UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Automatic and attentionally controlled processing in the cerebral hemispheres Eglin, Susan Mirjam 1982

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

Item Metadata

Download

Media
831-UBC_1982_A8 E44.pdf [ 3.3MB ]
Metadata
JSON: 831-1.0095348.json
JSON-LD: 831-1.0095348-ld.json
RDF/XML (Pretty): 831-1.0095348-rdf.xml
RDF/JSON: 831-1.0095348-rdf.json
Turtle: 831-1.0095348-turtle.txt
N-Triples: 831-1.0095348-rdf-ntriples.txt
Original Record: 831-1.0095348-source.json
Full Text
831-1.0095348-fulltext.txt
Citation
831-1.0095348.ris

Full Text

AUTOMATIC AND ATTENTIONALLY CONTROLLED PROCESSING IN THE CEREBRAL HEMISPHERES by SUSAN MIRJAM EGLIN The University of B r i t i s h Columbia, 1982 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS in THE FACULTY OF GRADUATE STUDIES (Department of Psychology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1982 Susan Mirjam E g l i n , 1982 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. I t i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of PS<dCtfOLO€!j The University of B r i t i s h Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 DE-6 (3/81) Page i i Abstract The thesis describes research investigating differences between the two hemispheres i n automatic and i n a t t e n t i o n a l l y controlled processes. It i s suggested that the interaction between these two processes may be a source of hemispheric differences. Three d i f f e r e n t paradigms that each imply d i f f e r e n t d e f i n i t i o n s of automatic and atte n t i o n a l l y controlled processes are used: A paradigm used to demonstrate i l l u s o r y conjunctions, a modified priming paradigm and a modified Stroop-task. Converging evidence from a l l three paradigms indicates that automatic processes are common to both hemispheres. Lateral asymmetries only emerge i n attentional e f f e c t s . For verbal information, s e l e c t i v e attention mechanisms i n the l e f t hemisphere are found to be selective for l e f t hemisphere items only, whereas right hemisphere mechanisms are sensitive to information from both hemispheres. The right hemisphere appears to be able to give some automatic support to attended verbal processing i n the l e f t hemisphere, while the reverse seems to be more d i f f i c u l t . Page i i i Table of Contents Abstract i i L i s t of Tables.... i v L i s t of Figures v Acknowledgement v i I. INTRODUCTION 1 1.1. Automatic and attentionally controlled processing: Evidence from three paradigms 5 1.1.1. I l l u s o r y conjunctions 5 1.1.2. The priming paradigm.... 6 1.1.3. The Stroop-task 9 1.1.4. Converging results? 12 1.2. Hemispheric processing: Functional and capacity models 14 1.2.1. Functional dichotomies 14 1.2.2. Capacity models 18 1.2.3. Functional l o c i 21 II. EXPERIMENT 1: I l l u s o r y conjunctions 22 III. EXPERIMENT 2: A priming task 34 IV. EXPERIMENT 3: A Stroop-task 5 0 V. GENERAL DISCUSSION 64 REFERENCES 71 Page iv L i s t of Tables I. Mean number per t r i a l of d i f f e r e n t types of responses for a l l subjects (n=30) in both visual f i e l d s 26 II. Mean number per t r i a l of di f f e r e n t types of responses for both visual f i e l d s 31 III. The d^s for a l l three target word conditions as well as priming and interference effects i n a l l s p a t i a l arrangements 45 Page v L i s t of Figures 1. Example of a stimulus display with t i l t e d "S"-and >$'-signs and arrows 29 2. Predictions for priming and interference effects derived from d i f f e r e n t hemispheric models 39 3. Mean reaction times (RTs) for correct responses and error rates for consistent (CO) and incon-sistent (INCO) t r i a l s in condition 1 (5 0/50) for both v i s u a l f i e l d s 54 4. Mean reaction times (RTs) for correct responses and error rates for consistent (CO) and incon-si s t e n t (INCO) t r i a l s for both v i s u a l f i e l d s i n the two extreme conditions 56 5. Stroop-effect (inconsistent-consistent) for both visu a l f i e l d s and the two extreme conditions 57 6. Effects of attentional and automatic weights for both vi s u a l f i e l d s 60 Page v i Acknowledgement I would l i k e to thank my committee members Dr. R. Tees and Dr. L. H. Ward for constructive c r i t i c i s m on an e a r l i e r draft of t h i s thesis. I am very grateful to Jackie Burkell for help in running and analyzing Experiment 2 and to Dr. D. Hahnemann and Diane Chajczyk for writing the programs. Special thanks go to my supervisor Dr. A. Treisman for great support and patience and an invaluable education in experimental psychology. Page 1 I. INTRODUCTION The d i s t i n c t i o n between two modes of processing, automatic and attentionally controlled ( S c h i f f r i n and Schneider, 1977) has been an important focus for much research i n recent years. Typ i c a l l y , these modes of processing are viewed as h i e r a r c h i c a l l y organised, with attentional processes either operating on or being at least strongly influenced by the outcomes of p r i o r , automatic processes. S h i f f r i n and Schneider (1977) gave their subjects d i f f e r e n t target sets of l e t t e r s or d i g i t s on each t r i a l . They found that i f targets and d i s t r a c t o r s were mapped consistently across t r i a l s , i . e . none of the targets was ever used as a d i s t r a c t o r , target detection seemed to become automatic and independent of load. If the assignment of targets and dis t r a c t o r s was inconsistent across t r i a l s , however, search seemed to be a t t e n t i o n a l l y controlled and strongly dependent on load. Different and sometimes new c h a r a c t e r i s t i c s of the two modes have been emphasized by other writers. The terms automatic and attentionally controlled have been used to describe involuntary versus voluntary processing (e.g. Posner and Snyder, 1975), preattentive versus attentive processing (e.g. Neisser, 1967, Treisman and Gelade, 1980), unconscious versus conscious processing (e.g. Posner and Snyder, 1975, Marcel, in press) or p a r a l l e l versus s e r i a l processing (e.g. Treisman and Gelade, 1980) . Page 2 Dichotomous descriptions of information processing also p r e v a i l in research on the cerebral hemispheres, gaining i n t u i t i v e support from the hemispheres' anatomical structure. Thus, hemispheric functioning has been described i n terms of two d i f f e r e n t modes of processing, as, for example, a n a l y t i c versus h o l i s t i c (e.g. Gardner, 1974) or s e r i a l versus p a r a l l e l processing (e.g. Cohen, 1973). Another approach to hemispheric differences has emphasized differences in the d i f f e r e n t types of information processed i n each hemisphere, as, for example, verbal versus visuospatial information (e.g. Kimura, 1966). In a rather recent and very interesting approach Sergent (1982) claims that the hemispheres d i f f e r in t h e i r s e n s i t i v i t y to the s p a t i a l frequencies of v i s u a l percepts. Some of the older functional dichotomies are controversial, however. They are not s u f f i c i e n t for explaining the great d i v e r s i t y of experimental results. S p e c i f i c a l l y , the l e v e l or stage at which functional differences emerge, and the s t a b i l i t y of functional differences over d i f f e r e n t task s i t u a t i o n s , remained open questions. Attempts to answer such questions gave r i s e to a number of new hypotheses that specify a precise functional l e v e l , i . e . the " i n t e r f a c e ' between preattentive and attentional processes, at which hemispheric differences are believed to emerge (e.g. Kinsbourne, 1982, Moscovitch, 1979). Other models deal with s i t u a t i o n a l variables believed to produce asymmetries that are superimposed on basic asymmetries in hemispheric function. Page 3 Above a l l , attentional factors were thought to be important (e.g. Kinsbourne, 1975, Hellige, Cox and Litvac, 1979, Friedman and Poison, 1981). Evidence from c l i n i c a l studies strongly suggests that functional differences between the hemispheres do e x i s t ; yet, despite a vast number of experimental findings and numerous models, they s t i l l seem to evade a comprehensive and yet parsimonious description. The present study was undertaken in order to examine whether there are differences i n hemispheric functioning which can consistently be described i n terms of automatic and atte n t i o n a l l y controlled processing. For the sake of s i m p l i c i t y , a h i e r a r c h i c a l view of information processing i s adopted. It i s assumed that automatic processes occur prior to a t t e n t i o n a l l y controlled processes, i . e . attention i s thought to operate on evidence from the p r i o r automatic stage. The hemispheres are believed to be similar in their automatic stage, independent of whether t h i s stage involves only early sensory or also higher l e v e l s of information processing. However, the hemispheres are thought perhaps to d i f f e r in the way in which automatic and attentionally "controlled processes inte r a c t . The interaction between the two modes of processing may be characterized by emphasizing any of the following relations between automatic and attentional l e v e l s : 1) attention may integrate - i . e . information from the automatic stage may be synthesized into higher order units. Page 4 2) attention may select - i.e. information from the automatic stage may either be rejected or selected for further processing. 3) attention may weight - i.e. weights varying i n size and/or sign may be attached to evidence from the automatic stage Differences between the hemispheres could conceivably take any of the above mentioned forms and i t could prove d i f f i c u l t to decide conclusively where exactly they a r i s e . The present study was designed to y i e l d some information en each of these three p o s s i b i l i t i e s . Three d i f f e r e n t paradigms that deal with automatic and attentionally controlled processes were chosen. They a l l bear on the possible r e l a t i o n between the two processes, integration, selection and weighting. These re l a t i o n s imply d i f f e r e n t d e f i n i t i o n s and functions for attentional processing. By using d i f f e r e n t paradigms rather than variants of a single one, i t was hoped that converging or complementary evidence for hemispheric functioning would be found. The paradigm o r i g i n a l l y used by S h i f f r i n and Schneider (1977), however, which involves giving subjects memory loads and also extensive practice, was not chosen. It was f e l t that memory loads and practice could only complicate assumptions about processes occuring i n either one or the other hemisphere. Page 5 1.1. Automatic and attentionally controlled processing: Evidence from three paradigms 1.1.1. Il l u s o r y conjunctions : Quite recently a feature-integration theory of attention has been proposed (Treisman and Gelade, 19 80 , Treisman and Schmidt, 1982) . It assigns focal attention and thus s e r i a l processing a central function in'a defined stage of object perception. The theory distinguishes between an e a r l y p a r a l l e l and a subsequent s e r i a l stage of processing i n the perception of objects. During the early p a r a l l e l stage, features on separable dimensions are registered (where 'dimension' refers to the set of possible mutually exclusive states of a variable, e.g. the set of colors, and 'feature' refers to a particular value on a dimension, e.g. red). Features within a single dimension may be p a r t l y organized within their own s p a t i a l map at t h i s early, preattentive stage. However, they can only be integrated with features from other dimensions and formed into multidimensional objects by means of focal attention at a l a t e r , s e r i a l stage of processing. It i s important to note that t h i s focal attention need not lead to conscious awareness. Perception and integration of features may both occur unconsciously. One of the theory's predictions, namely that i f attention i s overloaded or prevented, feature integration w i l l be Page 6 interfered with to the extent that i l l u s o r y conjunctions of features may occur, has recently been v e r i f i e d i n a number of experiments (Treisman and Schmidt, 1982). Two t y p i c a l experimental paradigms contrast free report and search for objects defined by conjunctions of properties. In t h i s model attention operates on the outcomes of p r i o r automatic processes to ensure the correct perception of objects whenever top-down constraints are i n s u f f i c i e n t . Thus, conjunction errors, i f they occur, are preattentive in the sense that attention has f a i l e d , and they provide a means to investigate the effects of a preattentive, p a r a l l e l and an attentional, s e r i a l mode of processing at a defined functional l e v e l in object perception. More s p e c i f i c a l l y , the c h a r a c t e r i s t i c s of the integrative function of attention may be studied. 