TMP 948 B

This article was downloaded by: [TÜBTAK EKUAL]
On: 7 August 2009

Access details: Access Details: [subscription number 772814176]
Publisher Informa Healthcare
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
International Journal of Neuroscience

Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713644851
The Distribution of Hand Preference in Normal Men and Women

Üner Tan a
a
Ataturk University, Medical Faculty, Institute of Physiology, Erzurum, Turkey
Online Publication Date: 01 July 1988
To cite this Article Tan, Üner(1988)'The Distribution of Hand Preference in Normal Men and Women',International Journal of
Neuroscience,41:1,35 — 55
To link to this Article: DOI: 10.3109/00207458808985740
URL: http://dx.doi.org/10.3109/00207458808985740
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf
This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
Inrern. J. .Neuroscience. 191i8, Vol. 41, pp. 35-55 0 1988 Gordon and Breach, Science Publishers. Inc.
Repnnts available directly from the publisher Printed in the United Kingdom
Photocopying permitted b!i license only
THE DISTRIBUTION OF HAND PREFERENCE IN

NORMAL MEN AND WOMEN
m E R TAN
Ataturk Uni.versity, Medical Faculty, Institute of Physiology, Erzurum, Turkey
(Received October IS, 1987)
The distributions 01 the hand preferences (Geschwind scores) were studied in men and women. The
incidences of the right-, mixed-, and left-handers were 66.1, 30.5, and 3.4%, respectively. The left- and
right-handers (LHs, RHs) were represented by two bars located at the opposite ends of a J-shaped
histogram for the total sample, men, and women. The distributions for the total RHs, and male RHs were
not J-shaped, but negatively skewed, the mode being close to the mean. The frequencies for the females
increased linearly. The distributions for the LHs did not significantly deviate from normality. Except for
Downloaded By: [TÜBTAK EKUAL] At: 12:59 7 August 2009
the LHs, the percentage distributions of the Geschwind scores were J-shaped. The LHs were evenly
distributed by chancz to be expected in a random sample from a rectangular population. The incidence of
weak right-handedness was significantly higher in men than women. The total RHs were more lateralized
than the total LHs w -10showed a tendency toward weak left-handedness. In males, the incidence for weak
left-handedness was significantly higher than that for strong left-handedness. The opposite was
established for females. The probability of the consistent right-handedness was significantly higher in
women than men. The LHs were less lateralized, and more widely dispersed than the RHs. The statistical
analysis of the distributions indicated that females tended to be more lateralized, i.e., more right-handed,
and more left-handed. than males.
Keywords: handedmw, hand preference, manual asymmetry, lateralization, hand, man
Hand preference can be defined as the tendency to perform several tasks with one
hand rather than the other. To assess hand preference, several questionnaires were
proposed by several authors (Humphrey, 1951; Annett, 1970; Oldfield, 1971;
Raczkowski et ad., 1974; Porac & Coren, 1981; Beukelaar & Kronenberg, 1983).
The questionnaire chosen was considered to be of importance, since it was
believed that the exact measurement of hand preference would be essential to
understand the nature of handedness. In these studies, the main interest was
concentrated on the distribution of the hand preference. Consequently, two
approaches emerged about the distribution of the hand preference: the discrete and
continuity arguments. The supporters of the first argument regarded hand
preference as a natural dichotomy, which suggested that there are two groups in the
population as right- and left-handers (Beukelaar & Kroonenbers, 1983; McManus,
1984).The continuity argument suggested that the distribution of hand preference
should be regarded as continuous between pure right- and left-hand preferences
(Gillieset al., 1960; Annett, 1970; Oldfield, 1971).
The continuity hypothesis inspired the emergence of a new theory about the
nature of handedness: the right-shift theory (Annett, 1972). This theory arose from
the observation that the left-right differences between the hands in peg moving
times were distributed in a unimodal and roughly normal manner. That is,
handedness is distributed essentially by chance in the hand and paw preferences of
primates and other mammals, but this normal distribution moves along the abscissa
such that the mean to the right of zero under the influence of a factor (rs +gene),
This study was supported by the Turkish Scientific and Technical Council (Project Nr. TAG-575).
35
36 U. TAN
which facilitates the development of speech in the left hemisphere, and incidentally,
gives a slight advantage to the right hand. On the other hand, Annett (1976) has
reported that there was a close correlation between hand preference and hand skill.
Consequently, she argued that hand preference depends on a continuum of
differences between the hands in skill. The continuous variability of the hand
preference was the most important standpoint of the right shift theory. Therefore,
the distribution of the hand preference was systematically investigated in the
present study. Moreover, an attempt was made to elucidate the sex differences in
hand preference, since there are controversial opinions in the literature on this
subject. The right shift theory presumes that the expression of the rs+ gene could
be more effective in right-handed females than males, but reduced or absent in left-
handers (Annett, 1973, 1979). To assess hand preference, the Edinburgh
Handedness Questionnaire (Oldfield, 1971) modified by Geschwind & Behan
( 1982) was used. Following Geschwind’s suggestion (personal communication), the
laterality score was calculated instead of a laterality quotient. In Norman
Geschwind’s memory, this laterality score was called the “Geschwind score.”
METHODS
Subjects. The subjects comprised about 750 mean and 350 women who ranged in
age from 20 to 22 years (students in the faculties of nursery, dentistry, and
medicine). All were healthy, devoid of neurological or psychiatric signs and
