Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Interference and coding processes in verbal and visual short-term memory Ternes, Willi 1973

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1973_A1 T47_7.pdf [ 4.86MB ]
Metadata
JSON: 831-1.0101030.json
JSON-LD: 831-1.0101030-ld.json
RDF/XML (Pretty): 831-1.0101030-rdf.xml
RDF/JSON: 831-1.0101030-rdf.json
Turtle: 831-1.0101030-turtle.txt
N-Triples: 831-1.0101030-rdf-ntriples.txt
Original Record: 831-1.0101030-source.json
Full Text
831-1.0101030-fulltext.txt
Citation
831-1.0101030.ris

Full Text

INTERFERENCE AND CODING PROCESSES IN VERBAL AND VISUAL SHORT-TERM MEMORY by WILLI TERNES B.A., U n i v e r s i t y o f B r i t i s h C o l u m b i a , 1969 M.A., U n i v e r s i t y o f B r i t i s h C o l u m b i a , 1971 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n t h e Department of P s y c h o l o g y We a c c e p t t h i s t h e s i s as c o n f o r m i n g t o t h e r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA J u l y , 1973 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by hi s representatives. It i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Psychology Department of The University of B r i t i s h Columbia Vancouver 8, Canada D a t e July 2 0 1 9 7 3 ABSTRACT A d e t a i l e d e v a l u a t i o n of short-term verbal and v i s u a l coding e f f i c i e n c y as a f u n c t i o n of Interp o l a t e d a c t i v i t i e s , varying along the dimensions of a t t e n t i o n and s i m i l a r i t y , was undertaken i n t h i s research. The r e s u l t s show t h a t r e t e n t i o n losses due t o a t t e n t i o n demand changes are comparable In verbal and v i s u a l coding c o n d i t i o n s , r e g a r d l e s s of the modality of the Int e r p o l a t e d a c t i v i t y . In a d d i t i o n , r e t e n t i o n losses are l a r g e r when the same modality Is involved i n processing the stimulus and Int e r p o l a t e d task. Retention losses are Interpreted as demonstrating a c l e a r separa-t i o n of short-term losses due t o : a) a t t e n t i o n d i v e r s i o n ; and b) d i r e c t I n t e r f e r e n c e , w i t h a t t e n t i o n d i v e r s i o n accounting f o r a l a r g e r p a r t of t o t a l r e t e n t i o n l o s s e s . The data f u r t h e r suggest t h a t the maintenance of Information In verbal and v i s u a l STS r e l i e s t o a large extent on the a v a i l a b i l i t y of the common c e n t r a l processing c a p a c i t y , with modality s p e c i f i c coding processes determining the most app r o p r i a t e coding mode f o r a given stimulus s i t u a t i o n . ii TABLE OF CONTENTS ABSTRACT . I TABLE OF CONTENTS II L IST OF APPENDIX I TABLES I l l L IST OF APPENDIX II TABLES Iv L IST OF FIGURES . v ACKNOWLEDGEMENT vt GENERAL INTRODUCTION • EXPERIMENT I - I n t r o d u c t i o n 39 Method 42 R e s u l t s and D i s c u s s i o n 47 EXPERIMENT II - I n t r o d u c t i o n 55 Method 56 R e s u l t s and D i s c u s s i o n . . 57 DISCUSSION - EXPERIMENTS I and I I 61 EXPERIMENT III - I n t r o d u c t i o n 67 Method 70 R e s u l t s and D i s c u s s i o n 72 GENERAL DISCUSSION • • • 83 SUMMARY • • • 97 REFERENCES 99 APPENDIX I - TABLES OF MEANS 105 APPENDIX II - ANOVA TABLES H I i i i L IST OF APPENDIX I TABLES T a b l e I. R e c a l l S c o r e s f o r E x p e r i m e n t s I and II 106 T a b l e 2 , Mean V a l u e s f o r V i s u a l C o d i n g G r o u p s In E x p e r i m e n t III 107 T a b l e 3 . Mean R e c a l l f o r V e r b a l C o d i n g G r o u p s i n E x p e r i m e n t M l 108 T a b l e 4 . Mean P e r c e n t a g e s o f C o r r e c t R e c a l l f o r V i s u a l and V e r b a l S t i m u l i In E x p e r i m e n t III 109 T a b l e 5 . Mean P e r c e n t a g e o f C o r r e c t R e c a l l tn E x p e r i m e n t s I and II 110 iv LIST OF APPENDIX II TABLES Table I. ANOVA for Correct Reproductions In Experiment I 112 Table 2. Newman-Keuls Test for Treatment Effects for Correct Reproductions in Experiment I 113 Table 3. ANOVA for d' Scores in Experiment I 114 Table 4. Newman-Keuls Test for d* Scores in Experiment I 115 Table 5. ANOVA for Recall Scores in Experiment II 116 Table 6. Newman-Keuls Test for Treatment Effect in Experiment II 117 Table 7. ANOVA for Correct Reproductions of Visual Patterns in Experiment III 118 Table 8. ANOVA for d' Scores of Visual Pattern Reproductions In Experiment III 119 Table 9 . ANOVA for Verbal Recall Errors In Experiment III . . . 120 Table 10. Analysis of Simple Effects for AB Interaction of Verbal Recall Errors in Experiment i l l 121 V LIST OF FIGURES F i g u r e I . Mean S c o r e s f o r C o r r e c t R e p r o d u c t i o n s o f V i s u a l P a t t e r n s In E x p e r i m e n t I 48 F i g u r e 2. Mean d' S c o r e s f o r V i s u a l C o d i n g i n E x p e r i m e n t I 49 F i g u r e 3. Mean C o r r e c t R e c a l l S c o r e s i n E x p e r i m e n t II 58 F i g u r e 4. Mean V a l u e s o f C o r r e c t R e p r o d u c t i o n s o f V i s u a l P a t t e r n s i n E x p e r i m e n t I I I ., 74 F i g u r e 5. Mean d' S c o r e s f o r V i s u a l C o d i n g In E x p e r i m e n t Ml •••• 75 F i g u r e 6. Mean V a l u e s o f V e r b a l R e c a l l S c o r e s i n E x p e r i m e n t I I I 76 F i g u r e 7. P e r c e n t a g e V a l u e s f o r C o r r e c t R e c a l l i n E x p e r i m e n t s I and II 86 v i ACKNOWLEDGEMENT I would l i k e to express my a p p r e c i a t i o n f o r the h e l p f u l comments and sugges t ions of D r s . Ray C o r t e e n , Dennis F o t h , Ron Mar ten iuk , and John Y u i l l e in the w r i t i n g of t h i s paper . I am e s p e c i a l l y indebted t o Dr . John Y u i l l e f o r h i s cont inuous suppor t which made t h i s research p o s s i b l e . I GENERAL INTRODUCTION The e m p i r i c a l and t h e o r e t i c a l I s s u e s e v a l u a t e d In t h i s I n t r o d u c t i o n c e n t e r around two a s p e c t s o f t h e c u r r e n t s t a t e o f knowledge a b o u t human memory. In t h e I n i t i a l t o p i c , r e s e a r c h e v i d e n c e and t h e o r e t i c a l p o s i t i o n s a r e d i s c u s s e d w h i c h a r e d i r e c t l y c o n c e r n e d w i t h p roblems I n h e r e n t In t h e arguments u n d e r l y i n g t h e c o n c e p t u a l i z a t i o n o f memory a s e i t h e r a u n i t a r y o r a dichotomous system. W i t h i n t h e c o n t e x t o f t h i s d i s c u s s i o n t h e emphasis i s p l a c e d on r e s e a r c h I n v e s t i g a t i n g s h o r t - t e r m memory (STM) p r o c e s s e s . I s s u e s c o n c e r n e d w i t h s e n s o r y memory (SM) and l o n g - t e r m memory (LTM) a r e o n l y p e r i -p h e r a l l y I n t r o d u c e d t o p r o v i d e an o v e r a l l framework f o r a g e n e r a l model o f memory. W i t h r e s p e c t t o t h e arguments p e r t a i n i n g t o a u n i t a r y v e r s u s d i chotomous memory, I . e . , t h e need t o p o s t u l a t e q u a l i t a t i v e l y s i m i l a r o r q u a l i t a t i v e l y d i f f e r e n t p r o c e s s e s f o r STM and LTM, two arguments a r e f o r w a r d e d . F i r s t , r e s e a r c h d a t a o b t a i n e d i n s t a n d a r d STM and LTM paradigms a r e r e v i e w e d , and t h e c h a r a c t e r i s t i c f e a t u r e s o b s e r v e d as a f u n c t i o n o f t h e d i f f e r e n t p a radigms a r e u t i l i z e d i n s u p p o r t o f a dichotomous memory sy s t e m . Second, STM r e s e a r c h d e m o n s t r a t i n g o t h e r t h a n s h o r t - t e r m p r o c e s s i n g c h a r a c t e r -i s t i c s a r e e v a l u a t e d w i t h r e s p e c t t o t h e n a t u r e o f t h e STM paradigms u s e d , and t h e s p e c i f i c p a r a m e t r i c c o n s i d e r a t i o n s employed i n such e x p e r i m e n t s . I t I s argued i n l i g h t o f t h i s e v a l u a t i o n t h a t a s t r i c t a d h e r e n c e t o p a r a -m e t r i c c o n s i d e r a t i o n s o f t h e s t a n d a r d STM paradigm p r o v i d e s an i n t e r p r e t a t i o n o f STM d a t a c o n s i s t e n t l y i n s u p p o r t o f a dichotomous memory s y s t e m . Once I t i s t e n t a t i v e l y a c c e p t e d t h a t q u a l i t a t i v e l y d i f f e r e n t p r o c e s s e s a r e i n v o l v e d i n STM and LTM s i t u a t i o n s , t h e second i s s u e i s i n t r o d u c e d w h i c h p r o v i d e s t h e g e n e r a l framework f o r t h i s r e s e a r c h . Of m a j o r c o n c e r n In t h i s t o p i c I s t h e d e t a i l e d e v a l u a t i o n o f s h o r t - t e r m p r o c e s s e s . E v i d e n c e I s o u t l i n e d w h i c h s u p p o r t s t h e e x i s t e n c e o f q u a l i t a t i v e l y d i f f e r e n t c o d i n g 2 processes in STM exper iments , I . e . , ve rba l and v i s u a l c o d i n g . In the review of the l i t e r a t u r e i t f u r t h e r becomes apparent t h a t in fo rmat ion about ve rba l sho r t - t e rm processes i s more complete than in fo rma t ion about v i s u a l p r o c e s s e s . Desp i te the incomplete In fo rma t ion , however, t he re appears to be s u f f i c i e n t ev idence to p o s t u l a t e separa te verba l and v i s u a l sho r t - t e rm p r o c e s s e s . As a r e s u l t of t h i s rev iew the f o l l o w i n g t h e o r e t i c a l ques t i on i s posed as the b a s i s of t h i s r e s e a r c h : t o what ex ten t are verba l and v i s u a l sho r t - t e rm cod ing processes i n t e r - r e l a t e d o r independent from each o ther? The e m p i r i c a l approach proposed t o I nves t i ga te t h i s t h e o r e t i c a l problem makes use of a f a c t o r i a l combinat ion of t h ree major d imensions of the STM paradigm: a) ve rba l and v i s u a l s t imu lus cod ing s i t u a t i o n s ; b) I n te rpo la ted a c t i v i t i e s d i f f e r i n g w i th respec t t o the moda l i t y of the In te r fe rence t a s k , i . e . , verba l versus v i s u a l ; and c) In te rpo la ted a c t i v i t i e s d i f f e r i n g w i th respec t t o d i f f i c u l t y of t ask demands w i t h i n each moda l i t y o f the i n t e r p o l a t e d a c t i v i t y . The dimension of t a s k d i f f i c u l t y , i n t e r p r e t e d in terms of a t t e n t i o n demands, Is Introduced to eva lua te the e f f e c t s of the c e n t r a l p rocess ing component in verba l and v i s u a l c o d i n g . Verbal and v i s u a l I n te rpo la ted t a s k s are Introduced to f u r t h e r e x p l o r e the nature of the cod ing p rocesses f o r each s t imu lus cod ing s i t u a t i o n . Evidence f o r STM Cur ren t approaches t o human memory inco rpo ra te a r a t h e r w ide ly accepted framework of seve ra l b a s i c p rocesses and mechanisms which are cons ide red as necessary aspec ts of an adequate memory model . Though d i s t i n c t i v e d i f f e r e n c e s can be de tec ted among approaches of v a r i o u s t h e o r i s t s , there appears t o be a t l e a s t a genera l agreement t o p o s t u l a t e 3 separate memory systems for STM and LTM functions. Certainly, research u t i l i z ing the operational dist inctions underlying STM end LTM has generated consistent evidence supporting the concept of a dichotomous human memory. The need to distinguish between memory processes along a loosely defined temporal continuum had already been realized by William James (1890). James proposed a dichotomous memory system based on speculations about di f ferent ia l attentional demands underlying what he cal led primary and secondary memory processes. Within this system, primary memory processes were said to be involved when information was recalled which had not yet lef t consciousness, i . e . , which had been attended to continuously. Recall of information which had been absent from consciousness for some time, In contrast, relied on secondary memory processes. The nature of the recal l process provided a further dist inct ion between the two systems. The stored information in primary memory, for Instance, was readily available for r e c a l l , while recall from secondary memory required active search and retrieval processes. Common elements between the descriptive model of William James and current models appear unmistakable. Isolated instances supporting his ideas for a dichotomous memory appeared in the following decades in the work of Wundt and Ebblnghaus on memory span capacity ( c f^K ln tsch , 1970). However, i t was not unti l these descriptive notions could be investigated by the operational dist inct ions Introduced by Brown (1958) and Peterson and Peterson (1959) that the current trend in memory research fu l ly emerged. The current operational dist inct ions between STM and LTM include the following major considerations of basic experimental manipulations. Experi-ments investigating memory after s ingle , brief stimulus presentations, with 4 retention intervals of up to 15-30 seconds, are usually considered as displaying STM processes. In contrast, experiments using repeated t r i a l s , long stimulus presentations and retention Intervals ranging as long as hours, or weeks, are defined as LTM research. Research based on this operational dist inct ion has In turn generated results which suggest qual i tat ively different underlying mechanisms and processes in LTM and STM. Long-term memory characterist ics include, among others, an unlimited storage capacity, indefinite retention of stored information, and elaborate coding, storage, and retrieval mechanisms. In contrast, the basic characterist ics proposed for STM include a limited storage capacity, a coding mechanism which re l ies predominantly on verbal-acoustic features of the stimulus material, and an active rehearsal mechanism. Evidence for d ist inct STM characterist ics has been obtained in a variety of experimental manipulations. An outline of supportive evidence for these characterist ics presents a rather convincing picture for the verbal acoustic interpretation of STM. Research demonstrating factors not included in a verbal acoustic STM framework, such as nonverbal or semantic factors, is excluded from the discussion at this time, and will be added at a later point. Storage capacity, as already mentioned, has been investigated by Ebblnghaus and Wundt around the turn of the century and found to be around six items (Kintsch, 1970). Similar results were obtained by Mi l ler (1956)' who found the immediate memory span to be seven items, plus or minus two, whether employing d i g i t s , letters or words. Further evidence for a limited STM capacity can be extracted from a comparison of tr iad (Murdock, 1963) and 12 item (Glanzer and Cunitz, 1966) free r e c a l l . 5 STM characterist ics were obtained in the tr iad study but appeared only for the last few items in the free recall task. In other words, only the last few items had been recalled from STM, while the ear l ier items were either lost from STM or recalled from LTM. A s l ight ly different conceptua-l ization of storage capacity is suggested by Atkinson and Shi f f r ln (1968), who treat the limited storage capacity in terms of a rehearsal buffer, i . e . , the capacity of the rehearsal mechanism. The size or capacity of the rehearsal buffer in this model is in turn determined by input and reorganizational factors of a given experimental s i tuat ion. In other words, the evidence for a limited or fixed STM capacity appears to be rather consistent, and, furthermore, would have to be expected simply in light of the temporal defines of the STM paradigm. The second important feature of STM centers around the predominantly verbal-acoustic effects obtained In STM experiments. Probably the f i r s t indication of this characterist ic was reported by Sperling (I960, 1963), who postulated an "auditory information stage" in immediate memory as a result of observing a large number of auditory confusion errors with tachistoscopically presented st imul i . The Implication of this finding from a theoretical viewpoint appears rather important since i t suggests that coding in short-term recall re l ies strongly on auditory-acoustic features of the stimulus material. These findings are quite in contrast to coding effects in LTM, where semantic characterist ics are of more importance. A more detailed evaluation of verba I-acoustic effects in STM was subsequently derived from a basic paradigm which investigates types of error scores when acoustically similar or dissimilar items are used as stimulus material. The generally consistent findings in this 6 experimental situation can be summarized by the following conclusions: a) substitution errors tend to be acoustically confusing Items; and b) the error score Is higher with acoustically similar than dissimilar l i s t s . These detrimental effects of acoustic s imi lar i ty were consistently demonstrated under a variety of changes In experimental manipulations. For instance, acoustic interference effects appeared when using strings of consonants, whether the letters were auditorally presented and embedded in noise (Conrad, 1964) or presented visual ly (Wickelgren, 1965). Further-more, the acoustic similar i ty effect was also obtained with words, i . e . , homophones, in a recall (Kintsch and Buschke, 1969) or In a recognition task (Wickelgren, 1966). Thus, the consistency of these results strongly supports the proposition that coding in STM re l ies largely on verbal acoustic properties of the stimulus, in contrast to LTM. The third c r i t i c a l aspect of STM is the rehearsal mechanism, a concept closely tied to memory capacity on the one side and verbal-acoustic coding on the other side. Evidence for the functional significance of the rehearsal mechanism derives from comparing two experimental operations, i . e . , rehearsal versus interference ac t iv i t i es during the retention... interval In an STM task. Murdock (1963) found recall of tr iads and trigrams to decrease to a probability of .08 after 15 to 18 seconds of verbal interference. In contrast, if rehearsal is allowed during the retention interval , recall for let ters, words and sentences has been shown to increase over a ten second delay interval (Crawford, Hunt, and Grahame, 1966). The fac l l i t a t i ve effects of rehearsal, in contrast to verbal Interference, have been further substantiated with verbal material in recall and recognition (Ternes and Y u i l l e , 1972). Furthermore, by 7 combining the two operations within a single experimental sett ing, recall can be effect ively manipulated by introducing repetition (HeI Iyer, 1962) or rehearsal (Stonner and Muenziger, 1969) before the onset of the inter-ference task during the retention interval . The current dist inct ion between STM and LTM i s , however, not only based on behavioral data. Neurological evidence has also contributed a convincing argument for a dichotomous memory. In fact , theorists such as Atkinson and Shi f f r in (1968) consider the effects of hippocampal lesions as the most convincing demonstration of a dichotomy in the memory system. Milner (1968), for instance, found that patients with hippo-campal lesions appeared to display normal immediate memory functions as tested by d ig i t span and dichotic listening tasks. At the same time, there appeared to be no loss of preoperative Iy acquired s k i l l s . However, the patients appeared incapable of adding new information to the long term store. Buschke (1968) confirmed these findings using a missing scan procedure on patients with bilateral hippocampal lesions. He concluded from his results that the patients suffered from a Iearningdeficlt , but displayed normal immediate memory functions. Also in support of these results is a later study of Baddeleyand Warrington (1970) with amnesic patients. In e f fect , the implications of the neurological evidence is quite consistent with a dichotomous memory model. Since STM functions as well as preoperative LTM information appear to be normal, while no new information can be added to LTM, the notion of some type of memory compartmentalization appears quite feasible. In summary, data from two widely divergent research approaches, such as behavioral and neurological research, present consistent evidence for 8 the concept of a dichotomous memory. However, this in i t i a l Interpretation has to be qualif ied by additional research data contradicting this con-clusion. For instance, there has also been data questioning the consistency of the above reported neurological findings (Robbins and Meyer, 1970). Since neurological evidence is based on only isolated research examples, however, i t appears premature at this time to consider the unitary versus dichotomous controversy fu l ly resolved on the basis of this line of research. In addit ion, the dichotomous interpretation has also come under attack by researchers stressing the effects of semantic factors in behavioral research of STM functions (e .g . , Wickens, 1970). The argument in this context is the following: if semantic factors, i . e . , LTM character ist ics, play an important role in STM functions, there is no need to postulate separate memory systems. Behavioral data of this nature point out a crucial issue underlying the interpretation of general STM research within the context of the unitary versus dichotomous memory controversy. As stated at the beginning of this introduction, there are two interdependent factors underlying the general concepts of STM and LTM. On the one hand, there are operational definit ions describing the speci f ic experimental situations under which memory is tested, i . e . , STM versus LTM paradigms. On the other hand, there are qual i tat ively dist inct characterist ics ascribed to memory mechanisms and processes which are operative as a result of the speci f ic experimental paradigms, i . e . , STM versus LTM character ist ics. In light of the number of variables involved, and the rather vaguely defined boundaries of parametric values of these variables in the two operationally d is t inct paradigms, the interpretation of a given experiment In terms of 9 the usually obtained characterist ic processes of the paradigm is not always unequivocal. For example, in