1.1.2. The priming paradigm : In a t y p i c a l priming task, the subject makes a judgement (usually a l e x i c a l decision) about a target stimulus (target), which i s preceded by a cueing or priming stimulus (prime). Depending on the r e l a t i o n s h i p between the two s t i m u l i , subjects' responses to the target may be influenced by the cue. The nature of t h i s relationship between target and cue may be investigated by varying the following three c h a r a c t e r i s t i c s : The c u e - v a l i d i t y ( i . e . i t s predictive value), the temporal relation between target and cue ( i . e . Page 7 the stimulus onset asynchrony (SOA) ) , and the type of association between target and cue, which can either be p r i o r and habitual or novel and experimentally defined. The e f f e c t s of such variations have been studied by means of the cost/benefit analysis (Posner and Snyder, 1975) . T y p i c a l l y , with short SOAs (< ca. 300 msec) , p r i o r and habitual associations between prime and target w i l l lead to f a c i l i t a t i o n (benefit), independent of prime v a l i d i t y ; with longer SOAs (> ca. 300 msec) t h i s e f f e c t i s reduced. Instead, i n the case of high cue v a l i d i t y , novel experimentally defined associations may lead to f a c i l i t a t i o n (benefit), i f subjects' expectations are confirmed, or to interference (cost) , i f subjects' expectations are not confirmed (Posner and Snyder, 1975, Neely, 1977). Early f a c i l i t a t i o n e f f e c t s are thought to be produced by ' i n h i b i t i o n l e s s spreading a c t i v a t i o n ' (Posner and Snyder, 1975), which takes place i n v o l u n t a r i l y and i s not under subjects' control. In the case that prime and target are words, for example, activation i s thought to spread from the long-term memory node or logogen of the prime to those of semantically associated words, which include the target. These words are thus activated to a l e v e l closer to their threshold and need less information to reach i t , which i s reflected i n f a c i l i a t i o n for the l e x i c a l decision (Morton, 1969). Late f a c i l i t a t i o n and i n h i b i t i o n e f f e c t s are thought to be produced by a 'slow limited-capacity conscious-attention mechanism' (Posner and Snyder, 1975). Page 8 Typi c a l l y , cue and target are presented successively and there i s no positional uncertainty for either of them. Subjects always know when and where cue and target appear. Thus, only one kind of selection i s involved: an internal target i s selected by priming or expectancy. This may occur either automatically or v o l u n t a r i l y . There i s another kind of sele c t i o n , however, i n which an external stimulus relevant to the response i s selected ( f i l t e r i n g ) . The priming paradigm may be modified to introduce such a target selection by presenting a target with a simultaneous d i s t r a c t o r . The paradigm w i l l then be interference dominant, since selecting the appropriate target i s interfered with by an irrelevant ncntarget. Priming i n t h i s case i s characterized as a reduction of interference rather than a pure f a c i l i t a t i o n e f f e c t . Any ncntarget w i l l primarily cause interference, but a prime w i l l do less so than an unrelated ncntarget. Thus, the priming paradigm can be used to study the following forms of select i o n : automatic selection of an internal target (automatic priming), attentional selection of an internal target (priming by expectancy) or attentional selection of an external target ( f i l t e r i n g ) . Page 9 1.1.3. The Stroop-task : In a Stroop-task, subjects have to make a judgement about one dimension of a multidimensional stimulus. T y p i c a l l y , the color of a word i s used as the relevant or reported dimension and the i d e n t i t y of the word i s used as the irrelevant or unreported dimension. Information from the unreported dimension may influence performance on the reported dimension. If the r e l a t i o n s h i p between dimensions i s consistent (e.g. the word 'RED' printed i n red ink) f a c i l i t a t i o n r e s u l t s . An inconsistent relationship (e.g. the word 'RED' printed i n green ink), however, w i l l i nterfere with the judgement about the reported dimension. T y p i c a l l y i t had been thought that the processing of the unreported dimension was automatic i n the sense of being strategy-invariant. Recently i t has been shown, however, that the Stroop-task may also involve a t t e n t i o n a l l y controlled, s t r a t e g i c processes (Logan and Zbrodoff, 1979, Logan, 1980). By varying the r e l a t i v e frequency of consistent (reported and unreported dimension specify the same meaning) and inconsistent t r i a l s (the two dimensions specify a d i f f e r e n t meaning) , they were able to show that subjects may use a strategy of dividing their attention between the two dimensions. Their results indicate that when inconsistent t r i a l s are more frequent than consistent ones, subjects are able to s t r a t e g i c a l l y adjust to t h i s s i t u a t i o n : they are a c t u a l l y faster responding to inconsistent than to consistent Page 10 t r i a l s . This is a reversal of the usual Stroop-effect and strong evidence against the notion of i t s pure automaticity. Subjects must be attending to the unreported dimension in order to show such a strategy e f f e c t . Logan and Zbrodoff (1979) and Logan (1980) suggested that t h i s e f f e c t can be described by a model of weighted decision making. In the two-choice situation of the simplest Stroop-task (the reported dimension has two possible a l t e r n a t i v e outcomes) evidence for one al t e r n a t i v e i s evidence against the other. Information about the unreported dimension may be viewed as s h i f t i n g the i n i t i a l state of evidence about the reported dimension toward one decision threshold or the other. The current state of evidence bearing on the decision i s expressed as a weighted sum of the evidence a v a i l a b l e about the reported dimension and the evidence a v a i l a b l e about the unreported dimension (Logan and Zbrodoff, 1979). In t h i s framework, d i v i d i n g attention between the two dimensions means that subjects a t t e n t i o n a l l y assign weights to evidence a v a i l a b l e from each dimension. If the two dimensions are consistent, each of these attentional weights w i l l have the same ( i . e . a positive) sign. If the two dimensions are inconsistent, the weights w i l l have opposite signs; the one attached to evidence from the reported dimension w i l l be po s i t i v e , the other one, which i s assigned to the unreported dimension, w i l l be negative. In addition, evidence from the unreported dimension may also be weighted automatically. Thus, i f responses to inconsistent t r i a l s are found to be Page 11 faster when inconsistent t r i a l s are more frequent than consistent ones, the weight assigned at t e n t i o n a l l y to the unreported dimension must be larger than the automatic weight i n order to overcome habitual response tendencies, but i t must remain small enough that i t does not produce a response without some information from the reported dimension. Automatic weights are assumed to be constant in sign and magnitude, whereas attentional weights may vary i n sign and magnitude r e f l e c t i n g the current strategy that allows for optimal performance. The ef f e c t s of attentional and automatic weights are assumed to combine a d d i t i v e l y (Logan and Zbrodoff, 19 79, Logan, 19 80). In the Stroop-task, then, when consistent and inconsistent t r i a l s are equally frequent, processing may be automatic i n the sense of being strategy-invariant. In addition, when the r e l a t i v e frequency of consistent and inconsistent t r i a l s i s varied, attentional effects r e f l e c t i n g s t r a t e g i c control may be involved. Thus, the paradigm allows one to investigate automatic (strategy-invariant) and attentionally controlled effects under divided attention. In p a r t i c u l a r , e f f e c t s of attentional (strategic) control can be described i n terms of weights attached to evidence from automatic processes. Page 12 1.1.4. Converging results? In the present study, i l l u s o r y conjunctions were expected to shed l i g h t on e a r l y perceptual operations occuring i n the absence of focal attention. Conjunction errors were not used to investigate the integrative function of attention, however. The priming paradigm was used to investigate automatic priming as well as selective attention ( f i l t e r i n g ) . F i n a l l y , the Stroop-paradigm was used to examine automatic, strategy-invariant processes on the one hand, and effects of attentional or s t r a t e g i c control on the other hand. The three'par ad igms implicate three d i f f e r e n t d e f i n i t i o n s of attention that may i n turn implicate d i f f e r e n t l e v e l s i n information processing at which automatic and attentional processes in t e r a c t . I l l u s o r y conjunctions would c e r t a i n l y r e f l e c t a f a i l u r e of attention at a very early stage i n perception. Interference and f a c i l i t a t i o n found i n a Stroop-task are believed to arise rather late i n the processing sequence when response selection takes place. However, the a t t e n t i o n a l and automatic processing of the unreported dimension may occur at an e a r l i e r l e v e l . Priming e f f e c t s i n a l e x i c a l decision presumably occur at a l e x i c a l or semantic l e v e l of word recognition. Interference effects of selective attention, however, are more d i f f i c u l t to locate. If selection of the target i s based on very simple physical properties of the s t i m u l i , i t may occur very early i n processing. It i s not known, however, what kind of a Page 13 selection strategy subjects w i l l actually use. They may choose to divide their attention between the s t i m u l i , i n which case selection would only take place at a l a t e r stage i n processing. Thus, the locus of interference due to selective attention i s d i f f i c u l t , to determine i n t h i s paradigm. As to the three d i f f e r e n t aspects of interaction between automatic and attentionally controlled processes, the present study investigated only two of them: selection and weighting. To summarize, the three paradigms should y i e l d evidence about automatic processes at d i f f e r e n t l e v e l s in information processing and should emphasize d i f f e r e n t c h a r a c t e r i s t i c s of the re l a t i o n between the two processes. Page 14 1.2. Hemispheric Processing: Functional and Capacity Models 1.2.1. Functional dichotomies A c r u c i a l role f o r language had long been claimed f o r the l e f t hemisphere (LH) by Broca and Wernicke. It was not u n t i l f a i r l y recently, that the r i g h t hemisphere (RH) was also credited with the c a p a b i l i t y to process l i n g u i s t i c material. Kimura (1966) s t i l l characterized the LH and RH as exclusively specialized f o r verbal and s p a t i a l processing, respectively. Supportive evidence f o r t h i s dichotomy came from a vast number of studies, showing, for example, a RVF(LH) su p e r i o r i t y f o r word recognition (e.g. E l l i s , 1974 , Hines, 1976), and a RH superiority f o r shapes or faces (e.g. R i z z o l a t i et. a l . , 1971). C l i n i c a l studies on neurosurgical patients demonstrated, however, that the RH is quite capable of comprehending words, although i t i s mute and has no access to speech production. Only recently Zaidel (1978a) , employing novel techniques that permit prolonged u n i l a t e r a l stimulation, showed that while the RH has no speech, i t has some writing, substantial visual vocabularies and s u r p r i s i n g l y r i c h auditory lexicons. The RH seems to have very l i t t l e syntax, however. He suggested, that the semantic structure of the RH vocabulary i s more d i f f u s e and connotative than in the LH. Since his subjects are s p l i t - b r a i n or hemispherectomized patients, they may actually display more sophisticated language functions i n the RH than Page 15 would be found i n the intact brain. It i s clear, however, that describing the RH as nonverbal and the LH as verbal i s an oversimplification. In an attempt to l i n k t h i s verbal/visuospatial d i s t i n c t i o n to a more general theory of information processing, Cohen (1973) proposed a s e r i a l versus p a r a l l e l d i s t i n c t i o n . These two modes of processing had been suggested to be basic to a l l information processing. Since verbally mediated matching had been found to be generally s e r i a l , whereas p a r a l l e l processing i s usually confined to matching on the basis of physical c h a r a c t e r i s t i c s (Beller, 1970), Cohen chose a matching task with both verbal and nonverbal s t i m u l i . She examined reaction times to judge a set of items 'same" ( a l l identical) or "d i f f e r e n t " (one item d i f f e r i n g from the r e s t ) . She found that i f the stimuli were l i n g u i s t i c ( i . e . l e t t e r s ) , increasing t h e i r number produced an increase i n reaction time in the LH, as in s e r i a l processing, but not in the RH, as i n p a r a l l e l processing. When the items were unnameable shapes, however, both hemispheres seemed to process i n p a r a l l e l . Thus, she suggested that l i n g u i s t i c material may be analyzed either verbally or v i s u o s p a t i a l l y and she proposed that the LH employs a s e r i a l , verbal mode of processing, whereas the RH employs a p a r a l l e l , v i s uospatial mode of processing. Even i f r e s t r i c t e d to matching tasks with verbal material, the resul t s could not be consistently replicated (e.g. White and White, 1975). In addition, there i s evidence that the RH might be specialized f o r perceiving faces (e.g. R i z z o l a t i et. Page 16 a l . , 1971) and memorizing melodies ( M i l n e r , 1962), both of which i t does i n a s e r i a l manner. Thus, the s e r i a l / p a r a l