Adaptation and habituation of the vestibulo-ocular reflex in intact and inferior olive-lesioned rats.

The gain of the vestibulo-ocular reflex (VOR) of intact pigmented rats was adaptively modified by training protocols that created a visual-vestibular conflict. For training, head restrained animals were oscillated on a turntable in front of an optokinetic pattern projected onto a cylindrical wall. The optokinetic pattern either moved the same amplitude with the animal ('in-phase': 0.05 Hz +/- 20 degrees/s) or opposite in direction ('out-of-phase': turntable and pattern 0.05 Hz +/- 10 degrees/s each). VOR responses were tested in darkness before and after each 8 min training period for a duration of 40 min. During 'out-of-phase' training the gain of compensatory eye movements measured in light was close to 2 from the beginning on and the VOR tested in darkness increased in gain progressively from 0.48 (+/- 0.12) to 0.9 (+/- 0.3; P less than 0.05) in 5 out of 7 rats. Two rats did not adapt their VOR gain. Phase values decreased slightly by about 10 degrees. During 'in-phase' stimulation compensatory eye movements were almost completely suppressed (gain close to 0) from the beginning on and the VOR tested in darkness decreased gradually in gain from 0.62 (+/- 0.17) to 0.13 (+/- 0.1; P less than 0.001) in all 6 trained rats. Phase values decreased in parallel from 151 degrees to 119 degrees (P less than 0.01). The effectiveness of the 'in-phase' training paradigm in the absence of compensatory eye movements indicates that retinal image slip is the relevant signal for adaptation. In seven rats with histologically verified almost complete inferior olive (IO) lesions (chemically induced at least 45 days prior to training), 'out-of-phase' and 'in-phase' stimulation evoked compensatory eye movements with gains comparable to those in intact rats. VOR parameters measured in darkness were altered with respect to those of control rats. Gain differed extremely between individuals and phase lag re acceleration was in all IO-lesioned rats larger than in intact rats. The time constant of the VOR in response to table velocity steps was significantly longer (17 s +/- 4) than in intact rats (11 s +/- 3). Training did not alter the gain of the VOR in 5 out of 7 IO-lesioned rats. One rat increased its gain during 'out-of-phase' training in the first, but not during a second training session (and not during 'in-phase' training) and another rat decreased its gain during 'in-phase' training (but not during 'out-of-phase' training).(ABSTRACT TRUNCATED AT 400 WORDS)