THE EFFECTS OF VARIOUS VERBAL REINFORCEMENT PROGRAMS UPON THE SIMPLE REACTION LATENCIES OF PRE-PUBERTAL CHILDREN Thesis for the Degree of M. A. MICHIGAN STATE UNIVERSITY STEPHEN. COLE 1971 uw. ' ‘4‘ W ‘ "’INC. LIBRARY BINDER ”MAGNET. mam» K\< ‘ ,5 : La ’— RETURNING MATERIALS: PVIESI_J Place in book drop to IJBRARJES remove this checkout from 4...!— your record. FINES wiII be charged if book is returned after the date stamped below. ABSTRACT THE EFFECTS OF VARIOUS VERBAL REINFORCEMENT PROGRAMS UPON THE SIMPLE REACTION LATENCIES OF PRE-PUBERTAL CHILDREN BY Stephen Cole This experiment was designed to study the effects of various verbal reinforcement programs upon the simple reaction latencies of pre-pubertal children of both sexes. A secondary objective of this study was to determine what type of stimulus (light, sound, or light + sound) produced the fastest reaction time for children within the educa- tional level of kindergarten through second grade. The experiment was conducted to test the following hypotheses: 1. Reaction time responses decrease with age (in this case between kindergarten and Grade two). 2. There is no improvement in reaction time if no verbal reinforcement is given. 3. Positive verbal reinforcement is more effective than negative verbal reinforcement in reducing reaction time. c) 7 a (o f/ :j Stephen Cole 4. The fact that males are regarded as having gener- ally faster reaction times than females will be accentuated by the reinforcement programs (for a particular age level). 5. Reinforcement relating directly to performance, i.e., fast responses warrant positive verbal encouragement, is more effective than a set pat- tern of encouragement or censure. To test the above hypotheses, a sample of 367 sub- jects was drawn by convenience from the Battle Creek Cur- riculum Project. The subjects were divided into sex-age groups and assigned to one of the three experimental groups. Each subject was given twenty trials of a simple index finger lift response, after receiving one of three stimulus modes. The first ten trials were given with no verbal cues; the second ten trials were conducted accord- ing to the experimental group the subject was assigned. The results of these experiments provide support for the hypothesis that, with an increase in age, there is a correSponding increase in the speed of reaction time. The presence of multiple stimuli appeared to accelerate a subject's reaction time in relation to the simple light and sound stimuli. Both the boys and girls produced significantly faster response for the light + sound stimulus than for the other two stimulus modes. Stephen Cole The results provided support for the hypothesis that verbal reinforcement increases the speed of reaction time. All three experimental groups had significantly fas- ter responses when compared to the control group, which received no verbal cues. However, no clear pattern emerged as to the relative importance played by the various rein- forcements. There was no significant difference between those subjects receiving the reinforcement which related directly to performance and those receiving a set pattern of reinforcement. From the results obtained, there was a suggestion that the girls' performance was adversely affected by nega- tive reinforcement. The boys, on the other hand, were not adversely affected and may even have improved the perform- ance when compared to the mean of their first ten trials. The results of this experiment clearly demonstrated that both boys and girls reSpond to reinforcement of both a positive and negative type. It is suggested that the control group with no verbal cues experienced boredom and fatigue far earlier in the test than the experimental groups. The results indicate verbal feedback in relation to a test of this kind will have some arousal effect on the performance of the subject. THE EFFECTS OF VARIOUS VERBAL REINFORCEMENT PROGRAMS UPON THE SIMPLE REACTION LATENCIES OF PRE-PUBERTAL C HILDREN BY Stephen Cole A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Health, Physical Education and Recreation 1971 ACKNOWLEDGMENTS I wish to express my gratitude to Dr. V. Seefeldt for his guidance and help in the production of this study. I would also like to thank Dr. w. Heusner for his guidance in the preparation of the statistical treatment of the data. ii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . iv LIST OF FIGURES. . . . . . . . . . . . . vi Chapter I. INTRODUCTION TO THE PROBLEM . . . . . . 1 Statement of the Problem. . . . . . . 3 The Need for the Study . . . . . . . 5 Limitation of the Study . . . . . . . 6 II. REVIEW OF LITERATURE. . . . . . . . . 8 I I I O METHODOLOGY 0 O O O O O O O O O O O l 7 IV. ANALYSIS AND PRESENTATION OF DATA . . . . 26 ‘VQ SUMMARY AND CONCLUSIONS. . . . . . . . 60 BIBLIOGRAPHY. . . . . . . . . . . . . . 65 APPENDIX . . . . . . . . . . . . . . . 68 iii 10. ll. 12. 13. LIST OF TABLES Group experimental conditions . . . . . . One way analysis of variance, grade x sex x stimulus type for first and second ten trials 0 O O O O O O O O O I O 0 One way analysis of variance, boy x girl for first and second ten trials . . . . . . One way analysis of variance, kindergarten x first grade x second grade for first and second ten trials . . . . . . . . . Mean data for each grade. . . . . . . . Mean data for grade + sex, first ten trials . Duncan Multiple Range Test, second ten trials by stimulus type. . . . . . . . . . Duncan Multiple Range Test, second ten trials by sex + stimulus type. . . . . . . . One way analysis of variance, time by rein- forcement type . . . . . . . . . . Duncan Multiple Range Test, time by rein- forcement type . . . . . . . . . . Duncan Multiple Range Test, time by rein- forcement type and grade . . . . . . . (One way analysis of variance time by rein— forcement type + sex . . . . . . . . IJuncan Multiple Range Test, time by rein— forcement type and stimulus type . . . . iv Page 22 29 30 31 33 34 35 36 40 41 43 46 49 Table Page 14. Duncan Multiple Range Test, ranked means by reinforcement type and stimulus type . . . 51 15. One way analysis of variance, time by rein— forcement type + grade + stimulus type . . 52 16. Duncan Multiple Range Test, time by rein- forcement type + grade + stimulus type . . 53 17. Duncan Multiple Range Test, time by rein— forcement type + sex + stimulus type . . . 54 18. Duncan Multiple Range Test, time by rein- forcement type + sex + stimulus type . . . 56 LIST OF FIGURES Figure Page 1. The Experiment Setting . . . . . . . . 24 2. A Comparison of Mean Scores of the First Ten Trials Compared with the Second Ten Trials for Each Grade + Sex + Stimulus Group . . 27 3. A Comparison of Stimulus Types by Grade Using Mean Scores . . . . . . . . . . . 37 4. A Comparison of Reinforcement Types by Sex Using Mean Scores . . . . . . . . . 45 5. A Comparison of Reinforcement Types by Sex and Stimulus Type Using Mean Scores . . . 