a serial recall task, i . e . , a standard LTM paradigm, the recency effect is interpreted in terms of STM processes. In other words, i t is possible to obtain STM character ist ics, in addition to LTM character ist ics, in a standard LTM paradigm. A similar argument can be made for results obtained in STM experiments. Given, for instance, an STM experiment which ref lects data contrary to the standard STM character ist ics, such as semantic coding ef fects , two types of interpretations are possible: a) STM contains other than standard STM character ist ics; or b) the given experiment tested other than STM processes. While both interpretations appear to be equally j u s t i f i e d , there is a certain c i rcu lar i ty in the interpretation of STM data wtth respect to the unitary versus dichotomous controversy. With a predisposition towards a unitary model, a researcher may argue that evidence for LTM characterist ics obtained in STM experiments eliminate the need to postulate separate memory systems. Conversely, with a pre-disposition towards a dichotomous system, a researcher u i a y argue that LTM characterist ics obtained in STM experiments merely provide examples showing that LTM processes may occur under certa'n experimental conditions within an STM paradigm. In a subsequent part of the introduction this writer will argue that the c i rcu lar i ty of both positions may be resolved by a s t r i c t adherence to parametric considerations of the STM paradigm. Theoretical Conceptualizations of STM Based on the above types of experimental evidence several memory models have been proposed. Examples of these models will be outlined next. While some of these models contain descriptions of other aspects 10 of memory, i . e . , sensory memory and LTM, only STM functions of the models wil l be considered in this context. Sensory memory and LTM functions wil l be mentioned only If relevant to the discussion of particular STM processes. One of the ear l ier memory models was developed by Waugh and Norman (1965, 1968), who proposed a simple quantitative method to separate long-term and short-term functions. Br ie f ly , their aim was to obtain a s ta t is t ica l estimate of memory performance in a probe-procedure and then apply th is estimate to the data of a series of published free-recalI and paired associate studies. The success in predicting STM functions in these two standard long-term memory paradigms led them to postulate two dist inct memory processes, primary and secondary memory, in the line of William James (1890). In this model, primary memory was hypothesized to contain a limited storage capacity in which retention was attributed to overt or covert rehearsal. Rehearsal was explained in terms of attention and was attributed a dual role in primary processing: a) re-entering items into primary memory; and b) transferring items into secondary memory. Loss of information from primary memory was described in terms of input and output interference, such that input Interference was determined by the number of Items presented during the retention Interval and output interference by the number of items recalled before the c r i t i c a l test item. In ef fect , a basic framework for STM was established displaying the limited storage capacity, as well as the rehearsal and interference functions which applied to processes observed in an STM task. At the same time, the model succeeded in separating STM and LTM processes in standard LTM paradigms. 11 Neisser (1967) proposed separate visual and auditory temporary memory systems, stressing the functional significance of these systems as transitory types of memory processes lying between the perception of an item on one side and the active memory on the other. Within the temporal defines of the standard STM paradigm, i . e . , 15 to 30 seconds, Neisser distinguished between echoic memory and the active verbal memory. Echoic memory in this system can be considered as analogous to sensory memory. The active verbal memory, in contrast, appears to describe what is usually called STM. In other words, with respect to the concept of primary memory, as proposed by Waugh and Norman, Neisser subdivides the concept into two parts. The f i r s t part is explained in terms of attention and the second part described in terms of the three STM character ist ics, i . e . , predominantly acoustic coding, the importance of the rehearsal mechanisms, and limited memory capacity. In e f fect , Neisser incorporates STM characterist ics into a d is t inct system within a broader model of human memory. While these two models show a certain amount of agreement on the general conceptualization of STM functions, several models opposing this general framework have also emerged. Based on a different line of research methodology, Laughery (1972) proposed a computer simulation model of STM. By rejecting the notion of item displacement in STM, and making use of decay as well as rehearsal functions, Laughery proposes an STM model of unlimited storage capacity. However, the model js restricted to one speci f ic task and a total of only 36 input Items. In light of these rest r ic t ions, the appl icabi l i ty of this model for human memory at this time suffers from crucial shortcomings, though optimistic predictions 12 with an extension of tasks and Input Items are offered. A complete rejection of the dichotomous memory hypothesis has recently been expressed by Murdock (1972). While he admits that short-term effects must be explained by any memory model, he proposed that this could be accomplished using a f in i te -s tate decision model. However, Murdock's approach appears to be more applicable In evaluating STM functions in free-recal l and pal red-associate learning, than in the investigation of processes observed in standard STM research. Furthermore, his main objec-tion "why such modal models ( i . e . , two store models) must be wrong is that in a very general sense they are incompatible with what we know about grammatical utterances" ( p . 9 1 ) . While It will certainly be necessary to interpret STM processes within a broader f ie ld of cognitive processes, an attempt already made by Neisser (1967), the lack of knowledge about both problems make an integration at this stage rather premature. Murdock's evaluation then appears to be more of an exposition of problems Inherent in a dichotomous memory, than a convincing argument for a unitary memory system. Probably the most f ru i t fu l current model using the concept of separate memory systems has been developed by Atkinson and Shi f f r in (1968). Sub-divided into the sensory memory (SM), short-term store (STS), and long-term store (LTS), the model presents a comprehensive account of memory research within a single theoretical framework. For instance, an attempt is made to distinguish between structural features and control processes within each subsystem. The basic invariant memory mechanisms, such as the rehearsal buffer, are defined as structural features, and situation speci f ic task requirements, such as Instructional set , are considered as 13 control processes. Though the potential usefulness of these concepts Is undeniable, the lack and d i f f i cu l ty of experimental validation make these additions to memory models s t i l l highly tentative. Furthermore, a clear dist inct ion Is made in this model between STM and STS, a dist inct ion based on the operational versus the theoretical aspects in memory experi-ments. Hence, the term STM applies to the operational definit ion of the experimental s i tuat ion, and STS pertains to mechanisms and processes which may be displayed In an STM experiment. In light of this d is t inc t ion , i t becomes possible to treat the results of an STM task in terms of STS and LTS functions, depending on the characterist ic features obtained in the given experiment. Within this general framework, Atkinson and Shi f f r in incorporated the previously mentioned STM character ist ics, i . e . , limited capacity, verbal-acoustic coding, and a rehearsal mechanism. STS processes within this system are described In the following way. When a stimulus Is presented, auditorally or v isual ly , it is registered in the sensory memory and immediately coded in a verbal-acoustic form and maintained in the rehearsal buffer. From there i t is transferred to LTS, where organizational and semantic features of the stimulus are stored. The rehearsal buffer, therefore, has the dual role of maintaining items in STS, as well as some undetermined transfer function in moving information to LTS. Items in the rehearsal buffer can be lost by displacement if the information exceeds the buffer capacity, and can also be lost if attention is diverted from the rehearsal process by an interpolated ac t iv i ty . Thus, the ava i lab i l i ty of a given amount of Information in STS Is fac i l i ta ted by the rehearsal process and reduced: a) by an information overload; and b) by an interruption of the rehearsal process by attention diverting 14 ac t i v i t i es . The diversity of these memory models creates an interesting question as to the just i f ica t ion of the different theoretical positions taken by various researchers. In other words, how is i t possible that from the broad f ie ld of STM research different theoretical conceptualizations of short-term processes emerge. Probably an obvious reason l ies in the particular experimental paradigm which is preferred as the research tool for investigations of these processes. For instance, In free recall and PA learning (Murdock, 1972), different aspects of short-term processing might be investigated than by the probe or distractor technique (Atkinson and S h l f f r i n , 1968). However, if one considers the latter as a more refined research tool to investigate short-term processes, contra-dictory results can also be obtained which, in turn, just i fy different theoretical posit ions. In the following discussion, a detailed evaluation of the standard STM paradigm (Peterson and Peterson, 1959) will be under-taken in order to expose several sources within the experimental situation which underlie contradictory results and subsequently iead to the diversity of theoretical positions. Variations on a Theme by Peterson and Peterson In the standard Peterson and Peterson paradigm (1959) the following sequence of events occurs. A brief stimulus is presented, followed by a retention interval during which S_may be required to perform an additional task. F ina l ly , a memory test of the stimulus is made. Within this experimental setting there are three major factors contribu-ting to the results and interpretation of a given experiment. F i r s t , the stimulus presentation contains a series of variables which affect 15 response probabil i ty: a) the characterist ics of the stimulus material; b) the modality of presentation; c) the duration of presentation; and d) the information load. The second major factor l ies In the retention interval . The basic effects in this category are due to: a) duration of the retention interval; and b) the type of interpolated act iv i ty to be performed by S_. The third source of response var iab i l i ty l ies in the response measure. Different response patterns can be obtained as a function of the response demands, e . g . , recall versus recognition, and the temporal constrains of the response task. Examples of each of these factors, and their interactions, will be considered next in an attempt to evaluate their relationship to theoretical interpretations of STM experiments. It should be noted that basic s imi lar i t ies can be identified between operational variables in the Peterson and Peterson task and other paradigms Investigating STM processes. For example, response trends generated by the manipulation of the stimulus duration variable in a Peterson and Peterson task are not unlike the effects obtained due to changes In the presentation rate in an STM serial recall experiment. Therefore, con-clusions can in many instances be generalized to problems inherent in other paradigms used to investigate STM functions. I The Stimulus Condition a) Stimulus Material Some of the examples of the effects of the stimulus material dimension have already been mentioned. Differential error scores with acoustically similar and dissimilar material have been influential in 16 considering STS as a predominantly verbal acoustic memory system (Atkinson and S h i f f r i n , 1968). However, when words are contrasted with pictures in an STM task (Ternes and Yu i l l e , 1972) evidence is obtained which cannot be Interpreted within a verbal STS framework. In fact , a considerable amount of data has accumulated supporting the existence of a v isua l , i . e . , non-verbal, STS. This evidence has been obtained under a variety of experimental tasks, such as Tversky's (1968) demonstration of visual and verbal coding preferences after a one second retention interval and a demonstration of visual coding effects by Parks, Parkinson, and KrolI (1971) after a 25 second retention interval . Research of this nature does not detract from the general assumption that a verbal acoustic memory system ex is ts . Rather, i t suggests that another type of coding can be demonstrated within the confines of an STM framework. Details about the necessary additions are, however, s t i l l controversial . While a visual STS currently receives a great amount of attention, evidence for kinesthetic STM (Posner, 1967) also has to be considered. b) Presentation Modality The method of presentation of the stimulus material, i . e . , auditory or v isua l , has been shown to result in similar or different response patterns, depending on the experimental situation under study. Error scores for acoustically similar and dissimilar consonants display a consistent trend whether obtained under auditory presentation (Conrad, 1964) or visual presentation (Wickelgren, 1965). Murdock (1972) in con-t rast , reports a consistent auditory over visual superiority in STM when employing the free recall paradigm. Furthermore, response differences can be obtained with visual presentations if the visual presentation 17 consists of either verbal or pictor ial material of familar objects (Ternes and Yu i l l e , 1972) or face drawings (Tversky, 1969). Thus, the inter-action between presentation modality and other task variables again generates contradictory resul ts , providing evidence for verbal as well as non-verbal coding in STM experiments. c) Stimulus Duration Apart from peripheral problems pertaining to cei l ing or f loor effects as determined by the duration of a stimulus presentation, the major theoretical problem inherent in this discussion pertains to the type of coding processes which can occur under given stimulus durations. Paivio and Czapo (1969), for instance, have demonstrated that the presentation time for a word or picture has to be reduced to around 200 m seconds in order to prevent cross-modality coding, that i s , in order to reduce verbal coding of pictures and visual coding of words. In addit ion, qual i tat ively different types of verbal coding processes can be identified with verbal material due to changes in the stimulus duration dimension. The focal point of the argument here pertains to the question of whether acoustic as well as semantic types of coding occur in STS. Baddeley (1964) has clearly demonstrated qual i tat ively different coding processes within the s t r i c t confines of the operational manipula-tions of the STM and LTM paradigm. He concludes that verbal coding in STM re l ies predominantly on acoustic features of the stimulus, whereas coding in LTM re l ies predominantly on semantic features of the stimulus information. However, semantic coding in STM experiments has also been reported (e .g . , Dale and Gregory, 1966; Wickens, 1971; Wickens and Eckler, 1968). Evidence of this nature shows that there can be l i t t l e argument 18 that semantic effects can be obtained in an STM experiment. However, within the s t r i c t confines of the STM paradigm, semantic effects can quite readily be attributed to the parametric considerations of a given experi-mental s i tuat ion. In a comprehensive review of this problem, Shulman (1971) concludes that semantic coding can be demonstrated in STM tasks, i f S_ is given suff ic ient time to make use of other than acoustic coding processes. For example, if the stimulus duration in a verbal coding STM experiment is changed from a minimum to a maximum within the vague defines of the STM paradigm, the parametric changes can lead to results supporting acoustic as well as semantic coding. While the basic paradigm can be considered in terms of STM, the theoretical evaluation of processes underlying the results can be interpreted in terms of STS and LTS processes. That i s , with a minimum stimulus duration, the predominant coding mode should be acoustic, and the effects can be interpreted in terms of STS functions. With a maximum stimulus duration, the possible influence of LTS processes cannot be ignored. If semantic coding effects do occur under this condition, the actual involvement of LTS processes must be acknowledged. However, the c i rcu lar i ty of this argument cannot be denied. After a l l , a researcher with a predisposition towards a unitary memory model can equally argue that there is no need to postulate a dichotomous system since similar coding characterist ics can be obtained in LTM as well as STM experiments. The latter posit ion, though, ignores one important issue: parametric changes in the experimental paradigm lead to quantitative as well as qualitative changes in the experimental results. Semantic coding effects in STM, therefore, can be largely attributed to LTS involvement as a function of parametric changes in the 19 STM paradigm. In summary, changes in stimulus duration can lead to cross-modality coding in a comparison of verbal and visual material, thereby influencing the argument of verbal versus visual STS. Furthermore, stimulus duration changes with verbal material can lead to quali tat ively different coding processes, i . e . , acoustic versus semantic. Unless parametric values of the STM paradigm are careful ly considered, the clear dist inct ion between coding processes In STS and LTS with respect to acoustic and semantic coding is diminished. However, if i t is admitted that parametric changes are instrumental in generating not only quantitative but also qualitative differences, a conceptualization of STS and LTS in terms of acoustic and semantic coding is supported. d) Memory Load Changes in memory load are ultimately restricted by the immediate memory capacity (Mi l ler , 1956). It is therefore not surprising that 12 item free recall will generate different response patterns over retention intervals than triad recall (Glanzer, 1972). Even within the defines of the immediate memory capacity, di f ferent ial response patterns are observed between single item, t r i ad , and trigram recall (Murdock, 1963). In addit ion, with a constant information load di f ferent ia l response patterns can be obtained as a function of the type of stimulus presenta-tion and the presentation times per Item. Paivio and Czapo (1969) demonstrated this effect by presenting either words or pictures of familar objects at two different presentation rates. Hence, given a constant information load, di f ferent ia l results can be obtained by changing 20 other relevant dimensions of the experimental task. In other words, if the information load across experiments is not comparable, di f ferent ia! response patterns can be expected in STM research. As long as interactions of various crucial variables with memory load are not c l a r i f i e d , contradictory conceptualization of STS characterist ics can be expected. I I Retention Interval a) Duration The length of the retention interval may affect the store S_ uses to retrieve the response. For instance, if a relat ively "long" rehearsal interval is employed, It becomes impossible to attribute the response level to a speci f ic retention mechanism, i . e . , STS or LTS. A description of the postulated processes involved i l lustrates this point. F i r s t , i t is assumed that the information enters the verbal rehearsal buffer, where i t is temporarily stored and available for re t r ieva l . Second, there is an unknown transfer function between the buffer and LTS, such that after a "long" rehearsal interval the stimulus information may be either s t i l l in the buffer or already in LTS. With respect to the previously mentioned issue regarding acoustic versus semantic coding in STM, i t should not be d i f f i c u l t to obtain evidence for semantic coding if the information has been retrieved from LTS (Shulman, 1971). In ef fect , the interpretation of responses obtained after a relat ively long retention interval cannot be confidently interpreted in terms of STS processes alone. Consequently, other than STS characterist ics might be attributed to STM, unless parametric considerations are careful ly evaluated. A similar problem exists on the other end of the duration continuum 21 with regard to very short retention intervals. Ternes and Yui l le (1972, 1973) have found that in a six item recognition sequence the f i r s t recogni-tion item, if presented immediately after the stimulus, decreases the recognition level of the subsequent items in the sequence. This observa-tion is not unlike the masking effect obtained in sensory memory experi-ments. Thus, by changing the duration of the retention interval in an STM experiment, the results can be effect ively influenced at both ends of the continuum by factors inherent in sensory memory or LTS. b) Type of Interpolated Act ivi ty Since this particular problem will be discussed in more detail In the next section, only a general statement concerning the diversity of results due to interpolated act iv i ty effects will be presented at th is point. Rehearsal act iv i ty during the retention interval has been shown to increase, decrease, or result in no change with respect to an immediate memory test , depending on the type of material and type of performance