|
Differences
in Language Asymmetries Between Left- and Right-Hander's Brains: Jeremiah J.
House Advisor: Dr. Christine Chiarello University of California, Riverside
Abstract Visual
half-field tachistoscopic measures were used to evaluate differences
between left-handers and right-handers in the effects of case alternated
stimuli and lowercase stimuli on lexical decision asymmetries. Twenty-four
left-handed and 24 right-handed native-English speakers with normal or
corrected-to-normal vision participated in a 2x2x2x2 (Handedness x
Lexicality x Case Type x Visual field) mixed-factorial design lexical
decision task study. Case alternation had a greater effect in the RVF than
in the LVF. Case alternation was found to affect word recognition much
more than nonword recognition for both left- and right-handers due to a
whole-word processing mechanism. Left-handers did not differ from
right-handers in VF asymmetries. This suggests that left-handers may not
be as different from right-handers with respect to language lateralization
as previously thought. Differences in Language Asymmetries Between Left- and Right-Hander's Brains: A Cognitive Neuroscience Approach Left-handers may live shorter lives because they experience more accidents
resulting from using tools, equipment, and technology designed for
right-handers (Coren & Halpern, 1991). Not only do left-handers differ
from right-handers in how accommodating society is to them, but they also
differ on cognitive task performance. One of the most striking differences
between left- and right-handers' brains is the lateralization of language.
It has been claimed that if a certain functional asymmetry (defined as the
anatomical hemisphere of the brain in which a certain function, like
language, is localized) is observed in right-handers, the corresponding
function in left-handers will be less lateralized and possibly lateralized
in the other direction (Springer & Deutsch, 1998). In other words,
language localized to the left hemisphere (LH) for right-handers would be
less localized to the LH, or even dominant in the right hemisphere (RH),
for left-handers. Specifically,
among those who exhibited some RH language during a sodium amobarbital
experiment, a significantly higher proportion were non right-handers. This
suggests that left- and mixed-handedness may be indicative of mixed
laterality. The study reported that 95 percent of right-handers and 70
percent of left-handers have speech localized to the LH (Loring et al.,
1990). Other studies reveal that
left-handers have even more language abilities in the RH. Prognosis for
recovery from aphasia (due to a stroke) is much better for left-handers
than for right-handers (Subirana, 1958; Zangwill, 1960). Behavioral studies (i.e., tachistoscopic visual and dichotic listening)
have also found left-handers to be less asymmetric for language than
right-handers (Peters, 1995). In visual half-field tachistoscope tasks,
the visual stimulus is rapidly presented to either the left or right
visual field and travels from the retina, via the ocular nerve, to other
nerves in the form of a chemical message, until it reaches its final
destination in the hemisphere opposite the visual field to which it was
presented. In sum, evidence suggests that language is less lateralized in
left-handers. However, less data exists surrounding lateralization of
language on left-handers as compared to right-handers (Peters, 1995).