l e l d i s t i n c t i o n d i d not prove to be a very u s e f u l one. In a r e c e n t and v e r y i n t e r e s t i n g model, Sergent (1982) suggests that a v e r b a l / n o n v e r b a l d i s t i n c t i o n does not grasp an aspect of v i s u a l information t h a t i s e s s e n t i a l f o r i n f o r m a t i o n p r o c e s s i n g i n the hemispheres. Instead, she proposes that a more b a s i c dichotomy emerges from the f a c t t h a t v e r b a l s t i m u l i ( i . e . l e t t e r s ) r e p r e s e n t a f i n i t e set of h i g h l y f a m i l i a r , o v e r l e a r n e d and p r e c i s e l y s t r u c t u r e d s t i m u l i . ' V i s u o s p a t i a l ' m a t e r i a l , however, as used i n some l a t e r a l i t y s t u d i e s , r e p r e s e n t s a p o t e n t i a l l y i n f i n i t e and u n f a m i l i a r s e t of s t i m u l i . Consequently, g i v e n b r i e f exposure and l a t e r a l viewing c o n d i t i o n s , v e r b a l and nonverbal s t i m u l i may d i f f e r as to how completely o r a c c u r a t e l y they can be encoded, and they may not achieve a q u a l i t a t i v e l y s i m i l a r v i s u a l r e p r e s e n t a t i o n . Sergent (1982) proposes that the RH i s more e f f i c i e n t at p r o c e s s i n g e a r l y - a v a i l a b l e l o w - s p a t i a l - f r e q u e n c y c o n t e n t s , whereas the LH i s b e t t e r at d e a l i n g with l a t e r - a v a i l a b l e high frequency contents of a v i s u a l image. Her model leads to i n t r u i g i n g l y p l a u s i b l e e x p l a n a t i o n s f o r a v a r i e t y of p r e v i o u s l y p u z z l i n g f i n d i n g s : F o r example, t h a t very b r i e f exposure d u r a t i o n s o r stimulus degradation, which w i l l prevent higher f r e q u e n c i e s from becoming a c c e s s i b l e , w i l l t y p i c a l l y produce a RH advantage; o r t h a t f a m i l i a r i t y of a stimulus w i l l lead to a LH advantage, by a l l o w i n g f o r more r e f i n e d and d e t a i l e d a n a l y s i s of higher s p a t i a l f r e q u e n c i e s . She even Page 17 speculates on the gradual ontogenetic development of a LH dominance: the more detailed visual processing of increasingly familiar material becomes, the more i t w i l l tend to be l a t e r a l i z e d i n the LH. One problem in demonstrating hemispheric s p e c i a l i z a t i o n i s to distinguish genuine cerebral asymmetries from other factors that may contribute to vi s u a l f i e l d asymmetries., Reading order i s one obvious candidate. Schwartz and Kirsner (19 82) showed that attentional effects may also play a c r u c i a l role. They were able to produce l e f t / r i g h t visual f i e l d asymmetries by varying stimulus p r o b a b i l i t y , and they showed that the same asymmetries could be observed i n v e r t i c a l l y defined vis u a l f i e l d s ( i . e . above and below the f i x a t i o n ) . They conclude that i t may often be unnecessary to invoke d i f f e r e n t i a l hemispheric s p e c i a l i z a t i o n i n order to account for visu a l f i e l d differences. Some of the dichotomies could also not be integrated into more general theories of perception and cognition. It remains unclear to what extent alternative modes of processing are l a t e r a l i z e d and at what le v e l or stage i n processing they occur. For example, i f the RH processes more i n p a r a l l e l and the LH more s e r i a l l y , i s t h i s true f o r both preattentive and attentional processes? Does attentional processing which general cognitive theories t y p i c a l l y claim to be s e r i a l , occur in p a r a l l e l in the RH, or not exist? Such questions can not be answered by these dichotomies of hemispheric functioning. Page 18 1.2.2. Capacity models: The oldest and most simple model states that wherever information has d i r e c t access to the hemisphere specialized f o r processing i t , superior and/or faster processing w i l l r e s u l t . It quickly became evident, however, that t h i s model could not account for the observed v a r i a b i l i t y i n l a t e r a l i t y e f f e c t s . More dynamic attentional models had to be developed i n order to account for fluctuations i n asymmetry. One such approach i s the selective activation hypothesis by Kinsbourne (1975). He claims that the involvement of a hemisphere i n a task w i l l result i n a maximum of attention being directed to the contralateral v i s u a l f i e l d . Any stimulus presented con t r a l a t e r a l l y to the more activated hemisphere should thus be processed more e f f i c i e n t l y than a comparable stimulus presented i p s i l a t e r a l l y . This e f f e c t w i l l be independent of, or rather w i l l overwhelm small asymmetries due to hemispheric s p e c i a l i z a t i o n . The model was subsequently revised, since i t could not account for the interference frequently found i n dual tasks. Interference effects are now incorporated into the revised model of functional cerebral distance (Kinsbourne and Hicks, 1978) , i n which f a c i l i t a t i o n e f f e c t s are conceived of as one of two possible predictions. However, as long as the theory can not predict in advance which e f f e c t , f a c i l i t a t i o n or interference, should occur i n any given task, i t w i l l retain a ce r t a i n post hoc q u a l i t y . Page 19 Another example of research showing attentional factors to be of importance was proposed by H e l l i g e , Cox and Litvac (1979) . Using tasks with a concurrent memory load of two to six words, they found that the memory load shifted a l e f t v i s u a l f i e l d (LVF(RH) ) supe r i o r i t y f o r a memory match of polygons to a r i g h t v i s u a l f i e l d (RVF(LH) ) advantage. The same load s h i f t e d a RVF(LH) s u p e r i o r i t y to the LVF(RH) for letter-name matching. As neither the d i r e c t access nor the functional cerebral distance model can explain these e f f e c t s , Hellige,Cox and Litvac (1979) suggested that the two hemispheres function as separate information processing systems to a certain degree, but that they cooperate to maximize processing e f f i c i e n c y . Thus, i f the LH is more activated than the RH by a verbal memory load, i t may be more e f f i c i e n t at visuospatial processing than the RH. If however, the LH i s overloaded by a verbal task concurrent with a verbal memory load, the RH may be more e f f i c i e n t at performing the verbal task. According to t h e i r view, hemispheric activation and hemisphere-of-presentation interact to determine the observed l a t e r a l i t y pattern. The l a s t model to be introduced here i s the multiple-resources model proposed by Friedman and Poison (1981). They suggest, that the two hemispheres comprise a system of two mutually inaccessible and f i n i t e pools of resource supplies. Furthermore, they claim that these two pools of resources cannot be made av a i l a b l e i n d i f f e r i n g amounts. If one hemisphere i s activated by a task, the same amount of Page 20 resources i s available in both hemispheres. Thus, i f two concurrent tasks draw on resources from only one hemisphere, they are l i k e l y to int e r f e r e with each other; i f each of them draws on resources from a d i f f e r e n t hemisphere, both tasks w i l l be f a c i l i t a t e d . This model makes assumptions as to how l a t e r a l i z e d a given task i s ( i . e . what resources i t w i l l draw on) and when i t w i l l reach i t s capacity l i m i t s . These assumptions can, i n some tasks, p l a u s i b l y be made i n more than cne way. Thus, the model does not always make clear predictions. Summarizing, i t seems that i n an e f f o r t to accommodate complex patterns of visual f i e l d asymmetries, more and more complicated models were developed. Unforunately, they have to rely on functional dichotomies i n order to predict or explain attentional demands a given task w i l l make on cne or the other hemisphere. Such assumptions about hemispheric functional s p e c i a l i z a t i o n and l a t e r a l i z e d task performance are controversial in themselves, however. Thus, these models have to be used very cautiously. It i s probably best to apply them only to well established l a t e r a l i z e d functions, l i k e , for example, language. Page 21 1.2.3. Functional Loci: This section introduces two hypotheses that specify a precise functional l e v e l at which hemispheric differences s t a r t to emerge. Kinsbourne (19 82) looks at hemispheric s p e c i a l i z a t i o n from an evolutionary point of view and claims that only processes pertaining to focal attention are l a t e r a l i z e d , whereas a l l preattentive processes are represented b i l a t e r a l l y . He proposes that under focal attention two processes proceed i n p a r a l l e l in the two hemispheres: a s e r i a l feature extraction i n the LH, and a concurrent registering of feature locations on a c e n t r a l l y represented feature map i n the RH. The tr ansmitted-later al ization hypothesis proposed by Moscovitch (1979) and Moscovitch and K l e i n (1980) also holds that an e a r l y sensory and pr ea t tent ion a l stage of information processing i s common to both hemispheres and that differences between them only occur at the l e v e l of a central processor beyond the i n i t i a l feature extraction. Within his framework, the LH concentrates primarily on functional and nominal aspects of the input, whereas the RH processes and encodes information cn the basis of appearance. Page 22 I I . EXPERIMENT 1: Conjunction errors In this experiment early, automatic and preattentional processes were investigated. From both Kinsbourne's (1982) and Moscovitch's (1979) theories one would predict that there should not be any hemispheric differences i n the number of conjunction errors, "since these errors are assumed to be preattentional. In terms of the s e r i a l / p a r a l l e l d i s t i n c t i o n , one would expect the more s e r i a l hemisphere, i . e . the LH, to produce more conjunction errors when attention i s overloaded and the s e r i a l integration of features impaired. Experiment l a : In a f i r s t experiment (la) colored l e t t e r s were used as s t i m u l i , with c o l o r and shape representing the two dimensions of each stimulus. A detection task with free verbal report was used. Since neither reaction times nor accuracy were analyzed and since conjunction errors are believed not to depend on verbal coding (Treisman and Schmidt, 1982), the verbal report was not thought to introduce a confound with any hemispheric differences. Page 23 Method : Subjects: 13 female and 17 male students from UBC served as subjects. Each was paid $4.00 f o r a 1-hour session. A l l subjects were righthanded as asessed by a l a t e r a l i t y questionnaire developed by Coren et a l . (1979). Apparatus and s t i m u l i : The st i m u l i consisted of a v e r t i c a l line of three colored uppercase l e t t e r s ; they were chosen from a set of f i v e possible l e t t e r s : I,N,0,S and X, and from a set of f i v e possible colors: yellow,green,pink,blue, and brown. Each l e t t e r subtended a v i s u a l angle of 1.10x1.38 deg. The whole configuration subtended a vi s u a l angle of 1.10x5.09 deg. By mistake, the l e t t e r s were moved out too far from the center, and the closest edge appeared 4.82 deg of visu a l angle to the ri g h t or l e f t of the center. In the center, two black digits,1,6,8 or 9, were presented. Each d i g i t subtended a vis u a l angle of 0.69x1.10 deg and one of them was positioned 0.82 deg of vi s u a l angle above, the other one at the same distance below the center. Each c o l o r and l e t t e r appeared equally often in each position. Each of the d i f f e r e n t c o l o r - l e t t e r combinations appeared between 3 and 5 times i n each po s i t i o n . 30 cards were made. The stimuli were drawn by hand, using colored inks and s t e n c i l s on white cards. A black and white noise mask, consisting of equal numbers of randomly arranged black or white 2-mm squares, and subtending the whole v i s u a l f i e l d was used. It had a black f i x a t i o n dot i n the center. Page 24 Procedure: Alternating which way up, the set of 30 cards was shown four times to each subject. The order of cards was randomized f o r each block and each subject. Examples of the stimulus cards with a l l possible colors, lettershapes and d i g i t s were shown to each subject before the experimental t r i a l s started. Subjects were instructed to report the two central d i g i t s f i r s t and subsequently as many of the colors, shapes and their positions as they could remember. They were tol d to be as accurate as possible on the numbers. They were asked only to report colors and shapes i f they were sure they had seen them, or else to indicate, i f they reported something they were uncertain about, and th i s was noted by the exper imenter. The stimulus cards were presented i n a Cambridge two-field tachistoscope. The experimenter gave a verbal 'ready' signal and i n i t i a t e d a t r i a l by pressing a button. Subjects f i r s t fixated on a black dot in the center of the noise mask, which also apeared again immediately after each t r i a l . The exposure duration f o r the stimulus cards was i n i t i a l l y set at 300 msec for each subject. Since they had to have their eyes focused on the center f o r the d i g i t naming task, 300 msec seemed short enough to prevent eyemovements. Exposure duration was then adjusted f o r each subject according to the following rule: i f a feature error (reporting a color o r a shape that was not on the card), or i f less than cne color and one shape was reported on two consecutive t r i a l s , exposure duration was increased by one step, but only to a maximum of 300 msec; i f Page 25 at least one color and one shape were reported on 7 successive t r i a l s , exposure duration was reduced by one step. For the f i r s t 20 subjects the steps were 300,200,15 0,130,115,100,90 and 80 msec. For the l a s t 10 subjects a new timer was introduced i n the hope of reducing the error rate. The following smaller steps could then be used: 2 0 msec steps down to an exposure duration of 150 msec, and then the same steps as for the f i r s t 20 subjects mentioned above. If subjects made a mistake on the d i g i t s , that t r i a l was discarded and rerun at least 5 t r i a l s l a t e r . A l l subjects were given 10 practice t r i a l s . On these t r i a l s feedback was given for feature errors. After the practice t r i a l s no feedback was provided. Results : The following types of responses were of interest: conjunction errors (two correct features wrongly recombined from two di f f e r e n t items) and feature errors (an incorrect feature either conjoined with a correct or an incorrect feature). Neither o v e r a l l nor for the separate v i s u a l f i e l d s were there s i g n i f i c a n t l y more conjunction than feature errors. From the t o t a l of . 