NONLANGUAGE CEREBRAL MECHANISMS
IN A VISUAL FIELD TASK
BRUCE J. SCHELL
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DECREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
The author wishes to express his appreciation of his committee
for the people they are. Particular thanks are due to his Chairman,
Dr. Paul Satz, for whose many forms of assistance over the years, I
can never adequately acknowledge. Special thanks are also due to
Dr. Madelaine Raimey for assistance in the statistical analysis of the
data. For the author, the most significant aspect of this study was
that he was allowed to do his own "thing," with its unique rewards
and pitfalls. For this I am truly grateful to my entire committee.
I would also like to thank my parents, Mr. and Mrs. B. J. Schell
of Las Cruces, New Mexico, for the wonderful support they have given
me throughout the years.
TABLE OF CONTENTS
LIST OF TABLES......................................................... iv
LIST OF FIGURES.................................................... v
INTRODUCTION........................................ ......... ...... 1
METHOD............................................... ............... 11
RESULTS .......... .............. .............. ......... ............ 14
DISCUSSION,..... .................... ..................... ......... 20
SUMMARY................................................................. .. 25
1 STIMULUS AND ERROR TYPES EXAMPLE .......................... 27
2 SUMMARY ANALYSIS OF VARIANCE................................ 28
BIBLIOGRAPHY ................. ....................................... 29
BIOGRAPHICAL SKETCH................................................ 31
LIST OF TABLES
TABLE 1 WAN CORRECT RECALL BY VISUAL FIELDS AND FAMILIAL HANDEDNESS... 16
LIST OF FIGURES
FIGURE 1 DISTRIBUTION OF ERROR TYPES............................. 17
FIGURE 2 PERCENT ERROR TYPES BY VISUAL FIELD ..................... 18
FIGURE 3 FAMILIAL HANDEDNESS BY ERROR TYPES...................... 19
Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy
NONLANGUAGE CEREBRAL MECHANISMS
IN A VISUAL FIELD TASK
Bruce J. Schell
Chairman: Paul Sats, Ph.D.
Major Department: Psychology
In recent years, considerable literature has been devoted to
laterality differences in visual perception. The basic rationale for
these experiments is twofold. First, in vision, each visual half field
(VHF) projects directly to the contralateral cerebral hemisphere.
Second, clinical studies of brain-injured adults have demonstrated a
dual functional asymmetry between the cerebral hemispheres in man,with
the left hemisphere primarily subserving speech and language function,
while the right hemisphere has been Implicated in higher integrative
nonverbal visual-spatial functions.
The demonstration and extension of these clinical findings, through
examination of VHF asymmetries in normals, hae been hampered by method-
ological inadequacies. Recently a new paradigm seemingly obviating these
methodological shortcomings has been presented in two studies by Hines.
This paradigm, Incorporating the principle of simultaneity from the
previous approaches, has departed from them in two major ways. First,
a new method to ensure macular fixation was devised. Second, a short-
term memory variable was introduced.
The present study was designed to test the efficacy of the paradigm
with a nonverbal stimulus, to investigate the type of errors made by
the Ss, and to test what effect familial sinistrality had on these factors.
Fifty-one right-handed Ss were presented block design stimuli via a
16 mm. projector. Ten of these Ss were eliminated on the basis of their
low performance level. The stimuli were presented 3 to the left or
right of fixation. Fixation was maintained through sequential presen-
tation and initial recall of four single digit numerals. The Ss were
then required to identify the correct block design from a five-item
multiple choice array. The array consisted of the correct choice, its
lateral mirror image (LMI), its vertical mirror image (VMI), a different
shaped design (DSD) occupying the same general area, and an unrelated
A significant left visual field (LVF) superiority in terms of mean
VHF differences (F1,80 4.58, p .05) and the number of Ss with an over-
all LVF superiority werefound (X 14.9, df=l, p--.001). Familial
sinistrality was found to have no significant effect on the mean VHF
The typesof errors made on missed trials were found not to.be
distributed randomly (X 2r= 17.28, df= 3, p<.001). There were more LMI
and fewer URD type errors than there were VMI or DSD types.
Significantly more URD errors were made in the right visual field
(RVF) than in the left (Zr2.03, p!.025).
Those Ss with a positive history of familial sinistrality made
significantly more VMI type errors (Z=1.86, p4.05) and significantly
less URD type errors (Z=2.02, p .025) than the other Ss.
The present findings, through demonstration of a LVF superiority
for a nonverbal stimulus, lend support to the feasibility of the use
of normal Ss to demonstrate brain laterality difference, and demonstrate
the efficacy of a new technique.