Additional research is needed. Degree of Handedness The handedness variable is measured differently across studies (Springer & Deutsch, 1998). Some experimenters rely on self-report. Still others use the raw handedness score, calculated from one of many handedness inventories, whereas others use cut-off handedness scores to establish two handedness groups. This process is subjective and results in varying groups across studies. A key issue in the study of handedness and language is the argument that language laterality should not be considered to be a discrete variable when handedness is a continuous variable (Loring et al., 1990). Word Specific Visual Pattern Some studies suggest that each hemisphere uses a somewhat different
strategy when recognizing a word (Lavidor & Ellis, 2001; Zaidel et
al., 1999). Generally, the LH is able to recognize words in their whole
form, as if they were a basic unit unto themselves, whereas the RH
recognizes a word in more of a letter-by-letter fashion (Lavidor &
Ellis, 2001). However, these same length effects were shown in both
hemispheres when the words were case alternated (Lavidor & Ellis,
2001). Besner and Johnston
(1989) suggest the existence of a word specific visual pattern (WSVP)
defined as the complete set of visual characteristics in a word. They
propose a mechanism within the brain by which a person is able to
recognize or name a word when reading, simply by its unique set of visual
features. Critical
to the hypothesis of a WSVP is that word processing is more disrupted by
case alternation than nonword processing. If case alternation interferes
with the use of WSVPs, then only words should be affected. Nonwords, which
by definition have no WSVP, should show little or no effect of case
alternation. However, in the case of a word, what would be quickly
recognized in upper or lowercase via WSVP could not turn up as a word when
case alternated, because under normal conditions, the word is not case
alternated. Therefore, case alternation changes the usual set of visual
features of a word, making it unrecognizable using WSVP. For example, a person sees the word "purple" and takes 500
milliseconds to identify it as a word by searching the mental lexicon for
a match to its visual features. Then, the person sees "pUrPlE"
and takes 1 second to respond �word� this time. The delay occurs
because letter-by-letter identification was necessary after WSVP found no
match. This is a relatively large 500 millisecond difference in response
latency. Then, suppose the person sees "scerds". Again, it takes
1 second to respond �nonword," because letter-by-letter
identification was necessary after WSVP found no match. Finally, the
person sees "sCeRdS," and it takes 1 second for a response
because letter-by-letter identification is needed. However, this time the
response is �nonword� and the response latency difference between
"scerds" and "sCeRdS" is 0 seconds. Therefore, Besner and Johnston (1989) theorized that the WSVP hypothesis
is supported when case alternation affects words more than nonwords. In a
lexical decision task experiment, word processing was more disrupted
(i.e., accuracy and latency) by case alternation than nonword processing
(Allen, Wallace, & Weber, 1995; Besner, 1983; Besner & McCann,
1987). Other researchers suggest that even though some variation of WSVP may exist, it cannot be necessary to successful reading because subjects are still able to read visually distorted (i.e., case alternated) words (Coltheart & Freeman, 1974). It might be the case that as one reads lowercase or uppercase words, one relies on letter identification in conjunction with WSVPs (Besner & Johnston, 1989). Current Study Reliable
data on lateralization of language in left-handers are needed.
Furthermore, whole-word processing theories need to apply left-handed
data. Differences in the lateralization of language between left- and
right-handers were predicted. The experiment utilized a lexical decision
task to examine the effects of case alternation on word recognition in
each hemisphere across left- and right-handers. The following predictions
were examined: 1.
For right-handers, the RVF/LH (right visual field/left hemisphere)
will react more quickly and perform more accurately on a tachistoscopic
lexical decision task than the LVF/RH (left visual field/right
hemisphere). 2.
For right-handers, the RVF/LH will show greater decrement in
accuracy and reaction time as a result of case alternation when compared
to the LVF/RH. 3.
Left-handers will show a reduced RVF/LH advantage for the lexical
decision of lowercase words/nonwords compared to right-handers. 4.
As the degree of left-handedness increases, the RVF/LH advantage
for lexical decision of lowercase words/nonwords will decrease. 5.
Case alternation will affect word recognition much more than
nonword recognition in the RVF/LH for both left- and right-handers. 6.
