'IIIEI; . S ‘fi'ol h: It)“; :I‘rii‘ij IL'I: ".;‘.'.;I ”Ill“ I ‘V “Vim V I I l‘ "IW ' "I”; I VI I , V I I‘. I I: Alt}. V)“ V, q, .'VV' {his V ”I“! . . .Wfi’ WWW, { 9%)? 2 I"! ". [1015-5 ‘*' A A)" “A“? flfilgém A} ‘5‘ V'urml it” 'l‘ I zit; '::l ’1" A ‘. ,' -‘“ ‘3!“ I!!!" V5,} If ' u l IL. A |. . Inn. . ,2 . Iltlr‘:J“. !‘*‘;.§&T‘:f}°\3” 'kewl -: .I 1;, Win“. 'w VbflZVV-fi“ “ K 2 ”If“ 2 (H . .I' 2" 1'3“}. '13 1" "i" , l-:_‘ "VI'."':'I I ' ' ['0’ ,FII'fiIAVM \‘EJ. ' 'I, .J" 145‘.“ Jaw I 65"“:‘3 |l’)’\.}.‘ :kéR-FJ l_ . AV‘IUT'VX’VNI ”Jr‘R'Q‘MMA, "I'fifit12égE-d¢5 IV"OV:‘1_‘l; v I‘VV’VVVVVII'V'V'VEV ‘4," Iciiipfi .V... I g‘" :V\ V'I_I: "3"" ,r “NJ 1; KCII' :43: VV’V'VVVVfiiuV .' .; ’V,V. V. I' "'1 3‘... V'V.‘ HI.I' -V- l ' . VI .. I‘ .. .' 0 II . V I‘-' I '1 - 07;“; _ , .m; u: .w I -- ‘1'.“ II .'."' .'l-T';' Il‘ - ' - ‘-- 3'" ' .n A‘ F‘r-Il IIH :‘VVW s-"I.'-. 'III~"‘ "I 'I .‘ "' I' "2' . :‘ ‘2 , " '-I I'D I I " ' "'~ I' !' “XVII. .' III -‘ - - I'I'. "‘ I -" .Y'.'|‘| -' " I - .' " ' “12.5 I‘). . I... '<' ‘31:“ II '3" '." t'x -': I" .I . - .". I'I'I. II‘II‘. . I:‘. fit?! . '-\‘£'AI‘I‘:5:C‘I_.~:"-‘ 'ifl'l‘rvlhymfz'vViVV I3. '.2.'I':: VI I “I I V V_lk-| | :i I. I -- "l. ”IV“, ‘2 V ‘V'I. ' ' 3- ‘II 'V " V t" V VI‘ A h'i'fl‘ "It. '5... "jaunt-2. II . 'Vll' q». ‘ .III'JI‘AI'I ' 2' fI 332' " I'II'IIVI I .V.' " ,; , ' I-I "--II""'I"l"'. IV"3"I4""'7*I’I 'I‘ II" “2.3 -§I'“’i"‘NVI"'3I l I '|‘- I..,, ’ . \‘H‘.’V I ’ V‘u VII 'V I" l‘ 4 ' D 3 I .' It; It V. VI" V V I ‘ "1:118, -"'V L, '?: VI V? V:VIVVI,V VVII‘VV ‘l cVIV VVVVIIV 'fi‘uVV' IV. f. I) ,Ifl,’ KIIVIV'3 . IV' EA ' 634:”. ' V» " , I A I . A. 'l’ '1. k l’. I ”V ii!“ 'Hr": " "' 1“) r “U ' v I; 'v ' Ln" ‘2'; \ ' (“fl lf’:r‘:& ch: H ,8 I'VII\:I ' I' H ”II: 'I‘I V V'VIVVV' V . {V‘I \IIVIVVVVVIVV ‘I -~ . I I ' Irkéié ! )I :25? 511““ I V, 2. V' :ll [:III . . VVw'IV '5' > I! ,L'“' I",I' 2.‘I. '. ' :I‘"' I FMI' ‘ A. ‘3‘."I3 ”k '9" 'I’ II .I I . ’ \‘HVV‘V‘VII :‘|i:’IVVV'.mVI‘:V-IIVV V, :2}. VVNV‘E‘VV”? "'v’lej'i‘ IEVVV.I',2“|VVIV'V"-'V;1V' f9 :HPSVA \‘VIII‘V‘E‘fi W “Too“ VVVVVV ”’IEVIV drk }‘ 5"):an '¥‘ 'unV V I"I., , l‘ V lI-l "' V'V'I‘ {",' "It. Ic°:‘.“ II] l‘l ‘ l.‘ I ~' ‘ .IIIVr . ‘3 "g IV'3'VV .. W45 "I IIIV I3 ‘l'II :IIVVVVVVW 3M“, w:"Ii.-.I"I"l3 :”,; jI'rI‘ I 1'1, IV‘ E 32.13.: \(K rt? ' . ;I' ",I. ‘II. unh-MI I. I. I I'M I1 .3: ,. _I.V.'3 .~; “km: :3. '2' I. .\.I I: ~,VV ‘ 'IM'IJI: f'I‘l' A,“ RV'IVQH N II. V} 'Vl‘_§:IV | V'zrnzl :fl‘” r \ "f. Vj“'u-':’q |"""I'V‘r.'.‘ .“"(..‘|‘I 1‘ V I E , -. I I, I. V . . 4| ,5,‘ l- 41:“ VVI '59”: .V .. .- -I’.5.I‘.IIII I"; ,33; 3° "2 “$3: ~.E;~ I; #5” ' “" :.,2!! "1:;..I:..I35:3i?.f '. ~23” II 3‘33 Hg“ ( II II' I l. ‘ IAN "H N" ..\l "Wt" - ' "‘ 'II'I 'III ..I‘ I' ' . " ‘ I ' II-IIIIIV; : ("III " ' 313'!!- -.-‘I‘3333II-3fi33'.-I~1e‘I3II , I? (“2,33 .VI'I II“ I I W . "I"I’:-!I.' WIIILII-S‘ II' V, ML .,,.‘.' 1, V-HfV r3 VIVVIV. "II: V iz‘fi “24:33ng ~ 33) null ' .V%'1‘:l I .l h IV’V:l II I“ :5 William If” " "I".l"" ‘,' l', ':V|:|".' II, “In; 33." "3}: I33 I'IIII'II .. III 3 ';I':' I! I::'.!I,,- 333' II, III I,“' fill“ ' ll'hqt II V. V .‘IP'V “1‘93 V‘V“ l:‘II VII' ‘ 1'50!“ ,‘UH 41-0%.- l S I . , .2 ,,. ,2. V VI! LIN; ".I V [,OII‘ - {I III V(VV5 I'I .. l. I. {JV-55;“ VnV/V'IVV In,“ 3 VV'A'AI'I *3“ IVI ." .r . EXP-III ”' 1.6:...1‘III/ I‘,II " I".IIKII " ”2.132-“ IL: .253! iLHELSIS ‘M Date 0-7 639 LIERARY Michigan grate University This is to certify that the thesis entitled The Relationship Between Cerebral Hemisphere Specialization and Cognitive Ability in Young Adults of Superior Academic Achievement presented by Richard S. Lewis has been accepted towards fulfillment of the requirements for Ph.D. degnflgh,Neuroscience-Psychology jor professor Lauren J. Harris 8/l4/84 MSU is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES -—. V RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. THE RELATIONSHIP BETWEEN CEREBRAL HEMISPHERE SPECIALIZATION AND COGNITIVE ABILITY IN YOUNG ADULTS OF SUPERIOR ACADEMIC ACHIEVEMENT BY Richard Stephen Lewis A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1984 ABSTRACT THE RELATIONSHIP BETWEEN CEREBRAL HEMISPHERE SPECIALIZATION AND COGNITIVE ABILITY IN YOUNG ADULTS OF SUPERIOR ACADEMIC ACHIEVEMENT BY Richard Stephen Lewis A central assumption in current research on lateral specialization of cerebral functions is that cognitive ability is, at least in part, a reflection of how cognitive functions are laterally organized in the cerebral hemispheres. To test this assumption, an investigation was conducted of the relationship between performance on various cognitive tests and measures of lateralization, both indirect (as indexed by the subject variables, handedness and sex) and direct (as indexed by performance on tachistoscopic recognition tests). The subjects were 56 freshmen college students, including an equal number of right- and left-handed men and women. On the hypothesis that the cerebral organization--cognitive ability relationship is more pronounced in persons of high reasoning ability, only students were included who had American College Test entrance examinations scores of 27 or more (93rd percentile). All subjects were administered three verbal tests (the Vocabulary and Similarities subtests of the WAIS-R, and the Controlled Oral Word Association Test) and three spatial tests (Block Design subtest from WAIS-R, Piaget and Richard Stephen Lewis Inhelder's Water-Level test, and a mental rotation test). A tachistoscopically administered lexical decision task and face discrimination task were given to derive indices of cerebral lateralization. On the cognitive tests, the result was a sex by handedness interaction: On spatial measures, right-handed men outperformed left-handed men, whereas left—handed women outperformed right-handed women. On the verbal tests, left-handed men outscored right-handed men, and left-handed women outperformed right-handed women. The magnitude of the difference between left- and right-handers varied with sex and the type of task. On the face discrimination task, all the groups showed a left visual field (right hemisphere) advantage, with no apparent sex or handedness differences. On the lexical decision task, there was an overall tendency for subjects to show a right visual field (left hemisphere) advantage, but this effect varied with the sex of the subject; the men showing a significant right visual field advantage, the women failing to show a significant visual field advantage. This pattern of results for the verbal tasks is consistent with previous reports suggesting greater asymmetrical cerebral organization of language functions in males than in females. To test the hypothesis that cerebral lateralization underlies cognitive ability, each cognitive measure was regressed on each verbal and spatial lateralization index. The results indicate complexly related to and magnitude of the particular cognitive the subject. Richard Stephen Lewis that cerebral lateralization is cognitive ability with the direction relationship depending on the task, and on the sex and handedness of To my wife, Laura H. H- ACKNOWLEDGEMENTS I want to express my deepest gratitude to my entire committee for their input, support, and patience throughout my many dissertation projects. I am most appreciative to the Chair of my committee, Lauren Harris, for continuing discussions of issues that are reflected in this dissertation, and over which many bottles of wine were shared. I am appreciative and obliged to my other committee members: Tom Carr for his many insightful and thought- provoking discussions; Tony Nunez, not only for his input and his valuable contributions, but also for his supportiveness and many sets of tennis; and Lynn Clemens, who besides offering valuable criticism, provided a laboratory of students who were responsible for much of my enjoyment at Michigan State. I also want to thank Brad Rakerd, who is one of the most interesting individuals I have met, for serving as an ad hoc member of my committee. I want to take this opportunity to thank Peter Arnett, Susan Palin, and Brian Sandler for their invaluable assistance in data collection. Finally, I want to thank my many friends at Michigan State, without whom my four years in Michigan would have been less enjoyable. iii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O O O O O O O O 0 LIST OF FIGURES 0 O O O O O O O O O O O O O O O O 0 INTRODUCTION 0 O O O O O O O O O O O O O O O O O O O Handedness Differences . . . . . . . . . . . . . . Lateral cerebral specialization of right— and left-handers o o o o o o o o o o o o 0 Cognitive differences between right- and 18ft-hander8 o o o o o o o o o o o o o o o o o Handedness Subject Variables . . . . . . . . . . . Familial handedness . . . . . . . . . . . . . . Strength of handedness . . . . . . . . . . . . . Handwriting posture . . . . . . . . . . . . . . Studies controlling for handedness subject variables . . . . . . . . . . . . . . . . . . Sex Differences . . . . . . . . . . . . . . . . . Lateral cerebral specialization of males and females 0 O O O O O O O O O O O O O O O O O 0 Cognitive differences between males and females 0 O C C C C C O O O O C O O O O O O O Handedness by Sex Relationships . . . . . . . . . Confounding Variables . . . . . . . . . . . . . . Differences in methods . . . . . . . . . . . . . Reasoning ability . . . . . . . . . . . . . . . Direct Studies of the Relationship Between Cerebral Lateralization and Cognition . . . . . . . . . . . Requirements for a Proper Test . . . . . . . . . . conCIUSionS O O O O O O O O O O O O O O O O O O O The current Study 0 O O O O O O O O O O O O O O 0 vii viii 12 12 12 14 16 20 20 22 25 29 41 42 METHOD 0 O O O O O O O O O O I O O I SUbjects O O O O O O O O O O O O 0 Subject recruitment . Handedness measure . . . . . . . Familial handedness . Handwriting posture . Cognitive Tests . . . . . . . . . . verbal tests 0 O O I O O O O O 0 Spatial tests . . . . . . . . . American College Test . . . . . . Visual Half-Field Test . . . . . . Stimuli for visual half-field task . . Apparatus . . . . . . . . . . . . Procedure . . . . . . . . . . . . . Visual half-field test . . . . . RESULTS . . . . . . . . . . . . . . . Achievement and Intelligence Level Cognitive Variables . . . . . . . . Cerebral Lateralization Index . . . Face discrimination . . . . . . . Lexical decision . . . . . . . . Relationship Between Lateralization variables 0 O O O O O O O O O O O O ACT-English subtest . . . . . . . ACT-Mathematics subtest . . . . . ACT-Social Studies subtest . . . ACT-Natural Science Subtest . . . Mental Rotation . . . . . . . . . Embedded-Figures Test . . . . . . Water Level . . . . . . . . . . . Controlled Oral Word Associatio Similarities . . . . . . . . . . DISCUSSION 0 O O O O O O O O O O O 0 Test . Handedness and Sex Differences in Cognitive Ability I O O O O O O O O O O O O O 45 45 45 46 47 47 47 47 48 48 48 48 49 49 50 53 53 53 61 61 61 63 65 65 65 65 67 67 67 68 68 69 69 Visual Half-F1811} Measure 0 o o o o o o o o o o o o o 73 Face discrimination lateralization index . . . . . 73 Lexical decision lateralization index . . . . . . . 75 Relationship Between Cerebral Lateralization and cognitive Ability O O I O O C O O ’0 C O C C O O O O O 79 Increasing Left Visual Field Advantage-Increasing Cognitive Performance . . . . . . . . . . . . . . . . 82 Spatial lateralization index . . . . . . . . . . . 82 Verbal lateralization index . . . . . . . . . . . . 83 Increasing Right Visual Field Advantage-Increasing Cognitive Performance . . . . . . . . . . . . . . . . 84 Spatial lateralization index . . . . . . . . . . . 84 Verbal lateralization index . . . . . . . . . . . . 85 Remaining relationships . . . . . . . . . . . . . . 86 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . 88 APPENDIX A Cerebral Lateralization Measures . . . . . . . . . . 95 APPENDIX B Figures . . . . . . . . . . . . . . . . . . . . . . . 98 LIST OF REFERENCES 0 O O O O O O O O O O O O O O O O O 114 vi LIST OF TABLES ACT Composite Scores for Right-Handed Men and Wome Left- and n O O O O O O O O O O Congitive Test Performance for Left- and Right-Handed Men and Women, Adjusted for Composite ACT Score MANOVA for Sex Difference MANOVA for Hand Preference . . . . . . . . . . Discriminant Analysis of Handed Interaction . . Lateralization Index for Test, Using Accuracy and Scores as Covariates . Lateralization Index for Test, Using Accuracy and ACT Scores as Covariates a Sex by Facial Discrimination Composite ACT Lexical Decision Composite vii 54 56 57 S8 60 62 64 10. 11. 12. LIST OF FIGURES Regression of ACT-English Test on the Spatial Lateralization Index . . . . . . Regression of ACT-Mathematics Test on the Verbal Lateralization Index, for Left'handers o o o o o o o o o o o 0 Regression of ACT-Social Studies Test on the Spatial Lateralization Index . . Regression of ACT-Social Studies Test on the Spatial Lateralization Index, for Left-Handers . . . . . . . . . . . . Regression of ACT-Social Studies Test on the Verbal Lateralization Index, for Left-Handers . . . . . . . . . . . . Regression of ACT-Natural Science Test on the Spatial Lateralization Index . . Regression of ACT-Natural Science Test on the Spatial Lateralization Index, for women 0 O O O I O O O O C C O O O 0 Regression of ACT-Natural Science Test on the Verbal Lateralization Index, for Left-Handed Women . . . . . . . . . Regression of Mental Rotation on the Spatial Lateralization Index . . . . . . Regression of Embedded Figures Test on the Verbal Lateralization Index, for women 0 O O O O I O I O O O O O I 0 Regression of Water Level on the Spatial Lateralization Index . . . . . . . . . . Regression of Water Level on the Verbal Lateralization Index . . . . . . . . . . viii 98 99 100 101 102 103 104 105 106 107 108 109 13. 14. 15. 16. Regression of Water Level on the Verbal Lateralization Index, Regression of Oral Word Association Test for Women on the Spatial Lateralization Index, for Left-Handers Regression of Oral Word Association Test on the Verbal Lateralization Index . . . Regression of Oral Word Association Test on the Verbal Lateralization Index, for Women ix 110 111 112 113 INTRODUCTI ON The central aim of neuropsychological research is to elucidate the nature of the functional organization of the brain, particularly in the case of human beings, for such "higher" cognitive functions as language, memory, abstract reasoning, and visual-spatial perception. To date, much of the human neuropsychological research has focused on the left—right axis of the brain because of the striking asymmetrical functional organization of the cerebral hemispheres. The left hemisphere is specialized for solving verbal tasks, the right hemisphere for solving visual- spatial tasks (Bradshaw & Nettleton, 1981). The specialization of each hemisphere for solving particular tasks has been proposed to be due to qualitative differences in processing information (Bradshaw & Nettleton, 1981), with the left hemisphere specialized for analytic or sequential strategies, the right hemisphere for holistic or synthetic strategies. This division of hemisphere functions is one of degree rather than absolute, there generally being some overlap in functions between the hemispheres. The evidence for a verbal/nonverbal or analytical/ holistic distinction between the hemispheres comes mainly from two sources: clinical studies of patients with 1 2 unilateral brain lesions and experimental studies of normal individuals. The clinical evidence indicates that most patients with a lesion in the left hemisphere typically will have a deficit in verbal or analytical functions, whereas most patients with a right hemisphere lesion will have a deficit in nonverbal or holistic functions. In experimental studies with verbal stimuli, most normal individuals typically show a right ear, right visual field, and right hand advantage on dichotic listening, visual half-field, and dichaptic discrimination tasks, respectively. With nonverbal stimuli, a left ear, left visual field, and left hand advantage are usually obtained with the same techniques (see Appendix A for a description of these techniques). An assumption of much of the current research on lateral specialization is that cognitive ability is, in part, a reflection of how these two broad categories of cognitive functions are laterally represented in the brain. One prominent theory relating cognitive functions and cerebral lateralization has been proposed by Levy (1974). She suggested that lateralization provides for, or makes possible, the mutual functioning of the two qualitatively different methods of processing information——analytic and holistic--so that in the normally lateralized brain, analytical functions are lateralized to the left hemisphere, holistic functions to the right hemisphere. The implication of this model is that any deviation from this conventional design would lead to a change in cognitive strategy. Levy 3 predicted that this change in cognitive strategy would reflect itself in a difference in cognitive ability. Specifically, she assumed that individuals having a greater degree of bilateral representation of language, therefore more cortical area devoted to language functions, would have better language ability but poorer spatial ability, due to incomplete specialization of the right hemisphere. To test such a model, variability of functional asymmetry must exist among individuals, making it possible to test for covariance of cognitive ability as a function of lateral brain organization. The two dimensions most often associated with variations in patterns of cerebral lateral- ization are handedness (Herron, 1980) and sex (McGlone, 1980). These variables consequently have become the major parameters for testing the hypothesized relationship between lateralization and cognitive ability. Handedness Differences Lateral cerebral specialization of right- and left- handers. At one time, left-handers were assumed to be simply mirror-reversals of right-handers with respect to language functions (Broca, 1865; as cited in Harris, 1980). If this were the case, left-handers would differ from right-handers only in the direction, not the degree, of language lateralization, and therefore would not provide an opportunity to test for the relationship between degree of lateralization and cognitive ability. Current research, however, indicates that left-handers are not the simple 4 mirror-reversal of right—handers. Rather, left-handers display greater variation of functional lateralization than right—handers (Goodglass & Quadfasel, 1954; Herron, 1980). As a group, left-handers show a greater degree of bilateral representation of language function. This has been supported by clinical research indicating that language dysfunction is more frequent in left-handers than right- handers after hemisphere lesions (Hecaen, DeAgostini, & Monzon—Montes, 1981; Hécaen & Sauguet, 1971; Satz, 1980), and after right hemisphere electroconvulsive shock therapy (Warrington & Pratt, 1973). This group result could be obtained in different ways: (a) by two groups of left- handers displaying opposite lateralization patterns (i.e., one group showing left hemisphere dominance for language, the other showing right hemisphere dominance for language); (b) by individual left-handers showing less asymmetric representation of language; or (c) by a combination of both patterns. Support for the third alternative comes from a study using the sodium amytal technique (Rasmussen, & Milner, 1977; see Appendix A for a description of this technique). It was found that approximately 70% of left-handers were left-dominant for language, 15% were right-dominant for language and 15% showed signs of bilateral representation. The clinical findings have been supported by experimental studies with normal individuals using dichotic listening (Higenbottam, 1973; Kimura, 1967, Satz, Achenbach 5 & Fennell, 1967; Zurif & Bryden, 1969) and visual half-field procedures (Piazza, 1980; Zurif & Bryden, 1969). With both procedures, using verbal stimuli, left-handers, as a group, show smaller right ear (left hemisphere) and right visual field (left hemisphere) advantages than right-handers. The differences between left- and right—handers in language lateralization are reasonably well understood, at least in general terms. By contrast, the differences for nonverbal functions are much less clear. Clinical studies provide no evidence of greater bilateral representative of nonverbal functions in left—handers than in right-handers (Hecaen & Sauguet, 1971; Masure & Benton, 1983), indicating that verbal and nonverbal functions may be independently organized. Two experimental studies, one in the tactile modality (Gardner et a1., 1977), and one in the visual modality (McGlone & Davidson, 1973) support this conclusion. However, in one study of normal individuals, right-handers, compared with left-handers, showed a significantly greater left ear (right hemisphere) advantage for dichotically- presented environmental sounds, and melodies, as well as a greater left visual field (right hemisphere) advantage on face recognition (Piazza, 1980). Cognitive differences between right- and left-handers. To test her hypothesis that bilateral representation of language would be associated with poorer spatial ability, Levy compared left- and right-handed men on the Verbal and Performance scales of the Wechsler Adult Intelligence Scale 6 (WAIS) (Levy, 1969). According to Levy's reasoning, if left-handers, as a group, have a greater incidence of bilateral representation of language, they should not perform so well as right-handers on tests of spatial ability, such as the Performance Scale of the WAIS. The results confirmed her expectation; right- and left—handers did not differ on Verbal I.Q. (right-handers, 142; left- handers, 138) but right-handers had significantly higher Performance I.Q.s than left-handers (right-handers 130; left-handers, 117). In addition, the left-handers had a significantly larger discrepancy between Verbal and Performance I.Q. (25 points) than right-handers (8 points). Levy concluded that the left—handers' bilateral repre- sentation of language had interfered with visual-spatial functions. In her 1969 analysis, Levy had simply assumed that the Verbal and Performance Scales of the WAIS constituted tests of left- and right-hemisphere functions. Levy (1974) re-analyzed her data using purer verbal and spatial measures, items that grouped into a verbal or spatial factor on a factor analysis of WAIS items. As expected, the right-handers outscored the left-handers on the spatial items, but now the left-handers significantly outperformed the right-handers on the items from the verbal factor. The subjects in Levy's (1969) study were graduate science students at the California Institute of Technology, with expectedly high I.Q. scores. Although Levy argued that 7 there was no a priori reason that these students should not provide for a valid test of her hypothesis, her study has been criticized on precisely these grounds. Other studies, using less of a specialized sample, also have found that left-handers differ from right-handers in cognitive ability, but the direction of these differences is not very consistent among studies. In studies of reading ability, where left—handers have been compared with right-handers, the left-handers have been found to be superior, thus consistent with the Levy model (Jones, 1944), but also inferior (Orton, 1937; Wussler & Barclay, 1970), and no different (Coleman & Deutsch, 1964; Sabatino & Becker, 1971). Studies of visual-spatial ability show similar discrepancies. Again, left-handers have been found to perform worse than right-handers, thus consistent with the Levy model (Johnson & Harley, 1980; Miller, 1971; Nebes, 1971; Nebes & Briggs, 1974; Yen, 1975), but also have been found to perform better (Herrmann and Van Dyke, 1978, 1978; Peterson & Lansky, 1974; Sanders, Wilson, & Vandenberg, 1982), and no different (Annett & Turner, 1974; Briggs, Nebes, & Kinsbourne, 1976; Gilbert, 1977; Hardyck, Petrinovich, & Goldman, 1976; Nebes, 1976). Finally, in one study of pitch recognition, "mixed" left-handers (i.e., left-handers with weak left-handed preferences) outperformed "pure" left-handers (left-handers with strong left-handed preferences) as well as right-handers (Deutsch, 1980). Handedness Subject Variables Given that left-handers appear to be more heterogeneous in lateral organization than right-handers, much effort has been devoted to the identification of possible sub-types of left—handers underlying this heterogeneity and thereby perhaps enabling more accurate prediction of functional lateralization. Three variables have been named as potential indices: familial sinistrality, strength of handedness, and hand-writing posture. Familial handedness. Familial handedness is a reasonable index if handedness is inherited (Annett, 1981). However, even though the evidence of the heritability of handedness is itself strong, the relationship between familial handedness and cerebral lateralization is equivocal. A higher incidence of bilateral representation of language functions among individuals with familial sinistrality (FS+) has been reported in several types of studies, including clinical (Hecaen & DeAgostini & Monza-Montes, 1981; Hécaen & Sauguet, 1971), visual half-field (Andrews, 1977; Zurif & Bryden, 1969), and dichotic listening (Zurif & Bryden, 1969). In contrast to these positive reports, other studies report that the absence of familial sinistrality (FS-) in left-handers is associated with bilateral representation of language (Higenbottam, 1973; McKeever & VanDeventer, 1972; Newcombe & Ratcliff, 1973), or is unrelated to the direction of cerebral lateralization (Briggs & Nebes, 1976; Bryden, 1973; 9 Warrington & Pratt, 1973). In a study of 472 normal individuals (largest sample size to date), no main effect of FS was found, using a dichotic listening task (Orsini, Satz, Light, & Soper, 1984). Similarly, Annett (1982), in a review of visual half-field studies of handedness, concluded that familial handedness was not associated with visual field advantage. Strength of handedness. Hand preference is not a dichotomous variable. Instead, it is represented along a continuous dimension (Annett, 1972).' People are not always easily classified as either right- or left-handed; instead they display varying degrees of lateralized preference depending on the task and frequency of hand use. Measures of performance of hand skill, such as hand strength or speed of a coordinated hand movement, tend to result in a normal distribution of right- to left-handed ability (Annett, 1976). For this reason, it has been inferred that cerebral lateralization may vary as a function of degree of hand preference. The evidence, however, is conflicting. According to some reports, the stronger the left-hand preference, the more likely the right hemisphere will be dominant for language (as indexed by the degree of left visual-field and left ear advantages; (Lishman & McMeekan, 1977; Shankweiler & Studdert-Kennedy, 1975). Other reports show the reverse pattern, i.e., the weaker the left-hand preference, the more likely the left hemisphere will be dominant for language (Dee, 1971; Gilbert, 1977; Hecaen & 10 Sauguet, 1971). And still other investigations have failed to find any relationship between degree of handedness and lateral cerebral organization (Knox & Boone, 1976; McKeever & VanDeventer, 1977; Satz et a1., 1967). Handwriting posture. Perhaps the most unlikely proposed index of functional lateralization in left-handers is handwriting posture. Among left-handers, it has long been noted that many use an "inverted" posture. In contrast to the normal posture, where the hand is held below the line of writing and the pencil points toward the top of the page, the "inverted“ writer holds the hand above the line of writing, with the pencil pointing toward the bottom of the page. 'Common sense' would suggest that this posture simply represents the left-hander's accommodation to a left-to- right writing sequence. That is, the hand is held above the line so as not to block the view of the written line, or to rub across the fresh ink. However, because some right- handers and not all left-handers use an inverted posture, because the inverted style is far commoner among men than women, and because the inverted posture also occurs among left-handed Israelis, whose alphabet (Hebrew) is right- to-left (Levy & Gur, 1980), it seems unlikely that accommodating to a left-to-right writing sequence would be the fundamental factor affecting handwriting posture. Levy and Mandel (1974) suggested that handwriting posture may be related to cerebral lateralization (Levy, 1974). Levy and Reid (1976, 1978) followed up this 11 suggestion and investigated the relationship between handwriting posture and lateralization, using a visual half-field test of dot location and syllable identification. They found that language, as indexed by performance on a visual half-field test, was dominant in the hemisphere contralateral to the dominant hand for writing in individuals with non-inverted writing posture (i.e., the right hemisphere for left-handers, the left hemisphere for right-handers), but was dominant in the hemisphere ipsilateral to the writing hand in individuals with inverted writing posture (i.e., the left hemisphere in left-handers, the right hemisphere in right-handers). The reversed organization was found for spatial functions (i.e., spatial functions were represented in the ipsilateral hemisphere for non-inverted writers, and the contralateral hemisphere for inverted writers). Support for Levy and Reid's hypothesis has been mixed at best, with some studies supporting it (Herron, Galin, Johnstone & Ornstein, 1979; Parlow, 1978; Smith & Moscovitch, 1979), but many more failing (see Weber & Bradshaw, 1981, for a critical review, and Levy, 1982, for a reply). The validity of Levy and Reid's hypothesis thus can be questioned, but there most likely is some relationship between hand posture and brain organization, if not exactly of the sort originally envisioned (Allen & Wellman, 1980; Gregory & Paul, 1980; Parlow, 1978; Segalowitz, 1983; p. 148; Todor, 1980; Weber & Bradshaw, 1981). For example, 12 Smith and Moscovitch (1979) found support for Levy and Reid's hypothesis on a visual half-field measure but not on a dichotic listening measure. They suggested, therefore, that differences in neural organization between inverted and non-inverted writing postures are primarily in the visual or visual-motor domain. Studies controlling for handedness subject variables. One drawback in many studies of cerebral lateralization is that they fail to control simultaneously for familial handedness, strength of handedness, and handwriting posture. In response to this failing Bryden (1982, p. 173) examined the relative effects of all three variables on a visual half-field procedure with left-handed subjects. He found that none of these variables was particularly useful at predicting cerebral lateralization for left-handed subjects, although familial handedness appeared to be the best of the three predictor variables. Sex Differences The second potential source of major variation in cognitive performance and brain organization is sex. Lateral cerebral specialization of males and females. Evidence that the sexes differ in cerebral organization comes from a variety of sources (see Bryden, 1979; Harris, 1978; McGlone, 1980, for reviews). In clinical studies, men tend to show language dysfunction only after left hemisphere lesions, and women after both left and right hemisphere lesions. This suggests that language is bilaterally 13 represented in women, but asymmetrically represented in men. Similar conclusions have been made for visual-spatial functions. Experimental studies using standard lateral- ization techniques (i.e., visual half-field, dichotic listening, and dichaptic touch) with normal individuals also have found evidence of bilateral representation of language and spatial functions in women. For example, in a discrimination study of tachistoscopically projected faces, Rizzolatti and Buchtel (1977) found that women's reaction times were similar whether faces were presented to the right or to the left visual field. In other words, women showed no evidence of greater right-hemisphere processing for these visuospatial stimuli. Men, however, showed significantly faster reaction times when faces were presented to the left than to the right visual field, suggestive of a more asymmetrical representation of spatial functions. Many other studies using similar techniques, however, have failed to find sex-related differences (see Fairweather, 1982; Satz & Zaide, 1982, for reviews). Nevertheless, when sex-related differences are observed, they are almost always found in the same direction, i.e., indicating more bilateral representation of function in females than males. Thus, in contrast to differences in language functions between left- and right-handers, which are robust, the conflicting literature on sex differences suggests that any sex differences in cerebral organization are probably modest in size and comparatively fragile. There is, though, more 14 evidence for sex differences in the organization of spatial functions than for handedness differences. A further complication is that sex-related differences in lateral organization may be present for only particular cognitive tasks. For example, in a study of patients with unilateral brain lesions, sex differences were found on Block Design, a task requiring the manipulation of spatial coordinates, but not on the Street Gestalt Test, which is a visuoperceptual measure of closure (Lewis & Kamptner, 1983). On Block Design, men with left hemisphere lesions performed normally, whereas men with right hemisphere lesions showed significant deficits in performance. Women, however, performed similarly, irrespective of side of lesion, and their scores were between the extremes of the right- and left-hemisphere lesioned men. On the Street Gestalt Test, men and women with right hemisphere lesions were impaired to a similar degree, whereas men and women with left hemisphere lesions were unimpaired. Cognitive differences between males and females. Although the evidence for differences in cerebral lateralization between the sexes is not overwhelming, sex differences in cognition have been well documented. In an early review of the literature, Maccoby and Jacklin (1974) concluded that females outperform males on verbal tasks, whereas males outperform females on spatial tasks. Females' verbal superiority is most evident on productive speech tasks such as fluency, articulation, and grammar, but does 15 not seem to extend to verbal reasoning (Hutt, 1972). More recent reviews have supported these general conclusions but have gone further in identifying the kinds of spatial tasks where sex differences are most evident. The new evidence indicates that males are superior on spatial tasks that require manipulation of spatial coordinates such as mental rotation, and judgement of horizontality on the Water Level test, but not on spatial visualization measures such as Block Design (Linn & Peterson, 1983). One prominent explanation of sex differences in cognitive ability is that they result from differences in life experiences (see reviews in Harris, 1978; Maccoby & Jacklin, 1974). The male's superior spatial ability thus is hypothesized to result from greater freedom and encourage- ment to explore the environment, among other factors, whereas the female's greater verbal ability is said to result from a greater degree of verbal stimulation of parents and others. The evidence of sex differences in cerebral organization suggests that the cognitive differences instead are related to cerebral organization. We thus are faced with an apparent paradox. In the case of sex differences, the evidence for cognitive differences is stronger than the evidence for lateralization differences, whereas in the case of handedness differences, the evidence for lateralization differences is stronger than cognitive differences. 16 Handedness byASex Relationships Given the number of variables that might account for variation in cerebral organization (handedness, familial handedness, hand posture, strength of handedness, and sex), it is difficult to evaluate research that fails to take these variables into account, which consists of the majority of extant studies. Fortunately, several studies have examined the relationship of at least the two major variables-~sex and handedness--on measures of laterality within the same experiment. Bryden, Hécaen, and DeAgostini (1983) assessed deficits in language and spatial task performance in patients with unilateral brain lesions and found that sex, handedness, and familial sinistrality (FS) were all associated with cerebral organization. In right-handers, sex was found to be the variable most strongly related to cerebral lateralization, with women showing less left-hemispheric asymmetry for language and also less right-hemispheric asymmetry for spatial functions. For left-handers, however, FS was found to be significantly more influential than sex for predicting cerebral lateralization. The presence of FS was associated with greater bilateral organization of both verbal and spatial functions. Similarly, in a study of normal college students, Dagenbach (1983) found a complex relationship between sex and handedness variables on both a visual half-field and a dichotic listening measure of language lateralization. A four-way interaction was found between 17 sex, handedness, familial handedness, and hand posture. The direction of the interaction, however, was dependent on the modality of the task. For the visual field measure, simple effects tests indicated that left-handed FS- men had a larger right visual field (left hemisphere) advantage than FS+ men. However, left-handed and ambidextrous women (the latter showing equal or weak hand preference) showed an approximately equal degree of right visual field advantage, irrespective of the presence or absence of FS. Left-handed and ambidextrous men with a normal handwriting posture showed a smaller degree of right visual field advantage than men with an inverted writing style, whereas the opposite was true for left-handed and ambidextrous women. For the dichotic listening task, the relationship between sex and handwriting posture was similar to the visual half-field measure. However, among right-handers, FS- men had larger right ear (left hemisphere) advantages than FS+ men, whereas FS+ women had larger right ear advantages than FS- women. Among ambidexters, FS was related to larger right ear advantages for men and women. Among left-handers, FS+ men had larger right ear advantages than FS- men, whereas FS- women had larger right ear advantages than FS+ women. Bryden et al.'s (1983) and Dagenbach's (1983) results disclose obviously complicated but systematic relationships between sex, handedness and familial handedness. Other investigators, however, have found different patterns, or no patterns at all. Searleman (1980) for example, failed to 18 find any relationship between these same variables and a verbal dichotic listening task. Piazza (1980), however, found a complex relationship between sex and handedness. On a dichotic listening task, with words as stimuli, right- handed men showed a stronger right ear advantage than left handed men, whereas women showed no relationship between handedness and ear advantage. However, on nonverbal dichotic listening tasks, with melodies and environmental sounds (e.g., ringing telephone, bird singing), women showed a significant ear by handednesss interaction (left-ear advantage for right-handers, right-ear advantage for left—handers), but men did not show a significant ear or handedness effect. Piazza (1980) also used visual half- field measures of laterality and found a right visual field advantage for word recognition but no effect of handedness or sex. On a face recognition test, there was no effect for sex, but left-handers significantly outperformed right- handers, irrespective of visual field. In addition, there was a significant visual field by handedness effect. Right-handers showed a left-visual field advantage, whereas left-handers did not show a visual field advantage. In contrast to Piazza's results, Jones (1980) found that both sex and handedness were significant factors in a tachistoscopic face categorization task with subjects of undesignated age. Among males, FS- right-handers and left-handers showed a right visual field advantage, whereas l9 FS+ left-handers showed a left visual field advantage. Females did not show any visual field advantage. Birkett (1980), tested subjects, ranging in age from 16 to 42 years, on a tachistoscopic presentation of a form recognition, and did not find a main effect of handedness but did find a sex difference, with men showing a greater left visual field (right hemisphere) advantage. McGlone and Davidson (1973) found similar results on a tachistoscopic presentation of a dot enumeration task with secondary school and university students. McKeever and VanDeventer (1977), on the other hand, found a main effect of handedness on a verbal dichotic listening task, indicating a greater right ear (left hemisphere) dominance for right-handers. There also was a sex by handedness interaction on a verbal visual half-field task. A greater right visual field superiority was found for right-handed women and right-handed men. Briggs and Nebes (1976), using college undergraduates, failed to find a sex or handedness difference, on a verbal dichotic listening task, which led them to suggest that left-hemisphere dominance for language was more pervasive among left-handers and females than has been generally supposed. In support, Jones and Anuza (1983) using under- graduates, did not find a sex or handedness difference on a visual half-field presentation of mental rotation. In contrast, Lake and Bryden (1976) did not find a main effect of handedness on a verbal dichotic