The lower incidence of URD errors in the LVF provides evidence not
only for the role of the right cerebral hemisphere in the processing
of nonverbal stimuli but also for the viability of using types of errors
to provide information about cerebral function. Thus, the lower rate
of URD type errors related to a positive history of familial sinistrality
suggests the possibility of a more diffuse cortical representation for
nonverbal function in those Ss.
While the results offer a new contribution to our knowledge of brain-
behavior relations, their primary significance relates to the viability
of these new methodologies.
The specialization of speech and language functions in the left
cerebral hemisphere in man has in recent years focused attention on the
role of other higher integrative functions which may be mediated by the
right cerebral hemisphere. Clinical studies, using heterogeneous brain-
Injured populations, have suggested that the right cerebral hemisphere
may be dominant for special visual-spatial functions (Teuber, 1962).
Unfortunately, many of these studies are fraught with methodological
problems (e.g. lesion specificity and variability, age, I. Q., and
edema effects) which tend to obscure possible brain-behavior relation-
,ships. In an attempt to control for these methodological considerations
the use of normal Ss to investigate brain-behavior relations has pro-
liferated. A principal approach has been the use of laterality differences
in visual perception as an indicant of underlying cerebral hemispheric
differences. However, the use of normal Ss has introduced new method-
ological difficulties (principally habitual directional scanning effects
and uncontrolled eye movements) which have again obscured possible brain-
If these methodological considerations do tend to obscure the
results, why bother to use the visual system? The primary advantages of
the visual modality for investigation of laterality differences lie in
the potential range of stimuli that may be investigated, the division of
each retina allowing specificity of the hemisphere to which the material is
projected, and the direct connections between the visual half fields
(via crossed and uncrossed pathways) and the contralateral cerebral
The present paper shall present a rationale for a new methodological
approach to the question of right cerebral hemispheric function. This
rationale will be expurgated through a review of relevant clinical,
experimental, and theoretical papers.
Two primary methodologies have been used in the visual research.
The first, historically, was sequential tachistoscopic presentation
(Type I) of verbal and nonverbal materials in the left (LVF) and right
(RVF) visual fields. The second approach (Type II) has involved simul-
taneous presentation of disparate stimulus pairs to the visual half
The Type I approach, in which Ss attend to a fixation point with
the stimulus materials presented a few degrees of visual angle to the
left or right of the fixation, has resulted in a predominant RVF super-
iority for verbal materials (White, 1969). The primary explanation for
these findings has been that the RVF has more direct connections to the
language centers in the contralateral left hemisphere. This finding
would seemingly demonstrate, in normals, the efficacy of the technique
and the predominant role of the left hemisphere in verbal processing.
Unfortunately, the value of these results is called into question by the
failure to demonstrate a corresponding LVF superiority for nonverbal
materials. The major demonstration, within the technique, of a LVF
superiority for "nonverbal materials" was by Kimura (1966). Attempts to
replicate this finding have either failed (Kinsbourne, 1970) or been
demonstrated only under special conditions (Van Nostrand, 1970).
Indeed, it has been questioned whether the task itself, enumeration of
dots, constitutes a nonverbal task.
The Type II design, simultaneous visual half field stimulation,
was based conceptually on analagous research in audition and on results
obtained from some clinical lesion groups. The clinical studies found
that patients with lesions of, for example, the right parietal lobe,
would, under simultaneous stimulation, often fail to report the stimulus
that was going to the right hemisphere (Teuber, 1962). As these patients
were fully capable of responding to sequential stimulation it was reasoned
that in the simultaneous condition somehow the "weaker" processor falls
to respond or responds inadequately to the incoming signal. Demonstration
of the phenomenon is then dependent on a "weaker" processor and the
stimulus condition of the simultaneous presentation. This phenomenon,
the extinction effect, has been demonstrated in all sensory modalities
and is routinely used in the screening of patients for cerebral insult.