symptoms. They did not object to participation in the study, i.e., a volunteer bias
was avoided. The subjects originated from different cities, and had all ranges of
intelligence quotients, constituting a random sample.
Hand preference. To assess their hand preference, all subjects received a Turkish
adaptation of Oldfield’s questionnaire (Oldfield, 1971)modified by Geschwind and
Behan (1982).The questions pertained to which hand was used by the subject for
writing, throwing, scissors, toothbrush, knife (without fork), spoon, holding the
handle for a shovel, striking a match, and twisting off the lid of a jar. The columns
“always right,” “usually right,” “either hand,,’ “always left,” “usually left” were scored
as + 10, + 5 , 0, - 10, and - 5, respectively. Following Geschwind’s suggestion
(personal communication), the laterality score was taken as the sum of all these
scores, and no quotient was calculated. In memory of Norman Geschwind, this
laterality score was called the “Geschwind score.” A score of - 100 indicated that
the subject responded “always left” on all items, and a score of + 100 indicated
“always right” on all items.
Data analysis. The distribution of the Geschwind scores (GSs) was analyzed by
frequency histograms constructed by a program, which computed the descriptive
statistics, and compared the distribution of the observed frequencies with the
normal frequency, displayed the distribution as a histogram, and assessed its
skewness and kurtosis. The histogram actually showed the distribution of the
standard deviations about the mean GS. The program first counted the number of
scores less than two standard deviations (S.D.) below the mean, and then the
frequency lying between 2 S.D and 1.5 S.D.below the mean, and so on. In all, there
were 10 compartments in the histogram; the width of each block was 0.5 standard
deviation. The percentage distributions of the GSs were also analyzed.
Distribution-free techniques (nonparametric statistics) were usually used for the
statistical tests and inferences. The arbitrary thresholds for the incidences of the
DISTRIBUTION OF HAND PREFERENCE 37
left-, right-, and mixed-handed subjects were drawn on the normal distribution
curves by looking at the table of the normal distribution function.
RESULTS
Total sample. 'The distribution of the Geschwind scores (GSs) in the total sample
( N = 1057) is illustrated by a bar graph in Figure 1, which actually shows the
distribution of the standard deviations about the mean. The observed and expected
frequencies differed from each other highly significantly. In this histogram, an
approximately J-shaped distribution is recognized. By a close inspection, three
subgroups can bme distinguished in this histogram: strong right-handers represented
by two bars at 0.5 and 1.0 S.D., left-handers represented by a single bar at
- 2.5 S.D., and mixed-handers (the bars between right-, and left-handers). The
incidences of the strong right-handers, mixed-handers, and left-handers described
above were 66.3., 28.4, and 5.5%, respectively. The incidence of the left-handers
having GSs from -35 to -90 was 3.4%. Then the proportions for the right-
handers, mixed-handers, and left-handers become 66.1, 30.5, and 3.4%,
respectively. Thiese subgroups occurred in binominal proportions. Considering
hand preference as normally distributed, two thresholds could be found by
referring to the table of the normal distribution function. For an incidence of 3.4%
for the left-handers the threshold was found to be - 1.8 S.D. from the mean of the
total sample, i.e., the division between the left-, and mixed-handers. The threshold
400 I
1 I
0
STANGARG DE'AATIGNS
FIGURE 1 Distribu.tion of Geschwind scores in the total sample ( N = 1057). Ordinate: number of
subjects (frequency). Abscissa: standard deviations about the mean (0.5 S.D.= 1 bar). The standard
deviations corresponding to each bar are shown on the right side of the histogram Chi-squared with
7 d.f.='656.2; median = 80; intequartile range = 60 and 90; range = - 90 to 100; skewness = - 2.3
(S.D.=O.O8,p<.Ol);k.urtosis=9.3 (shouldbe 3).
U. I A N
5@
&$j
-100
-a0
-60
5 -40
- 20
2o
!El 4o
60
El Bo
100
0.
SESCHWII4D SCORES
FIGURE 2 Percentage distribution of Geschwind scores for the total sample ( N = 1057). Ordinate:
percentage of the subjects. Abscissa: Geschwind scores; class size = 20; class sizes corresponding to each
bar (upper limits) are shown on the right of the histogram (negative scores to the left of the “0” line).
for the strong right-handers, i.e., the division between the right-, and mixed-
handers, was at - 0.4 S.D. from the mean. The important point to notice here is
that these proportions and their thresholds were obtained and calculated by using
the histogram in Figure 1.
The consistent right-handers (group 1 ) were represented by the bar on the far
right of the histogram (1.0 S.D.).They comprised 32.4% of the total sample (GSs
from 90 to 100). The sum of the bars at -2.5, -2.0, -1.5, and -l.OS.D.
comprised 17.88% (group II) of the total sample (GSs from - 90 to + 35). The
remaining subjects (group III; GSs from 35 to 85) represented by the bars at - 0.5
and + 0.5 S.D.,comprised 49.52% of the total sample.
Figure 2 illustrates the percentage distribution of the GSs in the total sample
( N = 1057). Of these subjects, 5.5% exhibited some degree of left-handedness
(GSs < 0), and the remaining 94.5% was due to the subjects having various degrees
of right-hand preference (GSs>O). The percentage values of the GSs for the left-
handers were evenly distributed, exhibiting merely chance variations to be expected
in a random sample from a rectangular population ( x 2 with 4d.f.=0.33,
.99 > p > .98). By contrast, the percentage values of the GSs for the subjects with
right-hand preference increased following a J-shaped distribution.
Total men and women. Figure 3 shows the distributions of the GSs in the male
(above) and female (below) subjects, which appeared to be similar in shape. In men
( N = 704), and women ( N = 365), the observed frequencies deviated from the
expected ones highly significantly, and the distributions had significant negative
40 U. TAN
skewness. The left-handers (GSs between - 10 and - 100) constituted a distinct