listening task but 20 did find a sex by familial handedness interaction. In women, FS+ increased the tendency to exhibit a left—ear advantage. For men, however, FS- increased the tendency to exhibit a left-ear advantage. McKeever (1984) reviewed eight studies of right-handed subjects carried out in his laboratory, and also found F8 to be a significant factor. In two studies using a visual half-field presentation of an object-naming task, he found a sex by handedness interaction in the same direction as Lake and Bryden (1976). However, this interaction seems to be task dependent, since he found F8 to be associated with greater lateralization for both sexes on word naming, color naming, clockface reading, object discrimination, object recognition and lexical decision (for nonwords) tasks. In contrast, Orsini (personal communication, 1984) not only failed to find a main effect of FS on a dichotic listening task, but also failed to find a sex by familial handedness interaction. Confounding7Variables In reviewing the evidence for variation in cerebral lateralization as a function of the variables of sex and handedness, the variable most consistently associated with cerebral variation is hand preference. The data are conflicting, however, for the other variables, familial handedness, degree of handedness, hand posture, and sex. Differences in methods. Many possible factors might have contributed to the conflicts and inconsistencies in the studies cited above. Where normal populations have been 21 used, these include the use of small sample sizes, failure to control for strategy differences, and the use of different tasks. The use of brain-injured populations creates further problems, including failure to control for the location, etiology, and extent of the lesion. And in much of the research, with both normal and clinical populations, a particularly troublesome problem is that lateralization is implicitly assumed to be a unitary phenomenon, meaning that all functions are lateralized to the same extent within an individual. There is accumulating evidence, however, that this is not the case, and that different functions are lateralized in varying degrees (Dagenbach, 1983; Gloning, Gloning, Haub, & Quatember, 1969; Hines & Satz, 1974; Herron et a1., 1979; Bryden, Hecaen, & DeAgostini, 1983). It thus is extremely difficult to integrate studies that differ in modality, nature of the task, and method for measuring laterality, since part of the variability in results across studies may be due to the fact that the lateralization of different functions is being measured. To make matters worse, it appears that depending on the stimulus parameters chosen, either "conventional" or reversed directions of visual field advantage can be obtained for both verbal and nonverbal stimuli (Sergent, 1983). This "reversibility" of laterality implies that cerebral specialization varies with the specific demands of the particular task and that it is consequently impossible to infer cerebral lateralization along all dimensions from 22 only a single measure of lateralization. Although this obviously complicates the assessment_of cerebral laterali- zation, it may ultimately provide clues as to how the cerebral hemispheres are specialized. For example, Sergent (1983) has proposed that the left hemisphere is specialized for processing higher spatial frequency information, the right hemisphere for lower spatial frequency information. Changing the spatial frequency, therefore, can affect the visual field advantage, independent of its verbal or nonverbal nature. It is also difficult to integrate results of studies using different techniques for investigating cerebral lateralization, such as performance of patients with unilateral lesions and lateralization measures in normal persons. Differences between studies can be attributed not only to differences in the laterality measures used but also todifferences in the task demands. For instance, clinical studies of lateralization tend to focus on speech production and comprehension, whereas visual half-field studies tend to focus on identification of a letter or a short, simple, familiar word presented visually. Only when task demands are more similar will it be possible to compare results using different techniques for studying brain laterali— zation. Reasoning ability. A second potentially confounding variable in most of the research on cerebral lateralization in normal individuals is reasoning ability. Although some 23 research has suggested that reasoning ability is an important factor in such studies (Briggs & Nebes, 1976; Geffen & Hochberg, 1971), it is usually not taken into consideration. Recent work by Harshman, Hampson, and Berenbaum (1983) suggests that it may indeed be a critical dimension. These investigators examined the relationships among cognitive ability, sex, and handedness in three different sample populations. The three factors proved to be related in a complex way. Each of the three populations, ranging in age from 16 to 38 years, was divided in half on the basis of reasoning ability. A different measure of reasoning ability was used for each sample: Raven Progressive Matrices (Raven, 1960), Inference Test (French, Ekstrom, & Price, 1963) and Nonsense Syllogisms (French et a1., 1963). In "high reasoners," right-handed men performed better than left-handed men on every one of 15 tests of spatial ability, including measures of mental rotation and spatial transformation, closure, and disembedding, whereas right-handed women performed worse than left-handed women on 12 of the same 15 spatial tests. In "low— reasoners," there was a trend for a reversed pattern of performance; now, right-handed men tended to perform worse than left-handed men, and right-handed women tended to perform better than left-handed women. For verbal ability, including measures of convergent production (e.g., vocabu- lary) and divergent production (e.g., verbal fluency), the reverse pattern was found. This time, in "high reasoners," 24 right-handed men tended to perform worse than left-handed men, whereas right-handed women performed better than left-handed women. Again, there was a trend for this pattern to be reversed for "low reasoners" so that the right-handed men performed better than the left—handed men, and the right-handed women performed worse than the left- handed women. Harshman et a1. (1983) did not propose any explanation of this complex relationship among sex, handedness, and reasoning ability. But whatever the explanation, the results at least suggest that a neurological factor must be at work at some level. Certainly, it is difficult to see how socialization influences alone could account for these findings inasmuch as we would have to postulate the highly unlikely situation in which the environment would favor spatial more than verbal ability in right-handed, high reasoning males, left—handed, high reasoning females, left-handed, low reasoning males, and right-handed, low reasoning females, but would favor verbal more than spatial ability in left-handed, high reasoning males, right-handed, high reasoning females, right-handed, low reasoning males, and left-handed, low reasoning females. Although the complex relationships found in the Harshman et a1., study have not yet been explained, they at least suggest that the use of intelligence as a moderator variable may help to clarify some of the discrepancies in previous research. For example, Yen (1975) and Sanders, 25 Wilson, and Vandenberg (1982) reported sex by handedness interactions in studies of spatial ability involving large sample sizes. Yen reported that for high school students, left-handedness was associated with low spatial performance in males but not in females. Sanders et a1., however, found that in subjects 14-20 and 35-60 years old, left-handedness was associated with high spatial scores in males, and with low scores in females. Harshman et a1., (1983) presented evidence suggesting that the Yen study used individuals more like the "high reasoners" in the Harshman et a1., study, whereas the sample in the Sanders et a1., study more closely resembled Harshman et al.'s "low reasoners." Thus, left- handedness appears to be associated with lower spatial scores in males but not females, in both Yen's subjects and in the "high reasoners" in Harshman et al.’s study. However, left-handedness is associated with high spatial scores in the males in Sanders et al.'s study and in Harshman et al.‘s "low reasoners." Note that the results of Levy‘s (1969) study with highly intelligent graduate students now can be seen to be consistent with the pattern of results found among the high reasoning males in Yen's and Harshman et al.'s studies. Direct Studies of the Relationship Between Cerebral Lateralization and Cognition Most of the studies reviewed so far have had either one or the other of two objectives: to assess, by means of clinical or experimental tasks, the functional 26 specialization of the cerebral hemispheres as a function of subject sex and handedness, or to assess the cognitive ability of individuals presumed (on the basis of their sex or handedness) to differ in cerebral lateralization. Few studies, however, make both determinations at the same time, i.e., to determine empirically whether cerebral laterali- zation, as indexed by conventional laterality measures in normal individuals (visual half-field, dichotic listening, and dichaptic touch), is actually related to cognitive ability. We turn, then, to a review of those studies that have specifically addressed this question. As noted earlier, Levy's (1969, 1974) cognitive crowding hypothesis, based on her research with right- and left-handed men, has been the most prominent of the various lateralization-cognition models. More recently Levy and Gur (1980) have refined the model as applied specifically to the question of sex differences. According to this new formulation, when the left hemisphere is dominant for language (the model case), verbal functions will tend to be bilaterally represented in females (as in the original model), but spatial functions will tend to be bilaterally represented in males. However, in the statistically rarer case when the language-dominant hemisphere is on the right, the reverse pattern of organization is predicted, with bilateral representation of spatial functions in females and bilateral representation of language functions in males. Whether there is any such complementary relationship between 27 lateralization of language and lateralization of spatial functions still remains to be seen, however. For example, Bryden, Hecaen, and DeAgostini (1983), in a study of the relationship between aphasia and spatial dysfunction in patients with unilateral lesions, concluded that language and spatial functions are organized independently, not complementarily. In any case, within this new theoretical framework, bilateral representation of either verbal or spatial functions would be expected to lead to superior ability of that function, and a relative deficit of the other function due to incomplete specialization (incomplete because of less cortical area devoted to that function). Cognitive ability thus is seen as the joint product of degree and direction of lateralization rather than as the product of degree of lateralization alone. Note that within this new framework, if handedness is an indicator of the language-dominant side of the brain, as Levy and Gur (1980) suggest, then a sex by handedness interaction exists that predicts the same direction of verbal and spatial superiority found among Harshman et al.'s I"high reasoners." That is, on tests of verbal ability, non-inverted, right-handed women, with bilateral repre- sentation of language, will outperform non-inverted, left-handed women, with bilateral representation of spatial functions, whereas on spatial tests, the left-handed women will outperform the right—handed women. The reverse direction will hold true for men, such as on tests of verbal 28 ability, non-inverted, left-handed men, having bilateral representation of verbal functions, will outperform non- inverted right-handed men on verbal tasks. Right-handed men, having bilateral representation of spatial functions, will outperform the left-handed men on spatial tasks. From the time the first of these models of the lateralization-cognition relationship was proposed (Levy, 1969), there have been numerous attempts to put it to test, involving several different methods. One method, which was reviewed earlier, involves comparisons of right- and left- handers (pp. 5-6), males and females (pp. 12-13), or both groups in combination (pp. 20-23) on measures of cognitive skill. This method thus uses sex and handedness as indices of cerebral lateralization. A second method uses a more direct index of cerebral lateralization such as performance on a dichotic listening or divided field test, and relates this index to cognitive performance, irrespective of sex or handedness. A third method combines the previous two methods by comparing male and female left-and right-handers on lateralization and cognitive performance measures. By now, the literature is vast. The results, however, are best described as mixed, with some studies reporting support, others not (see Harris, 1978). Among the negative studies, for example, is a recent report by Nichelli, Manni, and Faglioni (1983). These authors, seeking to test the hypothesis that sex differences in cognitive performance reflected sex differences in lateralization, concluded 29 instead that cognitive proficiency and lateralization were dissociated. After examination of at least a major part of the literature, it seems that the inconsistencies, to a large degree, may reflect less on the merit of the hypothesis than on the lack of any clear methodological requirements for a proper test of the hypothesis. The extant studies, that is, vary radically in procedures, choice of tasks, and other potentially critical variables. What is needed are criteria for what constitutes a proper test of the lateralization- cognition hypothesis (Lewis & Harris, in preparation). Requirements for a Proper Test To properly test the relationship between lateraliza- tion and cognitive performance, the following criteria should be met. In each instance, it is assumed that minimal psychometric standards (e.g., adequate test reliability and validity, sensitive measures of cerebral lateralization, and adequate sample size) have already been met (e.g., see Berenbaum & Harshman, 1980). Although sex and handedness measures have been used as indices of cerebral lateraliza- tion, these subject variables provide weak and indirect indices, at best. Therefore, attention will be directed to studies using more direct measures of cerebral laterali— zation such as dichotic listening and divided visual field techniques. The necessary criteria are as follows: 1. Assessment of cognitive ability. The first requirement is that the tests of cognitive ability must be 30 appropriate for assessment of individual differences, so that subjects will show an adequate range in ability. Without an adequate range of performance on a cognitive test, there is no hope of finding differences in cerebral lateralization underlying cognitive performance. The requirement of variability in performance can be made with respect to individual differences or group differences. The analyses used must be able to reflect variability in cognitive ability. For example, if there are differences in cerebral lateralization between different groups of individuals (e.g., between sexes or handedness groups), then the cognitive task chosen must be sensitive to cognitive differences between these groups. 2. Independence of cognitive test and lateralization Eggt. The assessment of subjects' cognitive performance must be independent of the tests of lateralization. Otherwise, the independent (lateralization) and dependent (cognition) variables are confounded, meaning that factors such as memory load (Moscovitch, 1979) and quality of the stimulus (Sergent, 1983) may be related to both laterali— zation and cognitive proficiency, thereby providing a spurious relationship between cerebral lateralization and cognitive performance. Another reason for independent measures is that laterally administered measures may provide information regarding only the performance of individual hemispheres, which may bear no relation to proficiency when both hemispheres are working together. Presumably, the 31 appropriate measure of cognitive ability involves a measurement when one is allowed to use all of one's cognitive and cerebral resources. 3. Variation in absolute magnitude of lateralization. For a proper statistical test of the relationship between degree of lateralization and cognitive ability, the subjects must show an adequate range of performance on the lateralization measure, i.e., from more bisymmetrical to more asymmetrical. It is for this reason, as already mentioned, that most tests of the hypothesis have compared left-handers with right-handers, or women with men (often without independent assessment of lateralization). The comparison, however, can be within a sex or handedness group, which might constitute an even more powerful test, since it would preclude the operation of other, potentially confounding factors (Harris, 1980). As with criterion 1, depending on the type of analysis used, variation of absolute magnitude of lateralization could apply to either individual or group effects. 4. Variation in direction of lateralization. To take account of Levy and Gur's (1980) reformulation according to which both degree and direction of lateralization affect the relationship between cerebral lateralization and cognitive ability, subjects should vary in direction as well as degree of lateralization. Again, this criterion could apply toward either individual or group analyses. Given criteria numbers 3 and 4, summary statistics must allow the joint analysis of 32 direction and absolute magnitude of cerebral lateralization. Arranging data in a scatter plot would seem to be a reasonable first step. 5. Independent assessment of language and spatial lateralization. Finally, in light of evidence that language and spatial functions may be independently, rather than complementarily, organized (Bryden et a1., 1983), cerebral lateralization of language and spatial function would have to be independently assessed; language from a lateralized measure of language functions, spatial from a lateralized measure of spatial functions. In other words, it would be inappropriate to study only the relationship between cerebral lateralization of language and performance on a non-lateralized test of spatial ability, since the measure of language lateralization would not necessarily provide any measure of how spatial functions are laterally represented. A similar problem is that different spatial or verbal functions may be lateralized to varying degrees (Samar, 1983). That is, the relationship between the lateralization of a particular spatial function may or may not be related to a particular spatial ability. Additional considerations. Furthermore, since different modalities appear to be independently organized (Dagenbach, 1983; Hines & Satz, 1974), the modality of lateralization and ability measures should be the same. Finally, since verbal and spatial ability are not unitary concepts, attention should be directed toward the type of 33 verbal and spatial task used in both lateralization and ability measures. With these criteria as guidelines, let us review some of the recent literature so that we might see to what extent the proposed requirements have been met. It should be noted that many of these experiments could not meet certain criteria inasmuch as they were not designed to do so in the first place. We begin with the Nichelli, Manni, and Faglioni (1983) study, whose authors have claimed to find evidence against the hypothesized relationship between lateralization and cognitive ability. At the outset, we should note that the authors are unclear whether their aim was to test a degree or a direction of lateralization model. The method used was to measure men's and women's accuracy and reaction times on a visual half—field dot discrimination task. The result was a significant visual field by sex interaction. Consistent with previous literature, men showed a significant left visual field advantage, and women showed no visual field advantage. The null result for women, however, was found to have resulted from the mutually cancelling effects of two subgroups showing opposite visual field advantages rather than a weaker or smaller visual field difference for individual women. That is, whereas most of the individual men showed a left visual field advantage (13 out of 14), half the women showed a right visual field advantage (7 out of 14), while the other half showed a left visual field 34 advantage, the magnitude of the difference in each case equal to the difference found for the