It would then be expected, if in normals there are functional differences
between the two cerebral hemispheres, that the simultaneous condition
would tend to accentuate this difference. Indeed, this was found in the
auditory research (Kimura, 1967; Satz, 1968). Simultaneous presentation
of disparate auditory information resulted in a clear cut difference in
accuracy of recall between the two ears. That is, when verbal materials
were presented simultaneously to both ears in normal right handers, a
significantly greater amount of the information that was presented to the
right ear was recalled correctly. The Kimura (1967) study further demon-
strated that when nonverbal materials (melodic patterns) were presented
simultaneously to the two ears a significantly greater amount of the
material presented to the left ear was recalled correctly. With the
application of the simultaneous technique in audition we then find clear
cut differences between the two ears (cerebral hemispheres), dependent
upon type of stimulus materials, that have not been seen with sequential
presentation (Kimura, 1967), The direct extension of the simultaneous
technique to vision was not, however, particularly fruitful in the
elucidation of underlying cerebral hemispheric differences. Almost
invariably a LVF superiority has been found for verbal materials. This
is directly counter to any central nervous system (CNS) hypothesis
(White, 1969). This predominant LVF superiority for verbal materials
has been primarily explained as due to the directional scanning effects
(Heron, 1957), selective training of the right hemiretina (Orbach, 1967),
and the intrinsic directionality of the stimulus (Orbach, 1967). Aside
from these artifactual elements producing differences between the visual
and auditory research another major difference is apparent. That is,
the auditory research typically presents three to four digit pairs per
trial thus introducing a short term memory (STM) variable that has
typically not been a factor in the visual studies.
In an attempt to control for these factors and to introduce a STM
variable, a new methodological approach (Type III) has been developed.
While incorporating the principle of simultaneity from the Type II
design, the Type III design deviates otherwise by ensuring more adequate
control of macular fixation and by including an STM component. This
approach has demonstrated, within a STM paradigm, a significant RVF
superiority for verbal stimuli (Hines et al., 1969; Hines and Sats, 1970).
Further, this RVF superiority was observed in over 80% of the right
handed Ss that were tested. It remains to be seen, however, whether this
technique is superior to the previous methodological approaches, as the
demonstration of a RVF superiority for verbal materials is but half the
task. Within this methodology a LVF superiority for nonverbal materials
must also be demonstrated before its value in exploring functional
cerebral laterality differences may be adequately assessed. The present
study is designed to test the efficacy of this new paradigm for nonverbal
From the above decision three immediate questions arise. What type
of nonverbal materials should be used? How does one measure the perform-
ance of the Ss? Is it possible within the STM paradigm to demonstrate a
LVF superiority for nonverbal stimuli? Each of these questions, of
course, rests on the separate and more general assumption that the right
cerebral hemisphere is differentially specialized for "nonverbal-spatial"
Previous experimental and clinical literature can provide some
guidelines for the type of nonverbal material used. The first criterion
as negatively demonstrated by the Kimura (1966) study, Is that other
investigators should be in agreement that the stimuli are indeed non-
verbal. The second guideline is provided by the research of Bryden and
Rainey (1963), Wyke and Ettlinger (1961), and Glanzer and Clark (1964).
The first two experiments demonstrated that the use of highly
familiar objects produces a RVF superiority. This RVF superiority is
probably due to the Ss verbally identifying the stimulus, thus producing
primarily a language mediated response for this obstensibly nonverbal
task. Clinically, Kimura (1963) has further demonstrated the importance
of the familiarity-nonfamiliarity dimension. In this study, using left
and right temporal lobectomy patients, she found inferior performance for
the right lobectomy patients (compared to the left lobectomy patients) on
an overlapping nonsense figures recognition test, while the converse was
seen on a familiar overlapping figures test. The Glanzer and Clark
(1964) study suggests further that Ss often verbally encode the visual
stimuli in "nonverbal visual memory" tasks. Therefore, the second guide-
line is that the nonverbal stimuli should be unfamiliar and not readily
accessible to verbal codification.
An appropriate type of design matching the above criteria is the
block design. Historically, block designs of varying complexity have
been used in the detection of general non-specific cerebral insult
(Satz, 1966). More recently, it has been shown that lesions invading
the right cerebral hemisphere differentially impair performance on block
design tasks (Parsons, 1970). It has also been demonstrated that
patients with split brains (callosal disconnection) are able to construct
design patterns significantly better with their left hands, presumably
because of the "dominance" of the functions in the contralateral right
hemisphere (Cazzaniga, 1967). Block designs have the advantage of
professional accord with respect to their nonverbal loading and as stimuli
represent objects low in familiarity. Technically, they provide the
additional advantage of being a standard size stimulus of which specified
manipulations may be made. How, then, using block design stimuli, can
one measure the Ss performance? Three possible alternatives would appear
to be available; recall expressed by describing the stimulus, recall
expressed by drawing the stimulus, and finally, recall demonstrated
through correct identification of the stimulus from a multiple stimulus
array. The first alternative, through measuring a verbally encoded
response, would defeat the purpose of or bring Into question the value
of the study. Requiring the Ss to draw the stimuli suffers from a
number of drawbacks, the principal ones being the introduction of a
subjective scoring element and the time consumed in drawing the stimuli.
The third alternative is preferable as it interferes least with the non-
verbal quality of the task, provides an objective scoring technique,
and allows for investigation into the types of errors made when the
correct stimulus is not identified.
