Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effect of changing task complexity on the psychological refractory period Chamberlin, Craig John 1979

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

Item Metadata

Download

Media
831-UBC_1979_A7_5 C53.pdf [ 4.02MB ]
Metadata
JSON: 831-1.0077340.json
JSON-LD: 831-1.0077340-ld.json
RDF/XML (Pretty): 831-1.0077340-rdf.xml
RDF/JSON: 831-1.0077340-rdf.json
Turtle: 831-1.0077340-turtle.txt
N-Triples: 831-1.0077340-rdf-ntriples.txt
Original Record: 831-1.0077340-source.json
Full Text
831-1.0077340-fulltext.txt
Citation
831-1.0077340.ris

Full Text

THE EFFECT OF CHANGING TASK COMPLEXITY ON THE PSYCHOLOGICAL REFRACTORY PERIOD by CRAIG JOHN CHAHBERLIN B.P.E. , University of B r i t i s h Columbia, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF PHYSICAL EDUCATION in THE FACULTY OF GRADUATE STUDIES School of Physical Education and Recreation We accept t h i s thesis as conforming to the required standards THE UNIVERSITY OF BRITISH COLUMBIA A p r i l , 1979 © Craig John Chamberlin, 1979 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of . The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 n*tp msi. it, , i i ABSTRACT By manipulating the r e l a t i v e complexity of each response i n a p s y c h o l o g i c a l r e f r a c t o r y p e r i o d paradigm, i t was hoped t h a t a d i s t i n c t i o n would be made between s e r i a l and p a r a l l e l models o f i n f o r m a t i o n p r o c e s s i n g . 2 exps were performed with 14 r i g h t -handed male Ss/exp., the apparatus used was a RT d e v i c e c o n s i s t i n g p r i m a r i l y of a joy s t i c k s e t i n t o a diamond shaped opening c u t i n a t a b l e top and a hand h e l d response button. In exp 1, the complexity of the second response was manipulated and i n exp 2 t h e complexity of the f i r s t response was manipulated. B e s u l t s g e n e r a l l y supported a s e r i a l p r o c e s s i n g model, such as the one proposed by H e l f o r d (1967). Changing the complexity of the second response had no e f f e c t on the ST to the f i r s t s t i m u l u s , a l s o , changing the complexity of the f i r s t response i n c r e a s e d the delay i n responding t o the second s i g n a l . The e x i s t e n c e of an a d d i t i o n a l delay i n responding t o the second s i g n a l , p o s s i b l y due to a f b or a t t e n t i o n - p a y i n g element, was shown by t h e data. However, p r a c t i c i n g the f i r s t task i n exp 2 d i d not reduce t h i s a d d i t i o n a l d e l ay. I t was concluded t h a t i n c r e a s i n g task complexity may not always have a c o r r e s p o n d i n g i n c r e a s e i n RT and t h a t s u b j e c t i v e response s t r a t e g y c o u l d p o s s i b l y produce e i t h e r s e r i a l o r p a r a l l e l p r o c e s s i n g a r t i f a c t s thereby i n d i c a t i n g t h a t a more f l e x i b l e model r a t h e r than the r e s t r i c t i v e s e r i a l or p a r a l l e l model i s needed. i i i TABLE OF CONTENTS Page ABSTRACT ...... i i LIST OF TABLES .......................................... V ACKNOWLEDGEMENT .-.- v i Chapter 1. INTRODUCTION .. 1 STATEMENT OF THE PROBLEM 5 Subproblems ................................... 5 DEFINITION OF TERMS ..... ..................... 6 DELIMITATIONS \. ^ v . «.r 6 ASSUMPTIONS AND LIMITATIONS 7 HYPOTHESES ..........................-....... .... , 7 2. REVIEW OF SELECTED LITERATURE . . .................... • . . 9 PSYCHOLOGICAL REFRACTORY PERIOD .................. 10 THE NATURE OF THE SECOND RESPONSE IN THE PRP ..... 11 Single Channel Theory ......................... 11 Response C o n f l i c t Theory ...................... 17 Expectancy Theory ............................. 19 THE EFFECT OF THE PRP ON THE FIRST RESPONSE ..... 20 S e r i a l Processing Approach .................... 22 P a r a l l e l Processing Approach .................. 24 THE EFFECT OF TASK COMPLEXITY ON REACTION TIME ... 27 PRACTICE, LEARNING AND FEEDBACK 28 3. METHODS AND PROCEDURES .............................. 30 SUBJECTS 4 . . 3 0 APPAR ATUS ....................... ..... ... ..... 30 Subject*s Response Apparatus .................. 30 Experimenter*s Control Eguipment 31 PROCEDURES 32 Response Tasks ................................ 32 . i v EXPERIMENTAL CONDITIONS 33 Experiment 1 .................................. 33 Experiment 2 ............ ........ ........... ... 34 EXPERIMENTAL CONTROLS ...............,, .... ... .... 36 EXPERIMENTAL DESIGN .............................. 37 ANALYSIS OF DATA 37 Test of Hypothesis 1 .......................... 38 Test of Hypothesis 2 .......................... 38 Test of Hypothesis 3 .......................... 38 4. RESULTS AND DISCUSSION . , . . . . . . . 3 9 RESULTS AND DISCUSSION OF THE PREPLANNED COMPARISONS OF THE HYPOTHESES 40 Experiment 1 40 Experiment 2 .................................. 44 Hypothesis 2 , - , 4 4 Hypothesis 3 44 SOMMARY OF HYPOTHESES - - - - - - , 4 6 ,5. SUMMARY AND CONCLUSIONS ....................... .... •• 50 SUMMARY 50 CONCLUSIONS , ........... 51 SUGGESTIONS FOR FURTHER RESEARCH ................. 52 BlBLIOGHAPHY i . . . . 54 APPENDIX ......... . . . . , , . . . . . . 6 1 V LIST OF TABLES TABLE PAGE 4.1 Bean Reaction Times and Standard Deviations f o r both Experiments ....................40 4.2 One-way Repeated Measures Analysis Of Variance Tables f o r Experiment 1 and Experiments ...........41 4.3 Differences Between the Means of C r i t i c a l Pairs for Experiment 1 .............................42 4.4 Difference Between Predicted and Actual Values for Mean RT2 under PHP Ef f e c t f o r Experiment 1 by S 11 & G t •••••••• • • • • •••••• • • • • • ••••••• • ••••• • • • • • • ^  3 4.5 Comparison of Reduction i n Differences Between Predicted and Actual RT2s from T r i a l 1 to T r i a l 2 under both Experimental Conditions f o r Experiment 2 ..... .................-..-.-.-......45 A.I Experiment 1- Mean RT Data by Subject ................62 A.2 Experiment 2- Mean RT Data by Subject ...63 A.3 Differences between Means for Experiment 1 ...........64 A.4 C r i t e r i o n f o r Hassey-Dixon R r a t i o f o r I d e n t i f i c a t i o n of Ou t l i e r s .........................65 a . 5 Subject*s Response apparatus .........................66 ACKNOWLEDGEMENT The author would l i k e to thank a l l those who assisted him in the production of t h i s thesis, i n pa r t i c u l a r h i s committee, Dr-„ G. S i n c l a i r , Dr. . B. Schutz and Dr. L. Ward, and a l l of the subjects who participated i n t h i s study. A spe c i a l thank-you to the committee chairman. Dr. G. S i n c l a i r , whose support and encouragement throughout the author*s academic career was much appreciated and needed. 1 Chapter 1 INTRODUCTION An area of investigation that has received a great, deal of attention since the early 1940s i s the study of the nature of the underlying processes of s k i l l e d behaviour. The concept of s k i l l , and the study of human performance i n terms of s k i l l e d behaviour, i s of extreme importance because, as Legge and Barber (1976) contend, without perceptual-motor s k i l l s , our intentions can never become actions.,Successful responses to st i m u l i coming from the environment can be measured only: i f , through the vehicle of s k i l l , our selected responses can be put into action. When performing a s k i l l , man can be thought of as functioning i n the capacity of an information processing device (Legge, 1970). In order to function e f f e c t i v e l y , he must s e l e c t and code appropriate incoming s t i m u l i , i n t e r p r e t t h i s information and se l e c t a response according to an i n t e r n a l p r i o r i t y scheme and then successfully put the selected response into e f f e c t . I t i s the central component of interpretation and sel e c t i o n that d i f f e r e n t i a t e s s k i l l e d behaviour from habitual behaviour. In order to understand more c l e a r l y the workings of the central processors which perform the service of inte r p r e t a t i o n and s e l e c t i o n and to investigate the nature of the underlying processes of s k i l l e d behaviour, researchers have often employed the reaction time (RT) paradigm. A c e n t r a l assumption of t h i s paradigm i s that a subject*s RT i s an accurate measure of his central processing time. An advance in the design of the RT experiment has permitted studies to incorporate the 2 psychological refractory period (PRP) phenomenon (Craik, 1947; Vince, 1948; Hick, 1948). The PEP design presents a subject with two rapidly successive s t i m u l i , under conditions that require a separate response to each of the two s t i m u l i . By measuring the subject's RT to each signal, i t has been noted that a delay i n responding to the second signal t y p i c a l l y e x i s t s (selford, 1952; Fraisse, 1955; Davis, 1956). This s i t u a t i o n , then, can be referred to as a double stimulation s i t u a t i o n , as compared to a single stimulation s i t u a t i o n i n which only one stimulus i s presented to the subject. Craik*s (1947) and Vince's (1948) observations and experimentation with the intermittency of s e r i a l responses, the phenomenon which Telford (1931) had named the "psychological refractory period", produced a number of theories purporting to explain the nature of the processes underlying s k i l l e d behaviour. The most prominent of these theories i s the Single Channel Theory (SCT) which was proposed by Helford (1952) and i s now accepted as a "unifying concept" in the f i e l d of information processing and human performance (Legge and Barber, 1976) . In essence, the theory states that the c e n t r a l processors e x i s t i n the form of a single channel with a li m i t e d capacity, such that only one uni t of information can be processed at any given time. One aspect of SCT, which i s of concern to t h i s study, i s the concept of the storage hypothesis, which contends that i f a signal a r r i v e s to be processed while the c e n t r a l processors are occupied by a previous s i g n a l , then the second s i g n a l w i l l be stored pre-attentively u n t i l the f i r s t s i g n a l has been dealt with. In other words, the second s i g n a l must wait the amount of 3 time that i s required to completely process the f i r s t s ignal and that the processing of the f i r s t s i g n a l w i l l be unaffected by the a r r i v a l of the second. An addendum to t h i s tenet holds that the second sig n a l not only has to wait for the f i r s t s i g n a l to be cleared by the c e n t r a l processors, but i t must also wait for any feedback r e s u l t i n g from the i n i t i a t i o n of the f i r s t response to be analyzed as well. This entire relationship i s represented mathematically by the equation: BT2 = BT1 BT2n + FB1 - ISI where BT2n i s the normal BT to the second s i g n a l alone; FB1 i s the feedback to the f i r s t s i g n a l ; and ISI i s the i n t e r v a l between the occurrence of s i g n a l one and that of si g n a l two (Welford, 1968). The concept of feedback as used in t h i s i n v e s t i g a t i o n can be related to Welford*s (1952) idea of "feedforward" or feedback -from .the beginning of a response. He stated, b a s i c a l l y , that two types of feedback operate in the performance of a response: feedback from the beginning of the movement and feedback from the end of the movement. The l a t t e r type of feedback i s involved i n the PBP paradigm only when very long inter--stimulus i n t e r v a l s (ISI) are used. The former type of feedback appears to capture the c e n t r a l processors momentarily at the beginning stages of a movement and seems to operate as a type of checking mechanism to ensure that the movement has actually begun and i s i n the desired d i r e c t i o n . What appears to be involved, then, i s the focussing of attention on the f i r s t response f o r a b r i e f period of time before i t i s shifted to the second s i g n a l . When the term "feedback" i s used i n t h i s paper, i t i s t h i s concept of feedback from the beginning stages of a 4 response that i s being referred to. Using the concept of the storage hypothesis, two predictions were made and were d i r e c t l y tested by t h i s i n v e s t i g a t i o n . The f i r s t prediction was that regardless of the a r r i v a l time of the second sig n a l and regardless of the complexity of the movement i n response to the second s i g n a l , the processing of the f i r s t s i g n a l would be unaffected and should proceed normally. This prediction has been put to a direct test by several researchers (Hay and Gottsdanker, 1968; Smith, 1969; Herman and Kantowitz, 1970; Tolkmitt, 1972; Mcleod, 1977) and t h e i r conclusions indicated that, contrary to SCT and the storage hypothesis, the processing of the f i r s t signal was affected by the a r r i v a l time of the second s i g n a l and by the complexity of the required response, as evidenced by a lengthening of BT1. In l i g h t of these findings, the lack of support for t h i s aspect of the storage hypothesis i s one of the main weaknesses i n SCT. The second