men. (The reason for this finding is unclear; possibly it reflects the particular demands of the task or the nature of the sample selected.) Then, to investigate the relationship between lateralization and cognitive proficiency, Nichelli et a1., analyzed the errors (omissions and false alarms) on the visual half-field task, as a function of sex of subject and visual field. The result was that neither the main effect for sex nor the sex by visual field interaction was significant, meaning that men and women did not differ in proficiency on the visual half-field test. Nichelli et a1., therefore concluded that the dissociation between proficiency and lateralization is challenging for proponents of the lateralization-cognition hypothesis, and that their "data in fact demonstrate that females, whilst being as proficient as males in a given task, can nevertheless show a different asymmetry in performance" (p. 630). According the the criteria suggested earlier, to what extent does Nichelli et al.'s experiment constitute a proper test of the lateralization-cognition relationship? As a test of the lateralization-cognition relationship as applied to the question of sex differences, the dot localization test was clearly not an appropriate choice for the assessment of cognitive ability (criterion 1), since 35 there is no independent evidence of any sex differences in absolute performance on this measure. The use of one and the same measure--proficiency on the visual half-field test for dot localization-~for assessment of lateralization and of cognitive proficiency violated the criterion of independent measurement (criterion 2). The requirement of a range of lateralization from low to high was not met (criterion 3). As already noted, the women consisted of two groups showing opposite visual field advantages of equal magnitude to each other and to the men as well. Since the two groups of women and the group of men also did not differ in proficiency (accuracy), Nichelli et a1., felt justified in rejecting the hypothesis that spatial lateralization is related to spatial proficiency. In fact, since all three groups were equally lateralized (only the direction of lateralization differed), then even if the other criteria had been met, Nichelli et a1., would have succeeded in testing only the direction, not the degree, part of the lateralization-cognition hypothesis (i.e., demonstrating only that direction of lateralization is not related to proficiency on a lateralized dot discrimination test, not that there is no relationship between degree of lateralization and cognitive ability). Conceivably, degree of lateralization, irrespective of direction, is associated with proficiency on dot discrimination. In this respect, their study can be said to have met criterion 4, although this evidently was not their aim at the time. (That is, 36 they did not construe their experiment as a test of a direction model.) Finally, criterion 5 is not applicable here, since Nichelli et al., used one and the same measure for cerebral lateralization and cognitive performance. The Nichelli et a1., experiment thus fails to meet several of the criteria suggested for a proper test of the hypothesized relationship between lateralization and cognition, either in its original or its more recent formulation. Do any other investigations meet the proposed requirements more successfully? Let us consider a few other studies for purposes of illustration. Sasanuma and Kobayashi (1978) found that men performed better than women on a lateralized presentation of a line orientation task (Benton, Hamsher, Varney, & Spreen, 1983) and on the same task had a left visual field advantage compared to the women's lack of a visual field advantage. The Sasanuma and Kobayashi experiment satisfies criteria 1 and 5. However, like Nichelli et a1., Sasanuma and Kobayashi fail to meet criterion 2 by not using a measure of spatial ability independent of the lateralization measure. Nor were criteria 3 and 4 met, since Sasanuma and Kobayashi were not specific about the direct relationship between visual field preference and absolute performance on the lateralization measure (presumably because their intention was not to assess the relationship between the two measures). 37 Fennell, Satz, Van Den Abell, Bowers, and Thomas (1978) examined the relationship between verbal dichotic listening and visual half-field lateralization measures and performance on Block Design and Thurstone's Primary Mental Abilities Spatial Abilities Test (PMA). High school and college students were divided into three categories based on the results of the lateralization tests. Group 1 consisted of subjects demonstrating the strongest left hemisphere advantage; group 3 of subjects showing the weakest left hemisphere or a right hemisphere advantage. Group 2 consisted of subjects showing a left hemisphere advantage between the extremes of groups 1 and 3. This experiment satisfies criteria 1 and 2. However, it fails to meet criterion 5 because only a verbal lateralization measure was related to spatial ability. It also fails to meet criteria 3 and 4 because degree and absolute magnitude of lateraliza- tion are confounded. Groups 1 and 2 differ in absolute magnitude of lateralization, whereas group 3 differs from the other groups in absolute magnitude and direction of lateralization. Fennell et al., found no relationship between cerebral lateralization and performance on the spatial tasks. However, when the means are analyzed, trends seem to appear between either degree or direction of lateralization even though group 3 contained subjects with both visual field advantages (albeit weak right ear and right visual field advantages). 38 Piazza (1980) and Samar (1983) correlated performance on visual half-field procedures with performance on measures of cognitive ability. This method would not provide an adequate measure if the lateralization-cognition relation- ship was a complex non—linear function, or if only absolute magnitude or direction were important. Therefore, criteria 3 and 4 are not met. Both studies, however, meet criteria 1, 2, and 5. Although neither Piazza nor Samar found significant correlations using this method of analysis, Samar did find significant correlations between evoked responses on a spatial task and other cognitive measures. McGlone and Davidson (1973) satisfied criteria 1, 2, 4, and 5 in an experiment which investigated the relationship between performance on a divided visual field test of dot enumeration and performance on the PMA. They found that subjects with a right visual field advantage had the lowest scores on the PMA. However, they failed to consider the absolute magnitude of cerebral lateralization (criterion 3). Hannay (1976) found that women with a left visual field advantage on a nonverbal tachistoscopic presentation had significantly higher scores on the Block Design subtest of the WAIS than women with a right visual field advantage. For men, direction of visual field advantage and Block Design score were not significantly related. This study satisfies criteria 1, 2, 4, and 5 but fails to meet criterion 3 by not considering the relationship between absolute magnitude of lateralization and spatial ability. 39 Zoccolotti and Oltman (1978) found that among 18- to 30-year-old men, those who did well on the Rod-and-Frame and Embedded-Figure tests (field-independent subjects) showed the expected significant right visual field advantage in reaction time to tachistoscopically-projected letters, whereas men who did poorly (field-dependent subjects) showed no hemifield differences. A second study confirmed this finding and revealed a similar difference for a "right- hemisphere" task. Men with high perceptual disembedding scores showed a significant left visual field superiority in tachistoscopic face discrimination, whereas men with low scores showed no significant hemifield difference. The authors concluded that perceptual disembedding ability (with its presumptive reliance on visual-spatial ability) is related to the degree of segregation, or separation, of functioning between the hemispheres, as indexed by the degree of perceptual asymmetry on the tachistoscopic test. This experiment satisfies criteria 1, 2, 3, and 5, but not 4 (effect of direction of lateralization). Waber (1977) correlated the absolute magnitude of an index of lateralization, derived from a verbal dichotic listening task, independent of direction, with performance on verbal and spatial tasks. The correlations were described as being very small. Her study satisfies criteria 1, 2, and 3. It also meets criterion 5 by relating the lateralization index to verbal ability, but it violates criterion 5 by relating language lateralization to 40 performance on spatial tasks. Criterion 4 is not met, since the experiment fails to consider the relationship between direction of lateralization and cognitive ability. Hines and Shipley (1984) correlated performance on verbal and nonverbal tasks with percentage of correct responses from both the left and right ear on a dichotic listening test. This experiment satisfies criteria 1, 2, and 5 (when dichotic listening performance is related to verbal ability, although it violated criterion 5 when language lateralization is correlated with spatial ability). This analysis failed to meet criteria 3 and 4, since Hines and Shipley ignored the dimension of lateralization altogether by looking separately at the proficiency of each hemisphere on dichotic listening. A common failure in all of the aforementioned studies is not analyzing the data so as to include information about both direction and absolute magnitude of lateralization. There is nothing inherently wrong with analyzing for either direction or degree of lateralization, but by doing so it necessitates ignoring negative results of the lateralization-cognition model. In this survey of studies relating cerebral lateralization to cognitive ability, only two studies satisfied all five criteria. One is by Birkett (1980), who found both linear and non-linear relationships between a nonverbal measure of cerebral lateralization and various spatial tasks. The other is by Kraft (1983), who, in a 41 longitudinal study, also found linear and non-linear relationships between a verbal and nonverbal dichotic listening index of lateralization and subtests from the Wechsler Intelligence Scale for Children-Revised. These later investigations, therefore, strengthen the suggestion that both direction and absolute magnitude of lateralization should be evaluated in the same study. In summary, when investigations designed to test the hypothesized relationship between cerebral lateralization and cognitive ability are examined against the criteria named above, the evidence certainly does not justify the negative conclusion reached by Nichelli et a1., (1983). Rather, those investigations meeting most or all of these criteria can be seen to lend the hypothesis at least a modest measure of support. Conclusions In trying to integrate the conflicting results found in studies of individual differences in cerebral lateralization and cognitive ability, this review has identified several problems. First, very few studies have simultaneously controlled for potentially relevant variables such as sex and handedness. In those studies that have done so, conflicting results have been reported. A promising development in this area, however, is the identification of reasoning ability as a moderator variable. Reasoning ability has largely gone uncontrolled in neuropsychological research using normal individuals, and this may have 42 contributed to the discrepancies found among studies (Harshman et a1., 1983). In addition, no studies have investigated how cerebral organization might vary as a function of sex, handedness, and reasoning ability; it is hoped that controlling for reasoning level will produce more consistent results in studies of individual differences. Second, very few studies have adequately addressed the issue of how cerebral lateralization is related to cognitive ability. Although conflicting data have been reported, improvements can be made by following the criteria suggested for a proper test, and by controlling for reasoning ability. Third, it will not be possible to integrate experi- mental studies with normal individuals and clinical studies with brain-lesioned patients until it can be assured that the tasks used among studies make comparable cognitive demands. The Current Study The existing literature is obviously confusing and inconsistent to an extreme degree. To try to clarify the major issues, the current study addressed the following objectives: 1. The first objective was to determine whether the relationship found by Harshman et a1., (1983) between sex, hand preference, and reasoning ability on verbal and non- verbal measures of cognitive ability could be confirmed in a new sample. Only the relationship among "high reasoners" 43 was addressed, since the interaction between sex, hand preference and cognitive ability was most prominent among “high reasoners," and since it was expected that low reasoning subjects (i.e., subjects with below—average I.Q. scores) would have been hard to obtain from currently available subject populations. Therefore, the subjects were Michigan State University freshmen with high academic achievement, as indexed by ACT (American College Test) college entrance examination scores. Only students scoring in the top 15% of entering freshmen at M.S.U. were used. The subjects were tested on a variety of verbal and non- verbal measures. Familial handedness, strength of handedness, and handwriting posture were measured, but the primary independent variables were sex and hand preference. 2. The second objective was to investigate the relationship of sex and handedness to functional cerebral lateralization, as indexed by performance on a verbal and nonverbal visual half-field test. If intelligence interacts with sex and handedness on measures of cognitive ability, as Harshman, et al.'s (1983) findings suggest, then it will be difficult to account for this complex pattern of results as being exclusively a product of environmental experience. If, however, this complex relationship is, in part, the result of neurological factors, as Harshman et a1., (1983) suggested, then sex, handedness, and reasoning level may also relate to indices of cerebral lateralization. Controlling for reasoning level, therefore, might clarify 44 the study of individual differences and cerebral lateralization. 3. As stated in the introduction, the ultimate goal of neuropsychology is to understand the relationship between the functional organization of the brain and behavior. Toward this goal, the third objective of this study was to investigate the relationship between cerebral lateralization and cognitive ability. As we have seen, the few studies that have directly investigated this relationship have reported conflicting results. Again it is hoped that by controlling for reasoning level, a more consistent picture will emerge. It is also hoped that the current study will serve as the basis for addressing the third limitation mentioned above (i.e., the lack of similar cognitive demands in tasks used in experimental and clinical studies). METHOD Subjects Subjects were fifty-nine 18- and 19-year-old M.S.U. freshmen. Two subjects were excluded from the study because of incomplete data, and one subject because he was not focusing on the central fixation point during the visual half-field task. Fifty-six subjects, therefore, comprised the final sample, including equal numbers of right- and left-handed men and women. All subjects reported normal or corrected to normal vision. Subject recruitment. Subjects were recruited from letters written to on-campus freshmen residents having ACT composite scores of 27 or above (the upper 15% of entering M.S.U. students), corresponding to the 93rd percentile of the national norms. There were 6,689 freshmen at M.S.U., of whom 820 (upper 15% and living on campus) comprised the pool. Initial contact with the students was made through the Office of the Provost for Admissions and Curriculum in order to respect the confidentiality of these scores. Of those, 337 students (41%) responded, of whom 301 (31%) agreed to participate in the study. After applying the criteria for determination of handedness, it was possible to fill each sex by handedness cell with 14 subjects. 45 46 To verify that high ACT scores were an index of high reasoning ability, I.Q. scores were estimated from the three subtests of the WAIS used in this study (Vocabulary, Similarities, and Block Design). The average estimated I.Q. score was 123, which places the subjects, on average, 1 1/2 standard deviations above the mean, or at the 93rd percentile of I.Q. scores. Handedness measure. The subjects' handedness was measured by Briggs and Nebes' (1975) modification of Annett's (1970) hand preference questionnaire. The original questionnaire asks for information about hand preference for 12 action items. The modification asks for further information about the strength of hand preference on each item. The test is scored as follows: An "always" response receives a score of 2, a "usually" response 1, and a "no preference" response 0. Right hand preference is scored as a positive value and left-hand preference a negative value. The total score is obtained by the sum of scores for all 12 items. Subjects with a positive score of 20 or above and with no familial history of left-handedness were classified as right-handers. Subjects who wrote with their left hand and had a negative score were classified as left-handers. The criterion was made less stringent for left-handers because they tend to display greater variability in their hand preferences. 47 Familial handedness. Familial handedness was assessed by having each subject report the hand each parent and sibling uses for writing. Handwriting posture. Handwriting posture was assessed by having each subject identify, from a set of four pictures depicting left- and right-handed inverted and non-inverted handwriting postures (from Levy & Reid, 1976, p. 337 (Figure 1)), the posture that best described his/her own posture. Cognitive Tests The following verbal and nonverbal tasks were selected so as to cover a range of cognitive functions. Several of these tasks were selected because they previously have been shown to be sensitive to subject variables thought to be associated with variations in brain lateralization. All cognitive measures were administered using the standard administration procedures found in the reference listed for each task. Verbal tests. Three verbal tests were administered: (a) the Controlled Oral Word Association Test (Benton & Hamsher, 1976), a verbal fluency test, which requires the subject to produce as many words as possible that start with C (first trial), F (second trial), and L (third trial), with each trial lasting 60 s; (b) the Vocabulary subtest of the WAIS-R (Wechsler, 1981), a test of word knowledge; and (c) the Similarities subtest of the WAIS-R (Wechsler, 1981), a test of verbal abstract reasoning. 48 Spatial tests. Four spatial tests were administered: (a) the Vandenberg and Kuse (1978) paper and pencil version of the Sheperd-Metzler Mental Rotation test, a test of 3-dimensional mental rotation; (b) the Block Design subtest of the WAIS—R (Wechsler, 1981), a test of manipulo—spatial skill (Gazzaniga & LeDoux, 1978); (c) four items from a paper-and-pencil version (Harris, Hanley, & Best, 1977) of the Water Level test (Piaget & Inhelder, 1956), requiring judgement of the correct orientation of fluid with respect to the horizon; (d) the first six items of the Embedded- Figures Test (Witkin, 1971), requiring subjects to find a geometrical figure embedded within a complex design. American College Test. In addition to the cognitive test given during the testing session, all subjects were asked to sign waivers allowing access to their ACT scores. The ACT consists of four subtests: (a) English Usage; (b) Social Studies Reading; (c) Natural Sciences Reading; (d) Mathematics Usage. Visual Half-Field Test Stimuli for visual half-field task. The verbal stimuli for the tachistoscopic visual half-field procedure consisted of 20 words, four letters in length, selected from the high frequency portion of the Francis-Kucera (1982) list (e.g., come, know, good) and 20 legal non-words formed by changing the order of letters of the 20 chosen words (e.g., moce, wonk, doog). The letters were presented horizontally, subtending 2.9 degrees in width, and .70 degrees in height. 