Relatively uninvestigated, particularly with regard to the influence
of functional cerebral laterality differences, is the question of what
types of errors Ss make when they are wrong. That is, will different
errors be made depending on the visual field to which the stimuli are
presented? Are particular types of errors more common across visual
fields than other types? Some suggestions as to possible factors to
investigate are provided by the studies of Noble (1968), Goodnow (1969),
and Mello (1965).
Noble (1968), using primates in the Wisconsin General Test Apparatus
(WGTA), has investigated some of the possible types of errors. In this
study, optic chiasm sectioned monkeys had one eye occluded and were
trained on a two choice angular discrimination. The nonreinforced element
of the discrimination was either a lateral mirror image (LMI) or a vertical
mirror image (VMI) of the reinforced stimulus. In this condition, due to
the chlasm section, one cerebral hemisphere receives direct visual input
while the other receives visual information indirectly via the corpus
callosum. After demonstrating that the discrimination had been learned,
he then reversed which eye was occluded and, thereby, which hemisphere
received direct visual stimulation. He found that the nonrelnforced LMI
was responded to at a higher level than the previously reinforced stimulus.
Thus, the information which was transmitted via the corpus callosum during
the training trials resulted in an orientational "confusion" of the
correct stimulus during the test trials. This effect was not seen with
Similarly, Mello (1965) found, after training pigeons on an angular
discrimination with an opaque goggle over one eye, that upon switching
eyes the pigeon responded maximally to the LMI of the reinforced stimulus.
Goodnow (1969) and Boone and Prescott (1968) have reported in children
that discrimination of changes in the vertical direction are easier than
changes in the horizontal direction.
Finally, the misperceptionsof the LMI and VMI as the correct stim-
ulus are common errors in children learning to read (e. g. reporting or
writing b for p (VMI) or d (LMI)).
The LMI and VMI errors both represent errors of direction (orienta-
tion). That is, the shape of the stimulus is preserved, but the direct-
ionality angulationn) of the stimulus is lost. This suggests, as does
the research of Hubel (1963),that, possibly, the shape and direction of
a stimulus are dissociable elements of the total perception. Hubel (1963)
has demonstrated, in the cat, with single cell recording from the visual
cortex, that there are cells that respond maximally to the shape of the
stimulus while others respond maximally to its orientation. Thus an
additional possible error is to correctly perceive the orientation
(direction) of the stimulus and yet to have an incorrect perception of
its shape. To test for this error it is necessary to have a different
shaped design (DSD) occupying the same general area.
Two principal theories are relevant to the basic question of
whether, within this paradigm, a LVF superiority for nonverbal materials
can be demonstrated. Both theories (Kimura, 1966; Kinsbourne, 1970)
are in agreement that in normal right handers language is predominately
represented in the left cerebral hemisphere and that the right hemisphere
plays a major role in nonverbal visual-spatial tasks. Kimura's (1966)
theory further proposes that stimulus input transmitted on the most
direct pathway will better maintain the integrity of the stimulus signal
and will therefore be more efficiently processed. Her theory then
predicts, barring artifacts of technique, a LVF superiority for non-
verbal materials and a RVF superiority for verbal materials. Kinsbourne
(1970) has recently proposed an alternative model: If a subject is
engaged in language behavior, prior or during a test trial, a left hemis-
pheric activation will take place and that accompanying this will be a
directional bias to the RVF regardless of the nature of the stimulus
input (e. g. verbal or nonverbal). "Such orientation will characterize
not only overt language use, but also covert (subvocal) language behav-
ior, including the state of expectancy to verbal response." His theory
thus predicts that, as long as the left cerebral hemisphere is activated
by any language behavior, verbal and nonverbal materials will be more
accurately responded to when they are in the RVF.
An additional separate variable relevant to the basic assumption
that the right cerebral hemisphere is "dominant" for nonverbal functions
is the possibility of differences in the degree of cerebral lateralis-
ation of function in the Ss.
Clinical studies by Subirana (1958) and Zangwill (1960) have found,
following cerebral insult, a more rapid remission of aphasia in right-
handed patients with left-handed relatives. This finding suggests the
possibility of a more diffuse representation of cortical function in
those right-handed subjects with a positive family history (+FH) of left-
handedness. Indeed, Hines and Satz (1970) have demonstrated, using the
Type III design with verbal materials, a difference in VHF performance
between those Ss with a +FH of left-handedness and those Ss with a
negative family history (-FH) of it. This difference, an attenuation in
the RVF asymmetry (i. e. superiority) in +FH Ss, is suggestive of a
difference in degree of cerebral lateralization. A preliminary investi-
gation of the influence of this factor on nonverbal materials will be
included in this research.
