i

.Iuiirg‘u

». ":95

 

an! . 3
. 151351.142!

.
95mm ‘5
‘

 

 

 

 

 

I: (1.1.3 . 1. .2311.

éfix, L? @iui

 

l ' LIBRARIES
q , v" MICHIGAN STATE UNIVERSITY
0 i) :— EAST LANSING, MICH 48824-1048
'4" ’ I". ‘74.!

This is to certify that the
dissertation entitled

AN INVESTIGATION OF LATERAL HOLDING BIASES IN
PRE-SCHOOL-AGE CHILDREN

presented by

Jason Bradley Almerigi

has been accepted towards fulfillment
of the requirements for the

Doctoral degree in Psychology

 

 

MW

MajEFProfessor’s §gnature

0W 2‘!" 2004-

Date

MSU is an Affirmative Action/Equal Opportunity Institution

---.-.-.-.-

-l- -.-o—-«-.--u-nn-n--

PLACE IN RETURN Box to remove this checkout from your record.
10 AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

AN INVESTIGATION OF LATERAL HOLDING BIASES
IN PRESCHOOL-AGE CHILDREN

By

Jason Bradley Almerigi

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Psychology

2004

ABSTRACT

AN INVESTIGATION OF LATERAL HOLDING BIASES
IN PRESCHOOL-AGE CHILDREN

By

Jason Bradley Almerigi

Most adults hold infants asymmetrically and in the same direction, with the
infant’s head and upper body to the left of the holder’s midline. Although many studies
have explored the bias in adults, especially in mothers with their newborn infants, no
adequate explanation of the bias has been supported. The current study approaches the
question from a developmental standpoint, assessing the earliest manifestation of the bias
to determine its origins and covariates, specifically sex and handedness. A community
sample of ninety-seven pre-school children, who were tested in their homes, were asked
to hold a realistic-looking infant doll on three occasions to assess degree of bias. Three
main concerns were addressed. First, prior developmental studies assessed the bias in
children from one of three different countries, Sweden, South Africa, and Brazil. The
current study extended this work to children from the United States. Second, prior
studies came to different conclusions about age of onset: one report first finding the bias
in 2-year-olds, another in 4 year olds, and the last in 6-year-olds. The current study
therefore sampled children between the ages of 3 and 5 to cover the possible period of
transition. It also used improved methods to repeatedly assess the strength of the bias as
reflected in the stability of the hold, its closeness and duration, and its flexibility when
the child had to perform a concurrent manual task. Using realistic dolls, the results

suggested that while the bias is not yet established in the group as a whole, girls but not

boys showed repeated, albeit non-significant, preferences to hold on the left. Girls also
were significantly more likely to hold the dolls closer to their body and for longer periods
of time. Finally, while handedness had no clear association to the bias in and of itself,
when there was concurrent manual task, where the child was asked to “feed the baby,”
left-holders continued to hold on the left while middle- and right-holders switched to the
left, presumably to free their dominant hand. In sum, findings provide a preliminary
indication that the 3-to-5-year age range is an important transition period for the

development of the bias and for the first appearance of sex differences in characteristics

of infant holding.

ACKNOWLEDGMENTS

I would like to extend my thanks to friends, colleagues, professors, and staff of
the psychology department at Michigan State University. Throughout my undergraduate
and graduate career, I have been fortunate to experience many years of intellectual
stimulation, scholarly debate, and scientific rigor in the classroom, at colloquia, in the
hallways, and especially during happy hour.

I am indebted to my committee members, Lauren Harris, Hi Fitzgerald, Alex von
Eye, Alessandra Passarotti, and Laura Symonds, for their flexibility and comments and
suggestions during the various stages of this ordeal. Without their guidance, this
document would be much less than what it is now. I owe special thanks to Hi Fitzgerald
for providing access to the population on which this document is based and to Jessica
Barnes for overseeing the collection of the data.

Nothing I am able to write here will do justice to the gratitude I have for Lauren
Harris. He has been my advisor, mentor, collaborator, role model, and best friend
throughout it all. He has been unconditionally supportive and encouraging of my
professional and personal development. Any measure of success I may enjoy in my
career I owe primarily to his nurturance of my intellectual development (not to mention

the innumerable nutty donuts and fresh-baked caramel rolls he brought to the office).

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... vii
LIST OF FIGURES ......................................................................................................... ix
INTRODUCTION ............................................................................................................. 1
A Left-Side Bias for Holding Infants ................................................................................. 1
Features of the Bias ............................................................................................................ 2
Near-Universality ................................................................................................... 2
Reliability ............................................................................................................... 3
Proximity ................................................................................................................ 3
Selectivity .............................................................................................................. 4
Sex Differences. . ............................................................................................ 4
Explanations of the Bias .................................................................................................... 6
Handedness ............................................................................................................ 6
Posture ................................................................................................................... 7
Maternal Heartbeat ................................................................................................ 8
Infant’s Resting Head Position .............................................................................. 9
Attention and Selective Hemispheric Activation ................................................ 10
The Complexity of the Bias ............................................................................................. 11
Development of the Bias .................................................................................................. 12
Questions for the Current Study ...................................................................................... 15
Age of Onset ........................................................................................................ 15
Role of Handedness ............................................................................................. 15
Relation between establishment of handedness and
establishment of side hold as a style of holding ................................ 16
Flexibility of holding-side bias when one hand is required
for a skilled act ................................................................................... 16
Role of Footedness and Posture .......................................................................... 17
Role of Sex of Child ............................................................................................ 18
Advantages that the Study of Young Children Offer for Understanding
the Origins of the Holding-Side Bias ........................................................................ 20
Hypotheses and Predictions ............................................................................................. 21
Hypotheses .......................................................................................................... 21
Prediction 1 .......................................................................................................... 22
Prediction 2 .......................................................................................................... 22
Prediction 3 .......................................................................................................... 22
Prediction 4 .......................................................................................................... 22
Prediction 5 .......................................................................................................... 22
Prediction 6 .......................................................................................................... 22
METHODS ...................................................................................................................... 23
Participants ....................................................................................................................... 23
Recruitment and Consent ................................................................................................. 24

Data Collection and Design ............................................................................................. 25

Infant-Holding Tasks ........................................................................................... 26
Handedness and Footedness ................................................................................ 27
Missing Data ........................................................................................................ 29
RESULTS ........................................................................................................................ 30
Hold Location .................................................................................................................. 30
Trial 1: Preferred Doll ........................................................................................ 3O
Trial 2: Non-Preferred Doll ................................................................................ 31
Trial 3: Preferred Doll, Holding and Feeding .................................................... 33
Hold Proximity ................................................................................................................. 35
Hold Duration .................................................................................................................. 39
Trial 1: Preferred Doll ........................................................................................ 39
Trial 2: Non-Preferred Doll ................................................................................ 41
Handedness and Footedness ............................................................................................ 44
Handedness .......................................................................................................... 44
Footedness ........................................................................................................... 50
DISCUSSION .................................................................................................................. 54
Summary of Results ......................................................................................................... 55
Prediction 1 .......................................................................................................... 55
Prediction 2 .......................................................................................................... 56
Prediction 3 .......................................................................................................... 57
Prediction 4 .......................................................................................................... 57
Prediction 5 .......................................................................................................... 58
Prediction 6 .......................................................................................................... 58
Consistency of Hold Side .................................................................................... 60
Limitations and Future Directions ....................................................................... 61
APPENDICES ................................................................................................................. 64
Appendix A ......................................................................................................... 64
Appendix B .......................................................................................................... 67
Appendix C .......................................................................................................... 70
Appendix D ......................................................................................................... 74
Appendix E .......................................................................................................... 76
REFERENCES ................................................................................................................ 78

vi

LIST OF

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

Table 11.

Table 12.

Table 13.

Table 14.

Table 15.

Table 16.

Table 17.

TABLES

Participant Demographic Information .............................................................. 24

Handedness Tasks with Motor Characteristics ................................................. 28

F ootedness Tasks with Motor Characteristics .................................................. 29

Trial 1: Preferred Doll — Hold Location Frequencies for Boys and Girls ....... 31

Trial 2: Non-Preferred Doll — Hold Location Frequencies for

boys and girls ................................................................................................... 32

Contingency Table for Hold Location on Trials 1 and 2 .................................. 33

Trial 3: Preferred Doll While Feeding for Boys and Girls

According to Hold Location ............................................................................ 34

Contingency Table for Hold Location on Trials 1 and 3 .................................. 35

Frequencies of Hold Proximity for Boys and Girls for Trials 1 to 3 ................ 36
Contingency Table for Hold Proximity on Trials 1 and 2 .............................. 37
Contingency Table for Hold Proximity on Trials 1 and 3 .............................. 38
Hold Proximity as a Function of Hold Location for Trials 1 to 3 .................. 38

Trial 1: Preferred Doll Hold Duration for Boys and Girls
According to Hold Location ........................................................................... 39

Trial 1: Preferred Doll Hold Duration for Boys and Girls
According to Hold Proximity ......................................................................... 40

Trial 2: Non-Preferred Doll Hold Duration for Boys and Girls
According to Hold Location ........................................................................... 42

Trial 2: Non-Preferred Doll Hold Duration for Boys and Girls
According to Hold Proximity ......................................................................... 42

Mean Hold Duration Scores for Boys and Girls According to
Hold Location for Trials 1 and 2 .................................................................... 43

vii

Table 18. Mean Hold Duration Scores for Boys and Girls According

to Hold Proximity ........................................................................................... 44
Table 19. Hand Use Frequencies Across Tasks .............................................................. 45
Table 20. Correlations and Scale Characteristics for Handedness Items ........................ 47

Table 21. Mean Handedness Scores for Boys and Girls According to Hold

Location and Trial Number ............................................................................ 49
Table 22. Foot Use Frequencies Across Tasks ............................................................... 50
Table 23. Correlations and Scale Characteristics for F ootedness Items ......................... 51

Table 24. Mean Footedness Scores for Boys and Girls According to
Hold Location and Trial Number ................................................................... 53

viii

LIST OF FIGURES

Figure 1. Trial 1: Interaction Between Sex and Proximity for Hold Duration .............. 41
Figure 2. Distribution of Total Handedness Scores ........................................................ 48
Figure 3. Distribution of Total Footedness Scores ......................................................... 52

ix

INTRODUCTION

The act of holding an infant serves many functions beyond simple transportation.
It is in the arms of the caregiver that, among other things, the infant is whisked away
from danger, carried to shelter, fed when hungry, and calmed when upset. The act of
holding is the primary means by which the parent-infant bond (infant-parent and parent-
infant attachment) is established (e. g., Bowlby, 1958; De Casper & Fifer, 1980; Feldman,
2003; Robson, 1967) and by which behaviors known to influence the infant’s cognitive
and socio-emotional development are facilitated. Examples include mutual gazing and
face-to-face communication (e.g., Cohn & Tronick, 1988; Fogel, Messinger, Dickson, &
Hsu, 1999), infant touching and caressing (e.g., Field, 2001), synchronous behaviors
(e.g., F eldman, 2003), and mother-infant vocalizations (e.g., Papousek & Papousek,
1986, 1989).

Given its importance for infant development, an understanding of the factors
underlying the act of holding represents an important goal for research. Other than
research on nursing behavior (e.g., Lonstein & Stern, 1999; Nightingale, 1860; Stables &
Hewitt, 1995) and an early investigation of holding styles by Rheingold and Keene
(1965), infant-holding, with one exception, has not been well-studied.

A Left-Side Bias for Holding Infants

The exception is a small but vigorous body of work by researchers who study
laterality. These researchers became interested in infant-holding on discovering that most
adults hold infants asymmetrically and in the same direction, with the infant’s head and
upper body to the left of the holder’s midline. The bias occurs whether the infant is held

in one arm or both arms in the cradling position, so that the infant faces the holder. This

left-side bias was reported as early as the mid-18th century (Harris, 2002) and can be seen
in relics and artifacts dating back nearly 2000 years (Gritsser, 1983).
Features of the Bias

Near-Universality

While the bias is most often reported for mothers with their newborn infants,
where about 80% hold to the lefi (e.g., Salk, 1960; Trevathan, 1982), it is not restricted to
this group. It is seen in fathers (Bogren, 1984; Dagenbach, Harris, & Fitzgerald, 1988)
and men and women in general (Harris & Fitzgerald, 1985) and in nulliparous college
students of both sexes (Bundy, 1979; Saling & Tyson, 1981). It also is culturally
ubiquitous, appearing in Western as well as Eastern countries, and in people fi'om
industrialized and non-industrialized cultures alike (Britsser, 1981; Harris, Almerigi, &
Kang, 2003; Harris, Jentoft, & Almerigi, 2002; Saling & Cooke, 1984). It even has been
observed in non-human primates (Hatta & Koike, 1991; Manning & Chamberlain, 1990;
Salk, 1960). Under what Manning and Chamberlain (1990) call “the most harmonious
assumption that left-side holding is homologous in apes and humans,” it would indicate
that “a strong left-side cradling preference may have originated as early as the common
ancestors of African apes and humans (about 6-8 million years ago)” (p. 1226).

Remarkably, elicitation of the bias does not require an actual infant to be held. It
occurs when participants are asked to hold a doll that looks like an infant (e. g., Manning
& Chamberlain, 1991; Saling & Bonert, 1983) or to hold a pillow or ball and to imagine
that it is an infant (Weiland & Sperber, 1970; Sperber & Weiland, 1973). It is even found
when adults are asked to imagine themselves holding an infant, without the aid of a

physical object (Almerigi, Carbary, & Harris, 2002; Bogren, 1984; Harris, Almerigi, &

Kirsch, 2000; Nakamichi & Takeda, 1995). Furthermore, when mothers’ imagined-
holding is directly compared to the actual holding of their own infant, the two are in “full
agreement” (Bogren, 1984, p. 15), meaning that the mothers imagined themselves
holding their infants on the same side as they actually held their infants.