49 The nonverbal stimuli consisted of four unfamiliar male faces used by Rizzolatti, Umilta, and Berlucchi (1971). Rizzolatti et a1., chose these faces on the basis of similarities in physiognomy. All of the models are shown wearing white caps in order to mask differences in hair color and style. Two faces were designated as positive stimuli, the other two as negative stimuli, following Rizzolatti et al.'s procedure. The faces subtended 1.8 degrees in width, and 2.9 degrees in height. As noted earlier, when Rizzolatti and Buchtel (1977) used these stimuli in a visual half-field study, sex differences in response times were observed, indicating that this procedure is effective for eliciting sex differences in lateraliza— tion. Apparatus. Stimuli were projected on a white screen 2 meters in front of the subject with the aid of a three- channel projector tachistoscope (Gerbrands model 300-6T). Response times were measured with a Gerbrands (model G1271) timer. Procedure After subjects were classified into the sex by handedness groups, they were randomly called by telephone in order to confirm their willingness to participate in the experiment, and to answer any questions they might have. If, after being contacted, they were still interested in participating, they were scheduled for a two-hour testing session. All testing took place in the Psychology Research 50 Building. Before starting the visual half-field procedure, subjects were administered the cognitive tests in the following order: familial handedness questions, handwriting posture, Block Design, Controlled Oral Word Association Test, Vandenberg and Kuse Mental Rotation Test, Vocabulary subtest of the WAIS-R, Water Level test, Similarities subtest of the WAIS-R, and Embedded Figures Test. This order of administration was chosen (simple tasks first, alternation of verbal and spatial tasks) because it seemed to cause the least anxiety and to provide the most interest. Visual half-field test. At the beginning of the visual half-field procedure, subjects were familiarized with the four faces (two designated as positive stimuli, the other two as negative stimuli) for 3 minutes. Before each trial, the subject was signaled to fixate on the outline of a box located on the projection screen in the subject's central visual field. The signal used was a click that occurred 500 ms prior to the presentation of the stimuli. For each trial an arrow appeared in the center of the fixation box. The arrow pointed toward either the left or right visual field, indicating from which visual field the subject was to report. Fixation was assured, since the subject must focus on the central visual field in order to detect the direction of the arrow and consequently which visual field is to be reported. Simultaneously with the arrow, stimuli were projected bilaterally to each visual field, 2 degrees to the right and left of central fixation. A bilateral 51 presentation of stimuli was used in light of evidence that this method increases the likelihood that the two cerebral hemispheres would function as independent channels (Hines 1975). One word and one non-word, or one positive and one negative face, were presented to each visual field on each trial. This was done so that accuracy of responding could be measured. However, the subjects were told that the stimuli were randomly presented to each visual field and that they would not be able to determine the type of stimulus without focusing on the central fixation box. Reaction time was measured on a go/no-go task. If the stimulus in the to-be-reported visual field was a word or positive face, the subject's task was to press two buttons, one with each hand, as quickly as possible. The response time of the faster hand was recorded. If the stimulus in the to-be-reported visual field was a non—word or negative face, the subject was not to respond. Accuracy of responding was also recorded. Stimuli were presented for a duration of 100 ms. The direction of the arrow, type of stimulus, positive and negative value of the stimulus, and stimulus duration were randomized across trials, with the constraint that there would be two blocks of trials. Kinsbourne (1975) has proposed that attentional factors, influenced by the nature of stimuli, determine the direction and degree of visual half-field advantages. Verbal stimuli were therefore proposed to activate the left hemisphere, and spatial stimuli the right hemisphere. 52 Randomizing the presentation of verbal and spatial stimuli helps to insure against attentional priming effects, along with order of presentation effects. The verbal section consisted of 80 trials; each word and non-word was presented one to each visual field. The nonverbal section also consisted of 80 trials. Each face was presented to each visual field 10 times. The entire session, therefore, consisted of 160 trials presented in two blocks of 80 trials, plus an additional 40 practice trials. A lateralization index was calculated for each subject by subtracting the subject's median reaction time of right visual field responses from the median reaction time for left visual field responses. This index was corrected for differences between subjects in the ACT composite score and the effect of a second degree polynomial equation of Bryden and Sprott's (1981) lambda, a lateralization index for accuracy. RESULTS Achievement and Intelligence Level Since the composite ACT score was used to select subjects, it is important to know whether any of the sex by handedness cells differed in composite ACT scores. This was tested with a 2-way ANOVA. The means and standard devia- tions are presented in Table 1. There was no main effect sex, F(l,55) = 2.33, p = .133, or handedness, F(l,55) = .522, p .473, and no sex by handedness interaction, F(1,55) = .006, p = .936. However, since the main effect for sex, although non-significant, suggested the presence a trend for males to have higher ACT scores, sex was used a covariate in the remaining analyses. Cognitive Variables Before examining the results of the relationship between cerebral lateralization and cognitive ability, it necessary to find out whether the cognitive ability tasks and cerebral lateralization tasks are valid measures. of of as is First, the cognitive ability tasks will be analyzed for sex, handedness, and sex by handedness interactions. Second, visual field advantages, and sex and handedness differences in visual field advantages will be analyzed. Differences in cognitive performance were found between left— and right—handed men and women. The direction of 53 54 Table 1 ACT Composite Scores for Left- and Right-Handed Men and Women Men N Mean Left-handers 14 29.2 Right-handers 14 28.8 Women Left-handers 14 28.5 Right-handers 14 28.2 ANOVA Source of Variation gf MS Sex 1 6.446 Handedness 1 1.446 Sex X Handedness Interaction 1 .018 lm 2.329 .522 .006 . .133 .473 .936 55 these differences was dependent on the nature of the cognitive task. Two multivariate ANOVA's (MANOVA) were used to analyze the relationship between sex and handedness for the cognitive measures. Means and standard deviations are presented in Table 2. The MANOVA for all cognitive variables indicated that the main effect for sex was significant, F(1l,55) = 3.036, p = .005 (Table 3). Women outscored men on the English subtest of the ACT, F(1,55) = 18.636, p = .001), whereas men outscored women on Mental Rotation, F(1,55) = 7.01, p = .011. The men also did better than the women, although not by a statistically significant margin, on the Mathematics subtest, F(1,55) = 2.68, p = .108, and Natural Science subtest of the ACT, F(1,55) = 2.97, p = .091, on the Water Level test, F(1,55) = 2.56, p = .116, and on The Controlled Oral Word Association Test, F(1,55) = 2.96, p = .091. None of the other sex differences approached significance (all p's >.150). The MANOVA also indicated that the main effect for handedness approached significance, F(11,55) = 1.775, p = .091 (Table 4). Separate univariate ANOVAs showed that the left-handers significantly outscored the right-handers on Block Design, F(1,55) = 5.00, p = .030, and Vocabulary, F(1,55) = 4.54, p = .038. The left-handers also outscored the right-handers on Water Level by a marginally significant degree, F(1,55) = 3.48, p = .068. Table 2 56 Cognitive Test Performance for Left- and Right-Handed Men and Women, Adjusted for Composite ACT Score Test Right-Handed Left-Handed Right-Handed Left-Handed Men Men Women Women ACT—English Mean 24.16 24.45 26.59 26.00 S.E. .46 .45 .45 .46 ACT-Math Mean 29.10 29.59 27.45 28.65 S.E. .78 .78 .77 .78 ACT-Soc.Sci. Mean 29.20 29.01 29.66 29.55 S.E. .58 .57 .57 .58 ACT-Nat.Sci. Mean 31.64 30.96 30.64 30.27 S.E. .49 .48 .48 .48 Similarities Mean 22.57 23.05 21.74 23.00 S.E. .62 .61 .61 .62 Vocabulary _ Mean 51.17 56.73 54.21 55.46 S.E. 1.61 1.59 1.59 1.61 Oral Word Assoc. Mean 39.41 45.14 37.52 38.64 S.E. 2.42 2.39 2.39 2.41 Embedded-Fig. Mean (sec.) 207.55 247.21 306.51 215.44 S.E. 38.13 37.69 37.71 38.07 Block Design Mean 40.28 42.37 38.16 42.97 S.E. 1.55 1.53 1.53 1.55 Water Level Mean 3.24 3.63 2.44 3.32 S.E. .34 .34 .34 .34 Mental Rotation Mean 13.58 11.57 8.42 10.01 S.E. 1.26 1.24 1.24 1.25 Table 3 57 MANOVA For Sex Difference MANOVA Univariate ANOVAs with (1.51) df Variable ACT-English ACT-Mathematics ACT-Natural Science ACT-Social Science Embedded-Figures Block Design Water Level Mental Rotation Similarities Vocabulary Oral Word Association as 53.325 22.561 9.589 3.364 15120.280 7.734 4.092 151.282 2.587 10.449 236.064 I111 3.034 § 18.636 2.678 2.969 .728 .762 .235 2.256 7.010 .498 .295 2.961 .005 .001 .108 .091 .398 .387 .630 .116 .011 .484 .589 .091 Table 4 58 MANOVA For Hand Preference MANOVA Univariate ANOVAs with (1.51) df Variable ACT-English ACT-Mathematics ACT-Natural Science ACT-Social Science Embedded-Figures Block Design Water Level Mental Rotation Similarities Vocabulary Oral Word Association gs .308 9.962 3.817 .343 9160.919 164.465 5.553 .600 10.420 160.536 162.986 I”! 1.775 I": .108 1.182 1.182 .074 .462 5.001 3.478 .028 2.005 4.538 2.044 .091 .744 .282 .282 .786 .500 .030 .068 .868 .163 .038 .159 59 Since the presence of sex and handedness differences was dependent on the nature of the task, a discriminant function analysis was carried out to test for the presence of a sex by handedness interaction for the verbal and spatial tasks. This analysis is able to address whether some weighting of the tasks is able to produce a sex by handedness interaction. For the spatial tests, the results were positive. A significant (p <.05) discriminant function was found for a sex by handedness interaction. Inspection of the standardized discriminant coefficients (Table 5) indicates that Embedded-Figures made the most positive contribution to the discriminant function. Mental Rotation made the next largest positive contribution to the discriminant function. Water Level made a small positive contribution, and Block Design acted as a suppressor variable (i.e., Block Design correlated with a non- discriminating component of a positive contributor). The direction of the means (Table 2) indicates that the right- handed men performed better than the left-handed men, whereas the left-handed women performed better than the right-handed women. For the verbal tests, a significant discriminant function ( p < .05) also was found for the sex by handedness interaction. The discriminant coefficients (Table 5) indicate that Similarities made a positive contribution, whereas the Vocabulary test, the ACT-English subtest, and the Controlled Oral Word Association Test acted as 60 Table 5 Discriminant Analysis of a Sex by Handedness Interaction Standard Discriminant Function Coefficients for Spatial Variables: p < .05 Variable 23133 Embedded-Figures +.74l Mental Rotation .566 Water Level .007 Block Design -.158 Standard Discriminant Function Coefficients for Verbal Variables: p < .05 Variable 23123 Similarities .646 ACT-English -.249 Oral Word Association -.318 Vocabulary -.802 61 suppressor variables. The means indicate that the left- handed men outscored the right-handed men, and the left-handed women outscored the right-handed women, but this difference was very small. Note that the direction of both the verbal and the spatial discriminant functions was in the same direction as in the Harshman et a1., (1983) study. Cerebral Lateralization Index. Each lateralization index was analyzed for its relationship to sex and handedness with a 2 (sex) X 2 (handedness) ANOVA, using a second degree equation of lambda (an accuracy measure), and the composite ACT score as covariates. Face discrimination. On the visual half-field presentation of the facial discrimination task there was a significant left visual field advantage (mean lateralization index = -39.66 ms; T(l) = 2.25, p = .014), indicating that the test was a valid measure of right hemisphere functions. There were, however, no main effects of sex F(1,55) = 1.92, p = .172, or handedness, F(1,55) = .215, p = .645, and no sex by handedness interaction, F(1,55) = .159, p = .692 (Table 6). Lexical decision. On the lexical decision task, there was a right visual field advantage, although the effect was not significant (mean lateralization index = 14.37 ms; T(l) = 1.04, p = .150). There was, however, a significant sex effect, F(1,55) = 6.48, p = .014, such that males showed a significant right visual field advantage, T(l) = 2.55 p = Table 6 Lateralization Index for Facial Discrimination Test, Using Accuracy and Composite ACT Scores as Covariates Mean Performance Men Left-handers Right-handers Women Left-handers Right-handers ANOVA Source of Variation Sex Handedness Sex by Handedness Interaction [Z 14 14 14 14 Mean -15.83 —50.06 fl§ 34764.28 3901.42 2877.51 E P 1.918 .172 .215 .645 .159 .592 63 .007, whereas the women did not show a significant visual field advantage, T(1) = .700, p = .487 (Table 7). Again, this finding lends credibility to the task as a measure of language functions, since the results are consistent with previous reports that language functions are more asym- metrically organized in males than in females. There was no handedness effect, however, F(1,55) = .072, p = .790, and no sex by handedness interaction, F(1,55) = .298, p = .588. Relationship Between Lateralization Index and Cognitive Variables The preceding analyses indicated that variation in cognitive ability and cerebral lateralization was present and was consistent with previous studies. To test the hypothesis that variation in cerebral lateralization is related to cognitive ability, all of the cognitive variables were regressed on polynomial equations, progressing from one to three degrees for each lateralization index. The lowest degree significant relationship was accepted as the most representative regression equation describing the relation- ship between the cognitive variable and the lateralization index. A second degree polynomial equation of accuracy on the visual half-field procedure, together with the ACT composite score, were used as covariates. Each relationship was also tested for sex by lateralization index, hand preference by lateralization index, and sex by hand preference by lateralization index interactions. When an interaction was present, regression equations were analyzed Table 7 64 Lateralization Index for Lexical Decision Test, Using Accuracy and Composite ACT Scores as Covariates Mean Performance Men Left Handers Right—handers Women Left-handers Right-handers ANOVA Source of Variation Sex Handedness [Z 14 14 14 14 Sex by Handedness Interaction Mean 55.95 48.61 -27.50 - 6.00 8.8. 58927.87 652.96 2709.77 I"! 6.483 .072 .298 .014 .790 .588 65 separately for the appropriate groups. The results indicated that cerebral lateralization is related to cognitive ability. The nature of the results, however, depends on the nature of the task, and the sex and handedness of the subject. Only significant or near- significant relationships will be discussed here. ACT-English subtest. There was a near-significant trend for subjects with greater left visual field advantages on the spatial lateralization index to have higher English scores on the ACT (Beta = -.186; p = .067; Figure 1). There also was a significant sex by spatial lateralization index interaction on the English subtest of the ACT (Beta = -.437; p = .003). However, separate analyses for each sex failed to disclose a significant relationship between the spatial lateralization index and English performance on the ACT for either sex. ACT Mathematics subtest. A significant handedness by verbal lateralization index interaction was found for the Mathematics subtest of the ACT (Beta = .355; p = .027). No significant relationship was found for right-handers, but there was a trend toward a significant relationship for the left-handers (Beta = .300; p = .105), with higher Mathematics scores being associated with greater right- visual advantages on the visual half-field test (Figure 2). ACT-Social Studies subtest. Higher scores on the Social Studies subtest of the ACT were associated with greater left visual field advantages on the spatial 66 lateralization index (Beta = -.232; p = .035; Figure 3). There was a trend toward a significant handedness by spatial lateralization index interaction (Beta = -.349; p = .097. Separate analyses for each hand preference group indicated that this relationship was significant only for the left- handers (Beta = -.4090; p = .005; Figure 4). A significant hand preference by lateralization index interaction also was found for the verbal index (Beta = -.378; p = .016). A linear relationship between Social Studies performance and the verbal lateralization index also was found, but only for left-handers (Beta = -.580; p = .002), with higher Social Studies scores being associated with left visual field advantages (Figure 5). ACT-Natural Science subtest. Higher Natural Science scores were associated with greater right visual field advantages on the spatial lateralization index (Beta = .257; p = .022; Figure 6). A sex by lateralization index interaction approached the .05 significance level (Beta = .318; p = .051), indicating that the positive association between Natural Science score and visual field advantage approached significance only for the women (Beta = .289; p = .104; Figure 7). A significant hand preference by sex by verbal lateralization index interaction was found on the Natural Science subtest (Beta = .381; p = .044). The only subgroup approaching significance was the left-handed women (Beta = .525; p = .075), with higher Natural Science scores 67 being associated with right visual field advantages (Figure 8). Mental Rotation. Mental Rotation scores were significantly related to the cube of the spatial lateralization index (Beta = .278; p7= .041) with higher scores on Mental Rotation being associated with right visual field advantages (Figure 9). There was a significant sex by hand preference by cube of verbal lateralization index interaction on Mental Rotation performance (Beta = .451, p = .052). However, none of the individual analyses were significant. Embedded-Figures Test. A non-significant trend for a sex by cube of verbal lateralization index interaction was present on the Embedded-Figures Test (Beta = .272; p = .085). Only the women's group approached a significant relationship (Beta = .393; p = .062), with lower times (better performance) being associated with left visual field advantages (Figure 10). Water Level. Performance on the Water Level test was significantly associated with the spatial lateralization index (Beta = .293; p = .030), with higher Water Level scores being associated with right visual field advantages (Figure 11). Water Level performance was also significantly associated with the square of the verbal lateralization index (Beta = -.272; p = .043). The best scores were associated with equal visual field performance (Figure 12). 