9", [11 ' 7’. vs. : l0- I; V" . . . .tf. . . ”w.a.uw_.w.-w . ..-... ch. ... b). ..I nunvmwdh 0. .....vHoMR . Emfiun _ . in...“ a»; 2 1w...” .. ,. .m...‘..,..h..m,. ...... .... .5... .37.... . mmflm¢fiwzm 3.....c5n - tan: ...}..vxnpvwfi‘ ... ... What. . ...tvkfl-t. .v < . I. \ v .6... r“. .urflmwwm... . . ... ......kuwhw. M... ...: r in .....th. ...... . 1.12... .. .. ...... 3 ...»...r. . .3... n -. 16- v . 0.. .u. .... ...... 3...... 1| \v' Q“ ~ UL . . . ..u. . EEK ....-.. . _ . . : : . I}:....~..M..:£.........:..SE}... .. . ...qu. . $3.... . . . , . ...... 1.1.2.5.. ..........”5... . . I. . In..- ... . x.-. ......d... ... ...... m4. . ...... c ... . ... utc 0.. o‘cr...;.. V Id. 1‘4... .. - 9‘ . rt... 1’" .- ... ac . tow...v‘,rcv. .4v I! h: t. .0‘. -HUV.’ . Ifu. 500%» 7 Y... o .w..¥.v...“. V\A¢N1¢w\ NR wh.u~u~L‘r-..n‘qu~....n huh}. n71. ... .rI‘l.Ivr\.flt.. u A... a p ,. .I o. 3.l l\ f .. ll wq - (V . ‘1. - ~‘nt‘\. .3”? W: I .137“. ..xwruu... 51%.qu . . . .1. .. a . . 4...... ‘1. a y . ..z,.«...u...... ..." ......dwlsuhfid . . .. u . ... . . u... 1.. ind. «uni...» . . .. . ....xt ., .. . . : . ... . . l 4 so)... .10....34 hunt]? -. .. . ..f. o , ...”.mm- nu. Nun . . . . . . . y .. . .. .. . . I“ * ml? “......vhh‘Mwousx’h}... m.vII—A~.uv\bm\..- ...“.OHFt ...”... 2.1L! . Sui!!! :- .c. 5.. . . . . .. . u .. ; . . ._ . . .... . y . . .. .. no! 1.. - I (.37.. {rankfltuhufti I. .41.... :1}: ..."...q . . . . . . . . . , . - .n HID. ‘RV. U. .\ I!’t.o...o! \.0.vt\tg l I Q” I. 1V0“? . . 1 1...! fi..|.r.\..llu.uc.... l ......) . 3.....-) .1954 . x (3...? V - hr. 1111s. I..- 7‘ ....nuchdrai.-.x ... . ... ...; .. . . -- . . .' «Lu-fin...“ ....Il.v..l...1¢i ...}..uo... I . to. "5A... ..a .93 .35.. v. iii“... . . . . - 3:14. a ‘ (I ”‘3': -.vt..000«..ovv 3 o. 73.1 - 1:. .1 v n T. . Mu... .d..1..r\v..-.f¢.. 2.1.". Inwnniv .. .Jhku: n.....t.1.l.n..... . ...dhflr . . Iu‘ lurlv. . Vnkl. u. I . ....n . v \.-v . .II «.60- vdav .0... 1... u. _ ...-lf...! uri rrl..1 - ..rrmnm. ... . an: n... ......ur.._.»u..k..fi_.. ...... ..Wi... . ..u... Harmon» .. .n. I. I I h . ... uhnbr . . 011.11“. NH}... .... flmfi..ytdcl.fl. n n 3114...“: b2. . mm“. .... . .. I 5 . .. ... .. ., - Mt! l ‘0“...523331 I...” 4...... - 361...“. .o x ... . ... . - .3331. run...“ ... . ...... nu. . . . .r. . . . -.. u o. a 1.00... h. 09”....” ..l L» {11.4 «HRH—"...“. .uh’ v3.4: . .... -. 6...... . I . . .7: v n c n 01 . -‘a.u..vv. .«uv. . .n... . in. . u I. .II .I»-\‘} I . ..."le.r|nl ‘13... Lalul...’ glblg. 5gg1.9 on g. I 1411-..“.‘d... ‘ D 1.. ... L11 {viz} . x 1 . - - t 1 V. a! 1 I t A. 1 v It... . “1....-. ......w... ...- .....Ittflr......twnw.! ..rrx..............u 5.34.35...” . H... ..I’I......~ ... . u‘. .3 .. 1!“... II! 31:111.... 334.}... «gran .l .01. .lfia... .. (Iii. f7. Fulfilw..n.1l.rflt.....t..h.¥1..V.-.?..!.S\..v.2. .7. ... . U; ..I.VL I}...«v.~a\‘. . . ....u?‘ I E l If” H: ff? ‘r' 5“. 5.; Ih‘ :I I: - . ... .... .l 1 14. . . . . . u... ..1 .... r A i 3' _ I? s' . ... I: .. ...... t - 1 ...R) . {Mun ......r «. .fl .. .....wl...“ than- .IL . .. . M93 .1. I IV 1... .\ . v ..l..... v: a . . u .‘2 HIM.“ V0. ”.2392! ,v' v .l} ‘." l ' I new" il’l’ ' 61‘ .l . ‘ - "w, . 0.... .. W. . .0 J . .0 . . I: n ... .--. ...; .-. ..-....1...:... . .3. ...n!’ I41! .l.lv.¢.v... ‘ a. This is to certify that the thesis entitled CONJUGATE LATERAL EYE MOVEMENTS AND PERFORMANCE ON SPATIAL AND VERBAL REASONING TESTS presented by Eric T. Olsen has been accepted towards fulfillment of the requirements for M.A. degree inPsychology Major professor Date fly,7: $7 0-7639 CONJUGATE LATERAL EYE MOVEMENTS AND PERFORMANCE ON SPATIAL AND VERBAL REASONING TESTS By Eric T. Olsen A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1977 ".2. 7 {7' VrMJ e1 0 5 ABSTRACT CONJUGATE LATERAL EYE MOVEMENTS AND PERFORMANCE ON SPATIAL AND VERBAL REASONING TESTS By Eric T. Olsen Research suggests that in humans the left cerebral hemisphere mediates analytic and sequential thought processes, while the right hemisphere processes information simultaneously and mediates the perception of Gestalt—like relationships. One technique of studying this lateralization of functions is the analysis of conjugate lateral eye movements which are inferred to reflect activation of the contralateral cerebral hemisphere. In this experiment it was hypothesized that individuals who shift their gaze consistently in one direction on the horizontal plane also prefer the information processing mode(s) of the contralateral hemisphere. Seventy-six right-handed male undergraduates were defined as LM, RM or BD based on their eye movement responses to 15 high imagery and 15 low imagery questions. Subjects completed the Differential Aptitude Verbal Reasoning test, the Differen- tial Aptitude Space Relations test, and the Closure Speed test. Neither LM, RM, nor BD, regardless of how they were defined, showed significantly different performance on the tests. ACKNOWLEDGMENTS I would like to thank Fran Levine, Liz Pease, and Wendy Shapiro for their assistance in testing the subjects in this study. The members of my thesis committee, Dr. Terry Allen, Dr. Lauren Harris, and Dr. Albert Rabin provided valuable guidance as well as autonomy for me to progress at my own pace. Finally, Sam Plyler aided me with my "education" during the course of this research and to him I owe my continuing gratitude. ii TABLE OF CONTENTS Page LIST OF TABLES. . . . . . . . . . . . . . . . . . . . v LIST OF APPENDICES. . . . . . . . . . . . . . . . . . Vi Chapter I. INTRODUCTION. . . . . . . . . . . . . . . . . 