group on the far left of both histograms, and the right-handers were concentrated
to the right of zero standard deviation. The mode for the right-handers was at
0.5 S.D. in men (GSs 85 to 90),and 1.0 S.D. in women (GS = 100).
The incidence of the strong right-handers (to the right of 0 S.D.) was 64.2% for
men (GSs from 85 to 100). The women having the same GSs comprised 70.1% of
their total sample. The incidences of the left- (GSs between - 35 and - 90), and
mixed-handers were 3.0 and 32.8% for men, and 3.8 and 26.1% for women. The
threshold for the strong right-handedness was - 0.36, and - 0.54 from the mean
for men and women, respectively. The threshold for the left-handedness was - 1.8
and - 1.76 from the mean for men and women, respectively. The incidence of the
consistent right-handedness (bars at 1.0 S.D in men and women) for women
(38.1%)was found to be significantly higher than that for men (29.1%; chi-squared
with 1 d.f. = 9.25, p < .005).
The incidence of the consistent right-handers (group I) was 29.1%, and 38.1% in
the male and female subjects, respectively. These subjects were represented by the
bars at 1.0 S.D. in both histograms (above and below). The sum of the subjects
represented by 4 bars at - 2.5, - 2.0, - 1.5, and - 1.0 S.D. (group 11) comprised
18.75%, and 16.3% of the male and female samples, respectively. Group I11
represented by the bars at - 0.5 and +0.5 S.D. comprised 52.15, and 45.6% of the
male and female samples, respectively.
Figure 4 illustrates the percentage distributions of the GSs for the total male
(above) and female (below) subjects. The subjects with left-hand preference
constituted 5.1%, and 5.2% of the total male and female subjects, respectivley. The
chi-squared one-sample test showed that the percentages of the GSs for the male
left-handers were evenly distributed by chance as expected in a random sample
from a rectangular population (chi-squared with 4 d.f. = 1.24; .90 > p > 30). The
same result was also obtained for the female left-handers (chi-squared with
4 d.f.= 0.67; .98 > P> .95). By contrast, the percentage distributions of the GSs for
the right-handers (to the right of zero) exhibited nearly a linear increase. Of the male
subjects, 40.6% had the highest GSs ranging from 85 to 100 (strong right-handers),
whereas 47.1% of the female subjects exhibited strong right-handedness (GSs from
85 to 100). The difference between these proportions was found to be statistically
significant (chi-squared with 1 d.f. = 4.61; p < .03). Of the remaining scores, 47.6"h
comprised the women, and 54.1% the men. The difference between these
proportions was found to be statistically significant (chi-squared with 1 d.f. = 3.58;
p < .05).
Right-handed men and women. The histograms in Figure 5 illustrate the
distributions of the GSs in the right-handed men ( N = 656) and women ( N = 347).
The goodness-of-fit test for normality indicated that the observed and expected
frequencies for both samples deviated from each other highly significantly. These
histograms were unimodal, but not J-shaped, and had significant negative skewness
(p<.001).The mean (75), median (80), and mode (85) did not coincide in the
histogram for men, confirming the negative skewness, and deviation from
normality. In women, the mean (go), median (80), and mode (100) also did not
coincide, confirming the negative skewness, and deviation from the expected
normal distribution. However, the mode in the histogram for men is seen at 0.5
standard deviation (GS= 65), whereas the mode for women is seen at 1.5 S.D.
(GS = 100).In the sample for men, 45.6% had GSs below the mean (relativelyweak
right-handers); 54.4% had GSs above the mean (relatively strong right-handers).
The number of the strong right-handers significantly exceeded the weak right-
handers ( z = 2.26; p < .01). In women, there were 42.1% weak, and 57.9% strong
right-handers. ‘The number of the strong right-handers was found to exceed the
weak right-handers highly significantly ( z = 2.89; p < .001). The difference between
the probabilities of occurrences for the weak right-handedness in the male and
female subjecta was not found to be significant (chi-squared with 1 d.f. = 1.00;
P< .25). However, the incidence of the consistent right-handedness (GSs form 95
to 100, at 1.0 and 1.5 S.D.)was significantly higher in the female (40.1%) than the
male (30.9%) subjects (chi-squared with 1 d.f. = 8.8; P< .01), indicating a right shift
of handedness ~nwomen. By contrast, there was no significant difference between
the incidences of the extemely weak right-handers (GSs<35, at -2.0 and
- 2.5 S.D.)in the samples for men (8.5%),and women (8.9%).
Figure 6 illustrates the percentage distributions of the GSs for the right-handed
male (above) arid female (below) subjects. The distributions were clearly J-shaped.
Of 656 males, 85.5% had GSs above S O (relatively strong right-handedness); of
347 females, 88.5% had GSs within the same range. Consequently, there were
14.5% weak right-handed men, and 11.5% weak right-handed women (GSs < SO).
The incidences of strong and weak right-handedness in men and women did not
differ from each other significantly (chi-squared with 1df. = 1.96; 20 > p > .lo). On
the other hand, the strong right-handers (GSs from 85 to 100)constituted 40.1% of
the male subjects, and 49.6% of the female subjects. The proportion of the
consistent right-handers was found to be significantly higher in women than men
(chi-squared with 1 d.f. = 8.69; p < .005). The incidence of the nonconsistent right-
handedness was 59.9% for men, and 50.4% for women. The difference between
these proportions was found to be highly significant (chi-squared with 1 d.f. = 7.92;
p < .005), i.e., the incidence of nonconsistent right-handedness was significantly
higher in men than women.
Total Right- and Left-Handers. Figure 7 illustrates the distributions of the GSs in
the total right- (above), and left- (below) handed subjects. In the sample for the
right-handers (iV=999), the distribution of the GSs is not J-shaped, and not
bimodal. It is rather continuous and unimodal with a significant negative skewness,
the mode being at 0.5 S.D. (GS = 65). Statistical analysis indicated that the observed
frequencies deviated from the expected normal distribution highly significantly
(chi-squared with 7 df.= 209.2; p < .001).The median value (80)was very close to
the mean (76.4). Confirming the negative skewness, the number of the subjects
above the mean (relativelystrong right-handers: 56.1%) significantly exceeded that
(relativelyweak right-handers: 43.9%) below the mean (z= 3.8; p < .001).
In contrast to the right-handers, the observed and expected frequencies of the
GSs for the left-handers did not differ from each other significantly at the
previously set significance level (chi-squared with 7 d.f. = 12.74; p > .05), indicating
that the sample comes from a normal population. The skewness to the left was not
significant (0.25; p > .05). Thus, the distribution for the right-handers was not
normal, but was normal for the left-handers. The number of subjects below the
mean (relatively strong left-handers, 44.8%) did not significantly differ from that
above the mean (55.2%; relatively weak left-handers; z = 0.73, p > .05). Another
prominent difference between the distributions of the GSs for the right-, and left-
handers was that the coefficients of variation were 26.4 and 58.9 for the former
and the latter, respectively.
Figure 8 shows the percentage distributions of the GSs for the total right- and
left-handers. The histogram for the total right-handers (above) is clearly J-shaped.
The subjects having GSs between 5 and 50 (weak right-handers) comprised 13.1%
of the total sample. The strong right-handers GSs between 55 and 100 comprised
86.9% of the to1a1 sample. The percentage of the strong right-handers significantly
exceeded the weak right-handers (t=32.9, p<.OOl). In contrast to the right-
handers, the percentage distribution of the GSs for the left-handers (below)was not
J-shaped. The weakly lateralized left-handers having GSs between - 5 and - 50
comprised 64.0% of the total left-handers, and the strong left-handers having GSs
between - 5 5 and - 100 included the remaining 36.0% of the sample. The
difference between these percentages was found to be statistically significant
( f = 3.26, p < .005). It is also seen in the histogram for the left-handers that the
percentages of Ihe weak and strong left-handers are almost evenly distributed.
Indeed, the chi-squared one-sample test indicated that the differences among the
observed percentages of the GSs for the weak left-handers were merely chance
variations (chi-squared with 4 d.f. = 3.2; .70 > p > .50). The percentage distribution
of the GSs for the strong left-handers was also found to exert only chance
variations (.20> p > .lo).
Left-handed men and women. The histograms in Figure 9 illustrate the

