Optokinetic circular vection: a test of visual-vestibular conflict models of vection nascensy

The propensity to experience circular vection (the illusory perception of self-turning evoked by a rotating scene, CV) as reflected by its onset latency exhibits considerable interindividual variation. Models of CV nascensy have linked this delay to the time it takes the visual-vestibular conflict to disappear. One line of these conflict models (Zacharias and Young in Exp Brain Res 41:159-171, 1981) predicts that, across individuals, CV latency (CVL) correlates positively with the vestibular time constant (TC) and negatively with the vestibular motion detection threshold (vTHR). A second type of models (Mergner et al. in Arch Ital Biol 138:139-166, 2000) predicts only an increase in CVL with TC. We here examine which of these predictions can be experimentally substantiated. Also, we ask whether the relative weight W (O) of the optokinetic contribution to the perception of real self-turning could also be a factor influencing CVL. We conducted 5 experiments in 29 subjects measuring: (1) CVL, (2) the TCs of velocity perception and of accompanying nystagmus during rotation in darkness and (3) likewise for displacement perception, (4) vTHR, and (5) W (O) as revealed by discordant visual-vestibular stimulation. CVL correlated with the nystagmus TC recorded during velocity estimation but with none of the other vestibular TCs nor with vTHR. Confirming earlier findings, CVL shortened with rising scene velocity. Finally, CVL correlated inversely with W (O): the larger an individual's optokinetic weight, the shorter was his CVL. Taken together, our data favour the second type of models which invoke an antagonism between CV inhibition by the optokinetic-vestibular conflict and disinhibition by optokinetic stimulation. Idiosyncratic factors appear to strongly modulate the balance between inhibition and disinhibition, thus increasing CVL variability and obscuring the expected relation between CVL and TC.