The present experiment is designed to investigate the following
questions. Will recall accuracy for nonverbal materials be influenced
by the VHF to which the stimulus is presented? Are some type errors
more common than other error types? Will right-handers with a +FH of
left-handedness perform differently than right-handers with a -FH of
left-handedness? That is, will the +FH Ss perform differently in regard
to either VHF asymmetry or types of errors made?
Fifty-one right-handed Ss from introductory Psychology classes
were used. The ages of the Ss ranged from 17 to 30 years with a mean
age of 20.14 years. No Ss were used whose visual acuity was not at
least 20/40 or who had a discrepancy of more than 1/2 minute of visual
arc between the eyes. Visual acuity was established through use of a
Snellen Eye Chart.
The stimuli vere presented via a 16 i.s. Kodak Analyst Projector
onto a rear-view screen. The stimuli were presented at about eye level,
All Ss had their heads positioned on a commercial chin rest. The test
film was administered in a dimly illuminated room with a single light
source directly behind the projector.
All Ss were given a self-administered handedness questionnaire.
They were then screened for visual acuity and finally viewed the test
The test stimuli consisted of 56 trials with 4 numbers and 1 block
design per trial. These numbers were the single digits O through 9.
No number appeared more than once in a trial. At the beginning of each
trial a central fixation point appeared for 605 msec. followed sequen-
tially by the 4 digits at fixation with the block design appearing
either to the left or right of the fixation digits. The exposure time
for each digit was 182 msec. (no interstimulus interval). The block
design was projected 30 from the fixation numerals for a total exposure
time of 606 msec. Each digit subtended approximately 45' of visual area
in height and 30' in width. The block design subtended approximately
90' of visual area in height and 60' in width. The first digit of the
sequence preceded the onset of the block design by 60.6 msec. with the
last digit remaining on the screen 60.6 msec. after the block design
was no longer present. This was done to insure initial fixation and to
reduce shift in fixation from the last digit over to the block design.
A between trials interval of 10 seconds was used, during which the Ss
reported the digits from the preceding trial (fixation control) and
then identified the design (dependent variable) from a 5 item multiple
choice array. The 5 item multiple choice array consisted of the
correct choice, its lateral mirror image (LNI), its vertical mirror
image (Vil), a different shaped design (DSD) occupying the sare general
area, and an unrelated design.
The first 12 trials were practice and included 6 trials in which
the block design was in the LVF and 6 in the RVF. Of the remaining 44
trials, 10 were not included in the investigation of error types as it
was not possible to -abk all the error types for those designs. There
were 5 of these trials in each visual field.
If, on any trial, the center fixation digits were not reported
correctly, the identification of the design was not subsequently counted.
The s were told to respond to every trial whether or not they knew the
See Appendix 1 for example.
correct response. A criterion of 23% correct (3% above chance) was
chosen for inclusion in the study. Ten Ss were eliminated on this
criterion. With the exception of their poor performance, they were
not distinguishable from the other Ss.
This paradigm, developed by Hines et al. (1969) and modified for
this study, provides two major advantages over the previous approaches.
(1) It insures fixation which the Type I and Type II designs could not
adequately provide. (2) It utilizes STM processes which, in audition,
have uncovered the most striking asymmetries. Finally, this modification
offers a direct and timely test of Kinsbourne's (1970) model which
predicts that verbal processing during or before a test trial will result
in a directional bias to the RVF.
Recall by Visual Half Field
These analyses represent the major test of the present thesis. The
first was concerned with the magnitude of the mean differences between
the RVF and LVF. The second dealt with the directional frequency diff-
erences in recall between the VHFs among the Ss. The mean correct recall
for designs in the RVF and LVF is presented in Table 1. An Analysis of
Variance (liner, 1962) based on the combined group of Ss revealed a
significant LVF superiority (F71,0= 4.58, p .05). The designs in the
LVF were reported correct more often than those from the RVF in 33 of the
41 Ss. This frequency difference was significant (X= 14.9, df= 1, p.001).
VHF Recall by Family History
Inspection of Table 1 reveals that family history of left-handedness
had no significant effect on the mean VHF differences.
VHF by Error Types
The following analyses were computed to see if error types were
distributed randomly and if presentation of the stimulus to a particular
VHF produced a difference in the relative incidence of the error types.
Figure 1 presents a breakdown of types of errors made on missed trials.
A Friedman Analysis (Hays, 1965) of the data revealed that the error
types were not distributed randomly (Xr= 17.28, df=3, p .001).
SSee Appendix 2 for Summary Analysis of Variance.