For left-handers, case alternation will also affect words more than
nonwords in the LVF/RH, due to more of a WSVP in the RH for left-handers. Although Levy and Reid (1976) proposed that handwriting posture could
provide information about cerebral laterality, later research has not been
supportive (Springer & Deutsch, 1998). For this reason, we did not
evaluate our participants� handwriting posture. Method Design The experiment consisted of a lexical decision task in
a 2x2x2x2 mixed-factorial design (lexicality x visual field x case type x
handedness). Handedness was included as a between-subjects variable. The
dependent variables were reaction time and accuracy. Participants Twenty-four right-handed (12 male, 12 female) and 24 left-handed (12 male,
12 female) participants participated. All were native English speakers and
had normal or corrected to normal vision. The participant�s handedness
was evaluated with a modified version of the Edinburgh handedness
inventory containing five items (Bryden, 1982). Participants with scores
of + .30 or greater were considered right-handed (M = +.8), and
participants with scores less than 0 were considered left-handed (M =
-.6). Participants either received course credit or were paid seven
dollars. Stimuli The
experiment consisted of 120 words and 120 nonwords. The words were equated
across lists for mean familiarity, imageability, nouniness, and length
(4-6 letters) (Chiarello, Shears, & Lund, 1999). The familiarity
ratings were from unpublished norms collected in Dr. Chiarello's
laboratory. Nonword stimuli were created by changing one letter in a real word that was comparable to the word stimuli (i.e., highly imageable 4-6 letter nouns). However, the nonword had to be pronounceable and orthographically legal. Each letter position was changed equally often in creating the nonwords. The
stimuli were counterbalanced across conditions and horizontally presented
in lowercase or case alternated (e.g. cAsE aLtErNaTeD) form using Times
New Roman 26 point font. Stimuli ranged in length from 1.5-3.1 cm.
Uppercase and lowercase letters had heights of .6 cm and .4 cm,
respectively. Apparatus
and Procedure Participants were seated 60 cm in front of a Power
Macintosh computer with a 17-inch monitor. The computer presented stimuli
and recorded responses with the PsyScope program (Cohen et al., 1993). An
adjustable headrest was used to stabilize head position.
A practice set of 35 words and 35 nonwords was given before the
first block. There were a total of 4 blocks consisting of 60 trials each.
Rest periods between blocks counteracted fatigue effects. The order of
trials was independently randomized for each subject. The medial edge of
each stimulus was positioned 1.81 degrees eccentric to a central
"+". This "+" served as the fixation point. A 300
millisecond (ms), centrally displayed "+" followed by a 50 ms
flickering red "+" preceded each trial. When the stimulus
appeared to either the LVF or RVF, it remained for 130 ms. Meanwhile, the
"+" remained for 325 ms after stimulus onset. The next trial was
initiated 1300 ms after the participant's response or after the timeout
interval (3000 ms) elapsed.
Participants were instructed to keep their head in the headrest, to
maintain their focus on the central "+" as long as it remained
visible, and to respond as quickly as possible. Right-handed participants
were told to use their right hand to press the designated word
"." or nonword "0" keyboard button on the number pad.
Left-handers were told to use their left hand. In both cases, one-half of
the subjects used their index finger to press the "nonword"
button and the middle finger to press the "word" button. The
remaining one-half used the opposite configuration. The PsyScope program
recorded reaction time, measured from stimulus onset, and accuracy. Results Analyses
of variance (ANOVAs) were performed on the data for both reaction time and
accuracy (% correct), examining the variables VF, Lexicality, Case Type,
and Handedness. Mean accuracy values
for left-handers and right-handers are shown in Table 1. Mean reaction
times values for left-handers and right-handers are shown in Table 2. In
these tables the mean accuracy and reaction times for each of the
conditions (e.g., case alternated word in LVF) were calculated for all
participants within each handedness group.
There was no significant main effect for handedness for either
accuracy, F (1, 46) = 2.58, p
< .1148, or reaction time, F
< 1. Main effects for case were significant for both accuracy, F
(1, 46) = 206.11, p < .0001, and reaction time, F (1, 46) = 98.13, p <
.0001; responses were quicker (914 ms) and more accurate (80.9%) for
lowercase stimuli than for case alternated stimuli (reaction time = 999
ms; accuracy = 70.6%). Significant main effects for VF were found for both
accuracy, F (1, 46) = 20.59, p < .0001, and reaction time, F (1, 46) = 14.00, p <
.0005; RVF responses were made more quickly (938 ms) and accurately
(78.1%) than LVF responses (reaction time = 975 ms; accuracy = 73.4%).
There was a significant main effect for lexicality under the reaction time
measure, F (1, 46) = 152.37, p < .0001; reaction time was quicker for words (859 ms) than for
nonwords (1054 ms).