48 Vi . NICHIGAN SIAIE UNIV‘I-jHSETY ILLIJUCHJHAL REVOUICCS CENTER 133 ERICKSON HALL L—AQT IAMQIAII‘ AllnlunAAl ‘AnA/x CHAPTER I INTRODUCTION TO THE PROBLEM In the struggle for survival, primitive man of necessity focused attention upon developing those abilities required for obtaining food, shelter, and protection. Modern man no longer lives in such intimate contact with toil, sweat, and physical violence. His world is one of implements, machines, and gadgets. When work is to be done, a labour sav- ing device is available (Rarick [22]). It is evident that as man's technology advances, rue no longer relies entirely on his motor mechanisms and gflrysical fitness. However, it should not be forgotten that even in a highly mechanized culture man must call upon his neuromuscular system in order to accomplish many of the manjqoulative skills found in industrial labor and the gross IMHKDI'SkillS found in recreative activities. Thus, as part Of a child's development, attention must be paid to the learwring of motor as well as mental skills. Total develop- ment :is dependent upon the mastery by the individual of skills; that will facilitate the fulfillment of his physi- cal arui psychological needs and assist him to avoid harm- ful eXperiences (Crow and Crow [4]). If we SUbSCIibe to this Statement, we also assume that it is necessary to provide the opportunity for children to learn the skills of their culture. Some of the problems facing children with poor motor development will affect their later life unless they are not remedied or corrected. A child's proficiency in motor skills determines to some degree the extent to which he or she will be successful in social, educational, and vocational areas. ~The child's judgment of himself is .shaped in part by his association with his peer group. If lie cannot compete on an equal basis, regardless of the skill involved, it is likely to be reflected in his estima- tion of self-worth. Havighurst [8] has emphasized a child's concept of himself is related to the motor skills he pos- sesses. As a child becomes part of an activity group, he contributes certain skills and knowledge. This is the time when he has a chance to test his skills against those of his peers. Beginning with middle childhood, enhanced motor competence becomes a major source of primary status. The child's world is abruptly changed when the ability to move about with freedom has been attained. The child is able to move independently and freely explore its environment with some feeling of competence. Children with motor handicaps not only feel more timorous in this respect but also find it difficult to maintain a level of aspiration when faced with motor tasks (Ausubel [1]). Rarick and McKee [23] have shown that third grade children who showed proficiency in motor skills were found to be more active, popular, calm, resourceful, attentive, and cooperative than children who lacked such proficiency. Thus, it may be surmised that the development of adequate motor skills is an important goal in the education of young children. Statement of the Problem Simple reaction time may be defined as the time interval between the onset of a stimulus and the initia— tion of a response, with standard instructions to respond as quickly as possible. This study will attempt to determine if verbal reinforcement is effective in the instrumental conditioning of reaction time in children, with the following specific purposes in mind: 1. To study the relationship between positive verbal reinforcement and response speed. 2. To study the relationship between negative verbal reinforcement and reSponse speed. 3. To determine whether children learn to respond faster with practice even though no reinforcement is administered. To determine if age is a factor in the relative success of positive or negative reinforcement in the lowering of the subject's reaction time. To study the differences between male and female response speeds under positive and negative rein- forcement programs. With reference to research findings, the following hypoth- eses were made: 1. Reaction time will not improve if no reinforcement is given. Positive reinforcement following fast reaction times will increase the overall speed of reaction time in contrast to the speed of the reaction time if no reinforcement is given. Negative verbal reinforcement will increase the Speed of reaction time. Reaction time is indirectly related to age (as age increases, reaction time decreases), regard- less of reinforcement given. The difference between male and female reaction times will be accentuated by the reinforcement programs. Reinforcement given according to performance (i.e., fast response warrants positive verbal encouragement), is more effective than a set pat- tern of encouragement or censure. There are no conclusive data for this age group pertaining to the type of stimulus (light, sound, or light and sound simultaneously) which ellicites the fastest response. Therefore, this study will also attempt to deter— mine which type,or combination of stimuli, produces the fastest response for this age group. The Need for the Study The importance of motor skills in the life of a child has already been indicated. What, then, is the role of the teacher in this critical aspect of a child's devel- Opment? This study is concerned with a component of motor performance; namely, simple reaction time. As this may be taken as a representative motor function, what is the effect upon performance of both positive and negative ver- bal reinforcement? It may be trite to suggest that words influence our actions. However, no one has yet fully accounted for this phenomenon and only recently has this problem received attention. As a result of the work conducted by Skinner [29] and Greenspoon [7], there has developed an area of research called verbal conditioning. This form of condi- tioning is an integral consideration in the establishment of a theory for the socialization process in children. However, most of the research in the form of reinforcement has been concerned not with words but with material rein— forcers. Similarly, during the past few years there has been a considerable amount of educational and psychological research directed towards aSpects of reaction time. How- ever, what research has been completed has been mostly directed to adults and adolescents. It is questionable whether one can infer that a similar relationship between stimulus and reaction will exist across the span of years between childhood and maturity. Basically this study is concerned with the influence of verbal reinforcement on a child's performance of a motor skill at a critical period in the development of a motor performance repetoire. Limitation of the Study This particular study is part of a larger curricu- lum project carried out in the Battle Creek school system (Vogel [31]). The sample for the whole project was drawn randomly from all the schools in the system and consisted of six hundred pupils. The sample utilized in this study is a part of that larger sample and selected according to the availability of the subjects