Inspection of the figure reveals there were more LMI and fewer URD errors
than there were VMI or DSD type errors.
The effect on error types of the VHF to which the design is presented
is shown in Figure 2. The significance of the differences between the
VHFs for the LKI errors and the URD errors was tested with the Wilcoxen
Test for Matched Samples (Hays, 1965). The results of the first analysis
for the LMI error typewere nonsignificant (Z=.45, p-.33). The differ-
ence between the VHFs for the URD error type was significant,
Error Types by Family History
The difference in error types made by +FH Ss and -FH Ss is presented
in Figure 3. The difference in incidence of VMI error types and URD
error types between the two groups of Ss was examined with the Mann-
Whitney Test (Hays, 1965). The VMI error was found to occur significantly
more often in the +FH Ss than in the -FH Ss (Z= 1.86, pC .05). The URD
error occurred significantly less frequently in the +FH Ss than in the
-FH Ss (Z=2.02, p .025).
Mean Correct Recall by Visual
Fields and Familial Handedness
Left Field Right Field
-FH (n 32) 13.06 (59.36%) 11.28 (51.27%)
+FH (n 9 ) 13.55 (61.59%) 11.66 (53% )
Distribution of Error Types
LMI VMI DSD URD
Percent Error Types by Visual Field
LMI V.I DSD URD
Right Field=---N=41, Trials Rated 295
Left Field ----- N=41, Trials Rated 272
Familial Handedness by Error Types
LMI VMI DSD URD
+FH= -- N=9, Trials Rated 123
-FH=-- - N=32, Trials Rated 444
The present findings confirmed the hypothesis that recognition of
nonverbal visual designs is facilitated by side of VHF input. Visual
designs presented to the left VF were more accurately recognized than
those presented to the right VF. This directional asymmetry, moreover,
was observed in over 80% of the Ss. The fact that a reversal in the VHF
asymmetry has been long reported for verbal stimuli (Kimura, 1966)
strongly suggests that brain laterality mechanisms probably underlie the
VHF asymmetry. This superior recognition for design patterns presented
to the left VF suggests that the right cerebral hemisphere may be differ-
entially specialized for processing visual-spatial information. Clinical
studies on brain injured patients have already suggested this possibility
This experiment, with the presentation during the test trials of
the fixation digits, would appear to have been an ideal test of
Kinsbourne's (1970) theory of VF asymmetries. His theory predicts a RVF
superiority regardless of the nature of the VHF stimuli (verbal or non-
verbal) whenever the subject is engaged in verbal activity concurrent
with or immediately preceding the test trial. The LVF superiority seen
in this study would then appear to disconfirm Kinsbourne's theory at
least in its present form. It is, of course, possible that in the
absence of adequate control of central fixation, Kinsbourne's theory
would be maintained. A review of the evidence offered by Kinsbourne
suggests that indeed this may be the case.
The failure to find a statistically significant difference, in
terms of VHF recall, between those Ss with a +FH and those with a -FH,
is at variance with the Hines and Satz (1970) study. That study, using
the Type III design with verbal materials, found a decreasing difference
between the two groups as the difficulty level of the task increased. At
the speed of 177 msec. per digit pair they found no difference between
the groups. It may be that the difficulty level, for nonverbal materials,
in this study was similar to that of the Hines and Sate (1970) study with
verbal material at its faster speeds. Alternatively, it may be that
the influence of familial sinistrality upon cerebral lateralization of
nonverbal function is more subtlely felt than are differences in degree
of language lateralization. Consonant with this hypothesis is the
difference in types of errors made by the two groups. A significantly
lower incidence of URD errors among the +FH as compared to the -FH group
is suggestive of the possibility of a more diffuse representation for
nonverbal function in the +FH group. That is, the URD error is the only
error that has no relationship to the original stimulus. Thus, when
that error was committed no portion of the stimulus gestalt was utilized.
This implies that the +FH group was able to utilize more consistently
at least a portion of the stimulus gestalt. The greater incidence of VMI
errors in the +FH group suggests that the more diffuse representation
seen in this group is primarily related to some "recognition of shape"
function rather than a directional or orientational factor.
The finding that the LMl error was the preferred error for all Ss
supports the infrahuman research findings of Noble (1968) and Mello (1965).
The demonstration, within this study, of a higher rate of LMI type errors
does not represent a new discovery about humans. This, however, is the
first study of types of errors made within the visual field research.