Reliability

As previously noted, the bias is repeatedly found across diverse samples of
individuals, indicating its reliability at the population level. It is spatially reliable, in that
it is found across diverse contexts and locations, and temporally reliable, in that it has
been observed over hundreds of years.

The bias also is reliable at the individual level. Most people (> 85%) are
consistent in the side to which they hold, whether the testing interval spans 20 minutes
(W eatherill, Ahnerigi, Levendosky, Bogat, & Harris, 2002), an hour (Almerigi, Van
Hooren, Bulman-Fleming, & Harris, 2002; Trevathan, 1982), a week (Bogren, 1984), 4
months (Harris, Almerigi, Carbary, & Fogel, 2001), or even 12 months (Dagenbach, et
a1., 1988). As just noted, people remain consistent even across different methods of
holding (imagined and actual) (Bogren, 1984). Taken together, the left bias is
remarkably consistent.

Proxian

Another noteworthy feature of the left-side hold is its proximity to the body. de
Chateau (1991) compared mothers who held on the left with mothers who held on the
right. The former held their infants closer, the latter farther away. Inasmuch as
proximity relates to such aspects of the holder-infant relationship as attachment, maternal

depression, and infant self-regulation (e.g., Feldman, 2003), side of hold may serve as an

important index of the psychosocial welfare or interaction of the infant and/or holder (de
Chateau, 1991; Weatherill et al., 2002).
Selectivity

The left-side bias does not seem to be merely an expression of a general bias for
holding on the left. Adults show it for holding infants, pretend-infants (dolls, pillows),
and imagined infants but not for inanimate objects such as books (Alley & Kolker, 1988;
Scheman, Lockard, & Mehler, 1978), balls (Weiland & Sperber, 1970), or bags of
groceries (Rheingold & Keene, 1965). Depending on the object, a bias is usually absent
or to the right. For example, in a study of book-carrying style, Alley and Kolker (1988)
found that a majority of adults held to the right. For another example, Ahnerigi et a1.
(2002) asked two groups of college students to imagine themselves holding an object and
then an infant. For one group, the object was a simple shoebox; for the other it was a
fragile vase. Both groups showed significant opposite-side biases for the infant and the
object: to the left for the infant, to the right for the object. Even more compelling is that
the same physical stimulus can elicit opposite-side biases depending on how the holder
views it. In the study by Weiland and Sperber (1970), cited above, which instructed
adults to hold a pillow close to their chest, the left-bias was found only when the adults
were asked to imagine that the pillow was an infant. Otherwise, most adults held the
pillow in the middle or to the right.
Sex Differences

Finally, although the left-side bias is found in both sexes, men’s and women’s
biases differ in strength and reliability. In some studies, men’s bias is as strong as

women’s (e.g., Harris et al., 2001), but just as often it is weaker (Dagenbach et al., 1988),

absent (e. g., Lockard, Daly, & Gunderson, 1979), or even reversed, so that the bias is to
the right (e.g., Britsser, 1981; Manning, 1991). One investigator who found large sex
differences in his sample even proposed calling left-side holding a female adaptation for
the role as primary caretaker (Manning, 1991). Consistent with these differences, there is
some evidence that women are more selective in their display of the bias. For example,
in the shoebox-vase study mentioned previously (Almerigi et al., 2002), the women’s
holding-side patterns differentiated between the infant and the objects more clearly than
the men’s. That is, women, regardless of their side of hold for the infant, were more
likely than men to switch to an opposite-side hold for the object.

The description of the left-side bias as a female adaptation fits with psychological
and anthropological studies showing that women usually are more involved in infant care
(e. g., Barry & Paxson, 1971). They also are more aroused by infants, more likely to pick
them up and hold them, better at recognizing their cries, and better at soothing and
quieting them (Fogel, 2001). In addition, women are more emotionally expressive and,
as Maccoby (1990) concluded after reviewing the literature, more “relation ” and
“reciprocal” in their style of social interaction. This difference is perhaps reflected in
what Rheingold and Keene (1965) found for infants under 1 year of age — that more
mothers hold in the chest-to-chest, or facing-toward, position, whereas more fathers hold
facing away. For women, holding an infant may be a more social-interactional act, even
when the hold is primarily for transport.

Certainly, these sex differences may arise from social-cultural practices that teach
us how to think and behave as males and females. They also may be partly biological,

what Panksepp (1998) called a difference in “nurturance circuits,” implying that they are

adaptive and rooted in evolutionary selection. If so, and if the left-side hold itself is
adaptive, it makes sense that it is stronger and more reliable in women.
Explanations of the Bias

In sum, the left-side bias is not an obscure, isolated phenomenon, occurring under
unusual circumstances. It is ubiquitous and robust. What is especially intriguing about the
bias, however, and what has contributed to its being studied for over four decades (since
Salk, 1960), is that it is not easily explained. A variety of explanations have been
proposed and tested. Five of them are summarized below.
Handedness

The most obvious explanation is that the bias is the direct product of handedness,
that is, that most people hold on the side that keeps their dominant hand free for other
activities, for example, for engaging the infant, for holding and manipulating feeding
utensils, or for performing other duties. By this reasoning, right-handers would hold on
the left, left-handers on the right. This explanation itself has a long history, and it appeals
to common sense. According to one nineteenth-century investigator who reported that
left-side holding was the mode, “an intelligent mother. . .will tell you that she must keep
her right arm free for the other children, and for her various household duties”
(Buchanan, 1862, pg. 162; cited in Harris, 2003). Likewise, in Bali, where left-side
holding also is the mode (Harris, Jentoft, & Almerigi, 2002), the anthropologist Margaret
Mead (1942) noted that for the Balinese mother, left-side holding would leave “her own
right hand free” (p. 13). The handedness explanation, however, was abandoned after
repeated demonstrations that right- and left-handers either had nearly identical left biases

or that the bias in lefi-handers was only slightly weaker but not reversed (e. g., Ginsburg,

Fling, Hope, Musgrove, & Andrews, 1979; Harris, Almerigi, & Kirsch, 2000; Salk, 1960;
Weiland, 1964). It is also difficult to see how handedness can account for those studies
where men show a weaker left bias than women, much less a reversed bias, considering
that handedness patterns are very similar in men and women, the only difference being
that left-handedness is slightly more common in men (~12% vs. 9%). The handedness
explanation also cannot explain the selectivity of the bias. Inasmuch as holders require
the use of their dominant hand whether holding infants or objects, a left bias should occur
for both. It is unlikely that handedness can explain how the same individual can show a
left-bias and right-bias for the same object, depending on the task (e. g., Weiland &
Sperber, 1970).

.13_O_§_tu_r§

When college students are asked to explain their bias, many have referred not to
handedness but to comfort, that is, that they preferentially hold on a certain side because
it feels more “comfortable” or “natura ” on that side (Harris, Kang, & Almerigi, 2003).
When asked to hold on both sides and to compare their comfort levels directly, nearly all
rate their preferred-side hold as more comfortable (Almerigi et al., 2002).

It is unclear what would make one side feel more comfortable or natural. One
possibility is that the act of holding is less an act of hand skill than it is what Saling and
Cooke (1984) called a “tonic postural configuration” (p. 335): how individuals normally
arrange, or configure, the parts of their body, especially while in a state of repose. For
the act of holding, some aspects of posture might be more important than others,
depending on whether one is standing or sitting. For example, most people, when they

stand in a relaxed position while not holding anything, shift their weight slightly to one

side; that is, they extend one leg for support and slightly flex the other. Erber, Almerigi,
Carbary, and Harris (2002) reasoned that people more comfortable with weight on one
side might be more comfortable holding a baby on that side. If so, andif most people
“balance” on the left, that might explain the left-bias for holding. It also might explain
why the women in one study not only rated their preferred side as more comfortable, but
why left-holders’ overall comfort ratings were significantly higher than right-holders’
(Almerigi et al., 2002).

Using an imagine-hold test with right-handed college students and a test of foot
preference for balance to measure postural bias, Erber et a1. (2002) found a modest
relation between posture and holding-side bias. Fifty-five percent of the men and 68% of
the women held on the left (only the women’s score was significantly different from
chance), with left-holders showing a weaker bias for balance than the right-holders. The
correlation between scores, however, was significant only for women (,1; = .252, p < .05).
The authors suggested that, inasmuch as women generally have more experience with
infants, postural bias may play a greater role for people with more experience because
with more experience comes a clearer sense of what is the more comfortable side.
Maternal Heartbeat

Salk (1960, 1961) proposed that the bias is a product of the left-side location of the
mother’s heart. He called the left-side hold “an instinctive response evolved from a need
on the part of the infant to experience a continuation of the maternal heartbeat rhythm, a
familiar sensation from intrauterine life” (Salk, 1960, p. 170). The left side therefore is
preferred because the infant is nearer the precordial area where the heartbeat is stronger,

soothing not only the infant but also the mother, who, “by virtue of having contact in this

area,” has the feelings of her own heartbeat reflected back (Salk, 1961, p. 745). Salk
(1973, p. 27) went on to suggest that “the time immediately after birth is a critical period
during which the stimulus of holding the infant releases a certain maternal response,” a
period he likened to the imprinting phenomenon found in such creatures as ducks and
chickens as demonstrated by Lorenz (1964) and other ethologists.

The heartbeat explanation has been questioned on several grounds (e. g.,
Dagenbach et al., 1988; Tumbull & Lucas, 2000). Given its focus on pregnancy and the
birth experience, it cannot explain why the bias appears not only in mothers but also in
individuals with little or no experience with infants, including nulliparous women and
male college students. Still, Salk must be credited with being the first to propose that the
left-side hold is adaptive, or beneficial, not only for the infant but for the holder as well.
Furthermore, consistent with the hypothesis, the left-side hold is at its peak for newborns
and young infants (Rheingold & Keene, 1965). A more theoretically useful way to
describe the role of the maternal heartbeat therefore may be as a “trigger” for a causal
bio-regulator related to comforting and feeding (Fitzgerald, Harris, Barnes, et al., 1991).
Infant’s RestingHead Position

So far, the focus has been on the attributes of the holder. Attributes of the infant
also deserve consideration. One possible way for the infant to contribute to the bias is by
its resting head position. While lying supine, most infants tend to turn their head to the
right, even when the head is initially held at the midline for a brief time and then released
(Harris & Fitzgerald, 1983; Turkewitz & Birch, 1971; Turkewitz, Gordon, & Birch,
1965). As such, it has been suggested that an infant in the rightward head position would

predispose the holder to hold on the left. That would facilitate face-to-face contact with

the holder as well as the initiation of breast-feeding, since the infant’s head already would
be turned toward the holder’s face and body. Although this explanation has found some
support (Ginsburg et al., 1979), the left bias persists independent of the infant’s head
position (Bundy, 1979; Dagenbach et al., 1988; Saling & Tyson, 1981). Indeed, in the
study by Dagenbach and colleagues (1988), the greatest continuity in lateral preferences
for holding occurred between the first and second months after birth, a time when infants
themselves were often inconsistent in their head orientation (Cornwell, Barnes,
Fitzgerald, & Harris, 1985).
Attention and Selective Hemispheric Activation

The last explanation proposes that the bias is a product of selective neural
activation accompanying the act of holding an infant. Drawing on evidence for lateral
differences in hemispheric specialization for the perception and recognition of faces (e.g.,
De Renzi, Faglioni, & Spinnler, 1968; Kanwisher, McDermott, & Chun, 1997; Sergent,
Ohta, & MacDonald, 1992) and emotion (e.g., Best, Wormer, & Queen, 1994; Borod,
Andelrnan, Obler, et al., 1992), this explanation posits that the acts of perceiving the
infant and responding emotionally to its facial expressions, vocalizations, and cries
activate a greater network of neural systems in the holder’s right hemisphere than the left
hemisphere (e.g., Natale, Gur, & Gut, 1983), thereby driving the holder’s attention to the
opposite, left side of space (Renter-Lorenz, Kinsboume, & Moscovitch, 1990). The left-
side hold thus is compatible with the direction of the holder’s attention. The explanation
further assumes that this same state of activation is adaptive by enhancing the holder’s
ability to interpret and monitor the infant’s emotional state. So far, this hypothesis has

found modest support from visual tests of attention (e.g., Harris, Almerigi, Carbary, &

10

F ogel, 2001; Vauclair & Donnot, 2003, unpublished data) and mixed support from
auditory tests. The latter have compared left- and right-side holders’ recognition of the
emotional tone of dichotically-presented speech. Tests using emotionally-toned
sentences found no differences (Sieratzki, Roy, & Woll, 2002; Tumbull, 2002; Tumbull
& Bryson, 2001), but a test using emotionally-toned words gave more promising results,
finding that only left-side holders showed a significant left-ear (right-hemisphere)
advantage and were more accurate overall (Almerigi et al., 2002). Even in this case,
however, only 9% of the variance was explained by hemispheric asymmetry for emotion
perception.

The Complexity of the Bias

In summary, the holding-side bias is not easily explained by any single
mechanism. The fundamental reason is probably that, like most psychological
phenomena, it is multiply-determined, so that although any one factor may be necessary
for its occurrence, no one factor is sufficient. This is to be expected if the bias is
evolutionarily adaptive (e.g., Cosmedes & Tooby, 1994). That is, it makes sense that a
biologically and psychologically important phenomenon is redundantly specified (cf.
Fodor, 1983).