A Handedness x Lexicality interaction was found on the accuracy
measure. Specifically, there was a difference between left- and
right-handers for nonwords, F (1, 46) = 7.09, p <
.0106, but not for words, F <
1. Right-handers were more accurate (77.6%) than left-handers (72.2%) in
rejecting nonwords. There was a Case x Lexicality interaction for both accuracy, F
(1, 46) = 114.44, p < .0001, and reaction time, F (1, 46) = 26.81, p <
.0001; words were more affected by case alternation than were nonwords
(see Figures 2 and 3). There was a main effect of case for words for both
accuracy, F (1, 46) = 249.09, p < .0001, and reaction time, F (1, 46) = 120.01, p <
.0001; responses were quicker (795 ms) and more accurate (88.3%) for
lowercase words than for case alternated words (reaction time = 922 ms;
accuracy = 64.8%). A significant main effect of case was also found for
nonwords for both accuracy, F
(1, 46) = 4.25, p < .0450,
and reaction time, F (1, 46) =
13.53, p < .0006; responses
were quicker (1033 ms) for lowercase nonwords than for case alternated
nonwords (1076 ms). However, responses were more accurate (76.3%) for
case alternated nonwords than for lowercase nonwords (73.5%). There was a significant interaction between case and visual field for reaction time, F (1, 46) = 7.86, p < .0074; as predicted, the RVF had greater decrement in reaction time for case alternated stimuli than did the LVF (see Figure 4). A main effect of VF for lowercase stimuli was found for reaction time, F (1, 46) = 17.67, p < .0001; responses for lowercase stimuli were quicker (886 ms) in the RVF than in the LVF (942 ms). A main effect of VF for case alternated stimuli was marginally significant for reaction time, F (1, 46) = 3.51, p < .0672; responses were quicker (989 ms) for case alternated stimuli in the RVF than for case alternated stimuli in the LVF (1009 ms). A significant interaction between case and VF was also found for accuracy, F (1, 46) = 12.63, p < .0009. There was an accuracy difference between the LVF and the RVF for lowercase stimuli, F (1, 46) = 41.45, p < .0001, but not for case alternated stimuli, F (1, 46) = 2.38, p < .1297. The RVF accuracy advantage (84.6%) over the LVF (77.2%) was significant for lowercase stimuli but not for case alternated stimuli (RVF = 71.7%; LVF = 69.5%)(see Figure 5). No other interactions were statistically significant. In addition, reaction time and accuracy laterality coefficients were calculated as follows: the reaction time formula was (LVF-RVF)/(LVF+RVF) and the accuracy formula was (RVF-LVF)/(LVF+RVF), such that a positive value would indicate a RVF/LH advantage. These coefficients were used to compute correlations with the handedness index scores. No statistically significant correlation was found between handedness index score and degree of VF laterality. Discussion
Handedness was not as closely related to differences in hemispheric
asymmetries for language as prior studies have reported (Cohen &
Freeman, 1978; Peters, 1995; Springer & Deutsch, 1998; Zangwill,
1960). In addition, right-handers did not have an overall accuracy and
reaction time advantage over left-handers, which is inconsistent with
prior studies showing right-handers� advantage on language tasks (Cohen
& Freeman, 1978). Given this finding, it is likely that another
variable operated on the data. Many
studies have included ambidextrous participants in their left-handed
group, whereas ours did not. Perhaps the inclusion of the ambidextrous
group inaccurately represents language asymmetry differences between
right-handers and true
left-handers. The degree of handedness was not significantly correlated with language
asymmetries. Handedness was treated as a continuous variable (Loring et
al., 1990). The lack of significant correlation may indicate that there
was not enough participants in each handedness index score group for
sufficient power.
A RVF/LH advantage was found for the lexical decision task for both
left- and right-handers, confirming the well-established idea that the RVF/LH
is dominant for language. For accuracy on the Case x VF interaction, the
RVF/LH advantage was for lowercase stimuli. This pattern seems to suggest
that case alternated stimuli effectively disrupt the processing mechanism
that the LH employs. Otherwise, the advantage should be shown for case
alternated stimuli, as well.
