Intonation can convey information about how lists are structured into groups, as well as about specific item positions within a group. In Bari Italian, this function is expressed by three different tunes a) a rising contour, signalling that the list has not yet been completed; b) a high-rising contour, marking the penultimate item, i.e. signalling that the end of the list is approaching; c) a falling contour, marking the last item, i.e. cueing the end of the sequence. In this paper we explore the effects of such intonational information on working memory. In particular, we demonstrate that when listeners are requested to recall spoken nine-digit sequences by strictly following their serial order, their performance is significantly better when lists are characterised by tunes of the type described above, compared to sequences whose items are marked by a neutral, peak accent and/or are grouped by inserting a silent pause. We also observed that recall of items marked by specific contours at positions 3, 6 and 9 is particularly enhanced at these positions, whereas in sequences also containing intonational cues to items in penultimate position (2, 5 and 8) recall of those items is not equally improved. Therefore, it appears that in serial recall of spoken sequences, even when a large number of specific intonational cues to serial positions are available, listeners can use only a selection of them.
Intonational cues to item position in lists: evidence from a serial recall task
SAVINO, Michelina;BOSCO, Andrea;
2014-01-01
Abstract
Intonation can convey information about how lists are structured into groups, as well as about specific item positions within a group. In Bari Italian, this function is expressed by three different tunes a) a rising contour, signalling that the list has not yet been completed; b) a high-rising contour, marking the penultimate item, i.e. signalling that the end of the list is approaching; c) a falling contour, marking the last item, i.e. cueing the end of the sequence. In this paper we explore the effects of such intonational information on working memory. In particular, we demonstrate that when listeners are requested to recall spoken nine-digit sequences by strictly following their serial order, their performance is significantly better when lists are characterised by tunes of the type described above, compared to sequences whose items are marked by a neutral, peak accent and/or are grouped by inserting a silent pause. We also observed that recall of items marked by specific contours at positions 3, 6 and 9 is particularly enhanced at these positions, whereas in sequences also containing intonational cues to items in penultimate position (2, 5 and 8) recall of those items is not equally improved. Therefore, it appears that in serial recall of spoken sequences, even when a large number of specific intonational cues to serial positions are available, listeners can use only a selection of them.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