Aside from the complexity of the phenomenon, the difficulty of explanation also
has to do, at least in part, with methodological limitations in how it has been assessed. In
many studies, investigators have simply reported the existence and magnitude of the bias
and have made only modest attempts to measure its association with any behavior or
attribute of the holder or the infant. Moreover, where these efforts have been made, it

often has been by oversimplifying the attribute through dichotomous classification. This

11

is especially true for the handedness of the holder. Most studies have treated handedness
as a dichotomous variable by classifying individuals into two groups, right-handers and
left-handers, thus not taking variations in the degree, or strength, of handedness into
account, as measured by a multi-item questionnaire or a perforrnance-based measure of
hand use. This practice is especially problematic with respect to left-handers because of
their greater heterogeneity. As a group, they are less strongly lateralized than right-
handers, but they also comprise at least two roughly equal-size subgroups varying in
degree of handedness. On directional tests of attention, only strong left-handers have
shown reverse biases from right-handers (Peters & Servos, 1989). Conceivably, then, a
larger, if still limited role, for handedness might be found if, especially for left-handers,
degree of handedness were taken into account.

In holding-side studies that have assessed the holder’s handedness, many also
have simply classified handedness by self-report (e. g., Bundy, 1979; Salk, 1960) or by
assessing the hand used for writing (e. g., Bogren, 1984) instead of by using a multi—item
questionnaire. Handedness classifications sometimes also have been reported without
describing the decision rule used (e.g., Manning & Chamberlain, 1990). Finally, so far,
no study has used perforrnance-based measures that assess hand differences on tests of
proficiency. With all these considerations in mind, the role of handedness in the holding-
side bias should not be ruled out.

Development of the Bias

With the left bias repeatedly confirmed in adults, at least in women, the question

arises as to its development. When does it first appear, how stable is it, what are the

contributing mechanisms, and is the process the same or different in boys and girls? So

12

far, only three studies have been reported. The first and most comprehensive was by de
Chateau and Andersson (1976), who assessed 305 Swedish children (135 girls, 170 boys)
in 8 age groups ranging from 2 to 16 years. The children were asked to enter an
observation room where a doll had been placed on a bed, to pick up the doll and pretend
it was a newborn baby, to sit in a chair while holding the doll, and finally to carry the doll
to an adult observer seated across the room. For each child, the observer noted how the
child held the doll while sitting and while carrying it. The three possible ratings were
“left” (doll in arms with head turned to the left), “right” (doll in arms with heatumed to
the right), and “in hands” (doll in hands, not in arms, regardless of head direction).

The two-year olds were reported to have difficulty understanding the directions.
The four-year-olds had no difficulty but nonetheless failed to show a bias in either
direction; instead, the majority (~70%) held and carried the doll in their hands
(“in-hands”) and away fi'om their body. The first appearance of the bias was in
six-year-old girls, with 79% holding on the left , 10% on the right, and only 11% in
hands. In six-year-old boys, 51% held in-hands, only 32% held on the left, and 17% held
on the right. For girls, a strong left bias persisted for all age groups; for boys, it appeared
only by age 16, and then by a smaller margin. For both sexes, increases in left-side holds
came largely as in-hand holds declined. Unfortunately, handedness was not assessed.

In the second study, Saling and Bonert (1983) asked 53 South African preschool
children (mean age = 59 months) to hold a life-like doll “so that it can go to sleep” (p.
149). Sixty-six percent of the children spontaneously held the doll to their left while the
remaining held to the right. That no children held in the middle or “in-hands” leaves

open the question of whether these children spontaneously held only on the left or the

13

right (either as a function of the experimental instructions or as a function of their own
natural tendencies) or whether the authors simply removed these children from analyses.
Further, the authors reported that the correlation between handedness and holding was
not significant (r = .02), but they did not identify their handedness items or decision rules.
In this study, boys were not tested.

In the last study, Souza-Godeli (1996) assessed holding-side bias in 520 Brazilian
children between the ages of two and six. Each child was presented with a doll and was
asked to hold it “as if it were a baby.” Along with the holding task, the child’s hand
preference was assessed on 4 tasks (hammering, drawing, holding a spoon, and throwing
a ball), and the child’s foot preference was assessed on a single task (kicking a ball). The
result was that 87% of the children “showed dexterity,” presumably meaning that they
were right-handed, and the group as a whole showed a left-bias for holding the doll. Side
of hold was “unrelated” to either age or sex. Based on the results, Souza-Godeli (1996)
called the holding bias an “innate behavior or at least a behavior emerging in the early
stages of human ontogenesis” (p. 1422).

This last study presents more problems than the others. First, data were not
reported for each age group or sex. Instead, the author simply reported a non-significant
Chi-square statistic for age by hold side and noted the absence of any age differences,
even though she reported in the discussion that, similar to what de Chateau and
Andersson (1976) found, the two-year-olds had difficulty following the instructions. The
statement that the results were “unrelated” to sex, likewise, is difficult to interpret in the
absence of separate statistics for boys and girls and the different age groups, especially

considering de Chateau and Andersson’s (1976) finding of an interaction between sex

14

and age. Finally, for the handedness measure, the Chi-square statistic that was reported
indicated that the author dichotomously classified the sample into right-handers and left-
handers. Scores for individual laterality items and score ranges also were not reported,
and no mention was made of the single footedness measure.
Questions for the Current Study

Prior studies show a holding-side bias in children, but they leave many questions
about the development of the bias unanswered. The current study sought answers to
some of these questions, four sets of questions, in particular.
Age of Onset

The first set of questions has to do with age of onset. In combination, the
evidence suggests that the left-side bias first appears somewhere between three and six
years of age, but why was it substantially earlier in the Brazilian and South African
children than the Swedish children, and why were there age differences in the Swedish
children but not the Brazilian children? The answers to both questions may have less to
do with nationality and more to do with the greater complexity of the method used. In
the current study, therefore, an age range was chosen that covered the period of first
reported appearance of the bias, and the method used was designed to be understandable
to this age group.
Role of Handedness

The second set of questions pertains to the role of handedness. For a study of the
role of handedness in young children as distinct from a study of its role in adults, two

different kinds of questions must be distinguished.

15

Relation between establishment of handedness and establishment of side hold as a
style of holding. In studies with adults, the only possible comparison, understandably,
has been between individuals with already established handedness. Unlike young
children, nearly all adults show a lateral hold, that is, to one side or other, instead of an
“in hands” hold, with the infant (or doll) held in the middle, or midline, of the holder’s
body (e.g., de Chateau & Andersson, 1976). Weatherill and colleagues (2002) found that
of over 500 holds observed in mothers with their one-year-old children, fewer than 10%
were in the middle, and most of these were observed while the mother was sitting rather
than standing.

Inasmuch as holding an infant is a motor act, a reliable holding-side bias,
irrespective of direction, presumably would not be possible until the child has developed
a reliable lateral bias for hand and hand-and-arrn acts of the kind that conventionally
comprise measures of handedness. For the current study, then, the first question is
whether a necessary condition for holding the doll to one side, in contrast to holding “in
hands” at the body midline, is the establishment of a lateral bias on the handedness
measure. If it is a necessary condition, the next question is whether the direction of
handedness is related to the direction of hold.

Flexibility of holding-side bias when one hand is required for a skilled act. For
adults, as already noted, the similarity of holding-side patterns for right- and left-handers
suggests that the choice to keep the dominant hand free for other tasks is not a major
contributor to side of hold. But circumstances also were noted where it presumably
would play a role, prominently, when the infant must be fed with a utensil requiring

motor control for which the dominant hand is more suited. The role would be different

16

for right-handers and left-handers. For right-handers, the large majority of whom hold on
the left, the right hand would already be free; it is the minority who hold on the right who
would have to shift to the left for feeding. For left-handers, a majority of whom also hold
on the left, a proportionately larger number therefore would have to change sides to free
the dominant hand. Only right-holders therefore would have their dominant hand free.
This means that a minority of right-handers and a majority of left-handers would have to
be flexible in their choice of holding side, depending on the circumstances. For children,
the same questions about flexibility can be posed. For example, for right-handed children
(based on their predominant hand use on a variety of tasks), if the majority hold the doll
on the left, it seems reasonable to suppose that they will continue to hold on the lefi if
asked to feed the doll. It seems less reasonable to suppose that those who hold on the
right will change to the left. That may require a measure of flexibility still beyond young
children’s capabilities.
Role of Footedness_ and Posture

The third set of questions pertain to footedness and posture. The score for the
single measure of footedness — kicking a ball - used in Souza-Godeli’s (1996) study may
not have been reported separately because it was treated as a test of skill like the tests for
handedness and thus was incorporated into the author’s overall estimate of laterality. As
such, it may have added little to the mix for the purpose of testing the relation of laterality
to side of bold. For the current study, as in the adult study by Erber et al. (2002), some of
the footedness items were used for this same purpose, but other items were used for a
different purpose - to assess the role of posture. These items were those designed to

assess foot and leg preference for stability and balance. As in the adult study, for the

17

doll-holding task, the children were tested in a standing position in order to assess the
relation between directional bias for stability and balance and the adoption of a lateral-
style hold as well as the direction of the hold. In addition, for one of the stability-balance
measures, a test of proficiency was added.

Role of Sex of Child

The last set of questions pertain to sex differences. What role does sex play in the
development of the bias, and why were sex differences pronounced in the Swedish
children but absent in the Brazilian children? The latter question cannot be answered
without more information about the results of the Brazilian study. But in light of
evidence reviewed earlier of the breadth of social-emotional-perceptual functions served
by infant-holding along with the evidence of sex-related differences in adults in interests
and skills important for infant care, it is reasonable to predict that similar differences will
be found in children.

There is, in fact, abundant evidence of sex differences in young children having a
possible relevance for the doll-holding task. Generally speaking, these are studies
showing differences in the general orientation of boys and girls toward other people. For
example, in a review of these differences, Haviland and Malatesta (1981) noted that
whereas “girls and women establish and maintain eye contact more than boys and men”
(pg. 189), boys and men gaze-avert more frequently, a difference found as early as 6
months. Further, in the first few days of life, girls orient to faces and voices more
frequently on average than do boys (Haviland & Malatesta, 1981) and, by at least 6
months of age, girls have better memory for faces and are more skilled than boys in

discriminating between similar faces (e.g., Fagan, 1972). These latter differences,

18

however, are more consistent in older children (Hall, 1984; Haviland & Malatesta, 1981;
Maccoby & J acklin, 1974; McGuinness & Pribram, 1980).

Girls and boys also differ in the nature of their social interactions. Several lines
of research have found that infant girls are more responsive, and even more sensitive, to
social cues than are boys (Gunnar & Donahue, 1980; Gunnar & Stone, 1984). Gunner
and Donahue (1980) found that mothers were equally likely to initiate social interactions
with their sons as with their daughters, but daughters were more responsive than sons to
their mother’s verbal requests. While the origins of these differences lie beyond the
scope of the current study, it is interesting to note that these differences have been
observed across several diverse cultures (Whiting and Edwards, 1988).

Of perhaps special relevance for the study of lateral biases in doll-holding in
children are studies of alloparenting, or “pretend parenting” during play. Although this
form of play can occur in both sexes, girls are far more likely to be observed in such play
(Lever, 1978; Meyer-Bahlburg, Sandberg, Dolezal, & Yager, 1994; Pitcher & Schultz,
1983). That alloparenting has been observed in non-human primates, especially in young
females who have not yet had their first offspring (Nicolson, 1987; Pryce, 1993),
evolutionary explanations for the origins of the sex differences in social interactions in
the context of infant care become more likely.

In light of this evidence, the current study asked whether and to what extent boys
might differ from girls not only in the strength of the side bias but in its flexibility in a
concurrent task requiring use of one hand for infant care. It also asked whether any such

differences are related to age and to the direction and degree of handedness and posture.

l9

Given the sex role differences between boys and girls, it also sought to assess any
differences in the proximity and duration of the hold.

Advantages that the Study of Young Children Offer for Understanding

the Origins of the Holding-Side Bias

In contrast to studies of adults, studies of young children — those in the 3- to 6-
year period — offer three main advantages for helping to understand the origins of the
holding-side bias. First, whereas handedness is already established in adults, in 3- to 6-
year-olds it is still undergoing development at least in degree if not necessarily in
direction (of. McManus et al., 1988). As a result, even though adults show variations in
strength of handedness, the variations will be greater in young children, since some will
be further along in the process than others. In fact, for degree of handedness as measured
by the number of different tasks for which the same hand leads, some longitudinal studies
find changes until ages eight or even nine (Bruml, 1972). Assuming that the underlying
neural bias is relatively constant, it may be that improvements in motor control allow for
better expression of the bias and that it takes time for children to discover and consolidate
their preferences across tasks, especially for new, complex tasks like writing and tool-use
(see Harris & Almerigi, 2001).

Second, individual differences in the development of general motor control
necessary for lateralized hand use therefore would be expected to contribute to young
children’s overall greater variability. There also is evidence that variability or instability
will be even greater in young children with a left-handed pattern. For example, where
direction of handedness was based on hand-use for writing, a cross-sectional study of 5-,

7-, and 10-year-olds showed that left-handers in every age group were less stable than

20

right-handers on a variety of other preference as well as proficiency measures (Bruml,
1972). Similarly, in a longitudinal study at 5, 8, and 11 years, hand preference at 5
predicted preference at 11 for only 74% of left-handers versus 97% of right-handers
(Fennell, Satz, & Morris, 1983).