Across visual fields the RVF showed greater decrement in reaction
time for case alternated stimuli with respect to lowercase stimuli than
did the LVF for both left- and right-handers. The hypothesis regarding
this result was supported by the data with the exception that the effect
was shown for both right- and left-handers. Left-handers usually tend to
show the same functional asymmetries as right-handers, but to a lesser
degree (Springer & Deutsch, 1998).
For accuracy, the greater RVF/LH advantage was only for lowercase
stimuli. These results suggest that the LH is able to recognize words in
their whole form, whereas the RH recognizes words in more of a
letter-by-letter fashion (Lavidor & Ellis, 2001). The discussion of
the next prediction lends further support to case alternation affecting
words more than nonwords. Thus, further evidence suggests global
processing of words.
Consistent with the fifth prediction case alternation did affect
word recognition more than nonword recognition. However, contrary to the
prediction was the result that the effect occurred in both visual fields.
Because case alternation affecting words more than nonwords is an
indication of WSVP (Besner & Johnston, 1989), this processing
mechanism might exist in both hemispheres to some degree. This would be
inconsistent with a laterality study that found evidence for the global
pattern word recognition mechanism solely in the RVF/LH (Zaidel, Bloch,
& Arguin, 1999).
The
final prediction was not supported because case alternation affected words
more than nonwords across both VFs for both left- and right-handers. It is
likely that if a WSVP does exist primarily in the LH for right-handers,
that it would be less lateralized to the LH for left-handers. However,
given our results that support whole word processing in both hemispheres
for right-handers, it would be unlikely that left-handers would show a
different pattern because the right-handers are really not
showing a VF asymmetry along this dimension.
Future Studies Future
studies might further explore language processing in the LH for
left-handers. It would also be interesting to evaluate language
asymmetries for ambidextrous participants. Indeed, these participants
could be responsible for many studies suggesting language asymmetry
differences between left- and right-handed participants because they have
often been included with the left-handed group. Further studies might also
evaluate whether degree of handedness is correlated with language
asymmetries by obtaining enough participants in each handedness index
score group to attain sufficient power. References Allen, P.A., Wallace, B., & Weber, T.A. (1995). Influence of case
type, word frequency, and exposure duration on visual word recognition. Journal
of Experimental Psychology: Human Perception and Performance, 21,
914-934. Besner, D. (1983). Basic decoding components in reading: Two dissociable
feature extraction processes. Canadian
Journal of Psychology, 37, 429-438. Besner, D., & Johnston, J.C. (1989). Reading and the mental lexicon:
On the uptake of visual information. In W. Marslen-Wilson (Ed.), Lexical
representation and process (pp. 291-316).
Cambridge, MA: MIT Press. Besner, D., & McCann, R.S. (1987). Word frequency and pattern
distortion in visual word identification and production: An examination of
four classes of models. In M. Coltheart (Ed.),
Attention and performance XII: The
psychology of reading (pp. 201-219). Bradshaw, J.L., & Nettleton, N.C. (1983). Human cerebral asymmetry. Englewood Cliffs, NJ: Prentice-Hall. Bryden, M.P. (1982). Laterality:
Functional asymmetry in the intact brain. New York:
Academic Press. Chiarello, C., Shears, C., & Lund, K. (1999). Imageability and
distributional typicality measures of nouns and verbs in contemporary
English. Behavioral Research
Methods, Instruments, & Computers, 31, 603-637. Cohen, G., & Freeman, R. (1978). Individual differences in reading
strategies in relation to cerebral asymmetry. In J. Requin (Ed.), Attention
and performance VII (pp. 411-426).