measure of the particular experimental situation (Milner, 1968; Posner, 1967; Ternes and Yu i l l e , 1972). S imi lar 'y , interpolated act iv i ty effects defined as interference conditions interact with material and response task to produce either increases, decreases, or no changes in the response measure (Glanzer, 1972; Kintsch, 1970). In addition, inter-polated task effects are influenced by the similar i ty of the interpolated act ivi ty to the stimulus material and by the d i f f i cu l ty of the interpolated act iv i ty . Differential results of this nature therefore have to be evaluated in light of the speci f ic experimental manipulations introduced in a given experiment, in order to obtain some type of consistent pattern of interpolated act iv i ty ef fects , 22 I 11 Response Measures In addition to quantitative response changes due to temporal con-straints on the response task, qual i tat ively different responses are said to underlie recall and recognition performance. In r e c a l l , storage and retrieval processes are involved, and recognition is considered as a purer estimate of storage (Kintsch, 1970; Murdock, 1972). Again, examples of contradictory response patterns due to response measures can be found by a comparison of Murdock (1972) and previously reported results investigating acoustic similar i ty effects (Baddeley, 1964; Wickelgren, 1966). Murdock reports a higher response level in recognition experiments, while Baddeley as well as Wickelgren find an equally strong acoustic similar i ty effect in both recall and recognition experiments. Further-more, in a series of experiments contrasting these two response measures, Ternes and Yui l le (1972) found evidence for di f ferent ia l coding under recognition and recall conditions. Words as well as pictures were shown to be coded verbally in free r e c a l l , whereas pictures were coded visually in the recognition task. In e f fect , qual i tat ively different coding strategies can be obtained depending on task demands. Changes in task demands therefore may contribute some information to the issue of verbal versus non-verbal characterist ics in STS. It should be noted though that response differences in this context are not concerned with immediate response levels, but with response trends over retention i ntervaIs. To summarize the impact of these examples with respect to theoretical arguments concerning the notion of STM, several tentative statements can 23 be made regarding two dist inct questions: a) does the evidence favour a dichotomous or a unitary memory modei; and b) to what extent can non-verbal processes be identified in an STM experiment? With respect to the f i r s t issue, i . e . , unitary versus dichotomous memory systems, i t has been shown that changes in a series of variables of the STM paradigm may lead to contradictory conclusions about this issue. Apart from quantitative differences in the information load between STS and LTS, as well as the postulated existence of a rehearsal mechanism, the crucial factor dif ferentiat ing the two systems is the qual i tat ively different coding system attributed to each. Unless STS and LTS are clearly differentiated along the acoustic versus semantic coding dimension, the need to postulate qual i tat ively different memory systems is greatly reduced. STM research pertaining to the dichotomous versus unitary question generates two possible viewpoints. The particular preference of one over the other in this case seems to be largely determined by the researcher's emphasis on a s t r i c t adherence to parametric considerations of the STM-LTM paradigms. After a l l , there are STM experiments reporting semantic coding ef fects. However, if these STM experiments are evaluated in light of STM parametric considerations, i t becomes evident that changes, such as in stimulus duration and retention interval , constitute experimental sources leading to LTS involvement in STM experiments. In other words, i t cannot be denied that semantic coding can occur in STM, however, these effects can be attributed to LTS functions. Simi lar ly , acoustic coding can be demonstrated in LTM experiments, which would detract l i t t l e from 24 the importance of the predominantly semantic effects in LTS. In light of the contradictory results which can be obtained due to changes in the basic paradigms, this writer prefers the conceptualization of STS and LTS in terms of qual i tat ively d ist inct coding character ist ics, thereby adhering to the idea of a dichotomous memory system. If the assumptions underlying a dichotomous memory system can be accepted, the speci f ic characterist ics underlying STS can be evaluated. The conceptualization of a verbal STS has been proposed in detail by Atkinson and Shi f f r in (1968). Even in th is model the addition of a visual STS has been acknowledged, depending on c la r i f i ca t ion by future research. Since then, STS research evaluating crucial variables, such as presentation mode, stimulus material, and response demands, provides strong evidence for the existence of visual processes in STS. A c l a r i f i -cation of speci f ic characterist ics of the visual system in STS is s t i l l outstandi ng. In light of this discussion of experimental variables in the STM task, two definite conclusions can be drawn about the current state of knowledge regarding memory models. F i r s t , contradictory results can be obtained within the general STM research area, which can be attributed to parametric changes and interactions of c r i t i ca l dimensions of the basic STM paradigm. As a result of these contradictory f indings, it can be expected that opposing theoretical positions will emerge. Second, i t is apparent that suff icient evidence is available which points out the incompleteness of each STM model, especially with respect to the evidence for non-verbal processes. 25 The second conclusion Is of importance in the context of this paper. If a visual STS can be postulated on the basis of current information, it is of importance to establ ish: a) to what extent this visual STS is independent from verbal STS; b) what i ts characteristic features are; and c) what, i f any, common elements can be identified between the visual and verbal STS. A detailed evaluation of research data with respect to these theoretical questions will follow In order to create the framework of the problem to be investigated in this research project. The Problem As noted above, the possible existence of other than verbal coding processes in STS was acknowledged by Atkinson and Shi f f r in (1968), but was at that time only considered as a possible extension of the model to be c la r i f i ed by further research. Since that time, convincing examples of visual coding processes in STM experiments have been demonstrated in a variety of experimental situations. Information from this type of research can be divided into two basic aspects: a) visual coding for immediate response conditions; and b) investigations of visual coding over short-term retention intervals. The emphasis in the latter case centers on the rehearsal functions of visual coding, and effects of various interpolated ac t iv i t ies on visual coding. Visual coding for immediate response conditions has been demonstrated by researchers such as Posner (1969) and Tversky (1969). For instance, in a letter matching task, reaction time was shown to be faster when judgements are based on physical rather than naming properties of the comparison st imul i , if the two letters were separated by an interval of 26 less than one second (Posner, Boles, Eichelman, and Taylor, 1969). In the same experiment, however, i t was also shown that the superiority of a physical match disappears when the interstimuI us interval is expanded to two or more seconds. In a similar experimental sett ing, Tversky (1969) investigated verbal and visual coding processes by using schematic faces and their well learned nonsense names as comparison st imul i . In each block of t r i a l s , Ss were presented with either word or picture st imul i , followed after one second by the comparison stimuli of either presentation mode. The presentation mode of the comparison st imul i , i . e . , words or pictures, within each block of t r i a l s wss always presented in a ratio of eight to one, such that in a given block of t r i a l s , S_could always build up expectations regarding the most frequently occurring presentation mode. The results of this experimental setting showed that reaction times for same-different judgements, regardless of the presentation mode of the f i r s t stimulus within each block, were always faster for the most frequently occurring presentation mode of the second stimulus. Thus, verbal material was shown to be coded verbally as well as v isua l ly , and pictor ia l material was shown to be coded visual ly as well as verbal ly, depending on S_'s expectations of the most frequent mode. Experiments of this nature present consistent data which cannot be interpreted in terms of a verbal STS. In addit ion, recent evidence from neurological Iy oriented research supports the idea of d is t inct verbal and visual aspects in short-term processes. The theoretical framework for this type of research is based on the general assumption of hemis-pheric asymmetry in the context described by theorists such as Lenneberg 27 (1967). That Is, language processes predominantly involve the left hemisphere, whereas non-verbal or spatial processing is evident in the right hemisphere. Assuming left hemispheric dominance of speech and language functions, and assuming right hemispheric dominance of non-verbal processes, i t can be expected that verbal and visual stimulus material presented in a matching task to the right or left visual f i e ld should result in reaction time patterns showing a relationship between the presentation mode of the stimulus and the respective hemispheric dominance. Geffen, Bradshaw,and Wallace (1971) confirmed these expectations. Verbal stimuli resulted in faster reaction times when processed in the left language hemisphere, and conversely, non-verbal s t imul i , such as faces, were processed faster in the right spatial hemisphere. Additional support for hemispheric asymmetry has been obtained in a memory scanning task (Klatzky, 1970; Klatzky and Atkinson, 1971). Both experiments agree on the general findings that test stimuli requiring spatial matching are processed faster by the right hemisphere and test stimuli using verbal processing are processed faster by the left hemisphere. A further invest i -gation of this problem (Geffen, Bradshaw, and Nettleton, 1972) points out that reaction time differences must be Interpreted in terms of hemis-pheric asymmetry as well as interhemispheric transfer of information. It should also be pointed out that in the experiments reported, the evidence for left hemispheric dominance is always more pronounced. In-complete as this line of research may be, i t contributes a strong argument for verbal and visual processing in STM functions. In light of these research examples, the existence of d ist inct visual 28 and verbal coding processes for immediate response conditions in STM experiments is consistently supported. If verbal and visual short-term processes are postulated in immediate response conditions, i t becomes of further interest to evaluate the functional characterist ics of these coding processes within the total temporal range of the STM paradigm. In order to determine characterist ic features of the verbal STS, the effects of rehearsal and interference conditions on verbally coded material were evaluated over short-term retention intervals. Conversely, character-i s t i c features of the visual STS should be obtained in an evaluation of similar interpolated ac t iv i t i es on visual ly coded material. The basic aim of this evaluation is to determine what types of common properties as well as d ist inct differences can be identified in the verbal and visual coding system. With respect to rehearsal effects in visual STS, two contradictory research findings are to be reported at this time. On the one side, there is evidence that unf i l led retention intervals lead to an improvement in recognition of faces (Milner, 1968) and accuracy of a visual location response (Posner, 1967). Furthermore, Cohen and Granstrom (1970) show recognition of pictor ia l material to be higher after retention intervals f i l l e d with rehearsal act iv i ty compared with visual interpolated ac t iv i ty . In contrast to the fac i l i t a t i ve effects of rehearsal on visual ly coded material, Ternes and Yui l le (1972) did not obtain performance differences of visually coded material after rehearsal or verbal interference. The lack of an e f f ic ient rehearsal mechanism in visual STM has also been demonstrated by Potter and Williams (1970). 29 The basic question posed in these experiments centers on whether visual information can be maintained over STM intervals without any retention losses, or even improvements in retention as has been shown in verbal STS. The assessment of this question i s , however, not easy. There are two possible evaluation methods available in this case. One, the rehearsal effect can be assessed with respect to the immediate recall or recognition level . Two, rehearsal effects can be assessed with respect to interference conditions using the same retention interval . In the f i r s t case, the assessment of the rehearsal effect can be distorted, since responding at zero delay has been shown to be a possible interference con-dit ion in i tse l f (Ternes and Y u i l l e , 1972). Therefore, the response level obtained at zero delay does not always constitute an accurate reference level in the evaluation of the rehearsal ef fect . In the second type of assessment, the problem of cross-modality encoding appears as a crucial factor. As already mentioned in the context of verbal STS, cross-modality coding can be reduced by a reduction in stimulus presentation time (Paivio and Czapo, 1969). In other words, unless cross modality coding is reduced, i t becomes necessary to distinguish between recognition of visual material and recognition reflecting predomi-nantly visual coding. That i s , if visual material is presented for a long duration, correct recognition of the information might be due to a combination of visual and verbal coding. With a long duration for stimulus presentation, a f ac i l i t a t i ve effect would have to be expected simply on the basis of the verbal rehearsal ef fect . With respect to this operational res t r ic t ion , only the Potter and Williams (1970) as well as 30 the Ternes and Yui l le (1972) experiments qualify as investigating visual coding ef fects, it is therefore not surprising to note the different conclusions drawn about the visual rehearsal mechanism by the latter two experiments in contrast toMi lner (1968) and Posner (1967). Though within this context the arguments favour a rejection of the rehearsal mechanisms in visual STS, the evidence appears by no means conclusive at this stage. Since the effects of cross-modality coding have been shown to under-ly contradictory conclusions about the existence of a visual rehearsal mechanism, similar problems can be expected in assessing interference effects when v isua l , pictor ial material is used. Hence, unless i t can be established that cross-modality coding of visual material is reduced, detrimental effects due to interference conditions have to be expected, and can be interpreted in terms of the famil iar loss in verbal STS. Interference effects in the visual STS have been investigated within a theoretical framework highly compatible with the neurological data pertaining to hemispheric asymmetry. That i s , if a given verbal stimulus is processed in the le f t , language hemisphere, storage of this stimulus is more affected when the interpolated act iv i ty requires verbal processing, i . e . , processing by the left hemisphere, than if the interpolated act iv i ty requires processing by the r ight, visual-spatial hemisphere. A similar prediction about stimulus processing can be made about the right hemis-phere. In other words, the notion of modality speci f ic interference effects is in agreement with the present knowledge about hemispheric asymmetry. Though research investigating this particular aspect of modality speci f ic interference has not been reported, general STM research 31 investigating the problem supports this proposition. It should be noted though, that differences in modality speci f ic interference should be relat ively small. Gel I en et al_. (1972) have already demonstrated the existence of interhemispheric transfer, even if the stimulus presentation is restricted to the left or right visual f i e l d . Subsequently, an even larger amount of transfer of information between visual and verbal pro-cesses can be expected if the stimulus is presented to both visual f i e l d s , i . e . , to the language and spatial hemispheres simultaneously. Research Investigating the problem of modality speci f ic interference effects within the behavioral research framework generates rather con-sistent support for th is proposition. Margrain (1967), for instance, reports a stronger interference effect for recall of v isual ly presented words follow-ing written than verbal interpolated ac t iv i ty . Furthermore, in an auditory shadowing task recal l suffered more for auditor ia l ly than for v isual ly presented letters, but only after a 25 second retention interval (Kro l l , Potter, Parkinson, Bieber,and Johnson, 1970). In an expansion of this paradigm (Parks, Parkinson,and K r o l l , 1971), recall performance for visual ly presented letters was shown to be superior to auditor ia l ly presented letters after auditory shadowing. Posner (1969), in summarizing some of his own research, establishes a type of hierarchy of interference factors of the visual code. The • introduction of a single auditory digi t as interpolated act iv i ty may, or may not, have an ef fect , depending on the amount of practice S_ has with the experimental s i tuat ion. Adding visual d igi ts as an interpolated act iv i ty has a signif icant detrimental effect on the visual code. The concept of hierarchial interference conditions has also been 32 demonstrated with nonverbal material (Cohen and Granstrom, 1970), in th is case with simple stimulus f igures. Apart from showing a larger recognition loss of p ictor ia l material due to visual rather than verbal interpolated ac t iv i ty , Cohen and Granstrom obtained an increase in recognition losses with a corresponding increase in the task demands of the visual inter-polated ac t iv i ty . Further support for a selective loss of information by means of visual and verbal interpolated act iv i ty has been reported by den Heyer and Barrett (1971) in an experiment where information regard-ing the identity or position of letters in a matrix had to be recal led. In summary, i t appears that despite the possib i l i ty of different degrees of cross-modality encoding, which might be underlying the above-mentioned visual coding examples, there appears to be consistent evidence for some type of speci f ic as well as general interference ef fects . An example of speci f ic interference can be observed in the visual interference effect on visual coding. Auditory interference effects on visual coding, in contrast, appear to be an example of a more general type of inter-ference. In the theoretical interpretation of interference effects in STS, two dimensions wil l be considered in this context as independent factors contributing to the losses in STS. One of these factors has been variously expressed in terms of similar i ty (Bernbach, 1969; Glanzer, 1972; Laughery, 1969), the other in terms of task d i f f i c u l t y , or generally, in terms of attentional demands of the interpolated act iv i ty ( c f . Atkinson and Sh i f f r in , 1968; Glanzer, 1972; Kintsch, 1970). Both of these dimensions appear to be instrumental in retention losses in verbal and visual STS. 33 The concept of s imi lar i ty as an interference factor has already been demonstrated by theorists such as Baddeley (1964) and Wickelgren (1966) by showing acoustically similar material to be more detrimental to verbal STS than acoustically dissimilar material. Glanzer and Cunitz (1966) defined similar i ty in terms of formal similar i ty between the stimulus material and the verbal material of the interpolated ac t iv i ty . However, they found that s imi lar i ty , as defined by formal properties of the material, was not a c r i t i c a l dimension in short-term losses. Yet in the same experi-ment Glanzer and Cunitz obtained a larger performance decrease over short-term retention intervals when the interpolated act iv i ty required process-ing of mathematical problems. In other words, processing words during the maintenance of word stimuli in STS is more damaging than processing mathematical problems. Thus, if s imi lar i ty is defined in terms of the degree of s imi lar i ty between the processes required to maintain informa-tion In STS and the processes required to perform the interpolated a c t i v i t i e s , the dimension of similar i ty can account for research findings as diverse as modality speci f ic interference effects as well as acoustic similar i ty ef fects . In summary, if short-term processing of the stimulus information and processing of the interpolated act iv i ty require the activation of similar c r i t i c a l coding mechanisms a tendency of increasing stimulus losses can be expected. The other dimension, i . e . , task d i f f i c u l t y , has been demonstrated as an effective interference variable with verbal (Posner, 1965) and visual (Cohen and Granstrom, 1970) material. Although this dimension can be divided into effects due to the d i f f icu l ty in performing a given task 34 and the effects due to the amount of information to be processed in a given time (Glanzer, 1972), the theoretical interpretation of STS losses due to these manipulations in both cases can be made in terms of attention demands of the interpolated act iv i ty (Atkinson and S h i f f r i n , 1968; Posner, 1968). The general assumption here seems to be that any act iv i ty during the retention interval will take up a given amount of central processing capacity. As a resul t , the central processing capacity will be divided between maintaining short-term information and performing the interpolated act iv i ty . As the attentional demands of the interpolated act iv i ty are increased, a larger part of central processing will be diverted from maintaining stimulus information and, subsequently, greater STS losses can be expected. In general, while the effects of attention demands of the interpolated act iv i ty have been well substantiated in verbal STS (Glanzer, 1972; Kintsch, 1970), only isolated examples investigating visual STS have been reported (Posner, 1969; Cohen and Granstrom, 1970). However, the response trends obtained In changes in attention demands appear to be consistent across the verbal and visual STS. It should be noted that within the context of this research, the terms task d i f f i c u l t y , attention, and central processing, are essential ly describing the same basic phenomena on different operational and theoretical levels ( e .g . , Atkinson and Sh i f f r in , 1968). That i s , the more d i f f i c u l t a task i s , the more attention is required to perform the task properly. The amount of attention ut i l ized in the performance of a task In turn describes the amount of central processing diverted from the total central processing capacity. 