at the time of testing. The subjects included in this study comprise approximately sixty-one percent of the original sample. The testing was done in a variety of schools and at various times of the day; however, an effort was made to standardize conditions as far as light intensity, seating positions, and amount of noise were concerned. When studying the age parameter in this problem, no attempt was made at grouping the subjects by mental, skeletal, or biological age. For ease of administration the grouping was determined by the grade of the subject. In addition, no attempt was made in this study to account for individual performances in a psychological context. However, the relationship between a subject's performance under varying forms of reinforcement and their psychologi— cal make-up may provide avenues of inquiry for further study. CHAPTER II REVIEW OF LITERATURE A tremendous volume of work has been completed in the field of reaction time in psychological, phsiological, and educational contexts. The literature to be reviewed here comes mainly from four areas of research related to reaction time: the effect of age, the effect of sex, the effect of differing stimuli, and the effect of differing types of reinforcement programs. Because of the tremen— dous number of experiments produced in these areas, a complete review would be impractical since it would far exceed the purpose and the scope of this study. Only literature that has direct pertinence to the problem will be reviewed. The simplest, most typical and time honored model for the objective observation of the effect of a stimulus upon a motor response is the classic simple reaction time experiment. Simple reaction time is defined as the time interval between the onset of the stimulus and the initia- tion of a response, with standard instructions to respond as quickly as possible. l1 [1 .J r! It is apparent from the outset that there are many influences acting upon reaction time performance at any one time. Even diurnal variations, as Elbel [5] discovered, have an effect on response times. His investigation com— pared time of day with the response time of twenty-three male subjects in a stimulus-hand response. Elbel was able to isolate the following factors: slowest responses are obtainable at twelve noon, maximum speed in early afternoon, and a near maximum in mid-morning. Thus, the time of the day in which the testing takes place has an effect on per- formance. PeOple supervising pre-adolescent physical activity may unknowingly have unattainable expectation levels set for their charges due to incompletely developed sensori- motor capacities. One of the capacities is reaction time. Therefore, for each age range, it is necessary to know how fast a child will react to a certain stimulus. Unfortu- nately, most of the research completed hitherto has been confined to studies of chronological age, leaving the more accurate measures of maturity, namely mental and skeletal age, largely unused. Pierson and Montoye [19] measured the reaction and movement time in four hundred male sub- jects, aged eight to eighty-three. They found that move- rmant time and reaction time are significantly related to cflrronological age. In both reaction and movement time, the fastest response was shown to be greatest around age 10 thirty. These findings suggest that one could expect con- sistent responses with young peOple (adolescents), but would expect wide inter-individual differences in children of kindergarten age. Botwinik and Thompson's [2] study on the components of reaction time,in relation to age and sex, would seem to substantiate the previous study. Unfortu- nately, the age range was post adolescence (eighteen years old) to eighty—seven and did not include young children. The reaction, pre-motor,and motor times, were calculated from data collected from a single finger lift reSponse. It was shown that the elderly group was slower in all three categories and also that both elderly and young males had faster response times than their female counter- parts. The research completed by Hodgkins [9] reported results similar to the two previously discussed topics. Nine hundred and thirty men, women, and children ranging from six to eighty-four years were tested to determine the differences between males and females of various ages in their speed of reaction and movement time. Results indi- cate that: (a) males were faster than females in both reaction and movement time; (b) speed of both functions increased up to early adulthood and then decreased; (C) peak Speed was retained longer by males in movement time and by females in reaction time. The age parameter appears more complex when con- sidering the study by Philip [18] on the reaction times 11 of children grades four to eight. Although he found that reaction time decreased with age, he also found that the improvement in coordination which enables this decrease to be brought about by maturation is not confined to infancy and early childhood. Further, he found that in the older age ranges, i.e., grade eight there was a defi- nite slowing down of the reactive processes apparent in girls but not in boys. Goodenough [6],in her study of the development of the reactive process from early childhood to maturity, selected subjects ranging in age from two and one-half years to eleven and one-half years and also fifty-six college students. She found that the development of reactive processes not only improves the speed of response, but even more marked is the degree of gaining voluntary control over motor activity, which manifests itself in fewer signs of bodily tension as the subjects increase in age. To summarize the above points, reaction time decreases with age during childhood and adolescence; how- ever, the rate of decrease is still open to question. The effects of involuntary motor movements in very young children further complicate this picture. Most of the studies dealing with reaction time h Masai-hull? venom + uzmnq ocsom osmflq HHHD >Om cmuummumocflx comma uma oomuo cam IIHN llll r—INM Nr-IO .qsouo msasfiflum+xom+mpmum some qu mamfluu no» Ucoomm who cuez ooquEoo mamfluu sou umuew era «0 mmuoom some mo coweummeoo c 0') CL :3 2I2 O 023 22l xumluv uocmolnv 2I| C3 mo 9.322.ka xwm um I2| 222 |22 022 Oil 02| .N musmfih oexb .sflum xwm occuo OUOU $44.1... zwb ozooum to 24m! mod-m... 2m... bmmr... no 24w: .10.... mm. 1 on. o D "Z n. J. c. saNooas NI la In A 1. 