Vestibular and neck proprioceptive signals are known to be used in jud
ging the locations of objects in space and relative to the body Given
that these signals are asymmetric in patients with spasmodic torticoll
is, one would expect such patients to have abnormal spatial perception
. We tested this idea by measuring patients' perception of visual stra
ight ahead (VSA) under various conditions: with the body in its primar
y position, i.e. with the head and trunk as closely aligned as possibl
e, and after well defined passive rotations of the head and/or trunk.
In the primary body position, patients' VSA direction showed considera
ble variations which were similar however to those of normal subjects;
it was independent of torticollis direction, of the head torque it pr
oduced, and of the weak spontaneous nystagmus recorded in seven of the
10 patients. After whole-body rotations, i.e. where head and trunk un
derwent the same motion, the VSA was shifted in both patients and norm
al subjects, and in both groups the shift was symmetrical after rotati
ons to the right or left. After motions whew the trunk rotated under t
he stationary head (neck proprioceptive stimulation) or the head on th
e stationary trunk (combined vestibular and neck stimulus), the VSAs o
f normal subjects coincided rather well with their head midsagittal pl
anes, whereas the VSAs of patients were shifted considerably towards t
he trunk again in a symmetrical way We suggest two mechanisms to expla
in the findings in patients: (i) a central compensation which restores
symmetry of the afferent inflow in the patients (unlike the motor eff
erence); (ii) shifting of the reference for the VSA from the head towa
rds the trunk, because the trunk is a more reliable egocentric referen
ce than the head in the patients. Our findings do not support the assu
mption that asymmetries in afferent inflow are responsible for the asy
mmetry of motor output in spasmodic torticollis.