35 A conceptual framework for the similar i ty and attention components of a given Interpolated act iv i ty may be construed in the following way. On the one hand, there are retention losses due to the simi lar i ty of processes required to perform the interpolated act iv i ty and the processes required to store the stimulus information. On the other hand, there are retention losses due to the diversion of the central processing capacity by the attention demands required to perform the interpolated ac t iv i ty . The relationship of these two components of a given Inter-polated act iv i ty appears, to some extent, to be addit ive. For instance, the attention demands in learning an acoustically similar versus an acoustically dissimilar l i s t appears to be constant, yet the simi lar i ty component in the f i r s t paradigm generates a larger error rate. Further-more, the visual interference condition in the experiment by Cohen and Granstrom (1970) is constant across verbal and visual coding conditions, yet leads to a greater retention loss in visual coding conditions. The additive nature of these two components therefore seem tentatively supported by research examples and should be pursued further. In light of this overview, several tentative statements can be made with respect to common and dist inct characterist ics of verbal and visual processes in STM experiments. The existence of visual coding processes in STM has already been suf f ic ient ly demonstrated. Though the relat ive eff iciency of the visual and verbal coding system cannot be direct ly investigated experimentally, in light of the inherent differences in the stimulus s i tuat ion, the general assumption that STS is a predominantly verbal system and probably a more e f f ic ient system, will be adhered to in this paper. 36 Research investigating rehearsal effects on visual coding have generated an equally unsatisfactory picture. Supportive evidence as well as data rejecting the existence of a visual rehearsal mechanism have been reported. Although a conclusion about the visual rehearsal mechanism cannot be made, the contradictory findings in visual STS, in contrast to the consistent f ac i l i t a t i ve effects in verbal STS, suggest that the different degrees of e f f ic iencies of the rehearsal processes can be con-sidered as differentiat ing characterist ics of the two systems. Interpolated act iv i ty ef fects , on the other hand, appear i n i t i a l l y to be not unlike those obtained in the verbal STS. Increases in attention demands of the interpolated act iv i ty lead to corresponding retention losses of visual ly coded material, a general trend consistently demonstrated in verbal STS. Furthermore, the dimension of s imi lar i ty , as defined in this discussion, applies equally well to verbally and visual ly coded information. However, because simi lar i ty effects appear to be operat-ing in both systems, a convincing argument for dist inct verbal and visual STS can be made. If visual interference act iv i ty is more detrimental to visual coding, qual i tat ively different coding processes must be involved in the storage of verbal and visual information. In summary, qual i tat ively different coding processes for verbal and visual STS can be inferred from the modality specif ic interference effects underlying the similar i ty dimension. The dif ferential eff iciency of the rehearsal mechanism between verbal and visual coding suggests an additional distinguishing characterist ic between the two systems. The effects of attention demands on retention losses clear ly point towards a common 37 characterist ic underlying verbal and visual STS. Given that common and dist inct characterist ics in the verbal and visual system can be ident i f ied, i t becomes of interest to evaluate the extent of independence or interdependence of the two systems within the STS framework. This theoretical issue can be investigated in a comparison of retention losses of verbal and visual coding situations requiring performance of interpolated act iv i t ies varying in the dimension of similar i ty and attention demands. If response patterns in both coding conditions do not d i f fer in l ight of changes in both similar i ty and attention demands of the interpolated task, the notion of separate systems for verbal and visual STS could be rejected. Conversely, i f changes along both dimensions of the interpolated task would generate dif ferential response patterns in verbal and visual coding, the need to postulate two quali tat ively d ist inct memory systems would be quite apparent. The other alternative, i . e . , only one of the dimensions generates a di f ferent ial ef fect , should result in a conceptual framework specifying the type and degree of interrelationship between the two postulated systems. It Is the general purpose of the present series of experiments to generate verbal and visual interpolated tasks with different degrees of attention demands in each modality, and to observe the effects of these interference conditions in experimental situations requiring verbal or visual coding. If the conceptual framework proposed in this discussion accurately des-cribes the processes in visual and verbal STS, the following general 38 trend of results should be obtained. Cross-modality interference effects should be less pronounced than same modality interference ef fects . Hence, verbal interference effects should be more damaging to the verbal code than to the visual code. Conversely, visual interpolated act iv i ty should be more damaging to the visual than verbal code. Furthermore, Inter-polated ac t iv i t i es with high attention demands should lead to a greater retention loss than interpolated act iv i t ies with low attention demands. This should occur whether processing the stimulus and the interpolated act iv i ty requires the same or different modalities. A general response pattern of this nature would support the following theoretical points: a) verbal and visual STS rely on quali tat ively different coding processes; b) both systems rely equally on attention demands in maintaining informa-tion in STS; and c) retention losses in verbal and visual STS are an additive function of the attention and similar i ty components inherent in the given interpolated ac t iv i ty . 39 EXPERIMENT I I introduction The f i r s t experiment was designed to evaluate the effects of inter-polated ac t iv i t ies varying on the dimensions of similar i ty and attention demands in an experimental setting requiring visual coding. The in i t ia l problem to be resolved was in the selection of the stimulus situation which could be interpreted in terms of visual coding requirements. As mentioned in the general introduction, there are several stimulus situations which fa l l into this category, among them: pictures of objects, drawings of schematic faces, and letters or d ig i ts placed on a spatial matrix. Since i t is d i f f i c u l t to completely eliminate possible influences of verbal coding effects in each one of these conditions, the final choice of the stimulus situation rested with a visual pattern previously used by Schnore and Partington (1967) in an investigation of visual def ic i ts in institut ionalized populations. The particular stimulus display consisted of a 4 x 4 matrix of black and white squares, where the pattern generated by the speci f ic arrangement of black squares had to be retained for subsequent reproduction. The advantage in using a pattern of this nature l ies in the absence of any verbal cues under a short presentation duration. The interpretation of the results in terms of visual coding therefore appeared quite j u s t i f i e d . The selection of appropriate interpolated act iv i tes varying along the similar i ty and attention dimensions appeared more d i f f i c u l t and was largely determined in a series of p i lo t research. Since the damaging effect of counting backwards by threes from a three d ig i t number had 40 been substantiated in previous experiments (e .g . , Peterson and Peterson, 1959), this particular task was chosen and defined in terms of a verbal interference task with high attention demands. With the intent of keeping the modality speci f ic task demands constant and decreasing the attention demands, counting forwards by one from the number one was chosen and defined as constituting a verbal interference task with low attention demands. The selection of appropriate visual interpolated a c t i v i t i e s , in contrast, turned into a more elusive project. Posner (1969) had already demonstrated that a visual matrix presented during the retention interval has no damaging effects on visual coding. This appears not too surprising, if one considers the amount and type of processing which might be required in looking at a given matrix display. Therefore, a task had to be generated requiring a substantial increase in visual processing. The final choice for the basic visual interference condition consisted of a visual pattern, similar to the stimulus display. The visual pattern contained six black squares which continuously changed at a rate of eight successive move-ments per second of two of the black squares. The visual effects obtained in this manipulation can be described as a constantly changing black pattern on a white background. The next step in the p i lot research was to create conditions dif fer ing with respect to attention demands required to process this visual information. It was assumed that different degrees of attention demands could be obtained by varying the speci f ic instructions about the nature of the visual interpolated task. Three types of instructions were i n i t i a l l y tested: one, "look at the 41 moving pattern, but let i t not disturb you in the retention of the stimulus"; two, "there will be a moving pattern", with no speci f ic instructions about the visual display; three, "concentrate on the moving pattern and follow i t with your eyes". Since the results under the f i r s t two instructions did not d i f f e r , but were s igni f icant ly less interfering with respect to the third instruction, the f i r s t and third instruction conditions were selected and defined in terms of visual interference conditions with low and high attention demands. One additional problem had to be resolved, which is inherent In the nature of a visual interference task. The issue here centers around the experimenter's lack of control and evaluation of the subject's task performance required by the interpolated task. Contrary to the verbal interference condition, where this type of assessment is readily available in S/s verbalization of the number sequence, any visual Interference lacks this bui l t in performance test . In other words, a visual Interference condition with an observable performance task had to be obtatned. Since i t was important to avoid the involvement of a verbal component as a performance check of the visual interference task, a visual-motor task was chosen to meet the above mentioned requirements. This condition consisted of the presentation of a 4 x 4 matrix of squares with successive changes in the black-white rat io of the display. In addit ion, the ratio changes were accompanied by pattern changes. The subject's task in this condition was to adjust a lever in accordance with the black and white ratio changes which had to be abstracted from the accompanying pattern changes. This Interference condition was defined 42 In t e r m s o f v e r y h i g h a t t e n t i o n demanding n o n - v e r b a l o r v i s u a l i n t e r f e r e n c e . In o r d e r t o e v a l u a t e t h e e f f e c t s o f t h e s e i n t e r p o l a t e d a c t i v i t i e s , r e s p o n s e m e a s u r e s f o r z e r o d e l a y , and f o r a r e t e n t i o n i n t e r v a l w i t h o u t any i n t e r p o l a t e d a c t i v i t y , were a l s o r e q u i r e d a s a d d i t i o n a l r e f e r e n c e p o i n t s . Wi th r e s p e c t t o t h e t h e o r e t i c a l i s s u e s wh ich were t o be r e s o l v e d i n t h i s e x p e r i m e n t , s e v e r a l p r o j e c t i o n s c o u l d be made a b o u t t h e e f f e c t s o f t h e p r o p o s e d m a n i p u l a t i o n s . O n e , i f a t t e n t i o n demands a r e a common c h a r a c t e r i s t i c u n d e r l y i n g v e r b a l and v i s u a l c o d i n g i n S T S , h i g h and low a t t e n t i o n c o n d i t i o n s w i t h i n e a c h m o d a l i t y s p e c i f i c i n t e r f e r e n c e t a s k s h o u l d r e s u l t in d i f f e r e n t i a l r e s p o n s e l e v e l s o f v i s u a l c o d i n g . Two, on t h e b a s i s o f t h e s i m i l a r i t y component i t c o u l d be p r e d i c t e d t h a t h i g h e r r e t e n t i o n l o s s e s s h o u l d o c c u r d u r i n g v i s u a l t h a n v e r b a l I n t e r f e r e n c e . However , t h e p r e d i c t i o n a b o u t the s i m i l a r i t y e f f e c t s h o u l d o n l y m a t e r i a l i z e i f t h e h i g h and low a t t e n t i o n c o n d i t i o n s o f t h e v e r b a l and v i s u a l i n t e r -p o l a t e d t a s k a r e c o m p a r a b l e . S i n c e no a s s e s s m e n t o f t h e d e g r e e o r amount o f a t t e n t i o n demands a c r o s s i n t e r f e r e n c e m o d a l i t i e s was p o s s i b l e , and s i n c e a t t e n t i o n demands and s i m i l a r i t y e f f e c t s were assumed t o be o f an a d d i t i v e n a t u r e , s p e c i f i c p r e d i c t i o n s a b o u t t h e s i m i l a r i t y e f f e c t c o u l d not be made. A c l a r i f i c a t i o n o f t h i s r e l a t i o n s h i p , however , c o u l d be e x p e c t e d i n a c o m p a r i s o n o f t h e r e s p o n s e p a t t e r n o b t a i n e d i n t h i s e x p e r i -ment w i t h t h a t o b t a i n e d i n an e x p e r i m e n t r e q u i r i n g v e r b a l c o d i n g . METHOD S u b j e c t s and D e s i g n One hundred and f i v e v o l u n t e e r s between t h e ages o f s e v e n t e e n and 43 twenty-two from the Introductory Psychology subject pool participated in the experiment and were randomly assigned to one of seven groups, except to keep the male-female ratio constant per group. Al l subjects were individually tested. The seven groups consisted of the following conditions: Group I: immediate recall (IMM). The remaining groups were tested after a ten second retention interval with the following interpolated ac t iv i t i es : Group II: rehearsal, no interpolated act iv i ty (REH); Group III: verbal interference, low attention (LVER); Group IV: verbal interference, high attention (HVER); Group V: visual interference, low attention (LVIS); Group VI: visual interference, high attention (HVIS); Group VII: visual-motor interference (VM). Material The stimulus material consisted of 4 x 4 matrices containing eight black and eight white squares (Schnore and Partington, 1967). The stimulus configuration to be remembered for subsequent reproduction was the particular pattern the eight black squares created in each matrix. Across the series of test t r i a l s the patterns were randomly generated with the restr ict ion that each cel l of the matrix was equally often f i l l e d with a black square. After randomly generating the test s t imul i , several changes had to be undertaken, according to the above mentioned rest r ic t ion , since several of the random patterns appeared as more cohesive or meaningful patterns than others. The final set of random, meaningless 44 patterns was then tested across a series of stimulus durations to obtain a recall level at zero delay meeting two c r i t e r i a : a) verbal coding should be reduced; and b) a response level should be obtained which would permit possible increases or decreases due to interpolated act iv i ty ef fects. A stimulus duration of 1.75 seconds was selected as an optimal value meet-ing the above c r i t e r i a . Four different sets of test t r i a l s , including a start s ign, retention interval , and recall cue, were filmed on standard eight mm fi lm with a single frame advance camera. The four test t r i a l s and their corresponding conditions were as follows: Film I contained the stimulus display immediately followed by the recall cue and was used for the immediate recall condition. Film II contained a retention interval which darkened out the screen for ten seconds between the stimulus and recall cue. This set was used for the rehearsal group as well as for the low and high attention verbal interference group. In film III, a ten second film str ip was inserted during the retention interval displaying a continuously changing black pattern on the screen. This condition contained a pattern of six black squares continuously changing position at a rate of eight times per second throughout the ten second retention interval . This fi lm was used for low and high attention visual interference groups. Film IV was used for the visual-motor interference group. The ten second interval displayed continuous changes in the ratio of black and white squares In the display, as well as pattern changes for each black 45 and white rat io . More spec i f i ca l l y , there were five levels of black and white rat ios, ranging from a l l black, three-quarters black, half black, one-quarter black, to no black, i . e . , a l l white. Ratio changes occurred once every second with an accompanying pattern change for a given ratio level after .5 seconds. Apparatus The films containing the material of the test t r i a l s were shown on a Kodak Instamatic movie projector. In addit ion, Ss in Group VII had to operate a lever along a seven inch continuum with end points marked black and white. The lever mechanism was connected to some inoperative test equipment to give the impression that an accurate measure for the lever response was obtained. Procedure The experimental procedure for the seven groups was identical with respect to the stimulus and response requirements. For instance, the immediate recall group was told that each t r ia l would begin with a start sign, followed immediately by a 4 x 4 matrix of black and white squares presented for 1.75 seconds. Subjects were instructed to retain the speci f ic pattern formed by the black squares, which had to be reproduced when the recall cue appeared on the screen. A scoring sheet was provided with 4 x 4 matrices. The stimulus pattern was to be reproduced by placing an X into the corresponding ce l l s on the scoring sheet for each black square on the stimulus display. For the remaining six groups, the following instructional changes had to be introduced regarding the nature of the interpolated act iv i ty . 46 Group II: (Rehearsal condition) "Following the stimulus presentation there will be ten seconds during which you are to rehearse the stimulus pattern and then reproduce i t when the recall cue is given." Group III: (Low attention, verbal interference condition) "Follow-ing the stimulus presentation, start counting forward by one, as soon as I cal l out 'one' , unti l the recall cue appears on the screen." Group IV: (High attention, verbal interference condition) "You will be given a three d ig i t number immediately following the stimulus presentation, and you are to count backwards by threes from this number until the recall cue appears." Group V: (Low attention, visual interference condition) "The stimulus will be followed by a continuously changing black pattern. Look at the changing pattern, yet try not to be disturbed by i t in retaining the stimulus pattern. However, do not close your eyes." Group VI: (High attention, visual interference condition) "The stimulus will be followed by a continuously changing black pattern. Con-centrate on this changing pattern and follow It with your eyes." Group VII: (Visual and motor interference condition) "The stimulus will be followed by a continuously changing matrix with changes in pattern and black and white ratios of the display. Adjust the lever as fast as possible along the black-white continuum in accordance with your perceptions of the ratio changes." Each S_ received 15 t r i a l s , of which only the last ten t r i a l s were considered as test t r i a l s . The f i r s t f ive t r i a l s were discarded in order to obtain a stable performance level free from distortions such 47 as practice effects and proactive interference. The following temporal sequence was maintained throughout each condition. The subject was told the t r i a l would star t , at which time the projector was activated. After one second, a Start sign appeared, followed by a .25 second blank Interval before the stimulus display. The stimulus was presented for 1.75 seconds, and was Immediately followed by the recal l cue in Group I, or the ten second interpolated act iv i ty in the remaining groups. For the delay groups, the recall cue appeared after the ten second interval at which time the projector was turned of f . Subjects were given 20 seconds to reproduce the stimulus pattern. A short rest period was given between t r i a l s for an inter-stimulus interval of 60 seconds. RESULTS AND DISCUSSION Responses were i n i t i a l l y scored with respect to the number of correct squares reproduced per stimulus pattern. The mean recall values for the ten test t r i a l s are plotted in Figure I for a l l seven conditions. In addition, the total number of correct and incorrect scores per subject were converted into d' scores, a correction factor for guessing (Kintsch, 1970). Figure 2 shows the mean d 1 values for a l l seven groups. The response trends in the two different response measures appeared to be in close agreement across the seven conditions (Appendix I, Table I), there-by indicating a relative absence of di f ferent ial guessing rates. The general trend in both figures showed a rather stable response level for the immediate and rehearsal conditions, as well as the low attention groups of both verbal and visual interference. Pronounced retention losses were 48 VERBAL VISUAL MOTOR 0-DELAY 10 SECOND DELAY ' FIG. 1 Mean scorea for correct reproductions of v i s u a l patterns i n Exp 0 I . 