Ni 6009

The final filtered fluorescence was measured with the two PDs mentioned before and the data acquisition was achieved with a NI6009 (National Instrument, United States) card. A Labview interface was implemented for controlling the data acquisition main parameters and to display the obtained signals in real-time for both PDs. During the experiments the sampling frequency was adjusted to 1 Hz and the recording time was 30 seconds for each sample. With the CPIII main emission peak being located in the lower spectral band and the PPIX peak being located in the upper spectral band, a stronger change of intensity in the lower band would indicate a change in CPIII concentration, while a stronger change in the upper band would indicate a change in PPIX concentration.

Participants performed two 5 s maximal voluntary contractions (MVCs) with the knee extensors of their self-reported dominant leg, interspersed with 1 min of passive rest [8 (link), 23 (link), 24 (link)]. The two MVCs were performed at baseline and as soon as possible after (77.50 ± 21.49 s) the RSA test. The measurements were carried out in a specific chair designed for maintaining hip and knee flexion at 90°, with participants secured by straps [8 (link), 23 (link), 24 (link)]. A metal rod was attached to the ankle and connected to a load cell with a maximum capacity of 100 kgf (MK Controle, São Paulo, SP, Brazil). The load cell signal was acquired by an analog data acquisition mode (NI 6009, National Instruments, Austin, TX, USA) sampling at 1000 Hz, using Labview software (National Instruments, Signal express, Austin, TX, USA). The recorded data were subsequently filtered by a second-order Butterworth filter and analyzed with custom designed software (MatLab R2015b, MathWorks, Natick, MA, USA). Before each test, the load cell was calibrated using known weights to create a linear regression model (r² > 0.99). The peak force of MVCs was defined as the mean force recorded during 100 ms of the force plateau [8 (link), 24 (link)].

The spherical treadmill was manufactured by custom milling with 6061 aluminum alloy. The treadmill has a concave surface at the end for placing the ball, which is supported by airflow. We fabricated foam balls (Last-A-Foam FR-7110, General Plastics, Burlington Way, WA USA) that are 10 mm in diameter using a ball-shaped file. We drew random patterns with black ink on the foam balls to provide a high-contrast surface for the ball tracking analysis. Fly behavior was videotaped from the side to capture any movement by a CCD camera (DCC1545M, Thorlabs) equipped with a machine vision camera lens (MVL6 × 12 Z, Thorlabs) and 950 nm long pass filter (FELH0950, Thorlabs). The acquisition frame rate for video recording was set to 8 Hz under IR light illumination at 970 nm (M970L4, Thorlabs). The stimulus signal from the Arduino is captured by NI-6009 (National Instrument) using a custom script written in MATLAB 2020b (Mathworks) to synchronize with the behavior video in data analysis.