TEMPORAL GROUPING EFFECTS IN IMMEDIATE RECALL - A WORKING-MEMORY ANALYSIS

The presence of temporal pauses during list presentation can markedly improve immediate memory for a sequence of verbal items. A series of e xperiments analysed this effect using Baddeley's (1986) model of worki ng memory. Experiment 1 showed that the effect of temporal grouping on memory for visual sequences was removed by either articulatory suppre ssion or reciting random digits. Experiment 2 indicated that effects o f temporal grouping were insensitive to the word length of the items. Experiment 3 showed that articulatory suppression did not remove the t emporal grouping effect for auditory lists. Experiment 4 showed that t he temporal grouping effect was insensitive to the phonemic similarity of the items. The effects of concurrent articulation suggest that gro uping influences the phonological loop component of working memory. Ho wever, the working memory model is insufficiently well specified to ac count for the insensitivity of grouping effects to word length and pho nemic similarity. The main findings could be simulated by a connection ist model of the phonological loop, which invokes a context timing sig nal (Burgess & Hitch, 1992, in press), This assumed that pauses during list presentation affect the timing signal in a similar way to the pa use before list presentation and made some novel predictions.