30 subjects, nine showed more feature than conjunction e r r o r s . The mean number per t r i a l of feature and conjunction errors are shown i n Table I, along with other types of responses. Conjunction errors are c e r t a i n l y not meaningful i f they occur less frequently than feature errors. Page 26 They may just represent feature errors for which the misperceived color o r shape happened to be among those on the card rather than among those not presented. Only four subjects had a substantial excess of conjunction over feature errors. One of them showed no asymmetry i n t h i s measure, one had a bigger excess of conjunction over feature errors i n the LVF(RH) and the other two i n the RVP(LH) . The main question of interest, whether conjunction errors would be more l i k e l y to occur i n one v i s u a l f i e l d than the other, could thus not be answered conclusively. However, the re s u l t s suggest that there are no hemispheric differences i n t h i s mea sure. Table I: Mean number per t r i a l of d i f f e r e n t types of responses for a l l subjects (n=30) i n both vis u a l f i e l d s . LVF(RH) ! RVF(LH) | Items correct 1.61 1 !- 6 7 1 Single features correct .59 I .62 | Feature errors .11 1 -11 I Single feature wrong .04 1 • 03 | Conjunction errors .17 1 -17 | Conjunction minus feature errors .06 | .06 Page 27 Experiment l b : In this experiment objects defined by d i f f e r e n t components of shape were chosen. Subjects performed a search task with dollarsigns as target s t i m u l i was chosen. The targets were i n a background of arrows and 'S'-signs, from which an i l l u s o r y d o l l a r s i g n could be formed. In this task a simple 'yes-no' response was required. Method : Subjects: 4 female and 6 male students from UBC volunteered to serve as subjects. A l l of them were right-handed as asessed by a questionnaire on behavioral l a t e r a l preference designed by Coren e t a l . (1979) . Apparatus and s t i m u l i : The st i m u l i were t i l t e d 'S'- and '$'-signs and arrows (see Figure 1) . There were 9 stimulus items on each card, arranged at equal distances from each other i n a square of 5.03x5.03 deg of visual angle. The arrows and '$'- signs subtended a v i s u a l angle of 1.36x1.36 each, the p l a i n 'S'-signs were s l i g h t l y smaller, subtending a v i s u a l angle of 1.09x1.09 deg each. The closest edge of any item was 2.05 deg of visu a l angle to the ri g h t or l e f t of the center. 64 cards were made. The stimuli were drawn by hand using red ink and s t e n c i l s . 32 Page 28 of the cards had the st i m u l i in the RVF, the other 3 2 had them in the LVF. - On 16 of the cards from each v i s u a l f i e l d the stimuli were t i l t e d 45 deg to the r i g h t , with the arrows pointing to the upper right-hand corner; on the remaining 16 cards the st i m u l i were t i l t e d 45 deg to the l e f t , with the arrows pointing to the upper left-hand corner. 12 cards showed 9 arrows, and 12 cards showed 9 'S'-signs. 24 cards showed 4 arrows and 5 'S'-signs in what appeared to be random positions. Thus, on these cards the slash i n the arrow and the 'S'-sign could be combined to form an i l l u s o r y '$'-sign. On 16 cards there was a '$'-sign: 8 of them showed 4 'S'-signs, 4 arrows and a '$'-sign i n what appeared to be random arrangements, the '$'-sign being i n a d i f f e r e n t position cn each card. 4 cards showed 8 'S'-signs and a '$'-sign and the remaining four cards showed 8 arrows and a '$'-sign. Again, .the '$'-signs were i n d i f f e r e n t positions on each card. A black and white noise mask, consisting of equal numbers of black and white 2-mm squares and subtending the whole v i s u a l f i e l d , was used. I t had a black f i x a t i o n dot i n the center. Page 29 Figure 1: Example of a stimulus display with t i l t e d 'S'- and $-signs and arrows. Procedure: Alternating which way up, the set of 64 cards was shown four times to each subject. The order of cards was randomized f o r each block and each subject. Examples of a l l types of stimulus cards were shown to subjects and each subject was given 10 practice t r i a l s before the experimental t r i a l s started. Subjects were instructed to look f o r the '$'-signs and say 'yes' i f they saw one or otherwise say 'no'. They were to l d to indicate i f they were uncertain about a response they were giving, and this was noted by the experimenter. On the practice t r i a l s , subjects were t o l d whether the i r response was correct or not. On the experimental t r i a l s no feedback was provided. The cards were shown i n a two-field Cambridge tachistoscope. The experimenter gave a verbal 'ready' signal and i n i t i a t e d a t r i a l by pressing a button. Subjects f i r s t fixated on a black dot i n the center of the noise mask, which also appeared again Page 30 immediately after each t r i a l . Exposure duration was i n i t i a l l y set at 80 msec. If subjects were uncertain with most of their answers, exposure duration was increased to 100 msec and then, i f possible, reduced to 80 msec again after one block. Results : The doubtful category was used on 31% of the t r i a l s . It seems that the subjects were thus not a l l that sure about what they had seen. Conjunction errors i n the present experiment were defined as reporting a "$'*-sign from a display of arrows and 'S'-signs. Feature errors were defined as reporting a "$'-sign when either only arrows or only 'S'-signs were presented. A l l subjects made more conjunction than feature errors i n the present task. A 2-way ANOVA (Sex x Visual f i e l d ) was done on C-F erors. The ov e r a l l mean for conjunction minus feature errors (C-F) was the same f o r both v i s u a l f i e l d s (see Table II) and the main e f f e c t f o r vi s u a l f i e l d was c l e a r l y not s i g n i f i c a n t (F(1,8)<1). Neither was there a s i g n i f i c a n t interaction (F(l,8) = 1.422, p>.10) . Females had more C-F errors o v e r a l l than males (see Table II) . The trend of the main e f f e c t for sex r e f l e c t e d t h i s fact (F(l,8)= 3.791, p<.10). In this experiment, then, there were c l e a r l y no differences between the v i s u a l f i e l d s . The only trend found was that females tended to have more C-F errors than males. Page 31 Table I I : Mean number per t r i a l of d i f f e r e n t types of responses for both v i s u a l f i e l d s . F = females, M = males. I LVF(RH) 1 RVF(LH) | Correct targets F 1 .82 .83 M 1 .86 .85 Feature errors F 1 .03 .02 M 1 .03 . 03 Conj unction F I .15 .15 errors M j .11 .10 Conjunction minus F 1 .12 .13 feature errors M ] .08 .07 Discussion : In the present search task, no vi s u a l f i e l d differences were found f o r conjunction minus feature errors. A trend was found ind i c a t i n g that females tend to show more conjunction errors than males. There were c l e a r l y not enough subjects of either sex, however, to be sure that t h i s i s not a random e f f e c t . The results from both i l l u s o r y conjunction experiments are taken as evidence that there are no differences between v i s u a l f i e l d s on the measure of conjunction e r r o r s . In terms of the s e r i a l / p a r a l l e l d i s t i n c t i o n i t had been hypothesized that the LH should show more conjunction errors than the RH. This was not the case, however. Rather, the above experiments lend support to the transmitted-l a t e r a l i z a t i o n hypothesis (Moscovitch, 1979) and to the hypothesis proposed by Kinsbourne (1982) , which claim that early sensory and prea t tent ion a l processes are represented Page 3 2 b i l a t e r a l l y . Conjunction errors supposedly occur at a very early, perceptual l e v e l p r i o r to object and event i d e n t i f i c a t i o n (Treisman and Schmidt, 1982), and i t seems that at t h i s l e v e l there exist no hemispheric differences. This conclusion means accepting the null hypothesis, however, and should have some further experimental support. Even though conjunction errors are not believed to depend on verbal report, the re s u l t s should be confirmed i n an experiment using, for example, a matching task and a manual response. Subjects could then be assigned two keys to each hand which would preserve task-hemispheric i n t e g r i t y (Wickens, Mountford and Schreiner, 1981). The results should further be corroborated by using tasks with dif f e r e n t features or d i f f e r e n t dimensions to specify the multidimensional objects. For example, geometric shapes ( c i r c l e s , squares etc.) instead of l e t t e r s would eliminate the p o s s i b i l i t y of a confound with 'verbal' s t i m u l i . As additional dimensions size or s o l i d i t y (Treisman and Schmidt, 1982) could be used. Since the RH i s often viewed as the hemisphere more concerned with s p a t i a l relations than the LH (e.g. Kinsbourne, 1982, Moscovitch, 1979), a conjunction experiment involving shapes of d i f f e r e n t sizes would y i e l d evidence as to whether the RH would be le s s l i k e l y than the LH to switch the 'size-features'. The feature-integration theory of attention (Treisman and Schmidt, 1982) also allows for top-down constraints on conjunction formation. So far these have not been Page 33 experimentally demonstrated. If such constraints can be shown to e x i s t , however, i t would be interesting to show that early perceptual stages are similar in the two hemispheres not only without top-down constraints, but also under the general constraints of the perception of the everyday environment. To summarize the two experiments, i t was found that the hemispheres do not d i f f e r i n an e a r l y , automatic and pr eat tent ion a l stage of perception. Page 34 I I I . EXPERIMENT 2: A priming task In t h i s experiment i t was investigated whether the finding from experiment 1 that there are no hemispheric differences i n an e a r l y preattentional stage would extend to automatic processes occuring later on in information processing. In addition, e f f e c t s of attentional selection were introduced. The experiment used a l e x i c a l decision task, with a target word or ncnword in lowercase l e t t e r s presented i n either the r i g h t or l e f t v i s u a l f i e l d . On some t r i a l s a neutral or a semantically associated word in uppercase was shown simultaneously with the target. Thus, the two possible kinds of selection could be investigated. Neutral words were expected to i n t e r f e r e with the selection of the target ( f i l t e r i n g ) . This interference e f f e c t could be determined by comparing the target plus neutral word to a single target condition. Primewords were expected to reduce some of t h i s interference (automatic priming). The target plus prime condition was thus expected to show f a c i l i t a t i o n when compared to the target plus neutral word condition. The spatial arrangements were such that target and ncntarget could be presented in the same (LH-S and RH-S) or in opposite (LH-0 and RH-O) hemispheres. Thus, interference and priming effects produced by i p s i - or c o n t r a l a t e r a l nontargets could be compared. For i p s i l a t e r a l arrangements (see Figure 2) the Page 35 s e r i a l / p a r a l l e l d i s t i n c t i o n (Cohen, i y / 3 ) applies. One might expect the RH to be able to process the two simultaneous words i n p a r a l l e l , whereas the LH might process them s e r i a l l y and select only the target. Thus, for i p s i l a t e r a l arrangements, the LH could i n i t i a l l y , at the perceptual stage, be expected to show more interference than the RH, since i t has to select the target from the two presented s t i m u l i . The amount of interference should depend on how simple a physical property the decision can be based on. Since the target was written in lowercase and the ncntarget in uppercase l e t t e r s , the r e s u l t i n g difference in size between the words provided quite a s a l i e n t physical cue. Thus, these early interference e f f e c t s i n the LH need not be substantial. At a later stage of response selection, the RH can be expected to show more interference than the LH. If i t has processed the two words