Immediate and final recall of pictures and words with written or oral tests

Immediate and delayed recall of pictures and words was examined as a function of semantic or nonsemantic orienting tasks and the type of test (written or oral). As expected, semantic tasks generally led to greater final recall than nonsemantic tasks, with semantic tasks even producing positive recency on the delayed test. The evidence for a picture-word difference was largely restricted to the final recall of items involved in negative decisions; for such items the advantage of semantic tasks was apparent only for pictures. This suggests that "congruity" may be an important factor in picture-word differences, with such differences more apparent for the weaker items from negative judgments. Type of test did not seem to be a major factor in determining level of recall, suggesting that reinspecting the recall protocol during a written immediate test does not contribute substantially to final recall performance.