University of Southern California Dissertations and Theses

Human Operant Eye Blink Conditioning, Awareness, And The Extraversion-Introversion Dimension Of Personality

(USC Thesis Other)

Human Operant Eye Blink Conditioning, Awareness, And The Extraversion-Introversion Dimension Of Personality

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been microfilmed exactly as received 68— 1685 KESSLER, Irving, 1920- HUMAN OPERANT EYEBLINK CONDITIONING, AWARENESS, AND THE EXTRAVERSION-INTROVERSION DIMENSION OF PERSONALITY. University of Southern California, Ph.D., 1967 Psychology, experimental University Microfilms, Inc., Ann Arbor, Michigan HUMAN OPERANT EYEBLINK CONDITIONING, AWARENESS, AND THE EXTRAVERSION-INTROVERSION DIMENSION OF PERSONALITY by Irving Kessler A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Psychology) August 1967 UNIVERSITY O F SOUTH ERN CALIFORNIA TH E GRADUATE SCHOOL. UNIVERSITY PARK LOS A N G ELES, CA LIFO RN IA 9 0 0 0 7 This dissertation, written by IRVING KESSLER under the direction of h.L$...Dissertation Com mittee, and approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R OF P H I L O S O P H Y <2. A Z h e ....... Dean Date September*.. 19.67..................... DISSERTATION COMMITTEE ■<T. ^ KjL.. ' Chairman ACKNOWLEDGMENTS I wish to convey my deepest feelings of gratitude and esteem to Dr. L. Douglas DeNike for his guidance, patience and dedicated scholarship in his supervision of my research. In addition, I wish to express my warm appreciation to Dr. Henry Slucki for his assistance with the planning and assembling of much of the apparatus, and to Dr. Newton S. Metfessel for his helpful comments and his careful analysis of the dissertation. My thanks to Morris P. Leibovitz, Henry N. Weber, and Arthur H. Weingaertner for their assistance as raters of awareness of the subjects' notes and post-experimental questionnaire protocols. I reserve the final acknowledgment for my wife, Shirley. Her devoted support, her unshakeable dependabil ity, and her unfaltering encouragement provided me with the resolution to persist. Her valuable assistance in adminis tering the post-experimental questionnaire is also acknowledged. TABLE OF CONTENTS Page ACKNOWLEDGMENTS...................................... ii LIST OF TABLES...................................... V LIST OF FIGURES...................................... vii Chapter I. THE PROBLEM.................................. 1 Introduction Background to the Problem Methodological Issues II. METHOD...................................... 12 Subjects Conditioning Procedure Post-Experimental Questionnaire Hypotheses III. RESULTS...................................... 26 Conditioning The Incidence of Awareness The Relationship between Awareness and Conditioning The Telegraph Key Masking Task IV. DISCUSSION................................. 58 V. SUMMARY...................................... 68 APPENDIX A. Table 13, Post-Experimental Question naire ................................ 73 iii Chapter Page APPENDIX B. Table 14, Basic Data................... 78 REFERENCES ................................. 81 LIST OF TABLES Table Page 1. Subject Data Tally Sheet.................. 20 2. Mean Eyeblinks Per Five Minute Period for Two Arbitrarily Chosen Subjects (#19 and #60) . 29 3. Summary of Analysis of Variance of the Eye- blink Performance Data for Males and Females (Experimental and Control Groups Combined).............................. 31 4. Summary of Modified Three Way Analysis of Variance of the Eyeblink Performance Data for the Experimental and Control Groups . . 34 5. Summary of Analysis of Variance and Linear Trend Test of the Eyeblink Performance Data for the Experimental Group............. 36 6. Summary of Analysis of Variance and Linear Trend Test of the Eyeblink Performance Data for the Control G r o u p ................. 37 7. Mean Eyeblink Performance Per Minute for Aware, Correlated Aware, and Unaware Sub jects of Experimental G r o u p ........... 42 8. Summary of Analysis of Variance of the Eye blink Performance Data for the Experimental Aware and Experimental Unaware Groups with Means Adjusted for Operant Rate Differences 44 9. Summary of Analysis of Variance and Linear Trend Test of the Eyeblink Performance Data for the Experimental Unaware Group.... 45 v Table Page 10. Summary of Analysis of Variance of the Eyeblink Performance of Unaware Subjects as a Func tion of the Extent to Which They Wanted the Reinforcement .............................. 49 11. Summary of Analysis of Variance of the Tele graph Key Performance Data for the Experi mental Aware and Experimental Unaware Groups...................................... 51 12. Mean Change in Eyeblink Performance under Reinforcement and Telegraph Key Response under Non-Reinforcement (Session 4 Minus Session 2) for Experimental Subjects in Relation to Awareness, Desire to Earn Points, and Belief or Disbelief in Efficacy of the Telegraph Key Response ................... 55 13. Post-Experimental Questionnaire ............. 73 14. Basic Data..................................... 78 vi LIST OF FIGURES Figure Page 1. Eyeblink Performance in Relation to Rein forcement Conditions and Extraversion- Introversion ................................ 33 2. Eyeblink Performance iri Relation to Subjects' Awareness of Correct or Correlated Response- Reinforcement Hypotheses ................... 46 3. Relationship between Reinforced Eyeblink and Non-Reinforced Telegraph Key Performance for Experimental Aware and Unaware Subjects 54 vii CHAPTER I THE PROBLEM Introduction Operant conditioning is the strengthening through reinforcement of a response which already exists at some strength before conditioning is attempted. Operant con ditioning results in the increase of the probability of a response by following the response with a stimulus that is reinforcing to the organism. There has been, in recent years, a growing interest in the application of operant conditioning techniques to human behaviors, verbal and non-verbal, normal and abnormal (Allyon and Haughton, 1962 Eriksen, 1962; Ferster, Nurnberger and Levitt, 1962; Flanagan, Goldiamond, and Azrin, 1958; Matarazzo, Saslow, and Pareis, 1960; Singer, 1961). There have been many studies (Greenspoon, 1955; Kanfer, in press; Krasner, 1958 McNair, 1957; Salzinger, 1959) which purported to demon strate that the increase in the emitting of a reinforced class of verbal responses was the result of "direct, auto matic, or, in other words, unconscious learning" (Dollard and Miller, 1950, p. 43). However, a great deal of evidence has been pre sented to suggest that verbal conditioning is dependent upon cognition or conscious mediation (DeNike, 1964; Dulany, 1962; Spielberger, 1962). Some reviewers have questioned whether the Law of Effect (Thorndike, 1913) is valid for human learning in general (Adams, 1957), or human non-verbal operant conditioning in particular (Eriksen, 1962). In addition, there has been a great deal of controversy as to whether particular dimensions of personality play a role in the acquisition of learning (Eysenck, 1947, 1960; Sweetbaum, 1963). The main goal of this experiment was to investigate the relationships among human non-verbal operant condition ing, awareness, and the extraversion-introversion dimension of personality. Background to the Problem Psychology as a science has been profoundly affected by E. L. Thorndike's Law of Effect: "Of several responses made to the same situation, those which are accompanied or closely followed by satisfaction to the animal will, other things being equal, be more firmly connected with the situ ation, so that, when it recurs, they will be more likely to recur," (Thorndike, 1911, p. 244). That is, that the strength of a response is directly and automatically in creased if it is immediately followed by a satisfying state of affairs, operationally defined as "one which the animal does nothing to avoid, often doing things as maintain and preserve it" (Thorndike, 1911, p. 245). The learning of a given response, for Thorndike, was mainly under control of its consequences and required only a temporal contingency, not necessarily the cognitions of the organism, to occur. The Law of Effect hypothesis has not always received complete acceptance, and in recent years consider able debate has developed over the implications of the problem of learning without awareness. It is interesting to note that there has been very little evidence that operant learning can take place in normal humans when the subject cannot describe the conditions of the reinforcement which influences his behavior (Adams, 1957; Eriksen, 1962; Keehn, 1967). Some of the early investigations in this area were concerned mostly with verbal conditioning. Thorndike and Rock (1934) found, in their investigation of the relation ship between the Law of Effect and awareness of the rein forcement contingency, that there was a gradual increase of correct responses without any sharp decrease in error performance; therefore they assumed that this result indi cated learning without awareness. Irwin, Kauffman, Prior, and Weaver (1934) immediately challenged this assumption and found evidence that there was no necessary relationship between the slope of a learning curve and a subject's induced awareness of the response-reinforcement contingency. Even when a subject was told the contingency between the response and reinforcement, the increase in performance was gradual. Postman and Jarrett (1952) asked their subjects to state the basis for their performance during a replica tion of the Thorndike and Rock experiment and they reported that there was a slow increase in the slope before the sub ject reported the correct contingency. They suggested that this was evidence of learning without awareness. Philbrick and Postman (1955) stated that in the absence of awareness, there was a steady increase in performance both prior to and after the verbalization of the principle of learning by the successful subjects as compared to the unsuccessful ones. This finding indicated to them that both learning without awareness and verbalization of the correct contin gency were affected by similar variables and that learning can take place when subjects cannot state the principle for themselves. Adams (1957), however, has questioned most of the studies related to learning without awareness asserting that subjects may have formed undetected correlated hypoth eses about the response-reinforced contingency, possibly different from the experimenter's hypothesis, but which would however result in better than chance performance. Further, Adams has sharply stated that there has been no convincing demonstration of learning without awareness in the laboratory, verbal or non-verbal. While DeNike (1964), Dulany (1962) and Spielberger (1962) have apparently demonstrated that in verbal condi tioning experiments performance gains were found essentially for subjects who showed a correct or correlational hypoth esis of response-reinforcement contingencies, few experi menters have attempted to investigate whether Thorndike's Law of Effect is direct and automatic for human operant non-verbal conditioning of a response. Studies by Heffer- line and Keenan (1963), Hefferline, Keenan and Harford (1959), Keehn (1965, 1967), Keehn, Lloyd, Hibbs and Johnson (1965), Sasmore (1966) , and Verplanck (1956) stand almost alone in demonstrating learning without awareness in human operant conditioning of motor responses, and these studies would be met with strong methodological criticisms by both cognitive and S-R theorists. Eriksen (1962) mentions an unsuccessful attempt to produce learning without awareness in a human motor task (Paul, Eriksen and Humphrey, 1962) and is not impressed by any purported laboratory demonstra tion of the learning without awareness effect up to that time. Since the cognitivist theoreticians dispute the methodological procedures that S-R theorists have used in their attempts to demonstrate learning without awareness in verbal or non-verbal investigations, there is then a need to demonstrate whether operant learning occurs without awareness in a human motor task under conditions that would meet the methodological criticisms of most cognitive theorists. Methodological Issues One of the main considerations regarding awareness has been the advantage of using questions during condition ing rather than after conditioning (Spielberger and DeNike, 1966). It is considered that asking questions in a post conditioning interview runs the risk of suggesting correct or correlated hypotheses of the response-reinforcement contingency. Moreover, such questioning does not permit an analysis of the temporal relationship between awareness and conditioning. However, when the subject is instructed to write "thoughts about the experiment" during condition ing, Spielberger and DeNike recognize that this may induce a "problem-solving set" to look for a contingency during conditioning, as would asking actual questions during con ditioning. It was decided in this experiment to use both the method of note writing of "thoughts about the experi ment" within conditioning, and post-experimental question ing as well. Since the focus of the investigation was the occurrence of learning without awareness, it was deemed more important to assure the detection of awareness than to avoid possibly suggesting the