An experiment is reported the object of which was to check whether a small amount of nonspecific experience in perceiving random-dot stereograms could facilitate the perception of a previously unseen stereogram. The mean stereopsis perception time of a group of totally naive subjects was found to be significantly slower than that of a group who had previously been shown two different stereograms. Closer inspection of the data showed that this difference was primarily due to approximately one third of the naive group who were much slower than the ‘experienced’ group. It is therefore suggested that nonspecific experience provides most initial help for relatively slow perceivers, since many naive subjects can perform as well as those with prior experience of other stereograms.