唯一标识代码 实验分组：受试者参与的实验设置（E1 = 行走与跑步，涵盖平地与跑步机场景；E2 = 行走与跑步，涵盖平坦与不平坦路面场景；E3 = 无扰动与有扰动行走，涵盖年轻与老年受试者） 组别：受试者所属组别（详见前文说明） 性别：受试者性别（M为男性，F为女性） 运动速度：数据采集时的运动速度，单位为[m/s]（若为两个以逗号分隔的数值，则表示分别在两种速度下完成采集，即行走与跑步状态） 年龄：受试者年龄（单位：岁，年龄大于65岁且小于80岁的受试者被划分为老年组） 身高：受试者身高（单位：[cm]） 体重：受试者体质量（单位：[kg]）。 包含步态周期分解数据的文件以RData（R Found. for Stat. Comp.）格式存储，文件名为"CYCLE_TIMES.RData"。此类文件采用数据框结构，包含30行（对应每个步态周期）与2列。第一列为触地累计时间（单位：秒），第二列为每个站立相的持续时间（单位：秒）。每组试验数据以单个R列表的元素形式存储，文件命名规则为"CYCLE_TIMES_P0020"，其中"CYCLE_TIMES"表示该文件包含步态周期分解时间数据，"P0020"表示受试者编号（本示例为第20名受试者）。请注意：受试者P0001与P0009的平地行走试验，以及受试者P0048的第二次不平坦路面跑步试验，仅分别包含22、27与23个步态周期。 包含原始、滤波与归一化EMG数据的文件以RData（R Found. for Stat. Comp.）格式存储，分别对应文件"RAW_EMG.RData"与"FILT_EMG.RData"。原始EMG文件采用数据框结构，包含30000行（对应每个采集的数据点）与14列。第一列为累计时间（单位：秒），剩余13列为原始EMG数据，其列名采用肌肉缩写，遵循本补充材料"材料与方法"部分的命名规范。每组试验数据以单个R列表的元素形式存储，文件命名规则为"RAW_EMG_P0053_OG_02"，其中"RAW_EMG"表示该文件包含原始EMG数据，"P0053"表示受试者编号（本示例为第53名受试者），"OG"表示运动模式（E1场景：OG=平地行走、OR=平地跑步、TW=跑步机行走、TR=跑步机跑步；E2场景：EW=平坦路面行走、ER=平坦路面跑步、UW=不平坦路面行走、UR=不平坦路面跑步；E3场景：NW=正常行走、PW=扰动行走），"02"表示试验编号（本示例为第2组试验）。每名受试者在每次平地运动实验时段中采集的10组试验数据（行走与跑步各10组）会被合并为一个文件。滤波与时间归一化的EMG数据命名规则相同，例如"FILT_EMG_P0053_OG_02"。 包含从滤波与归一化EMG数据中提取的肌肉协同数据的文件以RData（R Found. for Stat. Comp.）格式存储，分别对应文件"SYNS_H.RData"与"SYNS_W.RData"。肌肉协同文件分为运动基元（motor primitives）与运动模块（motor modules）两类，直接输出非负矩阵分解的结果，未按功能顺序排列。运动基元为数据框，包含6000行，列数等于协同因子数量（不同试验可能存在差异）加1。行代表随时间变化的系数（运动基元），每列对应一个协同因子，另有一列包含时间点（列名例如"Time, Syn1, Syn2, Syn3"，其中"Syn"为"synergy"的缩写）。每个步态周期包含200个数据点，站立相与摆动相各100个，结合30个记录的步态周期，总计6000个数据点，对应6000行数据。为提升用户可读性，该输出与前文描述的结构相比进行了转置。每组运动基元数据以单个R列表的元素形式存储，文件命名规则为"SYNS_H_P0012_PW_02"，其中"SYNS_H"表示该文件包含运动基元数据，"P0012"表示受试者编号（本示例为第12名受试者），"PW"表示运动模式（详见前文说明），"02"表示试验编号（本示例为第2组试验）。运动模块为数据框，包含13行（对应记录的肌肉数量），列数等于协同因子数量（不同试验可能存在差异）。行名为肌肉缩写，遵循本补充材料"材料与方法"部分的命名规范，行代表与时间无关的系数（运动模块），每个协同因子与每块肌肉对应一列。每组运动模块数据以单个R列表的元素形式存储，文件命名规则为"SYNS_W_P0082_PW_02"，其中"SYNS_W"表示该文件包含运动模块数据，"P0082"表示受试者编号（本示例为第82名受试者），"PW"表示运动模式（详见前文说明），"02"表示试验编号（本示例为第2组试验）。由于非负矩阵分解算法的特性，本补充数据集的结果与本论文成果所使用的数据集可能存在非显著性差异。 包含从运动基元计算得到的最大李雅普诺夫指数数据的文件以RData（R Found. for Stat. Comp.）格式存储，文件名为"MLE.RData"。最大李雅普诺夫指数结果以列表嵌套列表的形式呈现，每组试验包含三部分内容：1）发散度；2）最大李雅普诺夫指数值；3）发散度曲线与使用指定点数进行线性拟合得到的拟合曲线之间的决定系数R²。发散度以一维向量形式呈现，最大李雅普诺夫指数与R²均为单个数值。文件命名规则为"MLE_P0081_EW_01"，其中"MLE"表示该文件包含最大李雅普诺夫指数数据，"P0081"表示受试者编号（本示例为第81名受试者），"EW"表示运动模式（详见前文说明），"01"表示试验编号（本示例为第1组试验）。 所有用于EMG数据预处理、肌肉协同提取以及最大李雅普诺夫指数计算的代码均以R语言（R Found. for Stat. Comp.）格式提供。脚本中包含大量注释说明："SYNS.R"为肌肉协同提取脚本，"fun_NMF.R"包含非负矩阵分解函数，"MLE.R"为运动基元最大李雅普诺夫指数计算脚本，"fun_MLE.R"包含最大李雅普诺夫指数计算函数。