distribution of the GSs for the male (above) and female (below) left-handers. The
statistical analysis indicated that the observed and expected frequencies agreed at
p = .05 level of significance. There was no significant skewness in both histograms.
‘The variances ailso did not differ from each other significantly ( p < . 0 5 ) .These
results indicate that the GSs in the male and female left-handers tended to be
normally distributed.
The interquartile range for the women was - 80 and - 35, and for the men - 60
and -20. The Mann-Whitney LI test indicated that the female subjects have
significantly more negative scores (means = - 54.3 with S.D.= 28.9) than the male
(mean= -41.6 with S.D.=25.4)subjects(z= 1 . 7 8 ; ~ < . 0 4 ) .
Figure 10 illustrates the percentage distributions of the GSs for the left-handed
men (above) and women (below). In the male subjects, the distribution was not J-
shaped. There were significantly fewer strong left-handers (29.7%; GSs from - 55
to -100) than the weak left-handers (70.3%; GSs from - 5 to -50; z=2.56
p < .005). In women, the incidences of the weak (48.0%),and strong (52.0%) left-
handers were almost equally distributed. However, the proportions of the GSs
within each class did not show merely chance variations ( p < 901). It would seem
that more data are necessary before any definite conclusion can be made. The weak
beft-handed males were equally distributed ( p > .99),as were the strong left-handed
rnales ( 3 0> p > ..30).The incidence of the male left-handers with relatively strong
left-hand preference was found to be significantly less than the incidence of the
strong left-handed females, and there were significantly more weak left-handed
males than the weak-left handed females (chi-squared with 1 d.f. = 3.92, p < .03).
IIISCUSSION
Zotal sample. The distribution of the standard deviations about the mean GS for
the total sample could be regarded as J-shaped (see Figure 1).However, the left-
handers were represented by a single bar at - 2.5 S.D. (GSs from - 20 to - 90).
This subgroup comprised 5.5% of the total sample, following which there was an
abrupt decline in the frequency of the subjects with very weak lateralization at
--2.0 S.D. (GSs from 0 to - 15);0.8%.On the opposite side of the histogram, the
50 U. TAN
consistent right-handers were represented again by a single bar at 1.0 standard

