Hero 5

In a 50 m pool, all swimmers performed a 100 m front crawl at maximum effort starting from the water and after a visual–acoustic signal. Participants were instructed to breathe only by their preferred side. An experienced operator recorded each individual performance by a high-speed video camera (GoPro Hero 5, GoPro Inc., San Mateo, CA, USA) at 120 frames per second. To avoid parallax errors, the camera was fixed to a mobile trolley that followed the head of the swimmer along the course. Only the 20 central meters of the swimming course were used for the analysis of basic kinematic variables to prevent the influence of the start and the turn. Four external markers were allocated on both sides of the pool for that purpose, two at 15 m and two at 35 m (two markers on each side), allowing the caption of the head in line with the markers during the tests.

Testing was conducted similarly to that previously described (49 (link)). Marmosets were separated from their cage mate into the upper-right quadrant of their home cage 8 minutes before the end of the drug pre-treatment time. Following this, an unfamiliar human intruder entered the room and stood 40cm from the cage, maintaining eye contact throughout the 2-minute test period. The intruder was a researcher wearing a realistic human mask (Masks Direct) unfamiliar to the marmoset and wearing familiar scrubs. The order of masks and infusions were counterbalanced across marmosets with at least 2 weeks between each test. Behavior was recorded using a camera (GoPro Hero 5) and a microphone was used to record vocalisations (Sennheiser MKE 400).

Each participant had two visits to a home simulation laboratory at the KITE Research Institute. In the first visit, informed consent was obtained and clinical assessments, including the FMA-UE, the ARAT, the MoCA, and the Motor Activity Log-30, were carried out to ensure participant eligibility and provide an overview of their hand function. In the second visit, the researcher demonstrated the procedure for using a head-mounted egocentric camera (GoPro Hero 5, GoPro Inc., CA, USA), following a previously reported protocol [34 (link)]. After the demonstration, participants familiarized themselves with using the camera. They subsequently carried out a list of daily tasks (S1 Appendix) in six different room settings in the laboratory (i.e., living room, dining room, bedroom, washroom, kitchen, and hallway), while recording egocentric videos. Participants were asked to carry out the tasks as they normally would (i.e., the grasp type was not constrained). The researcher discussed with the participants their daily routines to agree on representative ADLs for home recordings. Participants were encouraged to record on different days or at different times during a day to capture diverse activities. After the two study visits, participants self-recorded their daily routines at home to collect three sessions of 1.5 hour-long recordings and returned the videos to the researcher.

All scenarios were video-recorded by 3 GoPro Hero 5 and 7 Black Edition (GoPro Inc) video cameras mounted on participants and dispatched around them for later analysis. The camera setup was standardized. The investigators double-checked whether the questionnaires were fully and accurately completed. Data was collected using a REDCap database web app (REDCap, Vanderbilt University) hosted at Geneva University Hospitals and interfaced on an iPad Pro iOS 12.4 (Apple Inc). Neither follow up nor retention plans were required.

Behavioral data were collected for each climb to compute the climber’s fluency scores. The participant wore a light and an IMU (HIKOB FOX^®, Villeurbanne, France) on the back of his harness. The sensor recorded the signal from an accelerometer, a gyroscope, and a magnetometer at 100 Hz. Ascents were filmed at 29.97 fps on 1920 × 1080 pixel frames with GoPro Hero 5^® (GoPro Inc., San Mateo, CA, United States) cameras covering each entire wall (i.e., one camera per wall). The holds of the route were instrumented with the Luxov^® Touch system¹ (Arnas, France) that uses a capacitive sensing technology to detect and record the time of contact on each hold. For the present study, these data (i.e., the signal of the touch) was used only to compute the climbing time (CT) (from the climber’s first movement – either with a hand or foot – from the starting position until he touched the last handhold).

During each active free play session, participants wore a MetaMax 3B portable indirect calorimetry unit (Cortex Biophysical Gmbh; Leipzig, Germany) and an ActiGraph GT3X+ accelerometer (ActiGraph Corporation; Pensacola, FL, USA) on their right hip and non-dominant wrist. During each session, a member of the research team used a Go-Pro Hero 5 (GoPro, Inc, San Mateo, CA) camera to video record participants for subsequent direct observation coding of activity type.