Finally, as already noted, whereas on tests of holding side, adults normally
comprise two unequal-size groups — left-holders and right-holders -— prior studies of
children, especially young children, show greater variability, with left- and right-side
groups more similar in size and with the appearance of a third group — “in hands” — less
often seen in adults. In sum, in young children, all of the bio-behavioral-action systems
that may be contributing to the holding-side bias are still developing, at diflerent rates, in
drflerent children. As such, fiom a purely statistical standpoint, the greater variability of
the systems allows for more latitude and more opportunity to examine the inter-
relationships. This is one of the major guiding assumptions for the current study.

Hypotheses and Predictions
Hypotheses

Inasmuch as infant holding is a motor behavior, the development of motor
systems must precede the behavioral manifestation of an infant hold. Likewise, the
development of lateralized motor biases must precede lateralized biases in holding. Both
conditions are necessary (but not necessarily sufficient) for the bias to occur. Further,
given the biological and psychological differences between boys and girls in behaviors
related to infant care (reflecting other contributing, possibly necessary, conditions), boys’
and girls’ holding-behavior patterns will reflect these differences. Six directional

predictions were made to test these hypotheses.

21

Prediction 1

Like adults, children as a group will show a left-side bias for holding a doll, but it
will be smaller in magnitude due to the presumably still incomplete development of the
underlying motor systems.
Prediction 2

Degree, or strength of handedness, as a proxy for motor development, as
measured by the extent of same-hand use across hand-use tasks, will predict lateralized
holding, regardless of side.
Prediction 3

As with adults, direction of handedness will be related to but not completely
predict direction of hold.
Prediction 4

Degree, or strength, of handedness, again as a proxy for motor development, will
predict flexibility in hold side on a task that requires holding an infant while performing
a skilled motor task.
Prediction 5

Postural bias, as measured by strength and direction of footedness, will augment
the predictive power of handedness for side-of-hold bias.
Prediction 6

Compared to boys, girls will show a stronger lateral holding-bias, a stronger left
bias, a closer and longer hold, and greater flexibility in arm choice for holding when

presented with a concurrent hand-use task.

22

METHOD
Participants
Participants were 97 pre-school-age children who were recruited as part of a
larger study by Michigan State University Outreach and Engagement1 of the
effectiveness of the Genesee County “Ready, Set, Grow! Passport initiative, a
community—based early childhood intervention program designed to enhance the well-
being of young children.2 The sample for the current study consisted of 47 boys and 50
girls ranging in age from 43 months to 64 months (mean = 52.6).3 As shown in Table 1,
age and ethnicity were nearly equally distributed between the sexes. Forty-eight children
(49.5%) were African-American, forty-one (42.3%) were Caucasian, seven (7.2%) were
multi-racial, and one (1.0%) was Hispanic. These percentages correspond well to the
racial distribution of the overall program, with Caucasian children slightly
underrepresented (6.3%) and Afiican-American children slightly overrepresented (7.3%).
The other race/ethnicities differed by less than one percent. For the overall community,
the differences were slightly larger but in the same direction (see Barnes, 2003, for more

information).

 

' The data collection for the evaluation was conducted by Jessica Barnes (and several undergraduate
students) under the guidance of Hiram Fitzgerald, Director of Outreach and Engagement, Michigan State
University.

2 This program is an ongoing, population-based program that attempts to enroll all families in Genesee
County, Michigan, with children under the age of 5 years (Barnes, Determan, & Fitzgerald, 2003). At the
time of testing, nearly 9,000 children and their caregivers were enrolled. For member families, the Passport
program provides a broad spectrum of information on healthy child development through the use of fact
sheets and timetables for prenatal care, well-baby check ups, and immunizations. The Passport program
also provides access information to social services related to child development and incentives (coupons)
for the use of such services and activities that promote positive social, emotional, cognitive, and physical
development.

3 Eight children who were tested for the evaluation were excluded from the current study because of prior
brain injury (n=2), clinical psychiatric disorder (n=3), language difficulty (n=2), or refusal to complete the
tasks (n=1).

23

Table 1. Participant Demographic Information

 

 

Boys Girls Total

N 47 50 97
Age (months)

Mean (SD) 52.7 (4.7) 52.5 (5.5) 52.6 (5.1)

Range 44-64 43-62 43-64
Ethnicity (%)

Caucasian 19.6 22.6 42.3

Afiican American 25.8 23.7 49.4

Hispanic 1.0 0.0 1.0

Multi-racial 2.1 5.2 7.2

Asian, Native American, Other 0.0 0.0 0.0

 

Recruitment and Consent

The children and caregivers were recruited via a mail flyer sent by the “Ready,
Set, Grow!” Passport program to all the individuals enrolled in the program. Contact
information for positive respondents was provided by the Passport program to university
research assistants who then telephoned the respondent to briefly explain the purpose of
the evaluation and to schedule an interview at the respondent’s home.

Individuals who agreed to be interviewed were informed that participation was
voluntary, that they or their child could withdraw at any time without suffering any
negative consequences, and that all information gathered was confidential. Prior to any
data collection, the research assistant read two consent forms to the caregiver: one for the
evaluation of the Passport program (Appendix A), another for the data collected for the
current study (Appendix B). Caregivers who agreed to participate then signed the

consent forms. All caregivers gave consent.

24

Data Collection and Design

Children and caregivers were separately interviewed in their own home by two
research assistants. One assistant interviewed the caregiver while the other interviewed
the child. Both interviews were conducted simultaneously and in separate rooms. The
caregiver interview usually took place at a kitchen table while the child interview usually
took place on the living room floor or another space large enough for the perforrnance-
based tasks. All information from questionnaires and tasks was recorded by the research
assistants on a prepared answer form. In the case of questionnaires, the research assistant
briefly described the questionnaire and read each question and corresponding response
scale aloud.

The caregiver interview consisted of questions about family demographics and
history and the use of Passport information and services. The caregiver’s knowledge of
child development, parent confidence and efficacy, and caregiver/child activities also
were assessed (Barnes, 2003). Finally, information about the child was assessed by
reading each of the items from the Child Behavior Checklist for Ages 1'/2 - 5 (Achenbach
& Rescorla, 2001).

The child interview mainly consisted of simple behavioral tasks that assessed the
child’s behavior regulation, emotion regulation, social problem solving, and ability to
produce and perceive facial emotions (Barnes, 2003). In addition, there were 23 tasks
assessing the child’s infant-holding bias, handedness, footedness, and motor
coordination. Appendix C provides a complete listing of the caregiver and child tasks in
the order they were administered. Below is a fuller description of the tasks used in the

current study.

25

Infant-Holding Tasks

Two realistic-looking infant dolls (approximate age representation: 6 months)
were used to assess side-of-hold. Both dolls were purchased from a local toy store and
were identical in material (“natural touch” soft plastic “skin”), size (approximately 40 cm
long x 15 cm wide), weight (.726 kilograms)’, and apparel (plain white diaper). Each
(1011 was sex-neutral but differed slightly in facial features, hair texture, and skin color to
represent babies of two different races (Black and White). The use of dolls of different
races was not integral to the study; the intent rather was to slightly alter the task in order
to make the two holding-trials more discrete.

For the first hold trial, the interviewer placed the dolls side-by-side and directly in
front of the child. The child was then told to pretend that the dolls were real babies and
to point to the “baby” that he/she wanted to pick up. Once the child made a choice, the
interviewer removed the other doll from the child’s sight by putting it into a plastic toy
bin and then put the chosen doll directly in float of the child. The interviewer then said,
“Pretend the doll is a real baby, pick up and hold the baby and give it love.” Once the
child picked up and held the doll, the interviewer recorded the side to which the child
held it (to the left, to the right, in the middle, or alternating between left and right sides),
the proximity of the hold (doll touching body, doll near but not touching body, or doll out
and away from body), and the duration of the hold, or hold time. Appendix D shows the
coding form used by the interviewer. Hold duration, recorded by stopwatch, was the time
between initially picking up the doll and ending the hold (e. g., putting the doll down or

playing with it in a way that did not involve an arm hold).

 

’ The choice of a doll that was smaller and lighter than a normals 6-month-old infant was made so that a
young child who wanted to could comfortably and naturally hold it for a sustained period of time.

26

The procedure for the second hold trial was the same as for the first, except that
instead of asking the child to choose a doll to hold, the interviewer placed the non-chosen
doll from the first trial in front of the child prior to repeating the instructions on
pretending that the doll was a real baby and to pick it up and “give it love.”

Following the second hold, the interviewer placed a small bowl and spoon
directly in front of the child and asked the child to pretend that “the bowl has cereal in it
for the babies.” The child was then asked to stir the cereal with the spoon. After this, the
interviewer placed both dolls side-by-side in front of the child but in reversed position
from the first trial and again asked the child to point to the “baby” that he/she wanted to
pick up and feed. After making the choice and following the procedure of the first trial,
the non-chosen doll was placed out of sight and the chosen doll was placed directly in
front of the child. Again, the child was asked to pick up, hold, and feed the “baby.” For
this third trial, hold side, hold proximity, and hand used to feed the baby were recorded.
Hold duration was not recorded.

Handedness: and Footedness

Handedness and footedness were assessed with perforrnance-based tasks, as
shown in Tables 2 and 3. For handedness, 14 tasks were administered (including some
repetitions for reliability); for footedness, 9 tasks were administered (again with some
repetitions). Tasks of the kind used have been shown to be reliable for assessment of
handedness and footedness in pre-school-age children (see Harris & Almerigi, 2001).

As shown in Table 2, the handedness tasks varied in the quality of movement

(fine versus gross) and degree of skill required.

27

Table 2. Handedness Tasks with Motor Characteristics

 

 

Fine vs. Relative
Gross Degree of
Item Task / Instructions Motor Skill
1. Hang a small monkey onto a string of monkeys Fine Less
2. Hang another small monkey onto the string of monkeys Fine Less
3. Throw the ball to me Gross More
4. Throw the ball to me, again Gross More
5. Draw shapes Fine More
6. Point to a Lego Gross Less
7. Place one Lego on top of another Fine More
8. Push a care quickly from here to there Gross Less
9. Push a car slowly from here to there Gross Less
10. Push a car quickly from here to there, again Gross Less
11. Push a car slowly from here to there, again Gross Less
12. Use a spoon to stir a bowl of imagined cereal Fine More
13. Use a spoon to feed a doll Fine More
14. Draw a picture Fine More

 

As shown in Table 3, the footedness tasks also varied according to whether they

constituted tests of target-oriented actions or balance. The first five items (kick, stomp,

point with foot) are target-oriented; the last four (standing on one foot, hopping on one

foot) assess balance. For the balance items, both the foot used and the length of time

performing the task were recorded.

For all handedness and footedness items, the interviewer used the same response

format to record whether the child used the left hand or foot, right hand or foot, both

hands or feet, or alternated hands or feet. Appendix B shows a sample response form for

the handedness tasks. For the four balance items on the footedness test, both the foot

used and the length of time performing the task were recorded.

28

Table 3. Footedness Tasks with Motor Characteristics

 

 

 

Item Task / Instructions Motor Action
1. Kick a ball to me Target Oriented
2. Kick a ball to me, again Target Oriented
3. Kick the Lego tower over with your foot Target Oriented
4. Stomp on the bag with one foot Target Oriented
5. Point to the red Lego with your foot Target Oriented
6. Stand on one foot for as long as you can Balance
7. Stand on the other foot as long as you can Balance
8. Hop on one foot for as long as you can Balance
9. Hop on the other foot for as long as you can Balance
Missing Data

Because the data for the current study were collected during an interview with the
child, there were very few missing data points, and where data were missing, it was
usually because the interviewer neglected to record the child’s response or action. In
most such cases, only 1 or 2% of the data were missing for each variable. The variable
with the greatest amount of missing data was length of time hopping on the non-preferred
foot (17%). Over 50 percent of the children had no missing data (there were no sex
differences in this regard). Furthermore, no child had more than 20% missing data, and
only six (6%) were missing more than 10% of their data.

In the current study, missing data were replaced with values calculated using the
EM algorithm (Little & Rubin, 1987). Missing data for each variable were computed
separately for each sex and by using behaviorally similar and maximally correlated (all
r’_s < .10) predictor variables. In addition, outlier estimates were downweighted using a
normal distribution, and the convergence criterion for the estimates was 0.001. No

estimates took more than 20 iterations.

29

RESULTS

For each dependent and independent variable, descriptive statistics will be
presented first, followed by subgroup comparisons. Results will be presented in the
following order: hold location, hold proximity, hold duration, handedness, and
footedness (posture). The hold side, proximity, and duration results will be described
separately for each of the three trials.

Hold Location

To recapitulate, the presence and degree of side bias were assessed with three
doll-holding trials. Prior to the first two trials, the child was shown two dolls that
differed only in color (black vs. white) and was asked to point to the doll that he or she
wanted to hold. Then, on Trial 1, the child was asked to hold the chosen doll (Preferred
Doll) and, on Trial 2, to hold the other doll (Non-Preferred Doll). Prior to Trial 3, the
child once again was shown both dolls and was asked to point to the doll that he/she
wanted to hold. On Trial 3, the child then was asked to hold that doll while pretending to
feed it with a spoon (Preferred Doll/Feeding).

There were three possible hold locations: on the left, middle, or right. As these
data are categorical, the analyses primarily consisted of non-parametric analyses (e. g.,
Chi-square) and group-based analysis of mean differences (e.g., Student’s T-test,
Analysis of Variance).
I_ri_al 1: Preferred Doll

Table 4 shows frequencies for hold location for Trial 1, the preferred doll. Sixty-

nine percent of the children held the doll in a lateral position, with middle holding

3O

significantly less often observed (1(96) = 4.04, p < .001). This was equally true for boys

(70% side-hold; t(46) = 3.00, p_ < .01) and girls (68% side-hold; t(49) = 2.70, p < .01).