1 Cerebral Lateralization of Cognitive Functions. . . . . . . . . . . . . . . . l Preferential Processing: Cultural and Individual . . . . . . . . . . . . . . . A Conjugate Lateral Eye Movements. Cerebral Basis of LEM. . . . . . \OONU'I Individual Consistency of LEM. Hemispheric Abilities of RM and LM . . . . 10 The Proposed Study . . . . . . . . . . . . 13 II. METHODS AND PROCEDURES. . . . . . . . . . . . 16 Selection of Subjects. . . . . . . . . . . 16 Experimental Setting . . . . . . . . . . . 17 Procedures . . . . . . . . . . . . . . . . 18 Statistical Analyses . . . . . . . . . . . 21 III. RESULTS . . . . . . . . . . . . . . . . . . . 25 Tests of Predictions: Full Sample . . . . 25 Active Mover Subsample . . . . . . . . . . 25 Proportional Score Based on High Imagery Questions. . . . . . . . . . . . . . . . 28 Comparison of LM and RM. . . . . . . . . . 31 iii IV. DISCUSSION. Experimental Hypotheses. Comparisons of LM and RM . The Closure Speed Test The Eye Movement Questions Bidirectional Subjects SUMMARY REFERENCES. APPENDICES. iv 35 35 38 39 A0 A2 AA A6 51 Table 7-10. 11. LIST OF TABLES Subjects' Total Number of Eye Movements Grouped by Direction. . . . . . . Performance of All Subjects on Verbal Reasoning, Space Relations, and Closure Speed Tests . . . . Analysis of Variance on Closure Speed Test Scores Performance on VR, SR, and CS tests by subsample of active movers. . . . . . Analyses of Variance of CS Scores for two methods of grouping subjects. . . . Performance on VR, SR, and CS, tests, eye movement score computed from spatial questions . . . . . . . . . . . . . . Analyses of variance, subjects grouped by Strong Mover Method . . . . . . Performance by LM and RM "Strong Movers," eye movement score computed for spatial questions . . . . . Page 23 26 27 29 3O 32 33 3A LIST OF APPENDICES Appendix A. Handedness Questionnaire. B. Feedback to Subjects. C. Selection of Eye Movement Questions D. The Closure Speed Test. E. Correclations of Eye Movement Scores with Aptitude Tests F. Eye Movement Questions. vi Page 51 52 5A 58 61 62 Introduction Cerebral Lateralization of Cognitive Functions Research with brain-damaged patients (Bogen, 1969), commissurotomy patients (Sperry, 1968; Nebes, 197A), and with normal persons employing EEG techniques (Galin & Ornstein, 1972) and tachistoscopic presentation techniques (Kimura, 1973) provides substantial evidence that the two cerebral hemispheres of the human brain play different roles in certain psychological functions. The evidence - suggests that the left cerebral hemisphere primarily mediates language and analytical thought processes, while the right hemisphere is specialized for visuo-spatial processes and the apprehension of Gestalt-like relationships. Experiments with commissurotomy patients shed light on the roles of the cerebral hemispheres in these cognitive processes. The primacy of the left hemisphere's role in language and analytical thought processes has been well documented (Gazzaniga & Sperry, 1967). The right hemiphere, once thought to play a minor role in most cognitive processes, has been shown to have some basic language capabilities (Gazzaniga & Hillyard, 1971) and to excel the left hemisphere in the performance of some tasks. Levy et al., (1972), based on their work with commissurotomy patients, suggested that the difference involves "... spatial synthesis for the right and temporal analysis for the left..." (page 61). Galin (197A) described the left hemisphere mode of information processing as "an analytical, logical mode for which words are an excellent mode" and right hemisphere processing as "a holistic, gestalt mode, which happens to be particularly suitable for spatial relations..." (page 573). In other words, the left hemisphere seems to analyze and differentiate in successive steps, while the right hemisphere seems to simultaneously combine information for its representation in visual imagery. Levy-Agresti and Sperry (1968) found the right hemisphere superior to the left in matching hand—held objects with representations of the objects laid out in two dimensions. The investigators reported that the left hemisphere (through its contralateral hand) seemed to analyze the details of the representation, while the right hemisphere seemed to visualize each representation as if folded up into a three dimensional object. Thus, the right hemisphere excels the left in processing spatial information and in constructing from partial sensory information a concept of the whole stimulus. Another research technique, the analysis of EEG activity during mental tasks, has recently provided further evidence for the lateralization of functions. Several studies (Galin & Ornstein, 1972; Doyle, Galin, & Ornstein, 197A; Morgan, McDonald, & MacDonald, 1971; Dumas & Morgan, 1975) have measured the relative levels of EEG alpha activity in the two hemispheres during tasks designed to engage primarily one or the other hemisphere. Alpha activity is associated with lowered cognitive activity, and its suppression ("alpha blocking") occurs during heightened cognitive effort. The studies cited above reported that the proportion of alpha in the right hemisphere to the total amount of alpha decreases during right hemisphere tasks and increases during left hemisphere tasks. In other words, alpha suppression, relative to the total amount of alpha, occurs