in p a r a l l e l , i t then has to select one of two processed words for a response. Late selection (as i n the RH) i s assumed to cause more interference than e a r l y selection (as i n the LH) . For the same reasons, namely that the prime i s processed to a higher degree i n the RH, t h i s hemisphere should also show more priming than the LH, which might not process the prime at a l l or only after having processed the target. Capacity models lead to d i f f e r e n t predictions for these i p s i l a t e r a l arrangements. From a d i r e c t access approach, which implies that a l l words are transferred to the LH for processing, two predictions could be made: i f c r o s s - c a l l o s a l Page 36 transfer of two words i s less e f f i c i e n t than transfer of just one word, then the RH should show more interference and less priming than the LH. If, however, c r o s s - c a l l o s a l transfer f o r two words i s as e f f i c i e n t as for a single word, both hemispheres should show equal amounts of interference and priming. The multiple resources model (Friedman and Poison, 19 81) would imply equal amounts of interference and priming i n both hemispheres, i f there are no capacity l i m i t s to t h i s task. If the task does reach capacity l i m i t s , however, i t would do so f i r s t in the less language specialized RH. The RH should then show more interference and less priming. For i p s i l a t e r a l arrangements, then, more interference i n the RH than in the LH can be accomodated by a l l of the above models. Two of them, the d i r e c t access approach and the multiple resources model (Friedman and Poison, 1981), can also explain equal amounts of interference i n both hemispheres. For priming e f f e c t s the models lead to contradictory predictions (see Figure 2) . For contralateral arrangements (see Figure 2) the cooperation of the hemispheres comes into play as a new variable. One can expect the LH to display a general dominance over the RH and to show a cert a i n degree of control over response sele c t i o n . Zaidel (1978b), for example, has suggested that the LH may be dominant in response selection even f o r tasks i t i s not specialized f o r . He showed that more interference a r i s e s i f a target stimulus i s presented to the RH and a response selected Page 37 by the LH than vice versa, which indicates a kind of ~ non-cooper at i v i t y ' of the LH. Another phenomenon that may be interpreted along the same line i s the l e f t - s i d e neglect syndrome re s u l t i n g from RH damage. The LH tends to attend to s t i m u l i on the r i g h t side only and ignore the l e f t side. The reverse i s true much less frequently for the RH (Heilman and Watson, 1977), however, since the RH monitors whether the LH has received information or not (Geschwind, 19 81) . Dominance of the LH should be p a r t i c u l a r l y pronounced i n a verbal task that the LH is more specialized f o r . Thus, i n the present experiment, interference and priming should be high for the RH-0 arrangement, since the RH monitors information i n the LH. In the LH-0 arrangement, however, l i t t l e interference and priming should be found, since the LH w i l l tend to ignore any information in the RH. The d i r e c t access approach also applies f o r the c o n t r a l a t e r a l arrangements. According to t h i s model, a target in the LH i s l i t t l e affected by a ncntarget i n the RH (LH-O), since the ncntarget, being transferred from the RH, w i l l reach the LH after the target. For the opposite arrangement (RH-O), however, substantial interference and priming should be found, since the ncntarget has d i r e c t access to the LH and i s processed p r i o r to the target. The multiple resources model (Friedman and Poison, iy8I) implies that the same amount of resources required to process the target i n cne hemisphere i s also available in the opposite hemisphere. Thus, processing Page 38 of the ncntarget should not draw resources away from the target. Less interference i s then expected i n these contralateral arrangements than in the i p s i l a t e r a l arrangements. The s e r i a l / p a r a l l e l d i s t i n c t i o n does not make any predictions for these contralateral arrangements. Page 3 9 Figure 2: Predictions for priming and interference effects derived from d i f f e r e n t hemispheric models. LH = target to LH, RH = target to RH, S = ncntarget to the same, 0 = ncntarget to the opposite hemisphere than target, I = interference e f f e c t s , P = priming e f f e c t s . S e r i a l / p a r a l l e l d i s t i n c t i o n P R H S LH-S RH-O , L H O 1 1 • 1 1 D i r e c t access a p p r o a c h R H S LH-S R H - O L H O — J 1 . i i 1 M u l t i p l e r e s o u r c e s m o d e l R H S LH-S RH-O L H - O 1 • • • L H dom i n a nee 1 P RH S LH-S RH-O L H - O - 1 ' i Page 40 Method: Subjects: 15 female and 15 male students from UBC volunteered to take part in the experiment. They were a l l right-handed, as assessed by a questionnaire (Coren et. a l . , 1979) and they were paid $ 4.00 for a 1-hour session. Apparatus and s t i m u l i : The stimuli were words, nonwords (scrambled l e t t e r s ) or blanks. On each t r i a l , a target that could either be a word (T) or a nonword (t) was shown simultaneously with a prime (P) (semantically associated word), a neutral (N) (semantically not associated) word or a blank (BI). Thus, there were the following 5 conditions that a l l appeared equally often: T+P, T+N, T+Bl, t+N and t+Bl. A l l targets were written in lowercase and a l l nontargets i n uppercase l e t t e r s . There were four s p a t i a l positions on the screen: Top r i g h t , top l e f t , bottom right and bottom l e f t . Excluding diagonal placement, a l l s t i m u l i could appear in a l l s p a t i a l positions. Ignoring differences between top and bottom rows, which were counterbalanced, 8 d i f f e r e n t possible s p a t i a l arrangements resulted for each condition. The target could be presented to the RH or the LH and the cue to the same (S) or opposite (O) hemisphere. This yielded the following arrangements of interest: RH-S, LH-S, RH-0 and LH-O. Target and prime words were taken from 'An a t l a s of normative free association data' by Shapiro and Palermo (1968). Their primary responses were chosen as target words, whereas their Page 41 target s t i m u l i were taken as prime words. Keeping word frequency and associative strength as high as possible, 320 word pairs were chosen. Each word was between 3 and 7 l e t t e r s long. The ncnwords were made by randomly scrambling a l l the l e t t e r s of each target word except the f i r s t one. Thus, the nonwords were easy to distinguish from the words, but the f i r s t l e t t e r s did not carry any information. One neutral non-associated word for each target word was selected from the same pool of words, approximately matched i n length and frequency. Each l e t t e r subtended a v i s u a l angle of about 0.71 x 1.33 deg. A 7-letter word subtended a v i s u a l angle of 6.95 x 1.33 deg. Since the task was very d i f f i c u l t to do, the stimuli were moved i n as close as possible to the center. Thus, the closest edge of any word appeared 1.78 deg of visual angle from the central f i x a t i o n point. This was considered peripheral enough since the f i r s t l e t t e r of each stimulus was not e s p e c i a l l y c r u c i a l and the second l e t t e r was already 2.82 deg of visua l angle from the center. The centers of the words on top and on the bottom were 3.33 deg of visua l angle apart. A pattern mask, made up of l e t t e r fragments, appeared i n a l l four s p a t i a l positions and subtended a v i s u a l angle of four times 7.07 x 1.78 deg. Subjects had four response keys, two for the r i g h t hand and sti m u l i in the RVF, and two for the l e f t hand and st i m u l i i n the LVF. This arrangement maintains what Wickens, Mountford and Schreiner (1981) have termed 'task hemisphere i n t e g r i t y ' , Page 4 2 i . e . processing and response occur in the same hemisphere. The displays were shown en a VT-11 graphic display processor, under the control of a PDP-11/34 computer (Dig i t a l equipment corporation). The stimuli were white on a dark background. A head-rest was used to ensure a constant viewing distance of 64 cm. Procedure: The 320 stimulus sets of 5 items each (T, t, N, P and BI) were always presented i n the same order. For each t r i a l a target and one of the other three items ( i . e . one of the 5 conditions) were chosen. The condition selected varied in each block and for each subject. Each subject had six blocks with 320 t r i a l s each. Each condition appeared equally often i n each of the 8 s p a t i a l arrangements. Within the above constraints, conditions and s p a t i a l arrangements were completely randomized for each subject and each block. Before each block, the message 'Press any key when ready' was shown and subjects started the t r i a l s themselves. Subjects were instructed to do a l e x i c a l decision task on the lowercase words. Accuracy was emphasized rather than speed, and error rates were the dependent variable. A l l subjects used the inside key of each hand to indicate words and the outside key to indicate non wo rds. Each t r i a l started with a central fix a t i o n dot for 900 msecs, followed by the stimuli for 250 msecs. Immediately afterwards the pattern mask appeared i n a l l four s p a t i a l positions. It went off again as soon as the subjects gave a response and the next t r i a l began. Page 43 Results: The o v e r a l l error rate i n the double word conditions was 32%, with a range of 17 - 39%. Subjects with an e r r o r rate higher than 40% were discarded, since an e r r o r rate of 50% r e f l e c t e d chance responding. The percent correct and percent false positives errors were used to calculate d's i n each of the three target word conditions. These are given i n Table III. Interference produced by showing two compared to only a single word was calculated by subtracting the d's of T+N from those of T+BL. Priming was calculated by subtracting the d's of T+N from those of T+P. These differences are also shown in Table III . A 2-way ANOVA (Sex x Visual f i e l d ) was done for the single target word condition. Performance was superior for the RVF(LH) (mean d": 1.72 for the LVF(RH) and 1.92 for the RVF(LH)), and the main e f f e c t for v i s u a l f i e l d was s i g n i f i c a n t (F(l,28) = 6.451, p<.02). Seven females and 11 males were more accurate i n the RVF(LH) than in the LVF(RH) . The opposite was true for 8 females and 4 males. There was neither a main e f f e c t for sex nor an interaction (both Fs(l,28)< 3, p>.10). The analyses for interference and priming i n the double word conditions examined the e f f e c t of target position (RH-S, RH-O versus LH-S, LH-O), and the e f f e c t of prime position (RH-S, LH-S versus RH-S, LH-O). Sex was used as an additional Page 44 factor. . Thus, a 3-way ANOVA (Sex x Target position x Prime position) yielded the following results (see Table I I I ) . Interference e f f e c t s : Showing a neutral word together with the target word caused more interference for the RH (RH-S and RH-O) than for the LH (LH-S and LH-O), and the main e f f e c t for target position was s i g n i f i c a n t (F(l,28) = 5.651, p<.05). This e f f e c t was mainly due to high interference i n the RH-O, and very low interference in the LH-0 arrangement. The means for RH-S and LH-S were almost i d e n t i c a l . RH-O showed more interference than RH-S, whereas the reverse was true f o r the LH, and the interaction targetposition x primeposition was s i g n i f i c a n t (F(l,28) = 6.510, p<.05). A Bonferroni t-test on the four means showed that three differences between means were s i g n i f i c a n t : RH-O, RH-S and LH-S were higher than LH-0 ( a l l ts(28) > 2.70, p<.01). There was neither a main e f f e c t for sex nor any interactions with sex ( a l l Fs(l,28)< 2, p>.10). Priming e f f e c t s : There were no main e f f e c t s or interactions for priming ( a l l Fs(l,28)< 3, p>.10) . t-tests showed that priming was highly s i g n i f i c a n t in the RH-O and LH-0 arrangements (both ts(29)> 3.0, p<.005), only just reached significance in the RH-S arrangement (t(29) = 1.889, p<.05) and was not s i g n i f i c a n t in the LH-S arrangement (t(29) = 1.657, p<.10). Substantial priming was. thus only found i n the RH-O and LH-0 arrangements. Page 45 Table III : The d's for a l l three target word conditions as well as priming and interference e f f e c t s i n a l l s p a t i a l arrangements. RH = target to RH, LH = target to LH, S = ncntarget on same side, 0 = nontarget on opposite side than target. RH-S LH-S RH-O LH-0 | Target plus prime .96 1.18 .96 | 1.52 | Target plus neutral word .84 1.08 .76 | 1.32 | Single target 1.72 1.92 1.72 1.92 | Interference .87 .83 .96 .60 | Priming .12 .10 .21 .20 | Discussion: Overall, a clear RVF(LH) s u p e r i o r i t y was found for the single target condition. This i s consistent with r e s u l t s , for example, by Bradshaw and Gates (1978) and Day (1977). The finding that 8 females and 4 males (n=30) showed a LVF(RH) advantage i s not surprising. In f a c t , Bradshaw and Gates (1978) concluded from a series of experiments that 'a RH verbal mechanism, which i s more strongly developed i n females than in males, i s associated with l e x i c a l decisions, i f both phonological and graphological c r i t e r i a may apply' (as i s the case with non-pronouncable ncnwords used i n t h i s experiment). Si m i l a r l y , Day (1977) found