68 The sex by square of verbal lateralization index almost reached the .05 level of significance (Beta = -.302; p = .051), indicating that the relationship between Water Level performance and square of verbal lateralization index was greater for the women (Beta = —.331; p = .078; Figure 13) than the men (Beta = -.ll9; p = .595). Controlled Oral Word Association Test. A significant hand preference by cube of spatial lateralization index was present on the verbal fluency test (Beta = .286; p = .043). Only the left-handers' performance approached a significant association with the lateralization index (Beta = .367; p = .064), with higher verbal fluency scores associated with greater right visual field advantages (Figure 14). Verbal fluency scores were also significantly related to the verbal lateralization index (Beta = .335; p = .016), with higher scores associated with right visual field advantages (Figure 15). The sex by lateralization index approached significance (Beta = .302; p = .083), indicating that this relationship was stronger for women (Beta = .546; p = .008) than for men (Beta = .013; p = .952; Figure 16). Similarities. A trend towards a significant sex by verbal lateralization index interaction existed on the Similarities subtest of the WAIS (Beta = .301; p = .098), but neither sex showed a significant relationship between Similarities score and the lateralization index. DISCUSSION Handedness and Sex Differences in Cognitive Ability The results disclosed several main effects of sex and handedness. Consistent with a vast literature (Harris, 1978, 1981), the men performed significantly better than the women on the Mental Rotation task. There was also a non- significant trend for men to outscore women on the Water Level test as well as the Mathematics and Natural Science subtests of the ACT. The women significantly outscored the men on one verbal test, the English subtest of the ACT. One unexpected finding was that the men outscored the women on the Controlled Oral Word Association Test, although this difference was not significant ( p = .091). As noted earlier, handedness differences on cognitive measures are less well documented than sex differences, with some studies showing differences, others not (Herron, 1980; Harshman et a1., 1983). The results of the current study contribute to the view that differences do exist, at least in young adults of high academic achievement. The left- handers outperformed the right-handers on the Vocabulary, Block Design, and Water Level tests. The left-handers' superiority on Vocabulary is consistent with Levy's (1969, 1974) findings that left-handed graduate science students 69 70 performed better than right-handers on verbal items from the WAIS. However, the results with Block Design (superior performance by left-handers) are in the opposite direction from Levy's results. Her results (better nonverbal performance by right-handers on intelligence measures) have been corroborated (Eme et a1., 1978; Hicks & Beveridge, 1978), although nine other studies have failed to find handedness differences (see Sanders et a1., 1982). The present study, evidently, is the only report of superior performance by left-handers on a nonverbal measure of intelligence, but it is consistent with reports of better spatial performance of left-handers on Mental Rotation (Sanders et a1., 1982). An increased percentage of left-handers in architecture school (Peterson & Lansky, 1974) and art school (Merbert & Michel, 1980) has also been found which is suggestive of greater spatial ability among left-handers. Harshman et a1., (1983) found a trend toward a sex by handedness by reasoning level interaction on verbal and spatial cognitive measures. In the present study, where only high academic achievers were considered, significant sex by handedness discriminant functions were also found. For the spatial measures, (Embedded-Figures, Block Design, Water Level, and Mental Rotation), Embedded-Figures and Mental Rotation made the largest positive contributions toward the discriminant function, and Water Level made a very small positive contribution, whereas Block Design acted 71 as a suppressor variable. The direction of the sex by handedness interaction found with the discriminant function is the same as predicted by Harshman et a1., (1983), with right-handed men performing better than left—handed men, but with left-handed women performing better than right-handed women. For the verbal measures, (ACT—English subtest, Controlled Oral Word Association Test, Vocabulary, and Similarities) Similarities made a positive contribution towards the discriminant function, whereas ACT-English, Vocabulary, and the verbal fluency test acted as suppressor variables. The direction of the interaction was dependent on the nature of the cognitive task. The left-handers tended to perform better than the right-handers. The difference in scores between left- and right-handers was dependent on sex, and the type of cognitive task. The pattern of results found in this study was very similar to the pattern found by Harshman et a1., (1983). However, the results of the discriminant analyses suggest that the sex by handedness interaction may be limited to particular verbal and spatial functions. For the spatial measures, Mental Rotation and Embedded-Figures, which made positive contributions toward the sex by handedness discriminant function, were the same type of measures included in the Harshman et a1., study. They used two Mental Rotation tests (Primary Mental Abilities test of Mental Rotation and the Vandenberg & Kuse adaptation of the Shepard and Metzler Mental Rotation test) and two 72 disembedding tests (the Educational Testing Service's Hidden Patterns and Copying). However, they did not include tests that were comparable to Water Level and Block Design, measures that in the present study had small positive and suppressor effects, respectively. The results for the verbal tests are less clear than the spatial measures but are consistent with the Harshman et a1., (1983) study. Harshman et a1., found that left-handed men outperformed right-handed men in seven out of the nine verbal tests, whereas the right-handed women outscored the left-handed women on only four of the nine tests. Similar results were found in the current study with the discri- minant function indicating that the handedness difference among men is stronger than among women. The direction of the interaction of the discriminant function indicated that left—handed men outperformed'right-handed men, and that left-handed women also tended to outperform right—handed women. The difference in scores between left- and right- handed women, however, was smaller than that between the left- and right-handed men. Measures of verbal fluency and vocabulary were used in both studies. In the Harshman et a1., (1983) experiment, left-handed men outperformed right-handed men on the Primary Mental Abilities (PMA) vocabulary test, whereas right—handed women outscored left-handed women, but the difference between the four groups was not significant ( p = .162). In the present study left-handers significantly outperformed 73 right-handers on the WAIS Vocabulary subtest ( p = .038). One difference between these two studies is the nature of the vocabulary test. The PMA is a multiple choice format, whereas the WAIS version is a free recall design. The pattern of performance on the test of verbal fluency in the present study was for left—handers to outscore right- handers, but the main effect of handedness did not reach significance ( p = .159). There was a greater difference in scores between left- and right-handed men (5.7 points) than between left— and right-handed women (1.1 points). However, the sex by handedness interaction was not significant ( p = .339). This pattern of performance (i.e., a greater left- handed superiority with men than women) was also found on a test of verbal fluency in two out of the three samples in the Harshman et a1., (1983) study. In summary, the present results confirm the presence of a sex by handedness interaction, occurring in opposite directions for spatial and verbal tests, but limit this interaction to particular verbal and spatial functions. In addition, the sex by handedness interaction is clearer with spatial functions than with verbal functions. Visual Half-Field Measure Face discrimination lateralization index. Levy and Gur (1980) predicted the same direction of results found in the present study and Harshman et al.‘s (1983) study for left- and right-handed men and women with non-inverted handwriting postures. As previously stated, their model proposes that 74 cerebral lateralization underlies the pattern of cognitive ability among left— and right-handed men and women. To test their hypothesis visual half—field indices of cerebral lateralization were measured, and then analyzed for their association to cognitive ability. As expected, a left visual field advantage was obtained on the face discrimination task (Davidoff, 1982). This signifies that the face discrimination task is a valid measure of right hemisphere nonverbal functions. However, there were no sex or handedness effects and no sex by handedness interaction on this task. The absence of a handedness effect is, perhaps, understandable. As noted earlier, although left-handers usually show evidence of greater bilateral representation of language functions, it is much less clear whether they also have bilateral representation of visual-spatial functions. Indeed, as noted earlier, there is little evidence of a handedness difference in the lateralization of visual spatial functions (Gardner et a1., 1977; Hecaen & Sauguet, 1971; Masure & Benton, 1983; McGlone & Davidson, 1973). The lack of a sex difference on this face discrimina- tion task is harder to explain, since a sex difference was previously found using the same procedure (Rizzolatti & Buchtel, 1977; Umilta et a1., 1978). This sex difference, however, has not been found outside of that particular laboratory, and it has been subsequently attributed to differences in strategies (Zoccolotti & Oltman, 1978). 75 Fairweather, in his review of visual half-field studies of sex differences, cites three experiments using unfamiliar faces as finding greater left visual field advantages with men (Rizzolatti & Buchtel, 1977; Umilta et a1., 1978), one study finding greater left visual field advantages for women, and six studies finding no evidence of sex differences. As previously stated, sex differences in cerebral lateralization are precarious, and, if present, seem to depend on conditions that we do not yet fully understand. Lexical decision lateralization index. There was a tendency for right visual field reaction times to be faster than left visual field reaction times on the lexical decision task, but the difference was not significant ( p = .150). A sex difference, however, was found on this task, with the men showing a significant right visual field advantage, but with the women showing no visual field advantage. These results are consistent with clinical and experimental evidence indicating a greater bilateral representation of language functions in women (McGlone, 1980) but are supported by only one of several other studies using a lexical decision task with normal subjects (Bradshaw, Gates, & Nettleton, 1977). Finding visual field advantages in the expected direction underscores the validity of this task for measuring language functions presumed to be lateralized in the right hemisphere for males, and bilaterally represented in females. 76 No difference between left- and right—handers was found on the lexical decision task. In comparison, of three other studies that also looked at handedness difference on a lexical decision task, one found a significant handedness difference (Bradshaw, Gates, & Nettleton, 1977), whereas the other two did not (Chiarello, Dronkers, & Hardyck, 1984; Leiber, 1976). In the current study, the means indicate that left-handers as a group tended to show greater visual field advantages than did right-handers. Although counter to the prevailing view that left-handers have a more bilateral representation of language, this same trend has been reported in clinical studies (Hecaen & Sauguet, 1971; Hecaen, DeAgostini, & Monzon-Montes, 1981) and in experimental studies (Dagenbach, 1983; McKeever & VanDeventer, 1977) of visual processing of language. In summary, the validity of the two lateralization measures was upheld by finding visual field advantages in the expected directions. For the face discrimination task, there was a significant left visual field advantage, and for the lexical decision task, the men showed a significant right visual field advantage, whereas the women did not show a visual field advantage, suggestive of greater bilateral representation of access to the lexicon for females than males. The failure to find a handedness difference on the verbal and spatial lateralization indices, together with the failure to find a sex—related difference on the spatial 77 lateralization index, is consistent with other studies, and can be explained in several ways. First, the possibility must be considered that these differences may not exist in the population. There is, in fact, enough evidence to support this conclusion for the lack of sex-related differences. Handedness differences in cerebral lateralization, however, are well established. Second, the handedness and sex differences may not exist among the subject sample studies. In the case of the handedness effect, if only 30% of left-handers display cerebral lateralization patterns different from right- handers (Rasmussen & Milner, 1977) then sampling error associated with small subject samples might easily mask a group difference existing in the population. With regards to sex-related differences, finding a sex difference on the lexical decision task argues for sex—related differences in cerebral lateralization within this subject pool. Third, handedness and sex differences in cerebral lateralization may exist but may escape detection due to measurement error. Undoubtedly, measurement error exists, but this cannot be corrected for simply with reliability measures, since this would assume that cerebral lateralization is a stable entity and ignores the possible effects of changes in functional lateralization due to changes in strategies and hemispheric activation. Fourth, assuming that the visual half-field measures reflect fixed structural lateralization patterns, the 78 presence or absence of individual differences on particular tasks may indicate how these particular cognitive processes are organized in the brain. However, there is a growing realization in neuropsychology that cerebral lateralization is not a single entity but rather varies with the particular task demands (Bryden, Hecaen, & DeAgostini, 1983; Dagenbach, 1983; Gloning, Gloning, Haub, & Quatember, 1969; Hines & Satz, 1974; Hellige & Cox, 1976; Herron et a1., 1979; Moscovitch, 1979; Samar, 1983). If so, then the present results suggest that left- and right—handers do not differ in their lateralization of access to the lexicon; nor are there sex or handedness differences in the lateralization of cognitive processes tapped by the facial discrimination task. Fifth, independent of fixed structural asymmetries, the visual half-field measures may be a reflection of activa- tional changes in functional lateralization. The most prominent model of hemispheric activation has been advanced by Kinsbourne (1975). He proposed that attentional factors can influence the direction and degree of visual field advantages. Stimuli serve to prime the activation of a hemisphere. For example, verbal material tends to activate the left hemisphere, and spatial stimuli the right hemisphere. Activation of one hemisphere also tends to inhibit the activation of the other hemisphere. Attentional factors such as expectations or preferred strategies also may affect the pattern of functional lateralization. One 79 problem with Kinsbourne's theory is that it is so flexible, it can be used to account for any set of data. An attempt was made in the current study, however, to limit the likelihood of occurrence of selective hemispheric activation by presenting the verbal and spatial tasks in random order. In general, this approach should not lead to selective activation of one hemisphere over the other, since the mixture of the stimuli should equally activate both hemispheres. One could argue, though, that in spite of the test order adopted, a subject nonetheless might adopt a strategy favoring the selective activation of one hemisphere. The number of possible explanations for the data attests to the complexity of the subject matter and the lack of a universally accepted theoretical model of cerebral lateralization. It is also quite likely that some combination of the explanations offered above account for the results of this study. Relationship Between Cerebral Lateralization and Cognitive Ability We have discussed the major cognitive data and the cerebral lateralization data. We turn now to the question of the relationship between cognitive ability and cerebral lateralization. But first, we should ask whether the current study has met the criteria suggested for a proper test of such a relationship. Criterion 1 (using cognitive tests that show a range of ability with the subject sample 80 used) was met because significant differences were found between the sexes by handedness groupings of subjects. Criterion 2 (independent tests of cognition and lateralization) was met by using different cognitive tasks for each measure. Criteria 3 and 4 (variation in degree and direction of lateralization) were met, since inspection of Table 6 and 7 indicates that subjects varied widely along these two dimensions. In addition, a significant difference in lateralization was found between the sexes on the lexical decision task. Criterion 5 was also met, since verbal and spatial tasks were regressed on both the facial discrimina- tion and lexical decision lateralization tasks. Therefore, all of the recommended criteria for a proper test of the lateralization-cognition hypothesis were met. As stated in the introduction (p. 36), only two other studies were found that met all five suggested criteria. This precludes exact comparison between the current study and most of the available literature. Of the two studies meeting the criteria, one, recall, was a study with 12 - l4-year—old children (Kraft, 1983), relating verbal intelligence scores to a dichotic listening lateralization index. Two problems exist in making a comparison between the two studies meeting the criteria and the current study: (a) the relationship between lateralization and cognition has been shown to change during childhood (Kraft, 1984), (b) there is a poor relationship between dichotic listening and visual half-field lateralization indices (Dagenbach, 1983; 81 Fennell et a1., 1977; Hines & Satz, 1974). The other study used an unselected sample from public high schools, colleges, and the general public, with subjects ranging in age from 16 years to 42 years (Birkett, 1980). In this study, Birkett did not find the same sex by handedness interaction on spatial tests, reported in this study and Harshman et al.'s study. Birkett's sample, however, would not be comparable to the current study or to Harshman et a1.'s study, since it did not select for high intelligence, "high reasoners," or high academic achievers. In support of the lack of comparability between studies, Birkett found a difference in score between sex and handedness groups on only one of three spatial tests administered. On that test of matching three dimensional shapes, the only difference was that right—handed males outscored right-handed females. In the current study, in order to test the hypothesis that cerebral lateralization is related to cognitive ability, regression equations were analyzed for possible relationships between the lateralization indices and each cognitive measure. Several important conclusions can be drawn from these analyses. First, cerebral lateralization is a major correlate of cognitive ability. Second, differences in cerebral lateralization are associated with differences in cognitive ability due to hand preference and sex. Third, since sex and handedness differences were still present in the lateralization-cognition relationships, sex and handedness must be indices of more than just cerebral 82 lateralization. They also might reflect other aspects of cerebral organization. With one exception, two patterns were found for the relationship between the lateralization index and cognitive measure: (a) increasing right hemisphere lateralization was associated with increasing cognitive performance, or (b) increasing left hemisphere lateralization was associated with increasing cognitive performance. These two patterns can be further broken down by the nature of the lateraliza- tion index. Keeping in mind the hypothesis that cerebral lateralization underlies cognitive ability, let us examine the ability of that hypothesis to account for the trends, significant or not, found in the current study. Increasing Left Visual Field Advantage—Increasing Cognitive Performance Spatial lateralization index. The linear relationship between the spatial lateralization index and the English subtest of the ACT almost reached significance ( p = .067), with increasing left visual field advantages being associated with higher English scores. The significantly better performance by women on the English test is consistent with their greater, although not significantly different from the men's, left visual field advantage on the spatial lateralization task. A significant sex by spatial lateralization index interaction indicated that the relationship between the English scores and cerebral lateralization was different for 83 the two sexes. However, neither of the separate regression equations of each sex were significant. There was virtually no difference in performance between the four sex by handedness groups on the Social Studies subtest of the ACT. However, differences in performance between individuals were related to differences in cerebral lateralization. High Social Studies scores were significantly associated with greater left visual field advantages on the facial discrimination task. This relationship was stronger for left-handers than for right-handers. Verbal lateralization index. Among left-handers, greater left visual field advantages on the lexical decision task were also