contingency artifactually to some subjects, since pseudo-aware subjects would be eliminated from the analysis for learning without awareness in any case. Another consideration is the question of using an extensive and intensive post-experimental questionnaire to make relatively certain that subjects who are aware are not undetected. Of course the danger is that with such a questionnaire there is a strong probability of providing some subjects with sufficient information and suggestions to produce awareness artifactually. However, the present experiment used an extensive and intensive post-experimental questionnaire to assure that all truly aware subjects verbalized their awareness. Criticisms have been made of post-experimental questionnaires on the ground that the questions were vague, unclear and inadequate. In the present experiment, an effort was made to provide specific and clear questions. Another objection to most experiments about learn ing without awareness has been that there has been an inadequate evaluation of correlated hypotheses. Dulany (1962) finds inadequate Adams' (1957) definition of a correlated hypothesis mentioned above. Dulany prefers to define it as verbalization of some response class which is positively, but imperfectly, correlated with the correct response class. The present experiment used this defini tion in analyzing for correct or correlated hypotheses of response-reinforcement contingencies. The defining of correlated hypotheses will be treated more fully in the discussion of instructions to the raters of awareness. There is objection, also, to extinction trials or interpolated tasks which may interfere with a subject's ability to provide awareness in an interview. The present experiment did not use extinction trials or interpolated tasks between the final conditioning sessions and the post- experimental questionnaire. Other specific methodological requisites mentioned by DeNike (1963) include the use of a non-reinforced oper ant period and of a randomly reinforced control group. DeNike criticizes Greenspoon's (1955) methodology of not using a non-reinforced base-line measure of the response class that is to be reinforced. This lack of control for operant level may lead to differences erroneously ascribed to reinforcement. Experiments are continually being reported which equate conditioning with significant gains from the operant base rate, but without controlling for time-correlated increases (Parton and Ross, 1965). In addition, Spielberger, DeNike, and Stein (1965) illustrated the importance of using a control group when they found that random reinforcement gives rise to performance gains in a Taffel sentence-construction verbal conditioning task. The present design avoided these methodological dangers. However, instead of a randomly reinforced control group, a yoked control group was used guaranteeing the exact fre quency and temporal distribution of non-contingent rein forcement for each control subject as there was for each 10 matched contingently reinforced experimental subject. The present experiment is in part essentially a replication of the Keehn, Lloyd, Hibbs and Johnson (1965) study "Operant eyeblink conditioning without awareness: A preliminary report," but with adequate controls for opera tions and subjects. While attempting to meet cognitive learning theorists' requirements, this present study, in addition, used Keehn's technique of non-verbal assessment of awareness along the lines of Skinnerian methodology, in order to assess the comparability of cognitive and behav ioral measures of awareness. Dulany (1962) points out that there are weaknesses in relying only upon verbal reports of private events because they are covert. As a cognitive theorist he has devised a theoretical network using inter connected joint hypotheses to delineate the subject's cognitive states. These derived variables, report of behavioral hypotheses, report of reinforcement hypotheses, and report of behavioral intentions, are still dependent upon subjective report. Thus, they are still subject to the objections of Eriksen (1960) who suggests that con vergent operations be introduced (Garner, Hake, and Eriksen, 1956) in order to separate the influence of a concept (awareness) from contaminating variables associated with 11 verbal report. Keehn et al. (1965) may have contributed by making available a masking task which is not reinforced. We can then attempt to evaluate whether the subject who conditions is aware by observing if his behavior on the non-reinforced masking task extinguishes. CHAPTER II METHOD The experiment dealt with the relationships among three variables: (1) subjects' scores on an extraversion- introversion test (Myers, 1962); (2) reinforcement (Experi mental group) versus non-reinforcement (Control group) of an eyeblink response, defined as a complete closure of the eyes; (3) judges' ratings of subjects' awareness or corre lated awareness of the response-reinforcement contingency as indicated by subjects' written "thoughts about the experiment" during the reinforcement periods, and by their responses to a post-experimental questionnaire. In addi tion, an analysis was made of changes in the masking task response (telegraph key press), as a possible convergent objective indicator of awareness. Subjects The subjects were 14 female and 30 male students enrolled in introductory psychology classes at the Univer sity of Southern California who were required to participate 12 13 in psychology experiments. They were drawn from the upper and lower quartiles of the distribution of extraversion- introversion scores obtained from a pool of 220 students who volunteered to take the Myers-Briggs Type Indicator (F) (Myers, 1962).1 The concepts of extraversion and introversion atti tudes as formulated by Myers are described as follows: The introvert's main interests are in the inner world of concepts and ideas, while the extravert's main interests are in the outer world of people and things. Therefore, when circumstances permit, the introvert directs both perception and judgment upon ideas, while the extravert likes to direct both upon his outside environment. A well-developed introvert . . . does his best work inside his head, in reflection. A similarly well- developed extravert . . . does his best work externally, in action. (p. 57) These patterns can be compared to Eysenck's des cription (Eysenck and Eysenck, 1963) of a "typical" extra vert and of a "typical" introvert. The typical extravert is sociable, likes parties, has many friends, needs to have people to talk to, and does not like reading or studying by himself. He craves excitement, takes chances, often sticks his neck out, acts on the spur of the moment and is gener ally an impulsive individual. He is fond of practical jokes, always has a ready answer, and generally likes 1 The 220 students were also given the Eysenck Per sonality Inventory (Eysenck & Eysenck, 1963) for experimen tal analysis of correlation between the two tests. The product-moment correlation between the tests was .66. 14 change; he is carefree, easygoing, optimistic, and likes to "laugh and be merry." He prefers to keep mov ing and doing things, tends to be aggressive and lose his temper quickly; altogether his feelings are not kept under tight control, and he is not always a reliable person. The typical introvert is a quiet, retiring sort of person, introspective, fond of books rather than people; he is reserved and distant except to intimate friends. He tends to plan ahead, looks before he leaps, and distrusts the impulse of the moment. He does not like excitement, takes matters of everyday life with proper seriousness, and likes a well-ordered mode of life. He keeps his feelings under close con trol, seldom behaves in an aggressive manner, and does not lose his temper easily. He is reliable, somewhat pessimistic, and places a great value on ethical standards. (pp. 4-5) The Myers-Briggs Type Indicator (F) was selected as the instrument for assessing extraversion-introversion factors because it has satisfactory reliability and valid ity (Ross, 1963; Strieker and Ross, 1963; Strieker, Schiff- man and Ross, 1965). The administration of the test took about fifteen minutes and was given in classroom groups independently of and prior to the experimental sessions. Subjects selected for high and low extraversion participated in the first experimenta.1 session, to be described below, in order to establish an operant rate of non-reinforced eyeblink responses during a masking task, pressing a telegraph key, which was reinforced for that ' " 15 session. At the conclusion of the first session, subjects were carefully paired on the basis of their operant level of non-reinforced eyeblink responses, their extraversion- introversion scores, and sex. Until this time, subjects were treated alike. At the conclusion of the first ses sion, without their knowledge, the matched pairs were randomly divided into experimental and control groups. This procedure resulted in a 2 x 2 design with an N of 11 in each cell. Conditioning Procedure The conditioning process was carried out in an experimental room containing a chair at a table in front of a one-way mirror through which the experimenter observed subjects' eyeblinks from an adjoining room. A telegraph key, a counter, and (housed in panels) a light, a buzzer, and a latching release button switch, were all mounted on the table in front of the subject. Facing the subject across the table there was a white light covered by frosted glass on subject's left side, which when illuminated pre sented the printed instructions "Earn points" on the glass. A red light behind the glass, on subject's right side, presented when illuminated the instructions "Stop writing." 16 In addition, there was a blank numbered examination book let and a pencil on the table in front of subject. The subject wore a PRO-4 Professional headset with fluid filled ear cushions designed to fit the head contour; this provided a very efficient seal for almost complete attenuation of ambient noise. A continuous "white noise" from a tape recorder was fed over the earphones during the task to further mask out any possible sounds from the relay and control gear in the experimental room. At the first session each subject heard the follow ing instructions over an intercom: This is an experiment designed to investigate cor relations between human non-verbal performance and personality. There will be four different sessions on four different days. Your task in this experiment is twofold. To begin with while remaining seated at your desk throughout the experiment, you are to make as many points as possible on the counter in front of you to your right. These points will be converted into money at a penny a point at the end of the fourth session if you are selected to continue in this experi ment, and only then. When the counter moves, you will also see a green light become illuminated directly below the counter, and you will hear a buzzer. These three signals, the counter, the light, and the buzzer, indicate that you have scored a point worth one cent. As soon as this happens you are to immediately press the red button to the right of the telegraph key to terminate these signals so that we can test your reaction time. It also serves to reset the equipment to enable you to make more points. Remember, you are to make as many points as possible. 17 In order to signal you that you can start earning points, the white light behind the frosted glass in front of you will come on. When it does start earning points. As soon as the white light goes off, as it is off now, the equipment is inoperative and you then have one minute to write your thoughts about the experiment in the blue book on the table. In order to more cor rectly understand what you are doing in the experiment it is important for us to take into account your thoughts about the experiment. So we would like you to write out briefly, but clearly, any ideas that come to you that have any relation to the experiment. We're not interested in just any thoughts you might have— you need not dwell on personal matters— but we would like to get any thoughts that you have about the experiment. So, when the white light goes off as it is off now, that will be your signal to pause and write down your thoughts on the blue book in front of you. You will have one minute to do this. Now open your book and inspect it. If between one pause and the next, no thoughts occur to you, write the word "none" in the space provided for that pause. Each page has a differ ent number for each period of writing. Before you begin to write your thoughts, jot down the number of points from your counter onto the left side of the page, as provided. And now you may write your name on the front cover . . . (Time given for this) Now, put the pencil and booklet aside. Ten seconds before your minute for writing is up, there will be a flashing red light on the frosted glass in front of you. When this happens, put the pencil and booklet aside as you have just done and get ready to earn more points as soon as the white light comes on again. The white light must be on to score points. Now, you've been given quite a mouthful of instruc tions; is everything clear? (Questions were answered by re-reading pertinent information from instructions.) During the task, a low-level static sound will come over your earphones. It is meant to tune out any extraneous noises which may interfere with the task. All right, make yourself comfortable and wait for the white light to signal you that you can start making as many points as possible to earn money. Points can be made by making certain responses. 