prediction to be investigated was the concept that the longer the processing of the f i r s t s ignal takes, then the longer w i l l be the delay i n responding to the second s i g n a l ; f o r example, i f the f i r s t s i g n a l reguires 200 usees to process, and the second s i g n a l occurs after 100 usees, then the delay i n responding to the second s i g n a l should be 100 usees. However, i f the f i r s t s i g n a l requires 400 usees f o r processing, and the second s i g n a l again occurs aft e r 100 msecs, then the delay would be 300 msecs. A thorough review of the l i t e r a t u r e has revealed that t h i s prediction has been tested on two occassions only (Kay and Weiss, 1961; Broadbent and Gregory, 1967) and, i n each case. 5 the test was merely an i n d i r e c t by-product of the main investigation- The r e s u l t s indicated that a larger increase than the expected un i t - f o r - u n i t increase, as predicted by a s t r i c t i n t erpretation of SCT, occurred. Welford*s (1952) suggestion of the c e n t r a l processors being occupied by feedback caused by attention being given to the f i r s t s i g n a l and i t s response has been generally accepted as accounting f o r the extra delay. Deductive reasoning would lead one to believe that, i f the feedback necessary for the performance of the f i r s t task was reduced, then the prediction of s t r i c t SCT- the unit-for-runit increase- should be confirmed. One aspect of learning theory has shown that by providing a subject with practice on a task, the amount of feedback required to perform the task i s , i n f a c t . This i n v e s t i g a t i o n i s primarily concerned with the nature of the second response i n the psychological refractory period paradigm; that i s , the eff e c t of the complexity of the response to the second s i g n a l upon the response time to the f i r s t s i g n a l . Subproblems  The subproblems are: 1. To determine i f an increase i n the BT to the f i r s t s i g n a l , as a r e s u l t of increased task complexity, w i l l cause a corresponding unit-for-unit increase in the BT to the second s i g n a l . 2. To determine i f there i s a learning e f f e c t upon t h i s r e l a t i o n s h i p . reduced. STATEMENT OF THE PROBLEM 6 DEFINITION OF TERMS Psychological Refractory Period (PRP)- The delay i n responding to a second signal when i t i s presented during the processing time of a previous s i g n a l . Reaction Time !_ • (RT1). The elapsed time from the presentation of stimulus l i g h t one (S1) u n t i l the i n i t i a t i o n of a response. Reaction Time 2 (RT2). The elapsed time from the presentation of stimulus l i g h t two (S2) u n t i l the i n i t i a t i o n of a response. Simple Movement _1 (SM 1). A single movement involving the depression of a response button. Simple Movement 2 CSH2). A single movement involving the movement of a joy-st i c k o f f the s t a r t i n g micro;-switch. Complex Movement (CM). A complex movement (re l a t i v e to SM1 and SM2) requiring a rapid movement, involving three d i r e c t i o n changes, of a joy-st i c k apparatus. Inter-Stimulus-Iaterval (ISI) - The amount of time elapsing between the presentation of S1 and the presentation of S2. DELIMITATIONS 1. The study i s delimited to an ISI of 150 msecs. 2. The study i s delimited to three discrete response tasks (SMI, SM2 and CM) at two l e v e l s of task complexity- a simple l e v e l (SM1 and SM2) and a more complex l e v e l (CM). 3. The study i s delimited to right-handed male subjects. 7 ASSUMPTIONS AND LIMITATIONS The following assumptions are made: 1. A more complex response program requires more processing time and, therefore, a longer RT than a simpler response program. 2. By practicing a response task, the amount of feedback . required to perform that task i s reduced. The i n v e s t i g a t i o n i s limited by: 1. The accuracy of the timing equipment. 2. The sample siz e of 14 subjects per experiment. HYPOTHESES The hypotheses are: 1. An increase in the complexity of the second response, and the corresponding increase i n RT to t h i s s i g n a l , results in an increase i n the RT to the f i r s t s i g n a l . Furthermore, the RT to the f i r s t s i g n a l i n both double stimulation s i t u a t i o n s w i l l be longer when compared to the RT to t h i s s i g n a l i n a si n g l e stimulation s i t u a t i o n . This hypothesis i s contrary to the prediction of the storage hypothesis of SCT and i s based on a review of l i t e r a t u r e which has shown a considerable amount of support f o r t h i s opposing view. I f t h i s hypothesis i s supported by the data collected, a more tenable alt e r n a t i v e to SCT might be found through the investigation of a p a r a l l e l processing explanation, that i s , either a type: of time-sharing system or a rapid alternation of attention scheme. 2. The increase in RT1 caused by increasing the complexity 8 of the f i r s t response w i l l cause a corresponding . increase i n ST2, but greater than the u n i t - f o r - u n i t increase predicted by SCT. As postulated by Helford (1952), following the processing of the f i r s t s i g n a l , the c e n t r a l processors are occupied by feedback to the f i r s t s ignal. This w i l l r e s u l t i n an extra delay in the processing of the second s i g n a l beyond that which i s caused by the processing of the f i r s t s i g n a l . By increasing the complexity of the f i r s t response, we are not only increasing the processing time involved, but we are also increasing the amount of feedback required f o r the i n i t i a t i o n of that response. 3. The effect of practice on the f i r s t task results i n a decrease i n BT2, with a larger decrease being noted f o r experimental condition 3 than f o r experimental condition 4. ,• By p r a c t i c i n g the f i r s t task, the amount of feedback necessary for the i n i t i a t i o n of that task w i l l be reduced, therefore, from a SCT point of view, i f less feedback i s being processed the t o t a l delay i n responding to the second s i g n a l w i l l be reduced. With a more complex f i r s t task, a greater amount of feedback i s involved i n i t i a l l y ; therefore, practice has a greater e f f e c t i n t h i s s i t u a t i o n than with a simple f i r s t task. , In order to test the predictions of the storage hypothesis, as outlined above, and the l i s t e d hypotheses, t h i s study was designed in the form of a two experiment sequence: Experiment 1, which was concerned with the main problem and was a t e s t of hypothesis 1, and Experiment 2, which was to explore the two sub-problem areas while testing hypotheses 2 and 3. 9 Chapter 2 REVIEW OF SELECTED LITERATURE Since Telford's (1931) observation of a delay i n responding to the second of two rapidly successive signals, a great deal of experimentation u t i l i z i n g what he, and l a t e r Craik, had termed the "psychological refractory period" has been performed. Generally speaking, t h i s research has been of two types. In the two decades following Welford's proposal of s i n g l e channel theory (SCT), that i s , the *50s and «60s, the work performed dealt mainly with the nature of the second response i n the PEP paradigm.. From t h i s , a number of theories evolved to explain t h i s phenomenon with the most widely accepted p o s i t i o n being Welford's o r i g i n a l formulation; the s i n g l e channel theory. In the l a t e '60s, a r e a l i z a t i o n began to occur that by concentrating on the nature of the second response, only part of the problem was being considered. Researchers began to concentrate on the nature of . the f i r s t response, and t h i s emphasis has continued through to the present day. Out of t h i s work came a new set of theories, a c t u a l l y variations of SCT, that can be termed time-sharing or p a r a l l e l processing approaches to information processing. I t i s the intention of th i s review of l i t e r a t u r e to f i r s t give a b r i e f introduction to the PRP phenomenon, and then to cover the two areas of investigation as outlined above and the theories that have developed, with emphasis on SCT. Because of the volume of work done i n the area, i t would be beyond the scope of t h i s review to cover i t a l l in d e t a i l ; however, the most s i g n i f i c a n t studies 10 have been included. PSYCHOLOGICAL REFRACTORY PERIOD The e a r l i e s t work with the PRP u t i l i z e d tracking tasks to demonstrate i t s ef f e c t (Craik, 1947,1948; Vince, 1948), however, the continuous nature of t h i s response task and the anticipatory behaviour involved resulted i n a change over to the use of a more discrete type of task. In most cases, the PRP paradigm has involved a pair of simple reactions, usually requiring the depression of response keys i n response to an auditory or v i s u a l s i g n a l (Hick, 1948; Fraisse, 1957; Davis, 1956,1957,1959; Kay and Weiss, 1961); however, variati o n s i n the response task, involving more complex movements, have been introduced (Gottsdanker and Way, 1966; Way and Gottsdanker, 1968; Williams, 1971,1973,1975). In some experiments, both responses were made with the same hand (Fraisse, 1957; Kay and Weiss, 1961). Fraisse demonstrated that the motor interference caused by these type of responses made only miminal contributions to the delays noted. L i t t l e work has been done using a choice reaction situation* Although a few investigators (Elithorn and Lawrence, 1955; M a r i l l , 1957; E l i t h o r n , 1961) have attempted to u t i l i z e a choice reaction time s i t u a t i o n , the complexity of the interpretation of the r e s u l t s has proven d i f f i c u l t . The general observation of a delay i n responding to the second of the two signals has been confirmed i n many of the studies and the most d i r e c t evidence that t h i s delay i s caused by a l i m i t a t i o n i n the central processing capacity has been presented by Davis (1957). In h i s experiment, the subject*s task 11 was to respond with simple key press movements i n response to two signals, one of which was auditory and the other being v i s u a l . The r e s u l t s revealed that the t y p i c a l delays found i n the PEP paradigm exi s t even in t h i s bimodal s i t u a t i o n . THE NATURE OF THE SECOND RESPONSE IN THE PRP The work completed i n t h i s area has primarily been concerned with the explanation of the delay i n responding to the second s i g n a l . Several theories have resulted from t h i s work, each of which w i l l be considered separately. Single Channel Theory Smith {1967) refers to the theory being considered here as a subclass of SCT that she terms "response selection delay". However, f o r the purpose of t h i s paper, SCT and response s e l e c t i o n delay theory w i l l be considered synonomous. One of the main attractions of SCT, and perhaps the reason i t i s s t i l l accepted at present, i s that i t appears to account for data from diverse areas of enquiry. The four main areas are choice amongst a l i m i t e d set of al t e r n a t i v e s , the control of movements, responding to successively presented stimuli (PRP) and responding to simultaneous inputs (Legge and Barber, 1976). I t was work done using the PRP paradigm that l e d d i r e c t l y to the theory's formulation. Welford*s postulation of the SCT i n 1952 was based on work., done by several investigators, the p r i n c i p a l ones being Te l f o r d , Craik, Vince and Hick. Telford (1931) was the f i r s t researcher to report the delay to a second s i g n a l . He had shown that subjects, performing a simple key pressing action in response to an auditory s i g n a l occurring at 12 unpredictable i n t e r v a l s ranging from .5 to 4 sees, responded more slowly to the second s i g n a l at a .5 sec i n t e r v a l than one of 1 sec or more- This, he explained, was due to a "refractory period", much l i k e the refractory period of physiological functions i n the muscles and nerves. Craik (1947,1948), using a tracking task, noted that subjects performed corrections at discrete i n t e r v a l s and not continuously as was anticipated. I t i s t h i s observation of the intermittency i n human sensory-motor a c t i v i t y that i s considered to be one of the most i n f l u e n t i a l ideas i n psychology of t h i s century (Bertelson, 1966). To explain t h i s intermittency, Craik u t i l i z e d Telford»s notion of r e f r a c t o r i n e s s , which was unfortunate in that the analogy between physiological refractoriness and psychological refractoriness i s a loose one at best. But, more importantly, Craik expressed the notion that the intermittency originated i n the c e n t r a l mechanisms. In his own words, Craik believed that once a unitary "computing process" had started, "new sensory impulses entering the brain while t h i s c e n t r a l computing process i s going on would either disturb i t or be hindered from disturbing i t by some switching system". Continuing Craik's work, Vince (1948), using a step-tracking technique i n which the course being traced jumped from one l e v e l to another at discrete i n t e r