Random dot patterns (RDPs) have been used to stimulate in isolation either of two binocular mechanisms for motion in depth: interocular velocity difference (IOVD), and changing disparity over time (CDOT). First, we examined how these stimuli isolate either mechanism using models based on motion/disparity energy detection. In the model, the IOVD mechanism calculates the difference in motion signal between the two eyes, and the CDOT mechanism calculates the difference in disparity signal between sequential stereo images. The simulation revealed that uncorrelated (0% correlation) RDPs are useful to isolate either of the two mechanisms, while the contrast-reversed version (anticorrelation) of RDPs may not. Second, we compared the IOVD model predictions with experimental data from a previous study for motion in depth with various contrasts, displacements and vertical shifts between the two eyes. The simulation showed that the model predicts the general trends of the effects of contrast, displacement and vertical shift shown in the data. This suggests the physiological plausibility of the energy-based model of motion in depth.