18 At the second session, subjects were given these brief instructions to insure that they understood the essen tial features of their task: Remember, we are investigating correlations between human non-verbal behavior and personality. Your task is to earn as many points as possible which will be converted into money at one penny for each point. When the counter moves and the green light comes on, you will hear a buzzer which you are to terminate immediate ly by pressing the red button so that we can test your reaction time. The white light behind the frosted glass signals that you can earn points. When it goes off, write your thoughts about the experiment in the blue book. Before you begin to write your thoughts, jot down the number of points from your counter onto the left side of the page. When a flashing red light illuminates the frosted glass, put the pencil and booklet aside and get ready to earn more points. Do you have any questions? . . . All right, make yourself comfortable and wait for the white light to signal you that you can start to earn points. Prior to Sessions 3 and 4, the subjects were merely asked if they had any questions about the experiment. The experimenter answered these questions over the intercom by reading pertinent information from the original instruc tions. At the beginning of Session 1, subjects were reinforced manually by the experimenter for pressing the telegraph key from a reinforcement schedule of continuous reinforcement (CRF) to a schedule of approximately fixed 19 ratio (FRIO), usually accomplished in about one to two minutes. At this point, the experimenter manually switched to the first programmed reinforcement schedule, TK/FR12. This period and all subsequent periods were automatically programmed for a five minute duration for conditioning fol lowed by one minute for writing. Excluding the initial brief telegraph key period, there were six task periods and six writing periods each session for four different days, providing for 24 task periods, and 24 writing periods (see Table 1). Ten seconds before the minute of writing was up, the flashing red light came on automatically to warn sub ject of the onset of the white light. All eyeblink responses were recorded by an experi enced experimenter whose reliability was checked out by two independent judges. Eyeblinks were recorded on a counter and event recorder whenever the experimenter pressed his key, while observing through the one-way mirror. All eye blinks and all telegraph key responses were recorded on separate counters and on different channels of the event recorder. Reinforcements were presented automatically by programmed relay equipment. The interval between a cri- terial eyeblink and its reinforcement was approximately 1/5 of a second. #_ °r_ Tim (Mi: 5 5 5 5 5 5 5 5 5 5 5 5 TABLE 1 SUBJECT DATA TALLY SHEET Schedule T.K. Response Count T.K. Reinforcement Count E.B. Response Count E.B. Reinforcement Count Session 1 Date_ Time TK/FR 12 TK/FR 24 TK/FR 48 TK/FR 48 TK/FR 144 TK/FR 144 Totals: TK/FR 48 TK/FR 144 EB/FR 3 EB/FR 3 EB/FR 3 EB/FR 3 Totals: Session 2 Date Time i 1 2 3 4 5 6 1 2 3 4 5 6 TABLE 1— Continued ^ Schedule T'K- espouse Beinforcement E'B' ResP°nse Reinforcement (Mm) Count Count Count Count Date Session 3 Time- 5 EB/FR 6 5 EB/FR 6 5 EB/FR 6 5 EB/FR 6 5 EB/FR 6 5 EB/FR 6 Totals: Date Session 4 Time- 5 EB/FR 12 5 EB/FR 12 5 EB/FR 12 5 EB/FR 12 5 EB/FR 12 5 EB/FR 12 Totals: 22 Session 1 established an estimated operant level of eyeblinks which was used for pairing subjects; subjects at this session were not reinforced for eyeblinks but for programmed schedules of telegraph key responding which were designed to produce very high rates of pressing. All in dividual sessions ran for six periods or 30 minutes of actual task performance. The first two five minute periods of the second session were scheduled so that telegraph key responses were still reinforced and eyeblinks were only recorded. This procedure was designed to maintain high telegraph key response rates. The six periods of Session 1 and the first two periods of Session 2 were employed to determine the operant rate of eyeblinks for all subjects. At this point the telegraph key became non-functional, and for the Experimental group, reinforcements now became contingent upon eyeblinks for gradually thinning reinforcement schedules ranging from EB/FR3 to EB/FR12. Each subject in the Control group was now reinforced on a schedule not con tingent on his own eyeblink rate, but matching the frequency and temporal distribution of reinforcements earlier given for eyeblinks to the subject who was his or her experimental yoke. Thus the matched control subject was reinforced 23 whenever the event recorder indicated that the Experimental subject had been reinforced. The reinforcement schedules for the Experimental group are shown in Table 1. Post-Experimental Questionnaire Immediately after the conclusion of the final ses sion on the fourth day, each subject was given a written questionnaire (See Table 13, Appendix A) in an adjoining 2 room in order to assess verbal hypotheses. Each question, typed on a separate sheet of paper, was presented with minimum verbal interchange. After the subject wrote his answer on the same sheet, it was given to the experimenter who then presented the next question to the subject. At the conclusion of the questionnaire the subjects were paid for the points earned in the experiment and were requested not to divulge any information about the experi ment. None of the subjects stated when questioned that he had received any prior information of the experimental procedures. The experimenter who administered the post- experimental questionnaire was the experimenter's wife, Mrs. Shirley Kessler. Hypotheses 1. It was hypothesized that the sex variable would not significantly affect eyeblink performance gains. 2. Because of conflicting evidence concerning the relationship of extraversion-introversion with conditioning (Eysenck, 1947, 1957; Field and Brengelmann, 1961; Franks, 1956; Gelfand and Winder, 1961; Sweetbaum, 1963) it was predicted that extraversion-introversion scores would not significantly produce an effect on conditioning of eye blinks . 3. Consistent with empirical studies of operant conditioning (Ferster and" Skinner, 1957; Kimble, 1961) it was hypothesized that the response-reinforcement contingent Experimental group would produce significantly more eye blinks than the non-contingent reinforced yoked Control group, and that this performance gain would increase over sessions. 4. Based on the results of investigations utiliz ing a verbal operant conditioning paradigm it was predicted that significant performance gains in eyeblinks would not occur for the Experimental subjects unaware of a correct or correlated contingency; that is, learning without aware ness would not be demonstrated. It was further 25 hypothesized that Aware subjects would not show performance gains prior to the time at which they first wrote a correct or correlated hypothesis, thus disconfinning the learning without awareness principle. 5. Based on verbal conditioning literature (DeNike, 1965) it was predicted that, for aware subjects, desire for the reinforcement would be related to awareness or perform ance gains. 6. Cognitive theory suggests that awareness of the ineffectiveness of the telegraph key response would not necessarily imply awareness of the correct response, or vice versa, and thus it was hypothesized that there would be no significant difference in the telegraph key responses (masking task) between the Aware and Unaware groups, indi cating that this manifest task cannot adequately predict awareness or unawareness of the reinforced response. 7. Following the above rationale it was further predicted that belief or disbelief in the efficacy of the masking task would not be significantly related to aware ness of the correct contingency. CHAPTER III RESULTS The sequence of the chapter on results is as fol lows: the first section reviews the findings on condition ing of the eyeblink response considering reinforcement, extraversion-introversion, and sex as the independent vari ables, and eyeblink as the dependent variable. The second section reports the incidence of awareness, the relationship between awareness and conditioning, and the relationship between motivation for the reinforcement and awareness. Section three reports the relations between the reinforced eyeblink response and the non-reinforced telegraph key response, and between the telegraph key response and aware ness. This section also reports the effect of belief or disbelief in the efficacy of telegraph key responding behavior in unaware subjects. The important independent variables under investi gation were: (1) the experimental treatment (contingent reinforcement) versus the yoked control treatment (non contingent reinforcement), (2) the extraversion-introversion 26 : 27 dimension of personality, (3) male versus female. Derived independent (possibly mediating) variables considered were: (4) awareness of a correct or correlated response-reinforcement contingency, (5) the desire of a subject to receive the reinforcement. The main dependent variable was the mean number of eyeblinks per minute per subject for each session. The secondary dependent variable was the mean number of tele graph key responses per minute per subject for each session, i.e., performance on the masking task that was employed to detect a relationship between awareness of the correct response class (eyeblink) and the frequency of a specific incorrect response class (telegraph key). Conditioning Reliability of eyeblink recording. Conditioning was assessed from eyeblink counts taken during each task session period. The reliability of the experimenter's count of eyeblinks was determined by correlating the experimenter's count with a judge's independent count for 1 arbitrarily chosen sessions of two subjects. The ■^The two judges, each of whom scored eyeblinks for one subject only, were Mrs. Shirley Kessler and the dissertation supervisor. 28 product-moment correlation for the eyeblink counts by the experimenter and judge 1 was .86 and for the experimenter and judge 2 it was .91. In addition the percentage of variation of interjudge estimate of eyeblinks was calcu lated from the same data and presented in Table 2. The range of variation varied from 0 to 12 per cent. The aver age percentage variation between the experimenter and judge 1 was 5 per cent and between the experimenter and judge 2 it was 3 per cent. It was concluded that the experimenter's reliability for counting eyeblinks was satis factory. Control for eyeblink operant level. Prior to the analysis of the conditioning data a simple analysis of variance was done on the mean number of eyeblinks per minute during the operant level sessions of the Experimental and yoked Control groups. The over-all means for the Experimental and Control groups respectively were 15.87 and 16.50 eyeblinks per minute. These means did not differ statistically, F (1,42) = 0.10, indicating that the Experi mental and Control groups were very closely matched with respect to their initial unreinforced eyeblink responses. Sex. The data from the combined Experimental and 29 TABLE 2 MEAN EYEBLINKS PER FIVE MINUTE PERIOD FOR TWO ARBITRARILY CHOSEN SUBJECTS (#19 and #60) Periods Experimenter Judge 3 Percentage of Variation of Interjudge Estimate Subject #19 Session 4 23 12% 25 4% 27 0% 26 4% 24 8% 23 0% Average 5% Percentage of Variation Periods Experimenter Judge 2 pf Interjudge Estlmate Subject #60 Session 3 1 44 46 5% 2 57 58 2% 3 71 72 1% 4 69 67 3% 5 72 68 6% 6 74 74 Average 0% 3% 1 26 2 26 3 27 4 25 5 26 6 23 30 Control groups were analyzed in a Lindquist (1953) type I design with sex as the between-subjects variable and ses sions as within-subjects variable. The sex classification did not produce a significant F ratio, indicating that there was essentially no difference in performance between males and females over sessions, F (1,42) = 0.77. On the basis of this finding (see Table 3), sex was dismissed as a classification variable and the data were collapsed across sex for all subsequent analyses. Extraversion-introversion matching. A simple analysis of variance was administered on the extraversion- introversion scores obtained from the subjects' performance on the Myers-Briggs Type Indicator, Form F (Myers, 1962) to evaluate the matching of subjects on this variable into Experimental and Control groups. The range for the Experi mental group was 49 to 143, and for the Control group 39 to 147. The mean for the Experimental group was 100.73, for the Control group 101.09. The results of the analysis of variance, reveal that there were no significant differences between groups for extraversion-introversion scores, F (1,42) = 0.0 01. It was concluded that the Experimental and Control groups were 31 TABLE 3 SUMMARY OF ANALYSIS OF VARIANCE OF THE EYEBLINK PERFORMANCE DATA FOR MALES AND FEMALES (EXPERIMENTAL AND CONTROL GROUPS COMBINED) Source df MS F Between Subjects 43 Sex (B) 1 241.86 n. s. Error (between) 42 13,223.89 Within Subjects 88 5,130.42 Sessions (A) 2 442.84 4.15* A x B 2 206.03 n. s. Error (within) 84 4,481.55 Total 131 < .05 32 very adequately matched on extraversion-introversion scores. Extraversion-introversion analysis. A modified three-way analysis of variance (Lindquist, 1953, Type III) was performed with reinforcement and extraversion-introver sion as the between-subjects variable and with sessions as the within-subjects variable. The F for the reinforcement by extraversion-introversion effect was not significant, F (1,40) = 0.31, indicating that conditioning was not a function of the extraversion-introversion dimension of personality. The nonsignificant main effect of extraver sion-introversion, F (1,40) = 0.11, also supports the con clusion that performance was unrelated to scores on the Myers-Briggs Type Indicator, Form F (Myers, 1962). Since extraversion-introversion in this analysis did not relate to the conditioning process, the data were combined across this classification for all later analyses. Reinforcement effects. The modified three-way analysis of variance referred to above yielded significant effects for Reinforcement, F (1,40) = 9.39, £ < .01 and for Sessions, F (2,80) = 4.06, £ < .05 (See Figure 1 and Table 4), thus indicating that the contingently reinforced Experimental group produced more eyeblinks than the non- MEAN EYEBLINKS PER MINUTE 33 32 EXPERIMENTAL HIGH EXTRAVERTS 30 28 EXPERIMENTAL LOW EXTRAVERTS^ 26 24 22 20 CONTROL LOW EXTRAVERTS fir— - CONTROL HIGH EXTRAVERTS 3 4 OP 2 TREATMENT SESSIONS Figure 1. Eyeblink performance in relation to reinforcement conditions and extraversion- introversion . 