v a l s , demonstrated a delay i n responding to the second step i f two step changes occurred at an i n t e r v a l of less than .5 sec. Showing s i m i l a r r e s u l t s , Hick (1948) used discrete responses by d i f f e r e n t hands to successive signals. This also demonstrated f a i r l y conclusively that the delay was not being caused by motor interference, as may be the 13 case i n a tracking experiment. As a r e s u l t of these experiments, Welford, i n 1952, proposed h i s theory which was further refined i n a paper published i n 1967, i n which he stated that "the data suggested a chain of c e n t r a l mechanisms containing three l i n k s - the f i r s t receives and stores data from signals u n t i l the second l i n k i s ready to deal with them; the second i s a decision mechanism concerned with the i d e n t i f i c a t i o n of signals, the choice of action i n r e l a t i o n to what i s perceived, and the issuing of •orders* to the t h i r d link which i s responsible f o r the detailed phasing of responding movements. I t i s the middle link which seems to form the single channel that gives r i s e to the delays i n responding to S2" (Welford, 1967). These delays are the r e s u l t of two signals being unable to co-exist within the middle l i n k of the c e n t r a l processors, such that a s i g n a l a r r i v i n g during the processing of a previous s i g n a l must be held i n store u n t i l t h i s mechanism has been cleared. According to these formulations then: RT2 = ET1 > BT02 - ISI (1) Where RT02 =central organizing time for S2. Normally t h i s w i l l be the same as RT f o r a single response alone, or, in other words, a normal RT2. Further i n v e s t i g a t i o n revealed that equation (1) could not accurately describe the e x i s t i n g data, s p e c i f i c a l l y the occurrance of the PRP even when ISI > fill. In order to account for t h i s , Welford postulated a revised formula, which stated that: RT2 = RT1 • T f b l * RT02 - ISI (2) 14 Hhere Tfb1 -the time f o r analysis of data fedback from the beginning of the f i r s t response. The assumption here was that the c e n t r a l decision mechanism i s momentarily occupied with feedback from the beginning of the f i r s t response, t h i s feedback being from kinaesthetic and other types of stimulation^ a type of checking mechanism . to assure that the f i r s t response has, indeed, begun, s e l f o r d stated that t h i s feedback i s necessary to c l e a r the decision mechanism. Several predictions that can be made from SCT, which have been the subject of varying degrees of experimentation since Welford's o r i g i n a l theorizing and are of the concern of t h i s experiment, are: 1) The delay in BT2 should be a d i r e c t function of the time reguired to select the f i r s t response. Very l i t t l e work has been done concerning t h i s prediction, which i s a rather large oversight considering that t h i s guestion i s an e s s e n t i a l one when distinguishing between the d i f f e r e n t theories formulated to explain the PBP phenomenon(Smith, 1967). Only two studies , that r e l a t e to t h i s problem, were located. Kay and Heiss (1961) attempted to increase the response selection time to S1 by manipulating the time uncertainty of S1. Their r e s u l t s showed generally that HT2 increased as the time uncertainty to S1 increased, thus lending support to SCT. However, t h e i r method of increasing ST 1 was rather i n d i r e c t , making any conclusions tentative at best. & more d i r e c t test was performed by Broadbent and Gregory (1967). In t h i s experiment, the length of BT1 was manipulated by the use of compatible or incompatible displays. Their r e s u l t s have demonstrated that a longer BT1 w i l l r e s u l t i n a longer BT2. D i f f i c u l t y with t h i s 15 experiment arises through eguipment l i m i t a t i o n s . In order to increase BT1 with an incompatible display, i t was also necessary to make the second response incompatible. Although Broadbent and Gregory have shown that the BT to S2 was increased more than the BT TO S1 i n t h i s s i t u a t i o n , i t could possibly be that the complexity of the s i t u a t i o n magnifies the BT increase to S2 such that the increase i n BT2 i s not due to a BT1 increase. A more precise study would be one in which BT1 was manipulated independently of BT2. 2) I f channel l i m i t a t i o n i s due to the response s e l e c t i o n process, then selection of the second response should be occuring during the performance of the f i r s t response. This position questions the existence of a feedback element i n the delay to the second response. The conclusions of work completed i n t h i s area are contradictory as some investigators (Telford, 1931; Vince, 1948) have found re s u l t s s i m i l a r to those of Welford (1952) and others have produced results that demonstrated no delays in the second response, i f the second sign a l occurs at an i n t e r v a l greater than the time taken for BT1. One author i n two separate experiments found, i n one case, no delays as mentioned above (Davis, 1956) and, i n the other case, r e s u l t s supporting Welford*s research (Davis, 1957). The most acceptable position i s that there i s some ad d i t i o n a l delay, but not as great as that found by Welford. Welford (1959) proposed that the feedback signals were grouped with S2, thus reducing the amount of delay that would normally be found. As Bertelson <1966) states, perhaps the only conclusion possible i s that occupation of the central channel by feedback st i m u l i i s 16 not unavoidable. Although conditions for such occupation are not clear, i t i s most l i k e l y a function of the degree of accuracy reguired and the l e v e l of practice attained. The l e v e l of practice appears to have a strong e f f e c t i n that Davis»s {1956) findings of no delay sere achieved with well practiced subjects. Another fa c t o r within the feedback guestion that has been only i n d i r e c t l y investigated i s the position that the nature of refractoriness as linked to feedback s t i m u l i i s d i f f e r e n t from the refractoriness associated with a response i n that the occupation of the channel does not involve a subsguent reaction. A group of experiments that were designed to t e s t the prediction that i f r e f r a c t o r i n e s s i s due to response s e l e c t i o n , then i f no response selection occuired no refractoriness should be noted, r e f l e c t on t h i s problem. In these experiments, no response was required i n reaction to the f i r s t s i g n a l . One group of researchers demonstrated that delays s t i l l occur i n t h i s s i t u a t i o n (Fraisse, 1957; Davis, 1959; Kay and Weiss, 1961; Nickerson, 1965) and they concluded that attention played a r o l e in the r e f r a c t o r i n e s s , that i s , the necessity of momentarily "paying attention" to a s i g n a l would occupy the channel. I t i s believed that since t h i s s i t u a t i o n ^ a signal occupying the channel with no subsequent reaction-? i s s i m i l a r to that ascribed to the feedback occupation, then the delays Welford i d e n t i f i e s as caused by feedback may be due to a delay caused by a required s h i f t i n attention from the f i r s t s i g n a l to the second (Smith, 1967). However, i t has also been documented that i f no response to the f i r s t s i g n a l i s required, then no delays are noted (Borger, 1963; Bubinstein, 1964).,It appears that both response 17 strategies and attention play a v i t a l role i n these types of experiment. I t i s possible that the locus of the single channel may l i e i n the attention system (Smith, 1967) and that r e s u l t s shown by Davis (1956) of no delay i n practiced subjects may be due to a strengthening of the S-H bonds between the signal and the response and a lessening of the amount of attention required. Response C o n f l i c t Theory This theory was f i r s t elaborated by Berlyne (1957,1960) and more s p e c i f i c a l l y applied to the PEP e f f e c t by Reynolds (1964,1966). Several investigators (Smith,1967; Bernstein, Blake and Hughes, 1968) include t h i s theory as a sub-class of SCT, whereas others (Reynolds, 1964; Herman and Kantowitz, 1970) have argued against t h i s l a b e l l i n g . Due to a desire to treat SCT as a d i s t i n c t theory from response c o n f l i c t theory (RCT) , RCT w i l l be considered i n t h i s review as separate from SCT. Ba s i c a l l y , RCT states that delays occur i n the selection of a response for execution from a number of competing response tendencies (Herman and Kantowitz, 1970). The greater the number of response tendencies, then the greater w i l l be the RT., In a double stimulation s i t u a t i o n as i s found in the PRP paradigm, the delay noted in the second response i s a function of the amount of c o n f l i c t e x i s t i n g between the two responses. Predictions from t h i s theory are as follows: 1) I f the delay i s a r e s u l t of c o n f l i c t i n g response tendencies, then i f no f i r s t response i s required, and therefore, no response tendency aroused, then no delay should be noted in responding to the second s i g n a l . 18 This prediction was discussed under the SCT heading and, as noted, the findings are contradictory. The majority of the experimentation found delays even with no f i r s t response, although a few dissenting voices were heard., A lack of an operational d e f i n i t i o n of what constitutes a response tendency makes interpretation of the r e s u l t s d i f f i c u l t . ^ 2) The greater the s i m i l a r i t y between the two responses, then the greater w i l l be the delays in RT2. This guestion has not been s u f f i c i e n t l y investigated to make any conclusions. The only study that has related to t h i s prediction i s one done by Greenwald and Shulman (1973). In t h i s experiment^ the two tasks being used i n the PEP paradigm di f f e r e d i n several ways; i . e . , i n modality of stimulus presentation, mode of response and the implication of a l i n g u i s t i c j as compared to motor, coding system. The responses U t i l i z e d were those of a movement of a lever i n response to a l i g h t and the r e p e t i t i o n of a l e t t e r that was a u d i t o r i l y presented. Their results showed no delays i n the second response, which would seem to i n d i c a t e support for the response c o n f l i c t theory. However, as mentioned, more work i s needed before any conclusive statements can be make. 3) Strengthening of the f i r s t response tendency through practice should increase the RT to the second s i g n a l . Reynolds has suggested that the r e l a t i v e degree to which a RT i s affected i s a function of the strength or prepotencies of the response tendencies. By practicing R1, the r e l a t i v e strength of this response should be increased, and i n ; the double stimulation s i t u a t i o n , should r e s u l t in R1 being prepotent and 19 therefore RT2 should be increased. To the writer's knowlege, thi s guestion has not received any systematic inves t i g a t i o n although Davis" (1956) findings of no delays i n a well practiced subject could be considered in opposition to t h i s hypothesis. 