When oscillating the upper limbs together in the parasagittal plane, movements coordination is lower (i.e., variability of the interlimb relative phase is higher) in antidirectional (ANTI) than in isodirectional (ISO) coupling. In contrast, we previously observed that for arm movements in the horizontal plane, the coordination was worse in ISO than ANTI and the energetic cost of postural activities was higher in ISO. Having hypothesised that the higher postural cost was one factor responsible for the coordination deficit in horizontal ISO, we measured the oxygen uptake ([Formula: see text]) in parasagittal movements, expecting that in this case too, the postural cost is higher in the less-coordinated mode (ANTI). Breath-by-breath metabolic ([Formula: see text], [Formula: see text]) and cardiorespiratory (HR, [Formula: see text]) parameters were measured in seven participants, who performed cyclic flexions-extensions in the parasagittal plane with either one arm or both arms, in ISO or ANTI coupling and at 1.4, 2.2 and 2.6Hz. In each condition, the intermittent exercise (12s movement, 12s rest) lasted 264s. A force platform recorded the mechanical actions to the ground. The exercise metabolic cost ([Formula: see text]) was found to be significantly higher in parasagittal ANTI than ISO. The movement amplitude being equal in the two modes, the ANTI-ISO difference should be ascribed to postural activities. This would confirm that the less-coordinated coupling mode requires the higher postural effort in parasagittal movements too. When rising the movement frequency, [Formula: see text] increased and linearly correlated with the coordination loss. Comparison of parasagittal with horizontal movements showed that [Formula: see text] was lower in parasagittal ANTI than in horizontal ISO (the less-coordinated modes), while it was not different between parasagittal ISO and horizontal ANTI (the more-coordinated modes).