significantly related to higher Social Studies scores. Relationships between both lateralization indices and Social Studies performance signify the complexity of this measure, with the organization of both verbal and nonverbal functions being relevant for Social Studies performance. Among the women, there was a non-significant trend ( p = .062) for increasing left visual field advantages on the lexical decision task to be associated with better perfor- mance on the Embedded—Figures Test. This is consistent with the left-handed women's greater, though not significantly different, left visual field advantage on the lexical decision task. However, neither the sex by handedness interaction on Embedded-Figure scores, nor the right-handed 84 men's better scores, can be explained by the pattern of cerebral lateralization. Increasing Right Visual Field Advantage-Increasing Cognitive Performance Spatial lateralization index. Increasing right visual field advantages on the facial discrimination task were significantly related to increasing performance on the Natural Science subtest of the ACT. The men's smaller left visual field advantage on facial discrimination task is consistent with their non-significant ( p = .091) trend toward better Natural Science scores. A sex by spatial lateralization index interaction ( p = .051) indicated that this relationship was stronger for women than men. Greater right visual field advantages on the facial discrimination task were significantly related to higher Mental Rotation scores. The men's smaller, although non- significant, left visual field advantage is consistent with their significantly better performance on the Mental Rotation Test. However, the relationship between spatial lateralization and Mental Rotation cannot explain the sex by handedness interaction on Mental Rotation, with right—handed men performing better than the left-handed men, and the left-handed women performing better than the right-handed women. Increasing right visual field advantages on the facial discrimination task were significantly related to increasing Water Level scores. The men's smaller, though not 85 significantly different, left visual field advantage on the facial discrimination task is consistent with their higher Water Level scores. Similarly, the left-handers smaller, but also not significantly different, left visual field advantage is consistent with their higher Water Level scores. On the Controlled Oral Word Association Test, perfor- mance was significantly associated with performance on the facial discrimination test for only left-handers, with higher verbal fluency scores being associated with greater right visual field advantages. The left-handed men's better performance on the Controlled Oral Word Association Test is consistent with their smaller, though not significantly different, left visual field advantage on the facial discrimination test. Verbal lateralization index. There was a trend towards a significant ( p = .105) linear relationship between the verbal lateralization index and the Mathematics subtest of the ACT for left-handers but not right-handers. Among the left—handers, higher Mathematics scores were associated with increasing right visual field advantages on the facial discrimination task. The men's trend toward significantly better performance on Mathematics ( p = .108) is consistent with their smaller, though not significantly different, left visual field advantage on the facial discrimination task. The better performance by the right—handed men compared to 86 the right-handed women apparently cannot be explained by cerebral lateralization patterns. The relationship for increasing right visual field advantages to be associated with increasing Natural Science scores approached significance ( p = .075) for left-handed women. Increasing right visual field advantages on the lexical decision task were associated with higher Controlled Oral Word Association Test scores. The men's significant right visual field advantage on the lexical decision task is, therefore, consistent with their trend toward signi- ficantly better performance on the Controlled Oral Word Association Test ( p = .091). This association was stronger for women than men ( p = .083.). Remaining relationships. The exception to the two patterns discussed above occurred on the Water Level task. The best scores on Water Level were significantly related to equal visual field advantages on the lexical decision task. This relationship is stronger for women than men. Although not a significant difference, left-handed women on the average had more of a left-visual field advantage than right-handed women. From this it would be expected that right-handed women would outscore left-handed women on Water Level, which was not the case. Significant relationships between Water Level performance and the two lateralization indices suggest the relevance of both verbal and nonverbal strategies towards Water Level solutions. 87 Finally, there were two significant interactions but the separate analyses did not result in significant relationships between the cerebral lateralization index and the cognitive measures. There was a significant sex by handedness by verbal lateralization index on mental rotation performance, indicating that the relationship between verbal laterali— zation and mental rotation was different between the four sex by handedness groups. There was also a trend toward a significant sex by verbal lateralization index interaction on the Similarities subtest of the WAIS, indicating a different relationship between verbal lateralization and Similarities for the two sexes. SUMMARY AND CONCLUS I ONS The results clearly show that the pattern of cerebral lateralization is related to cognitive ability. Just as clearly, the relationship is far from simple, since it seems to depend on the nature of the task as well as the sex and handedness of the subject. One striking finding is that the absolute magnitude of cerebral lateralization is not the most important aspect of the lateralization-cognition relationship. This is contrary to Levy and Gur's (1980) model according to which better performance is associated with the bilateral representation of that function. In the current study, this pattern was found only for the relationship between Water Level scores and the lexical decision task. The results thus imply that subjects with more bilateral representation of a given function show greater cognitive ability than the more "prototypical" asymmetric organization, not by virtue of being more bilaterally organized, but rather by differing in direction of lateralization. There are two trends for the lateralization-cognition data. First, the greater the right visual field (left hemisphere) advantage on the lexical decision task, the better the spatial task performance (see Figures 2, 8, and 10); and the greater the left visual field (right 88 89 hemisphere) advantage on the facial discrimination task, the better the verbal task performance (see Figures 1, 3, and 4). If verbal and spatial functions were organized complementarily, then increasing lateralization of one function might be associated with bilateral organization of the other. However, there are two points that argue against this interpretation: (a) as previously mentioned, Bryden et a1., (1983) found no evidence for complementarily organized functions, (b) there was no evidence in the current study that better performance on cognitive measures was associated with bilateral organization of the "complementary" function. Obviously, some aspect of the lateralization index is associated with cognitive performance, but it is unclear why there is a dominant trend for spatial lateralization to be associated with verbal ability, and verbal lateralization to be associated with spatial ability, in the absence of verbal ability-verbal lateralization and spatial ability-spatial lateralization relationships. The second trend was for increasing cognitive performance to be associated with increasing lateralization 3321 from the hemisphere normally associated with speciali- zation of that function (i.e., left hemisphere advantages for spatial functions tend to be associated with greater spatial ability, and right hemisphere advantages for verbal functions tend to be associated with greater verbal ability; see Figures 5, 6, 7, 9, and 11). One possible explanation is that everyone inherits a natural tendency for language 90 specialization of the left hemisphere and spatial speciali- zation of the right hemisphere. When other unknown factors alter this normal pattern of organization, they “overlap" with this normal tendency, so that if a reversed pattern of organization occurs (e.g., a right hemisphere advantage for language functions), the right hemisphere would have to develop language capabilities beyond the left hemisphere's normal specialization. This would result in superior ability for the reversed organized function. Note that this model proposes that increasing cognitive performance is associated with increasing representation of a function in both hemispheres. This model is different from Levy's such that cognitive ability is associated with the absolute ability of both hemispheres, rather than simply the bilateral representation of a function. There is no evidence at present for such a model, but to put it to test, indices of lateralization would have to take into account the absolute performance of each hemisphere instead of simply the relative differences of the two hemispheres. There is no implicit assumption in this model of complementary organization of verbal and spatial functions, so that no deficit in the "complementary" function is predicted. However, common sense would suggest that something must "give" in such a situation. Together, these two trends suggest that the best spatial ability is associated with left hemisphere representation of both spatial and verbal functions, and the 91 best verbal ability is associated with right hemisphere representation of both spatial and verbal functions. This pattern is opposed to predictions made by the "crowding hypothesis" (i.e., the representation of verbal and spatial functions in the same hemisphere results in a compromise of these functions and suggests, instead, that there are relatively unlimited resources for the representation of these two functions. If so, then the organization of the two functions in the same hemisphere might result in more efficient communication between these two functions. The "unlimited resources hypothesis" is directly supported by the results with the Social Studies and Natural Science subtests of the ACT. Increasing left hemisphere advantages on both the verbal and spatial laterality tasks are associated with increasing Natural Science performance, and increasing right hemisphere advantages on both the verbal and spatial laterality tasks are associated with increasing Social Studies performance. However, this pattern was not found on other cognitive tasks. Since verbal and spatial tasks are not unitary concepts, the results should not, perhaps, be organized within a verbal and spatial dichotomy. Instead, the results may be specific with respect to the nature of the particular cognitive task. The present results are not consistent enough to choose between these alternatives. There are three relationships that do not fit either of the two aforementioned trends: (a) for left-handers, 92 increasing right visual field advantages on the facial discrimination task were associated with better word production on the Controlled Oral Word Association Task; (b) for women, the best scores on Water Level were associated with no visual field advantage on the lexical decision task; (c) for all groups, increasing right visual field advantages on the lexical decision task are associated with better word production on the Controlled Oral Word Association Test. The first of the three "atypical" relationships is difficult to explain; the other two are more easily accounted for. The association of high scores on Water Level with bilateral organization, as indexed by the lack of a visual field advantage, is in accordance with Levy and Gur's prediction, but only if it is assumed that language plays a critical role on the Water Level test. The association of increasing left—hemisphere representation of language functions with superior word production has been assumed by those suggesting that stutterers' word production is impaired by the lack of unilateral control (bilateral representation) of language functions (Andrews, Quinn, & Sorby, 1972; Curry & Gregory, 1969). Lateralization by itself, however, cannot account for differences in ability on all of the cognitive measures, nor the sex by handedness interaction found on several verbal and spatial tasks, nor the differences found between certain sex or handedness groups. Those abilities and relationships between groups not explained by the lateralization indices 93 may instead be related to the lateralization of functions not tapped by the particular lateralization tasks used in this study. Another possibility is that sex and handedness may be more than just indirect indices of cerebral lateralization and may indicate other aspects of cerebral lateralization. Kimura (1983) has presented evidence indicating sex differences in the anterior-posterior organization of cerebral functions. By the failure of the lateralization indices to account for certain differences between left- and right—handers, the possibility exists that handedness may also be related to anterior-posterior organization of cognitive functions. Although the current data argue for the importance of cerebral lateralization in determining cognitive ability, the inability of the data to fully account for differences in performance suggests that lateralization is only one of a number of factors involved. Since the sex by handedness interaction is difficult to explain by socialization models, neurological models would seem to provide a potentially better explanation of these group differences. One possibility is that the anterior-posterior organization of verbal and spatial functions is related to the sex by handedness interaction found on cognitive tests. Without data about other aspects of cerebral organization, it seems premature to attempt to model the relationship between cerebral organization and cognitive ability using cerebral lateralization as the only parameter. 94 Future research should be aimed at: (a) verifying the results of this study in both clinical and experimental populations; (b) expanding the subjects to include both high and low intelligence groups; (c) expanding the number of tasks studied; (d) investigating other aspects of cerebral organization, such as the anterior-posterior dimension; and (e) examining absolute as well as relative levels of performance between the two hemispheres. APPENDICES APPENDIX A APPENDIX A Cerebral Lateralization Measures The following is an outline of the primary procedures used to measure lateralization of sensory functions in normal individuals. Dichotic Listening In a dichotic listening experiment, auditory stimuli are presented simultaneously to each ear. Subjects are later asked to recall the stimuli presented. Since the auditory pathways are such that the majority of auditory fibers decussate and terminate in the contralateral hemisphere, it is reasoned that fewer errors will be reported for stimuli presented to the ear contralateral to the hemisphere specialized for processing the stimuli. In a series of experiments, Kimura (1961 a,b) reported evidence supporting the validity of the dichotic listening procedure for determining the lateral organization of language functions. She used patients, undergoing surgery, in which the Wada procedure was used to locate language functions. In this procedure, sodium amytal, a reversible anesthetic, is injected into one of the carotid arteries. Contralateral hemiplegia results, along with cessation of functions lateralized ipsilateral to the injection site. 95 96 With the Wada procedure it is possible to determine in which hemisphere language functions are located. If language is lateralized to the left hemisphere, then speech will stop after a left-side injection, but not after a right-side injection. Kimura found that patients with language in the left hemisphere made fewer errors in recalling digits presented to the right ear. Conversely, patients with language functions in the right hemisphere made fewer errors in recalling digits presented to the left ear. Kimura (1964) also found that most subjects are better at recalling musical passages presented to the left ear, presumably reflecting right-hemisphere specialization for nonverbal functions. Visual Half—Field The visual half-field procedure is a visual analogue of the dichotic listening procedure. Subjects are asked to focus on a central visual field fixation dot. Stimuli are presented to the right and left of the fixation dot. The visual pathways are such that stimuli presented to the left visual field are initially transmitted to the right hemisphere and stimuli presented to the right visual field are initially transmitted to the left hemisphere. It is thought that greater accuracy and faster reaction times occur when detecting stimuli presented contralateral to the hemisphere specialized for processing the stimuli. In general, subjects show a right visual field advantage for 97 verbal stimuli and a left visual field advantage for nonverbal stimuli. Dichaptic Touch Dichaptic touch is used to measure lateralization of somatosensory functions. Stimuli are presented simultaneously to each hand. The somatosensory pathways are such that the majority of fibers terminate in the hemisphere contralateral to the side of the body sensing the stimuli. As expected, verbal stimuli are reported more accurately when presented to the right hand and line orientations are reported more accurately when presented to the left hand. APPENDIX B ACT - English Score 40 30 20 10 98 I l I L l l L -300 -200 -100 0 100 200 300 Spatial Lateralization Index (ms) Figure 1. Regression of ACT—English Test on the Spatial Lateralization Index. ACT Mathematics Score 40 30 20 10 99 L l l l l l J -300 -200 —100 0 100 200 300 Verbal Lateralization Index (ms) Figure 2. Regression of ACT-Mathematics Test on the Verbal Lateralization Index, for Left-Handers. ACT - Social Studies Score 40 30 20 10 100 1 1 1 I L 1 l -300 -200 -100 0 100 200 300 Spatial Lateralization Index (ms) Figure . Regression of ACT-Social Studies Test on the Spatial Lateralization Index. ACT - Social Studies Score 40 30 20 10 101 L l 1 l l I L -300 -200 —100 0 100 ‘ 200 300 Spatial Lateralization Index (ms) Figure 4. Regression of ACT-Social Studies Test on the Spatial Lateralization Index, for Left-Handers. ACT - Social Studies Score 40 30 20 10 102 1 In 4 1 1 J L, -300 -200 -100 0 100 200 300 Verbal Lateralization Index (ms) Figure 5. Regression of ACT—Social Studies Test on the Verbal Lateralization Index, for Left—Handers. Natural Science Score ACT- 40 3O 20 10 103 l J l L J l l -300 -200 -100 0 100 200 300 Spatial Lateralization Index (ms) Figure 6. Regression of ACT-Natural Science Test on the Spatial Lateralization Index. ACT- Natural Science Score 40 30 20 10 104 ¢ L 1 1 a L 1 -300 -200 -100 0 100 d 200 300 Spatial Lateralization Index (ms) Figure 7. Regression of ACT-Natural Science Test on the Spatial Lateralization Index, for Women. ACT- Natural Science Score 40 30 20 10 105 L I 1 L l L 1 —300 -200 —100 0 100 200 300 Verbal Lateralization Index (ms) Figure 8. Regression of ACT-Natural Science Test on the Verbal Lateralization Index, Women. for Left—Handed Mental Rotation Score 106 20 15 - / lol/ 5%. l l l l J 1 J -300 -200 -100 0 100 200 300 Spatial Lateralization Index (ms) Figure 9. Regression of Mental Rotation on the Spatial Lateralization Index. Speed on Embedded Figures Test 107 \ 1 I J J L 1 l i -300 —200 -100 0 100 200 300 Verbal Lateralization Index (ms) Figure 10. Regression of Embedded Figures Test on the Verbal Lateralization Index, for Women. Water Level Score 108 L 1 1 l 1 l 1 -300 -200 -100 0 100 200 300 Spatial Lateralization Index (ms) Figure 11. Regression of Water Level on the Spatial Lateralization Index. Water Level Score 109 1 1 1 I 1 l 1 -300 -200 -100 o 100 ' 200 300 Verbal Lateralization Index (ms) Figure 12. Regression of Water Level on the Verbal Lateralization Index. Water Level Score 110 l l 1 l l ‘ 1 -300 -200 -100 O 100 200 300 Verbal Lateralization Index (ms) Figure 13. Regression of Water Level on the Verbal Lateralization Index, for Women. Oral Word Association Test Score 60 50 4O 30 20 10 111 L 1 1 1 1 1 J -300 -200 -lOO O 100 200 300 Spatial Lateralization Index (ms) Figure 14. Regression of Oral Word Association Test on the Spatial Lateralization Index, for Left—Handers Oral Word Association Test Score 70 60 SO 40 30 20 10 112 1 l L I L l l -300 -200 -100 O 100 200 300 Verbal Lateralization Index (ms) Figure 15. Regression of Oral Word Association Test on the Verbal Lateralization Index. Oral Word Association Test Score 70 60 50 40 3O 20 10 113 1 1 1 1 1 1 1 -300 -200 -100 0 100 200 300 Verbal Lateralization Index (ms) Figure 16. Regression of Oral Word Association Test on the Verbal Lateralization Index, for Women. LI ST 01? REFERENCES LIST OF REFERENCES Allen, M., & Wellman, M. M. (1980). Hand position during writing, cerebral laterality and reading: Age and sex differences. Neuropsychologia, lg, 33-40. Andrews, G., Quinn, P. T., & Sorby, W. A. (1972). Stuttering: An investigation into verbal dominance for speech. Journal of Neurology, Neurosurgery, and Psychiatry, 32, 414-418. Andrews, R. J. (1977). Aspects of language lateralization correlated with family handedness. Neuropsychologia, lé! 769-778. Annett, M. (1967). The binomial distribution of right, mixed and left—handedness. Quarterly Journal of Experimental Psychology, l2, 327-333. Annett, M. (1972). The distribution of manual asymmetry. British Journal of Psychology, £3, 343-358. Annett, M. (1976). A coordination of hand preference and skill replicated. British Journal of Psychology, £1, 587-592. Annett, M. (1981). The genetics of handedness. Trends in Neurosciences, 3, 256-258. Annett, M. (1981). Handedness. In J. G. Beaumont (Ed.), Divided Visual Field Studies of Cerebral Organisation. London: Academic Press. Annett, M., & Turner, A. (1974). Laterality and the growth of intellectual abilities. British Journal of Educational Psychology, 53, 37-46. Benton, A. L., & Hamsher, K. D. (1976). Multilingual Aphasia Examination. Iowa City: University of Iowa. Benton, A. L., Hamsher, K. D., Varney, N. R., & Spreen, O. (1983). Contributions to Neuropsychological Assessment: A Clinical Manual. New York: Oxford. Benton, A. L., & Van Allen, M. W. (1968). Impairment in facial recognition in patients with cerebral disease. Cortex, 3, 344-358. 