4) Increasing the complexity of the f i r s t response should not affect the second response, i n that the c o n f l i c t between response tendencies has not been changed. Again, no t e s t of t h i s guestion u t i l i z i n g response c o n f l i c t theory has been performed. I t would seem, though, that Broadbent and Gregory»s (1967) work would refute t h i s prediction, i n that increasing BT1 caused a corresponding increase i n BT2. However, the problem with, t h i s study as mentioned previously applies to t h i s s i t u a t i o n as well. Expectancy Theory A theory that seems to have received more attention than i t merits i s the expectancy theory, which has been formulated by several d i f f e r e n t authors (Poul ton, 1950; Elit h o r n and Lawrence, 1955; Adams, 1962).For the mainpart, t h i s theory has r e l i e d on evidence c o l l e c t e d outside of the PBP s i t u a t i o n . The works ci t e d tend to be of the type that has shown BT increases with a varied foreperiod and with the shortest BT being noted at a mean value of the foreperiod range (fiowrer, 1940; K a r l i n , 1959; Drazin, 1961). In applying these findings to the PBP s i t u a t i o n , expectancy theorists have argued that S1 simply acts as a warning s i g n a l for the occurrance of S2. Thus, the delays noted for BT2 are a function of the ISI c h a r a c t e r i s t i c s , and not of the nature of the f i r s t response. Very l i t t l e support for t h i s theory has been found i n experimental situations u t i l i z i n g the 20 PEP paradigm. Several studies of t h i s type have been completed (Kay and Weiss, 1.961; Adams and Chambers, 1962; Borger, 1963; Creamer, 1963; Davis, 1965; Bertelson, 1966; Reynolds, 1966) and the i r r e s u l t s are more readily explained by a response s e l e c t i o n theory than by the expectancy theory. Expectancy may play a r o l e in the o v e r a l l delay noted, but t h i s r o l e seems to be miminal and t h e o r e t i c a l approaches u t i l i z i n g the expectancy rationale seem to be untenable. When the evidence concerning the nature of the delay i n the second response i s looked at the most tenable explanation seems to be that of the SCT as proposed by Welford. As Smith (1967) states, "the majority of the experiments strongly suggest the presence of some li m i t e d capacity single channel i n the system, most l i k e l y at the response selection or decision stage". As mentioned at the beginning of t h i s discussion, however, the majority of the experiments also ignored the e f f e c t of the PRP on the response to S1. When t h i s evidence i s considered, a di f f e r e n t picture begins to emerge, as w i l l be discussed i n the next section. THE EFFECT OF THE PRP OH THE FIRST RESPONSE. Since the early 1960s* a number of experiments u t i l i z i n g the PRP phenomenon have been performed whose r e s u l t s cannot apparently be explained by a s e r i a l processing view. These data led a number of workers to speculate that man may have the capacity to process information i n a p a r a l l e l manner. Before considering the p a r a l l e l processing approach, a b r i e f review of the work that has led to t h i s approach w i l l be presented. Although i t was i n i t i a l l y reported as an occassional 21 occurrence by e a r l i e r researchers (Vince, 1948; Ellson and H i l l , 1948), the f i r s t comprehensive work directed toward the question of the influence of the PEP on the f i r s t of successive si g n a l s was conducted by Gottsdanker, Broadbent and Van Sant (1963). They found that the mean BT f o r a sing l e choice s i t u a t i o n / that i s , f o r a s i g n a l occurring alone, was r e l i a b l y shorter than that for the f i r s t choice i n a double choice condition. In two l a t e r studies (Gottsdanker and Hay, 1966; Hay and Gottsdanker, 1968) i t was shown that by increasing response antagonism, BT1 was s i g n i f i c a n t l y increased, which led the two authors to conclude that the f i r s t response may be affected by the second s i g n a l , which was contrary to the storage hypothesis of the SCT. Herman and Kantowitz (1970) reviewed the r e s u l t s of the above experiments as well as several other related studies (Halliday e t . a l . , 1960; Sanders, 1964; Beynolds, 1966; Bertelson, 1967; Nickerson, 1967; Bernstein e t . a l . , 1968; Herman and, HcCauley, 1969) and found considerable evidence of an increase i n BT1 under conditions of psychological r e f r a c t o r i n e s s as compared to a single choice s i t u a t i o n . It has also been noted that the r e l a t i v e complexity of the second response may have an e f f e c t on the processing of the f i r s t s i g n a l . Several studies were directed toward t h i s problem (Hay and Gottsdanker, 1968; Karlin and Kestenbaum, 1968; Smith, 1969) and, using a variety of methods, a l l produced s i m i l a r r e s u l t s . Hay and Gottsdanker manipulated response antagonism of the second response towards the f i r s t , K arlin and Kestenbaum used event uncertainty of the second response and Smith increased the number of alternatives of the second response. 22 Each of these studies demonstrated that the r e l a t i v e complexity of the response required by the second s i g n a l had an a f f e c t on the processing of the f i r s t s i g n a l ; that i s , an increase i n processing demands of the second s i g n a l increased BT1. These two B1 ef f e c t s of the PBP, the increase in BT1 from a single to a double stimulation s i t u a t i o n and the increase i n BT1 associated with an increase in B2 complexity, have posed serious problems for s t r i c t SCT. The theory, in i t s s t r i c t i n t e r p r e t a t i o n , predicts no e f f e c t on B1. According to the storage hypothesis, the a r r i v a l of the second signal should be prevented from disrupting the processing of the f i r s t s i g n a l by a switching mechanism located preattentively. Therefore, the processing of the f i r s t s i g n a l should continue normally and BT1 should be the same, whether i n a single or double stimulation s i t u a t i o n . Conseguently, with the apparent f a i l u r e of t h i s aspect of SCT to account f o r the r e s u l t s of the studies reviewed above, alte r n a t i v e approaches which f a l l b a s i c a l l y i n t o two classes, SCT with modifications ( s e r i a l processing approach) and a departure from SCT e n t i r e l y ( p a r a l l e l processing approach) , have been presented by several investigators. S e r i a l Processing Approach Smith (1969) and Tolkmitt (1972) have both attempted to maintain the notion of a s e r i a l processor i n l i g h t of the challenging evidence. Smith suggested that the storage operation of S2 would take some capacity from the pool that i s involved i n the processing of SI. By postulating that the storage of a more complex s i g n a l would require more capacity than a simpler s i g n a l , she f e l t that both of the e f f e c t s on S1 noted in the 2 3 l i t e r a t u r e could be predicted. Tolkmitt proposes a modification of s e r i a l processing that incorporates three assumptions- that of single channel, additive stages and a perceptual interrupt system. The f i r s t two stages, the s i n g l e channel- and additive stages, assume that the processing of information occurs along a single channel which .is made up of successive components. Processing time within each i s additive and the speed and accuracy of processing within one stage i s dependent upon the processing of the p r i o r stages. The neural interrupt mechanism i s a type of safeguard system which can interrupt the processing of the s i g n a l at any stage.. More s p e c i f i c a l l y f o r the PBP, i f the neural a c t i v i t y at the receptor l e v e l caused by S2 exceeds a c r i t i c a l l e v e l , which has been predetermined and depends on factors such as set, expectancy and i n s t r u c t i o n s , the system w i l l be activated and w i l l i n t errupt the processing of S1. Following t h i s i nterruption the o r i g i n a l processing of S1 w i l l proceed. The length of the interruption i s said to depend upon the nature of the e x c i t i n g stimulation as well as the complexity of the required current processing. This modification of l e l f o r d ' s o r i g i n a l SCT appears to s u p e r f i c i a l l y explain the data quite well. Mcleod*s (1977) objection that t h i s model does not predict what i s happening to S2 during the processing of S1 seems somewhat premature. The contention i s held that when these data are plotted i n a novel form of ISI vs inter-response i n t e r v a l (IBI) an indication of the processing time f o r each of the two responses i s produced (Tolkmitt, 1972; Kahneman, 1973) and from t h i s analysis, i t appears that some processing of S2, during the processing of S1, i s occurring. 24 Mcleod expresses an opinion that the model does not account for t h i s occurrence. However, he seems to have overlooked the p o s s i b i l i t y that some processing of S2 may be occurring during the interruption time- such as preliminary i d e n t i f i c a t i o n and coding of the stimulus. Further work i s required before any conclusive statements can be made concerning t h i s p a r t i c u l a r model. P a r a l l e l Processing Approach The p a r a l l e l processing i n t e r p r e t a t i o n of information processing has been receiving a great deal of support recently as evidenced by the work of several investigators who have proposed models of t h i s nature (Herman and Kantowitz, 1970; Kahneman, 1973; Keele, 1973; Mcleod, 1977).,; As these models possess basic s i m i l a r i t i e s , Mcleod's elaboration of the basic premise w i l l be presented here, primarily because of i t s recency. Also discussed i n t h i s section w i l l be a more varied inte r p r e t a t i o n of the p a r a l l e l processing approach as expressed by Legge and Barber (1976). On the basis of his review of the l i t e r a t u r e done, Mcleod came to the following conclusions: (1) The decrease i n IRI with decreasing ISI shows that the longer S2 i s in the system with S1, the less processing S2 requires a f t e r the production of Hi. This indicates that the system i s a p a r a l l e l processor. (2) The increase i n RT1 accompanying the a r r i v a l of S2 implies that the p a r a l l e l processing system has a fixed and li m i t e d capacity. Although processing of S2 s t a r t s before R1 t h i s i s only achieved at the cost of decreasing the proportion 25 of capacity devoted to S1. (3) the dependence of RT1 on the d i f f i c u l t y of S2 i n d i c a t e s that the proportion of the available capacity allocated to S2, when i t a r r i v e s , i s determined by i t s d i f f i c u l t y . Mcleod's proposition, then, i s that information i s processed i n a p a r a l l e l manner, that the: t o t a l amount of capacity a v a i l a b l e f o r processing i s of a fixed nature and that the a b i l i t y f o r variable capacity a l l o c a t i o n e x i s t s . . This capacity a l l o c a t i o n i s made on the basis of task d i f f i c u l t y . For example, i t would be possible to give 100% capacity a l l o c a t i o n to S1'..until S1 has been completely processed. In t h i s case, processing would occur s e r i a l l y , and RT values would be produced that resemble s t r i c t SCT predictions. A c r i t i c i s m of t h i s model, which Mcleod himself acknowleges, i s that by judicious post hoc placing of the trade-o f f in capacity a l l o c a t i o n between two tasks almost any r e s u l t could be predicted. This model, then, i s more descriptive than p r e d i c t i v e , and precise, quantitative experimental measures, u t i l i z i n g his model, are d i f f i c u l t to obtain when applied to the PRP. I t i s most l i k e l y that experimentation using a d i f f e r e n t paradigm w i l l be necessary to test the effectiveness of t h i s p articular model. An a l t e r n a t i v e approach that i s not a s t r i c t p a r a l l e l processing view, but rather a more r a d i c a l modification of SCT, i s the multi-channel processor model proposed by Legge and Barber (1976) which i s based on the work of Kerr (1973), Sh a l l i c e (1972) and A l l p o r t , e t . a l . (1972). The basic idea expressed here i s that man u t i l i z e s three l e v e l s of channels 2 6 within the central nervous system capable of processing information. The bottom, or lowest, l e v e l channels consist of wired-in, genetically determined pathways u t i l i z e d to maintain vegetative functions. at the top l e v e l i s a super-ordinate channel characterized by unlimited f l e x i b i l i t y and adaptiveness. It i s within t h i s channel that problem solving occurs. In between these two l e v e l s are s p e c i a l purpose, mid-level channels that are affected by the l e v e l of learning. These channels are lim i t e d in t h e i r s u b j e c t i v i t y ; that i s , they are r e s t r i c t e d to the class of events f o r which they were established through the learning process. The advantage to these p a r t i c u l a r channels i s thei r extreme e f f i c i e n c y i n handling problems within t h e i r competence. In applying t h i s p a r t i c u l a r model to the data gathered i n PBP studies, Legge and Barber have proposed that each of the channels operate i n the manner of single channel processors, as outlined by Welford, and that a l l l e v e l s of channels are limited quantitatively, while the mid and lower l e v e l channels are l i m i t e d q u a l i t a t i v e l y as well. Under these terms, i t would be necessary f o r several passes to be made through the processors, f o r a l l but the simplest of processing problems, as each seperate, mid-level channel would not be capable of handling an entire problem. This task fragmentation would r e s u l t i n the e f f e c t s , such as the eff e c t of S2 on BTl, which were discussed e a r l i e r . In l i g h t of the r e l a t i v e recency of t h i s model, very l i t t l e work has been done to test the v a l i d i t y of i t s position. Some support has been found to qual i f y the statement that i f two tasks being u t i l i z e d i n the PBP paradigm vary s u f f i c i e n t l y , and 27 the l e v e l of performance i s high enough that completely diff e r e n t mid-level channels are u t i l i z e d , then there should be no delay i n responding to either of the tasks- Greenwald and Shulman (1973), as mentioned previously, used two such tasks and succeeded i n eliminating the PRP e f f e c t . I t would be necessary, however, f o r more work to be performed before any d e f i n i t i v e statements concerning the r e l a t i v e merits of t h i s p a r t i c u l a r model can be made. From t h i s review of l i t e r a t u r e , i t has become apparent that the atmosphere surrounding the PRP i s extremely clouded with a substantial number of models having been proposed to explain t h i s e f f e c t . The experiment that was performed here was an attempt to