while a particular hemisphere is engaged in the performance of tasks for which it is most specialized. Related evidence comes from the study of cortical evoked potentials in response to different types of stimuli. Verbal and nonverbal visual stimuli have been shown to elicit differential patterns of cortical evoked responses (Buchsbaum & Fedio, 1969; 1970) and a similar asymmetry appears to exist in auditory cortical evoked responses during speech perception (Wood, Goff & Day, 1971). The results of these varied lines of research apply primarily to right—handed persons in whom lateralization of functions is more nearly complete than in left-handed persons (Hecaen & Sauget, 1971). Moreover, the pattern of lateralization in females seems to differ somewhat from that of males. For example, there is some evidence that language and verbal processes appear to be more bilaterally represented in females than in males (Harris, 1975), possibly accounting for their greater verbal fluency yet poorer visuo-spatial abilities than males (Kimura, 1973). Thus, the variables of handedness and sex place some restrictions on the generalization of results from hemispheric lateralization studies. Preferential Processing; Cultural and Individual Bogen et a1. (1972) stressed a concept of preferential processing or the tendency to rely more on one mode of information processing than the other. They hypothesized that if the two hemispheres contribute in varying degrees to the solution of various problems, a person may come to depend more on one hemisphere than the other, depending on which cognitive mode his early cultural experience requires that he develop most fully. In a given culture or subculture, then, despite individual variation, there would be a central tendency in the relative reliance upon left and right hemisphere modes of information processing, which they term respectively the Propositional and Appositional modes. Bogen et al., propose that this relationship can be described in terms of a ratio of Appositionality to Propositionality, or the A/P ratio, a measure of relative ability in the two cognitive modes. In devising a study to test their concept of cultural hemisphericity Bogen et a1. suggested guidelines for testing these relative abilities. A test of the Appositional mode must be minimally verbal in presentation and response and must tap the right hemisphere's specialty in Gestalt-like relations and visuo-spatial performance. Conversely, a test of the Propositional mode must be highly verbal in presentation and response and must test ability in verbal and abstract thought processes. Of course, these guidelines only highlight the need for relatively "pure" tests of cognitive abilities. Moreover, the Appositional and Propositional modes are simply new labels for an old dichotomy. (See Bogen et al. for a review of various approaches to this dichotomy.) Bakan (1971) proposed a similar concept of hemisphericity based on studies of individual consistency in the direction of conjugate lateral eye movements, a phenomenon to be discussed in the next section. Conjugate Lateral Eye Movements Day (1964) first noted the phenomenon of conjugate lateral eye movements (LEM), a synchronized shift of gaze which occurs at the onset of reflective thinking. Day's early reports (1967; 1968) stimulated inquiries into the relationships between LEM and cognitive processes and personality variables. These lines of research rest somewhat on the assumption that LEM reflect activation of the cerebral hemisphere contralateral to the direction of eye movement. Cerebral Basis of LEM. Kinsbourne (1973) has been one of the principal proponents of the model in which LEM behavior is interpreted as an index of hemispheric activation. Human behavior is controlled by a structurally symmetrical nervous system and each of its two lateral divisions serves the contralateral half of the body and space for the input of stimuli and the output of responses. For example, the left hemisphere controls movement of the right hand. The operations of the two lateral divisions are coordinated by the flow of information along connecting commissures at various levels of the nervous system. When competing responses are generated by the two divisions, response priorities are established by reciprocal inhibitory interactions through the commissures. This inhibitory interaction amplifies the initial equivocal balance into a decisive one, thus insuring that the person's response is a unified and coordinated whole (Horridge, 1965). Within this model of the nervous system, each hemisphere (through its frontal eye field) controls the motions of looking in the opposite direction. The hemispheres operate in a reciprocal inhibitory relation- shit, i.e., looking in a particular direction results from the decisive resolution of the hemispheres' two opposing activities; eye movement in the vertical plane, however, is controlled equally by the two hemispheres (Crosby, 1953). In the case of damage to or destruction of one of the hemispheres, the intact hemisphere alone controls visual attention, resulting in neglect of attention to that part of the visual field contralateral to the impaired hemisphere (Kinsbourne, 1970a). Kinsbourne (1970a) incorporates into this model the evidence for the hemispheric lateralization of cognitive functions. Since the cerebrum is a highly linked system, any two given neurons are but a few synapses away from each other. This