This article was downloaded by: [TÜBTAK EKUAL]

On: 13 April 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

Paw Preference in Cats: Distribution and Sex Differences

Üner Tan a; Mevlüt Yaprak a; Necíp Kutlu a
a
Atatürk University, Medical Faculty, Institute of Physiology, Erzurum, Turkey

Online Publication Date: 01 February 1990

To cite this Article Tan, Üner, Yaprak, Mevlüt and Kutlu, Necíp(1990)'Paw Preference in Cats: Distribution and Sex
Differences',International Journal of Neuroscience,50:3,195 — 208
To link to this Article: DOI: 10.3109/00207459008987172
URL: http://dx.doi.org/10.3109/00207459008987172

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 . Neuro.;cience, 1990, Vol. 50, pp. 195-208 0 1990 Gordon and Breach, Science Publishers, Inc.
Reprints available directly from the publisher Printed in the United Kingdom
Photocopying permitted by license only

PAW F’REFERENCE IN CATS: DISTRIBUTION AND

SEX DIFFERENCES
UNER TAN, MEVLUT YAPRAK and NECiP KUTLU
Atatiirk 411niversity,Medical Faculty, Institute of Physiology, Erzurum, Turkey

(Received October 4 , 1989)

Paw preference assessed by a food-reaching test was studied in male and female cats. Of the total sample
( N = 66), 34 (51.5%) were right-preferent, 24 (36.4%) left-preferent, and 8 (12.1%) ambilateral. In the
total sample, tk.ere was evidence for an overall paw preference, general paw preference, right-, and left-paw
preference. The: distribution of the right- minus left-paw reaches was neither normal, nor U or J shaped.
Of the males (A’= 24), 10 (41.7%) were right-pawed, 12 (50.5v0) left-pawed, and 2 (8.3%) ambilateral.
In males, there was evidence for an overall, general, and right-, left-paw preference relative to no preference.
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

The right- minus left-paw reaches fitted to gaussian data with two prominent peaks due to the right- and
left-preferents. In females ( N = 42), 22 (52.4%) were right-preferent, 14 (33.3%) left-preferent, and 6
(14.3%) ambilz.teral. There was an overall, general, and right-preference but not a left-preference relative
to no preference. The distribution of the right- minus left-paw reaches was neither normal nor U or J
shaped. The female right-preferents showed a right-bias compared to males. The left-preferent males were
more left-preferent than the right-preferent males are right preferent. The mean right- minus left- paw
reaches for the female right-preferents were significantly higher than those for the male right-preferents.
There was no significant difference between the right- minus left-paw reaches of the male and female
left-preferents The paw preferences exhibited consistency over time; no learning tendencies were estab-
lished during testing periods for at least 10 days. Considering the mean right-paw reaches for each
successive day ( N = lo), the mean right-paw uses in the right-preferents was higher in females than males.
The mean left-paw uses in left-preferents was about the same for males and females. In males, the mean
left-paw uses for the left-pawed males were higher than the right-paw reaches for the male right-preferents.
In females, there was no difference between the right paw reaches of the right-preferents and the left-paw
reaches of the left-preferents. It was concluded that there is a right-bias in paw preference of cats, which
is caused by the. female right-preferents under the influence of a biological factor. It was suggested that the
paw preference groups should be considered separately in assessing the sex differences in cerebral lateraliza-
tion. By doing so, it was established that the brain of the female right-preferents tends to be more lateralized
than the brain of the male right-preferents. In contrast, the female left-preferents seem to be less lateralized
than the male left-preferents. The right-pawed males were found to be least lateralized than the male
left-preferents, and female right-.and left-preferents. The male brain is, in general, more lateralized than
the female brain.

Ke-words: paw preference, laterality. cerebral dominance. male, ,female. cat, sex differences

Right-handedness in conjunction with functions of the left brain such as language and
self-consciousness can be considered as one of the most prominent human character-
istics. Manual specialization seems to be of great importance in relation to cognition
and phyletic basis of human behavioral lateralization (cf. Tan, 1988b). It is now well
known that rodents, cats, at least one species of marsupial, and macaque monkeys
have individually consistent hand preferences for food reaching. In these species,
about as mz.ny animals are strongly left-preferent as strongly right-preferent, and
about 50% ambilateral, indicating an unskewed distribution (see for review Annett,
1985; Walker, 1980). However, group-level hand biases have also been found with
marmosets (Box, 1977), and Japanese macaques (Itani et al., 1963; Kawai, 1967;

Address for correspondence to Prof. Dr. Uner TAN, Director, Institute of Physiology, Medical Faculty,
Atatiirk University, Erzurum, Turkey

195
 196 U TAN, M. YAPRAK A N D N. KUTLU

Tokuda, 1969). McNeilage ( 1 987) have reported a significant sex-related hand bias
with chimpanzees. Sanford et al. (1984) found a distribution skewed toward a left-
hand bias in the vertical stance in the lesser bushbaby. Thus, there seem to be
indications for a phyletic basis of human right-handedness and hence neural laterali-
ration. Using a plaster-removing test, Tan (1987) found 57.1% right. 17.9% left
preference, and 25.0% ambilaterality in mongrel dogs. This right-bias in paw
preferences in dogs and above findings encouraged us to restudy the distribution of
paw preferences in cats by using appropriate statistical tests.