examine the PRP phenomenon from both sides, that i s , to look at the eff e c t of the PRP on the handling of both SI.and S2. I t was hoped that some support for a p a r t i c u l a r model would be obtained. THE EFFECT OF TASK COMPLEXITY ON REACTION TIME A great deal of experimental work, dating back to the studies of fierkel (1885), has been completed i n an e f f o r t to determine the e f f e c t of varying the complexity of a task on the reaction time to that particular task- I t i s now generally accepted that increasing task complexity w i l l also increase the task's RT for both SRT and CRT s i t u a t i o n s , although the e f f e c t appears to be greatest i n SET conditions (Hyman, 1953; Sidowski, e t . a l . , 1958; Henry and Rogers, 1960; Glencross, 1973). This phenomenon appears to be due to the organization of a response requiring two phases- a compiler or assembly phase and a running phase. A more complex response would require a 28 more complex motor program, t h e r e f o r e , more time would be required f o r the assembly and running of the more complex program {Glencross, 1973) . The a p p l i c a t i o n of the above mentioned s t u d i e s t o t h i s work i s t h a t the manipulation of task complexity provides an e x c e l l e n t method of i n c r e a s i n g or decreasing the ST to each task independently of each other, thus enabling the problem noted i n the Broadbent and Gregory (1967) study to be overcome. PRACTICE, LEARNING AND FEEDBACK An assumption made i n t h i s i n v e s t i g a t i o n was that motor responses r e q u i r e a c e r t a i n amount o f feedback i n order t o be performed s k i l l f u l l y and t h a t , with p r a c t i c e , the amount of feedback r e q u i r e d f o r performance of the task i s reduced. C u r r e n t l y , there are two major p o i n t s of view regarding the u t i l i z a t i o n of feedback i n motor l e a r n i n g : the close d - l o o p theory as expressed by Adams (1971), and the Motor Program approach, based on the work of Lashley and more r e c e n t l y being developed by Schmidt (1976). Although these two viewpoints are i n o p p o s i t i o n , they p r e d i c t s i m i l a r r e s u l t s when viewed with respect t o p r a c t i c e and l e a r n i n g . Adams maintains t h a t , with l e a r n i n g , a perceptual t r a c e i s developed t o which we can compare a response. As t h i s t r a c e i s being developed, a good deal of e x t e r n a l feedback i s r e q u i r e d but once the trace i s w e l l developed, the e x t e r n a l feedback i s not necessary and e r r o r i d e n t i f i c a t i o n i s made r a p i d l y through a comparison of the perceptual t r a c e and i n t e r n a l feedback sources. Therefore, Adam*s theory would p r e d i c t decreasing dependence on feedback with p r a c t i c e and l e a r n i n g . 29 The motor program approach does not maintain a feedback element, per se, but does require a necessary amount of attention i n the compilation and execution of a response, which could be interpreted as a feedback element. In the early stages of learning, the motor program i s not s e l l developed and a great deal of attention i s reguired f o r the execution of a motor s k i l l . With practice, the motor program s i l l become better established and would require l e s s and l e s s attention for i t s execution, as by d e f i n i t i o n , the motor program implies movement without moment-to-moment regulation. Therefore, through the learning process, the amount of attention, or feedback, necessary for performance i s reduced to a point where, eventually, feedback i s no longer required. I t becomes apparent that both major theories of motor learning would expect a reduction i n feedback, or attention, with practice, as i s assumed in t h i s study-30 Chapter 3 METHODS AND PROCEDURES Subjects Twenty-eight, right handed, male students from the undergraduate and graduate programs of the School of Physical Education and Recreation, University of B r i t i s h Columbia volunteered to p a r t i c i p a t e i n the study. Fourteen of the subjects were randomly assigned to the experiment 1 phase, and the remaining fourteen were assigned to the experiment 2 phase of the in v e s t i g a t i o n . Apparatus The apparatus u t i l i z e d was a reaction time device, of the experimenter's own design, which consisted of two component parts: the subject's response apparatus and the experimenter*s electronic control equipment. Subject's response apparatus. This response apparatus was composed of a joy s t i c k and a hand held response switch. The joy s t i c k was set into a diamond shaped opening cut into a table top. Two micro-switches, one located i n the vertex of the diamond c l o s e s t to the subject and the other i n s t a l l e d i n the ri g h t hand vertex, acted as s t a r t i n g and ending points, respectively. Metal contact elements were located i n the top and l e f t hand vertices of the diamond and on the joy s t i c k handle. The subject's display panel consisted of three colored l i g h t s situated i n the form of an e q u i l a t e r a l triangle (5cm) attached at a s l i g h t angle to the top of the table, A yellow l i g h t , 31 situated in the apex of the t r i a n g l e , served as a warning l i g h t and two red l i g h t s , located i n the l e f t and r i g h t base apices of the t r i a n g l e , were used as stimulus l i g h t s 1 and 2, respectively. 4t the beginning of each t r i a l the subject held the joy s t i c k , with his right hand, against the starting micro-switch such that the switch was depressed. The separate hand response button was held i n the S's l e f t hand. To mask a l l extraneous noises a pair of earmuffs were worn during a l l t r i a l s . The subject's response apparatus was located to the l e f t and s l i g h t l y ahead of the experimenter's contol equipment to ensure that the c o n t r o l equipment was outside the subject's f i e l d of v i s i o n . Experimenter's control equipment. The experimenter's control equipment consisted of two separate timing banks, an elec t r o n i c counter and three clocks. The two timing banks, one a 4-channel timer (model 52011, Lafayette Instrument Co.) and the other a 2-channel Universal timing module (CH. Stoelting Co.) were connected i n series and provided a t o t a l of six channels for use. Channel one controlled the "on" time of the warning l i g h t , channel four the "on" time of S1, and channel six the "on" time of S2. Channel three was used to manipulate the foreperiod for SI, and channel f i v e was used to manipulate the ISI. Channel two was not used. Channels one through four could be set to time to the nearest tenth of a second and channels f i v e and six to the nearest hundreth of a second. The e n t i r e sequence, beginning with channel one and running through to channel six was started by a c t i v a t i n g the i n i t i a t e switch on the 32 4-channel timing bank. Of the three clocks, two were d i g i t a l readout clocks (model 54517, Lafayette Instrument Co.; C1 and C2) , and one was a Klockkounter (model 120A, Hunter Mfg. Co.; C3). A l l three clocks were capable of measuring time i n t e r v a l s to the nearest millisecond: C1 was used to measure the subject's BT to SI,. C2 measured the subject's BT to S2 and C3 measured the subject's TBT, from the occurrence of Si to the completion of the movement with -the joy s t i c k . A l l times were recorded on data sheets by the experimenter. The counter (data recorder, model 5804, Lafayette Instrument Co.), was used to determine i f the subject had f u l l y completed the required movement and had contacted a l l v e r t i c e s of the diamond while executing the complex movement by recording two h i t s . Procedures Response tasks. Three d i s t i n c t response tasks were used i n this study.. On each te s t t r i a l , i t was necessary f o r the subject to perform two of the tasks selected beforehand. The three tasks were: 1. SM1- a simple movement completed by depressing the button of the hand held switch. 2. SM2- a simple movement involving the releasing of the sta r t micro-switch by moving the joy s t i c k off the switch. 3. CM- a rapid, three change of di r e c t i o n , more complex movement ( r e l a t i v e to SM1 and SH2) , involving the movement of the joy st i c k off the s t a r t micro-switch to touch the upper vertex of the diamond, then to the l e f t hand vertex and 33 f i n i s h i n g i n contact with the end micro-switch-The joy s t i c k * s movements (SM2 and CM) were executed with the r i g h t hand and the sequence of events for each test t r i a l was as follows: 1. Subject in ready p o s i t i o n . 2. Warning l i g h t illuminated. 3. Stimulus l i g h t one (S1) illuminated. 4. Subject responds with appropriate movement., During occurrence of steps 3 and 4 the following procedures occurred: 5. Stimulus l i g h t two (S2) illuminated. 6. Subject responds with appropriate movement. 7. BTs to S1 and S2 and THTs recorded. 8. Subject returns to ready position as experimenter resets instruments. Experimental Conditions (EC) As the study was designed to investigate the problem under two d i f f e r i n g conditions, the experimentation was conducted through the two following paradigms: Experiment 1. The three response tasks were ordered and paired together in such a manner that two ECs were u t i l i z e d . They were as follows: 1. EC 1:. SM1 - CM 2. EC 2: SM1 - SM2 During EC 1, the response task SM1 was completed i n response to S1 and CM was performed i n response to S2. EC 2 was s i m i l a r in that SHI vas performed i n response to SI, but SM2 was 34 the required response for S2-Only one testing session was required per subject, during which the subject was f i r s t required to perform 30 t r i a l s of SM1 only, 30 t r i a l s of SM2 only and 30 t r i a l s of CM only. During these t r i a l s , each response was paired with the stimulus l i g h t to which i t would be associated when performing under the required ECs. In addition, the order of performance of these three tasks was randomly assigned to each subject i n order to negate any order e f f e c t . The f i n a l 20 t r i a l s of each of these sets of 30 t r i a l s were used to determine the subject's representative RT to each of the respective tasks. The subject then performed 10 practice t r i a l s , followed by the 30 test t r i a l s , under each experimental condition.. The order of presentation of the two ECs was balanced so that hal f of the subjects performed EC 1 f i r s t , and the other half performed EC 2 f i r s t . Each testing session lasted approximately one hour. Experiment 2. As with Experiment 1, the three tasks were paired and ordered so that two d i f f e r e n t ECs were produced, as follows: 1. EC 3: CM - SM1 2. EC 4: SM2 - SH1 In EC 3, SI was to e l i c i t CM and S2 was to e l i c i t SMI... During EC 4, SM2 was the required response to the occurrence of S1, with SM1 again being the response to S2. This experiment required three sessions with each subject, that i s , one practice and two te s t i n g sessions which were conducted over a period of three separate days- The sessions were scheduled as follows and were controlled so that no more 35 than one day would elapse between any two of the sessions: Day 1: testing session 1- the subject performed 10 practice t r i a l s and 30 test t r i a l s of EC 3 and EC 4. Day 2: practice sessions consisted of the subject performing 100 t r i a l s of CM only and 100 t r i a l s of SM2 only. Day3: testing session 2- the subject i n i t i a l l y performed 30 t r i a l s of SM1 only. He then responded to 25 t r i a l s of CM only, followed by 10 practice t r i a l s and 30 test t r i a l s of EC 3 and 25 t r i a l s of SM2 only, followed by 10 practice t r i a l s and 30 t e s t t r i a l s of EC 4. To negate any order effect which could r e s u l t from the ECs, the order of presentation of the two ECs was balanced among the subjects. That i s , on each day, half of the subjects would perform one task f i r s t and the other half would perform the other task f i r s t . On day 1, then, seven subjects performed t h e i r t r i a l s of EC 3 f i r s t and the other seven performed t h e i r t r i a l s of EC 4 f i r s t . On day 3, t h i s order was reversed so that the seven subjects who performed EC 3 f i r s t on day 1 performed EC 4 f i r s t . The practice session on day 2 was s i m i l a r i l y balanced, with seven subjects performing their 100 t r i a l s of CM f i r s t and the other seven performing t h e i r 100 t r i a l s of SH2 f i r s t -Testing session 1 took approximately 40 minutes to complete, with the practice session and t e s t i n g session 2 requiring approximately 1 hour each. Experimental Controls To counteract the influence of several intervening variables, a number of experimental controls were u t i l i z e d . The grouping e f f e c t , that has been noted i n the l i t e r a t u r e , 3 6 was negated by highly emphasizing the f i r s t s i g n a l , as outlined by Way and Gottsdanker (1968). Through t h i s technique, the subjects were informed of the i r BT to the f i r s t s i g n