114 115 Berenbaum, S. A., & Harshman, R. A. (1980). On testing group differences in cognition resulting from differences in lateral specialization: Reply to Fennell et al. Brain and Cognition, ll, 209-220. Birkett, P. (1980). Predicting spatial ability from hemispheric "non-verbal" lateralisation: sex, handedness and task differences implicate encoding strategy effects. Acta Psychologica, fig, l-l4. Bradshaw, J. L., Gates, A., & Nettleton, N. C. (1977). Bihemispheric involvement in lexical decisions: Handedness and a possible sex difference. Neuropsychologia, lg, Bradshaw, J. L., & Nettleton, N. C. (1981). The nature of hemispheric specialization in man. The Behavioral and Brain Sciences, 3, 51-91. Briggs, G. C., & Nebes, R. D. (1975). Patterns of hand preference in a student population. Cortex, 11, 230-238. Briggs, G. C., & Nebes, R. D. (1976). The effects of handedness, family history, and sex on performance of a dichotic listening task. Neuropsychologia, 11, 129-134. Briggs, G. C., Nebes, R. D., & Kinsbourne, M. (1976). Intellectual differences in relation to personal and family handedness. Quarterlnyournal of Experimental Psychology, fig, 591-601. Bryden, M. P. (1973). Perceptual asymmetry in vision: Relation to handedness, eyedness, and speech lateralization. Cortex, 2, 418-435. Bryden, M. P. (1976). Response bias and hemispheric differences in dot localization. Perception & Psychophysics, l2! 23-28. Bryden, M. P. (1977). Measuring handedness with questionnaires. Neuropsychologia, l2! 617-624. Bryden, M. P. (1979). Evidence for sex-related differences in cerebral organization. In M. Wittig & A. C. Petersen (Eds.), Sex-Related Differences in Cognitive Functioning: Developmental Issues. New York: Academic Press. Bryden, M. P. (1982). Laterality: Functional Asymmetry in the Intact Brain, New York: Academic Press. Bryden, M. P., Hecaen, H., & DeAgostini, M. (1983). Patterns of cerebral organization. Brain and Language, 39, 249-262. 116 Chiarello, C., Dronkers, N. F., & Hardyck, C. (1984). Choosing sides: On the variability of language lateralization in normal subjects. Neuropsychologia, 33, 363-374. Coleman, R. I., & Deutsch, G. P. (1964). Lateral dominance and right—left discrimination: A comparison of normal and retarded readers. Perceptual and Motor Skills, l2! 43-50. Curry, F. K., & Gregory, H. H. (1969). The Performance of stutterers on dichotic listening tasks thought to reflect cerebral dominance. Journal of Speech and Hearing Research, lg, 73-82. Dagenbach, D. (1983). Hemispheric asymmetries in language processing: Effects of subject characteristics and perceptual modality. Unpublished masters thesis, Michigan State University, East Lansing, MI. Dee, H. L. (1971). Auditory asymmetry and strength of manual preference. Cortex, 3, 518-528. Deutsch, D. (1980). Handedness and memory for tonal pitch. In J. Herron (Ed.), Neuropsychology of Left-Handedness. New York: Academic Press. Eme, R., Stone, 5., & Izral, R. (1978). Spatial deficit in familial left-handed children. Perceptual and Motor Fagen-Dubin, L. (1974). Lateral dominance and development of cerebral specialization. Cortex, 12, 69—74. Fairweather, H. (1982). Sex differences: Little reason for females to play midfield. In J. G. Beaumont (Ed.), Divided Visual Field Studies of Cerebral Organisation. London: Academic Press. Fennell, E., Satz, P., Van Den Abell, T., Bowers, D., & Thomas, R. (1978). Visuospatial competency, handedness, and cerebral dominance. Brain and Language, g, 206-214. Francis, W. N., & Kucera, H. (1982). Frequency Analysis of English Usage. Boston: Houghton Mifflin. French, J. W., Ekstrom, R. B., & Price, L.A. (1963). Kit of Reference Tests for Cognitive Factors. Princeton, NJ: Educational Testing Service. Gardner, E. B., Denglish, A. G., Flannery, B. M., Hartnett, M. B., McCormicK, J. K., & Wilhelmy, B. B. (1977). Shape recognition accuracy and response latency in a bilateral tactile task. Neuropsychologia, lé, 607-616. 117 Gazzaniga, M. S., & LeDoux, J. E. (1978). The Integrated Mind. New York: Plenum. Geffen, D. S., & Hochberg, I. (1971). Ear laterality performance of children from low and middle socioeconomic levels on a verbal dichotic listening task. Cortex, 1, 193-203. Gilbert, C. (1977). Nonverbal perceptual abilities in relation to left-handedness and cerebral lateralization. Neuropsychologia, lg, 779-791. Gloning, I., Gloning, K., Haub, G., & Quatember, R. (1969). Comparison of verbal behavior in right-handed and non-right-handed patients with anatomically verified lesion of one hemisphere. Cortex, g, 43-52. Goodglass, H., & Quadfase, F. A. (1954). Language laterality in left-handed aphasics. Brain, 11, 521-548. Gregory, R., & Paul, J. (1980). The effects of handedness and writing posture on neuropsychological test results. Neuropsychologia, lg, 232-235. Hannay, H. J. (1976). Real or imagined incomplete lateralization of function in females. Perception & Psychophysics, lg, 349-352. Hardyck, C. (1977). Handedness and part-whole relationships: A replication. Cortex, lg, 177—183. Hardyck, C., Petrinovich, L., & Goldman, R. (1976). Left- handedness and cognitive deficit. Cortex, lg, 266-279. Harris, L. J. (1978). Sex differences in spatial ability: possible environmental, genetic, and neurological factors. In M. Kinsbourne (Ed)., Asymmetrical Function of the Brain. London: Cambridge University Press. Harris, L. J. (1980). Left-handedness: Early theories, facts, and fancies. In J. Herron (Ed.), Neuropsychology of Left-Handedness. New York: Academic Press. Harris, L. J. (1980). Lateralized sex differences: Substrates and significance. The Behavioral and Brain Sciences, g, 236-237. Harris, L. J. (1981). Sex related variations in spatial skill. In L. S. Liben, A. H. Patterson, & N. Newcombe (Eds.), gpatial Representation and Behavior Across the Life Span: Theory and Application. New York: Academic Press. 118 Harris, L. J., Hanley, C., & Best, C. T. (1978). Conservation of horizontality: Sex differences in sixth graders and college students. In Smart & Smart (Eds.), Adolescents and Relationships. New York: Macmillan. Harshman, R. A., Hampson, R., & Berenbaum, S. A. (1983). Individual differences in cognitive abilities and brain organization, Part 1: Sex and handedness differences in ability. Canadian Journal of Psychology, g1, 144-192. Hecaen, H., DeAgostini, M., & Monzon-Montes, A. (1981). Cerebral organization in left-handers. Brain and Language, lg, 261-284. Hecaen, H., & Sauguet, J. (1971). Cerebral dominance in left-handed subjects. Cortex, 1, 19-48. Hellige, J. B., & Cox, P. J. (1976). Effects of concurrent verbal memory on recognition of stimuli from the left and right visual fields. Journal of Experimental Psychology: Human Perception and Performance, 3, 210-221. Herrmann, D. J., & Van Dyke, K. A. (1978). Handedness and the mental rotation of perceived patterns. Cortex, lg, 521-529. Herron, J. (1980). (Ed.) Neuropsychology of Left- Handedness, New York: Academic Press. Herron, J., Galin, D., Johnstone, J., & Ornstein, R. E. (1979). Cerebral specialization, writing posture, and motor control of writing in left-handers. Science, 205, 1285-1289. Hicks, R. A., & Beveridge, R. (1978). Handedness and intelligence. Cortex, lg, 304-307. Higenbottam, J. A. (1973). Relationship between sets of lateral and perceptual preference measures. Cortex, g, 402-409. Hines, D. (1975). Independent functioning of the two cerebral hemispheres for recognizing bilaterally presented tachistoscopic visual half-field stimuli. Cortex, ll, 132-143. Hines, D., & Satz, P. (1974). Cross-model asymmetries in perception related to asymmetry in cerebral function. Neuropsychologia, lg, 239-247. Hines, M., & Shipley, C. (1984). Prenatal exposure to diethylstilbestrol (DES) and the development of sexually dimorphic cognitive abilities and cerebral lateralization. Developmental Psychology, lg, 80-94. 119 Hutt, C. (1972). Males and Females. Middlesex, England: Penguin. Johnson, 0., & Harley, C. (1980). Handedness and sex differences in cognition tests of brain laterality. cortex, l2, 73-820 Jones, B. (1982). Sex and handedness as factors in visual field organization for a categorization task. Journal of Experimental Psychology: Human Perception and Performance, g, 494-500. Jones, E., & Anuza, T. (1982). Effects of sex, handedness, stimulus and visual field on "mental rotation." Cortex, lg, 501-514. Kimura, D. (1961a). Cerebral dominance and the perception of verbal stimuli. Canadian Journal of Psychology, lg, 166-171. Kimura, D. (1961b). Some effects of temporal lobe damage on auditory perception. Canadian Journal of Psychology, lg, 156-165. Kimura, D. (1967). Functional asymmetry of the brain in dichotic listening. Cortex, g, 163-178. Kimura, D. (1969). Spatial localization in the left and right visual fields. Canadian Journal of Psychology, lg, 445-4580 Kimura, D. (1983). Sex differences in cerebral organization for speech and proxic functions. Canadian Journal of Psychology, g1, 19-35. Kinsbourne, M. (1975). The mechanism of hemispheric control of the lateral gradient of attention. In P. M. A. Rabbitt & S. Dornic (Eds.), Attention and Performance V. New York: Academic Press. Knox, A. W., & Boone, D. R. (1970). Auditory laterality and tested handedness. Cortex, g,, 164-173. Kraft, R. H. (1983). The effect of sex, laterality and familial handedness on intellectual abiliities. Neuropsychologia, gl, 79-90. Kraft, R. H. (1984). Lateral specialization and verbal/ spatial ability in preschool children: Age, sex and familial handedness differences. Neuropsychologia, 33, 319-336. 120 Kutas, M., McCarthy, G., & Donchin, E. (1975). Differences between sinistrals' and dextrals' ability to infer a whole from its parts: A failure to replicate. Neuropsychologia, lg, 455-464. Lake, D. A., & Bryden, M. P. (1976). Handedness and sex differences in hemispheric asymmetry. Brain and Language, Levy, J. (1969). Possible basis for the evaluation of lateral specialization of the human brain. Nature, 224, 614-615. Levy, J. (1974). Psychological implications of bilateral asymmetry. In S. Diamond and J. G. Beaumont (Eds.), Hemispheric Function in the Human Brain. London: Paul Elek, Ltd. Levy, J. (1980). Handwriting posture and cerebral organization: How are they related? Psychological Levy, J., & Gur, R. C. (1980). Individual differences in psychoneurological organization. In J. Herron (Ed.), Neuropsychology of Left-Handedness. New York: Academic Press. Levy, J., & Reid, M. (1976). Variations in writing posture and cerebral organization. Science, 194, 337-339. Levy, J., & Reid, M. (1978). Variations in cerebral organization as a function of handedness, hand posture in writing, and sex. Journal of Experimental Psychology: General, 107, 119-144. Lewis, R. S., & Harris, L. J. The relationship between cerebral hemisphere specialization and cognitive ability: A critical review, and suggested criteria for a test of the lateralization-cognition hypothesis. Manuscript in preparation. Lewis, R., & Kamptner, N. L. (1983). Spatial task performance and sex differences in normal and brain- lesioned patients. Paper presented at the 9lst meeting of the American Psychological Association, Anaheim, CA. Leiber, L. (1976). Lexical decisions in the right and left cerebral hemispheres. Brain and Language, g, 443-450. Linn, M. C., & Peterson, A. C. (1983). Emergence and characterization of gender differences in spatial ability: A meta-analysis. Paper presented at the 9lst meeting of the American Psychological Association, Anaheim, CA. 121 Lishman, W. A., & McMeekan, E. R. L. (1977). Handedness in relation to direction and degree of cerebral dominance for language. Cortex, lg, 30-43. Maccoby, E. E., & Jacklin, C. N. (1974). The Psychology of Sex Differences. Stanford, CA: Stanford University Press. Masure, M. C., & Benton, A. L. (1983). Visuospatial performance in left-handed patients with unilateral brain lesions. Neuropsychologia, gl, 179-181. McGee, M. G. (1982). Spatial abilities: The influence of genetic factors. In M. Potegal (Ed.), Spatial Abilities: Development and Physiological Foundations. New York: Academic Press. McGlone, J. (1980). Sex differences in human brain asymmetry: A critical survey. The Behavioral and Brain Sciences, g, 215-263. McGlone, J., & Davidson, W. (1973). The relation between cerebral speech laterality and spatial ability with special reference to sex and hand preference. Neuropsychologia, ll, 105-113. McKeever, W. F., & VanDeventer, A. D. (1977). Visual and auditory language processing asymmetries: Influences of handedness, familial sinistrality, and sex. Cortex, lg, 225-241. Mebert, C. J., & Michel, G. F. (1980). Handedness in artists. In J. Herron (Ed.), The Neuropsychology of Left-Handedness. New York: Academic Press. Miller, E. (1971). Handedness and the pattern of human ability. British Journal of Psychology, gg, 111-112. Moscovitch, M. (1979). Information processing and the cerebral hemispheres. In M. S. Gazzaniga (Ed.), Handbook of Behavioral Neurobiology, Vol. 2: Neuropsychology. New York: Plenum. Nebes, R. D. (1971). Handedness and the perception of part-whole relationships. Cortex, 1, 350-356. Nebes, R. D., (1976). The use of imagery in memory by right and left handers. Neuropsychologia, lg, 505-508. Nebes, R. D., & Briggs, G. C. (1974). Handedness and the retention of visual material. Cortex, lg, 209-214. Newcombe, F., & Ratcliffe, G. (1973). Handedness, speech lateralization and abilities. Neuropsychologia, ll, 122 Nichelli, P., Manni A., & Faglioni, P. (1983). Relationships between speed, accuracy of performance and hemispheric superiorities for visuo-spatial pattern processing in the two sexes. Neuropsychologia, gl, 625-632. Orsini, D. L., Satz, P., Light, R. K., & Soper, H. V. (1984). Evaluation of familial sinistrality as an indicator of cerebral organization. Paper presented at the 12th Annual Meeting of the International Neuropsychological Society Conference in Houston, Texas, February 1984. Orton, S. (1937). Reading, Writing, and Speech Problems in Children. New York: Norton. Parlow, S. (1978). Differential finger movements and hand preference. Cortex. lg, 608-611. Peterson, J. M., & Lansky, L. M. (1974). Left-handedness among architects: Some facts and speculation. Perceptual Motor Skills, gg, 547-550. Piaget, J., & Inhelder, B. (1956). The Child's Conception of Space. New York: Humanities Press. Piazza, D. (1980). The influence of sex and handedness in the hemispheric specialization of verbal and nonverbal tasks. Neuropsychologia, lg, 163-176. Rasmussen, T., & Milner, B. (1977). The role of early left-brain injury in determining lateralization of cerebral speech functions. Annals of the New York Academy of Sciences, 299, 355-369. Raven, J. C. (1960). Guide to the Standard Progressive Matrices. London: H. K. Lewis. Rizzolatti, G., & Buchtel, H. A. (1977). Hemispheric superiority in reaction time to faces: A sex difference. Rizzolatti, G., Umilta, C., & Berlucchi, G. (1971). Opposite superiorities of the right and left cerebral hemispheres in discriminative reaction time to physiognomical and alphabetical material. Brain, gg, 431-442. Sabatino, D., & Becker, J. T. (1971). Relationship between lateral preference and selected behavioral variables for children failing academically. Child Development, 42, 2055-2060. —_ 123 Samar, V. J. (1983). Multidimensional influences of cerebral asymmetries on visual half-field asymmetries and cognitive skills. Brain and Cognition, g, 355-382. Sanders, B., Wilson, J. R., & Vandenberg, S. G. (1982). Handedness and spatial ability. Cortex, lg, 79-90. Sasanuma, S., & Kobayashi, Y. (1978). Tachistoscopic recognition of line orientation. Neuropsychologia, lg, 239-242. Satz, P. (1980). Incidence of aphasia in left-handers: A test of some hypothetical models of cerebral speech organization. In J. Herron (Ed.), The Neuropsychology of Left-Handedness. New York: Academic Press. Satz, P., Achenback, K., & Fennell, E. (1967). Correlation between assessed manual laterality and predicted speech laterality in a normal population. Neuropsychologia, g, 295-310. Satz, P., Achenbach, K., Pattishall, E., & Fennell, E. (1965). Order of report, ear asymmetry and handedness in dichotic listening. Cortex, l, 377-396. Satz, P., Fennell, E., & Jones, M. B. (1965). Comments on: A model of inheritance of handedness and cerebral dominance. Neuropsychologia, 1, 101-103. Satz, P., & Zaide, J. (1982). Sex differences: Clues or myths on genetic aspects of speech and language disorders. In C. Ludlow (Ed.), Genetic Aspects of Speech and Language Disorders. New York: Academic Press. Searleman, A. (1980). Subject variables and cerebral organization for language. Cortex, lg, 239-254. Segalowitz, S. J. (1983). Two Sides of the Brain. Englewood Cliffs, NJ: Prentice Hall. Sergent, J. (1983). Role of the input in visual hemispheric asymmetries. Psychological Bulletin, gg, 481-512. Shankweiler, D., & Studdert-Kennedy, M. A continuum of lateralization for speech perception? Brain and Language, g, 212-225. Smith, L. C. & Moscovitch, M. Writing posture, hemispheric control of movement and cerebral dominance in individuals with inverted and non-inverted postures during writing. Neuropsychologia, l1, 637-644. 124 Todor, J. I. (1980). Sequential motor ability of left-handed inverted and non-inverted writers. Acta Psychologica, gg, 165-173. Umilta, C., Brizzolara, D., Tabossi, P., & Fairweather, H. (1978). Factors affecting face recognition in the cerebral hemispheres: Familiarity and naming. In J. Requin (Ed.), Attention and Performance VII. Hillsdale, NJ: Erlbaum Associates. Vandenberg, S. & Kuse, A. (1978). Mental rotations: A group test of three dimensional spatial visualization. Perceptual and Motor Skills, g1, 599—604. Waber, D. P. (1977). Sex differences in mental abilities, hemispheric lateralization, and rate of physical growth at adolescence. Developmental Psychology, lg, 29-38. Warrington, E. K., & Pratt, R. T. C. (1973). Language laterality in left handers assessed by unilateral ECT. Neuropsychologia, ll, 423-428. Weber, A. M., & Bradshaw, J. L. (1981). Levy and Reid‘s neurological model in relation to writing hand/posture: An evaluation. Psychological Bulletin, gg, 74-88. Wechsler, D. (1981). WAIS-R Manual. New York: The Psychological Corporation. Witkin, H. A. (1971). The Embedded Figures Test. Palo Alto, CA: Consulting Psychologists Press. Witkin, H. A., Dyk, R. B., Faterson, G. E. Goodenough, D. R., & Karp, S. A. (1962). Psychological Differentiation. New York: Wiley. Wussler, M., & Barclay, A. (1970). Cerebral dominance and its relation to psycholinguistic skills and reading disability. Perceptual and Motor Skills, gl, 419-425. Yen, W. M. (1975). Independence of hand preference and sex-linked genetic effects on spatial performance. Perceptual and Motor Skills, gl, 311-318. Zoccolotti, P., & Oltman, P. K. (1978). Field dependence and lateralization of verbal and configurational processing. Cortex, lg, 155-163. Zurif, E. B., & Bryden, M. P. (1969). Familial handedness and left-right differences in auditory and visual perception. Neurgpsychologia, 1, 170-187. "TITI'ITIQHILFIMIMEIMlllfylfll'!iflljlflfflfllll'“ w»s--—-“-_. , _ p_