organization allows significant "cross talk" between the lateralized cognitive functions and the ipsilateral mechanisms (frontal eye fields) which control the distribution of visual atten- tion. For example, activation of the left hemisphere through analytic thought processes spills over into the left frontal eye field and results in spontaneous lateral eye movements to the right. Considerable supporting evidence exists for this model. Kinsbourne (1972) found that in right-handed persons verbal problems elicit LEM primarily to the right while spatial problems trigger LEM primarily to the left. Numerical problems, however, failed to elicit LEM in a consistent direction and left—handed persons, reflecting their lesser lateralization of functions, showed an equal number of LEM to the left and right regardless of the problem type. Other studies have revealed a similar pattern of LEM in response to tasks which engage left and right hemisphere functions. Kocel et al. (1972) found that verbal and arithmetic questions elicited more LEM to the right than did spatial and musical questions; the data 3 analysis, however, failed to differentiate the effects of arithmetic questions separate from verbal questions. Weiten and Etaugh (1974a) reported similar findings. Ornstein and Galin (197A) studied lawyers and artists and a group of non—specialized §S. The specialized subjects were selected to test the assumption that the vocational groups differed in their cognitive preference and performance. The lawyers and artists failed to differ in the pattern of their eye movements. In all Ss, however, verbal questions evoked more LEM to the right than spatial questions. Finally, Oscar—Berman et a1. (1973) reported that Wechsler Information and Vocabulary subtests questions elicited more LEM to the right than the left. Individual Consistency of LEM. A different line of research has examined the personality and cognitive correlates of individuals who show consistency in the direction of their LEM. Day (196A) and Duke (1968) reported that many persons can be classified as 'right- movers' (RM) and 'left-movers' (LM) based on their LEM responses during dyadic interaction. Libby (1970) and Bakan and Strayer (1973) reported that the individual's characteristic direction of LEM showed a test-retest reliability of .80 after a period of three days. Bakan (1969a) and Bakan and Svorad (1969) found that LM have greater resting EEG alpha and susceptibility to hypnosis than RM. Weiten and Etaugh (197Ab) reported that RM and LM score higher on the Scholastic Aptitude Test than persons who show no directional consistency of LEM. Harnad (1972) studied mathematics professors and graduate students and found that LM manifested greater creativity than RM. The test of creativity was the subject's colleagues' ratings of his creativity as a mathematician, a test of questionable validity. Bakan (1971) has summarized much of the research on individual consistency in direction of LEM and proposes that the person's LEM may provide an index of his dependence 10 upon the cognitive modes specialized in the two cerebral hemispheres, a proposal similar to Bogen et al.'s (1972) concept of preferential processing. The fact of individual consistency in direction of LEM seems to contradict evidence that the nature of mental tasks greatly influences the direction of LEM. Gur, Gur, and Harris (1975) resolved this inconsistency in an experiment in which they observed the LEM of right-handed male Ss under two conditions. When the questioner sat behind Ss verbal questions elicited right LEM and spatial questions elicited LEM to the left. However, when the questioner faced the same Ss they showed individual consistency in the direction of their LEM regardless of the type of question. Gur et a1. hypothesized that when the questioner sits behind the S, the experimental tasks are the most salient stimuli in the situation. When the questioner faces the S the presence of the questioner may be the most salient stimulus which arouses anxiety and causes the S to rely on his preferred cognitive mode. Hemispheric Abilities of RM and LM. If, as Bakan (1971) proposes, the consistency of LEM reflects a relative dependence upon and development of a particular hemispheric mode of cognition, do RM and LM differ in the level of their visuo-spatial and verbal and abstract thought abilities? 11 A review of the literature reveals the few studies performed have addressed this question only obliquely and imperfectly. Bakan (1971) found that RM score higher than LM on the mathematics section of the Scholastic Aptitude Test and Weiten and Etaugh (1973) confirmed this finding. In the latter study the investiga- tors also used a concept identification task as a test of left hemisphere ability and an inverted alphabet printing task as a test of right hemisphere ability. LM performed better than RM on the printing task and RM performed better than LM on the concept identification task. The use of the inverted alphabet printing task may have confounded the results, since it involves the manipulation of a symbol which could evoke interfering responses from the left hemisphere. Weiten and Etaugh (197Ac) predicted that RM would perform arithmetic calculations more quickly and accurately than LM, who were hypothesized to rely on the "inappropriate" right hemisphere for arithmetic calculations. No significant difference was found in the speed and accuracy of response