34 TABLE 4 SUMMARY OF MODIFIED THREE WAY ANALYSIS OF VARIANCE OF THE EYEBLINK PERFORMANCE DATA FOR THE EXPERIMENTAL AND CONTROL GROUPS Source df MS F Between Subjects 43 Reinforcement (B) 1 2,538.82 9.39** Extraversion-introversion (C) 1 30.35 n. s. B x C 1 84.63 n. s. Error (between) 40 10,811.95 Within Subjects 88 Sessions (A) 2 221.42 4.06* A x B 2 141.95 n.s. A x C 2 17.32 n.s. A x B x C 2 3.32 n.s. Error (within) 80 54.53 Total 131 *p < .05 **£ < .01 35 contingently reinforced yoked Control group and that the performance difference between the groups increased over sessions. These results indicate that significant condi tioning occurred. In separate tests of trend (Winer, 1962, pp. 71-73) for the Experimental and Control groups, the linear trend was highly significant for the Experimental group, F (1,63) = 28.83, £ < .001, but not for the Control group, F (1,63) = 11.98 as presented in Tables 5 and 6 respectively. Thus, only the Experimental subjects significantly increased their reinforced eyeblink responses and, as a group, condi tioned; their yoked controls did not. The Incidence of Awareness Awareness ratings. Subjects' awareness was deter mined both from their written reports in the "thoughts about the experiment" (notes) written during the experiment, and from the responses which they wrote to the post- experimental questionnaire. Basis for raters' judgments. The subjects were rated separately on the basis of their notes and on their answers in the post-experimental questionnaire. A subject 36 TABLE 5 SUMMARY OF ANALYSIS OF VARIANCE AND LINEAR TREND TEST OF THE EYEBLINK PERFORMANCE DATA FOR THE EXPERIMENTAL GROUP Source df MS F Sessions (B) 3 778.51 9 .70*** Subjects (A) 21 A x B 63 80.24 Total 87 Linear Trend 1 2,313.36 28.83*** ***£ < .001 37 TABLE 6 SUMMARY OF ANALYSIS OF VARIANCE AND LINEAR TREND TEST OF THE EYEBLINK PERFORMANCE DATA FOR THE CONTROL GROUP Source df MS F Sessions (B) 3 4.35 n.s. Subjects (A) 21 A x B 63 8.26 Total 87 Linear Trend 1 11.98 n.s. 38 was rated aware of a correct response-reinforcement contin gency if he wrote a statement relating eyeblink response to reinforcement. Subjects were rated correlated-aware if they recorded a hypothesis involving body movements which result in reflexive eyeblinking. Observations during pilot work suggested that the following body movements increased the occurrence of eyeblinks: head movements, bringing a hand to the face, yawning, relaxing, and picking up objects on the table. Subjects who expressed no hypotheses except relating to the telegraph key, or wrote that reinforcement came for body movements not regularly associated with the reinforcement, or abandoned their hypotheses even if prom ising, or expressed only passing thoughts or notions with little confidence, were rated unaware. For marginal cases, the raters were instructed to err on the side of rating a subject correlated aware rather than unaware. Reliability of awareness ratings. Two raters independently rated the subjects in the Experimental and the yoked Control groups without any knowledge of their reinforcement group (Experimental or Control) or of their performance. Although the yoked Control group subjects were rated for awareness to determine the adventitious 39 occurrence of correct hypotheses, they were not included in the reliability figures cited below since their rein forcement was non-contingent upon their performance. Agreement between the two raters established a rating of awareness. Where there was a disagreement between the raters, one judging a subject aware and the other rater judging that subject unaware, a third rater resolved the 2 disparity. All of the raters had had previous experience in evaluating awareness in verbal conditioning. With regard to ratings of awareness, on both the notes and the post-experimental questionnaires there was unanimous agreement between the two raters. Only one sub ject of the 22 Experimental group subjects was aware and he was so judged by both raters. With regard to ratings of correlated-aware, on the basis of the notes, only one subject was so rated (by two of the three raters). No other judgments of correlated- aware were made on the basis of subjects' notes. Thus it appeared that there was satisfactory reliability between raters for awareness based on the notes. ^The writer is indebted to Arthur H. Weingaertner and the dissertation supervisor for their ratings of the notes and post-experimental questionnaires, and to Morris P. Leibovitz who was the third rater. 40 The ratings of correlated-aware from responses in the post-experimental questionnaires were not as clear cut. One rater judged that three subjects stated correlated hypotheses; the other rater singled out eight such subjects, two of which had been so judged by the first rater. A total of 13 out of 21 subjects (62 per cent) were judged unanimously to have been unaware. The third rater con cluded that four out of the six disparate ratings were correlated-aware, resulting in six subjects judged as having correlated hypotheses. For statistical purposes, the single aware subject was combined with the six corre lated aware subjects to result in a group of seven subjects, hereafter termed the "aware subjects," and 15 unaware, based on the post-experimental questionnaires. Subjects’ correlated hypotheses were generally uncertain and vague. Only one subject was unanimously agreed upon as aware or correlated aware on the basis of the notes. Only three subjects were unanimously rated out of seven considered aware or correlated aware on the basis of the post-experimental questionnaires. Thus it appeared that the 15 subjects judged unaware on the basis of the post-experimental questionnaire were truly such, and that if significant performance gains were found for them, it 41 would constitute very strong evidence for learning without awareness. Thus, in order to provide the most stringent test possible for the learning without awareness hypothesis, the ratings (seven aware, 15 unaware) from the post-experi mental questionnaire were adopted for all later analyses. The Relationship between Awareness and Conditioning Conditioning of Aware, Unaware, and Control groups. Following cognitive theorists' methodology, the influence of awareness as an inferred variable mediating eyeblink performance was evaluated by dividing the Experimental group into Aware (including Correlated Aware) and Unaware groups. Table 7 shows the mean number of eyeblinks per minute given by the aware and unaware subjects for each of the four sessions. An analysis of variance performed on the operant rate data for these two groups showed that they differed in operant rate, F (1,20) = 4.81, p < .05. Since the Aware group had a mean operant rate of 11.7 while the Unaware group1s mean operant blink rate was 17.8, the mean difference between the two groups, 6.1 was added to each Aware subject's mean eyeblink score for Sessions 2, 3, 42 TABLE 7 MEAN EYEBLINK PERFORMANCE PER MINUTE FOR AWARE, CORRELATED AWARE, AND UNAWARE SUBJECTS OF EXPERIMENTAL GROUP Subject O • • Session 2 Session 3 Session 4 Aware and Correlated Aware 7 5.1 7.5 11.5 18.1 8 9.9 10.4 26.8 52.1 9 17.1 19.0 33.2 95.9 13 13.5 25.6 20.5 31.2 42 11.2 21.0 21.5 19.4 57 14.5 20.6 33.8 31.9 60 10.7 9.8 12.9 13.3 Mean 11.7 16.3 22.9 37.4 Unaware 4 6.9 13.3 9.7 12.4 14 13.4 19.1 21.0 17.4 22 20.5 22.9 28.9 33.3 24 17.3 15.0 11.9 13.3 25 23.5 32.5 32.8 35.1 26 27.1 34.5 37.2 38.2 32 16.0 17.9 23.5 20.9 38 24.7 34.4 34.4 36.7 43 11.1 17.0 16.7 13.7 49 5.8 9.3 15.3 17.2 53 16.3 21.7 29.5 27.6 58 13.5 25.6 20.5 31.2 63 25.5 49.6 56.5 50.8 64 16.3 21.4 22.6 22.1 65 26.7 27.4 32.4 30.7 Mean 17.6 24.1 26.2 26.7 and 4. An analysis of variance (Type I) performed on the data with the adjusted means disclosed no significant dif ference between the conditioning performance of the Aware and Unaware groups, F (1,20) = 1.18, see Table 8. The fact that there was no significant difference between the Aware and Unaware subjects suggests that awareness did not markedly facilitate eyeblink performance under reinforce ment. However, since the performance of the Aware group started at a lower operant rate than the Unaware group and at the last session exceeded that of the Unaware subjects, it is evident that its slope is steeper than the slope for the Unaware group. In fact, the interaction F was signifi cant, F (2,40) = 6.49, £ < .01, indicating that the seven Aware subjects acquired performance gains at a significantly faster rate. Thus some facilitating effect of awareness on performance was found. In a test of trend (Winer, 1962), pp. 71-73) for the Unaware subjects, the linear trend was highly signifi cant, F (1,42) = 36.91, £ < .001, (See Table 9). This finding, probably the most important in this study, demon strates that significant learning took place without aware ness in this experimental paradigm (See Figure 2). The Control group's linear trend was found earlier to be 44 TABLE 8 SUMMARY OF ANALYSIS OF VARIANCE OF THE EYEBLINK PERFORMANCE DATA FOR THE EXPERIMENTAL AWARE AND EXPERIMENTAL UNAWARE GROUPS WITH MEANS ADJUSTED FOR OPERANT RATE DIFFERENCES Source df MS F Between Subjects 21 Groups (B) 1 469.76 n. s. Error (between) 20 398.03 Within Subjects 44 Sessions (A) 2 358.93 4.70* A x B 2 495.33 6.49** Error (within) 40 76.25 Total 65 *jd < .05 **£ < .01 45 TABLE 9 SUMMARY OF ANALYSIS OF VARIANCE AND LINEAR TREND TEST OF THE EYEBLINK PERFORMANCE DATA FOR THE EXPERIMENTAL UNAWARE GROUP Source df MS F Sessions (B) 3 268.14 16.60*** Subjects (A) 14 A x B 42 16.15 Total 59 Linear Trend 1 596.15 36.91*** ***£ < .001 MEAN EYEBLINKS PER MINUTE 32 EXPERIMENTAL HIGH EXTRAVERTS 30 28 EXPERIMENTAL ' LOW EXTRAVERTS^ 26 24 22 20 CONTROL LOW EXTRAVERTS JB------® CONTROL HIGH EXTRAVERTS 2 3 4 OP TREATMENT SESSIONS Figure 2. Eyeblink performance in relation to subjects' awareness of correct or correlated response-reinforcement hypotheses. insignificant (see Table 6). It may be concluded that while awareness seems to have facilitated learning, learning did take place without awareness. Temporal relationship between performance gains and awareness. One of the experimental goals was to determine whether Aware subjects began to show performance gains at the time when they first wrote a correct or correlated hypothesis (DeNike, 1963). However, because of the small number of subjects who were assessed as aware, this analy sis was not carried out. It is hoped that a future study involving a greater population would provide data which would be useful for such an analysis. Motivation for the reinforcement and awareness. In order to investigate the relation of desire for the rein forcement and awareness, the information provided by Question 11 of the post-experimental questionnaire was examined. This question required subjects to answer either "very much," "some," or "didn't care" with regard to how much they wanted to score points for money. To facilitate statistical analysis, the "some" and "didn't care" responses were combined into one category resulting in 11 experimental 48 subjects who were in the "very much" category and 11 in the "some" and "didn't care" category. A chi-square test per formed on motivation and awareness demonstrated that desire for the reinforcement was unrelated to awareness. To investigate whether motivation may have affected performance in the absence of awareness, the Unaware "some" and "didn't care" subjects were compared with the Unaware "very much" subjects. A simple analysis of variance on operant eyeblink rates for the two Unaware groups with different levels of motivation was significant. The "very much" group had a higher operant eyeblink rate (t < .001). The eyeblink means for Sessions 2, 3, and 4 were adjusted to compensate for operant disparities and a Type I analysis of variance was performed. As the summary of the data in Table 10 indicates, there was no significant group effect, F (1,13) = 1.22, nor interaction effect, F (2,26) = 0.09. In sum, it was concluded that motivation was unre lated to both awareness and conditioning performance. From a cognitive point of view, motivation would not have been expected to relate, to performance in the absence of awareness. However, the higher operant rate of the "very much" group suggests that, prior to reinforcement, these subjects' high motivation may have been accompanied by a 49 TABLE 10 SUMMARY OF ANALYSIS OF VARIANCE OF THE EYEBLINK PERFORMANCE OF UNAWARE SUBJECTS AS A FUNCTION OF THE EXTENT TO WHICH THEY WANTED THE REINFORCEMENT Source df MS F Between Subjects 14 Groups (B) 1 379.25 n. s. Error (between) 13 308.38 Within Subjects 30 Sessions (A) 2 35.23 4.01* A x B 2 .83 n. s. Error (within) 26 8.78 Total 44 *E < *05 50 general activation yielding higher eyeblink rates. As for the awares, there were too few subjects to permit analysis. The Telegraph Key Masking Task Relation between reinforced eyeblink responses and non-reinforced telegraph key responses. The non-reinforced telegraph key response data were analyzed in relation to the eyeblink data. Keehn et al. (1965) had reported that four subjects conditioned in reinforced eyeblink responses while maintaining a high rate of telegraph key responding. This relationship for Keehn et al. was taken as evidence for learning without awareness. That is, those investiga tors assumed that a subject who was unaware of the eyeblink contingency would not drop in rate of telegraph key responding. In the present experiment, the telegraph key responses of the seven aware subjects were compared with those of the 15 unaware subjects for Sessions 2 (periods 3 through 6) through 4 inclusive. Telegraph key responses for Session 1 and the first two periods of Session 2 were not analyzed because reinforcement was given for telegraph key responses during that session. An analysis of vari ance, Type I (Lindquist, 1953), as summarized in Table 11 TABLE 11 SUMMARY OF ANALYSIS OF VARIANCE OF THE TELEGRAPH KEY PERFORMANCE DATA FOR THE EXPERIMENTAL AWARE AND EXPERIMENTAL UNAWARE GROUPS Source df MS F Between Subjects 21 Groups (B) 1 7 f 373.6 n. s. Error (between) 20 19,303.0 Within Subjects 44 3,685.4 Sessions (A) 2 2,353.3 n. s. A x B 2 2,385.1 n. s. Error (within) 40 3,807.0 Total 65 n 52 shows no significant differences for the group, session, or interaction effects on telegraph key performance. The data disclosed that the eyeblink performance for the Unaware group increased approximately 9 per cent between Session 2 and Session 4 while the telegraph key responses decreased only one per cent, i.e., were essen tially constant. However, the eyeblink performance for the Aware group increased 129 per cent between Session 2 and Session 4 while the telegraph key performance decreased 23 per cent. These results suggest that where subjects were aware of the correct response there was decreasing perform ance of an incorrect response. Learning with awareness of the correct response-reinforcement contingency was not manifested in Keehn's study. However, the present results are consistent with the results of Keehn et al. and tend to support their assumptions regarding learning with and with out awareness. The Aware group had six out of seven subjects who increased their eyeblink responses between these sessions (86 per cent); the Unaware group had 10 out of 15 (67 per cent). The Aware group had five out of seven subjects who decreased their telegraph key responding (71 per cent); for the Unaware group it was six out of 15 (40 per cent). 53 Again there is strong suggestion that awareness of rein forcement contingencies facilitates the elimination of unreinforced responses. Figure 3 indicates the relationship between the eyeblink and telegraph key responses for the aware and unaware subjects for the three sessions where the eyeblink response was reinforced and telegraph key responding was not reinforced. Belief or disbelief in telegraph key responding in unaware subjects. The post-experimental questionnaires were examined for a relationship between belief or dis belief in reinforcement for telegraph key responding and 3 telegraph key response rate. Two judges agreed on 13 of 15 unaware subjects. Of these, three were judged to have expressed disbelief in the efficacy of the responding. All three of these subjects practically ceased their rate of telegraph key responding, decreasing far more than did any other unaware subject (see Table 12). Thus, these subjects were aware that the telegraph key response was ineffective even though they were unaware The judges were the experimenter and the disserta tion supervisor. MEAN REINFORCED EYEBLINKS P E R MINUTE 38 ' EXP. AWARE \ (Telegraph Key) 260 34 EXP. AWARE (Eyeblink) 30 26 240 EXP. UNAWARE (Eyeblink) 22 230 220 14 - EXP. UNAWARE (Telegraph Key) 210 200 2 3 4 TR EA TM ENT SESSIONS Figure 3. Relationship between reinforced eyeblink and non-reinforced telegraph key perform ance for experimental aware and unaware subjects. MEAN TELEGRAPH K E Y RESPONSES P E R MINUTE 55 TABLE 12 MEAN CHANGE IN EYEBLINK PERFORMANCE UNDER REINFORCEMENT AND TELEGRAPH KEY RESPONSE UNDER NON-REINFORCEMENT (SESSION 4 MINUS SESSION 2) FOR EXPERIMENTAL SUBJECTS IN RELATION TO AWARENESS, DESIRE TO EARN POINTS, AND BELIEF OR DISBELIEF IN EFFICACY OF THE TELEGRAPH KEY RESPONSE Subjects Belief or Eyeblink Disbelief (Sess. 4-Sess. 2) Telegraph Key (Sess. 4-Sess. 2) Aware, "Very Much" Desire 9 B 10.6 - 14.9 42 B - 1.6 - 90.5 Aware, "Some Desire" and "Didn't Care" 8 B 41.7 - 73.2 13 B 5.6 - 14.5 15 D 76.9 -179.4 57 B 11.3 7.2 60 D 3.5 21.0 Unaware, "Very Much" Desire 4 B - 0.9 39.7 14 D - 1.7 -193.5 22 B 10.4 73.8 25 D 2.6 74.6 26 B 3.7 79.6 38 B 2.3 8.1 58 D - 3.7 -142.0 63 B 1.2 223.5 64 B 0.7 - 22.3 Unaware, "Some" Desire and "Didn't Care" 24 B - 1.7 46.7 32 D 3.0 67.7 43 D - 3.3 -333.3 49 B 7.9 47.5 53 B 5.9 - 1.9 65 B 3.9 - 19.0 56 of the correct contingency. This finding appears to pro vide a way of usefully adding to the approach of Keehn et al. for investigating awareness. That is, a drop in tele graph key responding, even when accompanied by a gain in eyeblink performance, probably cannot be interpreted as adequate evidence of awareness of the correct contingency. Interestingly, these same three subjects showed a slight drop in the mean of their eyeblink response between Session 2 and Session 4. Thus this study provided no unaware sub jects who met the awareness criterion of Keehn et al. However, it is likely that performance measures are but imperfectly correlated with awareness (Irwin et al., 1934). Table 12 summarizes the data for this analysis. Resumd of Hypotheses and Results 1. The null hypothesis that the sex variable would not significantly affect eyeblink conditioning was consist ent with the data. 2. The null hypothesis that extraversion-introver sion scores would not significantly affect eyeblink per formance gains was also borne out. 3. The hypothesis that operant conditioning of eyeblinks would occur was supported. 57 4. The hypothesis that the unaware subjects would not show significant eyeblink performance gains was not supported. Learning without awareness was demonstrated. The additional hypothesis that aware subjects would not show performance gains prior to the time at which they first wrote a correct or correlated hypothesis was not investi gated because of the small sample of aware subjects avail able and because learning without awareness had been adequately demonstrated. 5. The hypothesis that desire for the reinforce ment would be related to awareness or performance gains was not confirmed. 6. The hypothesis that there would be no signifi cant difference in telegraph key responses between the Aware and Unaware groups was supported. This would indi cate that this masking task could not adequately predict awareness or unawareness of the reinforced response. 7. The hypothesis that belief or disbelief in the efficacy of the masking task response would not be related to awareness of the correct contingency was confirmed. CHAPTER IV DISCUSSION Learning and awareness. The results of the present investigation support findings (Keehn et al., 1965) that reinforcement increases rate of eyeblink responses. Al though note writing provided opportunities for subjects to express hypotheses and indeed probably increased the like lihood of developing awareness, and detailed post-experi mental questionnaires assured that awareness once developed would be detected, nevertheless there was a substantial Unaware group which showed increased performance gains and yet failed to report a correct or correlated response- reinforcement contingency. It is highly improbable that aware subjects in this experiment were not detected. Great efforts were made to use methodological procedures and controls to eliminate artifactual influences and aware subjects. There may have been a possible bias toward rating too many people aware, since the raters were instructed to rate correlated aware instead of unaware in marginal cases. Thus, very exacting 58 59 conditions were established to test the possibility that learning could occur without awareness. In order to comply with cognitivist methodology Keehn's procedures were refined. A yoked Control group was carefully matched, subject by subject, on operant rate of eyeblink with the Experimental group to guard against operant or time-correlated differences being mistakenly interpreted as conditioning effects. The results support Thorndike's hypothesis that the action of the reinforcer is direct and not mediated by cognitive processes; that this particular response was learned in the majority of subjects without awareness. All seven aware subjects increased their rate of eyeblink from their operant level to Session 4. This would suggest that awareness facilitated learning. Fourteen of the 15 unaware subjects increased their eyeblink performance during con ditioning. An examination of the original data tally sheet for the only subject (#24) who did not condition reveals that he engaged in "deep concentration" behavior, e.g., his eyes were fixated on one spot and his body was almost motionless, for most of the conditioning periods of Ses sions 2 and 3. This result was consistent with empirical evidence from the pilot work and the- finding by Walter 60 (1941) in that having a point of fixation inhibits the blinking rate. If this analysis would account for this subject's substantial decrease in eyeblink performance, then it might be appropriately noted that performance gains are automatic for conditioning subjects who do not for idiosyncratic reasons adopt inhibitory sets. Frustration and performance. After analyzing the conditioning results, it occurred to the author to inves tigate whether a frustration effect might have been respons ible for any increases in eyeblink response when the reinforcement contingency during Session 2 was abruptly changed from FR144/TK to FR3/EB. The mean number of rein forcements for these two periods for the Experimental group were compared. The mean reinforcement rates for the Experimental group were 8.7 for FR144/TK and 32.6 for FR3/EB. Since rate of reinforcement increased greatly, obviously it would be difficult to posit frustration to account for the observed increase in eyeblink rate. General activity and performance. In line with the above, an analysis was made to ascertain whether general activity, and hence eyeblinking, may have been stimulated by the change in contingencies. A comparison of eyeblink 61 mean rates for the Control group for the same two periods showed: FR144/-TK = 17.4, FR3/EB = 17.3. It is apparent that the eyeblink rate for the TK/FR144 period was very close to the EB/FR3 period. It is obvious that the change in rate of reinforcement was not paralleled by an increase in the critical response class; thus general activity can not be invoked to explain the obtained conditioning effect. Sex and performance. As indicated previously, sex had no significant relationship to eyeblink conditioning. Females had a 58 per cent increase in eyeblink rate from operant level to Session 4 (16.2 to 25.6); males had a 115 per cent increase (15.7 to 33.8) for the same comparison; however this difference was not significant. This differ ence was largely accounted for by the single (male) subject who verbalized the correct contingency (see Figure 1). Extraversion-introversion and performance. There was no evidence that extraversion-introversion factors influenced performance gains in this experiment. Actually, a very slightly higher mean performance was registered by the extrovert group rather than, as Eysenck (1947) and Franks (1956) would predict, the introvert group. However, those investigators' assessment of extraversion-introversion 62 was not based on the Myers-Briggs instrument used in this study, and this may account for the differing results. A correlation between the Myers-Briggs Type Indicator (F), (Myers, 1962) and the Eysenck Personality Inventory (Eysenck and Eysenck, 1963) produced a product-moment correlation of .66. Functional and non-functional responses and aware ness. The