deviation. These subjects comprised 66.1% of the total sample. The remaining
28.4% of the subjects were represented by the bars between the left-, and right-
handers. Consequently, three subgroups could be distinguished in the total sample:
left-, right-, and mixed-handers.
Annett (1985) argued that the J-shaped distribution of hand preference could be
fitted to the normal distribution of hand skill. She suggested that the 4% consistent-
left, and 66% consistent-right humans show no variability of hand preference, but
30% mixed-handers show all the variability. In agreement with this assumption, the
left-, and right-handers were represented by single bars (0.5 S.D.),but the mixed-
handers by four bars (2.0 S.D.), indicating that the highest variation in hand
preference occurs in the mixed-handers. Using a questionnaire, Annett ( 1972) has
established that 66.8% of the subjects exhibited consistent right-hand, and 3.7%
consistent left-hand preference; the remaining 29.5% were mixed-handers. In the
present work, these proportions were found to be 66.1, 3.4, and 30.5 per cent for
the right-, left-, and mixed-handers, indicating no cultural difference in hand
preference between the English and Turkish people.

Concerning the sociocultural influences on hand-preference, it can be argued
that people are forced especially to eat and write always with the right hand, which
could be the cause for shifting of handedness to the right. As a result, the majority
of the population would be consistent right-handers, leaving many mixed-handers
behind, the recruiting some left-handers to the side of the mixed-handers. Then the
whole pattern would create a right-handed world. Letters from left-handers usually
mention the strong forces from teachers and parents to change from writing and
eating with the left hand to the right hand. However, a great majority of the left-
handers rejected these instructions and remained left-handed. In spite of these
sociocultural forces, the incidence of the hand preference were about the same in
my country as in the West. Collins (1975) has reported that the left-handed mice
denied being right-handed despite forces from the “right-handed world.” It was
recently established that the incidences for the right-, left-, and mixed-pawedness
were 57.1, 17.9, and 25.0%, respectively (Tan, 1987). These results suggest that
hand preference could be biological rather than sociocultural, not a by product of
the linguistic functions of the human brain. In this respect, it is also interesting to
note that the incidences for the right-, mixed-, and left-pawedness in dogs (57.1, 25,
and 17.9%) are similar to those for the cerebral length in the same dogs: 56%
( R > L ) ,28%(R=L),and 16%(R<L)(Tan&Cali?kan,1987).
According to Annett’s (1985) right shift (rs) theory, the extremely right-handed
subjects should have r s + + genotype with a population frequency of 32.4%.
Interestingly, the consistent right-handers (the bar at 1.0 S.D.) comprised strictly
32.4% of the total sample in the present work. Moreover, the proportion of the
GSs at -0.5 and +0.5 S.D. (49.6%; GSs from 50 to 90) was also close to the
population frequency of the subjects having rs+ - genotype (49.0%).The remaining
subjects represented by the bars at - 2.5, - 2.0, - 1.5, and - 1.0 S.D. comprised
18.0%1of the total sample, which is close to the incidence of the r s - - genotype
( 18.5%).
As the total sample was described by a percentage distribution of the GSs (see
Figure 2), the continuous distribution of the hand preference could be seen more
clearly. However, the subjects with all degrees of the left-hand preference (GSs
from - 5 to - loo), and the subjects with all degrees of the right-hand preference
(GSs from 5 to 100) exhibited quite different distributions. The left-handers (to the
left of 0) were evenly distributed within a broad spectrum to be expected from a
rectangular population. By contrast, the right-handers (to the right of 0) exhibited a

continuous J-shaped distribution. In other words, the left-handers were less
lateralized and more heterogeneous than the right-handers. This finding is
consistent with some studies (Benton et al., 1962; Hicks & Kinsbourne, 1978;
Peters & Durding, 1979), but not with others (Satz et al., 1967; Peters & Durding,
1978). A group bias to the right was clearly seen in the percentage distribution of
the GSs, but not to the left as Annett's right shift theory assumes.
In agreement with the results above, Borod et al. (1984) have reported that the
distributions of hand preference for the left-handers were less lateralized, less
skewed, and wider than those for the right-handers. In studies using carotid
arteriograms, Hochberg and LeMay (1975) reported that the branches of the
middle cerebral artery leaving the Sylvian fissure showed a wider angle in the
majority of the right-handers (67%)on the right side, but equal angles on both sides
in the majority of the left-handers (72%), indicating that the left-handers are not
opposite of the right-handers, and the left-handers are about evenly divided for
biases to each side. These distributions are also similar to those found for skull
asymmetries exhibiting the typical pattern of greater protrusion of the left occipital,
and the right frontal regions (LeMay, 1977). The left-handers were again not the
opposite of the right-handers; the right-handers exhibited a more pronounced
asymmetry, whereas the left-handers were about evenly divided to each side. The
conclusion that the right-handers are more lateralized than the left-handers, and
the left-handers are evenly distributed merely by chance to be expected from a
random sample was found to be true also for dogs (Tan & CaliSkan, 1987): among
the right-preferent dogs, the left Sylvian fissure was significantly higher than the
right (opposite to the man), but it was equally distributed in the ambidexters and
left-handers.
Total men and women. The distributions of the standard deviations about the
mean GS for the total men and women were essentially similar to that for the total
sample (see Figure 3). However, there were some differences between these
samples. There were significantly more consistent right-handed women than men,
indicating that the women tended to be more right-handed, i.e., more lateralized
toward the right-handedness, than the men. Borod et al. (1984) reported that
females had larger dominance ratios than males for preference measures, indicating
stronger right-handedness in women. According to Annett (1985), the right shift
factor has a slightly greater effect in females than males shifting the degree of the
right-hand use fiirther to the right. It was frequently reported that females show
greater lateralization than males in hand preference ( Oldfield, 1971; Hardyck et al.,
1975; Bryden, 1977). However, it was also asserted that there are no sex
differences in hand preference (Dennis, 1958; Fleminger et al., 1977).
Concerning cerebral lateralization, the prevailing view is that the female brain is
more bilaterally organized than the male brain (Fairweather, 1976; McGlone,
1977; Sherman, 1978). With regard to the physical asymmetries, the asymmetries
of the skull (LeMay, 1977), and size of the planum temporale (Wada et al., 1975)
have been reported to be similar between the sexes. Thus, the conclusion that the
female subjects were more lateralized than the males, and the right-shift gene could
be more strongly expressed in women than men seem to contradict the results that
women show a lesser degree of cerebral lateralization. Thus, verbal processing in
the brain seems to be different from hand control. It is true that the mouth area is
near the hand area within the cerebral cortex, but the area for foot control is far
away from these areas, and still there is 90% correlation between handedness and
footedness. Acc'ording to Corballis ( 1983), the conflict between cerebral
57 U. TAN
dominance for language and handedness can be resolved, if it is supposed that