between RM and LM. Subjects in this study were classified as RM or LM based on their LEM responses to questions with the experimenter seated behind the S. The results of the Gur et a1. (1975) study suggest that the situation in which the questioner faces the S is the most valid measure of LEM consistency. l2 Galin and Ornstein (197A) studied lawyers and artists whose vocations they inferred to emphasize the verbal-analytic and spatial—holistic modes respectively. (This inference is questionable; perhaps one must perform well in SQSS cognitive modes in order to succeed in a specialized vocation). Eye movements were scored on their horizontal and vertical components following Kinsbourne's (1972) report of an upward movement (indicative of bilateral activation) associated with spatial problem solving. The artists made more upward movements than the lawyers but the two groups failed to differ on the horizontal component. The subjects were tested also on the Writing-from-Memory and Kohs Block Design tests and and A/P ratio (Bogen et al., 1972) was computed from scores on these tests. The artists performed better on the blocks test and poorer on the writing test than the lawyers. Thus, although performance differences on the cognitive tasks were consistent with occupational diff— erences, performance didn't relate strongly to the eye movement patterns. This study has two shortcomings: Ss were selected by occupation rather than on the basis of LEM consistency, and their different performances on the experimental tasks could be attributed to previous practice implicit in their occupational activities. 13 The Proposed Study The current study proposes to overcome the short— comings of previous studies by identifying Ss' LEM consistency in an appropriate experimental setting and by using dependent measures more likely to reflect those abilities attributed to the left and right hemispheres. Dependent Measures. The Differential Aptitude Tests (Bennett, Seashore, & Wesman, 1966) are a battery of tests designed to measure several cognitive abilities. Two of the tests, Verbal Reasoning and Space Relations, appear to measure ability in left and right hemisphere cognitive functions. The test manual (Bennett et al., 1966) states that the Verbal Reasoning test measures the "ability to understand concepts framed in words" (page 1-6). The Spatial Reasoning test measures "the ability to visualize a constructed object from a picture of a pattern ... and the ability to imagine how an object would appear if rotated in various ways..." (page 1-8). Although the Differential Aptitude Tests were standardized on large samples of high school students, the norms indicate that the tests in question have sufficient ceiling to measure the range of ability first year college students' at Michigan State University which does not use extremely selective admission procedures. The Verbal Reasoning test contains 50 items; the mean score for male high-school seniors is 27.9 and the standard 1A deviation is 10.3. The Space Relations test contains 60 items; the mean score for male high—school seniors is 35.7, and the standard deviation is l2.A. These tests should show sufficient ceiling if subjects in this study are restricted to first-year college students. One right hemisphere function which the Space Relations test seems not to measure is the ability to apprehend part-whole relations. The Street Test (Street, 1931) is a test which appears to measure this ability. Now published as the Closure Speed Test (Thurstone & Jeffrey, 1966), the test involves the presentation of partially obliterated silhouettes which interferes with their recognition. The test requires one to generate from part of the stimulus a concept of the whole stimulus, and performance is impaired in individuals who have suffered right cerebral hemisphere injury (DeRenzi & Spinnler, 1966). Twenty-four items appear on the test and three minutes are allowed to complete it; the score is the number of items correctly identified. Thus the Verbal Reasoning, Space Relations and Closure Speed tests can be used to measure the left and right hemisphere cognitive abilities of S identified as LM, RM or bidirectional (BD). The following hypotheses and predictions are proposed: 15 I. It is hypothesized that the LEM consistency of LM reflects a preference for right hemisphere cognitive modes, among which is spatial reasoning. It is predicted that LM will perform significantly better on the Space Relations test than on the Verbal Reasoning test. II. It is hypothesized that RM prefer verbal/ analytic cognitive modes and thus perform better on a test of reasoning with words. Therefore, RM should perform significantly better on the Verbal Reasoning test than on the Space Relations test. III. Since BD hypothetically have no preference for either left or right hemisphere cognitive modes, their performance on the Space Relations and Verbal Reasoning tests will not differ significantly. IV. Since performance on the Closure Speed test theoretically is mediated by the right hemisphere, it is predicted that LM will excel on this test. RM will perform poorly because their preferred cognitive mode is antagonistic to the apprehension of part-whole relations; BD will perform better than RM but not as well as LM. METHODS AND PROCEDURES Selection of Subjects Subjects were volunteers from introductory psychology classes who received extra points toward course grades for their