VERBAL MEMORY SPAN IN CHILDREN - SPEECH TIMING CLUES TO THE MECHANISMS UNDERLYING AGE AND WORD-LENGTH EFFECTS

According to the prevalent articulatory loop hypothesis of memory span (Baddeley, 1986), subjects recall items from a decaying phonological store that is refreshed by a covert articulatory process. The rate of covert articulation is said to determine memory span. The present research indicates that this account underemphasizes effects of (1) other covert mnemonic processes and (2) overt pronunciation in recall. Memory span and maximal speech rate for sets of one-, two-, and three-syllable words were examined in groups of children with mean ages of 4;5 and 8;8. Although the usual linear relation between the maximal speech rate and memory span was replicated, speech timing measurements based on the memory responses revealed that age and word length effects had different effects on the timing of responding. Whereas word length affected the duration of words in the response but not silent intervals, age affected the duration of silent, preparatory and interword intervals in the response but not the duration of words. These results are discussed in light of the hypothesis that the speeds of multiple, mnemonically relevant covert and overt processes affect memory Span. (C) 1994 Academic Press, Inc.