28 position to the groups faster and slower than it. Consid- ering Figure 2, it is obvious that the second ten trials do not follow that pattern. For example, in the case of second grade, girls, sound stimulus, the first ten trials rank fourteenth among all the groups. However, when one looks at the second ten trials, this group ranks ninth. Therefore, some other variable than grade, sex, and stimu- lus type must have had an effect on the second ten trials; ;possibly the type of reinforcement was responsible for the change. It will be seen from Table 2 that the between- ‘groups variance is higher for the first ten trials than :for the second ten. This would indicate that the subjects' rresponses are erratic during the first ten trials but huecome more stable during the second ten trials as the suflojects become accustomed to the experimental conditions. Sex differences apparently do not account for a great deal of variance in these data. In the first ten 'txzials, there is a significant difference between the mean art. the .002 level of significance. However, in the second ‘teern trials, at the .5 level, this variance is absent. {P1163 boys, in both cases, are slightly faster and appear to 13GB rnore stable over the twenty trials (see Table 3). The effect of age on the reaction times of the S“abjects is shown in Table 4. All three grades differ Sj—Qnificantly at the .005 level of significance. As 29 TABLE 2.--One way analysis of variance, grade x sex x stimulus type for first and second ten trials. 5°“r83 0f dF 55 MS F P Variance First Ten Trials Between groups 17 1.3955 0.0820 6.41 .0005 Within groups 349 4.4713 0.0120 Total 366 5.8669 Second Ten Trials Between groups 17 1.4932 0.0878 5.20 .0005 Within groups 349 5.8930 0.0168 Total 366 7.3863 Dependent variable--lst 10 trials Category variable--grade, sex, stim. type Dependent variable--2nd 10 trials Category variable—-grade, sex, stim. type 30 TABLE 3.--One way analysis of variance, boy x girl for first and second ten trials. Source of dF . SS MS F P Variance First Ten Trials Between groups 1 0.1523 0.1523 9.73 .002 'Within groups 365 5.7145 0.0156 'Total 366 5.8669 Second Ten Trials Between groups 1 0.0721 0.0721 3.60 .5 Within groups 365 7.3142 0.0200 Total 366 7.3863 Dependent variable—~lst 10 trials Category variable--sex Dependent variable--2nd 10 trials Category variable-—sex 31 TABLE 4.--One way analysis of variance, kindergarten x first grade x second grade for first and second ten trials. sourSe Of dF 53 MS F P Variance First Ten Trials Between groups 2 .0555 0.2776 19.02 .0005 Within groups 364 5.3166 0.0145 Total 366 5.8669 Second Ten Trials Between groups 2 .7147 0.3573 19.49 .0005 Within groups 364 6.6715 0.0183 Total 366 7.3863 Dependent variable-—lst 10 trials Category variable-~grade Dependent variable--2nd 10 trials Category variable—-grade 32 expected, the fastest age group was the second grade and the slowest, the kindergarten (see Table 5). When considering the interaction of sex and age, there is only one significant difference (P <.05) from the expected pattern. First grade boys exhibit a faster mean response than second grade girls in the first ten trial blocks (see Table 6). At face value, the analysis of the stimulus type would account for a very great degree of variance in these results: F(§%4) = 18.66 P <.005. However, on further inspection, it is apparent that the variance is almost totally due to one particular type of stimulus, the light + sound Mode. The Duncan Multiple Range Test shows that there is no significant difference between the two stimuli; namely, the light and sound, and the third, namely, the light and sound simultaneously. The latter elicits the fastest mean response (see Table 7). TTue interaction of sex and stimulus type also brings out tflris marked variance between the light and sound stimu- lus; (L + S) and the other two types. Both the boys and giJfls with the L + S stimulus are significantly different frcnn the other sex-stimulus groups (see Table 8). Even the! significant ranking of the grades (second, fast- est;.__+. Kindergarten, slowest) is masked when consider— ing the stimulus type and age (see Figure 3). Kinder- garten subjects with the L + S stimulus produce a faster 33 TABLE 5.--Mean data for each grade. Standard Mean . . Dev1ation First Ten Trials Kindergarten 0.47 0.13 First Grade 0.41 0.12 Second Grade 0.36 0.11 Kindergarten 0.43 0.18 First Grade 0.38 0.12 Second Grade 0 . 32 0 . lO Dependent variable--lst 10 trials Category variable--grade Dependent variable—-2nd 10 trials Category variable——grade 34 TABLE 6.--Mean data for grade + sex, first ten trials. Mean 3232233.. Kindergarten Girls 0.48 0.10 Kindergarten Boys 0.46 0.11 First Grade Girls 0.44 0.12 Second Grade Girls 0.39 0.13 First Grade Boys 0.38 0.10 Second Grade Boys 0.35 0.10 Dependent variable--1st 10 trials Category variable--grade + sex 35 TABLE 7.-—Duncan Multiple Range Test, second ten trials by stimulus type. Entries which differ signifi- cantly from the left hand entry Entry Mean Entry Mean Light + Sound 0.32 Sound 0.41 Light 0.40 Light 0.40 Light + Sound 0.32 Sound 0.41 Light + Sound 0.32 Note: Significant level = .05. 36 TABLE 8.--Duncan Multiple Range Test, second ten trials by sex + stimulus type. Entries which differ signifi- cantly from the left hand entry Entry Mean Entry Mean Boys, L + S 0.31 Boys, S 0.38 Boys, L 0.39 Girls, L 0.41 Girls, S 0.44 Girls, L + S 0.33 Boys, 8 0.38 Boys, L 0.39 Girls, L 0.41 Girls, S 0.44 Boys, S 0.38 Boys, L + S 0.31 Girls, L + S 0.33 Girls, S 0.44 Boys, L 0.39 Boys, L + S 0.31 Girls, L + S 0.33 GirJJs, S 0.44 Boys, L + S 0.31 Girls, L + S 0.33 Boys, S 0.38 k Note: Significant level = .05. lJGHT SOUND 37 IIIIIIIIIIIIIIIIIII oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo .......................................................................... oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo oooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo ............................................................................................................................................. ggggggggggggggggg A..-.gAAIAAQIIIIIIIl-IlllIII-IIOIIIOICI'D-IIIIIOOII WW / LI IlllllllllllllllllllljllIll H) C) 1) ct c '0. SONOOBS NI .lU STIMULUS TYPE BY GRADE types by grade usi lus l o 38 reSponse than one second grade group reacting to the sound stimulus and two first grade groups reacting to light and sound stimuli, respectively. To summarize the above results, it would appear that the L + S stimulus has accounted for the greatest amount of variance in the data. The other two stimulus types, namely, sound and light individually, did not pro— duce any significant variance. Sex was also found to be non-significant in its effect on reaction latencies. Chronological age was the other factor which contributed to the variance in the data. The reinforcement groups used in the following discussion are as follows: Control group-—the subject was given twenty trials with no verbal cues being given by the tester. Experimental group l——the mean was taken of the subjects' first ten trials. On the basis of this mean, the tester then administered "warranted" reinforcement for performance, i.e., if a subject did better than the mean score, he received positive reinforcement immediately after that trial. If the child did not attain the mean Score, the negative reinforcement was administered. EXperimental group 2A—-the first ten trials were Twininistered as usual. From the eleventh to the fifteenth triiils, positive reinforcement was administered, regardless 39 of the score