handedness is more susceptible to environmental or pathological influences than
the cerebral lateralization for language. On the other hand, Annett (1985) has
suggested that the relatively bilateral symmetrical appearance of women’s brains
might be accounted for by the overdependence on the left-hemisphere for both
verbal and visuospatial thinking. In fact, it was found in the studies of cerebral
blood flow (Gur et al., 1982) that the pattern of increased flow to the left
hemisphere during verbal tasks was clearer in right-handed women than right-
handed men.
The incidences of the consistent right-handers were found to be 38.1 and 29.1%
for the women and men, respectively. These subjects can be considered as having
rs+ genotype according to the RS theory. Actually, Annett (see 1985) suggested
+
that the proportion of the subjects having rs++ genotype would be 32.4%. This is
close to the mean proportion of the consistent right-handers mentioned above
(33.6°/0). The large deviation of this proportion for women (38.1%) from that
expected from the RS theory (32.4%) could be accounted for by the same theory
suggesting that the expression of the rs gene would be stronger in females than
males.
The incidences at -0.5 and +0.5 S.D were found to be 52.1, and 45.8% for
women and men, respectively. These values are in accord with that suggested for
the rs+ genotype (49.0%). Taking the average of these incidences, the exactly
same proportion of 49.0% is obtained. Thus, it is conceivable that the GSs at - 0.5
and +0.5 S.D. may be due to the subjects having the rs+- genotype. It can be
calculated from the mean and its S.D. that these subjects comprise those having
GSs from 50 to 85 for men, and from 55 to 90 for women. In the total sample too,
the limit between r s + - ana rs-- people corresponded to a GS of 50. Here again,
the stronger right shift is clearly recogmzed in the female subjects. The remaining
subjects could have rs- - genotype, since their proportion was very close to that for
the theoretical distribution ( 18.5%). In accordance with this expected proportion,
the remaining incidences for men and women were found to be 18.8, and 16.1%,
respectively (mean = 17.5%).Because of this close agreement between the expected
(18.5%) and observed (17.5%) proportions, one can conclude that the subjects
having GSs between + 35 and - 90 for men and women (S.D.sfrom - 2.5 to 1.O)
could have rs-- genotype. This analysis (18.8% for men and 16.1% for women)
indicates that the males exhibit more left-hand preference than females. Annett
( 1985) herself suggested that the incidence for left-handedness is expected to be
slightly higher in males than females.
The percentage distribution of the GSs showed that the incidences for the left-
handers did not differ in men and women (see Figure 4), indicating that there was
no effect of sex on the left-hand preference. However, it was frequently reported
that the incidence of the left-hand preference was higher in men than women
(Komai & Fukuoka, 1934; Clark, 1957; Hecaen 8z Ajuriaguerra, 1964; Oldfield,
1971; Annett, 1973; Teng et al., 1976). Annett (1985) hypothesized that sex
differences in hand skill should be clear at the dextral side of the distribution, but
absent at the sinistral side where the RS gene is expected to be absent; the
incidence for left-handedness could be slightly smaller in females than males, due
to the stronger expression of the rs+ gene in females. No difference between the
male and female left-preferent subjects may be due to the relatively small sample in
this study.
As in the total sample, the percentages of the male and female subjects with all
degrees of left-hand preference were evenly distributed in contrast to the right-
handers. The incidence of extreme right-handedness was found to be significantly

higher in women (47.1%) than men (40.6%). The incidence of weak right-
handedness was significantly higher in men (54.1%) than women (47.6%).Thus, the
women tended to be more lateralized and right-handed than men. In accord with
this conclusion, the RS theory suggests that the RS factor should be more strongly
expressed in women than men. However, this result is not consistent with some of
the literature on measures of hand preference (Dennis, 1958; Fleminger et al.,
1977), but consistent with the other literature (Annett, 1970; Oldfield, 1971;
Bryden, 1977). The percentage distributions for the subjects with all degrees of
left-hand preference did not exhibit any prominent lateralization; these percentages
were evenly distributed merely by chance to be expected in a random sample from
a rectangular population.
Right-handed inen and women. In these subjects, the distributions of the
standard deviations about the mean GS was continuous, and negatively skewed, but
not J-shaped (see Figure 5). The mode was located at 0.5 S.D. (GS=65) in men,
and 1.5 SD. (GS= 100) in women, indicating a stronger right shift in women than
men. In both samples, the frequency of the strong right-handedness significantly

