This study asked whether individual differences in the influence of vision on postural stability could be used to predict the strength of subsequently induced visual illusions of self-motion (vection). In the experiment, we first measured spontaneous postural sway while subjects stood erect for 60 s with their eyes both open and both closed. We then showed our subjects two types of self-motion display: radially expanding optic flow (simulating constant velocity forwards self-motion) and vertically oscillating radially expanding optic flow (simulating constant velocity forwards self-motion combined with vertical head oscillation). As expected, subjects swayed more with their eyes closed (compared to open) and experienced more compelling illusions of self-motion with vertically oscillating (as opposed to smooth) radial flow. The extent to which participants relied on vision for postural stability—measured as the ratio of sway with eyes closed compared to that with eyes open—was found to predict vection strength. However, this was only the case for displays representing smooth self-motion. It seems that for oscillating displays, other factors, such as visual–vestibular interactions, may be more important.