Hillsdale, NJ: Erlbaum. Cohen, J.D., MacWhinney, B., Flatt, M., & Provost, J. (1993). PsyScope:
A new graphic
interactive environment for designing psychology experiments. Behavioral Research
Methods, Instruments, &
Computers, 25,
257-271. Coltheart, M., & Freeman, R. (1974). Case alternation impairs word
identification. Bulletin of the
Psychonomic Society, 3, 102-104. Coren, S., & Halpern, D.F. (1991). Left handedness-A marker for
decreased survival fitness. Psychological
Bulletin, 109, 90-106. Lavidor, M., & Ellis, A.W. (2001). Mixed-case effects in lateralized
word recognition. Brain and
Cognition, 46, 192-195. Levy, J., & Reid, M. (1976). Variations in writing posture and
cerebral orginazation. Science, 194, 33. Loring, D.W., Meador, K.J., Lee, G.P., Murro, A.M., Smith, J.R., Flanigin,
H.F., Gallagher, B.B., & King, D.W. (1990). Cerebral language
lateralization: Evidence from intracarotid amobarbital testing. Neuropsychologia,
28, 831-838. Peters, M. (1995). Handedness and its relation to other indices of
cerebral larteralization. In R.J. Davidson & K. Hugdahl (Eds.), Brain
asymmetry (pp. 195-214). Cambridge, MA: MIT
Press. Springer, S., & Deutsch, G. (1998). Left brain right brain: Perspectives from cognitive
neuroscience (5th ed.). New York:
W.H. Freeman and Company. Subirana, A. (1958). The prognosis in aphasia in relation to cerebral
dominance and handedness. Brain, 81,
415-425. Zaidel, E., Bloch, M., & Arguin, M. (1999). Pattern perception in
hemispheric word recognition. Brain
& Language, 69, 379-382. Zangwill, O.L. (1960). Cerebral dominance and its relation to psychological function. Edinburgh: Oliver and Boyd. Table 1 Accuracy
Means (Standard Deviations) for Left- and Right-Handers _____________________________________________________________________________________________ Left-Handers
(N=24) _____________________________________________________________________________________________
Words
Nonwords
______________________
_________________________
Visual
Field
CA
LC
CA
LC _____________________________________________________________________________________________ LVF
63.4 (14.1)
83.1 (11.5)
72.7 (11.6)
67.1 (12.2) RVF 66.1 (15.2) 93.6 (5.5) 76.2 (8.1) 73.0 (10.3) _____________________________________________________________________________________________ Right-Handers
(N=24) _____________________________________________________________________________________________ LVF
63.7 (12.6)
84.2 (10.8)
78.2 (10.0)
74.4 (10.1) RVF
66.2 (14.0)
92.4 (7.8)
78.2 (7.5)
79.5 (9.7) _____________________________________________________________________________________________ Note.
CA= case alternated. LC= lowercase. Table 2 Reaction
Time Means (Standard Deviations) for Left- and Right-Handers _____________________________________________________________________________________________ Left-Handers
(N=24) _____________________________________________________________________________________________
Words
Nonwords
_______________________
__________________________
Visual
Field
CA
LC
CA
LC _____________________________________________________________________________________________ LVF
950 (215.9)
862 (246.0)
1113 (257.7)
1097 (246.1) RVF 915 (181.1) 778 (143.9) 1092 (247.2) 1048 (198.4) _____________________________________________________________________________________________ Right-Handers (N=24) _____________________________________________________________________________________________ LVF
927 (219.2)
792 (132.5)
1045 (257.1)
1016 (240) RVF
895 (196.1)
750 (158.6) 1055
(281.0)
970 (195.4) _____________________________________________________________________________________________ Note.
CA= case alternated. LC= lowercase. Figure Captions Figure 1. Case x Lexicality accuracy interaction. CA, case alternation; LC, lowercase. W, word; NW, nonword. Figure 2. Case x Lexicality reaction time interaction. Figure 3. Case x Visual Field reaction time interaction. LVF, left visual field; RVF, right visual field. Figure 4. Case x Visual Field accuracy interaction. Figure 1
Figure 2
Figure 3
Figure 4
Copyright 2002 by the Undergraduate Psychology Journal
|
