TITLE: 2-back task in split-belt adaptation GENERAL INFORMATION 1. Dataset for manuscript: Younger and late middle-aged adults exhibit different patterns of cognitive-motor interference during locomotor adaptation, with no disruption of savings. *IN REVIEW* Journal: Frontiers in Aging Neuroscience 2. Authors Cristina Rossi (2,3); cristinarossi@jhmi.edu Ryan T. Roemmich (2,4); rroemmi1@jhmi.edu Nicolas Schweighofer (1); schweigh@pt.usc.edu Amy J. Bastian (2,5); bastian@kennedykrieger.org Kristan A. Leech (1); kristanl@usc.edu (1) Division of Biokinesiology and Physical Therapy, University of Southern California, Los Angeles, CA, USA (2) Kennedy Krieger Institute, Baltimore, MD, USA (3) Dept of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, USA (4) Dept of Physical Medicine and Rehabilitation, Johns Hopkins University School of Medicine, Baltimore, USA (5) Dept of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, MD, USA 3. Correspondence: Kristan Leech Center for Health Professions 1540 Alcazar St #155 Los Angeles, CA 90033 kristanl@usc.edu 4. We studied split-belt treadmill adaptation and savings in young (21±2 y/o) and older (56±6 y/o) adults with or without a secondary 2-back task during adaptation. We here provide here kinematic data for split-belt treadmill walking - specifically, step length asymmetry, double support asymmetry, and limb excursion asymmetry measure. We provide raw response data (button presses and reaction time) for performance in the 2-back cognitive task. We also provide scripts that can be used to analyse motor and cognitive data using bootstrapping. DATA OVERVIEW 1. Motor data: a. File List: YAST_MOTOR.mat : motor data for YASingle group YADT_MOTOR.mat : motor data for YADual group OAST_MOTOR.mat : motor data for OASingle group OADT_MOTOR.mat : motor data for OADual group b. File Description: Each file contains the following kinematic measures (further description is provided in the manuscript): DSF: double support time of the fast leg DSS: double support time of the slow leg FSL: step length of the fast leg SSL: step length of the slow leg LEF: limb excursion of the fast leg LES: limb excursion of the slow leg Each of the kinematic variables is a 10x10000x7 matrix, where each row contains data for one subject, each of the 7 3rd dimensions contains data for a separate phase of the experiment (baseline, adaptation, 4x washout blocks, readaptation), and each column contains data for one stride (padded with nans at the end). Each file also contains a DT dummy variable, structured in the same way as the kinematic variables, that is 1 for strides when the participant was dual-tasking, and 0 for other trials. 2. Cognitive data: a. File List: YADT_COGNITIVE.mat : n-back task data for YADual group OADT_COGNITIVE.mat : n-back task data for OADual group OADN_COGNITIVE.mat : n-back task data for OADualNoise group b. File Description: Each file contains the following variables: groupScoreArray: contains response accuracy to the n-back task, where 1 is a correct response (press stimuli), 0 is a correct omission (do not press), -1 is an incorrect response (do not press), -2 is an incorrect omission (press) groupTimeArray: contains reaction time in seconds for responses Each variable is a struct with fields: "walk": a 125x10 matrix containing data from the baseline task "adapt": a 250x10 matrix containing data from the adaptation task in both matrices, each column is a subject and each row is a letter stimuli 3. Data generated using scripts below (not the original dataset; it is provided here as it takes a while to generate and it is needed to plot figures, but it can be generated de novo with the scripts and data provided here). a. Bootstrapped data: Provided in a zipped folder "bootstraps.zip", it must be unzipped before running figures code. b. Statistics: Provided in a zipped folder "statistics.zip", it must be unzipped before running figures code. SCRIPTS OVERVIEW All scripts are in MATLAB (.m extension). 1. Bootstraps: bootstrap_subjects: creates 10000 samples of 10 subjects, which will be used by the other boostrap scripts bootstrap_motor: bootstraps 10000 samples of bins for step length asymmetry, limb excursion asymmetry, double support asymmetry bootstrap_cognitive: bootstraps 10000 samples of logistic fits of error rate to press and do not press stimuli of the n-back task, and linear fits of reaction time boostrap_doubleexp: bootstraps 1000 samples of double exponential fits to step length asymmetry in adaptation and readaptation 2. Statistics: statistics_motor: computes confidence intervals related to the motor data (step length asymmetry, limb excursion asymmetry, double support asymmetry) statistics_cognitive: computes confidence intervals related to the cognitive data (error rate with press and do not press stimuli, and reaction time) statistics_doubleexp: computes confidence intervals related to the double exponential fits of adaptation and readaptation step length asymmetry 3. Figures: figure_2_5_6: creates figures 2, 5 and 6 figure_3_7: creates figures 3 and 7 figure_4_8: creates figures 4 and 8 supfigure1: creates supplementary figure 1 supfigure2: creates supplementary figure 2 supfigure3: creates supplementary figure 3 createallfigures: run this script to make figures 2 through 8 and suppl. figures 1 and 2 4. Helper functions: bssem: computes bootstrap standard error sem: computes standard error computeCI: computes confidence interval ccc: clears workspace and figures filldata, getaxis, getaxpos, getcolors, plotsproperties, setnox, setnoy: helper functions for figure-making METHODOLOGICAL INFORMATION 1. Description of methods used for collection/generation of data: Motor data was collected using Optotrak Motion Capture system at 100Hz, cognitive data was collected using MATLAB. Please refer to the manuscript for detailed methodology. 2. Methods for processing the data: For motor data, gaps in raw data were filled using the MATLAB function "interp1" with method "pchip". The data was filtered with a 4th order butterworth low pass filter with normalized cutoff frequency of 6/50. Then, heel strikes and toe offs were detected for each strided based on kinematic data. Heel strike was taken to be the position of the ankle furthest ahead in the direction of walking, and toe off was taken to be the position of the ankle furthest behind in the direction of walking. All other processing (including that for cognitive data) is described in the manuscript.