that an overall RVF(LH) advantage was reversed f o r some female subjects. So the r e s u l t s support the generally held b e l i e f s that language i s t y p i c a l l y l a t e r a l i z e d i n the LH and that t h i s Page 46 l a t e r a l i z a t i o n i s less pronounced i n females than in males Ox (see McGlone , 1980). The two effects investigated i n the double word conditions were interference and priming. Interference effects were substantial and larger than the priming e f f e c t s . Generally, the following picture seems to emerge from the r e s u l t s . For i p s i l a t e r a l arrangements interference and priming were the.same fo r both hemispheres. Thus, target selection probably takes place at the same stage in both hemispheres. In addition, influence from the automatic stage on the subsequent a t t e n t i o n a l l y controlled stage (the priming effect) i s the same in both hemispheres. So the finding from experiment 1 ( i l l u s o r y conjunctions) that there are no hemispheric differences i n e a r l y prea ttentional processes seems to extend to automatic processes involving higher l e v e l s of information processing. For c o n t r a l a t e r a l arrangements, i t seems that the LH represents a processing system that i s r e l a t i v e l y independent of the RH. Interference i s lowest with a ncntarget in the RH. The LH i s thus not interfered with much by verbal processing occuring i n the RH. Since priming was found i n the LH-O arrangement, however, there must be a transfer of l e x i c a l or semantic codes from the RH to the LH. This i s an example of benefit without substantial cost, which according to Posner and Snyder (1975) characterizes automatic processing. The res u l t s suggest that target selection i n the LH i s based on information that was o r i g i n a l l y presented to that hemisphere. Page 47 If evidence from the RH happens to be consistent with the semantic information i n the LH, i t may produce f a c i l i t a t i o n ; however, i f evidence from the RH i s unrelated to the information i n the LH, i t produces less f i l t e r i n g cost than in any other arrangment. The RH on the other hand i s influenced by a contralateral ncntarget at least as much as by an i p s i l a t e r a l ncntarget. It seems that the RH can not process verbal information independently of any processing occuring i n the LH. A ncntarget i n the LH seems to a t t r a c t attention away from a RH target, and produce both cost when unrelated and benefit when associated. Thus, i t seems that attention can be better focussed on verbal information presented to the LH and not be attracted away by verbal information in the RH. In other words, the RH appears to be able to give some automatic support to verbal processing i n the LH, while the reverse seems more d i f f i c u l t . With respect to the predictions derived from d i f f e r e n t models of hemispheric functioning, i p s i l a t e r a l arrangements w i l l again be discussed f i r s t . The predictions made f o r interference i n the framework of the s e r i a l / p a r a l l e l d i s t i n c t i o n were not precise. If one assumes that the amount of interference i s similar for early target selection (as i n the LH) or l a t e target selection (as i n the RH) , one could accommodate the absence of l a t e r a l differences i n interference. The predicted difference i n priming (more priming i n the RH than in the LH) was not found, however. Page 48 Thus, i t is reasonable to conclude that the s e r i a l / p a r a l l e l d i s t i n c t i o n i s not supported by the r e s u l t s . One of the possible predictions f o r i p s i l a t e r a l ncntargets derived from the multiple resources model (Friedman and Poison, 1981), i . e . equal amounts of priming and interference in both hemispheres, holds i f one i s w i l l i n g to assume that there are no capacity l i m i t s to t h i s task. Corss-callosal transfer of information, as suggested by a d i r e c t access approach, could produce the i p s i l a t e r a l interference r e s u l t s that were found i f c r o s s - c a l l o s a l transfer of two words i s assumed to be as e f f i c i e n t as for only a single word. Under the same assumption, t h i s model i s also compatible with the r e s u l t s found f o r priming. The pattern of results i n the contralateral arrangements was predicted from the d i r e c t access approach and from the notion of LH dominance. The multiple resources model (Friedman and Poison, 1981), however, was not supported. A contralateral arrangement only f a c i l i t a t e d processing of the target (as compared to i p s i l a t e r a l arrangements) i n the LH. Priming was the same fo r RH-O and LH-O. The fact that LH-O showed as much priming as RH-O i s surprising and can not be accomodated by any of the models so far discussed. Thus, except for priming e f f e c t s found f o r the LH-O arrangement, the contralateral results support the d i r e c t access approach and the notion of LH dominance. To conclude the discussion of hemispheric models, i t seems Page 49 that i f c r o s s - c a l l o s a l transfer of two words i s as e f f i c i e n t as for a single word, the d i r e c t access approach can accommodate a l l but one (priming i n LH-0) of the e f f e c t s found i n the present experiment. The notion of LH dominance only applies to c o n t r a l a t e r a l arrangements and i s supported by the r e s u l t s . The multiple resources model (Friedman and Poison, 1981) i n conjunction with an assumption about capacity l i m i t s was consistent with the i p s i l a t e r a l results only. It can not accommodate the effects found i n the contralateral arrangements. The s e r i a l / p a r a l l e l d i s t i n c t i o n could not accomodate the r e s u l t s , however. Summarizing, the r e s u l t s suggest that automatic processing i s shared and common to both hemispheres. Any f a c i l i t a t i v e e f f e c t s produced by automatic processes seem to be similar both within and across hemispheres.. Lateral asymmetries a r i s e , however, at an a t t e n t i o n a l l y controlled stage of processing. The data show that interference effects a r i s i n g from attentional selection are similar in both hemispheres for i p s i l a t e r a l target and ncntarget arrangements. In co n t r a l a t e r a l arrangements> however, the RH suffers much more interference than the LH. It seems that attention can be better focussed on verbal material in the LH than i n the RH. Page 5 0 IV. EXPERIMENT 3 : A Stroop task This experiment was undertaken in order to further compare hemispheric differences i n automatic and s t r a t e g i c e f f e c t s . The experiment involved three d i f f e r e n t conditions. In cne condition consistent and inconsistent t r i a l s appeared equally often. This condition represented the t y p i c a l situation f o r a Stroop-task. Thus, the unreported dimension yields inconsistent information about the reported dimension. Unless f a c i l i t a t i o n cn consistent t r i a l s greatly outweighs interference on inconsistent t r i a l s or vice versa, the best strategy i s to ignore the unreported dimension and s e l e c t i v e l y pay attention to the reported dimension cnly. The Stroop-effect found under such circumstances i s considered to be an automatic and strate.gy-independent e f f e c t . In two further conditions the r e l a t i v e frequency of consistent and inconsistent t r i a l s was varied. In these conditions subjects were expected to use a strategy of dividing t h e i r attention between the reported and the unreported dimensions. Thus, e f f e c t s due to weights assigned to the unreported dimension were expected to be found. Page 51 Method: Subjects: 12 female and 12 male student volunteers from UBC served as subjects. They were a l l right-handed as assessed by a questionnaire (Coren et a l . , 1979). They were paid S 4.00 for a 1-hour session. Apparatus and s t i m u l i : The st i m u l i were the word 'ABOVE' and the word 'BELOW, written i n c a p i t a l l e t t e r s and appearing either above or below a cross (a lowercase 'x'). The whole configuration (word plus cross) appeared either i n the RVF or in the LVF, thus making four possible s p a t i a l positions for the word and two for the cross. Each l e t t e r subtended a v i s u a l angle of .71x1.33 deg. The cross was s l i g h t l y smaller and subtended a vi s u a l angle of .71x.88 deg. Each word subtended a vi s u a l angle of 4.87 deg ho r i z o n t a l l y and 1.33 deg v e r t i c a l l y . The closest edge of any word appeared 2.66 deg of vi s u a l angle to the ri g h t or l e f t of the center. The cross was 5.31 deg of visu a l angle to the r i g h t or l e f t of a central f i x a t i o n point and appeared 2.22 deg of visual angle above or below the middle l e t t e r of the wo rds. A pattern mask made up of l e t t e r fragments appeared i n a l l four possible s p a t i a l positions and subtended a vi s u a l angle of four times 7.07x1.78 deg. The closest edge of the mask was 1.78 deg to the right or l e f t and 2.66 deg of visu a l angle above or below the f i x a t i o n point. There were 6 blocks of 320 t r i a l s each. Each of 3 d i f f e r e n t Page 5 2 conditions was run for two consecutive blocks. In condition 1 (50/50) 50% of the t r i a l s were consistent ('ABOVE' above and 'BELOW' below the cross) and 5 0% of the t r i a l s were inconsistent ('ABOVE' below and 'BELOW' above the cross). In condition 2 (80/20) 80% of the t r i a l s were consistent and 20% inconsistent. In condition 3 (20/80) 20% of the t r i a l s were consistent and 80% inconsistent. The two words and v i s u a l f i e l d s were completely balanced within subjects. Within the above constraints, the order of t r i a l s was randomized for each subject. The order of conditions was counterbalanced between subjects. As in experiment 2, subjects had four response keys, two for the r i g h t hand and sti m u l i in the RVF and two for the l e f t hand and stimuli in the LVF. The stimuli were shown on a VT-11 graphic display processor, under the control of a PDP-11/34 computer (D i g i t a l equipment corporation). They were white on a dark background. A head-rest ensured a constant viewing distance of 64 cm. Procedure: The subjects' task was to respond to the words' identity, independent of their s p a t i a l p o s i tion. Half the subjects used the inside keys of each hand to indicate 'ABOVE' and the outside keys to indicate 'BELOW', the other half did the opposite. Before each two blocks, they were to l d which condition would be presented and i t was stressed that they should t r y to avoid a l l errors. Before each block, the message 'Press any key when ready' appeared and subjects started the t r i a l s themselves. Each t r i a l started with a Page 53 central f i x a t i o n dot for 900 msecs, followed by the word-cross configuration for 250 msecs. Immediately afterwards the pattern mask came on i n a l l four s p a t i a l positions and stayed on for 1500 msecs or u n t i l the subject gave a response (whichever was longer), and then the next t r i a l began. The f i r s t 20 t r i a l s of each condition were discarded as practice t r i a l s . Results: Condition 1 (50/50) was analyzed separately from conditions 2 (80/20) and 3 (20/80) . For condition 1 (5 0/50) a 3-way ANOVA (Sex x Visual f i e l d x T r i a l type) was done on the reaction times (RTs) for correct responses (see Figure 3) . Subjects showed a Stroop-effect of 17 msec (inconsistent minus consistent t r i a l s ) and the main e f f e c t for t r i a l t y p e (Co/Inco) was highly s i g n i f i c a n t (F(l,22) = 9.868, p<.01). The main e f f e c t for v i s u a l f i e l d was also s i g n i f i c a n t (F(l,22) = 4.690, p<.05), r e f l e c t i n g the fact that o v e r a l l RTs i n the RVF(LH) were faster by 17 msec than those i n the LVF(RH). There was no main e f f e c t for sex (F(1,22)<1) and no s i g n i f i c a n t interactions ( a l l Fs (1,22) <2. 