a l on a l l t r i a l s , and were directed to attempt to maintain t h i s BT close to a normal BT of thi s task under a single stimulation s i t u a t i o n . Also, to maintain event uncertainty and to counteract any a n t i c i p a t i o n to the second s i g n a l , f i v e of the t h i r t y t e s t t r i a l s excluded S1, f i v e of the t r i a l s presented a varied ISI, and a variable foreperiod was employed. A f i n a l experimental control was an a p r i o r i application of the :Massey—Dixon B r a t i o f o r i d e n t i f i c a t i o n of o u t l i e r s to the MTs so that any t r i a l s on which the subjects slowed down th e i r f i r s t response could be i n d e n t i f i e d and discarded. The applicaton of these experimental controls produced ECs consisting of 20 test t r i a l s at a fixed ISI of 150 msecs. An ISI of 150 msecs was chosen f o r two reasons..Firstly, i n order to employ the PBP paradigm, i t i s necessary to have an ISI that i s within a normal BT to the f i r s t s i g n a l which, f o r the tasks SB 1 and SH2 was approximately 200 msecs, and secondly, Welford (1968) has found that with ISIs of 50 to 100 msecs, i t i s highly probable that grouping w i l l occur-Experimental Design For Experiment 1, the experimental design was a one-way repeated measures design. The independent variable was a complex variable of task complexity at two l e v e l s (simple to complex) under EC and experimental paradigm (single to double stimulaion) and the dependent variable was BT. For each subject, a t o t a l of 140 BT t r i a l s were recorded. The mean BTs for each of the three 37 tasks separately and under each of the two ECs were then calculated, to provide seven measures of the dependent variable for further data analysis. For Experiments, the same experimental design, dependent and independent variables, as applied to Experiment 1, were u t i l i z e d . For each subject, a t o t a l of 160 BT t r i a l s were recorded with the mean BT fo r each task under each of the two ECs for t r i a l 1 and t r i a l 2 being calculated, thus providing eight measures of the dependent variable f o r further data analysis. Analysis Of Data Prior to the commencement of any data analysis, the Massey-Dixon B r a t i o f o r i d e n t i f i c a t i o n of o u t l i e r s was employed to determine i f any abnormal scores had been included i n the data (see Table A.3). Then, four one-way repeated measures ANOVAs were calculated using computer program BMD:P2V.,The f i r s t ANOVA was to analyze the data from Experiment 1,; the second and t h i r d ANOVAs analyzed the data from t r i a l 1 and t r i a l 2 of the second experiment separately, and the fourth analyzed t r i a l s 1 and 2 together. The four ANOVAs were calculated to obtain the MS error term thus enabling preplanned comparisons to be used to tes t for si g n i f i c a n c e under each hypotheses. The three ANOVAs calculated for Experiment 2 allowed for analysis of t r i a l s 1 and 2 separately and then the two t r i a l s together. Test of Hypothesis J.. Hypothesis 1 predicted that the BT for SM1 would be longer under a double stimulation s i t u a t i o n than under a single stimulation s i t u a t i o n and that the BT to SM1 under EC 1 would be longer than the BT to SMI under EC 2. Three 38 pairwise non-orthogonal comparisons u t i l i z i n g Dunn's procedure sere performed to te s t for s i g n i f i c a n t differences between each of the pairs. Test of Hypothesis 2 . Hypothesis 2 predicted that the difference between the two fiTs under EC 3 would be greater than the difference between the pairs under EC 4 . An a p r i o r i orthogonal comparison using an F r a t i o was u t i l i z e d to test f o r si g n i f i c a n c e on t r i a l 1 and on t r i a l 2 Test of Hypothesis 3- Hypothesis 3 predicted that the relat i o n s h i p tested under Hypothesis 2 would be l e s s on t r i a l 2 than on t r i a l 1. Again, an orthogonal comparison using an F r a t i o was u t i l i z e d to test f o r s i g n i f i c a n c e . 39 Chapter 4 BESULTS AND DISCUSSION To determine representive BT to a task . i n a si n g l e stimulation s i t u a t i o n , the i n i t i a l 10 scores,which were regarded as orientation t r i a l s , were discarded, and the mean of the remaining t r i a l s was calculated. In addition, the Massey-Dixon B r a t i o for i d e n t i f i c a t i o n of o u t l i e r s was u t i l i z e d to i d e n t i f y any response that could be termed unusual and these t r i a l s were eliminated from the cal c u l a t i o n of the mean. For a l l 28 subjects, not more than one t r i a l under each EC was eliminated by t h i s process. Therefore, not less than 19 t r i a l s were used to calculate the mean BT under any of the four ECs. The subjects* individual reduced data are presented i n tables A.l and A-2 i n the Appendix. Table 4.1, which follows, presents the resultant accumulated mean BTs and t h e i r SDs f o r each task. Subsequent to the data reduction, computer program BMDrP2V was used to calculate four one-way repeated measure ANOVAs, one for Experiment 1 and three f o r Experiment 2. Of the three ANOVAs calculated for Experiment 2, the f i r s t analyzed t r i a l one, the second analyzed t r i a l 2 and the t h i r d analyzed t r i a l s one and two together. The ANOVAs were calculated i n order to obtain the MS error score which was necessary to complete the preplanned comparisons. The r e s u l t s of these ANOVAs are given i n Table 4.2. Due to the nature of the experiment, l i t t l e can be construed from an examination of the ANOVA tables. I t was necessary to u t i l i z e preplanned comparisons to test for sig n i f i c a n c e under each of the hypotheses i n d i v i d u a l l y , the resu l t s of which w i l l be discussed i n the next section. no Table 4. 1 Mean Reaction Times and Standard Deviations For Both Experiments EXPERIMENT 1 SM 1 NORMS SM2 PRP CM SH1 - CM SM1 - SM2 X SD 204 25.1 204 . 18.3 230 218 322 217 16.3 26. 1 23.9 24.4 313 32.8 EXPERIMENT 2 SM1 NORMS SH2 CM CM TRIAL 1 - SMI SM2 - SM1 CM TRIAL - SM 1 2 SM2 - SM 1 X SD 181 10.9 212 223 11.7 19. 237 5 27. 427 237 308 220 4 84-9 27.5 41.7 26.9 373 77.7 220 14.6 272 26.: RESULTS AND DISCUSSION OF THE PREPLANNED COMPARISONS OF THE HYPOTHESES Experiment _1 In order to test for significance between RT1 i n EC 1 and EC2, and with the RT to the same task within a single stimulation s i t u a t i o n , three pairwise, preplanned non-orthogonal comparisons using Dunn's procedure were employed. The r e s u l t 1 of this showed non-significant differences between a l l c r i t i c a l pairs at the .05 l e v e l (Table 4. 3). The term " c r i t i c a l pair" i s being used i n t h i s context to describe the three preplanned pairwise comparisons being made, that i s , between the RTs to SM1 (Norm) and SM1 (EC1); SM1 (Norm) and SM1 (EC2) ; and SM1 (EC1) and SM 1 (EC2) . The r e s u l t s obtained were somewhat unexpected as the more recent l i t e r a t u r e reviewed had found s i g n i f i c a n t increases i n 41 Table 4.2 One-way Repeated Measures Analysis of Variance Tables For Experiment 1 and Experiment 2 EXPERIMENT 1 SODRCE Subjects Conditions SxC (error) ss df 24690.78 13 218303.56 6 29326.21 78 ms 1899.29 36383.92 375.98 F 96.771 P <. 001 EXPERIMENT 2- TRIAL 1 Subjects Conditions SxC (error) 62462.39 13 335726.56 3 73601.19 39 4804.79 111908-81 1887.21 59.298 <.001 , EXPERIMENT 2- TRIAL 2 Subjects Conditions SxC (error) 49854.71 13 219212.06 3 50349.21 39 3834.98 73070.69 1291.01 56.599 <.001 EXPERIMENT 2-- BOTH TRIALS Subjects Conditions SxC (error) 102726.56 13 581755.44 7 133540.50 91 7902.04 83107.87 1467.48 56.633 <.001 RT1 from a single to a double stimulation s i t u a t i o n (Gottsdanker, Broadbent, and Van Sant, 1963; Herman and Kantowitz, 1970) and when the processing time of S2 was manipulated (Say and Gottsdanker, 1968; K a r l i n and Kestenbaum, 1968; Smith, 1969). A further examination of the data revealed that there was a s i g n i f i c a n t difference between the RTs of SM1 and CM under a single stimulation s i t u a t i o n which would in d i c a t e that CH was attended to as a more complex response by the Ss (Table A.3). Also, by applying the Massey-Dixon R r a t i o for the i d e n t i f i c a t i o n of o u t l i e r s , i t was determined that no extreme means were biasing the data. These data, then, cannot be explained by a p a r a l l e l 4 2 Table 4.3 Differences between The Means of C r i t i c a l Pairs for Experiment 1 NORMS EC 1 EC 2 sm1 sm1 . sm 1 ; NORM sm1 (x= =204) - 13 14 EC1 sm1 (x= = 217) - 1 EC 2 sml (x= =218) -D i f f . needed for si g n i f i c a n c e (.05) - 23.16 processing model but f i t s n i cely into a SCT explanation, as proposed by Belford. As was noted in Chapter 2, a p a r a l l e l processing model would predict increases in BT1 from a single to a double stimulation s i t u a t i o n , and an increase i n RT1 i f the complexity of R2 was increased. However, SCT predicts that BT1 would not be di f f e r e n t i n a double stimulation s i t u a t i o n when compared to the RT to t h i s task alone. Also, the storage hypothesis maintains that the processing of Si w i l l continue normally, regardless of the nature of the second response. Both these predictions were supported by the data c o l l e c t e d . To further investigate the re l a t i o n s h i p between Welford's SCT and the data c o l l e c t e d , predicted values f o r BT2 were calculated by using formula (1), (see p.13), and these values compared to the experimental values which had been obtained (Table 4.4) . The values for the actual RTs obtained were consistently longer than the predicted values, with only 4 of the 28 sets of data c o l l e c t e d showing a negative r e l a t i o n s h i p . The mean difference values of +40.4 msecs and +25.9 msecs f o r EC1 and EC2, repectively, are consistent with the values as noted by 43 Table 4-4 Difference Between Predicted and Actual Values For Mean BT2 Under PBP E f f e c t f o r Experiment 1 By Subject EC 1 . EC 2 SUBJECT ACT- PBED- DIFF- ACT. PBED. DIFF. 1 338 227 • 11.1 309 270 •39 2 290 254 • 36 287 265 +22 3 362 208 • 154 313 262 +51 4 328 235 +93 333 281 +52 5 305 303 +2 319 313 • 6 6 333 263 • 70 347 289 •58 7 262 252 + 10 326 294 • 30 8 362 334 • 28 356 379 -21 9 310 325 -15 332 321 • 11 10 335 313 • 22 340 335 • 5 11 296 235 +61 269 271 -2 12 277 251 • 26 376 289 • 47 13 322 310 + 12 306 294 • 12 14 259 304 -45 341 288 +53 X 312-8 272.4 • 40. 4 322.4 296.5 • 25.9 Bertelson {1966) i n his review of the l i t e r a t u r e . These elongated ETs would see m to indicate the operation of a checking mechanism or a s h i f t i n g of attention delay a f t e r the processing of SI and before the processing of S2 has begun, as : predicted by the single channel theorists- However, t h i s statement i s pure conjecture as i t i s not possible to determine the nature of the additional delay from t h i s experiment. I t i s towards t h i s question that Experiment 2 was directed . A p a r a l l e l processing model, as expressed by Mcleod (1977), would predict negative difference values. By comparing the actual to predicted scores. a d i r e c t evaluation of the nature of the IBI i s made. I f a p a r a l l e l processing model were to be adopted, t h i s c h a r a c t e r i s t i c of the IBI would have to be negative. However, Mcleod's contention that experimentally imposed response strategies could cause a s e r i a l processing a r t i f a c t . i n the data, 44 as was obtained, has been duly noted and w i l l be dealt with l a t e r in t h i s discussion. Experiment 2 Hypothesis 2. To determine i f an increase i n the complexity of the f i r s t response would cause a corresponding increase i n the delay i n responding to SI, an a p r i o r i orthogonal comparison by means of an F r a t i o was used. The application of t h i s procedure, as outlined by Kirk .. (1968) , revealed that the difference between the two tasks i n EC3 was s i g n i f i c a n t l y d ifferent from the differences between the two tasks i n EC4. This r e l a t i o n s h i p held for both t r i a l 1 and t r i a l 2 ( t r i a l 1, F=26.38, p<-001; t r i a l 2, F=26.87, p<.001). This r e s u l t demonstrated that the delay i n responding to S2 was much greater when the f i r s t response was more complex. An examination of the data showed that the increase i n delay was greater than the un i t - f o r - u n i t increase as predicted by s t r i c t SCT. This data would seem to indicate support f o r the hypothesis as stated, and for SCT with a feedback modification. The d i r e c t i o n of the r e s u l t s obtained i s consistent with those of the studies reviewed (Kay and Weiss, 1961; Broadbent and Gregory, 1967), however, int e r p r e t a t i o n i s made more d i f f i c u l t as a non-s i g n i f i c a n t difference existed between the simple and complex tasks employed. In f a c t , on both t r i a l 1 and t r i a l 2, the RTs for SM2 and CH were i d e n t i c a l . Despite t h i s , the BTs to S2 were s i g n i f i c a n t l y d i f f e r e n t , which would seem to i n d i c a t e that the subjects s t i l l attended to CM as being more