The precise role of cognitive control on optic flow processing during locomotion has rarely been investigated. Therefore, this study aimed to determine whether coping with unreliable visual inputs during walking requires attentional resources. Twenty-four healthy young adults walked on an instrumented treadmill in a virtual environment under two optic flow conditions: normal (NOF) and perturbed (POF, continuous mediolateral pseudo-random oscillations). Each condition was performed under single-task and dual-task conditions of increasing difficulty (1-, 2-, 3-back). In all conditions, subjective mental workload was estimated (raw NASA-TLX). For kinematic variables, mean, standard deviation, statistical persistence and step-to-step error correction were computed from gait time series in mediolateral and anteroposterior directions. For EMG variables of soleus and gluteus medius, the full width at half maximum and the variance ratio were calculated. Performance on N-back tasks was assessed using mean reaction time and d-prime. Cognitive performance was not affected by simultaneous walking in any optic flow condition. Gait variability was altered under POF compared to NOF, so that young adults sought to counteract those perturbations by adopting an effortful gait control strategy, independently of concurrent working memory (WM) load. Increasing WM load led changes first at the neuromuscular level, then at the behavioral level, with a prioritization of gait control in the mediolateral direction. Interestingly, dual-tasking lowered the effects of POF but in the anteroposterior direction only. These findings and their theoretical implications provide valuable insight into the complex interaction effects of cognitive and visual constraints on gait control during treadmill walking.

Subjects:

• Healthy young adults, kinematic data: N = 24, 12 men and 12 women, age = 21.67 +/- 2.28 years old.
• Healthy young adults, EMG data: N = 20, 10 men and 10 women, age = 21.80 +/- 2.42 years old.

The testing session was composed of three blocks performed in a randomised order: (1) three N-back tasks in a seated position (single-task cognitive performance, STC), (2) walking under normal (congruent) optic flow (NOF), and (3) walking under perturbed (continuous mediolateral pseudo-random oscillations) optic flow (POF). In the latter two blocks, the walking tasks were performed under both single-task (STW) and dual-task (DTW) conditions (i.e. walking while performing the N-back tasks). Participants were asked to walk naturally while looking straight ahead. The treadmill speed was adjusted to their preferred walking speed. The blocks (2) and (3) began and ended with a STW condition while the three DTW conditions were performed in a randomized order between the two STW conditions (Schaefer et al., 2015). Under dual-task conditions, no task priority instructions were given (Schaefer et al., 2015). Each condition lasted for 3 minutes. Hence, a total of thirteen experimental conditions were performed.

• "STC" = Single-task cognitive performance;
• "STW" = Single-task walking;
• "DTW" = Dual-task walking;
• "NOF" = Normal optic flow;
• "POF" = Perturbed optic flow;
• "1b" = 1-back task;
• "2b" = 2-back task;
• "3b" = 3-back task.

[See paper for exact definitions / descriptions]

Foot placement kinematics (200 Hz) and surface electromyography (EMG) of soleus and gluteus medius (1000 Hz) were recorded. For kinematic variables, mean, standard deviation (variability), statistical persistence and step-to-step error correction were computed from gait time series in mediolateral and anteroposterior directions. For EMG variables, duration and variability of muscle activation were calculated from the full width at half maximum (FWHM) and the variance ratio, respectively. Performance on N-back tasks was assessed with mean reaction time for target trials (RT, in second) and d-prime (d’, a.u.).

The dataset is composed of 5 files as follows:

• 1 text file (README_All-files.txt) which contains information about the database and how the files are organized;
• 1 Microsoft Office Excel spreadsheet (Cognition_All-participants.xlsx) which contains the results of the N-back task for each experimental condition; 
• 1 Microsoft Office Excel spreadsheet (RAW-NASA-TLX_All-participants.xlsx) which contains the results of the NASA-TLX questionnaire on perceived cognitive load for each experimental condition;
• 1 Microsoft Office Excel spreadsheet (Gait_All-participants.xlsx) which contains the results of the kinematic and electromyography analysis for each experimental condition; 
• 1 zip file (Gait-parameters.zip) which contains:
  • 1 text file (README_Gait-parameters.txt) which contains information about the gait parameters available in each of the 24 *.csv files;
  • 24 comma-separated value files (P01 to P24.csv) which contain the kinematic and EMG time series for each experimental condition.

Excel spreadsheets and *.csv files could be opened using open source alternatives such as LibreOffice (Calc).

Contact: Leslie M. DECKER, Associate Professor (PhD, HDR), Université de Caen Normandie (France): leslie.decker@unicaen.fr