Table 4. Trial 1: Preferred Doll - Hold Location Frequencies for Boys and Girls

 

 

Left Middle Right
Boys 15 14 1 8
Girls 22 16 12
Total 37 3O 30

 

Note. X2(2) = 2.57, ns.

Of those holding to the side, 55% held on the left and 45% on the right (t(66) = 0.853,
n.s.). When boys and girls were analyzed separately, with middle holds excluded, girls
were 2.20 times more likely than boys to hold on the left (z = 1.57, n.s.). Indeed, of those
holding to the side, left-side holding was observed in the majority of girls (65%; 3(33) =
1.77, p = .086) but in the minority of boys (45%; t(32) = 0.516, n.s.). Neither the within-
or between-sex differences, however, were significant.

Finally, there were no age differences for the three hold location groups (E(2,94)
= .169, n.s.). Age means and standard deviations for left holders were 52.95 months
(SD. = 5.52), for middle holders 52.47 months (SD. = 5.309), and for right holders
52.23 months (SD. = 5.12). Nor was there an interaction between age and sex for hold
location (E(2,91) = 2.610, n.s.).

DE 2: Non-Preferred Doll
Table 5 shows frequencies for hold location for Trial 2, the non-preferred doll.

Sixty percent of the children held on the side (left and right combined), compared to 40%

31

in the middle, a significant difference (t(96) = 1.96, p = .05). The difference was
significant for girls (64% held to the side; t(49) = 2.04, p < .05) but not for boys (55%
held to the side; t(46) = 0.73, n.s.). Of the side-holders, 50% held on the left, 50% on the
right. The difference in percentages between middle and left holders was not significant

(t(67) = 1.22, n.s.).

Table 5. Trial 2: Non-Preferred Doll — Hold Location Frequencies for Boys and Girls

 

 

Left Middle Right
Boys 1 l 21 l 5
Girls 18 1 8 14
Total 29 39 29

 

Note. X2(2) = 1.864, n.s.

 

When boys and girls were analyzed separately, with middle holds excluded, girls
were 1.75 times more likely than boys to hold on the left (; = 1.05, n.s.). For girls, left
and middle holds also were equally likely (Left 36%, Middle 36%, Right 28%), whereas
for boys, middle-holding increased to become the predominant style (Left 23%, Middle
45%, Right 32%). As on Trial 1, there were no age differences between the different
hold location groups.

Table 6 shows a cross—tabulation of fi'equencies for Trials 1 and 2. The diagonal
cells indicate the number of children who were consistent in hold location across trials,
and the off-diagonal cells indicate the number who changed location. The table shows
that most children (71%), regardless of hold location, were consistent across trials.

Where change occurred, both left-holders and right-holders were more likely to change to

32

the middle location than to the opposite side. However, middle-holders who changed

were more likely to change to the right side than to the left side.

Table 6. Contingency Table for Hold Location on Trials 1 and 2

 

Trial 2, Non-Preferred Doll

 

 

Left Middle Right Total
Left 25 9 3 37
Trial 1, Middle 2 23 5 30
Preferred Doll Right 2 7 21 30
Total 29 39 29 97

 

Note. X2(4) = 66.53, p<.001; bolded values represent numbers of participants who were consistent in
their hold-side across the two trials.

£131 3: Preferred Doll, Holding and Feeding

On Trial 3, as on Trial 1, when the child could choose the doll that he or she
wanted to hold and feed, in most cases (75%) the preferred doll was the same for both
trials. Consistency, however, varied, depending on which doll was chosen first: of those
who chose the white doll first, 82% were consistent; for those who chose the black doll
first, only 53% were consistent (X2(1) = 7.310, p = .007). Girls also were more consistent
than boys (82% vs. 68%).

Table 7 shows frequencies of hold location for Trial 3, the preferred-doll holding
and feeding condition. Nearly every child (95%) held the doll on the side, with middle
holding significantly less often observed (t(96) = 19.87, p < .001). This was equally true
for boys (94% side-hold; t(46) = 12.10, p < .001) and girls (96% side-hold; 1(49) = 16.43,

p < .001). Of those who held on the side, 60% held the doll on the left and 40% on the

33

right (t(9l) = 2.58, p = .012). When boys and girls were analyzed separately, with middle
holds excluded, girls were 2.02 times more likely than boys to hold on the left (2 = 1.61,
n.s.). Of those holding on the side, the majority of girls held on the left (71%; 1(47) =
3.14, p = .003) as did the majority of boys (55%; 1(43) = 0.599, n.s.). The bias was
significant only for girls, but the sex difference itself was not significant (t(90) = 1.62, p

=.110).

Table 7. Trial 3: Preferred Doll While Feeding for Boys and Girls According

 

 

to Hold Location
Left Middle Right
Boys 24 3 20
Girls 34 2 14
Total 58 5 34

 

Note. X2(2) = 2.893, n.s.

Table 8 shows a cross-tabulation of frequencies for Trials 1 and 3. As in Table 6,
the diagonal cells indicate the number of children who were consistent across trials, and
the off-diagonal cells indicate the number of children who changed their hold location.
The table shows that most left-holders (73%) on Trial 1 did not change hold location on
Trial 3. That is, they continued to hold on the left for the holding while feeding task.
Most middle- and right-holders, on the other hand, were not consistent with their hold
location and were observed holding the doll on the left while manipulating the feeding
spoon with their right hand. All together, of the children who changed, in the majority of

cases, it was to a left-side hold.

34

Table 8. Contingency Table for Hold Location on Trials 1 and 3

 

Trial 3, Preferred Doll/Feeding

 

 

Left Middle Right Total
Left 24 3 10 37
Preferred Doll Right 17 0 13 30
Total 58 5 34 97

 

Note. X2(2)=3.85, n.s.; bolded values represent numbers of participants who were consistent in their
hold-side across the two trials.

In sum, the results on the three hold location trials were significantly different
from each other. Trial 1 provided perhaps the purest measure of holding—side bias as it
was the first holding trial and involved the preferred doll. For children not consistent in
location across the trials, there are three possible reasons for the changes: Trial order,
doll characteristics (whether the doll was preferred or non-preferred), and change (on
Trial 3) in the task itself (holding while feeding). Considering the differences between
trials, it was decided that further analyses using a scale score for hold location (e.g., mean
score of the three trials) would not be useful.

Hold Proximity

Along with hold location, hold proximity was assessed for each of the three
holding trials. Each hold was rated on a three-point scale for proximity: Holding with
the doll touching the body, holding with the doll close to but not touching the body, and
holding the doll far from the body. Similar to hold location, analyses for the three

categories of hold proximity primarily consisted of non-pararnetric analyses (e.g., Chi-

35

square) and group-based analyses of mean differences (e. g., Student’s T-test, Analysis of
Variance).

Table 9 shows the distribution of frequencies and Chi-square values for hold
proximities on the three trials. Both boys and girls, on all the three trials, showed a
significant tendency to hold the doll in close proximity, that is, with the doll touching the
body. For all three trials, this effect was stronger for girls (that is, proportionately more

girls than boys held close to body), but the difference was not significant.

Table 9. Frequencies of Hold Proximity for Boys and Girls for Trials 1 to 3

 

 

D01! Do" Not Doll Away
Touching Touching From Body X2
Body Body
Trial 1
Boys 27 (57%) 6 (13%) 14 (30%) 14.34
Girls 37 (74%) 6 (12%) 7 (14%) 37.24
Total 64 (66%) 12 (12%) 21 (22%) 47.77
Trial 2
Boys 26 (55%) 8 (17%) 13 (28%) 11.02
Girls 34 (68%) 7 (14%) 9 (18%) 27.16
Total 60 (62%) 15 (15%) 22 (23%) 36.27
Trial 3
Boys 26 (55%) 12 (16%) 9 (19%) 10.51
Girls 37 (74%) 11 (22%) 2 (4%) 39.64
Total 63 (65%) 23 (24%) ll (11%) 45.86

 

Note. Chi-square values are calculated for each row; all p-values < .01.

Table 10 shows a contingency table for hold proximity on Trials 1 and 2. The

purpose of this table is to demonstrate the consistency of hold proximity over the two

trials. The majority of children (84%) held in the same proximity on Trial 1 as on Trial 2
(88% for those with doll touching body, 75% for those with doll not touching body, and

76% for those with doll away from body).

Table 10. Contingency Table for Hold Proximity on Trials 1 and 2

 

Trial 2, Non-Preferred Doll

 

 

Doll Doll Not Holding
Touching Touching Away From Total
Body Body Body
Doll Touching 5 6 3 5 64
Body
Trial 1, 13°“ N.“ 2 9 1 12
Preferred Touching Body
Doll Holding Away
From Body 2 3 16 21
Total 60 15 22 97

 

Note. Bolded values represent numbers of participants who were consistent in hold proximity across
the two trials.

Similarly, but by a smaller margin, 69% of the children were consistent for hold
proximity between Trials 1 and 3. As shown in Table 11, 81% of those holding the doll
so that it touched their body and 75% of those who held the doll close but not touching
the body were consistent with their hold proximity. For those children who held the doll
away from the body on Trial 1, only 29% were consistent with their hold proximity. Of
this latter group, 29% switched to a doll close but not touching the body proximity and

43% switched to a doll touching body proximity for Trial 3.

37

Table 11. Contingency Table for Hold Proximity on Trials 1 and 3

 

Trial 3, Preferred Doll/Feeding

D011 D011 Not Holding
Touching Touching Away From Total

 

 

Body Body Body
Doll Touching Body 52 8 4 64
Trial 1, D01] Not Touching Body 2 9 1 12
Preferred
Doll Holding Away From Body 9 6 6 21
Total 63 23 1 1 97

 

Note. Bolded values represent numbers of participants who were consistent in hold proximity across
the two trials.

Tables 12 is a contingency table comparing the frequencies for hold location and

hold proximity for the three trials. The table shows that location and proximity are

Table 12. Hold Proximity as a Function of Hold Location for Trials 1 to 3

 

 

Doll Touching Doll Not Holding Away
Body Touching Body From Body

Trial 1

Left 30 (81%) 4 (11%) 3 (8%)

Middle 12 (40%) 2 (7%) 16 (53%)

Right 22 (73%) 6 (20%) 2 (7%)
Trial 2

Left 23 (79%) 5 (17%) l (3%)

Middle 15 (38%) 5 ( 13%) 19 (49%)

Right 22 (76%) 5 (17%) 2 (7%)
Trial 3

Left 43 (74%) 10 (17%) 5 (9%)

Middle 2 (40%) O (0%) 3 (60%)

Right 18 (53%) 13 (38%) 3 (9%)

 

Note. Percentages are based on total frequencies for each row.

38

dependent upon each other. Specifically, on Trials 1 and 2, left-side holds were more
likely to be close (doll touching body) while middle holds were more likely to be far (doll
away from body). On Trial 3, on the other hand, only left-side holds were more likely to
be close.
Hold Duration

'_I‘_ri_a1 1: Preferred Doll

Hold duration for the preferred doll (Trial 1) ranged from 0.01 to 60 secondss.
For all children, the average duration was 26.19 seconds. Girls held appreciably and

significantly longer than boys (32.71 vs. 19.25 seconds; 1(95) = 3.933, p < .001).

Table 13. Trial 1: Preferred Doll Hold Duration for Boys and Girls According

 

 

to Hold Location

Standard

Hold Location Sex N Mean Deviation
Left Boys 15 22.37 14.17
Girls 22 31.64 17.14
Total 37 27.89 16.45
Middle Boys 14 18.33 17.08
Girls 16 33.41 21.68
Total 30 26.37 20.79
Right Boys 18 17.35 11.72
Girls 12 33.74 20.33
Total 30 23.91 17.43
Total Boys 47 19.25 14.1 1
Girls 50 32.71 19.06
Total 97 26.18 18.07

 

Note. Significance tests: location (E(2,91) = .07, us), sex (E(1,91) = 14.69, p < .001), location x sex
interaction (E(2,91) = .40, ns)

 

5 The score 0.01 seconds meant that the child picked the doll up and put it down immediately, literally
before the observer could start the stopwatch. The score 60 seconds meant that the child held the doll for
the maximum amount of time set for the trial.

39

When duration was assessed across hold location, no differences were found for
the group as a whole (E(2,94) = .398, us.) or for either sex alone (both F’s < 1.0). Table
13 shows the durations according to sex and hold location.

When duration was assessed across the different hold proximities, as shown in
Table 14, a significant difference was found (_E(2,94) = 3.51, p = .034), with “doll
touching body” holds being significantly longer than “doll away from body” holds

(Fisher’s LSD, p < .01).

Table 14. Trial 1: Preferred Doll Hold Duration for Boys and Girls According

 

 

to Hold Proximity

Standard

Hold Proximity Sex N Mean Deviation
Doll Touching Body Boys 27 21 .50 12.71
Girls 37 34.23 18.98
Total 64 28.86 17.68
Doll Not Touching Body Boys 6 28.92 19.26
Girls 6 26.40 17.98
Total 12 27.66 17.81
Doll Away From Body Boys 14 10.75 10.37
Girls 7 30.10 21.78
Total 21 17.20 17.31
Total Boys 47 19.25 14.1 1
Girls 50 32.71 19.06
Total 97 26. 19 18.07

 

This effect, though, was primarily driven by the boys. Figure 1 shows a line graph of the
interaction between sex and proximity for length of hold. When analyzed separately,
girls’ duration scores did not differ across the hold proximities (_E_(2,47) = 0.502,

2:608). Boys’ duration scores did differ (F(2,44) = 5.050, p = .011). For boys, post hoc

40

comparisons using Fisher’s LSD revealed that hold duration for “doll touching body” and
“doll not touching body” were not significantly different, but both were significantly

different from “doll away from body”.

