BackgroundBiomechanical studies of walk, especially walk on the circle, are scarce, while circles or curved tracks are frequently used during equestrian activities. To study horse–rider–circle interactions on the circle, the first steps would be to investigate how the unridden, freely walking horse is influenced by circular movement, and then add a rider. The aim was to study horse vertical trunk movements, and sagittal cannon angles (protraction–retraction) during walk in straight-line and on the circle without rider, and on the circle with a rider using minimal influence.MethodsTen horses were ridden by five riders, summing to 14 trials. Each trial included straight walk unridden (on concrete), and walk on 10 m diameter circles (left and right on soft surface) first lunged (unridden) and then ridden with minimal rider influence. Inertial measurement units (100 Hz) were positioned on the withers, third sacral vertebra (S3) and laterally on metacarpal and metatarsal bones (using self-adhesive bandage). Selected data were split in steps (withers and S3 vertical translations) or strides (cannon protraction–retraction) at maximum hind limb protraction, and range of motion (ROM), minima and maxima, and their timing, were extracted. Data were analyzed using mixed models with inner/outer/straight nested within unridden/ridden as fixed effect, and controlling for stride duration. Differences between: inner vs outer steps/limbs; the same step/limb unridden vs ridden; and the same step/limb straight vs inner/outer unridden; were examined for statistical significance at p < 0.05.ResultsInner limbs had smaller cannon ROM than outer limbs, for example, forelimbs when ridden (inner vs outer 62° vs 63°) and hind limbs when unridden (53° vs 56°). Forelimb cannon ROM was the largest for straight (65°). Hind limb ROM for straight walk (55°) was in-between inner (52–53°) and outer hind limbs (56–57°). Vertical ROM of S3 was larger during the inner (unridden/ridden 0.050/0.052 m) vs the outer step (unridden/ridden 0.049/0.051 m). Inner (0.050 m) and outer steps (0.049 m) unridden had smaller S3 ROM compared to straight steps (unridden, 0.054 m). Compared to when unridden, withers ROM was smaller when ridden: inner hind steps unridden/ridden 0.020 vs 0.015 m and outer hind steps 0.020 vs 0.013 m. When ridden, withers ROM was larger during the inner hind step vs the outer.ConclusionThe outer hind limb had greater cannon pro-retraction ROM, compared to the inner limb. Larger croup ROM during the inner step appears to be coupled to increased retraction of the outer hind limb. Knowledge of magnitudes and timing of the horse’s movements on the circle in unridden and ridden walk may stimulate riders to educate eye and feel in analyzing the execution of circles, and stimulate further studies of the walk, for example, on interactions with rider influence, natural horse asymmetries, or lameness.