A note on the masking of pictures.

An experiment is described in which photographs of everyday objects were masked by frequency-filtered random visual noise. The masking functions obtained were of the same type as those obtained by masking letters and words with random noise or a pattern mask. That is, the high-frequency mask produced a type A function while the low-frequency mask produced a type B function. This result is discussed in terms of the general applicability of models of visual information processing constructed on the basis of experiments with letter or word stimuli. It is suggested that spatial-frequency concepts may usefully be employed to describe the relevant features of different types of mask.