Figure 1. Trial 1: Interaction Between Sex and Proximity for Hold Duration

35.00 - . §423
30.00 -
25.00 -
20.00 -
15.00 1
10.00

 

 

Boys
- - - - Girls

 

 

 

 

10.75

Duration in Seconds '

 

 

Doll Touching Body Doll Not Touching Doll Away From Body
Body

T_rial 2: Non-Preferred Doll

Hold duration on Trial 2, for the non-preferred doll, was similar to hold duration
on Trial 1, for the preferred doll, and both were significantly correlated with each other (I
= .615, p < .001). Hold duration for the non-preferred doll ranged from 0.01 to 60
seconds, with the average duration dropping slightly to 22.39 seconds. As they had with
the preferred doll, girls held appreciably and significantly longer than boys (28.08 vs.
16.34 seconds; t(95) = 3.38, p = .001).

As shown in Tables 15 and 16, hold duration did not vary significantly for either
hold location (E(2,9l) = 1.026, n.s.) or hold proximity (F(2,91) = 1.773, n.s.) but did vary

significantly with sex of child, as noted above.

41

Table 15. Trial 2: Non-Preferred Doll Hold Duration for Boys and Girls According

 

 

to Hold Location
Hold Location Sex N Mean Standard Deviation
Left Boys 11 21.50 17.07
Girls 18 30.88 21.21
Total 29 27.32 19.97
Middle Boys 21 16.23 12.17
Girls 18 24.98 20.04
Total 39 20.27 16.64
Right Boys 15 12.71 13.39
Girls 14 28.47 17.79
Total 29 20.32 17.34
Total Boys 47 16.34 13.89
Girls 50 28.08 19.64
Total 97 22.39 18.00

 

Note. Significance tests: location (E(2,91) = 1.026, n.s), sex (E(l,91) = 9.998, p < .01), location x sex
interaction (_E(2,91) = .389, as).

Table 16. Trial 2: Non-Preferred Doll Hold Duration for Boys and Girls According

 

 

to Hold Proximity

Hold Proximity Sex N Mean Standard Deviation

Doll Touching Body Boys 26 19.16 15.47
Girls 34 29.60 21.27
Total 60 25.08 19.54

Doll Not Touching Body Boys 8 16.91 13.51
Girls 7 28.40 12.91
Total 15 22.27 14.07

Doll Away From Body Boys 13 10.35 08.82
Girls 9 22.11 17.95
Total 22 15.16 14.22

Total Boys 47 16.34 13.89
Girls 50 28.08 19.64
Total 97 22.39 18.00

 

Note. Significance tests: location (_E(2,91) = 1.773, n.s), sex (£(1,9l) = 7.387, p < .01), location x sex
interaction (E(2,91) = .014, us).

42

Table 17 shows the reduction at the group level in hold duration from Trial 1
(preferred doll) to Trial 2 (non-preferred doll) as a function of hold location and sex of
child. In all cases, there was a reduction in hold duration, but the magnitude of the
reduction varied according to hold location and roughly equally for boys and girls. The
greatest difference was in the middle location. On average, middle-holders on Trial 1
held 23% longer than their counterparts on Trial 2. Similarly, right-holders on Trial 1
held 15% longer than their counterparts on Trial 2. Left-holders on Trial 1, however,
held only 2% longer than their counterparts on Trial 2. In other words, the reduction was
trivial only for left-holders but was substantial for right- and middle—holders. It should be
noted that there are small variations in sample size for each of the hold locations on the
different trials.

Table 17. Mean Hold Duration Scores for Boys and Girls According to Hold Location
for Trials 1 and 2

 

 

Hold Preferred Doll Non-Preferred Difference %
Location Sex (Sec) Doll (Sec) (Sec) Reduction
Left Boys 22.37 21 .50 0.87 4%
Girls 31.64 30.88 0.76 2%
Total 27.89 27.32 0.57 2%
Middle Boys 18.33 16.23 2.10 11%
Girls 33.41 24.98 8.43 25%
Total 26.37 20.27 6.10 23%
Right Boys 17.35 12.71 4.68 27%
Girls 33.74 28.47 5.66 17%
Total 23.91 20.32 3.59 15%
Total Boys 19.25 16.34 2.91 15%
Girls 32.71 28.08 4.63 14%
Total 26.18 22.39 3.80 15%

 

Note: There are different numbers of individuals (also different individuals) between the cells. Table 6

reports the consistency of bold side (Left 68%, Middle 77%, Right 70%).

Table 18 shows the reduction at the group level in hold duration from Trial 1
(preferred doll) to Trial 2 (non-preferred doll) as a firnction of hold proximity and child’s
sex. Now, in all but one case, there was a reduction in hold duration, but no clear pattern
of reduction was obvious across the proximities. This is most likely the result of small

sample sizes in most of the cells.

Table 18. Mean Hold Duration Scores for Boys and Girls According to Hold Proximity

 

Preferred Non-Preferred Difference %

 

Hold Proximity Sex Doll (Sec) Doll (Sec.) (Sec) Reduction
Doll Touching.Body Boys 21 .50 19. 16 2.34 l 1%
Girls 34.23 29.60 4.63 14%
Total 28.86 25.08 3.78 13%
Doll Not Touching Boys 28.92 16.91 12.01 42%
Body Girls 26.40 28.40 -2.00 -8%
Total 27.66 22.27 5.39 19%
Doll Away From Boys 10.75 10.35 0.40 4%
Body Girls 30.10 22.1 1 7.99 27%
Total 17.20 15.16 2.04 12%
Total Boys 19.25 16.34 2.91 15%
Girls 32.71 28.08 4.63 14%
Total 26.19 22.39 3.80 15%

 

Note: There are different numbers of individuals (also different individuals) between the cells. Table 6
reports the consistency of bold side (Left 68%, Middle 77%, Right 70%).

Handedness and Footedness
Handedness
Table 19 shows the frequencies of hand use for the 14 tasks (9 different tasks)

comprising the handedness measure for the 97 children in the sample. For each task, the

44

right hand was used more often than either the left hand or both hands together. The

tasks also varied significantly in strength of bias. The draw-free and draw-shapes tasks,

for example, showed the strongest right bias, and the hang-monkey task showed the

weakest right bias, with a larger proportion of both hands use.

Table 19. Hand Use Frequencies Across Tasks

 

 

 

Hand Used
Task Left Both Right
Hang a small monkey onto the string of monkeys 31 16 50
Hang another small monkey onto the string of monkeys 27 19 51
Throw the ball to me 15 15 67
Throw the ball to me, again 12 16 69
Draw shapes 8 1 88
Point to a Lego 20 0 77
Place one Lego on top of another 18 6 73
Push a car quickly from here to there 15 1 81
Push a car slowly from here to there 17 1 78
Push a car quickly from here to there, again 16 0 81
Push a car slowly from here to there, again 16 1 80
Use a spoon to stir a bowl of imagined cereal 17 1 79
Use a spoon to feed a doll 27 l 69
Draw a picture 8 2 87

 

For each task, left-hand use was scored as -1.0, both-hands use as 0.0, and right-

hand use as +1.0. The construction of an overall handedness scale score can be

calculated by simply summing the scores for all of the tasks. As such, a child’s

handedness score could range from -14 to +14, where a negative value would indicate a

bias for left-hand use and a positive value a bias for right-hand use. Prior to analyses

using a scale score, task inter-correlations and scale statistics were evaluated.

45

Table 20 shows both the correlation matrix of the handedness tasks and the
statistics for the overall handedness scale. The overall reliability of the measure (alpha)
was 0.791 with only item #3, the hand used to feed the doll, having an item-scale
correlation (squared multiple correlation) lower than 0.5. Note that this item was paired
with the Trial 3 holding task so that it was confounded with preference for side of hold.
It, therefore, was removed from the construction of the overall handedness score,

resulting in an overall reliability (alpha) of 0.815.

46

moan. n AEQFQ banged 28m .802

 

 

 

$3 23 SS .83 82 $3 $3 83 33 $3 $3 Ed 88 33 325 80:325.
:3 $3 $3 $3 and 83 23 83s :3 was 32. Sec 82 :3 oauammfimwflm
82 8.3 88 $3 2.8 we; 23 a; :3 $2 33 :3 8g 53 Season caeasm
Sec 33 as 83 23 ago 33 $3 $3 was £3 £2 :3 83 as:

83 83 :2. 83 83 Ed 83 $3 83 :2 83 $3 83 End 050E389 :

83 36 ~36. ~85- $90. Rod- 83 mm; :3 $3- “8.? 89¢ Sec. :8 was 2

83 was 32 ~38 33 ~26 33 82 3.3 23 N25 a; 328 am 2

83 on; 83 23 Ed :5... 32 83 23 $3 5.3 3% so is :

83 :2 :3 83 :3 33 wt... $2. $3 53. Sam so :25 2

83 83 SS 83 £3 a; 9.3 E; :3- :85 30 =25 a

83 E3 55 23 $3 $2 :3 was- :5 55.5 m

83 83 R8 83 Rod e85 $3. 83 82a a

83 $2 83 :8 ”3c «86 83 SEE e

83 and fig 33 23 seam zen m

83 Ed 83 83 Sam 38:. a.

83 on; :3 .fimaefi m

83 :2. 13.82 mam N

83 1.382 as: a
z 2 2 : 2 a w a o n v m N _ saw as»:

 

web: $055283 co.“ 8553330 23m wee mecca—ebony .om £an

47

With item #6 omitted, the handedness scores ranged fiom ~11 to +13 with the
average being +7.64 (SD. = 5.45). Figure 2 shows a graphical display of the distribution
of scores. As expected, the distribution is skewed toward the positive end, indicating an

overall bias towards right-handedness.

Figure 2. Distribution of Total Handedness Scores

 

30

Number of
Children

 

 

Handedness Score

Note. Negative values indicate left hand preferences, positive values indicate right hand preferences.

When handedness scores were calculated separately for boys and girls; boys’ average
score was +7.12 (SD. = 5.73) and girls’ average score was 8.08 (SD. = 5.18). Using the
arbitrary cut-off value of 0.0 as the basis for classifying handedness, 7 boys (15%) and 3
girls (6%) were left-handed. This difference, however, was not significant (X2(1) =

1.145, n.s.).

48

Degree of handedness, independent of direction, was correlated with child’s age
in months but not significantly (1 = .166, p_ = .105). When the same correlation was
repeated for boys and girls separately, the correlation was stronger for girls (r = .237, p =
.098) than for boys (1' = .064, p = .670), but as neither correlation was statistically
significant, any interpretation of an interaction between sex and age for the development
of handedness should be preliminary.

Table 21 shows the mean handedness scores as a function of hold location, sex of
child, and trial. Overall handedness scores were not related to hold location on Trial 1
(E(2,94) = .176, n.s.) or Trial 2 (E(2,94) = 1.442, n.s.) but were related to hold location on

Trial 3 (£0.94) = 3.440, p = .036).

Table 21. Mean Handedness Scores for Boys and Girls According to Hold Location

 

 

and Trial Number
Hold Location Sex Trial 1 Trial 2 Trial 3
Left Boys 5.20 (7.48) 3.73 (8.21) 6.58 (4.76)
Girls 8.91 (4.70) 8.33 (5.29) 7.91 (4.85)
Total 7.41 (6.17) 6.59 (6.80) 7.36 (4.82)
Both Boys 7.50 (5.17) 7.76 (4.87) 0.00 (11.53)
Girls 7.36 (6.39) 7.06 (5.99) 6.00 (4.24)
Total 7.43 (5.75) 7.44 (5.35) ' 2.40 (9.04)
Right Boys 8.56 (4.12) 8.87 (3.62) 8.95 (5.13)
Girls 7.50 (4.40) 9.07 (3.89) 8.79 (6.24)
Total 8.13 (4.19) 8.97 (3.69) 8.88 (5.52)
Total Boys 7.17 (5.73)
Girls 8.08 (5.18)
Total 7.64 (5.45)

 

Post hoc analysis (Fisher’s LSD) revealed that left holders and right holders were

significantly different from middle holders on handedness scores but not from each other.

49

That is, although all hold locations were associated with right-handed biases, middle
holders had significantly lower handedness scores. Strength of handedness, irrespective
of direction, followed the same pattern of results, which was to be expected given the
small number of children with left-handed scores.

Strength of handedness, likewise, was not related to consistency of hold for Trials
1 and 2. Although consistent holders had slightly higher scores on the handedness
measure, this difference was not significant (1(95) = .182, n.s.).
Footedness

Table 22 shows the frequencies of foot use for the 7 tasks (5 different tasks)
comprising the footedness measure for the 97 children in the sample. For each task, the
right foot was used far more than either the left foot or both feet together. There also
were significant variations in foot preference across the different tasks. The kick ball
tasks, for example, showed the strongest right bias, and the balance on one foot task
showed the weakest right bias, with an almost equal number of right-foot and left-foot

preferences.