participation. Only right-handed males were invited to participate, because the lateralization of functions is less clear-cut in left-handed persons and in females. Inclusion of females or left-handed persons would have required a sample too large for the scope of this study. Handedness was measured using a questionnaire developed by Briggs and Nebes (1975) which can be found in Appendix A. The responses to the questionnaire were used to screen out left-handed subjects. No scoring cut-off point was used to exclude a subject, since there was little variance in the subjects' responses. The experiment was advertised as "Hemispheric Abilities." Potential subjects who wanted further information were told that they would take several standard aptitude tests and would answer a series of questions. They were also told that they would be given an explanation of the experiment at the end of their participation (See Appendix B). Eighty subjects participated in the experiment. Some subjects, however, suspected the importance of their eye movements and they were excluded from the sample. All subjects were routinely asked after 16 17 their participation if they suspected or knew of the importance of their eye movements. Four subjects reported awareness of the nature of the experiment and were excluded, leaving 76 subjects for the statistical analyses. Experimental Setting In the determination of subjects' eye movement patterns subjects were seated facing a female under- graduate research assistant. The subjects sat in an easy chair and rested their heads between two foam pads attached to the headrest. These pads kept their faces clearly visible over the closed—circuit television system. The lens of the television camera was just above the research assistant's head and, of course, was visible to the subjects. The walls behind and to the sides of the research assistant were covered with black fabric. Thus subjects faced a uniform visual background. A small hole was cut in the fabric through which the camera was pointed from an adjacent room. Observers sat in this room and recorded the subjects' eye movements while viewing them on a television screen. The subjects completed the Differential Aptitude tests and the Closure Speed test in a classroom near the experimental room prior to the observation of their eye movements. 18 Procedures A female undergraduate research assistant, different from the one mentioned previously, administered the tests to three to five subjects at a time. When the subjects arrived, the assistant explained that the experiment consisted of two parts and that the first part involved taking the standardized tests. The subjects then signed a form indicating their informed consent to participate. The Differential Aptitude subtests were then administered. These tests are timed. The Verbal Reasoning test was administered first; when all the subjects had completed the test the Space Relations test was given. After completion of this test, the subjects were given three minutes to do the Closure Speed test. After the tests were completed the subjects were tested individually to determine their eye movement patterns. Subjects who had to wait for their turn in the second part of the experiment were allowed to read or study in the room where the written tests had been administered. Each individual subject then was brought to the experimental room and introduced to the interviewing assistant (IA). He seated himself in the easy chair and the IA read the following instructions: In this part of the experiment I'm going to ask you a series of short questions, for example, word analogies and geographical questions. 19 As you see, this is an observation room; however, the windows have been covered. For the purposes of this study we will be monitoring your behavior, just through the video camera, but no recording will be made. We are simply interested in how people actually behave when answering these questions. We have you resting your head between the foam pads to facilitate viewing through the camera. You may glance around if you wish but try not to make any gross movements. 1 am going to ask you a series of thirty questions. Listen carefully to each question. You will probably find some of the questions easy and some more difficult. If you don't know the answer merely say, "I don't know." Keep your eyes open while you're thinking about and answering the questions. Are you ready to start? The IA then proceeded to ask the subject the series of thirty questions. These questions are found in Appendix F. They were developed from a pilot study in which participants rated the degree of imagery which the questions evoked. In the present study, half of the questions were categorized as low imagery or verbal questions. The other half were designated high imagery or spatial questions. Further information is contained in Appendix C. The IA was instructed to look up to the subject's face after reading each question. In the room adjacent to the interviewing room two observers recorded the direction of the subject's first eye movement after each question. The observers were 2O instructed to score eye movements only on the horizontal plane. Thus, a complete lack of movement and a perfectly vertical movement were both recorded as non—movement. For each subject three eye movement scores were tabulated: the number to the 1eft(L), to the right(R), and the number with no movement on the horizontal plane(N). For 