METHODS

Assessment of Paw, Prejerence

The animals were 66 adult cats collected randomly from different villages (42 female
and 24 male). They did not receive any other behavioral testing before or during hand
testing. To assess paw preference, the cats were placed in a cage; they could reach the
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

food by a small hole in the middle on one face of the cage. The right- and left-paw
reaches were counted by the observer. At least one hundred reaches were counted each
day. This was repeated at least for 10 days.

Data Analysis
The frequency of right plus left paw use was normalized so that the sum of both paw
uses was 50 for one trial. The mean frequency for each paw use and the significance
of the mean difference between the right- and left-paw use was evaluated by a
computer program (Dallal, 1988) which performs exact randomization tests including
Wilcoxon signed-rank test and Wilcoxon-Mann-Whitney U test (see Dallal, 1988). To
summarize the data, a computer program (Theodorsson. 1988) was used, which
performs nonparametric and parametric statistics and Anderson-Darling test for
normality (Anderson & Darling, 1952). According to the level of significance
0, < .01) of the difference between the frequencies of the right- and left-paw use, the
animals were designated as right-preferent, left-preferent, and ambilateral. Ciccetti
(1987)’s criteria for a comprehensive analysis of paw preference data were also
subjected to an analysis.

RESULTS

Distribution qf’ Pa\%.Preference

Totul sample. The Parametric and nonparametric tests to assess hand preference
indicated that of 66 cats studied. 34 (51.5%) were right-preferent, 24 (36.4%) left-
preferent. and 8 (12.1 YO)ambilateral. Chi square with expected values of 22, 22, and
22 (and 2 4’) produced a value of 14.6 (p < .OOl), indicating evidence for overall paw
preference in cats.
There were 58 (87.9% of the total sample) cats manifesting either right or left paw
preferences, 8 cats (12.1%) did not show any preference. The corrected chi square
with expected values of 33 and 33 (and 1 df) yielded a value of 37.6 (p < .001).
indicating evidence for general hand preference (i.e.. left or right) relative to no hand
preference.
 PAW PREFERENCE IN CATS 197

There were 24 cats (36.4%) exhibiting left-paw preference, and 8 cats (12.1Yo)with
no preference. The corrected chi square with expected values of 16 and 16 (and 1 df)
yielded a value of 7.04 (p < .Ol), indicating that there is evidence for left-paw
preference r1:lative to no paw preference.
Of the totid sample ( N = 66), 34 cats (51.5%) were found to be right-preferent, and
8 cats (1 2.194) ambilateral. A corrected chi square with expected values of 21 and 21
(and I d f ) produced a value of 14.88 (p < .001), which indicates evidence for the
right-paw preference relative to no paw preference.
There were more right-preferent cats ( N = 34) than left-preferent ones ( N = 24).
However, the corrected chi square test did not show a significant difference between
these numbers (chi square with 1 df = 1 . 4 0 , ~> .20). This indicates that the right-
preferent cat s do not significantly outnumber the left-preferent cats. Probably, more
cats are needed to be conclusive, since values of chi square are heavily influenced by
sample size. The same proportion of right-paw over left-paw preference is significant
when based on 200 cases instead of 66 cats.
Using the criterion of 75/100 reaches for food with the same paw, of 66 cats, 18
(27.3%) were right-preferent, 16 (24.2%) left-preferent, and 32 (48.5%) ambilateral.
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

These propcrtions were quite different from those found in this work.
Figure 1A shows the distribution of the right minus left paw reaches normalized to
a total of 50 reaches. In this histogram, four modes are distinguished, which could be
due to the strong right-preferents, weak right-preferents, ambilaterals including the
weak left-preferents, and strong left-preferents. The Kolmogorov-Smirnov one-
sample test showed that there was no significant difference between the expected and
observed nu:nbers of the cats in each class ( N = 66, D = 0.10, p > .05), i.e., the
observed differences are merely chance variations to be expected in a sample from the
rectangular population. The chi square one-sample test also yielded the same result.
The Anderson-Darling test indicated that the distribution was probably non-gaussian
(A2 = 1.05, .01 < = p < .025).

TABLE 1
Statistical results for Right Minus Left Paw Reaches in Male and Female Cats

Group N mean S.D. median Int. range A2

Total 66 4.5 30.6 11.0 - 20.1 to 29.3 1.05 (NG)

Males 24 - 2.1 29.6 5.6 -31.0 to 18.8 0.66 (G)
Females 42 7.1 31.1 12.4 - 15.4 to 34.2 0.77 (G)
_____ ~ _ _ _ ~

Int. range: in1 erquartile range (even); A2: Anderson-Darling test; NG: non-gaussian distribution; G:
gaussian distribution
TABLE 2
Percentage Distributions of Paw Preferences of Cats Assessed by Different Methods

Right-Preferents Left-Preferents Ambilaterals

Cats N N YO N % N %

Total ( I ) 66 18 27.3 16 24.2 32 48.5

Total (2) 66 34 51.5 24 36.4 8 12.1
Males (1) 24 4 16.7 8 33.3 12 50.0
Males (2) 24 10 41.7 12 50.0 2 8.3
Females (1) 42 14 33.3 8 19.0 20 47.6
Females (2) 42 22 52.4 14 33.3 6 14.3

(I), and (2) refer to the paw preferences assessed by the criterion of 75/100 reaches for food with the same
paw, and by the method used in this work, respectively.
 I08 II. TAN. M. YAPRAK A N D N.KUTLC

20r A
V V
(%I
16 -
12 -

8 -

4-

0-
50 - 3 0 -10 10 30 5 0
(R-L)

V
(70)
2L -
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

20-

16 -

12 -
8-

L-
n,
-50 -30 -10 10 30 5 0
IR-L I

FIGURE I Distribution of Right- Minus Left-Paw Reaches in Total Sample (A). Male ( B : straight lmes).
and Female (dashed lines) Cats Ordinate: % of the cats tested. Arrows indicate the modes.

