The differentiation of "non-organic" limb weakness from genuine paralysis is sometimes difficult in neurological practice. To address this problem, we developed a computerized quantitative method, based on the Hoover's test principle, that determines the extent of involuntary limb activation when contralateral movement is performed. Measurements of hip or arm extension isometric force are performed during direct maximal voluntary effort and during contralateral hip flexion. Maximal involuntary/voluntary force ratio (IVVR) is calculated. IVVR of the lower limbs in ten healthy subjects was 0.614, 0.044 (mean, SEM). Similar results were obtained from seven patients with genuine weakness and in the non-affected limbs of nine patients with "non-organic" mono- or hemiparesis. In contrast, IVVR in the affected limbs in the "non-organic" group was markedly increased (2.48, 0.61; P < 0.001). The same pattern was elicited in the upper limbs (2.27, 0.46 vs 0.406, 0.06; P < 0,001). We conclude that Hoover's sign in "nonorganic" paralysis is a preservation or increase of a normal synkinetic phenomenon. Quantitative measurement of the IVVR can serve as a useful ancillary test in diagnosing non-organic weakness in either lower or upper limbs.