51 of the 76 subjects two observers independently recorded the subject's eye movements. (Only one observer was present for the remaining 25 subjects.) The percentage of agreement between the two observers was computed for each of the three component eye movement scores. For example, each observer recorded the number of times a subject moved his eyes to the left. The total number of movements to the left was reported by each observer and the percentage of agreement was computed for left movement, or L. The same procedure was applied separately to the other component scores N and R. For L the observers achieved agreement 90% of the time, for N 85% of the time, and for R 83% of the time. There were two types of disagreement between the observers. The first type occurred when one observer reported no movement (N) while the other observer reported movement to the left or the right. This type accounted for 8A% of the disagreements between observers. The second type consisted of those instances in which one 21 observer reported left movement and the other reported right movement. This type accounted for the remaining 16% of the disagreements. For subjects whose eye movements were recorded by two observers, the component score reported by one observer, e.g. L, was averaged with the L score reported by the other observer. When the IA completed the series of thirty questions the subject was given a written explanation of the experiment (Appendix B) and sCme time to ask questions. Statistical Analyses The data on the subjects' eye movements afforded different ways of grouping the subjects for the statistical analyses. Researchers have typically established a cut-off point to exclude subjects who move their eyes in response to a few questions only. For example, a subject whose eyes moved in response to less than 67% of the questions would be excluded from the data analysis. For subjects included in the data analysis an eye movement proportional score was then computed using the formula R . L + R In the initial tests of hypotheses all subjects were included regardless of how often they actually moved their eyes. In the later statistical analyses only those subjects who moved their eyes in response to 2A or more of the questions were included. This cut-off point reduced the 22 sample from 76 to AA subjects. Table 1 summarizes the relevant characteristics of the sample. Initially the proportional eye movement score was computed by using subjects' responses to all 30 questions. Two alternative proportional scores were computed. The first alternative was based on the subjects' eye movement responses to the 15 high imagery questions only. The other alternative was based on the responses to the low imagery questions only. Two methods of grouping the subjects were used. One method involved ranking the subjects according to their eye movement score and then dividing this continuum into three equal groups. This method will be called the Equal Groups method. The second method involved the same ranking and then selection of an artitrary number of subjects at the extreme ends of the continuum. Barnat (1972) has described these subjects as "strong movers" and has suggested that they are most likely to show the characteristics of LM and RM. The remaining subjects are designated bidirectionals (BD). This grouping method is called the Strong Mover method. The Closure Speed Test score was divided into two subscores: Movement items score and Non-movement items score. The Movement items primarily depict human activities while the Non-movement items depict inanimate objects or 23 Table 1 Subjects' Total Number of Eye Movements Grouped by Direction g_ x 100% 0-33% 3A-66% 67-100% 0-33% 1A 12 2 3A-66% 18 7 2 fi§Ex100% 67-100% 12 9 O AA 28 A 2A other stationary scenes. The rationale for these subscores comes from the theory of Rorschach's test (Beck, 1961). RM have been described as inner-oriented and LM as externally—oriented (Barnat, 1972). In the language of Rorschach RM are introversive and LM are extratensive. Introversive persons are more likely than the extratensive person to perceive movement on the Rorschach cards. Thus, the subscores were used to test the possibility that LM also respond more quick than RM to movement items on the Closure Speed test. The specific breakdown of items can be found in Appendix D. RESULTS Tests of Predictions Full Sample The preliminary tests of hypotheses were based on the data from all 76 subjects who were grouped according to the Equal Groups method. The eye movement score for all 30 questions was used for ranking subjects. Subjects' scores on the Verbal Reasoning (VR), Space Relations (SR), and Closure Speed (CS) tests were converted to Z scores. The results are found in Table 2. Although LM did slightly better on the SR test than on the VR test, the difference between the means was not significant (t=.637, df=2A, p<.05). BD performed relatively poorly on both the VR and the SR tests but there was not a significant difference between the means (t=.35A, df=25, p<;05). RM did slightly better on the VR tests than on the SR tests; but, again the difference between the means was not significant (t=.816, df=2A, p<305). Finally, Table 2 shows subjects' performance on the CS test and Table 3 summarizes the analysis of variance of these scores. No significant differences among the groups were obtained (F=.l78; df=2.73; p<205). Active Movers Subsample In this part of the data analysis only those subjects who moved their eyes in response to 2A or more of the questions are included. The Equal Groups method and the 25 26 TABLE 2 Performance of All Subjects on Verbal Reasoning, Space Relations, and Closure Speed Tests R Verbal Space a Closure E_:_R x 100 N Reasoning Relations t test Speed x SD x SD t p 36 SD LM 251 1 (0-29%) .00 .961 .116 .886 .637 NS -.0A5 .983 BD L . l. —. 