Interference in Memory for Tonal Pitch: Implications for a Working-Memory Model
Total Page:16
File Type:pdf, Size:1020Kb
MeITWry de Cognition 1992. 20 (3), 314-320 Interference in memory for tonal pitch: Implications for a working-memory model THOMAS PECHMANN and all..BERT MOHR Unioersitiit des Saarlandes, Saarbriicken, Germany The degree of interference caused by different kinds of stimuli on memory for tonal pitch was studied. Musically trained and untrained subjects heard a sequence of two tones separated by an interval of 5 sec. The tones were either identical in pitch or differed by a semitone. Subjects had to decide whether the tones were identical or not. The interval was filled with tonal, verbal, or visual material under attended and unattended conditions. The results revealed clear group differences. Musically trained subjects' retention of the first test tone was only affected by the interposition of other tones. In contrast, the performance of musically untrained subjects was also affected by verbal and visual items. The findings are discussed in the framework ofBaddeley's (1986) working-memory model. In a number of experiments, Deutsch studied the ef Deutsch & Feroe, 1975). The explanation based on lat fect of short sequences of tones on the retention of tonal eral inhibition might be further specified by reference to pitch (see Deutsch, 1982, for an overview). These ex the concept ofthe critical bandwidth (cf. Plomp & Levelt, periments principally employed the same method, which 1965). Since the size of the critical band depends on its is illustrated in Figure 1. Subjects heard a first test tone center frequency, the distance between two tones later- . followed by a series of six interpolated tones varying in ally inhibiting each other might depend on the range of pitch. All tones had a duration of200 msec and were sep frequencies being investigated. However, this issue has arated by short pauses of300 msec. At the end, there was not been taken up in the literature. a longer pause of2,000 msec duration and then a second Iflateral inhibition is the crucial mechanism, the most test tone was presented that was either identical in pitch efficient pitch eraser should consist oftonal material. But with the first test tone or not. The subjects were asked what about speech or other acoustic stimuli? Do they have to compare the pitches of the first and second test tones. a detrimental effect as well, and is it within the same The dependent variable was the number of errors. Deutsch range? In one ofher earlier experiments, Deutsch (1970) mainly varied the frequency and the position ofa critical studied the effects ofinterpolated tones and spoken.num interpolated tone as independent variables, holding the bers on memory for tonal pitch using her standard para timing skeleton and the number ofintervening tones con digm (described above). Her results displayed a detri stant. She found maximal interference if the critical tone mental effect of tonal items but no effect of interpolated was the second one in the series ofintervening tones and speech. She concluded that to obtain interference effects on if it differed from the initial test tone by two thirds of a memory for pitch, one has to use highly specific material. tone. In this condition, subjects made 51% errors if the Lateral inhibition of tonal pitch is a function of both two test tones were different and 21% if they were iden frequency and temporal distance. In a number ofexperi tical (Deutsch, 1975). ments, Deutsch also varied the serial position of the crit Deutsch explained this effect in terms oflateral inhibi ical interpolated tone and found that it produced substan tion and provided additional evidence for this claim by tially more interference when it was included early rather showing disinhibitory effects on the first test tone by in than late in the sequence. Thus, if the memory for tonal hibition of the second interpolated tone (Deutsch, 1984; pitch is only disrupted by lateral inhibition, its represen tation should be firm and safe against any attempt offur ther interference once it has survived the stage sensitive This is a revised version of a paper presented at the Fourth Confer to lateral inhibition. ence of the European Society for Cognitive Psychology at Como, Italy, So far, we might assume that memory for pitch in a in September 1990. Both authors contributed equally to this research. We are grateful to Bernhard Morsch for collecting part of the data. recognition task like that designed by Deutsch is indepen Thanks are also due to Herb Clark, Johannes Engelkamp, Pim Levelt, dent of the musical expertise of individual subjects, since Antje Meyer, Piet Vas, and Pienie Zwitserlood, who commented on the mechanism of lateral inhibition can hardly be influ an earlier draft of this paper. In addition, we would like to thankRobert enced by training. But what about later stages? In Deutsch's Crowder, Robert Proctor, and two anonymous reviewers for helpful suggestions concerning revision. Correspondence should be addressed experiments, the tonal representation had to somehow to T. Pechmann, Universitiit des Saarlandes, Fachrichtung Psychologie, be maintained until the second test tone was presented. 