The research reported here was initially stimulated by a contradiction in the literature that suggested there was no significant performance advantage for manipulator operators employing stereo versus conventional TV systems, while typical direct-viewed results indicated that binocular performance is always superior to monocular performance in tasks requiring depth judgement or distance estimation. This paper presents an analysis of the interaction of several important variables involved in visual display research, with particular reference to our own research comparing mono and stereo TV displays in simulated underwater environments. Three experiments were conducted in our lab to assess the impact of degraded visibility on remote manipulator operator performance using either mono or stereo TV. Subjects were required to perform tasks which differed markedly in the number and type of depth cues available. As predicted, the results indicated that stereo performance was superior to mono under most conditions tested; however, the amount of improvement was shown to be a complex function of visibility, task, and learning factors. Conclusions and recommendations for further research aimed at understanding the relative contributions of these and additional display factors are presented for discussion.