The ability to voluntarily stabilise the head in space exhibited by two subjects during lateral rhythmic oscillations of the trunk has been investigated before, during and after a prolonged period of microgravity (microG) exposure. In flight acquisitions were performed onboard the Core Module of the Russian Space Station MIR as part of the T4 "Human Posture in Microgravity" experiment of the 179-days ESA-RKA mission EUROMIR-95. Data collection and kinematic analysis were performed by means of a space-qualified version of the automatic motion analyser ELITE. Head stabilisation in space strategy was estimated by means of the head anchoring index and cross-correlation analysis. Results show that head orientation may be well stabilised about the roll axis both with and without the presence of visual information. This was true despite the expected reduction in vestibular efficiency and muscular proprioception occurring in-flight. In one subject, however, vision was found to improve head stabilisation in space post-flight, presumably to recover from the postural deficiency induced by the long-term microG exposure. Head stability during trunk movements was achieved with either over-compensatory (out-of-phase), under-compensatory (in-phase) or mixed movement strategies, as was attested by the analysis of cross-correlation functions between head and shoulder movements. In weightlessness, vision occlusion seemed to influence the choice of the strategies to be used as well as the reduction of movement variability. The feedforward nature of compensatory head movements suggests that head stabilisation could be based in weightlessness on the internal postural body scheme, supposed to be adapted to the weightless environment within 5 months of microG exposure.