M u l e cats. Of 24 male cats, 10 were right-preferent (41.7%). 12 left-preferent

(50.0%). and 2 ambilateral (8.3%). There was an overall paw preference in these
animals (the corrected chi square with 2 df = 7.0, p < .05). There was also evidence
for a general paw preference, since there were significantly more cats exhibiting right
or left preferences ( N = 22) compared to two ambilaterals (the corrected chi
square = 16.66, df = 1, p < ,001).
The left-preferent cats ( N = 12) significantly outnumbered the ambilaterals
= 2. the corrected chi square = 7.14. p < ,001, df = 1). The right-preferent cats
(AV = 10) significantly outnumbered the ambilaterals ( N = 2). Thus, there was evi-
dence for a general paw preference. However, there was no significant difference
between the numbers of the right- and left-preferent cats (the corrected chi
square = 0.18, df’ = I , p > .05).
The distribution of the right minus left paw preferences in male cats is presented in
Figure 1B (straight lines). The Anderson-Darling test for normality indicated that
there was no evidence against the hypothesis of gaussian data (a2 = 0.66,
.05 < = p < .lo). but at a marginal significance level. In fact. there were two
prominent peaks in the histogram, due to moderately right-pawed cats, and the other
to the left-pawed cats.
 PAW PREFERENCE IN CATS 199

Females. Of 42 female cats, 22 were right-preferent (52.4%), 14 left-preferent

(33.3%), arid 6 ambilateral (14.3%). There was an overall paw preference in this
sample (the corrected chi square = 12.03, df = 2, p < .OOl). There was also evi-
dence for a general paw preference, since there were significantly more cats manifest-
ing either right or left preference ( N = 36) than cats showing no paw preference (the
corrected chi square = 20.02, df = 1, p < .OOl). There was no evidence for left-paw
preference (the corrected chi square = 2.45, df = 1, p > .05). The right-preferent
animals sigriificantlyoutnumbered the ambilaterals (the corrected chi square = 4.02,
dj”= I , p <I .05), indicating evidence for right-paw preference. There were more
right-pawed cats ( N = 22) than left-pawed cats ( N = 14), but the difference was not
significant ( z = 1.17, p = .12). The corrected chi square test for one sample also
yielded the same result.
Figure 1B8also shows the distribution of the right minus left paw use for the female
cats (dashed lines, shaded bars). The distribution fitted to a rectangular chance
distribution. There was, however, a prominent peak due to the strongly right-prefer-
ent cats. It is also seen in this diagram that the females values show a more right-
biased paw preference compared to males, and the male data show a more left-biased
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

preference compared to females. The whole picture seems to be right-shifted for

females and left-shifted for males. Table 1 summarizes the results concerning the paw
preference for the male and female cats. The Kruskal-Wallis test indicated that there
were significant diferrences between these three groups (total sample, males and
females, p <: .05).
Table 2 presents the percentage distributions of the male and female cats according
to their paw preferences evaluated by using the criterion of 751100 reaches for the
same paw (see Cole, 1955), and the criteria used in the present work. If our data are
analyzed by using Cole’s criterion of 75/100 paw reaches for the right- or left-prefer-
ence, 27.3% of the animals were found to be right-preferent and 24.2% left-preferent,
i.e., the righi.- and left-preferents were equally distributed. The remaining 48.5% was
due to the ambilaterals. The stastistical analysis used in this work revealed, however,
that 51.5% of the cats were right-preferent and 36.4% left-preferent: only a small
percentage ( 12.1YO)was due to the ambilaterals. Considering the sex differences.
Cole’s criterion showed that 16.7% of the males and 33.3% of the females were
right-preferent. In contrast, 33.3% of the males and 19% of the females were left-
preferent; atlout 50% of the animals comprised ambilaterals. Interestingly, the right-
preferent fernales significantly outnumbered the right-preferent males; the left-prefer-
ent males significantly outnumbered the left-preferent females. In our analysis, the
right-preferent females (52.4%) slightly outnumbered the right-preferent males
(41.7%); the incidence of the left-preferent males (50.0%) was higher than the
incidence of the left-preferent females (33.3%).

Right-, Left-Preferents, and Ambilaterals

Figure 2 shows the distributions of the right minus left paw reaches normalized to the
right plus left 50 reaches for each cat in the male and female right-, left-pawed, and
ambilateral cats. The right-pawed cats comprised the class intervals between zero and
50 (straight lines). There was a single mode within the class interval from 10 to 20.
Thus, the male right-preferent cats were, in fact, represented by moderately right-
pawed cats. The right minus left paw reaches for the female cats showed a right-bias
compared to males. The histogram for the female right-pawed cats comprised the
class intervals from 10 to 50 (dashed lines, shaded bars). The most prominent peak
 200 IJ. TAN, M . YAPRAK A N D N . KUTLC

26
(%I
2L

22

20
.
18

16

lr!