operational definition of unawareness by Keehn et al. is a performance gain in the reinforced eyeblink response and a maintained level of performance in the non- reinforced telegraph key response. Applying these criteria to this study should make it possible to predict which experimental subjects were unaware. It could be inferred from Keehn's definition of awareness that a rise in eye blink response and a drop in telegraph key responding would be adequate evidence of awareness. Subject #13's mean operant eyeblink of 13.5 per minute increased to 31.2 on Session 4; his mean telegraph key responses for these periods were 338.4 and 319.5, thus he essentially maintained his non-functional telegraph key responding. However, he reported in the post-experimental questionnaire, "Sometimes bringing the hand to the face 63 would bring a reward." This subject's eyeblink and tele graph key responding would have met Keehn's criteria of unawareness. The subject, was however, rated as correlated aware since in pilot work bringing a hand to the face was empirically associated with blinking. Subject #57 increased her mean operant eyeblink rate from 14.5 per minute to 31.0 in Session 4; her tele graph key responses increased from 276.2 per minute in Session 1 to 310.9 in Session 4. This subject also would have met Keehn's criteria of unawareness. In her post- experimental questionnaire she reported, "It seemed when I looked up I was rewarded." And again, "By looking up or around the room." But since these behaviors are a priori correlated with eyeblinking, this subject was also rated correlated aware. Moreover, some yoked Control subjects who could not possibly be truly aware, might be so judged by Keehn. For example, subject #10's mean operant eyeblink rose from 12.3 to 18.7 per minute in Session 4; his telegraph key responses dropped from 307.7 to 243.4. Subject #46's eye blink increase was from 11.5 to 20.6; for the telegraph key, the decrease was from 319.6 to 194.2. Finally, sub ject #23's eyeblinks rose from 18.8 to 28.6 while telegraph 64 key responses dropped from 250.3 to 12.7! It would appear that the performance-bound criteria of Keehn et ai. cannot make fine distinctions between awareness and unawareness. A drop in telegraph key re sponding does not necessarily mean that a subject is aware of the correct or correlated contingencies; he may just know that the telegraph key has nothing to do with the reinforcement. Similarly a subject who is aware of the correct or correlated contingency may or may not become aware that the telegraph key is ineffective, he may believe that both responses— eyeblink and telegraph key— are neces sary for reinforcement. Experimental subject #9, for exam ple, increased his eyeblink rate from 5.1 to 18.1 while his telegraph key responses increased from 261.9 to 279.3. He had been judged correlated aware because of his report of eyeblink-correlated hypotheses; he wrote that key tapping plus frustration was the main part of the experiment. Note writing and post-experimental assessment of awareness. Comparing the note writing and the post- experimental questionnaires, it appeared that the question naires were more sensitive in detecting awareness. From the notes only one subject was rated aware and one subject 65 was affirmed to be correlated aware by the deciding judge. The post-experimental questionnaires, however, provided two additional subjects rated aware, and four others arbitrated to be aware by the third judge. The increased frequency of detected awareness from the questionnaires does not neces sarily mean that the questionnaires are more accurate. The questionnaire may simply be too sensitive, and thus make unawares seem aware. It is somewhat likely that two apparently aware subjects were really unaware, since they maintained their frequency of telegraph key responding. Awareness and verbal, voluntary motor, and autonomic conditioning. The firmest evidence has been presented (DeNike, 1964; Dulany, 1962; Spielberger, 1962; Spielberger and DeNike, 1966; Spielberger, Levin, and Shepard, 1962) that verbal operant conditioning is cognitively mediated. Results reported by Hefferline and Keenan (1963), Heffer- line, Keenan, and Harford (1959), Keehn (1965, 1967), Keehn et al. (1965), and Sasmor (1966) all using Skinnerian paradigms indicate that operant conditioning of a skeletal- muscle response occurs without awareness. The present study using a larger subject population, a matched control group, and sophisticated awareness assessment methods, 66 generally confirmed the findings of Keehn et al. It appears then that operant conditioning of a skeletal-muscle response, frequent enough in operant occurrence, can take place without awareness. With increasing frequency, experimental investiga tions are being reported of operant conditioning of auto- nomically mediated responses, e.g., Fowler and Kimmel, 1962; Greene, 1966; Harwood, 1962; Kimmel and Hill, 1960; Shearn, 1960. This evidence has accumulated despite Kimble's earlier opinion (1961) that operant conditioning of smooth muscle and glandular responses cannot be modified by instrumental techniques. Little effort has been made as yet to determine whether operant conditioning of autonomically mediated responses requires cognitive mediation. However, evidence that classical conditioning of autonomic responses took place only with verbalized awareness has been reported by Dawson (1967) and Dawson and Grings (in press) . Investigations of conditioning have often been con taminated by cognitive influences. Classical eyelid conditioning, for example, had been thought to be free of cognitive effects; more recent evidence (Spence, 1966) clearly contradicts this assumption. The problem now becomes one of eliminating the cognitive contamination and thus producing "purer" unconscious learning studies, or else of studying conscious learning with the aware subjects DeNike, Dulany, and Spielberger have provided methods which discriminate between aware and unaware subjects in verbal operant conditioning, but the unaware subjects show no learning, pure or impure. Hefferline and Keehn have done the pioneer work which has given rise to seemingly valid paradigms of operant learning of smal1-magnitude skeletal-muscle responses without awareness. The present study may have contributed to this area by providing important methodological safeguards and controls. It should now be possible to use the operant eyeblink response with safeguards for the detection of cognitive processes, for the investigation of various parameters of learning without awareness. CHAPTER V SUMMARY There has been great controversy regarding the role of awareness in the mediation of human operant condition ing. Contrary to Thorndike's Law of Effect which states in part that such learning is direct and automatic, a great deal of evidence suggests that verbal operant conditioning is dependent upon the subject's awareness of the reinforce ment contingency. The very existence of operant condition ing without awareness in humans has been questioned. In addition, there has been disagreement about the role that the extraversion-introversion dimension of personality plays in the operant learning process. The main goal of this experiment was to investigate the relationships between human operant eyeblink condition ing and awareness; that is to say, whether or not there is learning without awareness for a simple motor activity frequent in base rate and essentially not monitored by the subject. 68 69 Cognitive theorists dispute the methodological procedures that stimulus-response theorists have used in their efforts to demonstrate learning without awareness in verbal operant conditioning. This study utilized extensive procedures designed to detect and control for cognitive influences on the learning process. Very exacting condi tions were established to test the possibility that learn ing could occur without awareness. The present investigation dealt with the relation ships among contingent (Experimental group) versus non contingent (Control group) reinforcement, awareness or correlated awareness of the response-reinforcement contin gency as indicated by subjects' written "thoughts about the experiment" during reinforcement, and by their responses to a post-experimental questionnaire. In addition, an analy sis was made of performance on a masking task (telegraph key pressing) as a possible non-verbal indicator of aware ness . The subjects were 14 female and 30 male college students. They were drawn from the upper and lower quar- tiles of a distribution of extraversion-introversion scores on a personality test. These subjects participated in an initial experimental session in order to establish their | 70 I : operant rate of non-reinforced eyeblink responses while being reinforced for telegraph key presses. At the con clusion of this session, subjects were carefully paired on the basis of their eyeblink operant levels, their extraver- 'sion-introversion scores, and sex, and then assigned to matched experimental and control groups. During the second session, the telegraph key became non-functional and, for the Experimental group, reinforcements then became contin gent upon eyeblinks for gradually thinning reinforcement schedules. Each subject in the Control group was reinforced f not for his own eyeblinks, but on a schedule matching the frequency and temporal distribution of reinforcements earlier given for eyeblinks to his reinforced Experimental yoke subject. Thus he was reinforced whenever the Experi mental subject was reinforced. Immediately following the fourth session, each subject was given a written question naire to assess his hypotheses of the response-reinforcement contingency. The results of this study support findings that reinforcement increases rate of eyeblink responses; signif icant conditioning occurred for the Experimental group but not for the yoked Control group. Significant performance gains occurred for the 15 Unaware subjects, indicating that for this non-verbal, non-monitor^d response the action of the reinforcer was direct and not mediated by cognitive processes. Performance gains were found in the absence of awareness although the aware subjects' performance increased at a greater rate. Frustration and general activity were not found to account for increase in eyeblink rate. Sex and extraver sion-introversion had no significant relationship to eye blink conditioning. The present study may have established a paradigm with a response class, eyeblink, uniquely suitable for the investigation of operant conditioning without awareness. APPENDICES A. TABLE 13, POST-EXPERIMENTAL QUESTIONNAIRE B. TABLE 14, BASIC DATA 72 APPENDIX A TABLE 13, POST-EXPERIMENTAL QUESTIONNAIRE 73 APPENDIX A TABLE 13 POST-EXPERIMENTAL QUESTIONNAIRE The interview questions, adapted from DeNike1s (1963) investigation, were planned to descry whether a sub ject had formed an hypothesis related to the eyeblink response-reinforcement contingency. Beginning with general questioning, the interview gradually became more specific with reference to bodily movements, some of which were correlated with the correct contingency. One question (#11) sought to evaluate the subjects' motivation for reinforcement during the conditioning period. In the original form the Interview questions were each typed on a separate sheet and presented to the sub jects one at a time. Post-Experimental Questionnaire 1. What do you think the experiment was all about? 2. What do you think you were supposed to do? 3. How did you go about deciding what to do? 74 TABLE 13— Continued While in the experiment, did you get the idea that you were supposed to behave in a certain way? Did you observe, at any time in the experiment, that you received a reinforcement (point on the counter, a buzz, and a light) when you did not press the telegraph key? (If yes) When did this occur? At any time in the experiment did you get the feeling that you were supposed to be earning points for any thing else than pressing the telegraph key? (If yes) What gave you the idea? Actually for some of the people who participate in the experiment, but not for all subjects, they earn points for making a certain kind of movement(s). Did you happen to notice if you were rewarded for making a certain kind of movement(s)? (If yes) For what kind of movement(s)? If the idea occurred, to you that a movement other than pressing the telegraph key produced points, check below which part of your body was involved in this movement(s): (a) ( ) whole body 76 TABLE 13— Continued (b) ) arm (c) ) head (d) ) leg (e) ) some other part (f) ) none [For those who answered Describe this movement (s) : At what point in the experiment did you get the notion that a movement was necessary to get points? 10. After you got the idea that you were getting points for some movement (s) , did you change the way in which you operated the telegraph key? (If yes) How did you change this? (If no) Why not? 11. How much would you say you wanted to earn points for money? ( ) Very much? ( ) Some? ( ) Or that you really didn't care? 12. If there's anything in your notes that you think we'll have trouble understanding later, I'd appreciate it if you'd clarify them now, using this red pencil, so that TABLE 13— Continued we'll know what was written after the experiment 1 proper. 