7, p>.10) . Since error rates were f a i r l y high, the same analysis was also done on these (see Figure 3). A l l the means went in the same direction as the RTs, but there were no s i g n i f i c a n t main e f f e c t s or interactions. It i s thus j u s t i f i e d to say that Page 54 there were no error trade-offs in condition 1 (50/50) and for th i s condition only the re s u l t s for RTs w i l l be discussed. Figure 3: Mean reaction times (RTs) for correct responses a n d error rates for consistent (CO) and incon-sistent (INCO) t r i a l s in condition 1 (5 0/5 0) for both v i s u a l f i e l d s . RTs for correct responses and error rates for the two extreme conditions are plotted i n Figure 4. The results for these two conditions b a s i c a l l y r e p l i c a t e Logan and Zbrodoff's (1979) findings. When inconsistent t r i a l s were r e l a t i v e l y rare, RTs Page 55 to inconsistent s t i m u l i were slower (by 97 msec for the LVF(RH) and by 86 msec for the RVF(LH)) than those to consistent s t i m u l i . When inconsistent t r i a l s were r e l a t i v e l y frequent, however, the opposite was true: RTs to inconsistent t r i a l s were faster (by 28 msec for the LVF(RH) and by 29 msec for the RVF(LH)) than those to consistent t r i a l s . Thus, compared to condition 1 (50/50) a much enhanced Stroop-ef feet was found i n condition 2 (80/20), whereas i n condition 3 (20/80) i t was reversed. A 4-way ANOVA (Sex x Condition x Visual f i e l d x Trialtype) was done on RTs for correct responses. Inconsistent t r i a l s were slower o v e r a l l than consistent ones and the main e f f e c t for t r i a l type was s i g n i f i c a n t (F(l,22) = 13.723, p<.001). The interaction for condition x triaLtype(Co/Inco) was also highly s i g n i f i c a n t , r e f l e c t i n g the fact that in condition 2 (80/20) there was the usual Stroop-effeet, whereas i n condition 3 (20/80) the Stroop-effeet was reversed ( i . e . inconsistent t r i a l s were faster than consistent ones). The difference in Stroop-ef feet between conditions 2 (80/20) and 3 (20/80) w i l l subsequently be referred to as the "strategy e f f e c t " . Again, l i k e in condition 1 (50/50) , the RVF(LH) showed o v e r a l l faster RTs than the LVF(RH) (by 35 msec) and the main e f f e c t for v i s u a l f i e l d was s i g n i f i c a n t (F(l,22) = 20.648, p<.001). Page 56 Figure 4: Mean reaction times (RTs) for correct responses and error rates for consistent (CO) and inconsistent (INCO) t r i a l s for both visual f i e l d s i n the two extreme conditions. The question of interest was, however, whether the pattern of Stroop-effects across conditions (enhanced Stroop-effect in condition 2 (80/20) and reversal in condition 3 (20/80)) would be d i f f e r e n t for the two v i s u a l f i e l d s . This was not the case and the interaction between the factors condition x visual-f i e l d x t r i a l type (Co/Inco) was not s i g n i f i c a n t (F(1,22)<1) (see Figure 5). It seems, however, that differences i n strategy effects Page 57 between v i s u a l f i e l d s tended to go i n opposite directions for females and males, and the interaction between a l l four factors reflected t h i s trend (F(l,22) = 3.742, p<.10). The individual data showed that o v e r a l l 9 females had a bigger strategy e f f e c t i n the RH than in the LH, wheras the reverse was true f o r 7 males. 1 female showed no asymmetry. Figure 5: Stroop-effect (inconsistent-consistent) for both visu a l f i e l d s and the two extreme conditions. C o n d i t i o n Page 58 The same analyses were done on e r r o r rates again (see Figure 4). The interaction condition x t r i a l t y p e (Co/Inco) was highly s i g n i f i c a n t l i k e for the RTs (F(l,22) 19. 323 , p<.001), r e f l e c t i n g the reversal of the Stroop-effeet i n condition 3 (20/80). No other s i g n i f i c a n t e f f e cts were found i n the error rates, although the means went in the same direction as the means for RTs: the error rates were higher for inonsistent than for consistent t r i a l s , and also higher for the LVF(RH) than the RVF(LH) . Thus, l i k e i n condition 1 (50/50), no error trade-offs were found i n these two conditions and only r e s u l t s for RTs w i l l be discussed. The effects of weights were calculated for both vis u a l f i e l d s and both sexes (see Figure 6). In condition 2 (80/20) attentional and automatic weights have the same sign. RTs to frequent consistent and infrequent inconsistent t r i a l s , then, r e f l e c t the summed effects of attentional and automatic weights with the same sign. For the frequent consistent t r i a l s the weights s h i f t the decision threshold i n the correct d i r e c t i o n . The opposite i s true for infrequent inconsistent t r i a l s , however. So the differences between infrequent inconsistent and frequent consistent t r i a l s r e f l e c t the effects of two positive attentional and two positive automatic weights. In condition 3 (20/80) attentional and automatic weights have opposite signs and the differences between infrequent consistent and frequent inconsistent t r i a l s r e f l e c t the effects of two positive attentional and two negative automatic weights. The effects of attentional weights can Page 59 thus be estimated as: 1/4((RT(I,In) - RT(F,Co)) + (RT(I,Co) - RT(F,In))/ and the effects of automatic weights as: l/4((RT(I fIn) - RT(F,Co)) - (RT(I,Co) - RT (F, I n) ) , where I = infrequent, F = frequent, Co = consistent, In = inconsistent t r i a l s (see Logan and Zbrodoff, 1979, Logan, 19 80) . The effects of the weights were analyzed with a 3-way ANOVA (Sex x Weight type x Visual f i e l d ) . As expected, the eff e c t s of attentional weights were larger than those of automatic weights, and the main e f f e c t for weight type was s i g n i f i c a n t (F(l,22) = 9.590, p<.01). For the females, the e f f e c t s of attentional weights tended to be bigger in the LVF(RH) than i n the RVF(LH), whereas the opposite was true f o r the males. This was reflected i n a trend f o r the interaction between a l l three factors (F(l,22) = 4.152, p<.06). Like for the RTs, the asymmetry was more consistent for females than f o r males: 9 females had bigger attentional weights i n the RH than in the LH, the reverse was true for 7 males. 1 female showed no asymmetry. Page 6 0 Figure 6: E f f e c t s of attentional and automatic weights for both vis u a l f i e l d s . Page 61 Discussion : Condition 1 (50/5 0) was analyzed separately from the two extreme conditions, since i n t h i s condition cnly automatic ef f e c t s were expected to occur. Subjects showed a substantial Stroop-ef feet, but there were no ef f e c t s for sex or v i s u a l f i e l d . Like i n the priming task, no hemispheric differences for an automatic e f f e c t were found. These re s u l t s are i n agreement with findings in a colored word naming task by Schmit and Davis (1974) and by Warren and Marsh (1978) . Along the l i n e s of a d i r e c t access approach, one could argue that i n spite of the response key arrangement which preserves task-hemispheric i n t e g r i t y (Wickens, Mountford and Schreiner , 19 81) , the words are always processed i n the LH and never in the RH. This could be what the slower RTs of the RH r e f l e c t , and t h i s i s how.the above mentioned authors argue. Since there i s no reason why the RH should also transfer any p o s i t i o n a l information ( i . e . the unreported dimension) to the LH in the present task, the LH would be predicted to show a bigger Stroop-effeet. This was not the case, however. With colored words as used by Schmit and Davis (1974) and Warren and Marsh (1978) , the assumption that the word i s transferred as a colored word, which would abolish a l l hemispheric differences i n the Stroop-ef feet, may be a more plausible one. In the present context the assumption that the hemispheres performed the task separately from each other i s preferred. Page 62 It seems, then, that there are no hemispheric differences i n the extent to which responses to the words (the reported dimension) are influenced by the automatic processing of position information (the unreported dimension) i n the present Stroop-task. The results from condition 2 (80/20) and condition 3 (20/80) showed that subjects used a strategy of dividing their attention between the unreported and the reported dimension. There were no e f f e c t s of vi s u a l f i e l d on t h i s strategic a l l o c a t i o n of attention. A trend was found i n the data, however, which suggested that for the females, the st r a t e g i c e f f e c t s were bigger for the LVF(RH) than for the RVF(LH), whereas the opposite was true f o r the males. This was also reflected i n a trend of the effects of the weights: the eff e c t s of attentional weights were bigger in the LVF(RH) than in the RVF(LH) for the females, whereas the opposite was true for the males. A consistent asymmetry of strategy effects was r e a l l y only evident i n the females, however. Thus, i t seems that in the present task the RH in females uses an optimized strategy more e f f i c i e n t l y than the LH and may be able to attach larger attentional weights to evidence from automatic processes than the LH. Since automatic weights are assumed to be constant in size and sign, the ef f e c t s of the automatic weights calculated f o r the two extreme conditions may te n t a t i v e l y be applied to condition 1 (50/50) . In t h i s condition the difference between inconsistent and consistent t r i a l s should r e f l e c t the e f f e t s Page 63 of two automatic weights. This was true f o r the males (see Figures 3 and 6). For the females, however, the Stroop-effect was smaller than would have been expected from the automatic weights. Females may thus have attached 'negative' attentional weights to the unreported dimension, similar to condition 3 (20/80) , which would reduce the Stroop-ef feet. This conclusion i s only very speculative, however. Page 64 V. GENERAL DISCUSSION At the beginning of t h i s study a framework f o r the interaction of automatic and attentionally controlled processes was outlined. Attention was described as operating on evidence from automatic processes. It was indicated that c e r t a i n aspects of the attentional operation, i . e . integration, selection or weighting of evidence from automatic processes, may be possible sources of hemispheric differences. Three experiments that imply d i f f e r e n t functions of attention and d i f f e r e n t l e v e l s where automatic and attentional processes interac t , were run i n order to f i n d evidence f o r two suggestions: f i r s t , that the automatic stage i s similar in both hemispheres, and second, that hemispheric differences arise at the "interface" of automatic and attentionally controlled processes. In addition, several models of hemispheric function were introduced and discussed with respect to the d i f f e r e n t experiments. To summarize, evidence was found that automatic processing i s indeed similar i n both hemispheres, independent of whether i t involves only early sensory or also higher l e v e l s of information processing. This conclusion i s based on the finding that differences between the hemispheres were neither present for a very early automatic stage of object perception ( i l l u s o r y conjunctions) nor for the e f f e c t s of automatic Page 65 processes believed to involve higher l e v e l s of information processing (priming e f f e c t and Stroop-ef feet i n condition 1 (5 0/5 0) ) . It seems that automatic stages are common to both hemispheres. This claim i s substantiated by the finding that i l l l u s o r y conjunctions are made 'across hemispheres' (Treisman and Schmidt, 1982) . P i l o t data from an i l l u s o r y conjunction experiment using colored l e t t e r s support t h i s finding. Two l e t t e r s were positioned v e r t i c a l l y i n cne v i s u a l f i e l d and a th i r d one horizo n t a l l y across i n the opposite f i e l d . The mean number of conjunction errors per t r i a l was .13 across and .09 within v i s u a l f i e l d s . Some suggestions on how the findings could be corroborated further have already been made for i l l u s o r y conjunctions. The findings in the priming task and the Stroop-task could be generalized to task situations involving, for example, words (the unreported dimension or ncntarget) and pictures or color bars (the reported dimension or target) . Two of the three aspects of attentional operation mentioned, namely selection and weighting, were explored i n t h i s study. The t h i r d one, an integrative function of attention, could be examined by manipulating attentional load i n an i l l u s o r y conjunction task. Any effects of load on the number o r type of conjunction errors should then r e f l e c t t h i s integrative function of attention. In a selective attention situation no hemispheric differences Page 66 were found f o r items w i t h i n each hemisphere (priming task w i t h i p s i l a t e r a l ncntargets) . However, when items from both hemif i e l d s were i n v o l v e d (priming task w i t h c o n t r a l a t e r a l ncntargets) , the RH s u f f e r e d more i n t e r f e r e n c e from a LH-item than the LH s u f f e r e d from a RH-item. I t i s p o s s i b l e t h a t the LH may focus a t t e n t i o n t o a high degree on i n f o r m a t i o n o r i g i n a l l y r e c e i v e d i n t h a t hemisphere. A t t e n t i o n i n the RH on the other hand, seems to be a t t r a c t e d away by items i n the LH. T h i s r e s u l t should also be extended to d i f f e r e n t task s i t u a t i o n s . I t would be i n t e r e s t i n g to f i n d out whether the same p a t t e r n of e f f e c t s would be found f o r more S t r o o p - l i k e t a s k s . So f a r , the Stroop-ef f e e t has been i n t e r p r e t e d as p u r e l y automatic. However, t h i s might be o n l y t r u e f o r S t r o o p - f a c i l i t a t i o n . S t r o o p - i n t e r f e r e n c e e f f e c t s are b e l i e v e d to a r i s e as a r e s u l t of response c o n f l i c t , which c o u l d be i n t e r p r e t e d as an a t t e n t i o n a l l y c o n t r o l l e d response s e l e c t i o n . By i n t r o d u c i n g n e u t r a l t r i a l s , f a c i l i t a t i o n and i n t e r f e r e n c e e f f e c t s c o u l d be anlayzed s e p a r a t e l y . A task w i t h words and c o l o r - b a r s , f o r example, would allow f o r i p s i - and c o n t r a l a t e r a l p r e s e n t a t i o n s . Thus, the question whether a s i m i l a r aymmetry as i n the s e l e c t i v e a t t e n t i o n task would a l s o be found f o r l a t e r i n t e r f e r e n c e e f f e c t s of a°more S t r o o p - l i k e task c o u l d be answered. I f the i n t e r p r e t a t i o n t h a t the LH tends to a t t r a c t a t t e n t i o n away form the RH i s c o r r e c t , the f a c i l i t a t i v e e f f e c t s found i n the priming task may not be p u r e l y automatic as has been Page 67 suggested so f a r . It would be interesting to manipulate the frequency of semantically associated t r i a l s and thereby the extent to which subjects divide their