49 to w « o o to 3 _ IMM 0-Delay REH LOW HIGH VERBAL 10 SECOND DELAY LOW HIGH VISUAL-VISUAL MOTOR FIG. 2 Mean d' scores for visual coding in Exp0 I. 50 displayed by the high attention verbal condition and the visua I-motor condition, with intermediate losses due to high attention visual inter-ference. Analyses of variance were performed with the data of both response measures (Appendix I, Tables I and 3). Both measures revealed a s i g n i f i -cant treatment e f fect , F(6,98)=26.67; p<.01, for number of correct scores, and F(6,98)=20.74; p<.0l, for d' scores. Post hoc tests by the Newman-Keuls method produced the following signif icant comparisons in both response measures (Appendix II, Tables 2 and 4). There was no signif icant difference between the immediate recall and the rehearsal groups. Both groups in turn, did not d i f fer from the low attention groups of verbal and visual interference. High attention verbal interference, in contrast, showed a signif icant difference with respect to every other condition, a l l p's<.0l, except in the d' analysis where a smaller difference was obtained with respect to the high attention visual condition, p<.05. The high attention visual condition, in turn, revealed a greater difference with respect to the rehearsal group, p<.01, than with respect to the immediate recall condition as well as both low attention conditions in verbal and visual interference, a l l p's<.05. On the other hand, the visual motor group showed a s igni f icant ly greater retention loss in comparison with every other condition, a l l p's<.0l. The interpretation of these data within the theoretical framework suggests several tentative statements regarding evidence for s imi lar i ty and attention components underlying retention losses due to interpolated a c t i v i t i e s , yet provides l i t t l e c la r i f i ca t ion with respect to the nature 51 or effectiveness of the visual rehearsal mechanism. The problems inherent in the assessment of rehearsal effects have already been mentioned in the general introduction. They include effects of cross-modality coding of the stimulus as well as the problem of establishing a valid reference point for the assessment of the rehearsal ef fect . Cross-modality coding effects appear to be negligible in light of the absence of verbal cues inherent in the stimulus pattern and presentation time of the stimulus. As a valid reference point, the immediate recall group seems to provide the most appropriate comparison. There is no difference between immediate recall and recall after a ten second retention interval with no interpolated act iv i ty . Stating that the rehearsal mechanism does neither fac i l i ta te nor lead to an appreciable loss in retention does l i t t l e in c lar i fy ing the nature of the postulated visual rehearsal mechanism, however, especially since i t becomes doubt-ful whether the immediate recall group is indeed a valid reference point for zero delay coding ef f ic iency. After a l l , responding at zero delay has been shown to be an effective interference condition (Ternes and Yu i l l e , 1972). The nature of the response task may even be construed as containing a certain interference component leading to the relat ively low retention level in this condition. For instance, Margrain (1967), among others, has shown that a written interpolated act iv i ty constitutes an interference effect in STM. In this experiment, Ss have to reproduce the stimulus pattern by putting X 's onto a response sheet, a task which required up to twenty seconds. It Is not unreasonable to suspect that the visually coded stimulus pattern might undergo some changes during the 52 time it takes to complete the reconstruction of the pattern on the response sheet. The same argument can be made for the rehearsal group. However, since sensory memory effects at zero delay cannot be ruled out, di f ferent ia l coding losses due to the task demands of the response task are quite l ikely to occur between immediate recall and recall after a ten second retention interval . Therefore, an evaluation of rehearsal effects should not be made based on these data. On the other hand, the rehearsal group does appear as a val id reference point in evaluating interpolated act iv i ty ef fects , since possible interference effects due to the response task demands should be constant for a l l groups having a ten second retention i ntervaI. Although the response patterns generated by the Interpolated act iv i t ies appear at certain points to be unexpected, it should be noted that a sub-sequent experiment using a three second stimulus duration essential ly replicated the response trend, thereby at least substantiating the con-sistency of the Interpolated act iv i ty effects in this visual stimulus si tuat ion. Despite the consistency of the response trends, the interpre-tation of these results within the proposed theoretical framework evaluating similari ty and attention components appears at times ambiguous; Several tentative statements can s t i l l be made to summarize the theoretical imp Iications. F i r s t , interpolated ac t iv i t ies dif fering in their degree of attention demands for each modality speci f ic interference task display consistent increases in reproduction losses corresponding to increases in attention demands. This Is true for verbal interference where counting forwards 53 by ones and counting backwards by threes d i f ferent ia l ly affect the recall level . A smaller difference is obtained in visual Interference between low and high attention groups, yet, in conjunction with the visual-motor group, the response trend due to attention demands is comparable in visual and verbal conditions. Second, retention losses due to the postulated similar i ty component are more d i f f i c u l t to isolate. There are no response differences between visual and verbal interference for the respective low attention groups. Furthermore, there is a difference between the high attention groups of both modalities, yet, i t is the verbal interference which leads to a higher retention loss. In other words, there is no apparent indication that visual interference is more detrimental than verbal interference on visual coding. To i l lustrate this problem, response differences in the two verbal interference conditions can be considered as a function of the increased central processing demands required to count backwards, since the modality speci f ic task component, i . e . , the verbal production of the number sequence, is relat ively constant across both groups. Hence, there is no retention loss due to the modality speci f ic task component in the verbal Interference condition, i . e . , there is no similar i ty ef fect . However, the same conclusion can be drawn about the visual interference groups, since the modality speci f ic component in the low attention group does not lead to a signif icant retention loss. Subsequently, the retention loss of the high attention visual group can be interpreted again in terms of losses due to attention demands. The visual-motor effects then appear as a mere 54 extension of the attention demand continuum. Thus, the postulated sim-i l a r i ty component, i . e . , visual interference should be more detrimental to visual coding than the effects of verbal interference, does not emerge in this experiment. However, since the amount of attention demands required to perform interpolated act iv i t ies in both modalities is unknown, the arb i t rar i ly defined low and high attention groups across modalities cannot be expected to be comparable on this dimension, in ef fect , while several alternative interpretations of similar i ty and attention effects can be forwarded, a detailed evaluation of these effects has to await a comparison of response trends in Experiment I with those obtained in an experimental setting requiring verbal coding. The rationale underlying this expectation centers around the following argument. If a comparable response trend is observed in verbal coding conditions, an Interpretation of interference effects can be made in terms of attention demands alone. If a di f ferent ial response pattern emerges, other than attention components must be incor-porated in rhe interpretation of re-i*?ntion losses. 55 EXPERIMENT I I Introduction Experiment II was designed to evaluate the effects of the interpolated act iv i t ies used in Experiment I in a stimulus situation requiring predom-inantly verbal coding. The theoretical value of this experiment was: a) in providing additional data to evaluate the postulated simi lar i ty and attention components of the interpolated ac t i v i t i es ; and b) In providing Information about the interrelationship or independence of the verbal and visual systems in STS. The particular paradigm selected for this experiment consisted of the presentation of three words per test t r i a l , in other words, a stimulus situation which has been frequently used in previous experiments invest i -gating verbal STM function (e .g . , Murdock, 1963; Peterson and Peterson, 1959). The advantage of using this method l ies in the well established response pattern obtained at zero delay as well as after retention intervals f i l l e d with verbal interference, i . e . , counting backwards by threes. Depending on the theoretical assumptions which are adhered to with respect to s imi lar i ty and attention components of the interpolated ac t iv i ty , the following predictions can be made about the expected results. If retention losses due to the interpolated ac t iv i t ies are assumed to be an additive function of the similar i ty and attention component of the inter-ference task, a different response pattern should be obtained in this experiment in comparison to Experiment I. More spec i f i ca l l y , verbal inter-ference conditions should lead to a larger retention loss with respect to 56 visual Interference conditions, relat ive to the response pattern obtained in visual coding conditions. On the other hand, if the postulated simi lar i ty effects are non-existent, and losses In a given interpolated act iv i ty con-dit ion are a direct function of the amount of central processing required to perform the interpolated task, an identical response pattern should be obtained in th is experiment as found in Experiment I. Regardless of the response pattern obtained, the results should lead to some informa-tion c lar i fy ing the extent to which the postulated simi lar i ty and attention components can be considered as crucial factors underlying retention losses during interpolated a c t i v i t i e s . METHOD One hundred and five Ss from the same subject pool participated in this experiment. Fifteen S_s were randomly assigned to each of seven groups except to maintain an equal male-female ra t io . Subjects were again tested individual ly. The general procedure was identical to that in Experiment I, except for the stimulus display and the response task. The stimulus display consisted of a consecutive presentation of three words, each word presented for 5/16 second, with no inter-item interval . The Individual stimulus items consisted of one and two syl lable concrete nouns with high frequency rat ing, selected from a standard l i s t of rated words (Paivio, Yui l le and Madigan, 1968). It should be noted that a l l stimulus words had been previously used in an STM task (Ternes and Yu i l l e , 1972). Further-more, Ss were given response sheets and instructions for a standard written recalI tes t . To br ief ly summarize the general procedure, the seven groups differed 57 with respect to Experiment I only In the type of stimulus material and stimulus duration, as well as the subsequent recall task. The following experimental factors remained constant with respect to Experiment I: one: the material for the Interpolated a c t i v i t i e s , includ-ing start and recall signs; two: temporal parameters of interpolated ac t iv i t ies and inter-t r i a l intervals; three: Instructions with respect to Interpolated ac t i v i t i es ; four: the number of practice and test t r i a l s ; f ive: general apparatus, e . g . , f i lm presentation. RESULTS AND DISCUSSION The mean numbers of words correctly recalled were tabulated for the last ten t r i a l s and are shown in Figure 3. An almost perfect recall score was obtained for the immediate recall and rehearsal groups (e .g . , Murdock, 1963). At the same time, both visual interference conditions displayed a similar ly high response level . The verbal interference conditions, in contrast, showed a decrease in word r e c a l l , with a substantial drop between the low and high attention conditions. The visual-motor group also showed a decrease in r e c a l l , yet, unlike in Experiment I, the v isua l -motor group was not the most damaging interference condition (Appendix I, Table I). An analysis of variance was performed on recall scores which revealed a signif icant treatment ef fect , F(6,98)=93.65; p<.OI. (Appendix II, Table 5). Post hoc tests by the Newman-Keuls method were used to further c la r i fy speci f ic comparisons between individual conditions (Appendix II, Table 6). 58 1 FIG. 3« Mean correct recall scores in Expe II» 59 These comparisons generated the following recall trend across conditions. There was no signif icant difference between the immediate recall and rehearsal group. Both, in turn, did not d i f fer from the low and high attention groups of visual interference. In contrast, both verbal inter-ference conditions as well as the visual-motor condition showed a decrease with respect to the rehearsal condition, and revealed s i g n i f i -cant differences with respect to each other, a l l p's<.OI. The results show a definite ce i l ing effect for the immediate r e c a l l , rehearsal, and low and high attention groups of visual interference. In e f fect , recall appears to be almost errorless for a l l four groups. How-ever, there is a highly signif icant dif ferentiat ion between the remaining three groups. Br ie f ly , recall differences due to attention demands appear in both modality speci f ic tasks. As central processing demands are increased from counting forwards to counting backwards, corresponding recall losses can be observed between the two verbal Interference con-ditions as in Experiment I. Though a similar retention loss is not obtained between the low and high attention groups in visual interference, a comparable trend emerges along the attention continuum In the visual modality, if the visual-motor condition is considered as an additional level of visual interference with very high attention demands. Further-more, since in the low attention groups verbal interference is more detrimental than visual interference, there is some indication that this loss can be attributed to the similar i ty component inherent in the verbal condition. However, this is not necessarily the case since i t could be argued that counting forward contains a larger attention component than 60 merely looking at a continuously changing visual pattern. In other words, the assessn.ant of s imilar i ty and attention components in Experiment II contains similar problems as in Experiment I. It appears therefore that as long as attention components cannot be specified across modalities of the interpolated task, s imi lar i ty effects within experiments cannot be clearly isolated. A detailed comparison of response trends across Experiments I and II are expected to c lar i fy the speci f ic relationship between similar i ty and attention components underlying retention losses due to Interpolated a c t i v i t i e s . 61 DISCUSSION EXPERIMENTS I and II The general purpose of the two experiments was to evaluate the extent to which the postulated similar i ty and attention components of interpolated act iv i t ies can be shown as the crucial factors underlying retention losses in STM tasks. The conclusions drawn from an evaluation of retention losses were in turn expected to provide information regarding the nature of verbal and visual STS. With respect to the f i r s t issue, i . e . , s imi lar i ty and attention components, several conclusions can be drawn from the results of the two experiments. Both experiments attest to the importance of attention demands underlying losses of information in STS. As a larger proportion of central processing is required in the performance of the interpolated ac t iv i ty , a larger retention loss can be observed. This trend appears in experimental situations requiring either verbal (Experiment II) or visual (Experiment I) coding and is true for interpolated ac t iv i t i es in the verbal and visual modality.- In addition, the dif ferential response trends across experiments clearly point out that attention demands are not suff ic ient to explain retention losses. Results of this nature suggest the need to postulate other than attention components in the interpretation of retention losses in STS. There are three speci f ic comparisons of interpolated act iv i ty effects across experiments which support the existence of the similar i ty component as an important inter-ference factor i n STS. One, the low attention verbal interference condition has no s i g n i f l -62 cant effect on visual coding (Experiment I), yet produces a signif icant recall loss in verbal coding (Experiment II). If the 'nterpolated act iv i ty for this condition is evaluated in terms of s imi lar i ty and attention components, Experiment I shows that counting forward by ones contains an insignif icant similari ty effect as well as an insignif icant attention component. After a l l , visual coding does not decrease in this condition. The same interpolated ac t iv i ty , however, leads to a loss in verbal r e c a l l . Since i t is safe to assume that attention demands inherent in counting forwards remain constant across the two experiments, recall losses in Experiment II must be attributed to the similar i ty ef fect . In other words, the modality speci f ic task component in the counting task does not affect the visual code, yet interferes with recall of verbal material. Two, the low attention visual groups do not lead to retention losses in either coding conditions. In other words, the effects of the simi lar i ty and attention components of the interpolated act iv i ty do not add up to generate a signif icant re'tention loss. However., the high attention visual 'condit ion d i f ferent ia l ly affects the two coding conditions. The retention loss.for this interference condition was i n i t i a l l y interpreted in Experiment I in terms of attention demands. If this interpretation is correct, a similar retention loss due to the identical attention demand component would have to be expected in Experiment II, yet this is not.the case. Consequently, i t can be argued that the retention loss in Experi-ment I must be due to the simiIarity component inherent in this experi-mental s i tuat ion, where visual processing is required in retaining the 63 stimulus information and in performing the interpolated task. The same visual processing component of the interpolated act iv i ty does not inter-fere with verbal coding. This evidence for modality speci f ic interference further confirms the postulated existence of the similar i ty component as an important factor in STS losses. Three, a comparison of the relative effects of the high attention verbal groups and the visual-motor groups across both experiments provides further support for the additive hypothesis of retention losses due to similar i ty and attention components. Again, if retention losses are due to attention demands alone, the visual-motor condition should lead to greater recall losses than the high attention verbal group in the results of Experiment II, as was observed in Experiment I. However, the order of increasing losses due to these conditions is clear ly reversed in Experiment II. Despite the lack of a direct s ta t is t ica l comparison, i t is apparent that counting backwards is the most damaging interference condition In verbal coding, and the.visual-motor group leads to the largest retention loss In visual coding. Honce, the attention, component by i tse l f is insuff icient to explain the retention losses in this comparison. In l ight of the results of both experiments, i t should be noted that the speci f ic labels supplied to the different Interference conditions with respect to attention demands can be rather misleading at this time. For instance, the low attention visual interference condition should not any more be considered as a comparison condition for the low attention verbal interference condition. In fact , the results strongly