13. What did you hear about this experiment, or other experiments like it, before coming in to participate in it? 14. Okay, I'll ask you to leave your notes with us now, and I will turn in your experimental class credit. Please sign for your cash now. Thanks very much for your participation. Please do not discuss this exper iment with anyone connected with USC until after you receive feedback about the experiment m your class. ^"A red pencil was provided on the table for subject to clarify any statement. An attempt was made by the interviewer to get the subject to agree to this with at least a head nod. If any information was requested, the interviewer said, "Sorry we can't give you any additional explanation now, but you will receive feedback about this experiment in your class." APPENDIX B TABLE 14, BASIC DATA 78 APPENDIX B TABLE 14 BASIC DATA Group: Experimental (Exp.) and Yoked Control (Y. C.) Subject Sex Extraversion-introversion Score Awareness Ratings: Aware (A), Correlated Aware (CA), Unaware (U) Belief (B) or Disbelief (D) in the Efficacy of the Telegraph Key Response Desire: Very (V), Some (S) , Didn't Care (D) for Reinforcement Session 1: Mean Eyeblink (EB) and Telegraph Key (TK) Response Per Minute Session 2: Mean Eyeblink (EB) and Telegraph Key (TK) Response Per Minute Session 3: Mean Eyeblink (EB) and Telegraph Key (TK) Response Per Minute Session 4: Mean Eyeblink (EB) and Telegraph Key (TK) Response Per Minute 79 TABLE 14 BASIC DATA 1 2 3 4 5* 6* 7* 8 9 10 11 E.B. T.K. E.B. T.K. E.B. T.K. E.B. T.K. Exp. 4 M 145 U B V 6.9 280.5 13.3 283.9 9.7 316.9 12.4 323. 6 Y. C. 1 M 145 - - - 8.2 240.6 8.1 299. 0 6.8 263.2 10.0 276.6 Exp. 8 M 49 CA B D 9.9 239.9 10.4 221.5 26.8 252.2 52.1 148.3 Y. C. 18 M 65 - - - 14.7 270.7 13.1 272.0 16.5 243.8 13.3 160.8 Exp. 9 M 143 CA B V 5.1 261.9 7.5 294.2 11.5 293.5 18.1 279.3 Y. C. 21 M 135 - - - 5.5 305.3 7.2 239.2 7.5 178.8 7.5 325.0 Exp. 13 M 61 CA B s 13.5 338.4 25.6 334.0 20.5 317.0 31.2 319.5 Y. C. 34 M 65 - - - 10.8 148.9 14.3 218.0 15.1 229.0 13.7 249.3 Exp. 14 M 61 U D V 13.4 249.8 19.1 228.4 21. 0 71. 8 17.4 34.9 Y. C. 10 M 61 - - - 12.3 307.7 14.2 197.0 18. 0 239.3 18.7 243.4 Exp. 15 M 129 A D s 17.1 180.3 19. 0 185.8 33.2 95. 0 95.9 6.4 Y. C. 44 M 129 - - - 16 . 6 216.1 21.7 263.2 16.4 262.0 19 .2 272.1 Exp. 22 M 41 U B V 20.5 231,3 ;22. 9 232.2 28.9 270.7 33.3 306.0 Y. C. 23 M 39 - - - 18.8 250.3 20.3 22.0 18.2 8.0 28.7 13.1 Exp. 24 M 141 u B D 17.3 328 . 9 15. 0 315.8 11. 9 347. 9 13.3 362.5 Y. C. 61 M 127 - - - 19.8 260.5 15.5 172.8 17.7 83.7 16.7 7.0 Exp. 26 M 129 u B V 27.1 269.7 34.5 121.7 37.2 266.8 38.2 301.3 Y. C. 48 M 143 - - - 25.3 259.4 23.8 116. 0 20.7 39.1 20.1 43.4 Exp. 25 F 143 u D V 23.5 218.1 32.5 152.1 32.8 234.7 35.1 226.7 Y. C. 7 F 139 - - - 20.8 248.1 18.1 207.0 22.1 167.2 18.8 249.2 Exp. 32 M 65 u D s 16.0 229.6 17. 9 167.7 23.5 221.5 20.9 235.4 Y. C. 52 M 67 - - - 19.0 213.8 14.2 348.0 20.3 336.0 22.5 311.1 Exp. 38 M 71 u B V 24.7 320.8 34.4 290.1 34.4 276.8 36.7 298.2 Y. C. 39 M 71 - - - 32.5 34 .1 37.3 38.8 241.5 264.3 239.2 219.3 Exp. 42 M 129 CA B V 11.2 314.0 21.0 341.6 21.5 334. 6 19.4 251.1 Y. C. 46 M 131 - - - 11.5 182.5 14.2 202.2 23.2 206.4 20.6 193.5 Exp. 43 M 143 u D s 11.1 303.8 17. 0 335.5 16.7 129.2 13.7 0.2 Y. C. 6 M 147 - - - 14.8 224.0 16.1 33.0 16.1 268.7 18.2 286.1 Exp. 49 F 139 u B s 5.8 189 . 2 9.3 237.7 15.3 214.5 17.2 236.7 Y. C. 19 F 137 - - - 6.1 286.5 6.1 268.0 6.0 294. 0 5.2 296.7 Exp. 53 F 139 u B D 16.3 218.2 21.7 182.4 29.5 188.1 27.6 180.5 Y. C. 45 F 137 - - - 18.3 279.0 21.8 309.5 15.9 305.0 12.5 306.2 Exp. 57 F 71 CA B s 14.5 276.2 20.6 303.7 33. 8 311.7 31.9 310.9 Y. C. 36 F 69 - - - 18.2 182.5 14.6 202.0 20. 3 206. 0 18.7 193.5 Exp. 58 F 81 u D V 16.1 122.3 27.4 142.0 36.2 33.3 23.7 0.0 Y. C. 50 F 73 - - - 14.5 253.8 10.7 186.0 9.3 171.4 8.1 286.0 Exp. 60 F 67 CA D D 10.7 208.1 9.8 202.9 12.9 64.8 13.3 223.9 Y. C. 28 F 57 - - - 9.4 106.8 8.0 145.5 6.9 124.0 6.7 138.3 Exp . 63 M 135 u B V 25.5 267.4 49.6 87.1 56.5 294.2 50.8 310.6 Y. C. 11 M 135 - - - 25.6 279.6 23.5 281.4 16.5 309.5 28.3 334.8 Exp. 64 M 61 u B V 16.3 254.2 21.4 223.5 22.6 211.7 22.1 201.2 Y. C. 27 M 65 - - - 18.0 235. 6 22 . 0 184.5 19.3 185.4 19.9 194.8 Exp. 65 F 73 u B s 26.7 182.6 27.4 171.0 32.4 180. 0 30.7 152. 0 y . c , . 55 F 87 — ~ 22.3 214.3 23. 0 200. 0 23.1 180.5 18.1 156.0 ♦Inapplicable to yoked Control subjects. REFERENCES REFERENCES Adams, J. K. Laboratory studies of behavior without aware ness. Psychological Bulletin, 1957, 54, 383-405. Allyon, T., and Haughton, E. Control of the behavior of schizophrenic patients by food. Journal of the Experi mental Analysis of Behavior, 1962, 5, 343-352. Dawson, M. E. Human GSR classical conditioning and aware ness of the CS-UCS relation. Unpublished Ph.D. dissertation, University of Southern California, 1967. Dawson, M. E. and Grings, W. W. Comparison of classical conditioning and relational learning. Journal of Experimental Psychology, in press. DeNike, L. D. Awareness in verbal conditioning: The assessment of awareness from verbal reports written by subjects during conditioning. Unpublished Ph.D. dissertation, Duke University, 1963. DeNike, L. D. The temporal relationship between awareness and performance in verbal conditioning. Journal of Experimental Psychology, 1964, 68, 521-529. DeNike, L. D. Recall of reinforcement and conative activ ity in verbal conditioning. Psychological Reports, 1965, 16, 345-346. Dollard, J., and Miller, N. E. Personality and psycho therapy . New York: McGraw-Hill, 1950. Dulany, D. E. The place of hypotheses and intentions: An analysis of verbal control in verbal conditioning. In C. W. Eriksen (Ed.), Behavior and awareness. Dur ham, N. C.: Duke University Press, 1962. Pp. 102-129. 82 83 Eriksen, C. W. Discrimination and learning without aware ness: A methodological survey and evaluation. Psychological Review, I960, 67, 279-300. Eriksen, C. W. Figments, fantasies, and follies: A search for the subconscious mind. In C. W. Eriksen (Ed.), Behavior and awareness. Durham, N. C.: Duke University Press, 1962. Pp. 3-26. Eysenck, H. J. Dimensions of personality. London: Rout- ledge & Kegan Paul, 1947. Eysenck, H. J. Dynamics of anxiety and hysteria. London: Routledge & Kegan Paul, 1957. Eysenck, H, J. The structure of human personality. Second edition. London: Methuen, 1960. Eysenck, H. J. and Eysenck, S. B. G. Manual for the Eysenck Personality Inventory. San Diego: Educational & Testing Service, 1963. Ferster, C. B., Nurnberger, J. I., and Levitt, E. B. The control of eating. Journal of Mathetics, 1962, 1, 87-109. Ferster, C. B., and Skinner, B. F. Schedules of reinforce ment. New York: Appleton-Century-Crofts, 1957. Field, J. G., and Brengelmann, J. C. Eyelid conditioning and three personality parameters. Journal of Abnormal & Social Psychology, 1961, 63, 517-523. Flanagan, B., Goldiamond, I., and Azrin, N. Operant stut tering: The control of stuttering behavior through response-contigent consequences. Journal of the Experimental Analysis of Behavior, 1958, 1, 173-178. Fowler, R. L., and Kimmel, H. D. Operant conditioning of the GSR. Journal of Experimental Psychology, 1962, 63, 563-567. Franks, C. M. Conditioning and personality. Journal of Abnormal & Social Psychology, 1956, 52, 143-150. 84 Garner, W. R., Hake, H. W., and Eriksen, C. W. Operation- ism and the concept of perception. Psychological Review, 1956, 63, 149-159. Gelfand, D. M., and Winder, C. L. Operant conditioning of verbal behavior of dysthymics and hysterics. Journal of Abnormal & Social Psychology, 1961, 62, 688-689. Greene, W. A. Operant conditioning of the GSR using par tial reinforcement. Psychological Reports, 1966, 19, 571-578. Greenspoon, J. The reinforcing effect of two spoken sounds on the frequency of two responses. American Journal of Psychology, 1955, 68, 409-416. Harwood, C. W. Operant heart rate conditioning. Psycho logical Record, 1962, 12, 279-284. Hefferline, R. F., Keenan, B., and Harford, R. A. Escape and avoidance conditioning in human subjects without their observation of the response. Science, 1959, 130, 1338-1339. Hefferline, R. F., and Keenan, B. Amplitude induction gradient of a small scale (covert) operant. Journal of the Experimental Analysis of Behavior, 1963, 6, 307-315. Irwin, F. W., Kauffman, K., Prior, G., and Weaver, H. B. On "learning without awareness of what is being learned." Journal of Experimental Psychology, 1934, 17, 823-827. Kanfer, F. H. Verbal conditioning: A review of its cur rent status. To be published in: Dixon, T. R., and Horton, D. L. (Eds.) Verbal behavior and its relation to general S-R theory. New York: Prentice-Hall, (in press). Keehn, J. D. Operant eyelid conditioning of severely retarded children. Psychonomic Science, 1956, 3, 585- 586. 85 Keehn, J. D. Experimental studies of the "unconscious": Operant conditioning of unconscious eyeblinking. Behaviour Research and Therapy, 1967, 5, 95-102. Keehn, J. D., Lloyd, K. E., Hibbs, M., and Johnson, D. Operant eyeblink conditioning without awareness: A preliminary report. Psychonomic Science, 1965, 2, 357-358. Kimble, G. A. Hilgard and Marquisf conditioning and learn ing. New York: Appleton-Century-Crofts, 1961. Kimmel, H. D., and Hill, F. A. Operant conditioning of the GSR. Psychological Reports, 1960, 1, 555-562. Krasner, L. Studies of the conditioning of verbal behavior. Psychological Bulletin, 1958, 55, 148-170. Lindquist, E. F. Design and analysis of experiments in psychology and education. Boston: Houghton-Mifflin, 1953. Matarazzo, J. D., Saslow, G., and Pareis, E. N. Verbal conditioning of two response classes: Some methodolog ical considerations. Journal of Abnormal and Social Psychology, 1960, 61, 190-206. McNair, D. M. Reinforcement of verbal behavior. Journal of Experimental Psychology, 1957, 53, 40-46. Myers, Isabel B. Manual (1962), the Myers-Briggs Type Indicator. Princeton, N. J.: Educational Testing Service, 1962. Parton, D. A., and Ross, A. O. Social reinforcement of children's motor behavior: A review. Psychological Bulletin, 1965, 64, 65-73. Paul, G. L., Eriksen, C. W., and Humphreys, L. G. Use of temperature stress with cool air reinforcement for human operant conditioning. Journal of Experimental Psychology, 1962, 64, 329-335. 86 Philbrick, E. B., and Postman, L. A further analysis of "learning without awareness." American Journal of Psychology, 1955, 68, 417-424. Postman, L., and Jarrett, R. F. An experimental analysis of "learning without awareness." American Journal of Psychology, 1952, 65, 244-255. Ross, J. The relationship between the Myers-Briggs Type Indicator and ability, personality and information tests. Research Bulletin 63-8. Princeton, N. J.: Educational Testing Service, 1963. Salzinger, K. Experimental manipulation of verbal behavior. Journal of General Psychology, 1959, 61, 65-94. Sasmor, B , . M. Operant conditioning of a small-scale muscle response. Journal of the Experimental Analysis of Behavior, 1966, 9, 69-85. Shearn, D. W. Operant conditioning of the heart rate. Unpublished doctoral dissertation. University of Indiana, 1960. Singer, R. D. Verbal conditioning and generalization of prodemocratic responses. Journal of Abnormal and Social Psychology, 1961, 63, 43-46. Spence, K. W. Cognitive and drive factors in the extinc tion of the conditioned eye blink in human subjects. Psychological Review, 1966, 73, 445-458. Spielberger, C. D. The role of awareness in conditioning. In C. W. Eriksen (Ed.), Behaviorism and awareness. Durham, N. C.: Duke University Press, 1962. Pp. 73-101. Spielberger, C. D., and DeNike, L. D. Descriptive behav iorism versus cognitive theory in verbal operant con ditioning. Psychological Review, 1966, 73, 306-326. Spielberger, C. D., DeNike, L. D., and Stein, L. S. Anxiety and verbal conditioning. Journal of Personality and Social Psychology, 1965, 1, 229-239. 87 Spielberger, C. D., Levin, S. M., and Shepard, Mary C. The effects of awareness and motivation towards the reinforcement on the operant conditioning of verbal behavior. Journal of Personality, 1962, 30, 106-121. Strieker, L. J., Schiffman, H., and Ross, J.. Prediction of college performance with the Myers-Briggs Type Indica tor. Educational and Psychological Measurement, 1965, 25, 1081-1095. Strieker, L. J., and Ross, J. Intercorrelations and reli ability of the Myers-Briggs Type Indicator Scales. Psychological Reports, 1963, 12, 287-293. Sweetbaum, H. A. Comparison of the effects of introversion- extraversion and anxiety of conditioning. Journal of Abnormal and Social Psychology, 1963, 66, 249-254. Thorndike, E. L. Animal intelligence: Experimental studies. New York: The Macmillan Co., 1911. Thorndike, E. L. The psychology of learning. Educational psychology, II. New York: Teachers College, 1913. Thorndike, E. L., and Rock, R. T. Learning without aware ness of what is being learned or intent to learn it. Journal of Experimental Psychology, 1934, 17, 1-19. Verplanck, W. S. The operant conditioning of human motor behavior. Psychological Bulletin, 1956, 53, 70-83. Walter, W. G. Observations and experiments concerning reflectory and voluntary eyelid movements. Archives Neerlandaises de Physiologie de 1'Homme et des Animaux, 1941, 25, 601-620. Winer, B. J. Statistical principles in experimental design. New York: McGraw-Hill, 1962.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Prognostic Expectancy Effects In The Desensitization Of Anxiety Over Invasion Of Body Buffer Zones