attention between the two simultaneous words. With more attention being paid to the ncntarget, more f a c i l i t a t i o n f or semantically associated words and more interference f o r unrelated words are expected. If RH support of processing i n the LH i s somewhat automatic, t h i s manipulation should have l i t t l e influence on targets i n the LH in contralateral arrangements. If LH support of processing i n the RH involves attentional e f f e c t s , on the other hand, the extent to which attention i s divided should affect f a c i l i t a t i o n and interference f o r the processing of contralateral targets in the RH. With attention viewed as a weighting t o o l , i t was found that females tend to attach larger attentional weights i n the RH than i n the LH. The results also suggested that females may have used attentional strategies i n a condition believed to involve only automatic effects (Stroop-effeet i n condition 1 (50/50)). This finding i s somewhat puzzling. If attention tends to be focused i n the LH, as suggested by the priming task, then cne would expect bigger attentional effects i n the LH rather than in the RH. It i s possible, that d i f f e r e n t functions of attention, i . e . integration, selection or weighting, show di f f e r e n t l a t e r a l asymmetries. The LH may be more specialized f o r s e l e c t i o n , while the RH may be more specialized f o r integration or weighting. Cl e a r l y , t h i s trendwise e f f e c t Page 68 needs further investigation. One manipulation that comes to mind i s to vary the r e l a t i v e frequency of consistent and inconsistent t r i a l s of a Stroop-task separately for each v i s u a l f i e l d . I t would be valuable to know whether a strategy can be controlled independently i n the two hemispheres, or whether i t s use i n one hemisphere automatically generalizes to the other hemisphere. As to the models of hemispheric function that were discussed, the results from the i l l u s o r y conjunction experiments supported Kinsbourne's (19 82) and Moscovitch's (1979) hypotheses claiming that there are no hemispheric differences in e a r l y , prea t tent ion a l and automatic processes. However, the findings i n the priming task and the Stroop-task suggest that the locus where l a t e r a l asymmetries emerge may be better described i n terms of automatic versus a t t e n t i o n a l l y controlled stages than in terms of early sensory and prea t tent ion a l versus attentional stages, as proposed by Kinsbourne (19 82) and Moscovitch (1979) . I t seems that not only early sensory processing, but also higher l e v e l processing, i f performed automatically, i s similar in both hemispheres. Also, i n disagreement with Ki nsbourne'*s (1982) and Moscovitch's (1979) hypotheses, i t i s suggested that i f l a t e r a l asymmetries do emerge, they may be described i n terms of attentional control rather than in terms of hemispheric sp e c i a l i z a t i o n of function. The results in the priming task also provided some, although not very strong support for a d i r e c t access approach and the Page 69 multiple resources model (Friedman and Poison, 1981). The s e r i a l / p a r a l l e l d i s t i n c t i o n , as proposed by Cohen (1973) was refuted, however. It seems that the results found i n the three experiments complement each other and provide a s u f f i c i e n t l y clearcut picture to make speculative claims about the functional locus of hemispheric differences. It i s speculatively concluded that the locus where hemispheric differences emerge i s the 'i n t e r f a c e ' between automatic and attentionally controlled processes. Automatic processes may involve d i f f e r e n t stages i n information processing, depending on the task requirements. It seems that for a l l stages explored i n the present study these automatic processes are shared and common to both hemispheres. Lateral asymmetries are believed to arise only from attentional operations. Three possible c h a r a c t e r i s t i c s of such attentional operation were introduced: integration, selection and weighting of evidence from automatic processes. For the weighting function of attention a trend f o r a sex difference emerged. The RH in females seemed to weight evidence from automatic processes more a t t e n t i o n a l l y than the LH. Males tended to show the opposite e f f e c t . Clear evidence was found f o r s e l e c t i v e mechanisms, however, which seem to d i f f e r in the two hemispheres when items from both v i s u a l f i e l d s are involved. It seems that the RH can support processing i n the LH in an automatic fashion with r e l a t i v e l y l i t t l e cost. The LH, on the other hand, seems to at t r a c t attention away from the RH and Page 70 produces substantial interference (cost) as well as f a c i l i t a t i o n f o r processing i n the RH. Page 71 REFERENCES B e l l e r , H. K. P a r a l l e l and s e r i a l stages i n matching. Journal of Experimental Psychology, 1970 , 84, 213-219. Bradshaw, J. L. Peripherally presented and unreported words may bias the perceived meaning of a c e n t r a l l y fixated homograph. Journal of Experimental Psychology, 1974, Vol. 103, No. 6, 1200 - 1202. Bradshaw, J. L. and Gates, A. Visual f i e l d differences i n verbal tasks: Effects of task f a m i l i a r i t y and sex of subject. Brain and Language, 1978, 51, 166-187. Brownell, H. H. and Gardner, H. Hemisphere s p e c i a l i z a t i o n : D e f i n i t i o n s , not incantations. An open peer commentary to: Bradshaw, J. L. and Nettleton, N.C. The nature of hemispheric s p e c i a l i z a t i o n i n man. The Behavioral and Brain Sciences, 1981, 4, 51-91. Cohen, G. Hemispheric differences i n s e r i a l versus p a r a l l e l processing. Journal of Experimental Psychology, 1973, 97, 349-356. Coren, S. , Porac, C. and Duncan, J. A behaviorally validated s e l f - r e p o r t inventory to assess four types of l a t e r a l preference. Journal of C l i n i c a l Neuropsychology, 1979, 1, 55-64. Day, J. Right-hemisphere language processing i n normal right-handers. Journal of Experimental Psychology, Human perception and Performance, 1977, Vol. 3, No. 3, 518-528. E l l i s , H. D. and Shepherd, J. W. Recognition of abstract and concrete words presented i n l e f t and r i g h t v i s u a l f i e l d s . Journal of Experimental Psychology, 1974, Vol. 103, No. 5, 1035-1036. Friedman, A. and Poison, M. C. The hemispheres as independent resource systems: limited-capacity processing and cerebral s p e c i a l i s a t i o n . Journal of Experimental Psychology, Human Perception and Performance, 19 81, Vol. 7, Nr. 5, 1031-1058. Gardner, H. The shattered mind. New York: Random House, 1974. Geschwind, N. The perverseness of the r i g h t hemisphere. An open peer commentary to: Puccetti, R. The case f o r mental duality: Evidence from s p l i t - b r a i n data and other considerations. The Behavioral and Brain Sciences, 1981, 4, 93-123. Page 72 Heilman, K. and Watson, R. T. Mechanisms underlying the un i l a t e r a l neglect syndrome. In: Hemi-inattention and hemisphere s p e c i a l i z a t i o n , E. A. Weinstein and R. P. Friedland (Eds.), New York: Raven Press, 1977. Hellige, J. B., Cox, P. J. and Litvac, L. Information processing i n the cerebral hemispheres: Selective hemispheric activation and capacity l i m i t a t i o n s . Journal of Experimental Psychology, General, 1979, Vol. 108, No. 2, 251-279. Hines, D. Recognition of words, abstract nouns and concrete nouns from the l e f t and right v i s u a l h a l f - f i e l d s . Neuropsychologia, 1976, Vol. 14, 211-216. Kimura, D. Dual functional asymmetry of the brain in vi s u a l perception. Neuropsychologia, 1966, 4, 278-285. Kinsbourne, M. The mechanism of hemispheric control of the l a t e r a l gradient of attention. In: Attention and Performance, V, P. M. A. Rabbit and S. Domic (Eds.), New York: Academic Press, 19 75. Kinsbourne, M. and Hicks, R. E. Functional cerebral space. A model for overflow, transfer and interference e f f e c t s i n human performance. In: Attention and Performance, VII, J. Requin (Ed.), H i l l s d a l e , N. J . : Erlbaum, 1978. Kinsbourne, M. Hemispheric s p e c i a l i z a t i o n and the growth of human understanding. American Psychologist, 1982, Vol. 37, No. 4, 411-420. Logan, G. D. Attention and automaticity i n Stroop and priming tasks: Theory and data. Cognitive Psychology, 19 80 , Vol. 12 , 23-53. Logan, G. D. and Zbrodoff, N. J. When i t helps to be misled: F a c i l i t a t i v e e f f e cts of increasing the frequency of c o n f l i c t i n g s t i m u l i in a Stroop-like task. Memory and Cognition, 1979, 7(3), 166-174. Marcel, A. J. Conscious and unconscious reading: The ef f e c t s of visual masking on word perception. Cognitive Psychology, i n press. McGlone, J. Sex differences i n human brain asymmetry: a c r i t i c a l survey. The Behavioral and Brain Sciences, 19 80, 3, 215-263. M i l l e r , J. Discrete versus continuous stage models of human information processing: In search of p a r t i a l output. Journal of Experimental Psychology, Human Perception and Performance, 19 82, Vol. 8, No. 2, 273-296. Page 73 Milner, B. L a t e r a l i t y e f f ects in audition. In: Inter hemispheric relations and Cerebral Dominance, Mountcastle, V. B. (Ed.), John Hopkins Press, Baltimore, 1962. Morton, J. Interaction of information in word recognition. Psychological Review, 1969, 76, 16 5-178. Moscovitch, M. Information processing and the cerebral hemispheres. In: Handbook of behavioral neurology, Gazzaniga, M. S. (Ed.), Vol. 2, Plenum Publishing Co., N.Y., 1979. Moscovitch, M. and K l e i n , D. M a t e r i a l - s p e c i f i c perceptual interference f o r v i s u a l words and faces: Implications for models of cpacity l i m i t a t i o n s , attention, and l a t e r a l i t y . Journal of Experimental Psychology, Human Perception and Performance, 19 80, Vol. 6, No 3, 5 90-603. Neely, J.H. Semantic priming and r e t r i e v a l from l e x i c a l memory: Roles of i n h i b i t i o n l e s s spreading activation and limited-capacity attention. Journal of Experimental Psychology, General, 1977, Vol. 106, No. 3, 226-254. Neisser, U. Cognitive Psychology. New York: Appleton-Century- Crofts, 1967 . Patterson, K. E. and Marcel, A. J. Aphasia, dyslexia and phonological coding of written words. Quarterly Journal of Experimental Psychology, 1977, 29, 307-318. Posner, M. I. and Snyder, C. R. R. Attention and cognitive control. In: Information Processing and Cognition: The Loyola Symposium, R. L. Solso (Ed.), H i l l s d a l e , N. J . , Erlbaum, 1975. R i z z o l a t i , G., Umilta, C. and Berlucchi, G. Opposite s u p e r i o r i t i e s for the rig h t and l e f t cerebral hemispheres i n discrimination RT to physiognomical and alphabetical material. Brain, 1971, 94, 431-442. Schmit, V. and Davis, R. The r o l e of hemispheric s p e c i a l i z a t i o n in the analysis of Stroop s t i m u l i . Acta Psycholigica, 1974, 38, 149-158. Sergent, J. The cerebral balance of power: Confrontation or cooperation? Journal of Experimental Psychology, Human Perception and Performance, 1982, Vol. 8, Nr. 2, 253-272. Shapiro, S. I. and Palermo, D. S. An atlas of normative free association data. Psychonomic Monograph Supplements, Vol. 2, No. 12, (Whole No. 28), Psychonomic Press, 1968. Page 74 S h i f f r i n , R. M. and Schneider, W. C o n t r o l l e d and automatic human information p r o c e s s i n g : I I . P e r c e p t u a l l e a r n i n g , automatic a t t e n d i n g , and a g e n e r a l theory. P s y c h o l o g i c a l Review, 1977, V o l . 84, No.2, 127-190. Schwartz, S. and K i r s n e r , K. L a t e r a l i t y e f f e c t s i n v i s u a l i nformation p r o c e s s i n g : hemispheric s p e c i a l i s a t i o n or the o r i e n t i n g of a t t e n t i o n ? Q u a t e r l y J o u r n a l of Experimental Psychology, 19 82, 34A, 61-77. Treisman, A. M. and Gelade, G. A f e a t u r e - i n t e g r a t i o n theory of a t t e n t i o n . C o g n i t i v e Psychology, 19 80 , 12 , 97-136 . Treisman, A. and Schmidt, H. I l l u s o r y c o n j u n c t i o n s i n the p e r c e p t i o n of o b j e c t s . C o g n i t i v e Psychology, 19 82, 14, 107-141. Underwood, G. Semantic i n t e r f e r e n c e from unattended p r i n t e d words. B r i t i s h J o u r n a l of Psychology, 197 6, 67 ,3 27-338. Warren, L. R. and Marsh, G. R. Hemispheric asymmetry i n the p r o c e s s i n g of Stroop s t i m u l i . B u l l e t i n of the Psychonomic S o c i e t y , 1978 , V o l . 12 (3), 214-216. White, M. J . and White, K. G. P a r a l l e l - s e r i a l p r o c e s s i n g and h emispheric f u n c t i o n s . Neuropsychologia, 19 75, 13, 377-381. Wickens, C. D. , Mountford, S. J . and S c h r e i n e r , W. M u l t i p l e r e s o u r c e s , task-hemispheric i n t e g r i t y , and i n d i v i d u a l d i f f e r e n c e s i n t i m e - s h a r i n g . Human F a c t o r s , 19 81, 23 , 2, 211-229. Z a i d e l , E. L e x i c a l o r g a n i z a t i o n i n the r i g h t hemisphere. In: C e r e b r a l C o r r e l a t e s of Conscious Experience, INSERM Symposium No. 6, Buser and Rougeul-Buser (Eds.), E l s e v i e r / N o r t h - H o l l a n d Biomedical P r e s s , 1978. (a) Z a i d e l , E. Concepts of c e r e b r a l dominance i n the s p l i t b r a i n . In: C e r e b r a l C o r r e l a t e s of Conscious Experience, INSERM Symposium No. 6, Buser and Rougeul-Buser (Eds.), E l s e v i e r / N o r t h - H o l l a n d Biomedical P r e s s , 1978 . (b) 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0095348/manifest

Comment

Related Items