The mechansim predisposing species ranging from humans to pigeons for
this type error remains obscure. Speculatively, it is as if the synaptic
signal for direction in the vertical dimension is somehow "weaker" or
more tenuously linked to the total. Ve do know that in ontology
discrimination in the vertical dimension is slower to develop and more
difficulties are experienced with this dimension than with the horizontal
dimension (Boone and Prescott, 1968). The lower incidence of URD errors,
the error with no relationship to the stimulus, emphasizes that Ss,
even when they do not correctly identify the stimulus, still have pro-
cessed part of the stimulus gestalt. That is, if the Ss had no inform-
ation about the stimulus, then the rate of response to the URD error
should have been the same as that of the DSD and VMI type error.
This study represents the first investigation of the influence of
the VHF upon types of errors made. The significantly lower incidence
of URD errors in the LVF thus provides evidence not only for the role
of the right cerebral hemisphere in the processing of nonverbal stimuli
but also for the viability of using types of errors to provide infor-
mation about cerebral function. Thus, even in the absence of a demon-
strable LVF superiority for block designs, this evidence, that the Ss
were able to utilize more of the stimulus information from the LVF,
would have, alone, been indicative of the predominant role of the right
hemisphere in nonverbal function.
This study, representing the fruition of several years of collab-
orative research, has demonstrated two new methodological breakthroughs
in the investigation of cerebral laterality differences. While the
results offer a new contribution to our knowledge of brain-behavior
relations, their primary significance relates to the viability of
these new methodologies. Thus, while the demonstration of a LVF super-
iority for a nonverbal stimulus represents the first unequivocal visual
research showing the unique functional specialization of the right
hemisphere in processing nonverbal stimuli in normal adults, its value
is overshadowed by the potential contribution that this paradigm (Type III)
offers in the study of cerebral mechanisms In perception. Similarly,
the new information about brain function garnered through investigation
of the types of perceptual errors has less significance as new data then
it does as a demonstration of the feasibility of this approach to the
question of cerebral laterality.
The potential extentlons of these techniques to the investigation
of cerebral laterality differences in normals are myriad. For example,
through examination of the types of errors it would be possible to
investigate the relative contribution of the orientation and shape of
the stimulus to the gestalt and to assess the relative contribution
of these two factors depending on the VHF in which the stimulus is
presented. We have completed a preliminary investigation, using the
Type III design with normal right-handed Ss, that demonstrated, in a
single experiment, a dissociation between the VHFs for verbal and non-
verbal stimuli between Ss. Currently we are attempting to demonstrate
this VHF dissociation for verbal and nonverbal materials within Ss and
to investigate the relative influence of varying degrees of left-and
right-handedness on VHF performance. The role of familial history of
left-handedness on performance is also being investigated in the current
study. Projected future studies involve variations in the stimulus
input and extension into new modalities.
The Type III design is, at present, the only approach with normal
right-handed Ss in which a RVF superiority for verbal materials
(iines et. al., 1969) and a LVF superiority for nonverbal materials has
been reliably demonstrated. These findings are thus consonant with
empirical findings on hemispheric asymmetry in brain injured patients.
The major factors unique to this design and, thus, probably responsible
for the results, involved control of macular fixation, the introduction
of a short term memory variable, and the fact that the experimental
conditions were more analogous to a sensory overload situation than
to a sensory threshold situation with its possible peripheral sensitivity
The present study investigated, through the use of two new method-
ologies, VHF recall and error differences to block design stimuli.
Fifty-one right-handed Ss were tested, of which 10 fell below a 23%
cutoff point (3% above chance) and were eliminated. These Ss did not
fall into any definable group. The block designs were presented, via
a 16mm. projector, 30 to the left or right of fixation. Fixation was
maintained through sequential presentation and initial recall of 4
single digit numerals. The Ss were then required to identify the correct
block design fronw a 5 item multiple choice array. A significant LVF
superiority was observed. This finding casts doubt on Kinsbourne's
activation theory of VHF asynnmetries.
The types of errors made by the Ss were also analyzed. This
analysis revealed that the majority of errors represented reversal
orientations of the original stimuli. Further analysis of the error
types revealed an effect attributable to the VHF to which the stimulus
was presented and an effect due to familial history of left-handedness.
Famillal history of left-hendedness was found to have no effect on the
recall scores of the Ss.
The significance of these methodologies for the assessment of
cerebral laterality mechanisms was discussed.
STIMULUS AND ERROR TYPES EXAMPLE
Sti mlus LMI
SUiMARY ANALYSIS OF VARIANCE
Source SS df ms F
Yethod 66 1 66 4.58*
Error 1152 80 14.4
Total 1218 81
* p .05
Boone, D. and Prescott, T., Development of left-right discrimination
in normal children. Percept. Mot. Skills, 1968, 26, 267-274.