complex even though the BTs would not indicate t h i s . Hypothesis 3. Again, an a p r i o r i orthogonal comparison by 45 means of an F r a t i o was used to determine i f the relationship investigated i n hypothesis 2 changed s i g n i f i c a n t l y from t r i a l 1 to t r i a l 2- The difference was found to be n o n s i g n i f i c a n t (F=.46, p>.05) and even though the F r a t i o was quite small, an inspection of the data revealed that there was a change i n the hypothesized d i r e c t i o n . This was shown by comparing the actual and predicted RT2s- the predicted RT2s being calculated by using Welford ,s formula (1). Through t h i s procedure, the amount of delay that can be accredited to a feedback element i s demonstrated guite c l e a r l y and the amount by which th i s element i s reduced from t r i a l 1 to t r i a l 2 can be noted. The results are shown i n Table 4.5. Table 4.5 Comparison of Reduction in Difference Between Predicted And Actual RT2s From T r i a l 1 to T r i a l 2 Under Both Experimental Conditions For Experiment 2 EC 3 EC 4 d i f f 1 d i f f 1 T r i a l 1 T r i a l 2 - d i f f 2 T r i a l 1 T r i a l 2 - d i f f 2 ACT 427 373 308 272 PRED 267 250 267 250 DIFF 160 123 37 41 22 19 For EC 4, the extra delay was reduced by 19 msecs and for EC 3 the extra delay was reduced by 37 msecs. These type of changes, and the magnitude of the numbers involved, would seem to indicate a SCT with feedback modification model as has been proposed by Welford. However, t h i s contention i s not supported by the s t a t i s t i c a l analysis, thus there i s only a s l i g h t i n d i c a t i o n that a change i n the hypothesized d i r e c t i o n did 46 occur. It would seen reasonable to assume that, for the more complex movement, a greater amount of feedback would be reguired for performance and that t h i s would produce a longer a d d i t i o n a l delay when compared to a simpler movement. I t would also appear reasonable to assume that, since the more complex movement has a larger feedback element, practice would have a greater e f f e c t on reducing the feedback necessary i n the more complex movement. I t may be that the length of practice used i n t h i s study was not s u f f i c i e n t for the anticipated changes to occur and,thus, non-s i g n i f i c a n t r e s u l t s were produced. SUMMARY OF HYPOTHESES The r e s u l t s from the data analysis of hypothesis 1, Experiment 1, and hypotheses 2 and 3, Experiment 2, appear to give consistent support to a s e r i a l processing view of information processing, more s p e c i f i c a l l y the SCT view as formulated by Welford. The r e s u l t s of the analysis of the data co l l e c t e d i n Experiment 1 have demonstrated that the processing of S1 can continue normally under a double stimulation s i t u a t i o n and i s unaffected by the a r r i v a l time of S2 or the complexity of the response reguired by S2. Also, i t was shown that the delay i n responding to S2 was not only equal to the length of time predicted by s t r i c t SCT but i t was delayed an ad d i t i o n a l 25 to 40 msecs, on the average. This s i t u a t i o n can be interpreted as a demonstration of Selford's feedback element, that i s , a type of checking mechanism which ensures that the f i r s t reponse has indeed been i n i t i a t e d . Testing of the hypotheses in Experiments indicated that 47 the delay in responding to S2 was s i g n i f i c a n t l y lengthened by the increase i n complexity of the response f o r S1. This elongated delay was greater than the u n i t - f o r - u n i t increase predicted by s t r i c t SCT, which seems to provide further evidence of Welford's feedback element., However, practice did not s i g n i f i c a n t l y decrease the feedback element, as i t was expected i t would, but, rather, a trend i n the hypothesized d i r e c t i o n was noted and i t i s possible that with a longer practice schedule the expected re s u l t s may have been achieved. Although the above r e s u l t s seem to indicate support for a s e r i a l processing view, i t i s not possible to t o t a l l y r e j e c t a p a r a l l e l processing approach as t»o problems which have come to l i g h t through the course of t h i s investigation present d i f f i c u l t i e s when attempting to make a judgement between the two views. The assumption that task complexity i s d i r e c t l y related to BT, though generally accepted, does not seem to consistently hold true. In Experiment 1, the difference i n BT to SM2 and CM was 26 : msecs, which was a s i g n i f i c a n t difference, but i n Experiment 2 the difference was only 11 msecs. Of the 28 subjects used i n both experiments, 22 responded more slowly to CM than SH2, 5 responded more slowly to SM2 than CM and one subject had the same BT to both tasks. Similar d i f f i c u l t i e s i n using task complexity to a f f e c t BT have been noted i n other studies undertaken at the University of B r i t i s h Columbia (Leech, 1977; B o l l o , 1978). It seems that i t i s necessary to u t i l i z e a screening process to i d e n t i f y subjects more responsive to task complexity i f t h i s p a r t i c u l a r methodology i s to be used. 48 Mcleod's contention that experimental instruction can impose a response strategy on the part of the subject may well be true. In t h i s p a r t i c u l a r study, the subject was instructed to respond immediately to the occurranee of S1 and not to delay the response in a n t i c i p a t i o n of S2. According to Mcleod, this would be an example of imposing a s e r i a l response strategy on the subject and the r e s u l t s obtained would be exactly as expected because of the experimental i n s t r u c t i o n s . , I f neither response was emphasized i n the i n s t r u c t i o n , then the subject would perceive the responses as being of equal importance and would probably display a p a r a l l e l processing type of behaviour. The question which would have to be answered then, i s exactly which type of responses are those that could be termed '•appropriate" responses on the part of the subject? I t i s the be l i e f of the investigator that man contains an extremely f l e x i b l e and adaptive information processing system and whether the processing of information i s of a s e r i a l or p a r a l l e l nature depends e n t i r e l y on the response strategy as invoked by the interpretation of the s i t u a t i o n . Attempting to distinguish between models by using a PRP paridigm i s d i f f i c u l t at best as t h i s p a r t i c u l a r approach appears to be extremely , sens i t i v e to reponse strategies and therefore can demonstrate a wide range of re s u l t s which could be used to support a diverse number of views. I f we accept an information processing system of a f l e x i b l e and adaptive nature, then a more f r u i t f u l approach to a model explaining the underlying nature of human information processing would be to develop a more adaptive model, such as Legge and Barber's (1976) multi-channel approach, rather than the more r e s t r i c t i v e s e r i a l or p a r a l l e l processing models. 50 Chapter 5 SUMMftBY AND CONCLUSIONS Summary The main purpose of t h i s i n v e s t i g a t i o n was to study the effe c t that changing the task complexity of the second response would have on the processing of the f i r s t response i n a psychological refractory period paradigm. Sub-problems related to the investigation examined the e f f e c t s of changing task complexity of the f i r s t response on the processing of the second s i g n a l , and, i n addition, i f practice and i t s related subsequent learning, would influence t h i s r e l a t i o n s h i p . I t was anticipated that examination of these problems would permit a d i s t i n c t i o n to be made between s e r i a l and p a r a l l e l processing views of human information processing theory. To achieve these ends, two experiments were performed. Experiment 1, which investigated the main problem, involved 14 male subjects who were required to respond to 30 t r i a l s of a simple movement followed by a complex movement and then, to respond to 30 t r i a l s of the same simple movement followed by a second, but d i f f e r e n t , simple movement. Bithin each block of 30 t r i a l s , 20 were performed at a fixed ISI of 150 msecs. During the testing session, the subjects* representative BT to each of the tasks was determined under a si n g l e stimulation s i t u a t i o n . One, hour long testing session was required. Experiment 2, which investigated the sub-problems, involved another group of 14 male subjects who were required to perform a number of tasks over a series of three testing sessions. In the f i r s t t e sting session, 51 the subjects responded to 30 t r i a l s of a complex movement followed by a simple movement and then responed to 30 t r i a l s of a simple movement followed by the same simple movement as used in the f i r s t set of t r i a l s - In testing session two, the subjects received practice on the f i r s t response of each of the experimental conditions. They were required to perform 100 t r i a l s of each task under a single stimulation s i t u a t i o n . Testing session three required the subjects to perform 25 "reminder" t r i a l s of the f i r s t tasks separately, and then a repeat of the procedure involved i n testing session one., Within each block of the four blocks of 30 test t r i a l s , 20 t r i a l s were performed at a fixed ISI of 150 msecs. Three, 45-minute testing sessions were used. Three experimental tasks were used: a complex movement involving a rapid, three-changes-of-direction movement of a joy st i c k apparatus, a simple movement reguiring the depression of a hand held response button, and a second simple movement consisting of the removal of a joy s t i c k o f f of a micro-switch. Each movement was i n i t i a t e d i n response to the illumination of a stimulus l i g h t . Subsequent to an analysis of variance, a number of pre-planned comparisons were administered to test each s p e c i f i c hypothesis. Conclusions The conclusions formulated from t h i s investigation are as follows: 1. The ET to the f i r s t s i g n a l i n a double stimulation s i t u a t i o n did not d i f f e r s i g n i f i c a n t l y from the ET to 52 t h i s s i g n a l i n a single stimulation s i t u a t i o n . 2. Increasing the complexity of the second response did not af f e c t the processing of the f i r s t s i g n a l . , 3. Increasing the complexity of the f i r s t response r e s u l t s i n a s i g n i f i c a n t delay i n the response to the second s i g n a l , but greater than a unit-for-runit increase as predicted by a s t r i c t SCT view. These conclusions support a s e r i a l view of human information processing, s p e c i f i c a l l y the Single Channel Theory as proposed by Helford (1952). 4. As a resu l t of prac t i c i n g the f i r s t task, the amount of delay when a complex task i s used i n response to S1 i s not reduced more than when a simple task i s used i n response to S1. Although the change noted i n the data was i n the hypothesized d i r e c t i o n , t h i s conclusion does not support Single Channel Theory due to the non-significant r e s u l t s . SUGGESTIONS FOR FURTHER RESEARCH In l i g h t of the problem a r i s i n g as a resu l t of the use of varying task complexity, i t i s recommended that i n further research employing t h i s method of manipulating RTs, either a pretest be administered to the subjects and only those who have a s i g n i f i c a n t l y slower RT to the more complex task be used or that tasks of various complexities that r e l i a b l y produce the hoped for changes i n RT be u t i l i z e d . In addition, i t i s suggested that Experiment 2 be re p l i c a t e d , using more t r i a l s 53 during the practice period, and that the added e f f e c t of a varied ISI be incorporated. Furthermore, i t i s recommended that a r e p l i c a t i o n of the Greenwald and Shulman (1972) study, using the Legge and Barber (1976) multi-channel processing model as a base, be undertaken. 54 BIBLIOGRAPHY 55 Adams, J . A., Test of the Hypothesis of the Psychological Refractory Period. Jounal of Experimental Psychology, 64 (1962) , 280-287. __, A Closed-Loop Theory of Motor Learning, Journal of Motor Behaviour, 3 (1971), 111-149. .„ • S R- H- Chambers, Response to Simultaneous Stimulation of Two Sense Modalities. Journal of Experimental Psychology, 63 (1962), 198-206. All p o r t D. A., B. Antonis, & P. Reynolds, On the Division Of Attention: A Disproof of the Single Channel Hypothesis, Quarterly Journal of Experimental Psychology, 24 (1972), 225-235. Berlyne, D. E., Uncertainty and C o n f l i c t : A Point of Contact Between Information Theory and Behaviour Theory Concepts. Psychological Review, 64 (1957). 329-339. • • C o n f l i c t , Arousal and Cur i o s i t y , Hew York: McGraw-H i l l , 1960. Bernstein, I. H., 8., R. Blake, & M. H. Hughes, Effects of Time and Event Uncertainty upon Seguential Information Processing, Perception and Psychophysics, 3 (1968), 177-184. Bertelson, P., Central Intermittancy Twenty Years Later, Quarterly Journal gf Experimental Psychology. 