12

10
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

0
-50-40-30 - 2 0 -10 0 10 2 0 30 40 50
(R-L)

F IG UR E 2 Distribution of Right- Minus Left-Paw Reaches for Male (straight line) and Fenialc (dashed
line) right- and Left-Preferent Cats Dotted line refers to the male and female (shaded area) ambilaterals.
Ordinate: 70of the cats tested.

occurred within the class interval from 40 to 50. The other less prominent peak was
due to the moderately right-pawed female cats. The left-preferent males comprised the
class intervals from 0 to - 50 (straight lines). However, the most prominent peak was
between - 30 and - 40, indicating that the left-preferent males were more left-prefer-
ent than the right preferent males are right-preferent. The right minus left paw reaches
for the female left-pawed cats were distributed between - 10 and - 50 (dashed lines.
shaded bars). The distribution seems to be rectangular, and the mostly encountered
left-preferences occurred within the class interval from - 10 to - 20. The ambilaterals
are shown by the dotted lines in this figure. The right- minus left-paw reaches for the
female ambilaterals extended from - 10 to 10 (dotted lines, inversely shaded bars) as
the male ambilaterals). However, the proportion of the female ambilaterals are higher
than the proportion of the male ambilaterals as seen in the bars (0 to - 10) due to
these two groups. The whole picture appears to be relatively right-shifted for the
females, and left-shifted for the males. Moreover, the bimodality of the right- minus
left-paw distribution seems to be a distinctive feature of the male cats. The right-
minus left-paw preferences of the female cats seem to be evenly distributed with a
probable exception of the strong right-preferents.
The data from the male and female cats are summarized in Table 3. A5 seen from
this table. the mean right- minus left-paw reaches for the female right-preferents are
 PAW PREFERENCE IN CATS 20 1

TABLE 3
Statistical Results of the Right-Minus Left-Paw Reaches in Male and Female Cats Including Right-,
Left-preferents and Ambilaterals

Pawedness N mean S.D. median int. range A2

Right-pawed 34 29.1 13.5 21.7 18.6 to 40.8 0.82 (NG)
males 12 22.7 12.6 18.8 11.8 to 34.8 0.74 (G)
females 22 32.6 12.8 31.8 22.0 to 45.6 0.58 (G)
Left-pawed 24 -31.0 13.4 - 33.7 - 39.8 to - 16.2 0.65 (G)
males 10 - 33.3 11.9 - 34.7 - 39.8 to - 26.8 0.24 (G)
females 14 - 29.4 14.5 - 33.5 -45.2 to - 15.4 0.76 (G)
~~

Int. range: interquartile range (even); A2: Anderson-Darling test for normality; NG: non-gaussian
distribution; G: gaussian distribution

higher than those for the male right-preferents in all statistical results. The left-
preferences h r the male cats seem to be slightly more than those for the female cats.
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

The Kruskal-Wallis test indicated that the right- minus left-paw reaches for the female
right-pawed cats were significantly higher than those for the male right-pawed cats
(p < .05). There was no significant difference between the right- minus left-paws
reaches of the male and female left-pawed cats.

Consistency of Paw Preferences in Male and Female Cats

To evaluate paw preference, cats were tested for at least 10 days. the diagram in Figure
3A shows the mean right-paw reaches for the right-pawed male (open circles) and
female (closed circles) cats in each successive day. The sum of the right- and left-paw
reaches was normalized to 50 reaches. The Kruskal-Wallis test indicated that there
was no significant difference between the right-paw reaches for each day either in male
or in female #cats.It is also seen in this diagram that the mean right-paw reaches for
the female cats are higher than those for the male cats. However, according to the
results of the Kruskal-Wallis test, the differences between the males and females were
found to be significant only for the 5th (p < .Ol), 6th, 7th, and 10th days (p < .05).
The diagram in Figure 3B show the mean left-paw reaches from 50 right plus left
paw reaches in the left-pawed male (open circles) and female (closed circles) cats. The
Kruskal-Wallis test indicated that there were no significant differences between the
mean left-paw reaches for each successiveday. The male and female data did not show
any difference.
In Figure 4, the male and female cats were compared in regard to pawedness. The
diagram in Figure 4A shows the mean right-paw reaches from the right-pawed males
(open circles), and the mean left-paw reaches for the left-pawed males (closed circles)
in each successive day. It is seen that the mean left-paw reaches for the left-pawed
males are higher than the mean right-paw reaches for the male right-pawed cats
(p < .05).
The diagram in Figure 4B shows the mean right-paw reaches for each successive
day in the right-pawed female cats (open circles), and the mean left-paw reaches from
the left-pawed female cats (closed circles). As clearly seen in this diagram, there were
no differences between the male and female data. The mean right-paw reaches for the
female ambilaterals were also shown in this diagram for each day (open triangles).
The Kruskal-Wallis test indicated that there were no significant differences between
the mean right-paw reaches concerning the testing days ( N = 10).
 102 U . TAN, M . YAPRAK AND N . KUTLU

20
1 2 3 4 5 6 7 0 9 1 0
DAYS
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

1 2 3 L 5 6 7 0 9 10
DAYS
FIGURE 3 Mean Right-Paw Reaches in Right-Preferent and Left-Preferent Cats Tested for 10 Days
Abscissa: days. Ordinate: the mean right-paw reaches with standard deviations (vertical lines). A: right-
preferent cats; open circles, males (y = 35.9 - O.Ix, r = .28); closed circles, females (y = 41.6 - 0 . 1 ~ .
v = ,231. €3: left-preferent cats: open circles, males (y = 42.9 - 0 . 3 ~ v. = ,381: closed circles. females
( y :r 4 0 . 8 ~ r. r= .11).