obtained. For the final five trials, negative reinforcement was given after each trial. Experimental group 2B--the first ten trials were administered as usual with the subjects receiving negative reinforcement from the eleventh to the fifteenth trials. For the final five trials, positive reinforcement was administered. These four reinforcement programs will be known hereafter as the control group, the warranted group, the + ve group, and the - ve group, respectively. The variance shown in the Analysis of Variance (see Table 9) for reinforcement types is quite marked, F(7%§) = 7.66, P = .005. However, on referral to the Duncan Multiple Range Test (see Table 10), it can be attributed to one of the groups rather than a uniform vari— ance between all four types. At the .05 level of signifi- cance, only the control group showed variance from the warranted and the + ve reinforcement groups. This would suggest that any verbal reinforcement program will accel- erate the subject's response speed when compared to those subjects receiving no verbal reinforcement. The three eXperimental treatments, the warranted, + ve, and - ve groups, were not significantly different from one another indicating that no one verbal conditioning program is more effective than the other two. 40 TABLE 9.-—One way analysis of variance, time by reinforcement type. Source of dF . SS MS F P Variance Between categories 3 .0505 0.1684 7.66 .0005 Within categories 729 16.0408 0.0220 Total 732 16.5461 Dependent variable--time Category variable--reinforcement type 41 TABLE lO.--Duncans Multiple Range Test, time by reinforcement type. Entries which differ signifi- cantly from the left hand entry Entry Mean Entry Mean Warranted 0.34 Control 0.41 + ve rein. 0.37 Control 0.41 - ve rein. 0.38 Control 0.41 Warranted 0.34 + ve rein. 0.37 Note: Significant level = .05. 42 Within each grade level, only the control group was significantly different from the warranted reinforcement type at the .05 level of significance (see Table 11). How— ever, it is interesting to note the relative order of the four groups for each grade. In the kindergarten and first grade, the order was as follows: slowest reSponse-control, followed by - ve, then + ve, and the fastest was the war- ranted group. Only in the second grade did this order vary, the - ve reinforcement obtained a faster reSponse than the + ve group. The overall difference between grade and rein- 11) 721 The significant factor in the relationship of sex forcement type was significant: F( = 8.42, P = .005. to reinforcement type is that only the control groups varied from the warranted groups for both sexes. The boys and girls acted as an homogeneous group in their reaction to the experimental treatments. There was no significant difference between the warranted, + ve, and - ve groups. The Duncan Multiple Range Test of Time vs. Reinforcement Type and Sex shows an interesting order of ranked means. Figure 4 illustrates that the girls under — ve reinforce- ment obtained the second slowest mean response, whereas the boys under the same reinforcement obtained the second fastest response; but these differences were not signifi- cant at the .05 level. The overall variance between Reinforcement Type and Sex is significant (see Table 12): 7 _ _ F(=,-—2—5) — 4.52, P — .005. 43 TABLE ll.--Duncans Multiple Range Test, time by reinforcement type and grade. Entries that differ signifi- cantly from the left hand entry Entry Mean Entry Mean Warranted - second 0.30 Control second .34 Warranted first .36 Warranted Kinder. .38 — ve first .38 Control first .40 + ve Kinder. .42 - ve Kinder. .44 Control Kinder. .47 - ve second 0.31 Warranted Kinder. .38 Control first .40 + ve Kinder. .43 - ve Kinder. .44 Control Kinder. .47 - ve second 0.33 Control first .40 + ve Kinder. .43 - ve Kinder. .44 Control Kinder. .47 Control second 0.35 Warranted second .30 Control first .40 + ve Kinder. .42 - ve Kinder. .44 Control Kinder. .47 Warranted — first 0.36 Warranted second .30 Control first .40 + ve Kinder. .42 - ve Kinder. .44 Control Kinder. .47 + ve first 0.36 — ve Kinder. .44 Control Kinder. .47 Warranted Kinder. 0.38 Warranted second .30 - ve second .31 Control Kinder. .47 44 TABLE ll.--Continued Entries that differ signifi- cantly from the left hand entry Entry Mean Entry Mean - ve first 0.38 Warranted second .30 Control Kinder. .47 Control first 0.40 Warranted second .30 - ve second .31 + ve second .33 Control second .35 Warranted first .36 Control Kinder. .47 + ve Kinder. 0.43 Warranted second .30 - ve second .31 + ve second .33 Control second .35 Warranted first .36 - ve Kinder. 0.44 Warranted second - ve second + ve second Control second - ve Kinder. 0.44 Warranted first .36 + ve first .36 Control Kinder. 0.47 Warranted second .30 - ve second .31 + ve second .33 Control second .35 Warranted first .36 + ve first .36 Warranted Kinder. .36 - ve second .38 Control second .40 —.n Note: Significant level = .05. 45 .mouoom cooE mcflm: xmm an mmmxu ucoEoou0mcflmu mo COmwquEoo 4 .v ousoflm moEo m>om awkzl 3+ Jomhzoo on. SUNOOHS‘ NI 033d$ .Lb' 46 TABLE 12.——One way analysis of variance time by reinforcement type + sex. Source of . dF SS MS F P Variance Between categories 7 0.931 0.0990 4.53 .0005 Within categories 725 15.8530 0.0218 Total 732 16.5461 Dependent variable--time Category variable—-reinforcement type + sex -4. ___III-_. - _ A “Y.” .1}. lfliA_! UM “ I __.l 47 The high degree of variability apparent when con- sidering Reinforcement Type and Stimulus Type: F(%%) = 8.77, P = .0005, is once again due to the effect of the L + L stimulus type. Even the control group with the L + S stimulus achieved a significantly faster response than three out of four L stimulus-reinforcement groups and the control—sound group (see Figure 5). The Duncan Multiple , Range Test at the .05 level of significance shows that there is no significant difference between all the reinforcement type groups having the L + S stimulus (see Table 13). Again, referring to Figure 5, it will be noted that the sound control - ve and + ve groups are significantly slower than all but one of the remaining stimulus type reinforce- ment groups; namely, the light control group. Within each grade there were significant differences for each reinforcement type-sex group at the .05 level of significance. The slowest group in each grade was the control-girls, and they were significantly different from the fastest group in their grade. The fastest group within each grade varied; in the second grade, the - ve reinforcement boys had the lowest mean; in the first grade, the warranted boys; and in the kindergarten, the slowest times were recorded by the warranted girls. The Duncan Multiple Range Test of ranked means shows clearly that there was no pattern to the order of the kindergarten and first grade mean responses; that is, the means of those : j‘lig‘ \um- 48 . MOHOUm CMOE ocem: momxu msadeflum an mmm>u ucoswou0wcflou Mo comflumesoo 4 Marv szsfiomomzmm 3 0>|0>+ o D>+ 3 3 0>I w)... o m>l o m .w I. ozoow Fro... wa>._. won. 32 Ch .. mm on on 0? mg on. .m gunmen SUNOQEIS NI 033d$ .IH LLIITEST L r, :— I i I I , I . '_ \3 J 1‘; I I STA E IWT I’JNAL R31)” 133 HHCKSO I MICHrr,z\:» CTRLJC- III HAL 49 TABLE l3.