exceeded that for the weak right-handedness. The difference between the
frequencies of the weak right-handedness for men and women was not significant.
However, the proportion of exteme right-handedness was found to be significantly
higher in women than men. Thus, women tended to be more right-handed than
men. This finding is in accord with the RS theory suggesting that the expression of
the rs+ gene is stronger in females than males. The percentage distributions of the
GSs were similar for the right-handed men and women (see Figure 6). Here again,
the incidence of the extreme right-handed women significantly exceeded that for
men. By contrast, the incidence of the weak right-handedness was significantly
lower in women than men. These results suggest that women tended to be more
lateralized and right-handed than men.
Total right-, and lef-handers. The distributions of the standard deviations
about the mean (3swere unimodal, continuous, and negatively skewed in the total
right-handers (GSs> 0). By contrast, the GSs for the left-handers (GS< 0) were
normally distributed at p=.O5 significance level (see Figure 7). That is, the left-
handers were not opposite to the right-handers. Borod et al. (1984) have also
;studied the distributions of hand preferences for right- and left-handers. They
Found that the distributions for the right-, and left-handers were J-shaped, and the
lleft-handers were less lateralized than the right-handers. However, the left-handers
‘were found to be normally distributed without any bias to either side in the present
.work. This is coiisistent with findings for other asymmetries observed in the left-
Ihanders (planum temporale, middle cerebral artery, Sylvian fissure, dogs), which
were discussed above. In summary, the findings for structural and functional
<asymmetriessuggested that left-handers do not show asymmetries as do right-
Iianders but, in general, no bias to either side. The results of the present work also
indicated that the left-handers were more widely dispersed in hand preference than
the right-handers, and the right-handers were more lateralized toward the right-
hand use than the left-handers to the opposite side, as also reported by Borod et al.
( 1984).
The right shift in the sample for GSs > 0 (all degrees of right-hand preferences)
might be caused by a RS factor essentially inducing left-hemisphere speech, and
incidentally right-handedness (see Annett, 1985). However, a right shift in the
distribution of the paw preferences was also established in dogs, similar to humans
54 U. TAN
(Tan, 1987).Despite the inability of the dogs to speak, they tended to exhibit a right
tendency in paw use. We do not know yet, what the factor is facilitating right-paw
use in the majority of the dogs who cannot speak but understand and express their
feelings.
The percentage distributions of the total right-handers (GSs > O), and the total
left-handers (GSs < 0) exhibited essential differences in the right- and left-hand
preferent subjects (see Figure 8). The histogram for the right-handers could be
described as J-shaped, the strong right-handers (GSs > 50) being significantly more
than the weak right-handers (GSs from 5 to 50). By contrast, the percentage
distribution of the GSs for the left-handers was not J-shaped; there were
significantly more weak left-handers (GSs from - 5 to - 50; 64%) than the strong
left-handers (GSs from - 50 to - 100; 36%). Thus, the right-handers showed a
tendency toward strong right-handedness, whereas the left-handers were toward
the weak left-handedness, i.e., the opposite tendencies for hand preference could be
distinguished among right-, and left-handers. However, the weak and strong left-
handers were evenly distributed, in contrast to the right-handers. These results
suggest that the left-handers were less lateralized, and more widely dispersed than
the right-handers. This conclusion is in accordance with Borod et al. (1984). The
finding that the weak left-handers exceeded the strong left-handers may be
accounted for by recruitment of some mixed-handers whose incidence is known to
be much higher than the true left-handers.
Left-handed men and women. The distributions of the standard deviations about
the mean GS were normal ( p = .05) in the male and female left-handers (see Figure
9). There was no significant skewness, and the variances did not differ from each
other significantly, indicating that the male and female left-handers distributed by
chance without any significant bias to either side as expected from the RS theory.
The interquartile ranges were found to be - 60 and - 20 for men, and - 35 and
-80 for women. The Mann-Whitney U test indicated that the women had more
negative GSs than men. These results suggest that the women were more left-
handed than the men. It was also shown above that the females were more
right-handed than the males. These results are in accord with Gur et al. (1982)who
suggested that females may be more asymmetrical than males in the cerebral
representation of functions. It is conceivable that females might be more prone than
males to the existence or absence of the RS factor.
The percentage distributions of the GSs also showed some differences between
men and women (see Figure 10). There were significantly less strong left-
handedness than weak left-handedness among men. In women, the incidence of the
female left-handers with strong left-handedness was signficantly higher than the
incidence of the strong left-handed men. These and the other results in this work
further support the conclusion that females tended to be more lateralized than
males. Borod et al. (1984) also reported that among right-handers, females had
larger dominance ratios than males for preference measures, and among left-
handers the same tendency emerged for preference, indicating that females were
more lateralized than males.
REFERENCES
Annett, M. ( 1 970). A classification of hand preference by association anaylsis. British Journal of

PSychology, 61,303-321.
Annett, M. (1972).The distribution of manual asymmetry. British Journal ofPsychology, 63, 343-358.
DISTRIBUTION O F HAND PREFERENCE 55
Annett, M. (1973).Handedness in families. Annals ofHuman Genetics, 37,93-105.