DISCUSSION

Totalsmiple. The cats were found, in this work, to show consistent paw preferences.
They constituted three groups: right-preferents (51.5%), left-preferents (36.4%), and
ambilaterals (12.1 YO).Cole (19%) also studied the pawedness in cats using a criterion
of 75/100 reaches for food with the same paw. He reported that 20% were right-
preferent, 38.3% were left-preferent, and 41.7 were ambilateral. We did not use Cole’s
criterion. and found a right-bias in pawedness of cats, which was produced by
preponderance of the right-preferent female cats. A left-bias relative to right-paw
preference similar to Cole’s cats could not be established in our study. Re-analyzing
our data by using the criterion of 75/100 reaches with the same paw, 27.3% of the
total sample were found to be right-preferent, 24.2% left-preferent, and 48.5%
ambilateral. So, we were not able to confirm the proponderance of left-preferents in
cats as in accordance with others, but the high proportion of ambilaterals appears to
be reliable (Forward, et al., 1962, Warren et al., 1967). The proportions found in the
present work by using the criterion of 75/100 reaches for the same paw fits the reports
 PAW PREFERENCE IN CATS 203

201 ' I I I I I I I I 1
1 2 3 4 5 6 7 8 9 10
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

10 11
1
1
2
1
3
1
4
1
5
1
6
11
7 8
11
9 1 0
DAYS

FIGURE 4 h k a n Right-Paw Reaches in Male and Female Cats Abscissa: days. Ordinate: Oh of the cats
tested. A: malei; open circles, the mean right-paw reaches fiom the right-preferents (y = 35.9 - O.Ix,
r = .28); closed circles, the mean left-paw reaches from the left-preferents (y = 42.9 - 0 . 3 ~ )B:. females;
open circles, the mean right-paw reaches from the right-preferents (y = 41.6 - O.Ix, r = .23); closed
circles, the mean left-paw reaches from the left-preferents (y = 4 0 . 8 ~ )Open
. triangles: the mean right-paw
reaches for the female ambilaterals (y = 23 + 0.2x, r = .41).

on hand or paw preferences in monkeys, chimpanzees, cats, rats, and mice (see for
review Annett, 1985).
In the present work, another statistical approach was used to assess the direction
and degree of paw preference in cats, and found a right-biased distribution, which was
neither U shaped nor J shaped. It was probably a rectangular distribution indicating
chance variai.ions of the observed frequencies to be expected in a random sample from
the rectangular population. In Cicchetti's (1987) terms, there was evidence for overall
and general paw preference, and left- and right-paw preferences relative to no prefer-
ence for the total sample. However, the right-preferents did not significantly outnum-
ber the left-preferents. Probably, more animals are needed to be conclusive, since
values of chi square are heavily influenced by sample size. The same proportion of
right-paw over left-paw preference would be significant if based on 200 cats.
The propclrtion of right-handedness in the human population seems to be higher
than that in cats. Using a preference test (plaster removing test) without visual
guidance, Tan (1987) has found that 57.1% of the dogs exhibited right-paw prefer-
 20.1 U . TAN, M. YAPRAK AND N. KUTLU

ence. In humans, about 66% of the population was estimated to be right-handed (see
Annett, 1985). Interestingly, the incidence of right-handedness was found to be 66.1'4
in Turkish students (Tan, 1988a). The higher incidence of right-handedness in humans
than cats (this work) and dogs (Tan, 1987) might be determined by social and genetic
factors. Using T-maze tests, Castellano et al. (1987) have found that practice facili-
tates the expression of the behavioral assymmetry in rats at both individual and
population levels. These authors have argued that the differences in behavioral
population laterality between humans and animals are related to characteristics of the
behavior used for each species. Human laterality is assessed usually by the behavioral
patterns receiving continuous practice during routine daily activity (e.g., writing,
drawing). In contrast, the tasks used for rats are usually novel (Castellano et al. 1989).
Thus, the lower incidence of right-sidedness or right-pawedness in animals such as
rats, cats. and dogs than the incidence of right-handedness in humans may also be due
to practice differences between humans and animals.
Collins (1988) has reported that mice allowed to observe a model demonstrate a
lateralized solution to a novel problem later exhibited the same behavioral asymmetry
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

