Control of the head in response to tilt of the body in normal and labyrinthine-defective human subjects.

1. Head movement responses to discrete, unpredictable tilts of the trunk from earth upright were studied in normal and labyrinthine-defective (LD) subjects. Tilts of the seated, restrained trunk, were delivered in pitch and roll about head-centred axes and approximated raised cosine displacements with peak amplitudes of 20-30 deg and durations of 1.5-2 s. Subjects performed mental arithmetic with eyes closed or read earth-fixed text. 2. At the onset of tilt the head momentarily lagged behind the trunk because of inertia. Subsequently, head control varied widely with three broad types: (i) head relatively fixed to the trunk (in normal subjects and some patients); (ii) head unstable, falling in the direction of gimbal tilt (typical of acute patients for pitch motion); (iii) compensatory head movement in the opposite direction to gimbal tilt (observed consistently in normal subjects and in well-adapted patients). 3. EMG was well developed in subjects with compensatory head movement and consisted of an initial burst of activity at minimum latencies of 25-50 ms (means 72-108 ms), followed by a prolonged peak; both occurring in the 'side up' neck muscles, appropriate for righting the head. These muscles are shortened during the initial head lag so the responses cannot be stretch reflexes. In normal subjects their origin is predominantly labyrinthine but in patients they may be an 'unloading response' of the neck. 4. Head stability in space was superior with the visual task for all subjects but vision only partially compensated for labyrinthine signals in unstable patients. 5. Modelling the responses to tilt suggests that, in LD subjects, the short-latency burst could be driven by signals from the neck of the relative acceleration between head and trunk tilt. The longer latency EMG could be driven by a signal of head tilt in space. Normally, this signal is probably otolithic. In patients it could be synthesized from summing proprioceptive signals of position of head on trunk with trunk tilt.