PMC

Joint torques indicate that campaniform sensilla on the trochanter encode the unloading of the leg during walking. (a,b) Torque and torque rate at the CTr joint of the right middle (RM) leg during stance. Bold black lines show the grand mean of all steps (n = 244 steps from N = 10 animals). Grey lines show means per animal. Magenta and blue lines indicate torques and torque rates above which G3/G4 afferent activity was confirmed in reduced leg preparations (‘sensory thresholds’, calculated from []). The vertical black line marks the mean onset of unloading, t_UL. (c) Schematic of expected G3/G4 response based on the mean torque and torque rate time courses. Leg unloading is expected to terminate G3 and initiate G4 afferent activity. (Online version in colour.)

Leg unloading is linked to loading of posterior neighbouring leg. (a) Touch-down events of all legs (top) and likelihood of touch-down per leg (bottom) relative to the onset of unloading, t_UL, in the right middle (RM) leg (n = 244 steps from N = 10 animals). Touch-downs of the right hind (RH) leg reliably precede t_UL in the middle leg with short latency (arrowhead). The box marks the step shown in b. (b) Exemplary stance phase showing the CTr torque of the right middle leg from the experiment (black) and the static model (grey). Horizontal bars on top indicate the stance phases of all legs. Vertical lines highlight the times of leg touch-downs. Touch-down induced changes in simulated torque provide a measure of a leg's unloading efficacy. (c) Simulated efficacy of legs in unloading the right middle leg, pooled across all steps. The 50% line corresponds to 50% of the mean torque measured during walking (a). (Online version in colour.)

Load-based inter-leg coordination in an insect. (a) Schematic side view of a stick insect during walking. Each leg is labelled as being a right (R) or left (L) and a front (F), middle (M) or hind (H) leg. If a leg touches down (e.g. RH), it may unload the leg in front (e.g. RM). (b) Example coordination pattern of stance (grey) and swing (white) phases for each leg. Legs are coordinated in a back-to-front sequence (black arrows). Stance phases of ipsilateral neighbouring legs overlap, providing potential time for load transfer (e.g. RM and RH). (c) Schematic front view of a stick insect leg during stance. The GRF induces high bending torques at the proximal CTr joint. Campaniform sensilla groups G3 and G4 on the dorsal trochanter are highly sensitive to the associated strain in the trochantero-femur. G3 is activated when dorsal bending torques increase (loading of leg), G4 when they decrease (unloading of leg). (d) Schematic of G3/G4 reflex pathways onto coxal muscles in active animals. Broken lines indicate functional motor effects. G3 afferent activity excites (+) the depressor (stance) muscle and inhibits (−) the levator (swing) muscle []. G4 afferent activity is assumed to have the opposite effect. Unloading induced by a neighbouring leg may reverse afferent activity from G3 to G4, thereby promoting the leg's stance-to-swing transition. (Online version in colour.)

Leg unloading coincides with a switch from stance to swing muscle activity. (a) Side view of an animal carrying a lightweight EMG backpack and motion capture markers (white circles) while stepping onto a force plate. (b) CTr torque of the right middle (RM) leg and simultaneously recorded activity of the levator muscle (light blue) and depressor muscle (light red) of an example step. Dots below EMG traces indicate muscle spikes detected based on amplitude. TD, touch-down; LO, lift-off. (c) Raster plot of detected muscle spikes (top) and likelihood of muscle activation (bottom) relative to t_UL of the right middle leg (n = 73 steps from N = 8 animals). Walking speed corresponds to the mean speed of the COM during stance. The box marks the step shown in b. (d) Time from t_UL to peak levator activity over unloading rate (r = −0.67, p < 0.001, n = 73 steps from N = 8 animals). Peak levator activity was determined from rectified and smoothed EMG signals (moving average filter with 80-ms window). The unloading rate corresponds to the mean CTr torque rate from t_UL to LO in each step. The blue line shows the corresponding predicted G4 afferent activity (calculated from []). (e) Leg movements during the stance phases in c plotted in joint angle space. Coloured dots indicate detected muscle spikes. Movement direction is from left to right (positive to negative protraction). Bold black lines show the mean movement trajectory. Black dots mark t_UL in individual stance phases. Histograms indicate the likelihood of muscle activation per 5° joint angle bins in the second half of stance (grey shaded area). Arrowheads mark the variable onset of the levator in two exemplary stance phases. At zero degree protraction and levation, the leg is perpendicular to the long body axis in the horizontal plane; at zero degree extension, the tibia is completely flexed. (Online version in colour.)