The Detection and Identification of Pictorial Changes between Successive Presentations

The present research examines the abilitv to detect and identify particular changes between successive pictorial scenes. This investigation focussed on two particular types of change in complex black and white pictures: the addition of objects or patterns, or the deletion of similar objects or patterns in a successivly presented picture. In Experiment 1, additions were found to be more easilv identified than deletions but only for pattern changes and not for objects. This difference decreased in magnitude with a longer interval between pictures of a pair. A potential problem with this experiment is that subjects were required to report five changes per picture pair and systematic output competition effects may have interacted with picture memory as a function of delav. In Experiment 2 the picture pairs contained only one change. The results showed that additions were more often detected than deletions and object changes more often than pattern changes. A planned comparison revealed that the detection of pattern-additions were significantly better than the detection of pattern-deletions. Experiment 3 again replicated this finding. The results of this research are discussed in terms of a dual-code theory and its utility in a variety of applied settings.