0-6600 Saarbriicken, Germany. Does pitch maintenance during this retention phase re- Copyright 1992 Psycho nomic Society, Inc. 314 MEMORY FOR TONAL PITCH 315 This claim might be substantiated by reference to the ....>-l working-memory model proposed by Baddeley (l986). 200 lIst Test 8 Tone I rt> This model is characterized by a device called the central executive and a number ofslave systems. The central ex 300 Pause ecutive is responsible for monitoring and controUing the smooth functioning ofthe slave systems and for distribut 200 ing attentional resources if necessary. The function ofthe Slot 1 slave systems is to preserve the incoming perceptual in =l formation as long as necessary for the task in which the 300 Pause subject is involved. Two such slave systems have been studied in detail by Baddeley and co-workers: the visuo 200 Slot 2 spatial sketchpad and the articulatory loop. They are de H signed as active devices: to keep the information in store, ::l 300 Pause rT it must be regularly refreshed. In the case of the visual CD I-f system, it is speculated that eye movements fulfill this func '0 200 Slot 3 --g. tion. In the case of the auditory system, the proposed ill mechanism is subvocal rehearsal. rT CD The articulatory loop is designed to process linguistic 300 Pause P- i-'- material. It consists of a phonological store and an ar rT ro ticulatory refreshing mechanism. The assumption of the 200 Slot 4-~ latter is supported, for instance, by experiments demon strating that suppressing the (subvocal) articulation of to be-learned verbal material by a concurrent (overt) articula 300 Pause tory task affects the retention of the material (Murray, 1967). 200 Slot 5 One might assume that musical pitch is retained by a very similar mechanism: a kind of "tonal" loop. Ifsub 300 Pause jects are asked to retain the pitch of a particular tone for some time, they usually report afterwards that they in ternally rehearsed the tone. However, the efficiency and 200 Slot 6 automaticity of a tonal loop may be crucially dependent on subjects' musical training. The articulatory loop is as 2000 Pause sumed to work rather automatically, that is, without the requirement ofcentral resources. In contrast, at least for musically untrained people, tonal rehearsal may be a much 200 /2nd Test Tone more controlled process that requires additional alloca tion ofattention. Thus, musical training may significantly affect the functioning of the loop. Figure 1. Timing skeleton used in Deutsch's experiments (see Deutsch, 1975). With respect to the processing of speech and music in working memory, the additional question arises whether they are processed in a common acoustic system or in quire a substantial amount of attention? If the answer is two different, specialized systems. Thebasic line ofargu positive, does subjects' experience with musical material mentation underlying memory research and the working play any role? memory model in particular is that if the working of a In a study in which spoken numbers were used as in system is not affected by the concurrent processing of a terpolated stimuli, Deutsch (l970) also included a condi particular kind ofmaterial, then this material is processed tion in which the subjects had to recall the list of num in another system. According to this argument, the find bers. But even in this condition, the spoken words did ings obtained by Deutsch (1970) that concurrent speech not affect subjects' memory for tonal pitch. Thus, one does not affect the retention of tonal pitch suggest that might conclude that the maintenance of tonal pitch does music and speech are processed in two different memory not require any substantial amount of central capacity. systems. But then the opposite should also be true: the However, we do not know whether Deutsch used musi processing of linguistic items should not be affected by cally trained or untrained subjects, yet this factor might the concurrent processing of musical stimuli. playa significant role. It is conceivable that musically This issue has been taken up in a recent study by Salame trained subjects differ from musically untrained subjects and Baddeley (1989). They investigated the effects ofcon in the amount of attention required to maintain a tonal current speech, noise, or music on the retention ofvisually representation over time and under different conditions. presented verbal material. They found that noise had no 316 PECHMANN AND MOHR detrimental effect. Music, however, significantly dis Materials and Procedure. We used the experimental procedure rupted short-term memory performance compared with si developed by Deutsch as a timing skeleton for all conditions, fill lence, although it was less disruptive than speech. For sub ing the slots for interpolated items with different classes of stim jects who had previously performed a separate memory uli.