Selective recall of surprising visual scenes. An experimental note on Seligman and Yellen's theory of dreams

Seligman and Yellen (Behav. Res. Ther. 25, 1-24, 1987) reported that dream fragments that are unrelated to the preceding fragments of the same dream are recalled as being more vivid and colorful and are remembered in more detail than dream fragments that are plot related and that lack an element of surprise. The authors take this as support for their hypothesis that while one dreams, two distinct types of visualization occur: one consisting of sudden, emotional bursts and the other occurring as a product of cognitive integration. An alternative explanation of Seligman and Yellen's data is that surprisingness of visual scenes is, in itself, sufficient to produce vivid, detailed, and colorful recall. In order to test this alternative hypothesis, 239 volunteers evaluated target slides that were preceded by either related or unrelated slides. When preceded by unrelated slides, the target slide was experienced as more surprising, was forgotten less often, and was evaluated as more colorful, more detailed, and more vivid. It is concluded that the data given by Seligman and Yellen do not constitute unequivocal support for their hypothesis.