of the model. Glick and Ross (1981) have reported that 54.8% (p < .25) of the rats
tested for amphetamine-induced or nocturnal rotation had right side biases; 57.5%
(p < .02) of rats tested in two-lever operant situations preferred the right lever. Glick
and Ross have concluded that female Sprague-Dawley rats as a population have a
right-sided bias. The incidence of right side bias in rats is about the same as the
incidence of right-pawedness in cats established in the present work. Glick and Ross
have also proposed that the increase in population bias in humans as compared to
rodents is related to the evolution and growth of the cortex in relation to subcortical
structures.
In contrast to earlier findings (Cole, 1955; Forward et al., 1962; Warren et al., 1967)
indicating an incidence of ambilaterality of about 50%, we found only 12.1YOof the
cats being ambilateral in paw preference. This was apparently due to different treat-
ments of data. We also obtained an incidence of 48.5% for ambilaterality using the
criterion of 751100 reaches with the same paw. The ambilaterality in humans was
reported to occur rarely in distributions of hand preferences (Porac & Coren, 1981).
So, the incidence of ambilaterality of 12.1YOin cats does also not in conflict with
ambilaterality in humans.
The incidence of the left-paw preference was found to be rather high in cats
(36.4%). In fact, the incidence of left-handedness in humans was reported to show
variations ranging from 1% to 30% according to several authors and methods of the
assessment of left-handedness as reviewed by Hecaen and Ajuriaguerra (1964). The
incidence of left-handedness in humans has been reported to show moderate varia-
tions between people of different cultures. Such a variability was argued to be
regulated by cultural coercion and the imposition of social sanctions (Dawson, 1977;
Teng et al., 1976). Collins (1988) suggested that even if forces of cultural coercion were
absent, processes of social learning alone may be sufficient to lead to local lateralized
traditions. Thus, social learning and cultural coercion may have forced many left-
handers toward right-handedness decreasing the incidence of left-handedness in
human populations. Collins (1985) has, in fact, demonstrated in genetically uniform
mice that environmental biases can exert strong influence on manifest laterality.

Mules andJenzules. The distribution of the right- minus left-paw reaches exhibited
important differences between males and females. The male distribution was clearly
bimodal due to the right- and left-preferents. The right- minus left-paw reaches for the
female cats was evenly distributed except for the mode for the strong right-preferents.
 PAW PREFERENCE IN CATS 205

The incidences for the right-, left-preferents, and ambilaterals were 50.0%, 41.7%,
8.3% in males, and 52.2, 33.3%, and 14.3% in females, respectively. Thus, the right-
preference is slightly higher in females than males. The male cats were more left-
preferent than females. Ambilaterality was more pronounced in females than males.
Using the criterion of 75/100 reaches for the same paw, the incidences of the right-,
left-preferents, and ambilaterals were 16.7% 33.3%, and 50% for males, and 33.3%,
19.0%, and 47.6% for females, respectively. Here, the ambilaterality is about 50% in
males and kmales. However, the incidences of the right-, and left-paw preferences are
different depending on sex. The incidence of right-preference is higher in females than
males. The incidence of the left-preference is lower in females than males. We see here
that Annett's (1985) proportions of the right- and left-preferences, i.e., 25% right- and
25% left-pi-eferences, are not necessarily correct. The distributions are strongly
sex-dependent.
Interestingly, the incidences of the right- and left-preferences are reversed in males
and females: 33.3% for the male left-preferents, 33.3% for the female right-preferents,
16.7% for the male right-preferents, and 19.0% for the female left-preferents. This
example also shows clearly that females are more right-preferent than males, and
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

males are more left-preferent than females. This is also seen in the histogram of the
right- minus left-paw reaches for males and females. The mode for the right-preferent
males represents the weak right-preferent cats, whereas the mode for the right-pawed
females represents the very strong right-preferents. These data are in accordance with
humans. Tan (1988) has studied the distribution of hand preference in Turkish male
and female students. He found that the incidence of weak right-handedness is signifi-
cantly higher in men than women; the probability of consistent right-handedness is
significantly higher in women than men.
Collins (1975) has reported that male mice are generally less lateralized than
females for paw preference. In a critical review, Robinson et al. (1985) have reported
a consisten1 sex difference favoring females in the degree of postural and motor
asymmetries. In the present work, the right- minus left-paw reaches were found to be
distributed evenly in females, except for the peak for strong right-preference. The
males distribution of the right- minus left-paw reaches exhibited two distinct groups
as the weak right-preferents and left-preferents. This result indicates a more pro-
nounced right-pawedness in females than males. The female distribution appears to
be more symmetrical than the male. Diamond (1980) has concluded that the male
brain is more asymmetrically organized than the female brain which appears as
bicerebral. The right-bias in females paw preference is apparently determined biolog-
ically. Because of these sex differences in cerebral lateralization and handedness, it is
conceivable that the selective hormonal control could be at least one of the factors
contributing to the development of the cerebral lateralization including handedness.
There are, indeed, studies indicating that sexual hormones play a role in the structure
and function of the central nervous system (see Reinisch 1974; 1976).
The resulfs of the present work suggest that the sex differences should be considered
for left-prefcrents and right-preferents separately. The distribution of the right- minus
left-paw reaches for the left-preferent male cats had only one prominent peak rep-
resenting rather strong left-preferents (see Figure 2). This indicates an asymmetric
motor organization of the male brain. The distribution of the right- minus left-paw
reaches for 1 he left-preferent females cats did not show any prominent peak except a
nonprominent peak for the weak left-preferents. Thus, the female brain is organized
less asymmetrically than the male brain for the motor functions. On the other hand,
the distribution of the right- minus left-paw reaches for the right-preferent males had
a prominen: mode for the weak right-preferents. The right-pawed females had a
 206 U. TAN, M . YAPRAK A N D N . KUTLU