indicate that the high attention visual condition and the visual motor condition 64 a r e t h e r e s p e c t i v e c o m p a r i s o n g r o u p s f o r low and h i g h a t t e n t i o n v e r b a l i n t e r f e r e n c e , In summary, a c o m p a r i s o n o f r e s p o n s e t r e n d s a c r o s s b o t h e x p e r i m e n t s c l e a r l y p o i n t s o u t t h a t a t t e n t i o n demands c o n s t i t u t e m e r e l y one component u n d e r l y i n g r e t e n t i o n l o s s e s in S T S . The d i f f e r e n t i a l r e t e n t i o n l o s s e s a c r o s s v i s u a l and v e r b a l c o d i n g e x p e r i m e n t s f o r t h e low a t t e n t i o n v e r b a l and t h e h i g h a t t e n t i o n v i s u a l g r o u p s , as w e l l as t h e r e v e r s e d o r d e r o f r e t e n t i o n l o s s e s due t o h i g h a t t e n t i o n v e r b a l and v i s u a l - m o t o r g r o u p s a c r o s s c o d i n g c o n d i t i o n s , s t r o n g l y s u g g e s t s t h e e x i s t e n c e o f o t h e r t h a n a t t e n t i o n l o s s e s . The i n t e r p r e t a t i o n o f t h i s a d d i t i o n a l f a c t o r i n t e r m s o f s i m i l a r i t y a p p e a r s t o be j u s t i f i e d , s i n c e i n a l l o f t h e above m e n t i o n e d i n s t a n c e s v e r b a l i n t e r f e r e n c e a p p e a r s more d e t r i m e n t a l t o v e r b a l c o d i n g t h a n t o v i s u a l c o d i n g , and c o n v e r s e l y , v i s u a l i n t e r f e r e n c e l e a d s t o a l a r g e r r e t e n t i o n l o s s o f v i s u a l c o d i n g t h a n o f v e r b a l c o d i n g . In e f f e c t , t h e i n t e r p r e t a t i o n o f r e t e n t i o n l o s s e s in te rms o f b o t h g e n e r a l and s p e c i f i c i n t e r f e r e n c e e f f e c t s becomes a p l a u s i b l e c o n c e p t u a l i z a t i o n o f STS p r o c e s s e s . The i n t e r p r e t a t i o n o f r e t e n t i o n l o s s e s in t e r m s o f s i m i l a r i t y and a t t e n t i o n c o m p o n e n t s , in t u r n , c o n t a i n s s e v e r a l i m p o r t a n t i m p l i c a t i o n s w i t h r e s p e c t t o t h e c o n c e p t u a l i z a t i o n o f STS in te rms o f v e r b a l and v i s u a l s y s t e m s . F i r s t , a t t e n t i o n demands in te rms o f c e n t r a l p r o c e s s i n g a p p e a r t o be g e n e r a l and can be c o n s i d e r e d as a common f a c t o r u n d e r l y i n g r e t e n t i o n l o s s e s In v e r b a l and v i s u a l c o d i n g . T h i s i s d e m o n s t r a t e d by c o m p a r a b l e r e t e n t i o n l o s s e s in bo th c o d i n g s y s t e m s due t o c h a n g e s i n a t t e n t i o n demands o f t h e i n t e r p o l a t e d a c t i v i t y , r e g a r d l e s s o f t h e m o d a l i t y 65 of the interference task. Second, if the maintenance of the stimulus information and the performance of the interpolated act iv i ty require qual i tat ively similar processes, a larger retention loss is obtained than if different processes are required in the performance of the two tasks. If i t can be postulated that the attention and simi lar i ty components are important factors underlying retention losses, i t becomes of interest to consider to what extent these two factors are Involved In the main-tenance of information in STS. For example, i t may be argued that reten-tion losses over STM intervals should not occur in an experimental setting u t i l i z ing a retention interval which contains no attention and no similar i ty component. The basic questions posed in th is example are the following: a) does the absence of the attention and similar i ty components during the retention interval describe the optimal level of STS functioning; b) can this state of optimal functioning be described in terms of processes underlying these two components? If these two questions could be answered in the affirmative, i t could be postulated that the maintenance of information in STS rel ies on the ava i lab i l i ty of the central processing capacity in conjunction with modality speci f ic coding processes. In experiments investigating verbal and visual functions in STM, common factors underlying the observations could then be con-sidered as defining c r i t e r i a for STS processes. Conversely, modality speci f ic factors which might be observed can be considered as defining c r i t e r i a for separate verbal and visual systems in STS. A detailed \ evaluation of this conceptualization, however, has to await the c l a r i f i c a -66 tlon of one assumption which has been consistently made so far. The conceptual framework proposed has been based on data considering the visual-motor group as a visual interference condition requiring relat ively high attention demands. While this Interpretation might be appropriate in so far as the visual-motor task does not contain a verbal component, the unexpectedly high retention losses in both experiments warrant a more detailed evaluation of the visual-motor task. Though it can be argued that the motor component, i . e . , one lever movement per second, does not constitute a signif icant interference component, and i t has been shown that looking at a changing pattern, i . e . , low attention visual group In Experiments I and II, does not lead to retention losses, i t Is not clear what speci f ic aspect or combination of performance demands generate this pronounced retention loss in both coding conditions. It becomes, therefore, of further interest to Investigate what speci f ic components, i . e . , attention, s imi lar i ty , or motor, might be underlying retention losses of the v I sua I r-motor task. 67 EXPERIMENT I I I Introduction The purpose of this experiment was to evaluate the relative con-tributions of separate components of the visual-motor Interference task to the retention losses obtained in Experiments I and II. It was postulated ear l ier that retention losses were an additive function of the similar i ty and attention components of the given interpolated act iv i ty . However, the presence of the motor task in this Interpolated act iv i ty appeared as a possible confounding factor in the interpretation of retention losses. The effects of the motor task as a potential inter-ference component in STM research had already been demonstrated by Posner (Posner, 1969; Posner and Konick, 1967). Two dimensions had been identified as possible sources of interference: a) response compati-b i l i t y ; and b) the d i f f i cu l ty of the motor task. The relationship between these two components of the motor task and the concept of attention as defined in this research appeared unmistakable. Motor task d i f f i cu l ty seemed to be related to the general concept of task d i f f i c u l t y . Response compatibil ity, on the other hand, also appeared to be related to central processing requirements. For instance, if the task input consists of discrete changes and a continuous motor adjustment is required, more central processing is required than if the input and motor task are in discrete units. A close evaluation of the experimental requirements inherent in the visual-motor condition resulted in the following tentative conclusions about motor task d i f f icu l ty and response compatibil ity. The motor task in both experiments consisted of lever adjustments 68 t o a v i s u a l p r e s e n t a t i o n which d i s p l a y e d p a t t e r n changes every .5 seconds, and black-white r a t i o changes once per second. There were f i v e d i s t i n c t black-white r a t i o l e v e l s . I t should be noted here t h a t , since p a t t e r n changes could not be produced f o r the a l l black and a l l white l e v e l s , a s i n g l e white or black square was introduced r e s p e c t i v e l y as a simula-t i o n of patter n changes f o r these two l e v e l s . In e f f e c t , S_ had t o a b s t r a c t the r a t i o changes from the patter n changes and adju s t the lever a c c o r d i n g l y . Although a response measure was not obtained on the motor performance, the task appeared t o be very d i f f i c u l t . In f a c t , the task appeared t o be impossible to execute with any degree of accuracy, as indi c a t e d byS_s' post experimental r e p o r t s . Thus, i t was safe to place t h i s task a t the extreme end of the motor task d i f f i c u l t y dimension. An ev a l u a t i o n of the response c o m p a t i b i l i t y dimension suggested t h a t the experimental c o n d i t i o n s were indeed not h i g h l y compatible, at l e a s t with respect t o one c r u c i a l aspect of the task. That i s , changes of d i s t i n c t v i s u a l l e v e l s had to be t r a n s l a t e d i n t o continuous motor adjustments. In summary then, the visual-motor task could be considered as co n t a i n i n g a high l e v e l of motor task d i f f i c u I t y as we 11 as a low l e v e l of response c o m p a t i b i l i t y . Hence, in order t o evaluate p o s s i b l e r e t e n t i o n losses due to these two components of the motor t a s k , an i n t e r f e r e n c e c o n d i t i o n c o n t a i n i n g a smaller motor task d i f f i c u l t y com-ponent as well as a lar g e r response c o m p a t i b i l i t y component had t o be created f o r an e f f e c t i v e comparison c o n d i t i o n . Pi l o t work i n d i c a t e d t h a t a response apparatus c o n t a i n i n g f i v e microswitches placed in a semi-c i r c u l a r arrangement, t o f i t the curvature of the hand, might c o n s t i t u t e 69 a less d i f f i c u l t and more compatible response condition. In this condition, Ss would have to press one of the buttons in accordance with one of the f ive levels of black-white rat ios. Assuming that a comparison between adjusting a lever and pressing a button could, at least to some extent, evaluate retention losses related to the speci f ic motor component of the interpolated ac t iv i ty , i t also became necessary to evaluate to what extent the visual input of the inter-polated task was responsible for retention losses. If an appropriate analogue of the visual input information could be found in the verbal modality, the s imi lar i ty component could be further evaluated. However, a direct translation of the visual information into verbal components appeared impossible. Although there were five different ratios which could be easi ly translated into f ive different numbers presented auditor-ial ly , an attempt to find an analogue for the superimposed pattern changes resulted in a rather fu t i le search. The f inal verbal analogue of the visual information to be processed during the interpolated act iv i ty consisted of the auditory presentation of a number sequence from one to f i ve , where one represented black and f ive represented white. The taped number system corresponded to the appropriate ratio changes in the visual condition. Apart from number changes which occurred once per second, each number was repeated once at .5 second in an attempt to create an analogue of pattern changes of the visual presentation. The obvious shortcoming of this auditory analogue was, of course, the absence of a comparable distort ion of information which was obtained in the visual presentation by the accompanying pattern changes. It should be noted here 70 that the shortcomings of this auditory analogue contain potentially serious implications with respect to the interpretation of resul ts . For example, any response differences between verbal and visual presentations of the interpolated condition would have to be considered in terms of modality speci f ic effects as well as corresponding changes in task demand requi rements. In order to obtain more information about the nature of retention losses obtained in the visual-motor condition of Experiments I and II, a factorial combination of both response demands and both presentation modalities of the interpolated task, tested under visual or verbal coding conditions, was considered as an appropriate assessment. Because of the exploratory nature of this experiment, speci f ic predictions were not made. Regardless of the resul ts , though, i t was expected that the data should provide some information to isolate s imi lar i ty , attention, and motor components underlying retention losses of the visual-motor task. METHOD Subjects and Design Eighty Ss from second year undergraduate Psychology courses par-ticipated in this experiment. Except to keep male-female ratios constant, ^s were randomly assigned to one of eight groups for individual test ing. Four groups received visual patterns and the other four groups received word st imul i , as in Experiments I and II. Each stimulus condition was tested under four different types of interpolated a c t i v i t i e s , consisting of a factorial combination of the two motor tasks and the two presentation modalities of the interpolated ac t iv i ty . In e f fect , for each stimulus 71 condition, two groups received the visual presentation of the interpolated act iv i ty and had to respond either by adjusting a lever or by pressing a button. Similar ly , two groups in each stimulus condition received the auditory analogue of the interpolated act iv i ty and responded with one of the two motor tasks. Material The stimulus material, i . e . , the visual patterns and sets of three words from Experiments I and II, were again used for the visual and verbal coding conditions. In addit ion, the visual pattern sequence used in the visual-motor conditions in Experiments I and II was used for a l l conditions tested with visual presentation of the interpolated act iv i ty . For the groups receiving an auditory presentation of the interpolated ac t iv i ty , a number sequence, representing the corresponding black-white rat io changes of the visual presentation, was recorded and presented on a tape recorder. The auditory presentation consisted of a recording of the numbers one to f i ve , arranged in accordance with the ratio changes. The numbers were presented at a rate of one per second, with each number repeated once after .5 second. Apparatus The stimulus material and the visual interpolated act iv i ty were presented on f i lm. For the auditory presentation of the interpolated ac t iv i ty , the screen was darkened out during the retention interval , and a tape recorder was activated to present the number sequence. The apparatus for the response task of the interpolated act iv i ty consisted of either a lever arrangement labelled "black to white" or "one to f ive" 72 for the visual or auditory presentation of the interpolated act iv i ty respectively, or consisted of a f ive button arrangement with similar labels. The response apparatus in each condition was hooked up to an elaborate set of inoperative test equipment, to give the impression that accurate measurements were being obtained on the performance of the interpolated ac t iv i ty . Procedure The general procedure was unchanged from the previous experiments. In fact , the visual-motor conditions requiring lever adjustments were essential ly replications of the visual-motor groups in Experiments I and II. The remaining groups differed only with respect to the speci f ic task demands of the interpolated ac t iv i ty . For instance, the v isua l -motor group requiring button press responses received the visual presenta-tion of the interpolated act iv i ty and was instructed to press the corresponding buttons labelled from a l l black to a l l white. The auditory interpolated act iv i ty groups, in turn, received the taped number sequence during the retention interval and had to either adjust the lever labelled one to f i ve , or press the corresponding buttons in the other group. These four interpolated act iv i ty conditions were tested under both visual and verbal stimulus conditions. The response task to the stimulus material consisted, as in Experiments I and II, of either the reproduction of the stimulus pattern or written recall of the verbal st imul i . RESULTS AND DISCUSSION Since the dependent measures obtained in the visual and verbal stimulus conditions di f fered, the data were analyzed as two separate 73 e x p e r i m e n t s . R e s p o n s e s f o r t h e v i s u a l s t i m u l u s c o n d i t i o n s w e r e , as i n E x p e r i m e n t I, r e c o r d e d w i t h r e s p e c t t o t h e number o f c o r r e c t r e s p o n s e s as w e l l as d ' s c o r e s . F i g u r e s 4 and 5 show t h e r e s p e c t i v e mean v a l u e s f o r t h e f o u r i n t e r f e r e n c e c o n d i t i o n s . The g e n e r a l r e s p o n s e t r e n d in bo th measures i n d i c a t e d a l a r g e r r e t e n t i o n l o s s f o r v i s u a l l y than f o r a u d i t o r i a l l y p r e s e n t e d i n t e r p o l a t e d c o n d i t i o n s , w i t h o n l y m i n o r d i f f e r e n c e s due t o t h e two motor t a s k s ( A p p e n d i x I, T a b l e 2). A 2 x 2 a n a l y s i s o f v a r i a n c e w i t h both r e s p o n s e measures was p e r f o r m e d w i t h f a c t o r s : a) m o d a l i t y o f t h e i n t e r p o l a t e d c o n d i t i o n , v i s u a l v e r s u s a u d i t o r y ; and b) r e s p o n s e t a s k , l e v e r v e r s u s b u t t o n . Both a n a l y s e s showed a s i g n i f i -c a n t m o d a l i t y e f f e c t o f t h e i n t e r p o l a t e d c o n d i t i o n , F(I,36)=I0.80; p<.OI, f o r c o r r e c t r e s p o n s e s , and F(I,36)=8.95; p<.OI, f o r d ' s c o r e s ( A p p e n d i x I I , T a b l e s 7 and 8). T h e r e Was no s i g n i f i c a n t e f f e c t w i t h c h a n g e s in motor t a s k s . In o t h e r w o r d s , r e g a r d l e s s o f t h e motor t a s k , t h e v i s u a l p r e s e n t a t i o n was more d e t r i m e n t a l t h a n t h e a u d i t o r y p r e s e n t a t i o n o f t h e i n t e r p o l a t e d a c t i v i t y t o t h e r e t e n t i o n o f v i s u a l l y p r e s e n t e d s t i m u l u s i n f o r m a t i o n . The mean r e c a l l s c o r e s f o r t h e v e r b a l s t i m u l u s c o n d i t i o n s a r e shown i n A p p e n d i x I, T a b l e 3, and r e v e a l e d a s i m i l a r , y e t more ambiguous r e s p o n s e t r e n d a c r o s s i n t e r p o l a t e d c o n d i t i o n s . T h e r e a p p e a r e d t o be a g e n e r a l d e c r e a s e i n r e c a l l between a u d i t o r y and v i s u a l i n t e r p o l a t e d c o n d i t i o n s . A t t h e same t i m e t h e r e a p p e a r e d t o be a d i f f e r e n c e between motor t a s k s , a t l e a s t f o r t h e v i s u a l i n t e r p o l a t e d c o n d i t i o n s ( F i g u r e 6). An a n a l y s i s o f v a r i a n c e s u p p o r t s t h i s i n i t i a l i m p r e s s i o n ( A p p e n d i x I I , T a b l e 9). T h e r e was a . s i g n i f i c a n t main e f f e c t due t o m o d a l i t y , F(1,36)=I7.30; p<.01, 7 5 CO w « o o to 2.0 1.5. 1.0 .5 0„0 -o5 BUTTON LEVER JL AUDITORY VISUAL PRESENTATION MODALITY OF INTERPOLATED ACTIVITY \ \ ' FIG. 5 Mean d« scores for visual coding in Exp0 III, \ \ PRESENTATION MODALITY OF INTERPOLATED ACTIVITY i FIG. 6 Mean values of verbal r e c a l l scores i n Exp. I I I . 