Bryden, M!. P. and Rainey, C. A. Left-right differences in tachistoscopic
recognition. J. Exp. Psychol., 1963, 66, 568-571.
Gazzaniga, M. S. Language after section of the cerebral conmissures.
Brain, 1967, 90, 131-148.
Glanzer, N., and Clark, W. H. The verbal-loop hypothesis: Conventional
figures. Amer. J. Psychol., 1964, 77, 621-626.
Goodnow, Jacqueline Eye and hand: Differential sampling of form and
orientation properties. Neuropsychologia, 1969, 7, 365-373.
Hays, W. L. Statistics for Psychologists. Holt, Rinehart and Winston,
N. Y., 1965.
Heron, W. Perception as a function of retinal locus and attention.
Amer. J. Psychol., 1957, 70, 38-48.
Hines, D., Satz, P., Schell, B., and Schmidlin, Sara Differential
recall of digits in the left and right visual half fields under
free and fixed order of recall. Neuropsychologia, 1969, 7, 13-22.
Hines, D. and Satz, P. Visual half field recall: Its relation to
varying rate, family history, and cerebral dominance. Neuropsychologia,
1970, in press.
Hubel, D. H. The visual cortex of the brain. Scientific American, 1963,
Kimura, Doreen Right temporal-lobe damage. Arch. Neurol., 1963, 8,
Kimura, Doreen Dual functional asymmetry of the brain in visual perception.
Neuiopsychologi, 1966, 4, 275-285.
Kimura, Doreen Functional asymretry of the brain in dichotic listening.
Cortex, 1967, 3, 163-178.
Kinsbourne, M. The cerebral basis of lateral asymmetries in attention.
Unpublished manuscript, 1970.
Mello, Nancy. Interhemispheric reversal of mirror-image oblique lines
after monocular training in pigeons. Science, 1965, 148, 252-254.
Milner, B. Brain mechanisms suggested by studies of temporal lobes. In
F. L. Darley (Ed.), Brain Mechanisms Underlying Speech and Language.
New York: Grune and Stratton, 1967.
Noble, J. Paradoxical interocular transfer of mirror-image discrimination
in the optic chiasm sectioned monkey. Brain Research, 1968, 10,
Orbach, J. Differential recognition of Hebrew and English words in right
and left visual fields as a function of cerebral dominance and
reading habits. Neuropsychologia, 1967, 5, 127-139.
Parsons, O. Clinical Neuropsychology in Speilberber, Charles (Ed.).
Current Developments in Community and Clinical Psychology,
Charles Thomas, 1970, in press.
Satz, P. A block rotation tasks the application of multivariate and
decision theory analysis for the prediction of organic brain
disorder. Psychol. Mon., 1966, 80, Whole No. 629.
Satz, P. Laterality effects in dichotic listening. Nature, 1968, 218,
Subirana, A. The prognosis in aphasia in relation to cerebral dominance
and handedness. Brain, 1958, 81, 415-425.
Teuber, H. L. Effects of brain wounds implicating right or left
hemisphere in man: Hemispheric differences and hemisphere inter-
action in vision, audition, and somesthesis. in V. B. Mountcastle
(Ed.), Interhemispheric Relations and Cerebral Dominance,
Baltimore: John Hopkins Press, 1962, 131-157.
Van Nostrand, G. Dual functional asymmetry of the brain in visual
perception: A reexamination. U. of Florida, Masters Thesis, 1970.
White, M. J. Laterality differences in perception: A review. Psychol.
Bull., 1969, 72, 387-405.
Winer, W. L. Statistical Principles in Experimental Design. McGraw-
Hill Series in Psychology, N. Y., 1962.
Wyke, M., and Ettlinger, G. Efficiency of recognition in left and
right visual fields. Arch. Neurol., 1961, 5, 659-665.
Zangwill, O. L. Cerebral Dominance and its Relation to Psychological
Function. Oliver and Boyd, London, 1960.
Bruce John Schell was born August 6, 1943, in El Paso, Texas.
He graduated from Las Cruces High School In Las Cruces, New Mexico,
in June, 1961. He obtained a Bachelor of Arts degree, majoring in
Psychology, in June, 1966, from New Mexico State University. In June,
1968, he received a Master of Arts degree, with a major in Clinical
Psychology, from the University of Florida.
Bruce Schell is married to the former Marcy Myers and has one
son, Eric Kendall.
This dissertation was prepared under the direction of the chairman
of the candidate's supervisory committee and has been approved by all
members of that committee. It was submitted to the Dean of the College
of Arts and Sciences and to the Graduate Council, and was approved as
partial fulfillment of the requirements for the degree of Doctor of
Dean, College of Art and Sciences
Dean, Graduate School