prominent mode for the strong right-preferents. These results indicate that the brain
of the right-preferent females is more asymmetrically organized for the motor func-
tions than the brain of the right-preferent males. In sum, the cerebral motor lateraliza-
tion is inversely organized in the male and female left- and right-preferents. That is,
the brain of the right-pawed females is more lateralized than the brain of the right-
pawgd males. On the contrary, the brain of the left-preferent females is less lateralized
than the brain of the left-preferent males. Annett (1972, 1978) has reported that
females were more lateralized in favor of the right hand than males on a peg-moving
task.
McGlone (1 977) has suggested that male right-handers may be more homogeneous
than female right-handers with respect to left hemisphere dominance for speech
functions. The same arguments may be applied to the motor asymmetries established
in cats in the present work. In the distribution of the right- minus left-paw reaches (see
Figure 2 ) , the right-pawed males were represented by one prominent peak due to the
weak right-preferents; the female right-preferents were represented by two modes due
to the weak and strong right-preferents. Thus, the male right-preferents were more
homogeneous than the female right-preferents. McGlone (1980) has reported in a
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

review article that the male brain is more lateralized for both verbal and spatial
cognitive processes which are more bilaterally organized in the female brain. This can
be confirmed for the cat brain if the whole histogram including the right- and
left-preferent cats is considered (see Figure 1): the males exhibit two distinct groups;
the females exhibit a continuous distribution. However, it is also seen in this histo-
gram that the female brain is associated with a greater tendency towards dextrality.
This diagram clearly shows that females are more right-pawed than males; males are
more left-pawed than females. These results are in accordance with human studies.
Glick et al. (1980) have argued that lateralization processess in human and animal
brains are more similar than different.
The paw preference in cats were found to be reliable and enduring. There was no
significant difference between the test results for at least 10 days. There was no
significant change in the degree of paw preference for at least 10 successive days.
Castellano et al. (1987, 1989) have found that practice facilitates the expression of the
behavioral asymmetry in naive rats at both individual and population levels. All the
animals exhibited consistent paw preferences in the present work. In right-preferent
cats, the mean right-paw reaches for each successive day were higher in females than
males (see Figure 3A). This again shows that the right-preferent females are more
right-preferent than the right-preferent males. In the left-preferent cats, there was n o
difference between the mean left-paw reaches for each successive day in males and
females (Figure 3B). This result is in accord with human data indicating that sex
differences are clear at the dextral side of the distribution, but absent at the sinistral
side, i.e., females are farther to the right at the dextral side (see Annett, 1985).
However. we cannot overlook the prominent peak in the histogram of the right-
minus left-paw reaches for the male left-preferents, and the evenly distributed right-
minus left-paw reaches for the female left-preferents. Thus, it is conceivable that the
male left-preferents are more left-preferent than female left-preferents. In fact, there
are slightly more male left-handers than female left-handers at the extreme left-side
of Annett’s histogram (see Annett, 1985, p. 212). In the male cats, the mean left-paw
reaches for the left preferents for each successive day was found to be higher than the
mean right-paw reaches for the right-preferents (see Figure 4A). Thus, the left-paw
use of the left-preferent males is better than the right-paw use of the right-preferent
males. In other words, the left-preferents are more left-preferent than the right-prefer-
ents are right preferent. This is also seen in the histogram showing the distribution of
 PAW PREFERENCE IN CATS 207

the right- minus left-paw reaches: the majority of the right-preferents are only weak
right-preferents (R-L = 10 to 20); the majority of the left-preferents are rather strong
left-prefermts (R-L = - 30 to -40). It can thus be concluded that the male left-
preferent animals are more lateralized than the male right-preferents. However, there
was no difrerence between the mean frequencies of the right-paw reaches from the
right-preferent females and the left-paw reaches from the left-preferent females (see
Figure 4B). This indicates that the right- and left-preferent females are equally
lateralized. The important feature of the male right-preferent cats is that the mean
frequencies of their right-paw reaches is always lower than the mean frequencies of
the right- and left-paw reaches of the female cats and the mean left-paw reaches of the
male cats. It can thus be concluded that the right-pawed males are least lateralized
among the others.

REFERENCES

Anderson, T. W. & Darling, D. A. (1952). Asymptotic theory of certain ‘goodness of fit’ criteria based on
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

stochastic: processess. Annual Mathematical Statistics, 23, 193-212.