-—Duncan Multiple Range Test, time by reinforcement type and stimulus type. Entries which differ signifi- cantly from the left hand entry Entry Mean Entry Mean Warranted L + S 0.30 Warranted Light .37 Warranted Sound .37 - ve Sound .38 + ve Light .41 Control Light .42 - ve Light .42 Control Sound .48 - ve L + S 0.31 + ve Light .41 Control Light .42 - ve Light .42 Control Sound .48 + ve L + S 0.32 + ve Light .41 Control Light .42 - ve Light .42 Control Sound .48 Control L + S 0.34 + ve Light .41 Control Light .42 - ve Light .42 Control Sound .48 Note: Significant level = .05. 50 two grades have no apparent hierarchy (see Table 14). The second grade, on the other hand, shows far more stability in performance. Eight of the t0p nine places in the ranked means are taken by the second grade; only the first grade warranted group differs from the pattern (they ranked sixth in the mean scores). Elm“. _ The Analysis of Variance for the variables rein— forcement type, grade, and stimulus type, shows that a significant variance does exist (see Table 15); but it is "e— EL."‘" “H" lai.' ? .. difficult to establish any clear pattern. There is signifi- cant variance with the L + S stimulus grOUps. The - ve reinforcement, warranted, and control second grade groups differ significantly from the kindergarten and first grade control groups for this stimulus type. Therefore, as far as the control groups are concerned, it would appear that grade is the prime cause of the variance in this set of results (see Table 16). The interaction of the variables, reinforcement type, sex, and stimulus type, again brings out the lack of within-group variance for those groups having the L + S stimulus (see Table 17). The girls' groups with the sound stimulus exhibited significant within-group variance. The warranted, + ve, and - ve reinforcement types were Significantly different from the control group at the .05 level (see Table 18). There was, however, no significant difference between the experimental groups. In the bOYS' 51 TABLE l4.-—Duncan Multiple Range Test, ranked means by reinforcement type, grade, sex. Entry Mean - ve second Boy .28 Warranted second Boy .29 + ve second Boy .31 Warranted second Girl .32 Warranted first Boy .32 - ve second Girl .34 + ve second Girl .34 Control second Boy .35 g Control second Girl .35 F + ve first Girl .36 I + ve first Boy .36 ; Warranted Kinder. Girl .37 i - ve first Boy .37 Control first Boy .38 Warranted first Girl .38 Warranted Kinder. Boy .39 - ve first Girl .39 + ve Kinder. Girl .42 - ve Kinder. Boy .42 Control first Girl .43 + ve Kinder. Boy .43 Control Kinder. Boy .46 - ve Kinder. Girl .46 Control Kinder. Girl .49 Note: Significant level = .05. 52 TABLE 15.-—One-way analysis of variance, time by reinforce- ment type + grade + stimulus type. sourSe 0f dF 55 MS F P Variance Between categories 35 3.430 0.098 5.21 .0005 Within categories 697 13.115 0.018 Total 732 16.546 Dependent variable--time Category variable--reinforcement type + grade + stimulus type 53 TABLE l6.-—Duncan Multiple Range Test, time by reinforce— ment type + grade + stimulus type. Entries which differ signifi- cantly from the left hand entry Entry Mean Entry Mean - ve second L + S 0.25 Control first L + 0.37 Control Kinder. L 0.40 Warranted second L + S 0.26 Control first L + 0.37 Control Kinder. L 0.40 Control second L + S 0.27 Control first L + 0.37 Control Kinder. L 0.40 Note: Significant level = .05. -- A 54 TABLE l7.--Duncan Multiple Range Test, time by reinforce- ment type + sex + stimulus type. Entries that differ signifi— cantly from the left hand entry Entry Mean Entry Mean Warranted Boy, L + S 0.28 Warranted, Girl, Light .36 Warranted, Boy, Light .37 - ve, Boy, Light .39 Control, Boy, Light‘ .41 Warranted, Girl, Sound .41 + ve, Boy, Light .41 + ve, Girl, Light .42 - ve, Girl, Sound .42 Control, Girl, Light .43 Control, Boy, Sound .44 - ve, Girl, Light .45 Control, Girl, Sound .52 - ve, Boy, L + S 0.30 Control, Girl, Light .43 Control, Boy, Sound .44 - ve, Girl, Light .45 Control, Girl, Sound .52 + ve, Boy, L + S 0.31 Control, Girl, Light .43 Control, Boy, Sound .43 - ve, Girl, Light .45 Control, Girl, Sound .52 - ve, Girl, L + S 0.31 Control, Boy, Light .41 Warranted, Girl, Sound .41 - ve, Girl, Sound .42 Control, Girl, Light .43 Control, Boy, Sound .44 - ve, Girl, Light .45 Control, Girl, Sound .52 Warranted, Girl, L + S 0.32 Control, Boy, Light .41 Warranted, Girl, Sound .41 + ve, Girl, Light .42 - ve, Girl, Sound .42 Control, Girl, Light .43 Control, Boy, Sound .44 - ve, Girl, Light .45 Control, Girl, Sound .52 r. urns}; Mimi; at. .w r: Nils-4‘53“?! I I TABLE 17 . --Continued 55 '7 hand entry Entries that differ signifi- cantly from the left Iintry Mean Entry Mean + Knee, Girl, L + S 0.34 Control, Boy, Sound .44 - ve, Girl, Light .45 Control, Girl, Sound .52 Corit:1:ol, Boy, L + S 0.34 Control, Girl, Light .43 Control, Boy, Sound .44 - ve, Girl, Light .45 Control, Girl, Sound .52 Cor11:3:ol, Girl, L + S 0.34 Control, Girl, Light .43 Control, Boy, Sound .44 - ve, Girl, Light .45 Control, Girl, Sound .52 Nod:ee:: Significant level = .05. he 56 ITLBLE 18.--Duncan Multiple Range Test, time by reinforce- ment type + sex + stimulus type. hand entry Entries that differ signifi- cantly from the left Entry Mean Entry Mean Ccuit:1:ol, Girl, Sound 0.52 Warranted, Girl, Sound 0.41 + ve, Girl, Sound 0.36 — ve, Girl, Sound 0.42 Cor1t:1:ol, Boy, Sound 0.44 - ve, Boy Sound 0.34 Warranted, Boy, Sound 0.34 Wairxrainted, Girl, Light 0.36 Warranted, Boy, L + S 0.28 Control, Girl, Sound 0.52 Warsrrainted, Boy, Light 0.37 Warranted, Boy, L + S 0.28 - ve , Boy, Light 0.39 Warranted, Boy, L + S 0.28 Control, Girl, Sound 0.52 Corlt:1:ol, Boy, Light 0.41 Warranted, Boy, L + S 0.28 - ve, Girl, L + S 0.31 Warranted, Girl, L + S 0.32 Control, Girl, Sound 0.52 + \re=,. Boy, Light 0.41 Warranted, Boy, L + S 0.28 Control, Girl, Sound 0.52 + ve , Girl, Light 0.42 Warranted, Boy, L + s 0.28 Warranted, Girl, L + S 0.32 Control, Girl, Sound 0.52 Control, Girl, Light 0.43 Warranted, Boy, L + S 0.28 - ve, Boy, L + S 0.30 + ve, Boy, L + S 0.31 - ve, Girl, L + S .0.32 Control, Boy, L + S 0.34 Control, Girl, L + S 0.34 Control, Girl, Sound 0.52 ‘QI- quABLE 18.