GROUPING IN PRIMARY MEMORY - THE CASE OF THE COMPOUND SUFFIX

Appending a nominally irrelevant item, or ''suffix,'' to the end of a short list of items impairs recall of the list. Appending a second suc h item, however, does not increase the impairment. The research report ed here shows that the impairment can in fact be increased if the suff ix items are physically dissimilar. Thus, Experiments 1-4 show that me mory for a sequence of digits is impaired more by the addition of two zeros uttered in different voices than by either a single zero or two zeros uttered in the same voice. Experiment 5 shows a similar pattern of results in the visual modality, with physical similarity defined by typefont. The findings are contrary to at least two theories of the s uffix effect but can be accounted for by assuming that physically simi lar items tend to form a cohesive group.