Annett, M. (I‘m). Left, right, handandbrain: the rightshft theory. London, Hildale, New Jersey: Lawrence
Erlbaum Associates.
Annett, M. (1972). The distribution of manual asymmetry. British Journal of Psychology, 63, 343-358.
Annett, M. (1978). Genetic and nongenetic influences on handedness. Behavior Genetics, 8, 227-249.
Box, H. 0. (1977). Observations on spontaneous hand use in the common marmoset (Callithrix jaccus).
Primates, 18, 395-400.
Castellano, M. A,, Diaz-Palarea, M. D., Rodriguez, M., & Barroso, J. (1987). Lateralization in mae rats
and dopaminergic system: Evidence of right-side population bias. Physiology and Behavior, 40,
607-612.
Castellano, M. A,, Diaz-Palarea, M. D., Barroso, J., & Rodriguez, M. (1989). Behavioral lateralization in
rats and dopaminergic system: Individual and population laterality. Behaviroal Neuroscience, 103,
46-53.
Cicchetti, D. ‘4.(1987). On viewing the evidence for primate handedness: Some biostatistical considera-
tions. Behavioral and Brain Sciences, 10, 268-269.
Cole, J. (1955). Paw preference in cats related to hand preference in animals and in men. Journal of
Compararive and Physiological Psychology, 48, 137-140.
Collins, R. L. (1975). When left-handed mice live in right-handed worlds. Science, 187, 181-184.
Collins, R. L. (1985). On the inheritance of direction and degree of asymmetry. In S. D. Glick (Ed.),
Cerebral iateralization in nonhuman species. New York: Academic Press.
Collins, R. L. I 1988). Observational learning of a left-right behavioral asymmetry in mice (mus musculus).
Journal oj- Comparative Psychology, 102, 222-224.
Dallal, G. E. (1988). PITMAN: A FORTRAN program for exact randomization tests. Computers and
biomedical Research, 21, 9-15.
Dawson, J. L. 1M. B. (1977). An anthropological perspective on the evolution and lateralization of the brain.
In S. J. Dimond & D. A. Blizard (Eds.), Annals of New York Academy of Sciences: 299. Evolution and
lateralization of the brain (pp. 424447). New York: Academy of Sciences.
Dimond, S. J. I 1980). Sex differences in brain organization. The Behavioral and Brain Sciences, 3,234-235.
Forward, E., Warren, J. M., & Hara, K. (1962). The effects of unilateral lesions in sensory motor cortex
on manipulation by cats. Journal of Comparative and Physiological Psychology. 55, 1130-1 135.
Glick, S. D., Schonfeld, A. R., & Strumpf, A. J. (1980). Sex differences in brain asymmetry of the
rodent. Thu Behavioral and Brain Sciences, 3, 236.
Glick, S. D., & Ross, D. A. (1981). Right-sided population bias and lateralization of activity in normal rats.
Brain Resmrch, 205, 222-225.
Hecaen, H., & Ajuriaguerra, J. (1964). Left handedness: Manual superiority and cerebral dominance. New
York: Grune and Stratton.
Itani, J . , Tokuda, K., Furuya, Y., Kano, K., & Shin, Y . (1963). The social construction of natural troops
of Japanese monkeys in Takasakiyama. Primates, 4 , 1 4 2 .
Kawai, M. (1967). Catching behavior observed in the Koshima troop: A case of newly acquired behavior.
Primates. 8, 181-186.
McGlone, J. (L977). Sex differences in the cerebral organization of verbal functions in patients with
unilateral brain lesions. Brain, 100, 775-793.
 208 U . TAN, M . YAPRAK A N D N. KUTLU

McGlone. J. (1980). Sex differences in human brain asymmetry: a critical survey. The Behavioral and Brain
Sciences, 3, 2 15-263.
McNeilage. P. F. (1987). The evolution of handedness in primates. In D. Ottoson (Ed.), Wennergrrn
International Symposium Serie.r: Vol. 47. Duality and unity of the brain: Unified functioning and
specialization of the hemispheres (pp. 100-1 13). New York: Macmillan.
Porac, C., & Coren, S. (1981). Lateral pwferences and human behavior. New York: Springer Verlag.
Reinisch, J. (1974). Fetal hormones, the brain and human sex differences: a heuristic integrative review of
the recent literature. .4rchives of Sexual Behavior, 3, 51-90.
Reinisch, J. (1976). Effects of prenatal hormone exposure on physical and psychological development in
humans and animals: with a note on the state of the field. In E. Sachar (Ed.) Hormones, behavior, and
psychopathology, pp. 69-94. New York: Raven Press.
Robinson, T. E., Becker, J. B., Camp, D. M., & Mansour, A. (1985) Variation in the pattern of behavioral
and brain asymmetries due to sex differences. In S. D. Glick (Ed.), Cerebral lareralizarion in nonhuman
.specier (pp. 185-23 1). New York: Academic Press.
Sanford, C., Guin. K., & Ward, J. P. (1984). Posture and laterality in the bushbaby (Galago .senegalensis).
Brain Behavior and Evolution, 25. 21 7-224.
Tan, U . (1987). Paw preferences in dogs. International Journal (?f .Ireurnscience. 32. 825-829.
Tan. U . (1988a). The distribution of hand preference in normal men and women. International Journal qf
Neuroscience, 41, 35-55.
Tan. U. (1 988b). The relationship between nonverbal intelligence, familial sinistrality and Geschwind
Downloaded By: [TÜBTAK EKUAL] At: 09:53 13 April 2009

scores in right-handed female subjects. International Journal of Neuroscience, 43, 177- 182.
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.
Theodorsson, E. (1988). BASIC computer program to summarize data using nonparametric and parame-
tric statistics including Anderson-Darling test for normality, Computer Methds and Programs in
Biomedicine, 26, 207-2 14.
Tokuda, K . (1969). On the handedness of Japanese monkeys. Primates, 10. 41-46.
Walker, S. F. (1980). Lateralization of functions in the vertebrate brain: A review. Brirish Journal of

Warren. J . M., Ablanalp. J. M.. & Warren, H. B. (1967). The development of handedness in cats and
rehesus monkeys. In H. W. Stevenson. E. H. Hess, & H. L. Rheinglod (Eds.), Early behavior:
Comparative and developmental approaches. New York: Wiley.