-—Continued 57 ’7 Entries that differ signifi- cantly from the left hand entry Entry Mean Mean - ‘76:, Girl, Light 0.45 Warranted, Boy, L + S 0.28 - ve, 0.30 + ve, 0.31 - ve, Girl, L + S 0.31 Warranted, Girl, L + S 0.32 + ve, Girl, L + S 0.34 Control, 0.34 Control, Girl, L + S 0.34 I .I. H__ ' Noi:€e:: Significant level = .05. a... .L - .AL I 58 case, the control group differed from the warranted and the - ve reinforcement, but not from the + ve group. The light stimulus groups showed little within-group variance, but were significantly different from many of the L + S groups (see Table 18). Three important trends emerged in the second por- ' 11 tion of the investigation. The largest part of the vari— h ance was attributed to the difference between control and experimental groups, as was expected. It would appear, h the concepts of arousal may be useful in explaining the L___P relationship between the level of Speed for the control I groups and those for the experimental groups. After the standard instructions, the groups con- tinued their twenty trials. The control group with no verbal cues may experience boredom and fatigue far earlier in the test than the experimental groups. It may be that any spoken word in relation to the test will have some arousal effect on the subject, hence the lack of significant difference between the experimental groups. There was, however, a suggestion of a difference between the relative success of the + ve versus the — ve reinforcement type between the sexes. The girls who received the - ve reinforcement in the third five trials, followed by the + ve reinforcement in the fourth five, obtained the second slowest mean response. The boys, on the other hand, under the same conditions had the second 59 highest mean response. This would suggest that the girls' performance was adversely affected when confronted with the - ve reinforcement after the first ten trials. The boys were not adversely affected and even improved slightly on their mean of the first ten trials. This was the only result which shows a differentiation between the boys' and girls' performance. The remaining important factor in the results was the apparent homogeneity of the kindergarten and first E grade children's performance for a particular reinforcement L~___ type. These two grades obtained the same order of ranked means; and as a group, were significantly different from the second grade groups. As already reported, this was not the case when considering the effect of age or reaction time with no account taken of reinforcement type. The reason for this difference is open to conjecture. One possible reason is that by the time the child has reached the second grade, his rate of socialization is accelerating. He may be increasingly able to cope with the - ve rein— forcement, as indeed the Duncan Multiple Range Test shows (see Table 16). (This is the only grade in which the - ve reinforcement has a faster response than the + ve, although not at a significant level.) CHAPTER V SUMMARY AND CONCLUSIONS This experiment was designed to study the effects of various verbal reinforcement programs upon the simple reaction latencies of pre-pubertal children of both sexes. With reference to research findings, the following hypotheses were advanced: 1. Reaction time responses decrease with age; (in this case between kindergarten and grade two). There is no improvement in reaction time if no ver- bal reinforcement is given. Positive verbal reinforcement is more effective than negative verbal reinforcement in reducing the subjects' reaction times. The fact that males are regarded as having gener- ally faster reaction times than females will be accentuated by the reinforcement programs (for a particular age level). Reinforcement relating directly to performance, i.e., fast response warrants positive encouragement, 6O 61 is more effective than a set pattern of encourage- ment or censure. A secondary objective of this study was to deter- mine what type of stimulus produced the fastest response for this age range. To test the above hypotheses, a sample of three hundred and sixty-seven subjects was divided into sex-age groups and were assigned to an experimental group. Each subject was given twenty trials of a simple index finger lift reSponse, after receiving one of three stimulus modes. The first ten trials were given with no verbal cues; the second ten trials were conducted according to the experi- mental group he or she was assigned. The results of these experiments provide support for the general hypothesis that with an increase in age there is a corresponding increase in the speed of reaction time for this age range. The performance of the grades agreed with the traditional concept that the older children produce the fastest response. This is the case for both boys and girls. Thus, kindergarten girls have slower reaction times than first grade girls, and these in turn have slower reaction times than second grade girls. The boys were generally faster than the girls for each age group, but this was not at a significant level (see Table 3). From the results, it would appear that for a 62 particular age level, the boys and girls act as a homoge- neous group. The major differences in reaction time per— formance arose as a result of differences in age. Possibly the physiological-psychological mechanisms needed to pro- duce a reaction response are at a similar stage of develop- ment for boys and girls at this early age. As Pierson and Montoye [20] point out, possibly in two or three years, the boys will improve their reaction time significantly faster than the girls. The effect of the L + S stimulus type on the sex- age groups was clearly significant. As already reported, the L + S stimulus provoked a significantly faster response than the other two stimulus modes. Both the boys and girls produced their fastest reSponse when stimulated by the combination of the light and sound stimuli (see Table 7). Stimulus complexity appears to enhance a subject's reaction time in relation to the simple light and sound stimuli. Even at this early stage of motor development, there was no confusion concerning the response to more than one stimulus mode presented simultaneously. The difference between the simple light and sound stimuli on the performance of the subjects proved inconclusive at all age levels studied. The results of this experiment provides support for the general hypothesis that verbal reinforcement increases the speed of simple reaction time. All three 63 experimental groups had significantly faster responses when compared with the control group, which received no verbal cues of any kind. No clear pattern emerged as to the relative success of the experimental groups. There was no significant dif- ference between those subjects receiving the warranted, + ve or — ve reinforcement. However, for each sex, only the warranted group proved significantly different from the control groups. The Duncan Multiple Range Test for reaction time vs. grade also showed that the warranted group obtained the fastest reSponse for the kindergarten and first grade, but this was not at a significant level. The results suggest that in the kindergarten and first grade, the reinforcer "good" following a fast response by the subject, tended to keep the subject's responses close to his or her maximum speed. Similarly, if the subject's response Speed fell below that of the mean of the first ten trials, the mild censure brought the response speed up to that.of the mean value for the subject's first ten trials. The results were that the girls' performance was adversely affected