18 (1966), 153-163. • • • - • •, The Refractory Period of Choice Reactions with Regular and Irregular I n t e r s t i m u l i Intervals, Acta Psychologica. 27 (1967), 45-56. Borger, R., The Refractory Period and S e r i a l Choice Reactions, Quarterly Journal of Experimental Psychology, 15 (1963), 1-12. , Broadbent, D. E. S M. Gregory, Psychological Refractory Period and the Length of Time Reguired to Make a Decision, Proceedings of the Royal Society: Series B« : 168 (1967), 181-193. Craik, K. H. J., Theory of the Human Operator i n Control Systems. I. The Operator as an Engineering System, B r i t i s h Journal of Psychology. 38 (1947), 56-61. , Theory of the Human Operator in Control Systems- I I . Man as an Element i n a Control System. B r i t i s h Journal of  Psychology. 38 (1948), 142-148. 56 Creamer, L. R. , Event Uncertainty, Psychological Refractory Period and Human Data Processing, Journal of Experimental  Psychology, 66 (1963), 187-194. Davis, B., The Limits of the Psychological Refractory Period, Quarterly Journal of Experimental Psychology, 8 (1956), 24-38. __, The Human Operator as a Single Channel Information System, Quarterly Journal of Experimental Psychology, 9 (1957) , 119-129. , The Role of Attention i n the Psychological Refractory Period, Quarterly Journal of Experimental Psychology. 11 (1959), 211-220. , Expectancy and Intermittancy, Quarterly Journal of Experimental Psychology, 17 (1965) , 75-78. Dixon, W. J. S F. J- Massey, Introduction to S t a t i s t i c a l  Analysis, 3rd ed- Men York: McGraw-Hill, 1969. Drazin, D- H., E f f e c t s of Foreperiod, Foreperiod v a r i a b i l i t y and Probabili t y of Stimulus Occurrence on Simple Reaction Time, Journal of Experimental Psychology, 62 (1961), 43-50. El i t h o r n , A., Central Intermittancy: Some Further Observations, Quarterly Journal of Experimental Psychology, 13 (1961), 240-247. . S C. Lawrence, Central I n h i b i t i o n - Some Refractory Observations, Quarterly Journal of Experimental Psychology, 7 (1955) , 116-127. Ferguson, G. A., S t a t i s t i c a l Analysis i n Psychology and Education. New York: McGraw-Hill, 1971., Fraisse, P., La Periode Ref r a c t a i r e Psychologigue, A'nee Psychologigue. 57 (1957), 315-328. Gleacross, D. J . , Response Complexity and the Latency of Different Movement Patterns, Journal of Motor Behaviour, 5 (1973) , 95-104. Gottskanker, R., The E f f e c t of superseding Signals, Quarterly Journal of Experimental Psychology. 18 (1966), 236-249. L. Broadbent, S C. Van Sant, Reaction Time to Single and to F i r s t Signals, Journal of Experimental  Psychology, 66 (1963), 163-167. - , 6 T. C. Way, Varied and Constant I n t e r - s i g n a l Intervals i n Psychological Refractoriness, Journal of Experimental Psychology, 72 (1966), 792-804-57 Greenwald, a . G., 6 H. G. Shulman, On Doing Two Things at Once: II . Elimination of the Psychological Refractory Period, Journal of Experimental Psychology. 101 (1973), 70-76. Halliday, a . , M . , M. Kerr, 6 a . E l i t h o r n , Grouping of Stimuli and apparent Exceptions to the Psychological Refractory Period, Quarterly Journal of Experimental Psychology. 12 (1960), 72-89. Harrison, J . S., Psychological Refractoriness and the Latency Time of Two Consecutive Motor Responses, Research Quarterly. 31 (1960V. 590-600. Henry, F* M., 6 J . S. Harrison, Refractoriness of Fast Movement, Perceptual and Motor S k i l l s , 13 (1961), 351-354. • • . & D. E..Rogers, Increased Response Latency f o r Complicated Movements and a 'Memory Drum' Theory of Neuromotor Reaction, Research Quarterly, 31 (1960) , 448-457. Herman, L.M., & B. H. Kantowitz, The Psychological Refractory Period E f f e c t : Only Half the Double Stimulation Story?, Psychological B u l l e t i n . 73 (1970), 74-88. & M. E. McCauley, Delay i n Responding to the F i r s t Stimulus i n the "Psychological Refractory Period" Experiment: Comparison with Delay Produced by a Second Signal Not Reguiring a Response, Journal of Experimental Psychology. 81 (1969), 344-350. Hick, W. E., The Discontinuous Functioning of the Human Operator i n Pursuit Tasks, Quarterly Journal of Experimental Psychology. 1 (1948), 36-51. Hyman, R., Stimulus Information as a Determinant of Reaction Time, Journal of Experimental Psychology. 45 (1953), 188-195-Kafray, D., 8 D. Kahneman, Capacity Sharing and Refractoriness i n Successive Reactions, Perceptual and Motor S k i l l s . 44 (1977). 327-335. Kahneman, P.. Attention and E f f o r t . Englewood-Cliffs, N. J - : Prentice-Hall, 1973. K a r l i n , L., Reaction Time as a Function of Foreperiod Duration and V a r i a b i l i t y , Journal of Experimental Psychology. 58 (1959), 185-191. , 6 R. Kestenbaum, E f f e c t s of the Number of alternatives on the Psychological Refractory Period, Quarterly Journal of Experimental Psychology. 20 (1968), 167-178. 58 Kay, H. , & A. D. Weiss, Relationship Between Simple and S e r i a l Reaction Time, Nature, 191 (1961), 790-791. Keele, S. W., Effects of Input and Output Modes on Decision Times, Journal of Experimental Psychology, 85 (1970), 157-164. , Attention and Human Performance. P a c i f i c Palisades, C a l i f . ; Goodyear, 1973. Kerr, B., Processing Demands During Mental Operations, Memory and Cognition. 1 (1973), 401-412. Kirk, R. E., Experimental Design: Procedures for the Behavioural Sciences. Belmont. C a l i f . ; Wadsworth, 1968. Leech, M., The Joint E f f e c t s of Task Complexity and Resonse Probability on Response Latency: A Test of the Existence of a Defensive Strategy i n a Two-Choice Reaction Time Task, Unpublished Master's Thesis, University Of B r i t i s h Columbia, 1977. Legge, D-, S k i l l s . Middlesex. England: Penguin Books, 1970. , S P. J . Barber, Information and S k i l l . London: Metheun, 1976. ~ Marcel, A., Some Constraints on Sequential and P a r a l l e l Processing Limits of Attention, Acta Psychologica, 33 (1970) , 77-92. M a r i l l , T., The Psycholigcal Befractory Phase, B r i t i s h Journal of Psychology. 48 (1957), 93-97. Marteniuk, B. G., Information Processing, Channel Capacity, Learning Stages and the Acquisiton of S k i l l , i n H. T. A-Whiting (ed.), Headings i n Human Performance. London: Lepus Books, 1975. Mcleod., P., P a r a l l e l Processing and the Psychological Befractory Period, Acta Psychologica„ 41 (1977), 381-396-Moray, N., Where i s Capacity Limited? A Survey and a Model, Acta Psychologica. 27 (1967), 84-92. Mowrer, 0. H., Preparatory Set (Expectancy)- Soae Methods of Measurement, Psychological Monographs. 52:2 (Whole No. 233) , 1940. Nickerson, B. S., Response Time to the Second of Two Successive Signals as a Function of Absolute and Relative Duration of i n t e r - s i g n a l I n t e r v a l , Perceptual and Motor S k i l l s , 21 (1965) , 3-10. 59 , Psychological Refractory Phase and the Functional Significance of Signals, Journal of Experimental Psychology. 73 (19 67), 303-312. Poulton, E. C., Perceptual Anticipation and Reaction Time, Quarterly Journal of Experimental Psychology f 2 (1950) , 99-112. Reynolds, D., Effects of Double Stimulation: Temporary In h i b i t i o n of Response, Psychological B u l l e t i n , 62 (1964), 333-34 7. • •- , Time and Event Uncertainty i n Unisensory Reaction Time, Journal of Experimental Psychology, 71 (1966), 286-293. Rollo, D., Thesis i n Progress, University of B r i t i s h Columbia, 1978. Rubinsteini L., Intersensory and Intrasensory E f f e c t s i n Simple Reaction Time, Perceptual and Motor S k i l l s , 18 (1964), 159-172. Sanders, A. F., Selective Strategies i n the Assimilation of Successively Represented Signals, Quarterly Journal of Experimental Psychology, 16 (1964), 368-372. •: , P r o b a b i l i s t i c Advance Information and the Psychological Refractory Period, Acta Psychologica, 35 (1971) , 128-137. Schmidt, R. A., Control Processes in Motor S k i l l s , Exercise and Sport Sciences Review, 4 (1976), 229-262. Sh a l l i c e , T., Dual Functions of Consciousness, Psychological  Review, 79 (1972), 383-393. Sidowski, J. B., R. Morgan, G G. Complexity and Instructions Reaction Times, Journal of (1958) , 163-166. Eckstrand, Influence of Task Upon Simple and Discrimination Experimental Psychology, 55 Slater-Hammel, A. T., Psychological Refractory Period i n Simple Paired Responses, Research Quarterly k 29 (1958), 468-481. Smith, M. C , Theories of the Psychological Refractory Period, Psychological B u l l e t i n . 3 (1967), 202-213. _, The Effect of Varying Information on the Psychological Befractory Period, Acta Psychologica. 30 (1969), 220-231. 60 Telford, C. W., Refractory Phase of Voluntary and Associative Responses, Journal of Experimental Psychology, 14 (1931)., 1 -35. Tolkmitt, F. J . , A Revision of the Psychological Refractory Period, Acta Psychologica. 37 (1972), 139-154. Triggs, T. J . , Capacity Sharing and Speeded Reactions to Successive Signals, Technical Report #9, University of Michigan, Contract No. AF 49 (678)-1736, U..S. Dept. of Defence, 1968- , Vince, M- A., Intermittancy of Control Movments and the Psychological Refractory Period, B r i t i s h Journal of Psychology. 38 (1948), 149-157. Bay, T. C., & R. Gottsdanker, Psychological Refractoriness with Varying Differences Between Tasks. Journal of Experimental Psychology. 78 (1968), 38-45-Welford, A- T-, The Psychological Refractory Period and the Timing of High speed Performance: A Review and a Theory, B r i t i s h Journal of Psychology. 43 (1952), 2-19. • • • , Evidence of a Single Channel Decision Mechanism Limiting Performance i n a S e r i a l Reaction Task, Quarterly Journal of Experimental Psychology, 11 (1959)193-210. • . Single Channel Operation i n the Brain, Acta Psychologica. 27 (1967), 5-22-- - • , Fundamentals of S k i l l y London: Metheun, 1968. Williams, L. R. T., Refractoriness of Long Movement, Research Quarterly. 42 (1971), 212-219. , Psychological Refractoriness of Two S e r i a l Motor Tasks, Research Quarterly. 44 (1973), 23-33., - • • . Ef f e c t of Complexity on Movement Refractoriness, Research Quarterly. 46 (1975), 177-183. 6 1 APPENDIX 62 Table A. 1 Experiment 1- Mean BT Data By Subject NO BMS PBP Subj SH1 SH2 CM SMI - SM2 SH1 - CM BK 192 172 212 205 338 217 309 GS 184 197 226 197 290 189 287 MG 169 181 220 177 362 192 313 BL 197 196 238 189 328 193 333 KL 172 221 265 232 305 198 319 JH 219 202 215 211 333 224 347 PX 186 188 233 214 262 213 326 BS 242 221 247 263 362 282 356 PJ 198 230 244 245 310 227 332 BK 255 206 248 257 335 237 340 BC 197 191 218 194 296 203 269 KT 211 200 222 201 277 217 336 SH 228 228 208 232 322 236 306 KF 209 224 228 230 259 210 341 X 204 204 230 218 313 217 322 63 Table A. 2 Experiment 2- Mean BT Data By Subject NORMS PBP T r i a l 1 T r i a l 2 Subj SMI SM2 CM SM2 - SMI CM - SM 1 SM2 - SMI CH - SM 1 ES 194 220 218 223 266 254 355 214 261 225 322 PF 168 196 194 230 346 221 455 217 298 193 425 BG 185 210 210 216 322 204 536 222 268 221 448 KC 174 225 260 300 355 278 374 214 31.1, 217 361 DC 175 212 223 223 297 245 443 213 285 217 380 LJ 187 218 228 226 314 227 362 21.1 297 20 2 363 GK 175 196 210 212 278 223 391 199 236 195 290 BB 202 209 "•"243""- 291 406 296 540 241 3,15 289 432 GS 192 207 243 242 279 242 567 248 278 255 569 GH 179 212 211 227 286 221 357 238 250 222 295 HH 187 205 221 234 258 253 282 209 246 198 357 BC 164 194 196 208 308 207 404 207 253 204 290 AH 176 228 220 236 263 204 394 21.1. 233 198 302 MB 170 230 249 255 330 244 514 234 ... 279 243 390 X 181 212 223 237 308 237 427 220 272 220 373 Table A. 3 Differences between Means for Experiment 1 NORMS NORMS SM1 (X=204) SM2 (X=204) CM (X=230) PRP (EC 1) SMI (X=217) CM (X=322) PRP (EC2) SMI (X=218) SM2 (1=313) PRP CM SM 1 - CM SM1 26» 142 118i 132 26» 14 118V 13 12 92 * 13 - 105* 12 - 104* *p<.05 ' c r i t i c a l pairwise comparison (non-significant at .05) 65 Table A. 4 C r i t e r i o n for Massey-Dixon R r a t i o For I d e n t i f i c a t i o n Of Ou t l i e r s No- Of S t a t i s t i c obs. , k P=70 P=80 P=90 P=95 P=98 P=99 P=99.5 X(2)-X<1) 3 .684 .781 .886 . 941 ... 976 .988 -994 r(10) = 4 .471 .560 .679 .765 .846 .889 -926 X(k)-X(1) 5 .373 .451 .557 .642 .729 .780 .821 6 .318 .386 .482 .560 . 644 .698 .740 7 .281 -344 .434 .507 -586 . 6 37 .680 X{2)-X(1) 8 .318 .385 .479 .554 .631 .683 .725 r(11) = — • 9 .288 .352 .441 .512 .587 .635 . 677 X(k-1)-X(1) 10 .265 .325 .409 . 477 .551 .597 .639 X{3)-X<1) 11 .391 .442 .517 .576 i 638 .679 .713 r(21) = . 12 .370 .419 .490 .546 .605 .642 .675 X(k-1)-X(1) 13 .351 .399 .467 .521 .. 578 .615 .649 X(3)-X{1) 14 370 .421 .492 . 546 . 60 2 .641 .674 r{22) = 15 .353 .402- .472 .525 .579 .616 - 647 X(k-2)-X(1) 16 .338 .386 .454 .507 .559 . 595 .624 17 .325 .373 .438 .490 -542 .577 .605 18 .314 .361 .424 .475 .527 .561 .589 19 . 304 .350 .412 .462 -514 .547 .575 20 .295 .340 .401 .450 . 502 .535 -562 21 . 287 .331 .391 .440 -491 .524 .551 22 .280 .323 .382 . 430 . 481 .514 -541 23 . 274 .316 .374 .421 - 472 . 505 .532 24 .268 .310 .367 . 413 . 464 .497 -524 25 .262 .304 .360 .406 . 457 .489 .516 66 

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}]}"
                            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.831.1-0077340/manifest

Comment

Related Items