PDF

The Effect Of Dissonance In Self-Esteem On Susceptibility To Social Influence

PDF

Effects Of Group Behavior Therapy Imagery On Basketball Performance

PDF

The Effects Of Anxiety And Threat On Self-Disclosure

PDF

Differences Between Cues In Effectiveness As Retrieval Aids

PDF

Human Gsr Classical Conditioning And Awareness Of The Cs-Ucs Relation

PDF

The Effects Of Nonreward Dissonance And Secondary Reward On Extinction And Attractiveness

PDF

Verbal Reports Of Emotional States And Onsets And Offsets Of Conditioned Stimuli

PDF

The Enhancement Of Eeg - Alpha Production And Its Effects On Hypnotic Susceptibility

PDF

Performance Of Retardates On Piagetian Tasks As A Function Of Ethnicity

PDF

Imagery And Response Styles In Desensitization

PDF

Ethnic Group Differences In Certain Personal, Intellectual, Achievement, And Motivational Characteristics

PDF

The Effect Of Discriminability On The Partial Reinforcement Effect In Human Gsr Conditioning

PDF

The Effects Of Justice, Balance, And Hostility On Mirth

PDF

Operant Discrimination Of An Interoceptive Stimulus In The Urinary Bladder Of Intact And Dorsal Root Transected Female Rhesus Monkeys

PDF

A Factor Analysis Of The Figural-Evaluation Abilities

PDF

Effects Of Video Tape Feedback Versus Discussion Session Feedback On Group Interaction, Self Awareness And Behavioral Change Among Group Psychotherapy Participants

PDF

The Effects Of A Self Shock Procedure On Hallucinatory Activity In Hospitalized Schizophrenics

PDF

A Comparison Of The Values Of High And Low Creative Seventh Grade Students In Selected Junior High Schools In The Los Angeles District

PDF

An Examination Of Positive And Negative Reinforcement In Classical And Operant Conditioning Paradigms In The Primary Psychopath

Asset Metadata

Creator Kessler, Irving (author)

Core Title Human Operant Eye Blink Conditioning, Awareness, And The Extraversion-Introversion Dimension Of Personality

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, experimental

Language English

Contributor Digitized by ProQuest (provenance)

Advisor De Nike, L. Douglas (committee chair), Metfessel, Newton S. (committee member), Slucki, Henry (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-590761

Unique identifier UC11360131

Identifier 6801685.pdf (filename),usctheses-c18-590761 (legacy record id)

Legacy Identifier 6801685.pdf

Dmrecord 590761

Document Type Dissertation

Rights Kessler, Irving

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA