Ocular torsion: rotations around the "WHY" axis.

BACKGROUND: Traditional teaching holds that there is a partial compensatory countertorsion after head tilt because the intorters in the eye on the side of the head tilt and the extorters in the contralateral eye are stimulated. This teaching is inconsistent with a number of clinical observations.
METHODS: Review of existing literature, reanalysis of data from the investigator's previous experiments, and inductive and deductive reasoning were used to reconcile inconsistencies and present a theory on why torsional movements occur.
RESULTS: The inconsistencies can be reconciled if one considers that during the dynamic phase of head tilt, there is an alternating series of intorsional and extorsional movements of both eyes. Each eye has slow dynamic compensatory counterrolling phases that serve as torsional "doll's-head" movements to stabilize the image during head tilt. This counterrolling is partially eliminated by a series of anticompensatory torsional saccades in the direction of head tilt, which is in contrast to traditional teaching.
CONCLUSION: Dynamic compensatory counterrolling occurs during head tilt. It is largely eliminated by anticompensatory torsional saccades in the opposite direction so that by the end of head tilt only minimal static countertorsion remains. The dynamic compensatory counterrolling motion is necessary to minimize peripheral visual movement during head tilt. The elimination of most of the counterrolling by the end of head tilt is necessary to preserve convergence and stereopsis.