77 and no differences due to the motor tasks. However, the interaction between modality and motor task was also s igni f icant , F(I,36)=4.08; p<.05. An analysis of simple effects c la r i f i ed the nature of this inter-action (Appendix II, Table 10). There was a greater retention loss for visual than auditory presentations of the interpolated act iv i ty with button press responses, F(I,36)=19.09; p<.0l, than with the lever response, F(I,36)=2.28; p<.05. In addition, there was a corresponding difference between motor responses to visual presentations, F(I,36)=6.93; p<.05. In other words, there was no difference between response levels for both auditory groups and the lever adjustment group in the visual presentation. However, a greater recall loss was obtained when the interpolated act iv i ty consisted of pressing the corresponding buttons to visual changes during the retention interval . An interpretation of these data within the theoretical framework proposed generates the following tentative conclusions about s imi lar i ty , attention, and motor components underlying retention losses. The s ta t is t ica l analysis for both coding conditions shows a s i g n i f i -cant main effect for interference modality. Visual input during the interpolated act iv i ty leads to a larger retention loss than verbal input. While this response trend in the visual coding condition i n i t i a l l y suggests that a similar i ty component is underlying retention losses, i . e . , visual interference Is more detrimental to visual coding than verbal interference, a similar main effect in the verbal coding condition contra-dicts this interpretation. Thus, as already mentioned in the introduction to this experiment, 78 response differences between visual and auditory presentations of the interpolated act iv i ty must be considered in light of modality changes as well as inherent changes in the attention demands. The recall differences in the verbal coding condition cannot be interpreted in terms of a main effect for s imi lar i ty , since the visual presentation leads to the greatest retention loss. Retention losses must therefore be considered as a function of di f ferential attention demands between modalities. This suggests that a motor response to visual ly presented black and white ratio changes, which have to be abstracted from pattern changes, contains an inherently higher central processing component than a response to a number sequence. S t i l l , there are two speci f ic aspects of the data which suggest response differences due to the similar i ty component. F i r s t , the response differences between auditory and visual presentations are relat ively greater in the visual than verbal coding condition (Appendix I, Table 4). The larger percentage losses in visual coding could be an indicator of an additional retention loss due to the simi lar i ty ef fect . Second, there is no verbal recall loss between modalities in the motor task requiring adjustments, whereas the respective comparison in visual coding shows a signif icant difference. However, an interpretation of these aspects of the data-in terms of similar i ty appears not j u s t i f i e d , since the data clearly point out that attention, s imi lar i ty , and motor components do not exist in isolat ion, but display a complex interaction between stimulus coding and the total interpolated act iv i ty demands. In other words, there is a relat ively strong indication for retention losses due to attention 79 demands which appears to be overshadowing any clear effects due to simi-la r i ty . The evaluation so far , however, ignores the third postulated inter-ference component, i . e . , the motor task, which appears to be equally ambiguous in i ts interpretation. There is no difference due to the motor task changes in the visual coding conditions. In contrast, there is a small s ignif icant drop between lever adjustment and button pressing in the visual interpolated task for verbal coding. In fact , this response trend is a definite reversal from predictions made on the basis of d i f f i cu l ty and response compatibility dimensions discussed in the intro-duction . Information gained in S_'s post experimental reports suggests several interesting points which might lead to some c la r i f i ca t ion of this problem. Responding to the auditor ial ly presented number sequence by either adjust-ing a lever or pressing a button appeared to be a task well within the capabil i ty of each S_, after a few practice t r i a l s . This is not too sur-pr is ing , since the task basically requires one response to a d ist inct number, once per second. Except for different latencies, the task could be performed with relative ease in both coding conditions. In other words, the motor component did not constitute a signif icant problem in the performance of the interpolated ac t iv i ty , and retention losses must be largely attributed to central processing requirements inherent in processing the rapidly presented information. Assuming that a similar motor component is involved in responding to visual information, the increases in retention losses during visual presenta-80 tions of the interpolated act iv i ty subsequently constitute an increase in central processing required in recognizing and abstracting the rat io changes from the total display sequence, in contrast to central process-ing demands required to respond to number sequences. The verbal recall difference in motor tasks for visual Interpolated material, however, shows that this is not the case in every comparison. Some c la r i f i ca t ion can again be forwarded from Ss' reports. When responding to the visual changes with lever adjustments, S_ is responding continuously, whether correctly or incorrectly, since the S_'s hand always remains on the lever. In the button press condition, in contrast, ^ has to make a d is t inct response in order to indicate that a given ratio level has been recognized. It seemed that the speed of ratio changes and the d i f f i cu l ty in recogniz-ing and abstracting the ratio changes from the pattern changes constitutes a situation where d is t inct motor responses cannot be made with any great degree of certainty. In other words, button pressing appears to be a less compatible response condition in this s i tuat ion. Thus, it is possible to expect a larger central processing component to be involved in this motor condition in visual ly presented Interpolated a c t i v i t i e s . Elusive as this explanation i s , the Immediate question arising i s , of course, why does this difference not emerge In visual coding conditions. The answer here Is rather easy to f ind , though does l i t t l e to c la r i fy the reason for motor task differences: the retention level for visual stimuli in visual interpolated conditions shows a definite f loor ef fect . The percentage of correct reproductions are 22.40$ and 21.84$ for button and lever responses (Appendix I, Table 4), and d' scores put the retention 81 levels in these conditions very near to zero (Appendix I, Table 3). In other words, there is really no room for di f ferent ial responding at this response level . Unsatisfactory, and at times rather subjective, as these explanations might be, the overall evaluation of this experiment in terms of isolating c r i t i c a l components of the interference task generates several tentative conclusions: a) changes in attention demands, though confounded with presentation modality, constitute a c r i t i c a l factor in retention losses; b) similar i ty effects do not clearly emerge in these conditions; and c) the motor task appears as a rather insignif icant aspect of retention losses. In e f fect , retention losses in visual-motor or verba I-motor conditions are not due to the motor performance, but instead are due to central processing demands required to process the information inherent in the visual or verbal presentation of the interpolated act iv i ty . In addition, attention and similar i ty components in these tasks appear to be confounded and cannot be clearly separated. In summary, retention losses due to the postulated similar i ty effect are largely overshadowed by corresponding changes in central processing demands between auditory and visual presentations. Furthermore, response differences due to the motor tasks did not emerge, except for one com-parison where the differences could be interpreted in terms of central processing demands. Though this experiment did not succeed in clear ly delineating the effects of attention, s imi lar i ty , and motor components, i t provides consistent evidence in support of the interpretation of the visual-motor groups in Experiments I and II. That i s , retention losses 82 due to the visual-motor condition in the previous experiments can be interpreted without reference to a signif icant retention loss due to the motor component. 83 GENERAL DISCUSSION The overall aim of this research project was to investigate STS processes, especial ly with respect to the question of verbal and visual systems. It seemed of particular interest to evaluate to what extent the two postulated coding systems could be considered as independent or interrelated aspects of STS. It was reasoned that common and dist inct characteristics of the verbal and visual STS could be explored by observ-ing the relative coding eff iciency in both modes after a series of inter-polated ac t iv i t ies varying along two dimensions, i . e . , attention and s imi lar i ty . The overall results confirm that common and specif ic retention losses can be obtained employing these experimental manipulations. Retention losses due to changes in attention demands appeared to be com-parable in the visual and verbal coding conditions, regardless of the modality of the interpolated act iv i ty . Therefore, attention diversion can be considered as a common factor underlying retention losses in verbal and visual coding. In addition, when the maintenance of the stimulus information and the performance of the interpolated task required processing of information of the same coding mode, retention losses were larger than if processing of the two tasks involved different modes. Retention losses of this nature were interpreted in terms of similar i ty ef fects . Since common and dist inct components underlying retention losses in verbal and visual coding can be ident i f ied, i t was suggested that the maintenance of information in STS could be described in terms of these 84 two retention loss components. That i s , given an experimental condition in which: a) no attention diversion is introduced, i . e . , the total of the central processing capacity is ut i l i zed for stimulus coding; and b) in which there is no processing of addit ional , same modality Informa-t ion , a state of optimal short-term processing should be observed. This state could then be described as consisting of central processing plus modality speci f ic coding processes. The effects of attention diversion underlying retention losses in STM experiments has been discussed in detail by theorists such as Atkinson and Shi f f r in (1968). In fact , this general concept of attention is equally applicable to the models of Waugh and Norman (1966) and Neisser (1967), and is also quite in agreement with the concept of attention in primary memory used by William James (1896). In James' terms, informa-tion in primary memory has to be continuously attended to for r e c a l l . Isolated examples of the effects of s imi lar i ty , in terms of modality speci f ic interference ef fects , have also been reported (e .g . , Parks, Parkinson, and Kro l l , 1971; Posner, 1969). Furthermore, there have also been attempts to combine both dimensions into a single experiment. For example, Cohen and Granstrom (1969) have investigated visual interference effects with varying degrees of attention demands. However, a detailed evaluation of similar i ty and attention effects in both visual and verbal coding conditions within a single experimental paradigm has not yet been available. The importance of the present research, therefore, l ies in the invest i -gation of attention and similar i ty manipulations of both interpolated 85 modalities in both verbal and visual coding conditions. The advantage of this research approach l ies largely in i ts potential usefulness in generating direct comparisons of attention and similar i ty effects across interference modalities and coding modes. Of special interest in this comparison is the evaluation of the relative importance of attention and simi lar i ty effects in STS losses. If a consistent trend between these two factors can be established, a more detailed conceptualization of verbal and visual STS can be obtained. For example, i f attention or s imi lar i ty effects account for a consistently larger retention loss, a more refined evaluation of the nature of STS processes will be avai lable. Indeed, if the responses in Experiments I and II are converted into percentages of correct r e c a l l , some indication of the relative importance of losses due to attention and similar i ty can be extracted from the data (Figure 7). The following example will i l lustrate the rationale under-lying this evaluation: a) the rehearsal group in Experiment II is the reference point for verbal coding eff iciency after the given retention interval; b) the task of counting backwards in Experiment II was defined in terms of an attention component and a similar i ty component; c) the task of counting forwards In Experiment II was interpreted in terms of the similar i ty component. A comparison of correct recall percentages for these three conditions should then lead to an evaluation of retention losses which, more or less, can be direct ly attributed to attention and simi lar i ty components inherent in the verbal interference in Experiment I I. Hence, the percentage difference between the rehearsal group and the 86 i-3 o w K EH O W « « o u w o g Pi 100 _ EXP. II 90 J 80 EXP. I 70 . 60 . 50 . ko 30 1 IMM 0-DELAY r B W 1 LOW HIGH VERBAL 10 SECOND DELAY LOW. HIGH VISUAL-VISUAL MOTOR FIG. 7„ Percentage values for correct r e c a l l in Ezp 0 I and I I 0 87 counting backwards group constitutes the retention loss due to the com-bined effect of attention and s imi lar i ty , i . e . , 97.56$ - 47.33$ = 50.23$ (Appendix I, Table 5). Furthermore, the difference between the rehearsal group and the counting forwards group should be an estimate of the retention losses due to the similari ty component alone, i . e . , 97.56$ -88.00$ = 9.56$. In other words, retention losses due to the simi lar i ty component by i tse l f constitute a loss of 9.56$, whereas retention losses due to the combined effect constitute a 50.23$ loss. Thus, by subtracting percentage losses due to the similar i ty component from the total losses, an estimate of retention losses due to attention demands inherent in counting backwards should be obtained, i . e . , 50.23$ - 9.56$ = 40.67$. Since attention is considered as a common factor in verbal and visual coding, and since the attention component in counting backwards should be constant across verbal and visual coding conditions, i t would be expected that percentage losses due to counting backwards in Experiment I should be comparable to those attributed to the attention component in counting backwards in Experiment II. Before pursuing this evaluation, i t should be noted that these per-centage losses cannot be considered as absolute values in a comparison. F i r s t , since response levels in verbal and visual coding show a difference for the respective rehearsal groups, i . e . , 97.56$ versus 82.66$, it cannot be expected that a given percentage loss with respect to one rehearsal level is comparable to that obtained in the other. Second, i t must be admitted that a given percentage loss due to, for example, a similar i ty component might contain other than retention losses due to the similar i ty 88 ef fect . Thus, losses due to simi lar i ty and losses due to attention, as used in this context, should be considered as merely describing the most pre-dominant component in a speci f ic comparison. So far then, i t has been shown in Experiment II that the s imi lar i ty component of verbal interference accounts for a 9.56$ loss, and the attention component accounts for 40.67$ of retention losses. A compar-able retention loss due to the attention component in counting backwards should therefore be obtained in the visual coding condition in the difference between the rehearsal and counting backwards group in Experi-ment I, i . e . , 82.66$ - 52.66$ = 30.00$. It can be seen that retention losses due to the attention component inherent in counting backwards are larger in verbal than in visual coding. A similar evaluation of s imi lar i ty and attention components is ava i l -able for visual Interference in Experiment I. The analogous comparison groups are: a) the rehearsal group; b) the visual-motor group, display-ing a s imi lar i ty and attention component; c) the high attention visual group, displaying the effects of the simi lar i ty component by i t s e l f . Hence, the percentage difference between the rehearsal and visual-motor group constitutes the retention loss due to the combined effect of attention and s imi lar i ty , i . e . , 82.66$ - 31.91$ = 51.75$. Furthermore, the difference between the rehearsal group and the high attention visual group should be an estimate of retention losses due to the simi lar i ty component alone, i . e . , 82.66$ - 66.41$ = 16.25$. In other words, reten-tion losses due to the similar i ty effect account for 16.25$ whereas losses due to the combined effect constitutes a 51.75$ retention loss. Subsequently, 89 retention losses due to the attention component inherent in the v isua l -motor task in Experiment I can be estimated to account for a 35.50$ loss, i . e . , 51.75$ - 16.25$ = 35.50$. Again, if attention is considered as a common factor in verbal and visual coding, a comparable percentage loss should be obtained in the visual-motor group in verbal coding. The percentage difference in Experiment II between the rehearsal group and the visual-motor group is 27.56$, i . e . , 97.56$ - 70.00$ = 27.56$. In other words, retention losses due to the attention component inherent in the visual-motor task are larger in visual than verbal coding, yet show a comparable trend to that obtained in the verbal interference evaluation. In addit ion, an evaluation of retention losses of interference conditions, interpreted in terms of s imi lar i ty , can also be undertaken. The speci f ic emphasis here centers around modality speci f ic versus cross modality interference ef fects . For example, the high attention visual group in Experiment I was interpreted in terms of losses due to simi-lar i ty and constituted a 16.15$ retention loss. The identical interpolated condition in verbal coding leads to a 2.00$ retention loss. Conversely, the similar i ty component in Experiment II, i . e . , low attention verbal group, resulted in a 9.56$ retention loss. In Experiment I this inter-polated task leads to a 4.65$ retention loss. Hence, retention losses of a given interpolated act iv i ty interpreted in terms of s imi lar i ty are greatly reduced if different modalities are involved in stimulus coding and in the performance of the interpolated act iv i ty . In e f fect , there is a clear dif ferentiat ion between modality speci f ic and cross-modality effects in both interpolated modalities and their respective coding conditions. 90 Despite the shortcomings of this evaluation, several consistent trends emerge in this numerical comparison. The f i r s t point l ies in the highly comparable retention loss in both experiments due to the interpolated conditions containing both simi lar i ty and attention components. Counting backwards leads to a 50.23$ loss in verbal coding. The visual-motor group displays a similar 51.75$ loss in visual coding. However, in light of the different speci f ic experimental manipulations underlying these resul ts , i t Is unclear what signif icance, if any, this unexpectedly comparable retention loss contains. Except for unfounded speculations about coding and retention loss functions, this comparison will be d i s -carded as a coincidence of no theoretical value. Second, there has been a consistent relationship between the estimates of retention losses due to the attention component in one experiment to that obtained in the other experiment. For example, retention losses due to the attention component of counting backwards were 10 to \\% higher under verbal coding than under visual coding1 conditions. Conversely, retention losses due to the attention component of the visual motor task were about 8% higher in visual coding than in verbal coding. In both comparisons then, the estimates of retention losses due to the attention component were higher when the same modality was involved in processing the stimulus and interpolated act iv i ty . The implications of this con-sistent trend are that, apart from the similar i ty and attention component, there must be some type of interactive effect between the modality of the two tasks which constitutes an addit ional, factor contributing to the total retention losses. In light of the data, the nature of the interaction 91 appears to be such that retention losses, with a constant attention com-ponent, are increased in a same modality experimental s i tuat ion. Though the existence of an interaction between similar i ty and attention appears as a reasonable description of these data, a convincing evaluation of this problem would have to be undertaken in an investigation of speci f ic parametric changes in attention and similar i ty components in both coding conditions. The third finding appears less ambiguous, yet at the same time of greater theoretical importance. This trend describes the relative reten-tion losses of s imi lar i ty and attention components in both experiments. The estimates of retention losses due to simi lar i ty and attention inherent in visual interpolated ac t iv i t i es of Experiment I constitutes 16.25$ and 34.60$ respectively. Conversely, in Experiment II the simi-lar i ty component accounts for 9.67$ and the attention component for 40.67$ of retention losses of the verbal interpolated ac t iv i ty . Hence, in both experiments, retention losses due to the attention component constitute by far the greater aspect of total retention losses than those accounted for in terms of s imi lar i ty . In other words, manipulations of the attention continuum generates a larger retention loss than changes along the s imi lar i ty dimension, at least for the manipulations in this research. The fourth trend in this evaluation is a complimentary aspect of the last part and is concerned with the clear dist inct ion obtained between modality speci f ic and cross modality effects of the two interpolated ac t iv i t i es interpreted In terms of s imi lar i ty alone. Counting forwards as well as the high attention visual group clear ly show comparable effects 92 across same modality versus different modality experimental situations. For each interpolated modality, modality speci f ic effects are always more pronounced than cross-modality effects on the retention of stimulus Information. The last two observations appear of importance in generating a more refined theoretical conceptualization of STS processes. F i r s t , the evidence gained about the relative retention losses due to the attention and similar i ty components can be used to specify the Importance of central processing and speci f ic coding processing components underlying the main-tenance of information in STS. Second, the evidence obtained about the similar i ty and attention component underlying retention losses should provide a clear separation of retention losses in terms of attention diversion and direct Interference in STS. The information gained about the relative importance of similar i ty and attention components In retention losses provides an interesting expansion of previous conceptualizations of STS processes. It has already been stated that the two types of retention losses obtained in STS could in turn be used to describe the processes involved In the main-tenance of information in STS. If the relative importance of the simi-lar i ty and attention components in retention losses can be considered as an indicator of the relative importance of central processing and the speci f ic coding processes in the maintenance of information in STS, the following important addition to the conceptual framework is warranted: maintenance of information In STS re l ies to a greater extent on central processing components than on modality speci f ic coding components. Hence, 93 STS can be conceptualized as consisting largely of the central processing capacity, u t i l i z ing the most appropriate modality speci f ic coding process for a given stimulus condition. The theoretical contribution of this evaluation to the concept of STS therefore appears to be two fo ld . F i r s t , a common central processing capacity is identified as underlying dist inct visual and verbal coding systems in STS. Second, the