2 .88 . A NS .0 2 1.0 RM (62-100%) 25 .100 1.063e.086 1.231.816 NS .099 .971 3The t test is between the VR and SR mean scores. b NS=not significant 27 TABLE 3 Analysis of Variance on Closure Speed test Scores for All Subjects Sum of Source Squares df MS F p, eta square Between Groups 7.2A 2 3.62 .178 .8A .005 Within Groups 1A83.6A 73 20.32 Total 1A90.88 75 28 Strong Mover method of grouping were used. The results are found in Table A. The performance of LM, defined under the Equal Groups method, did not differ significantly on the VR and SR tests (t=1.35, df=l3, p<105). LM, defined under the Strong Mover method, again failed to perform significantly differently on the VR and SR tests (t=1.66, df=9, p<.05). One can note, however, that the performance differences are more pronounced than when the full sample of subjects is used. Table A also shows that neither BD nor RM performed significantly differently on the SR and VR tests, regardless of which method of grouping the subjects was used. The group mean scores on the CS test are shown in Table A and the summaries of the analyses of variance are shown in Table 5. Again, no significant differences among the groups were found. Proportional Score Based on High Imagery Questions Appendix E shows the correlations of the three types of prOportional eye movement scores with performance on the tests. The proportional score based on the high imagery questions correlates moderately with some of the test scores. Thus, the data were analyzed again by grouping subjects on the basis of their eye movement responses to 29 .mpmop mQOHumHmm momam 6cm mcfizommom Hmnpm> pom memos tsp cmmzpmp mfi pmtpup teem Ame. mos. mz Ho.H mma.fi mmm. mew. mmm. OH Agooaumev 2m moo.a moo.u m2 mme. mmm. sea.u mmm. omo.u am Ammeummv am new. eom.u m2 mm.H ‘Amm. Hmfi. emo.fi Amm.- OH ARHNIHV so Amo.a 5mm. mz mm.H mmm.a moo. moo.H mom. ma ANOOHImmV 2m mHo.H mac. m2 mam. 3mm. omo.u mmm. mmH.u ma Aagmummv mm ppm. mmm.u m2 mm.H mmw. Hmo. Amo.fi m:m.l 3H Axmmnov an mm x a p mm x mm M I I z OOH x m m a Ummqm mpwop p mcoapmHom wcficommom whamoao momam Hmnpo> mhm>oz m>Hpo< mo mHQEmmnsm mp wumoe mo 6cm .mm .m> co mocMEnompom : mqm¢8 Jerw Buoaqg sdnoaa Ienbg TABLE 5 Analyses of Variance of CS Scores for Two Methods of Grouping Subjects Equal Groups Method Sum of Source Squares df MS F p eta square Between Groups 65.98 2 32.99 1.66 .20 .075 Within Groups 81A.66 Al 19.87 Total 880.6A A3 Strong Movers Method Sum of Source Squares df MS F p eta square Between Groups 76.38 2 38.19 1.95 .16 .086 Within Groups Total 8OA.25 A1 19.62 880.63 A3 31 the high imagery questions. Again the Equal Groups and Strong Mover methods of grouping were used to classify subjects. The results are shown in Table 6. Again, although performance differences are apparent, neither LM, BD, nor RM differed significantly in their performance on the VR and SR tests. Table 7 shows the summary of the analysis of variance of CS tests scores, which also failed to show significant performance differences among the groups. Comparison of LM and RM Hypotheses I, II, and III focus on differential performance within subjects. Of further interest is whether LM and RM show differences across groups. Subjects were ranked according to their eye movement score on high imagery questions and grouped using the Strong Mover method. The summaries of the analyses of variance of the scores on the dependent measures are shown in Tables 7—10. Table 11 shows the results of the t-tests between mean scores of LM and RM on the dependent measures. RM surpassed LM on the VR test (t=A.95, df=9, p<.001), the CS test (t=3.58, df=9, p<.01) and on the Non—movement subscore of the CS test (t=5.A8, df=9, p<.001). RM also showed a slight but non-significant superiority over LM on the SR test. 32 .mpmop mCOHpmHom mowmm paw wcficommom Hmnho> pom memos esp coozpmn ma pmmp p onem Jerw Buoaas sdnoag Ienba Ame. OOO. m2 OOO.H OmH.H Omm. ONO. mmm. OH AHOOHmmAO OH.H mHH.- Oz ONO. OOO. mHH.1 OHO. OOO.- Om siHHHHmme OOO. HOH.- m2 OO.N ONO. OHH. OO.H Omm.u OH HHmmMOO OmO. OON. m2 HO.H AH.H OOO. OmO.H Omm. mH HHOOmeOV mOm.H OOH. mz OOm. mOH.H OOO.1 ONO. HOH.- mH HHOmmva AHO. Omm.- mz OOH. HOO. WOO. mOH.H mOH.- OH HammmOv Om m O O Om m Om m poodm mCOHpmHmm wCHcomwom z OOH x MIHIM whamoao momma p momqm Hmnpo> m mCOHumosa Hmfiumom Eonm umuSQEoo opoom pcoEm>oz ohm .mpmoe mo Una mm nm> co mocmEHompom w mqm<8 m: oo.mmmmH Hence O0.00m HO OO.MOOOH mOOOLO chsz Omo. mm. mm. Om.Oom m Om.mOO mosopo cmozpom opmsvm mum Q m m: up moLOSOm mo 83m oopsom mCOOpmHom momom MO mm.mon Hmpoe mm.mmH HO om.Hmmm masons cfisufiz OMH. AOO. Om.m NO.mmO m mm.mOO OOOOHO cmmzpmm ogmamm «no a m m: up mopmsvm mo 85m condom mcficommom Hmppm> Hmpoe Om.m HO om.mmm mosopm cHssz OOH. OO. mO.m OO.mm m OH.OO mOOOtO cmmzpmm mumswm mum m. m m: up mmpmsvm mo Ezm oopsom oncom pcoEm>oEIcoz pomdm mHSmOHo mO Om.omm proe m0.0m HO mO.mmO masonO chsz .mmo. Om. mO.H om.om m om.wm mascuw cmmzpom matadm mum m m m: up mogmnwm mo 55m oopsom Ummmm opzmoao ponpmz Lo>oz wCOLum hp poadopc mpomwnsm mocprm> mo mmmszc< OHIO mmqm