Annett, M. (1979). i?arnily handedness in three generations predicted by the right shift theory. Annals of
Human Genetics, 42, 479-491.
Annett, M. (1985). Lefi, right, hand and brain: the right shift theory. London: Lawrence Erlbaum
Associates Ltd.
Beukelaar, 1.J., & Kroonenberg, P. M. (1983). Towards conceptualization of hand preference. British
Journal of Psychology, 74,33-45.
Borod, J. C., Caron, H. S., & Koff, E. (1984). Left-handers and right-handers on performance and
preference measures of lateral dominance. British Journal ofPsychology, 75, 177- 186.
Bryden, M. P. (1977 I. Measuring handedness with questionnaires. Neuropsychologia, 15, 617-624.
Clark, M. M. (1957).Left-handedness. London: University of London Press.
Collins, R. 1 . (1975).When left handed mice live in right handed worlds. Science, 187, 181-184.
Corballis, M. C. (1983). Human laterality. New York: Academic Press.
Dennis, W. (1958). E.arlygraphic evidence of dextrality in man. Perceptual and Motor Skills, 8, 147-149.
Fairweather, H. (1976).Sex differences in cognition. Cognition, 4,231-280.
Fleminger, J. J., Dalton, R., & Standage, K. F. (1977). Age as a factor in the handedness of adults.
Neurophsychologia, 15, 47 1-473.
Geschwind, N., & Behan, P. (1982). Left-handedness: Association with immune disease, migraine, and
developmental learning disorder. Proceedings of the National Academy of Sciences, USA, 79,
5097-5100.
Gillies, S. M., MacSweeney, D. A,, & Zangwill, 0. L. (1960). A note on some unusual handedness
patterns. Quarterly Journal of Experimental Psychology, 12, 113-1 16.
Gur, R. C., Gur, R. I:., Obrist, W. D., Hungerbuhler, J. P., Younkin, D., Rosen, A. D., Skolnick, B. E., &
Reivich, M. (19112).Sex and handedness differences in cerebral blood flow during rest and cognitive
activity. Science, 217, 659-661.
Hardyck, C., Goldman, R., & Petrinovich, 1 . (1975).Handedness and sex, race and age. Human Biology,
47.369-375.
Hecaen, H., & Ajuriaguema, J. (1964). Lefr handedness: Manual superiority and cerebral dominance.
New York Grune and Stratton.
Humphrey, M. E. ( I 95 1). Consistency of hand usage. British Journal of Educational Psychology, 21,
214-225.
Komai, T., & Fukuoka, G. (1934).A study of the frequency of left handedness and left footedness among
Japanese school children. Human Biology, 4.33-42.
LeMay, M. (1977). Asymmetries of the skull and handedness: Phrenology revisted. Journal of the
Neurological Sciences, 32,24 3-253.
McGlone, J. (1977). Sex differences in cerebral organization of cerebral functions in patients with
unilateral brain lesions. Brain, 100, 775-793.
McManus, I. C. (1984). Genetics of handedness in relation to language disorder. In F. C. Rose (Ed.).
Advances in neurology, VoL 42: Progress in Aphasiology, New York Raven Press.
Oldfield, R. C. (1971). The assessment and analysis of handedness: The Edinburgh Inventory.
Neuropsychologia, 9, 97- 1 14.
Porac, C., & Coren, S. ( 198 1). Lateral preferences and human behavior. New York Springer Verlag.
Raczkowski, D., Kalat, J. W., & Nebes, R. (1974). Reliability and validity of some handedness
questionnaire items. Neuropsychologia, 12,43-47.
:Sherman, J. A. (1978). Sex-rated cognitive differences: An essay on theory and evidence. Springfield,
Illinois: Charles (2.Thomas.
'Tan, U. ( 1987).Paw preferences in dogs. International Journal of Neuroscience, 32,825-829.
'Tan,U., & CaliSkari, S. (1987). Asymmetries in the cerebral dimensions and fissures of the dog.
Infernational Journal of Neuroscience, 32,943-952.
Teng, E. L., Lee, P-H, Yang, K-S, Chang, P. C. (1976). Handedness in a Chinese population: Biological,
social and pathological factors. Science, 193, 1148-1 150.
'Wada, J. A., Clarke, R., & Hamm, A. (1975). Cerebral hemisphere asymmetry in humans. Archives of
Neurology, 32,2.39- 246.

TMP 948 B

Cargado por

Información del documento

Descripción original:

Título original

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

TMP 948 B

Cargado por

Copyright:

Formatos disponibles

This article was downloaded by: [TÜBTAK EKUAL]

On: 7 August 2009

International Journal of Neuroscience

The Distribution of Hand Preference in Normal Men and Women

Online Publication Date: 01 July 1988

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

THE DISTRIBUTION OF HAND PREFERENCE IN

(Received October IS, 1987)

Keywords: handedmw, hand preference, manual asymmetry, lateralization, hand, man

skewness. The left-handers (GSs between - 10 and - 100) constituted a distinct

Left-handed men and women. The histograms in Figure 9 illustrate the

consistent right-handers were represented again by a single bar at 1.0 standard

preference between the English and Turkish people.

rectangular population. By contrast, the right-handers (to the right of 0) exhibited a

dominance for language and handedness can be resolved, if it is supposed that

handers. The incidence of extreme right-handedness was found to be significantly

men. In both samples, the frequency of the strong right-handedness significantly

Annett, M. ( 1 970). A classification of hand preference by association anaylsis. British Journal of

Annett, M. (1973).Handedness in families. Annals ofHuman Genetics, 37,93-105.

También podría gustarte