maintenance of information in both coding modes re l ies largely on the amount of central processing capacity available for verbal or visual coding. Apart from providing a more speci f ic conceptualization of STS pro-cesses, the data also contribute new information regarding the general concept of interference in STS. The issue raised here centers around a c la r i f i ca t ion of the nature of retention losses in terms of attention diversion and direct interference. For example, in previous verbal STM experiments using counting backwards as an interpolated ac t iv i ty , the typical interpretation of retention losses was made in terms of rehearsal prevention (e .g . , Atkinson and S h l f f r i n , 1968). To paraphrase Atkinson and Sh l f f r in , the arithmetic task plays the role of preventing rehearsal and has no direct interference ef fect . That i s , attempts at stimulus coding are terminated when attention is given to counting backwards. The interpretation of retention losses in the present research, however, clearly points out that attention diversion constitutes only one speci f ic aspect of retention losses. An evaluation of verbal interpolated act iv i ty effects will i l lustrate this argument. It was shown that counting backwards generated retention losses which were interpreted in terms of losses due to s imi lar i ty and 94 attention In Experiment II, and losses in terms of attention alone in Experiment I. Thus, the previous interpretation of the task of counting backwards in terms of attention diversion has to be modified. That i s , there is a component of attention diversion as revealed by the percentage losses in Experiments I and II. However, if the same modality is involved in the performance of the interpolated task and in stimulus coding, there is an additional retention loss based on the similar i ty ef fect . Retention losses due to the simi lar i ty component can then be interpreted in terms of a direct interference ef fect . This interpretation is further supported by an evaluation of the task of counting forwards in Experiments I and II. A previous interpre-tation of the results in Experiment II could have been: counting forwards prevents rehearsal through attention diversion, therefore, verbal coding is reduced. The absence of a comparable decrease In visual coding, however, indicates that an interpretation in terms of attention diversion is inappropriate in this case. In other words, verbal recall losses are a function of the similar i ty effect and therefore can be interpreted in terms of direct interference. A similar evaluation of retention losses due to visual interpolated ac t iv i t ies in terms of attention diversion and direct interference further supports the consistency of this interpretation. Thus, two qual i tat ively dist inct types of retention losses in STS can be ident i f ied. On one hand, there are retention losses due to the similar i ty ef fect , as defined in this research, which are modality speci f ic and are interpreted in terms of direct interference between STS processes. On the other hand, there are retention losses due to the attention component, 95 which display comparable effects in verbal and visual coding and are inter-preted in terms of attention diversion. Thus, this research has succeeded in providing a more refined evaluation of retention losses in STS, such that i t becomes possible to interpret the previously used concept of attention diversion (e .g . , Atkinson and Sh i f f r in , 1968) in terms of a combination of an estimate of attention diversion as well as an estimate of direct Interference. The data of this research might have far reaching implications for future empirical and theoretical evaluations of STS. To emphasize the contribution of this research, three speci f ic conclusions have to be considered: f i r s t , the separation of attention and simi lar i ty effects in STS retention losses; second, the interpretation of STS retention losses in terms of attention diversion and direct interference ef fects; and th i rd , the conceptualization of STS processes as a function of a pre-dominant central process component and modality speci f ic coding components, in this case, verbal and visual coding components. However, while these refinements of STS processes have been clearly exposed tn this research, the direct implications of these findings to current STS issues are d i f f i c u l t to assess. On one hand, i t would seem necessary to evaluate a l l previous research in light of the present, more detailed framework of STS processes to substantiate the general appl ic-ab i l i ty of these conclusions. On the other hand, the major interest generated by these data could probably l ie in the new research directions opened up, or at least suggested by these f indings. For instance, the concept of s imi lar i ty , as defined in this research, has been shown as a 96 crucial factor in STS. Within this framework i t would also become of interest to determine if the s imi lar i ty effect pertains merely to modality differences, or whether this concept could be generalized to manipulations within a given modality. For example, could a simi lar i ty effect be obtained in the visual modality by contrasting pictures and random patterns in a factorial combination of both types of material in the stimulus and interpolated act ivi ty? Of equal interest would be a more detailed evaluation with respect to central processing and speci f ic coding processes in STS. For example, i t is s t i l l not clear what speci f ic factors are underlying the recall level of words after an interval of counting backwards. It has been stated that attention diversion and direct interference account for the recall losses. However, this statement does not specify why about 50% of the words are s t i l l available for recall at this point. Depending on the assumptions made in this context, a series of rather unsupported interpretations could be forwarded. For example, assuming the total central processing capacity is used up in the task of counting backwards, then the recall level must be a function of the modality speci f ic coding process in STS. Conversely, i t can be assumed that counting backwards requires the use of only a portion of the central processing capacity, or on the other hand, that attention is constantly shift ing between the Interpolated and stimulus task. If this were the case, the f inal recall level would have to be interpreted with reference to a central processing component. From the above examples, i t can be seen that this research is 97 raising a number of relat ively unexplored aspects of STS processes. In ef fect , apart from the contributions to the present state of knowledge of STS processes, a major importance of these findings probably l ies in the potential usefulness in generating new research which can lead to a more comprehensive conceptualization of STS processes. SUMMARY The purpose of this research centered around an evaluation of STS processes with respect to the speci f ic relationship between verbal and visual coding systems. In order to evaluate verbal and visual coding character ist ics, the respective coding conditions were investigated as a function of a series of Interpolated ac t iv i t i es di f fer ing along two dimensions: a) changes in attention demands; and b) changes in the modality of the interpolated task. It was argued that changes in attention demands, regardless of the modality of the interpolated task, should exert a relat ively constant effect on both coding condtions. Changes in the modality of the interpolated task, in turn, were expected to generate results exploring differences between modality speci f ic and cross modality types of interference ef fects . The data supported these projections. Two separate components of retention losses were obtained. One, the attention component, which Is a function of changes in attention demands, has a comparable effect on verbal and visual coding. Losses due to attention therefore were considered as a common factor in verbal and visual STS. Two, when the processes of stimulus information and Inter-polated act iv i ty were simi lar , larger retention losses were observed than when unrelated processes were Involved in the performance of both tasks. 98 These findings in both modalities were interpreted in terms of the simi-lar i ty ef fect . The similar i ty effect, in turn, was interpreted as evidence for qual i tat ively different coding processes in verbal and visual STS. Furthermore, an evaluation of the relative retention losses attributed to attention and simi lar i ty components strongly suggest that changes of the attention component account for a larger percentage of coding losses than those obtained due to the similar i ty ef fect . The data suggested the following conceptualization of verbal and visual STS. Central processing is the most Important aspect underlying the maintenance of Information in STS and is common to both verbal and visual coding. Verbal STS is therefore conceptualized as consisting of central processing and verbal coding components, and visual STS consists of central processing and visual coding components. It was postulated that the maintenance of information in STS re l ies to a large extent on the ava i lab i l i ty of central processing capacity with the modality speci f ic coding component determining the most appropriate coding mode for a particular stimulus si tuat ion. The data were further Interpreted as providing evidence for two qual i tat ively d is t inct retention loss components. Losses due to simi-lar i ty were Interpreted in terms of direct interference between processing similar types of material for the stimulus and interpolated ac t iv i ty . Losses due to attention were considered as a more refined estimate of the effects of attention diversion. Thus, retention losses previously inter-preted in terms of attention diversion can now be interpreted as a function of an attention diversion component and a direct Interference component. 99 REFERENCES Atkinson, R. C. and Sh i f f r in , R. M. Human memory: A proposed system and i ts control processes. In K. W. Spence and J . T. Spence (Eds.) , The Psychology of Learning and Motivation: Advances In Research  and Theory, Vol . II, Academic Press, New York, 1968, pp. 89-195. Baddeley, A. D. Semantic and acoustic similar i ty in short term memory. Nature, 1964, 204., 11 16-1 117. Baddeley, A. D. Short-term memory for word sequences as a function of acoustic, semantic and formal s imi lar i ty . Quarterly Journal of  Experimental Psychology, 1966, JJ3, 362-365. Baddeley, A. D. and Warrington, E. K. Amnesia and the dist inct ion between long and short-term memory. Journal of Verbal Language and Verbal  Behaviour, 1970, 9_> 176-189. Bernbach, H. A. Replication processes in human memory and learning. In G. H. Bower and J . T. Spence (Eds.) , The Psychology of Learning  and Motivation: Advances in Research and Theory, Vol . I l l , Academic Press, New York, 1969, pp. 201-240. Brown, J . A. Some tests of the .decay theory of immediate memory. Quarterly Journal of Experimental1 Psychology, 1958, 10, 12-21. Buschke, H. Input-output, short-term storage. Journal of Verbal Learn- ing and Verbal Behaviour, 1968, 7_, 900-903. Conrad, R. Acoustic confusion in immediate memory. Br i t ish Journal of  Psychology, 1964, 55, 75-84. Cohen, R. L. and Granstrom, K. Reproduction and recognition in short-term visual memory. Quarterly Journal of Experimental Psychology, 1970, 22, 450-457. 100 Crawford, J . , Hunt, E. and Grahame, P. Inverse forgetting in STM. Journal of Experimental Psychology, 1966, J_2, 3, 415-422. Dale, H. C. A. and Gregory, M. Evidence for semantic coding in short-term memory. Psychonomic Science, 1966, 5_, 75-76. den Heyer, K. and Barrett, B. Selective loss of visual and verbal informa-tion in STM by means of visual and verbal interpolated tasks. Psychonomic Science, 1971, 25, 100-102. Geffen, G . , Bradshaw, J . L., and Wallace, G. I nterhemi spheric effects on reaction time to verbal and non-verbal visual st imul i . Journal  of Experimental Psychology, 1971, 87, 415-422. Geffen, G . , Bradshaw, J . L., and Nettleton, N. C. Hemispheric assymetry: Verbal and spacial encoding of visual st imul i . Journal of  Experimental Psychology, 1972, (in press). Glanzer, M. Storage mechanisms in r e c a l l . In G. H. Bower and J . T. Spence (Eds.) , The Psychology of Learning and Motivation: Advances  in Research and Theory, Vol . V, Academic Press, New York, 1972, pp. 129-193. Glanzer, M. and Cunitz, A. Two storage mechanisms in free r e c a l l . Journal of Verbal Learning and Verbal Behaviour, 1966, 5_, 351-360. Hellyer, S. Frequency of stimulus presentation and short-term decrement in r e c a l l . Journal of Experimental Psychology, 1962, 64, 650. James, W. Principles of Psychology, Holt, New York, 1890. Kintsch, W. Learning, Memory, and Conceptual Processes. John Wiley and Sons, New York, 1970. Kintsch, W. and Buschke, H. Homophones and synonyms in short-term memory. Journal of Experimental Psychology, 1969, 80, 403-407. 101 Klatzky, R. L. Interhemlspheric transfer of test stimulus representation in memory scanning. Psychonomic Science, 1970, 2J_, 201-203. Klatzky, R. L. and Atkinson, R. C. Specialization of the cerebral hemisphere in scanning for information in short-term memory. Perception and Psychophysics, 1971, J_0_, 335-338. Kro l l , N. E. A . , Parks, T . , Parkinson, S. R., Bieber, S. C , and Johnson, A. L. Short-term memory while shadowing: Recall of v isual ly and of aurally presented letters. Journal of Experimental Psychology, 1970, 85, 220-224. Laughery, K. R. Computer simulation of short-term memory: A component decay model. In G. H. Bower and J . T. Spence (Eds.) , The Psychology  of LearnIng and Motivation: Advances in Research and Theory, VoI. I l l , Academic Press, New York, 1969, pp. 136-200. Margrain, S. A. Short-term memory as a function of input modality. Quarterly Journal of Experimental Psychology, 1967, 12, 109-114. Mi l le r , G. A. The magical number, plus or minus two: Some limits in our capacity for processing information. Psychological Review, 1956, 63, 81-97. Milner, B. Neuropsychological evidence for dif fering memory processes. Abstract for the symposium on short-term and long-term memory. Proceedings of the 18th international congress of Psychology, Moscow, 1966, Amsterdam: North Holland Publ . , 1968. Murdock, B. B. , J r . Interpolated recall in short-term memory. Journal of Experimental Psychology, 1963, 66, 525-532. 102 Murdock, B. B. Short-term memory. In G. H. Bower and J . T. Spence (Eds.) , The Psychology of Learning and Motivation: Advances in Research  and Theory, Vol . V, Academic Press, New York, 1972, pp. 67-127. Neisser, U. Cognitive Psychology. Academic Press, New York, 1967. Paivio, A. and Czapo, K. Concrete image and verbal memory codes. JournaI of Experimental Psychology, 1969, 80, 279-285. Parks, T . , Parkinson, S. R, and KrolI , N. E. Visual and auditory short-term memory: Effects of phonemicaily similar auditory shadow material during the retention interval . Journal of Experimental  Psychology, 1971, 87, 274-280. Peterson, L. R. and Peterson, M. J . Short-term retention of individual items. Journal of Experimental Psychology, 1959, 58, 193-198. Posner, M. I. Characteristics of visual and kinesthetic memory codes. Journal of ExperimentaI Psychology, 1967, 67_, 103-107. Posner, M. I. Abstraction and the process of recognition. In G. H. Bower and J . T. Spence (Eds.) , The Psychology of Learning and  Motivation: Advances in Research and Theory, VoI. I l l , Academic Press, New York, 1969, pp. 44-100. Posner, M. I. and Konick, A. F. Short-term retention of visual and kinesthetic information. Organisational Behaviour and Human  Performance, 1966, j_, 71-86. Posner, M. I., Boies, S. J . , Eichman, W. H., and Taylor, R. L. Retention of visual and name codes of single letters. Journal of Experimental  Psychology Monograph, 1969, 79, No. I, Part 2. 103 Potter, M. C. and Williams, M. I. Is there an echo box in pictor ia l memory? Paper presented at the meeting of the Eastern Psycho-logical Association, A p r i l , 1970. Schnore, M. M. and Partington, J . T. Immediate memory for visual patterns: Symmetry and amount of information. Psychonomic Science, 1967, 8, 421-422. Shulman, H. G. Similari ty effects in short-term memory. Psychological Bul le t in . 1971, 75_, 399-415. Sperling, G. The information available in brief visual presentations. Psychology Monographs, I960, _74 (Whole No. 498). Sperling, G. A model for visual memory tasks. Human Factors, 1963, 5_, 19-31. Stonners, R. F. and Muenzinger, G. F. PronouncabiIity in STM. Journal of Experimental Psychology, 1969, 80, 359-363. Ternes, W. and Yu i l l e , J . C. Words and pictures in an STM task. Journal of Experimental Psychology, 1972, 96, 78-86. Ternes, W. and Y u i l l e , J . C. Additive interference processes in STM. Journal of Experimental Psychology, 1973, (in press). Tversky, B. Pictorial and verbal encoding in a short-term memory task. Perception and Psychophysics, 1969, 6, 225-233. Waugh, N. C. and Norman, D. A. Primary Memory. Psychological Review, 1965, 72, 89-104. Waugh, N. C. and Norman, D. A. The measure of Interference in primary memory. JournaI of Verba I Learning and Verbal Behaviour, 1968, 1_, 617-627. 104 Wickelgren, W. A. Acoustic s imi lar i ty and intrusion errors In short-term memory. Journal of Experimental Psychology, 1965, 70, 102-108. Wickelgren, W. A. Short-term recognition memory for single letters and phonemic simi lar i ty of retro-active interference. Quarterly  Journal of Experimental Psychology,1966, 181, 55-63. Wickens, D. D. Encoding categories of words: An empirical approach to meaning. Psychological Review, 1970, 77., 1-15. Wickens, D. D. and Eckler, G. R. Semantic as opposed to acoustic encod-ing in STM. Psychonomic Science, 1968, J_2, 63. 105 A P P E N D I X I 106 TABLE I Recall Scores for Experiments I and II. 0 Delay 10 Sec. Delay |MM REH Verbal Visual Visual-Low High Low High Motor Correct Reca11 - out of 80 63.80 66.13 62.33 42.13 60.80 53.13 25.53 Q_ B d' Scores 2.14 2.49 2.00 .76 2.12 1.35 -.02 — Correct Reca11 out of 30 29.47 29.27 26.40 14.20 29.40 28.67 21.00 Q. X LU TABLE 2 Mean Values for Visual Coding Groups in Experiment III. Presentation Modality Auditory Visual Correct d 1 Correct d' Reca11 Scores Reca11 Scores Button 39.70 .59 28.00 .12 Lever 36.60 .39 27.30 .02 TABLE 3 Mean Reca11 for Verbal Coding Groups In Experiment Presentation Modality Auditory Visual Button 23.70 15.90 Lever 23.30 20.60 TABLE 4 Mean Percentages of Correct RecaI I for Visual and Verbal Stimuli in Experiment III. Presentation of Interpolated Task Aud itory Visual  Stimulus Condition Stimulus Condition Verbal Visual Verbal Visual Button 79.00 49.62 53.00 35.00 Lever 77.66 45.75 68.66 34.12 110 TABLE 5 Mean Percentage of Correct Recall in Experiments I and II. 0 Delay 10 Sec. Delay JMM REH Verbal Visual Visual -Low High Low High Motor EXP I 79.75 82.66 77.91 52.66 76.00 66.41 31.91 EXP II 98.23 97.56 88.00 47.33 98.00 95.56 70.00 I l l APPENDIX I I 112 TABLE I ANOVA for Correct Reproductions In Experiment I. Source of Variance SS df MS F p Treatment 17,626.33 6 2,937.72 26.67 <.0I Exp. Error 10,795.07 98 110.15 Total 28,421.40 104 113 TABLE 2 Newman-Keuls Test for Treatment Effects for Correct Reproductions in Experiment I. Order Treatment in Order of VM HVER HVIS LVIS LVER IMM REH Position 25.53 42.13 53.13 60.80 62.33 63.80 66.13 Truncated Range 2 3 4 5 6 7 Sxq .95 7.56 9.07 9.69 10.58 11.07 11.44 Sxq -99 9.99 11.34 12.15 12.71 13.20 13.52 2 3 4 5 6 7 2 16.60** 27.60** 11.02** 4 35.27** 18.67** 7.65* 36.80** 20.20** 9.18* I .53 6 38.27** 21.67** 10.65* 3.00 I .47 7 40.60** 24.00** 12.98** 5.33 3.80 2.33 114 TABLE 3 ANOVA f o r d' Scores i n Experiment I. Source of Variance Treatment Exp. E r r o r Total SS df 73.45 6 57.82 98 131.27 104 MS F 12.24 20.74 .59 P <.0I 115 TABLE 4 Newman-Keuls Test f o r d' Scores i n Experiment I. Order 1 2 3 4 5 6 7 Treatment i n Order of VM HVER HVIS LVER LVIS IMM REH P o s i t i o n .98 1.76 2.35 3.00 3.12 3.14 3.49 Truncated Range 2 3 4 5 6 7 Sxq .95 .53 .63 .70 .74 .77 .80 Sxq .99 .70 .79 .85 .89 .92 .95 1 2 3 4 5 6 7 1 .78** 1.37** 2.02** 2.14** 2.16** 2.51** 2 - .59* 1.24** 1.36** 1.38** 1.73** 3 - .65* .77* .79* 1.14** 4 - .12 .14 .49 5 - .02 .37 6 - .35 116 TABLE 5 ANOVA for Recall Scores in Experiment II Source of Variance SS df MS F p Treatment 3,056.86 6 509.47 93.65 <.0I Exp. Error 533.34 98 5.44 Total 3,590.20 104 117 TABLE 6 Newman-Keuls Test for Treatment Effect in Experiment II. Order 1 2 3 4 5 6 7 Treatment in Order of HVER VM LVER HVIS REH LVIS IMM Position 15.80 9.00 3.60 1.33 .73 .60 .53 Truncated Range 2 3 4 5 6 7 Sxq .95 1.68 2.01 2.21 2.35 2.46 2.54 Sxq .99 2.22 2.52 2.70 2.82 2.92 3.00 2 6.80** 2 3 4 5 6 7 12.20** 5.40** 4 14.47** 7.67** 2.27** 15.07** 8.27** 2.87** .60 6 15.20** 8.40** 3.00** .73 . 13 7 15.27** 8.47** 3.07** .80 .20 .07 118 TABLE 7 ANOVA for Correct Reproductions of Visuai Patterns in Experiment I I I . Source of Variance SS df MS F p A (Visual vs. Verbal) 1,102.50 I 1,102.50 10.80 <.0I B (Lever vs. Button) 36.10 I 36.10 .35 A x B 14.40 I 14.40 .14 Error 3,674.60 36 102.07 Total 4,827.60 39 TABLE 8 ANOVA for d f Scores of Visual Pattern Reproductions in Experiment III. Source of Variance A (Visual vs. Verbal) B (Lever vs. Button) A x B Error Total SS df MS 1.79 I 1.79 .20 I .20 .03 I .03 7.38 36 .20 9.41 39 120 TABLE 9 ANOVA for Verbal Recall Errors in Experiment III Source of Variance SS df MS F p A (Visual vs. Verbal) 275.63 I 275.63 17.30 <.0I B (Lever vs. Button) 46.23 I 46.23 2.90 A x B 65.02 I 65.02 4.08 <.05 Error 573.50 36 15.93 Total 960.38 39 121 TABLE 10 Analysis of Simple Effects for AB Interaction of Verbal Recall Errors in Experiment III. A ( (Auditory) (Visual ) B (Button) 6.30 14.10 B 2 (Lever) 6.70 9.40 Source SS df MS A for B ( 304.20 I 304.20 19.09 .01 B 2 36.45 I 36.45 2.28 B for A ( .80 I .80 .05 A 2 110.45 I 110.45 6.93 .05 Error Between 573.50 36 15.93 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0101030/manifest

Comment

Related Items