by negative reinforcement. The boys, on thfii other hand, improved on the mean of their first ten trialfiS (although not at a significant level). From the results of this study, several avenues are Open for further research. Social status and its 64 interaction with reinforcement in the production of a reaction time, would seem a logical progression from the experiment described in this study. It seems that social and ethnic background may well have an effect on the way a subject reacts to a positive or negative reinforcement. In order to make a study feasible, it would be necessary to draw a small N from a variety of ethnic and social backgrounds. Firstly, the degree of socialization achieved and the psychological make«up of each subject, would have to be studied; then a test of their responses to a verbal reinforcement program could be conducted. In addition, the socio-psychological status of the subject could be correlated with the effectiveness of the reinforcement programs. Similarly, experiments using verbal reinforcement could be adapted to studies involving more complex motor skills where movement time and reaction time are measured. Such a study could have more practical value to physical educators who are teaching young children. In conclusion, the observed relationship between verbal reinforcement and reaction time has the likelihood of contributing to a more satisfactory theory of social reinforcement of children's behavior. BIBLIOGRAPHY 10. BIBLIOGRAPHY Ausubel, D. Theory and Problems of Child Development. New York: Greene and Stratton, 1958, p. 509. Botiwinik, J. and Thompson, L. W. "Components of Reaction Time in Relation to Age and Sex." Genetic Psychology, 1966, Vol. 108, pp. 175-183. Colgate, T. P. "Reaction Times of Individuals Reacting to Auditory, Visual, and Tactile Stimuli." Research Quarterly, 1968, Vol. 39, pp. 783-785. Crow, A. and Crow, L. Child Psychology. New York: Banes and Noble, 1959, p. 34. Elbel, E. R. "A Study in Variation in Response Times." Research Quarterly, 1939, Vol. 10, pp. 49-500 Goodenough, F. L. "The Development of the Reactive Process from Early Childhood to Maturity." J. Exp. Psychology, 1935, Vol. 18, pp. 431-450. Greenspoon, J. "The Reinforcing Effect of Two Spoken Sounds of the Frequency of Two Responses." Am. J. of Psych., 1955, Vol. 68, pp. 409-416. Havighurst, R. J. Human Development and Education. New York: Longmans, 1953, p. 139. Hodgkins, J. "Reaction Time and Speed of Movement in Males and Females of Various Ages." Research Quarterly, 1963, Vol. 34, pp. 454-461. JHolden, E. A. "The Effects of Rest, Reprimand, and Reward on Simple Reaction Time in Educable Retardates." Am. J. of Mental Deficiency, 1966, Vol. 1, pp. 427—432. 65 ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 66 Jones, H. E. "Reaction Time and Motor Development." Am. J. of Psych., 1937, Vol. 50, pp. 181-194. Lewis, R., Aikan, B., and Lichtenstein, M. "Reaction Times to Regularly Recurring Visual Stimuli." Perception and Motor Skills, 1964, Vol. 18, pp.’713-720. Mathur, V. K. "Sex Differences in Reaction Time." Indian J. of Psych., 1964, Vol. 39, pp. 31-34. McCullers, J. C. and Stevenson, H. W. "Effects of Verbal Reinforcement in a Probability Learning Situation." Psychology Rep., 1960, Vol. 7, pp. 434-445. Martinez, A. R. "An Investigation Into the Relation- ship Between Positive Verbal Reinforcement and Simple Reaction Time in Children." Ph. D. Thesis. University of Houston, 1966. Norrie, M. L. "Practice Effects on Reaction Latency for Simple and Complex Movements." Research Owen, W. A. "Effect of Motivating Instructions on Reaction Time in Grade School Children." Child DevelOpment, 1959, Vol. 30, pp. 261-267. Philip, B. R. "Reaction Times of Children." Am. J. of Psych., 1934, Vol. 46, pp. 379-396. Pierson, W. R. and Montoye, H. H. "Movement Time, Reaction Time, and Age." J. of Geron., 1958, Vol. 13, pp. 418-421. Pomeroy, J. E. "The Relationship Between Reaction Time and Intelligence." Child Development, 1938, Vol. 9, pp. 281-283. Pugh, L. A. "The Effects of Praise, Censure, and Noise on Electrodemal and Reaction Time Measures in Chronic SchiZOphrenic and Normal Women." Ph. D. Thesis. University of Oklahoma, 1965. IRarick, G. L. Motor Development During Infancy and Childhood. College PrinEing and Typing Co., Inc., Madison, Wisconsin, 1961, pp. 1-3. Pharick, G. L. and McKee, R. "A Study of Twenty Third Grade Children Exhibiting Extreme Levels of .Achievement of Tests of Motor Proficiency." .Besearch Quarterly, 1949, Vol. 20, pp. 142-152. 24. 25. 26. 27. 28. 29. 30. 31. 32. 67 Robinson, D. N. "Visual Reaction Time and the Human Alpha Rhythm: The Effects of Stimulus Liminance, Area, and Duration." J. of Exp. Psych., 1966, Vol. 71, pp. 16-25. Ribenstein, L Intersensory and Intrasensory Effects on Simple Reaction Time, Perception, and Motor Ryan, T. J. and Watson, P. "Children's Response Speeds as a Function of Sex and Verbal Reinforce— .1 ~ ment Schedule." Psychoanalytic Sci., 1966, Vol. 6, pp. 271-272. Shankweiler, D. P. "Effects of Success and Failure Instructions on Reaction Time in Patients with Brain Damage." J. of Comp. Physiol. Psych., 1959, Vol. 52, pp. 546-549. Sheikl, A. A. "Response Speeds as a Function of t”.— Different Reinforcement Conditions and a Ready Signal." Child Development, 1967, Vol, 38, pp. 857-867. Skinner, B. F. Verbal Behavior. Harvard University Press, 1948. Spence, K. W. Behavior Theory and Learning, Prentice-Hall, 1960. Vogel, P. "Battle Creek Physical Education Project." J. of Health, Physical Education, and Recreation, 1969, Vol. 40, pp. 25-29. Willgoose, C. E. Evaluation in Health, Education, and Physical Education. McGraw-Hill, 1961, pp. 15.-l6. APPENDIX APPENDIX RAW DATA PRINT OUT--EXPLANATION OF CODE BY COLUMNS Column 1 Subject Number Column 2 Grade - 0 = Kindergarten; l = lst Grade; 2 = 2nd Grade Column 3 Sex - 1 = Boy; 2 = Girl Column 4 Mean of lst 5 trials in l/100th's of a second Column 5 Mean of 2nd 5 trials in l/lOOth's of a second Column 6 Mean of 3rd 5 trials in l/lOOth's of a second; Control, Experimental Groups 3 and 4 Column 7 Mean of 4th 5 trials in l/lOOth's of a second; Control, Experimental Groups 3 and 4 Column 6 Mean of 3rd 5 trials in l/lOOth's of a second; Experimental Group 1 Column 7 Mean of 4th 5 trials in l/lOOth's of a second; Experimental Group 1 Column 8 Deck Number 68 Column 9 Column 10 69 Stimulus Type - l = Light; 2 = Sound; 3 = Light + Sound Reinforcement Type - l = Control; 2 = Experi- mental Group 1 (Warranted); 3 = Experimental Group 2A (+ ve); 4 = Experimental Group 2B (- ve) 7O N“ I II II I I -IIII I IIIn.-I.I-III "C(umxIC .I.-. .§(W~\(.UTIIII-I I (H m .cceLocMCbL(«Cll I.- Imm m ageing-.I--- dId-Iu den-III- MN. m“ L?CCM..O\IP.LQ.( 3 «Crux - IINHIIII III: --I I I II II --I I- IIII I II - - I I- II- II- I IINI -IIII- -II- I ”morn-U .INHHWI(4UNPI- I .II: NM rm CNCFMC “NCUN(f.(.«.l .IIIWNIIIIIIZILIIIIIIE-IIIIZIII;:IIII.E. II III. IME- - I I. COPMMCLMUHMHRU(F.III m“ N NUCCCCFVCFWCRCCQ Hm I m uflucuouworuccadfi. IIII; LN W .¢(R