Table 22. Foot Use Frequencies Across Tasks

 

 

 

Foot Used
Task Left Both Right
Kick 3 ball to me 6 O 91
Kick a ball to me, again 8 l 88
Kick the Lego tower over with your foot 13 O 84
Stomp on the bag with one foot 21 6 70
Point to the red Lego with your foot 22 2 73
Stand on one foot for as long as you can 45 0 52
Hop on one foot for as long as you can 32 1 64

 

50

For each task, left-foot use was scored as -1.0, both-feet use as 0.0, and right-foot

use as +1.0. The construction of an overall footedness scale score can be calculated by

summing the scores for all of the tasks. As such, an individual’s footedness score could

range from -7 to +7, where a negative value would indicate a bias for left-foot use and a

positive value a bias for right-foot use. Prior to analyses using a scale score, task

intercorrelations and scale statistics were evaluated.

Table 23 shows both the correlation matrix of the footedness tasks and the

statistics for the overall footedness scale. The overall reliability of the measure (alpha)

was 0.421.

Table 23. Correlations and Scale Characteristics for F ootedness Items

 

 

 

Task 1 2 3 4 5 6 7

1 Kick ball to me 1.000

2 Kick ball to me, again 0.842 1.000

3 Kick Lego tower over 0.025 0.094 1.000

4 Stonrp on bag 0.157 0.125 0.131 1.000

5 Point at Lego with foot 0.263 0.175 0.283 0.212 1.000

6 Stand on one leg 0.104 0.116 0.059 0.043 0.041 1.000

7 Hop on one foot -0.092 -0.027 -0.023 -0.055 -0.220 0.349 1.000
Mean 0.876 0.825 0.732 0.505 0.526 0.072 0.330
Standard Deviation 0.484 0.559 0.685 0.831 0.483 1.003 0.943
Squared Multiple Correlation 0.735 0.721 0.117 0.063 0.215 0.151 0.185
Alpha if Item Deleted 0.344 0.337 0.391 0.398 0.392 0.350 0.494

 

Note. Alpha = .421

F ootedness scores ranged from -5 to +7 with the average being +3.87 (SD. =

2.60). Figure 3 shows a graphical display of the distribution of footedness scores. As

expected, the scores are skewed toward the positive end.

51

Figure 3. Distribution of Total F ootedness Scores

25‘

Number of 20

Childre n 1 5

10

 

Footedness Scone

Note. Negative values indicate left foot preferences while positive values indicate right foot preferences.

Footedness scores were calculated separately for boys and girls: Boys’ average score
was 4.28 (SD. = 2.58) and girls’ average score was 3.48 (SD. = 2.60). This difference,
however, was not significant (1(95) = 1.516, n.s.). Using the arbitrary cut-off value of 0.0
as the basis for classifying footedness, 2 boys (4%) and 7 girls (14%) were left-footed.
This difference, however, was not significant (Xz(l) = 2.733, p = .098).

Strength of footedness, independent of direction, was not correlated with child’s
age in months for either boys (r = -.024, n.s.) or girls (r = .031, n.s.). It was, however,
related to the consistency of hold location for Trials 1 and 2 in girls (t(48) = 3.43, 9 =
.001) but not boys (t(45) = 1.26, n.s.).

Table 24 shows the mean footedness scores as a function of hold location, sex of

child, and trial. Overall footedness scores were related to hold location for Trial 1

52

(_13(2,94) = 4.315, p = .016), Trial 2 (130,94) = 2.783, p = .067), and Trial 3 (32,94) =
3.664, p = .029). Post hoc analysis (Fisher’s LSD) for the first and third trials revealed
that middle holders had significantly stronger right-footedness scores than left holders but

not right-holders. Footedness scores did not differ between left- and right-holders.

Table 24. Mean Footedness Scores for Boys and Girls According to Hold
Location and Trial Number

 

 

Hold Location Sex Trial 1 Trial 2 Trial 3
Left Boy 3.80 (3.19) 2.64 (3.08) 5.29 (1.71)
Girl 2.68 (2.82) 3.28 (2.91) 3.53 (2.62)
Total 3.14 (2.91) 3.03 (2.93) 4.26 (2.43)
Both Boy 5.00 (2.18) 5.29 (2.00) -0.33 (4.16)
Girl 4.88 (1.46) 3.61(2.57) 3.50 (3.54)
Total 4.93 (1.80) 4.51(2.40) 1.20 (4.03)
Right Boy 4.11 (2.30) 4.07 (2.37) 3.75 (2.40)
Girl 3.08 (2.78) 3.57 (2.38) 3.36 (2.65)
Total 3.70 (2.51) 3.83 (2.35) 3.59 (2.48)
Total Boy 4.28 (2.58) 4.28 (2.58) 4.28 (2.58)
Girl 3.48 (2.60) 3.48 (2.60) 3.48 (2.60)
Total 3.87 (2.61) 3.87 (2.61) 3.87 (2.61)

 

53

DISCUSSION

The discussion begins with a review of the background and purpose of the current
study. It then provides a summary of the results as they pertain to the major predictions
and some general conclusions. Finally, it identifies some limitations of the study and
outlines directions for further research.

As described in the Introduction, most adults hold infants to their left. This bias
has been observed in a wide range of individuals who vary in parental status, sex,
handedness, nationality, and even species. Moreover, it is stable over time — at both the
population-level and individual-level - and it is selective in that it occurs for infants but
not for inanimate objects of similar size. In virtually all studies of the phenomenon,
women have shown the bias, whereas men have been more variable, sometimes showing
the bias as strongly as women, sometimes more weakly, sometimes not all.

Each of the several of explanations of the bias has found some measure of
support, including explanations based on the holder’s handedness, postural asymmetries,
heart-beat, emotional monitoring / direction of attention, and on the infant’s own resting
head position. So far, however, no combination of explanations, much less any one
explanation, has proven to be sufficient to explain the bias, due, at least in part, to
inadequacies in the methods for assessing the bias and in the handling of data.

In contrast to the many studies of the bias in adults, there have been only a few
investigations in children. All have found the left bias, and, although the age of its first
appearance remains unsettled, the evidence suggests that it appears between the ages of 3
to 6 years and that when it does appear, it occurs earlier in girls than in boys. For these

reasons, the current study tested 3- to 5-year-old boys and girls to assess the emergence

54

of the bias. For testing the contribution of handedness, this age range had the further
advantage of bracketing the period when reliable handedness was emerging.

The current study expanded on prior studies by using improved methods for
assessing holding-side bias. Location, proximity, and duration of hold were assessed
using realistic-looking dolls in a repeated-measures design. Handedness and
footedness/posture were assessed through performance-based, multi-task measures.
Analyses of group and individual differences were performed to test the following
predictions.

Summary of Results
Prediction 1

The first prediction was that children would show a left-side bias for holding the
dolls but that the bias would be weaker than that typically found in adults, or at least in
adult women. In light of de Chateau and Andersson’s (1976) finding that the majority of
their youngest age groups, the 2- and 4-year-olds, held the doll “in-hands”, that is, in the
middle, the first consideration in answering this question was whether, in the current
sample, most children displayed a lateralized hold (i.e., side hold) rather than a middle
hold. Across the three trials, the results indicated that side holds were, in fact, more
common. The next question, then, for side-holders, was whether the bias was to the left
(as in adults), to the right, or to each side equally. The results for Trials 1 and 2 by and
large did not indicate a left-side bias. On Trial 1, of children holding to one side, 55%
held to the lefi, but this was not a significant majority, and on Trial 2, left- and right-side
holds occurred equally. In Trial 3, however, when the child was asked to hold and “feed

the baby,” 60% of the side—holders, now a significant majority, held to the left. There

55

were no significant differences across the age range of the sample. In sum, the children
displayed wide variability in their hold location preferences. Although more children (on
Trial 1) were side-holders than in de Chateau and Andersson’s (1976) 4-year-old group,
fewer showed a left-side bias than in Saling and Bonert’s (1983) and Souza-Godeli’s
(1996) samples. In combination with de Chateau and Andersson’s (1976) findings, the
current results thus do not show a clear left-side bias in young children. They therefore
throw into question Souza-Godeli’s (1996, p. 1422) conclusion that left-side holding is “a
behavior emerging in the early stages of human ontogenesis,” and, instead, suggest that
the bias follows a different and more gradual developmental course - from
undifferentiated middle-location holding in early childhood to lateralized holding in
either direction around the age of 4 years, to group-level biases to the left in later
childhood.
Prediction 2

The second prediction was that degree, or strength, of handedness, independent of
direction, would predict lateralized holding. This prediction was based on the hypothesis
that the emergence of lateralized holding would follow the development of lateralized
motor functioning. For Trials 1 and 2, no evidence for this relationship was found.
Strength of handedness was nearly identical for middle- and lateralized-holding boys and
girls. The predicted relationship did appear, however, on Trial 3, where handedness
strength was positively related to the existence of a lateralized hold. That no relationship
was found for Trials 1 and 2 therefore suggests that lateralized motor development in
children is not related to lateralized infant holding. The significant relationship on Trial

3, while supporting the hypothesis, is confounded, however, by the pairing of the

56

handedness task (feeding) with the holding task. That is, although it is clear that
handedness influenced hold location on Trial 3 (93% of the middle-holding children
switched to a side hold, and 57% of right-holders switched to a left-hold), it is not clear
whether handedness influenced hold side directly or indirectly through the feeding
component of the trial.
Prediction 3

The third prediction extended the second in that it predicted that strength of
handedness, again as a measure of motor development, would be related to flexibility in
hold side from Trial 1 to Trial 3. That is, for those who held the doll with the dominant
hand and arm for Trial 1, whether they were right-handers who held on the right or left-
handers who held on the left, the prediction was that children with more complete motor
development would be more likely to switch hold locations to free their dominant hand
for the feeding component of the holding-while-feeding condition. The results failed to
support this prediction as there were no strength of handedness differences between
children who switched (left to right, right to left, in-hands to left, in-hands to right) and
those who did not switch. Furthermore, handedness was as strong in children who
switched to a left hold as it was in children who switched to a right hold.
Prediction 4

The fourth prediction was that direction of handedness would predict direction of
hold. There was no direct evidence to support this prediction for any of the three trials.
That is, there were no indications that right-handers were more likely to hold on the left,
left-handers on the right. Curiously, however, on all three trials, the subgroup of children

holding on the right did have stronger right-handedness scores than the subgroup of left-

57

holders, which is the reverse of the predicted direction, but these differences were
significant only for the holding-while-feeding trial. One possible interpretation of this
finding is that children at this age do not yet understand the conditional nature of infant
holding. Rather than viewing the act of holding an infant as serving the larger goal of
interacting with the infant or providing it with care and nurturance, young children may
view infant holding as a skilled act, similar to the handedness tasks, where the only
relevant component of the task is the holding behavior itself.
Prediction 5

The fifth prediction was that postural biases, as measured by footedness, would
predict hold location above and beyond handedness by itself. As handedness was not
predictive of hold side for the first two trials, footedness was assessed independently.
Like handedness, the results did not clearly support the prediction that posture would be
related to side of hold. For Trials 1 and 3, left side holding was associated with weaker
footedness scores than was middle-holding. Footedness scores, however, did not
differentiate right-side holding from either middle-holding or left-holding. These results
thus provide tentative corroboration of Erber et al.’s (2001) results that left-holders
showed an increased tendency to balance (i.e., place their weight) on the left foot than on
the right foot. While posture may still add predictive power to a model of holding bias
that is above and beyond handedness, the present sample was too small to detect the
differences.
Prediction 6

The sixth prediction, pertaining to sex differences, had three parts: that girls

would show stronger left-side biases, closer holds, and longer holds than boys. The first

58

part was not supported: there were no significant sex differences in the likelihood of
lateralized holding on any of the three trials. Likewise, there were no significant sex
differences for holding to either the left or right side. Girls, however, did show a clear
but non-significant left-bias on all three trials, whereas boys showed a non-significant
right bias on Trials 1 and 2 and a left bias only on Trial 3. The second part of the
prediction, pertaining to proximity, also was not supported. Although proportionately
more girls than boys held the doll close, the difference was not significant. Unfortunately,
if these differences exist, their effect size is too small for the current sample size to detect
significant differences. Finally, the third part of the prediction was supported
unequivally: girls held significantly longer than boys (nearly twice as long on average).

There are several possible interpretations of these significant and trend-level sex
differences. The first is that the boys were less engaged in the holding tasks, as suggested
by their tendency to hold the doll farther away fiom their body and for shorter periods of
time. Anecdotal reports from the experimenters, however, were that the boys seems
equally engaged in the tasks and did not behave in a way that suggests they found the
tasks disagreeable (e.g., verbal taunts, throwing the doll) as was found in de Chateau and
Andersson’s (1976) study. Nevertheless, the possibility exists that the sex differences are
due to the social stigma against boys playing with dolls. Future studies may be well
advised to provide control holding conditions using gender-neutral stimuli such as stuffed
animals, footballs, small animals, such as kittens or puppies, or, perhaps, even real
infants.

A second interpretation is that girls have more experience with dolls and/or more

training from others in the act of holding or cradling dolls and infants. Based on the adult

59

data (Harris, Almerigi, & Kirsch, 2000), it is unlikely that experience plays a direct role
in the bias. It may, however, play an important role in its acquisition — if not generally,
then as a moderator for its expression (e. g., increased trajectory, earlier age of
acquisition). As such, it would be worthwhile for future studies to use methods to
ascertain the child’s level of experience with infants and/or dolls (e. g., using such
methods as ratings of the home environment, listings of personal effects or toys or time
with infants, or parents’ estimation of time with dolls or infants).

Consistencv of Hold Side

 

The results for hold location showed substantial differences across the three trials.
In the case of the shift from a weak left-bias on Trial 1 to no bias on Trial 2, there may be
several reasons. One is that, unlike adults, young children’s holding-side biases are still
unstable, meaning that they would be expected to show different patterns of results across
any number of repetitions of the task. Another reason may be that the temporal order of
the two tasks was the important factor. That is, a subgroup of children may have decided
to use the other arm to hold the doll as a way to introduce variety or novelty to the now-
familiar task.

Still another possibility may be related to doll choice prior to holding. The
children were allowed to hold the “preferred” doll on Trial 1 but then were asked to hold
the “non-preferred” doll on Trial 2. The increased frequency of middle holding therefore
may reflect the child’s preference for a certain doll rather than any instability per se. That
is, the children may have adjusted their holding style in accord with the values associated
with the doll. If so, then, as suggested previously, it is the results on Trial 1 that provide

the closest measure of the holding-side bias.

6O

The two doll-holding conditions also differed in the average amount of time the
doll was held in arms. The preferred doll was held 3.8 seconds longer, on average, than
the non-preferred doll. The difference was largest for middle holders, who, overall,
showed a 23% reduction in hold duration from the preferred to non-preferred doll. Right-
holders showed a 15% reduction, and left-holders showed only a 2% reduction. When
these difference were assessed separately for boys and girls, girls showed the greatest
reduction (at least among those children who were consistent in hold location for the two
trials). This may mean that the girls were more discriminating about the thing being held,
just as Almerigi et al. (2001) found with college students imagining themselves holding
infants and inanimate objects.

Limitations and Future Directions

The current study provides a good start in tracking the development of the
holding-side bias and in elucidating the contributions of the task and of the child’s
handedness, posture, and sex, but the study also is limited in several respects.

The first limitation was that a left bias was not clearly found in the present sample
of children. It is unlikely that this is the result of the methods used to assess the bias as
dolls have been used successfully in prior child studies. Rather, it is likely that the age
range was not sufficiently broad to clearly identify the age at which the left-side holding
bias appears. At best, the results suggest that the bias is just beginning to emerge in 4-
year-old girls. Increasing the upper age range would have allowed for a better estimation
of the point at which the bias is established, thereby providing the “baseline” on which to

compare facets of holding for the earlier ages.

61

The second limitation was that significant differences were observed across the
holding conditions. Although some variability was desired for analyses, the magnitude of
differences between the first two trials, for example, prevented richer analyses involving
scale scores, thereby reducing measurement error, or analyses of the factorial structure of
“infant holding,” in general. Further, administering only three trials of the task over a
short period may have capitalized on state characteristics of the child or situational
characteristics of the context. More repetitions over greater intervals of time would
provide a more exact measurement of the intra-individual continuity of the bias.

A third limitation was that only handedness and footedness were assessed to
account for the bias. Based on what is known from studies of adults, the model of the
current study is empirically and theoretically under-identified. That is, for a proper
assessment of the origins of holding bias, many additional variables, such as cerebral
asymmetries and emotional state, would need to be measured. A more proper analysis of
the contribution of handedness would then be conducted in the statistical “presence” of
these other variables.

A fourth limitation is sample size. Several analyses lacked the power to yield
significant results. Specifically, power to detect subgroup differences in the current study
was minimal. While power is a function of sample size, it is also a function of effect
size. For example, this study may indicate, like adult studies, that handedness is not a
major contributor to holding side biases. While the contribution of handedness may exist
and be measureable, its effect is too small for the current sample size to reliably estimate.
Furthermore, a larger sample would be needed for more complex analyses that would

explore such topics as subgroup differences, especially left-handers, and of the

62

relationship between children's handedness and their consistency of hold across the

tasks. One potential analysis would be symmetry analysis of contingency tables or turn-
over tables (e. g., Schuster & von Eye, 1998). In consideration of the categorical nature of
"side-of-hold," other possible analyses would be latent class analysis (e. g., von Eye &
Clogg, 1994) and/or item-response theory (Hambleton, Swaminathan, & Rogers, 2001).

Finally, the most critical limitation of the current study is that, being a cross-
sectional study, it did not assess the intra-individual development of the bias. Inasmuch
as the goal of the current study was to uncover the developmental origins of the left-side
holding bias, a prospective longitudinal study would have been required to observe intra-
individual change in holding pattems and to identify causal predictors of the left-side
holding bias. The current study is limited to non-directional associations between the
variables independent and dependent variables.

In summary, the results of the analyses of the relation between strength and
direction of handedness and style and side-of-holding indicate at best a modest role for
handedness in the development of the holding-side bias. In retrospect, these results are
compatible with the shape of the distribution of the children’s handedness scores, which
look similar to the distributions found in studies of adults. Inasmuch as the left-side
holding bias was not yet established in these children, it seems unlikely that handedness
and the holding-side bias co-develop, contrary to one of the main hypotheses of the study.
Instead, the results suggest that the major roles in the development of the holding-side

bias are more likely to be played by other variables.

63

APPENDIX A

64

Date Prepared: October 4, 2002

Project Title: Evaluation of Ready Set Grow Passport: Passport
Outcomes Study

Date of Approval: October 31, 2002

This informed Consent Form is valid for a period not to exceed one year from the date of approval
appearing above.

Purpose:

You are being asked to participate in a research study to help us find out how
helpful the Passport Program is to its members. The researchers are interested in
determining how being a part of Passport affects your behavior as a parent and
how it affects your child’s development.

Procedure and Duration:

You will be participating in a home visit that lasts an hour and a half. During the
home visit, one project staff person will ask you questions from surveys about
your child’s behavior, and your own beliefs about parenting and child
development. During this time, a second project staff person will work with your
child to get information about your child’s social and cognitive development.

Risk and Discomfort:
There are no significant risks involved in being a participant in this study, aside
from the amount of time the home visit will take away from your busy schedule.

Benefits:

Participation in this study will assist the Passport Program in determining how
effective their services have been and what types of improvements need to be
made.

Refusal or Withdrawal of Participation:

Taking part in this study is completely your choice. You can withdraw from the
study at any time for any reason. It will not affect your membership in Passport
or the services you receive from Passport.

Privacy and Confidentiality:

Your name will not be used in any reports on this study. The records will be kept
private to the maximum extent allowable by law. The information gathered will
be used to make recommendations to the Passport Program as to how they can
better serve their members and to inform others as to how programs such as
Passport impact the lives of children and families.

We will strictly avoid any disclosure that may identify you to any persons not

related to this research program. The consent form, which is the only identifier
linking participants to data, will be kept in locked files at the Passport office.

65

Therefore, not even the researchers will be able to connect your data with your
name.

Additional Information:

If you would like to discuss any questions related to this research, please feel flee
to contact Jessica Barnes or Dr. Hiram E. Fitzgerald. If you have any questions
regarding your rights as research participants, please contact Ashir Kumar, M.D.,
Chair of Michigan State University’s Committee on Research Involving Human
Subjects, at 517-355-2180.

Jessica Barnes Dr. Hiram E. Fitzgerald

6 Kellogg Center 22 Kellogg Center

East Lansing, MI 48824 East Lansing, MI 48824
Campus Tel.: (517) 353-3508 Campus Tel.: (517) 353-8977

I have read the information above and have been given explanations about my
participation and my child’s participation in the study. The project staff has
satisfactorily answered all questions concerning the study. I agree (hereby
consent) to participate in the home visit.

 

 

Participant Signature Date

 

 

Project Staff Date

66

APPENDIX B

67

Date Prepared: October 4, 2002

Project Title: An empirical evaluation of the theoretical model of affective social
competence

Date of Approval: October 31, 2002

This informed Consent Form is valid for a period not to exceed one year from the date of approval
appearing above. '

Purpose:

The purpose of this research is to understand the effectiveness of the Passport
program and to understand how children’s development of the understanding and
regulation of emotions is related to their behavior. We are asking that we be able
to use the data that is collected for the evaluation of Passport in order to inform
the scientific community about these processes. No additional data will be
collected for this purpose. Therefore, the risks involved in participating in this
study are limited to those involved in the participation of the Passport Evaluation
study. By participating in this study of affective social development and
parenting factors, you are helping us to understand more about children’s
development.

Refusal or Withdrawal of Participation:

Taking part in this study is completely your choice. You can withdraw from the
study at any time for any reason. It will not affect your membership in Passport
or the services you receive fi'om Passport.

Privacy and Confidentiality:

Your name will not be used in any reports on this study. The records will be kept
private to the maximum extent allowable by law. The information gathered will
be used to report on the effectiveness of Passport and children’s social
competence.

We will strictly avoid any disclosure that may identify you to any persons not
related to this research program. The consent form, which is the only identifier
linking participants to data, will be kept in locked files at the Passport office.
Therefore, not even the researchers will be able to connect your data with your
name.

68

Additional Information:

If you would like to discuss any questions related to this research, please feel free
to contact Jessica Barnes or Dr. Hiram E. Fitzgerald. If you have any questions
regarding your rights as research participants, please contact Ashir Kumar, M.D.,
Chair of Michigan State University’s Committee on Research Involving Human
Subjects, at 517-355-2180.

Jessica Barnes Dr. Hiram E. Fitzgerald

6 Kellogg Center 22 Kellogg Center

East Lansing, MI 48824 East Lansing, MI 48824
Campus Tel.: (517) 353-3508 Campus Tel.: (517) 353-8977

I have read the information above and have been given explanations about my
participation and my child’s participation in the study. The project staff has
satisfactorily answered all questions concerning the study. I agree (hereby
consent) to have the information collected for the Passport Evaluation used for
this study.

 

 

Participant Signature Date

 

 

Project Staff Date

69

APPENDIX C

70

10.

113.

11b.

12.

List of Parent Questionnaires

Consent Form for Study of Child Development

Consent Form for Evaluation of Community Intervention
Demographics

Passport Usage Survey

Knowledge of Child Development Inventory

Parenting Questionnaire

Child Behavior Checklist for Ages 1 ‘/2 - 5

List of Child Tasks (in order)

Three-Bag—Task: Gift 1 (neutral gift)

Handedness 1: Add a small plastic monkey to a string of monkeys
Handedness 2: Add another monkey to the string of monkeys
Behavioral Regulation: Keep the M&M on your tongue
Footedness 1: Kick the ball to me

Footedness 2: Kick the ball to me, again

Handedness 3: Throw the ball to me

Handedness 4: Throw the ball to me, again

Emotion Display Task

Emotion Perception 1: DANV A2 - first 8 items

Lollipop Test (Cognitive Tests)

Handedness 5: Hand used to draw shaped recorded in Lollipop Test

Handedness 6: Point to the red Lego

71

13.

14.

15.

16a.

16b.

17a.

17b.

18a.

18b.

19a.

19b.

20.

21a.

21b.

22.

23a.

23b.

24.

25.

Handedness 7: Put the red Lego on the black Lego

Footedness 3: Point to the red Lego with your foot

Footedness 4: Kick the Lego tower over with your foot

Handedness 8: Roll the car as quickly as you can from here to there (which hand)
Motor Dev. 1: Roll the car as quickly as you can from here to there (duration)
Handedness 9: Roll the car as slowly as you can from here to there (which hand)
Motor Dev. 2: Roll the car as slowly as you can from here to there (duration)
Handedness 10: Again, roll the car as quickly as you can from here to there
(which hand)

Motor Dev. 3: Again, roll the car as quickly as you can from here to there
(duration)

Handedness 11: Again, roll the car as slowly as you can from here to there
(which hand)

Motor Dev. 4: Again, roll the car as slowly as you can from here to there
(duration)

Wally Test 1 (Social Competence)

F ootedness 5: Stand on one foot for as long as you can (which foot)

Motor Dev. 5: Stand on one foot for as long as you can (duration timed)
Motor Dev. 6: Stand on your other foot for as long as you can (duration timed)
Footedness 6: Hop on one foot for as long as you can (which foot)

Motor Dev. 7: Hop on one foot for as long as you can (duration)

Motor Dev. 8: Hop on the other foot for as long as you can (duration)

Wally Test 2 (Social Competence)

72

26.

27.

28.

29.

30.

31a.

31b.

32.

33.

34.

35.

36.

37.

38.

Doll Choice 1: Pretend these dolls are real babies. Point to the baby you want to
pick up

Doll 1: Pretending the doll is a real baby, pick up and hold the baby (Side,
Closeness, & Duration)

Doll 2: Now, pretend this doll is a real baby. Pick up and hold the baby (Side,
Closeness, & Duration)

Handedness 12: Pretend the bowl has cereal for the babies in it. Stir the cereal
with the spoon.

Doll Choice 2: Point to the baby you want to pick up

Doll 3: Pick up, hold, and feed the baby (Side & Closeness)

Handedness 13: Hand used to feed doll.

Emotion Perception 2 : DANV A2 — second 8 items

Motor Dev. 9: Stop/Go Task

Motor Dev. 10: Whisper Task

Handedness 14: Draw a picture

Emotion Perception 3: DANV A2 — third 8 items

Three-Bag-Task: Gift 2 (bad gift)

Three-Bag-Task: Gift 3 (good gift)

73

APPENDIX D

74

Example Doll Holding Coding Form

 

 

 

TASK CODING
Pretend these dolls are real babies. Point to
the baby you want to pick up.
White Doll Black Doll
Pretending the doll is a real baby, pick up and
hold the baby (up to one minute). L R B A

 

CTB C~TB HOAB

 

 

L B .
sec m.sec
Now, pretend this doll is a real baby. Pick up
and hold the baby (up to one minute). L R B A

 

CTB C~TB HOAB

 

 

 

L B -
sec m.sec

Pretend the bowl has cereal for the babies in
it. Stir the cereal with the spoon.

9 .- - 9 L R B A

L R
Point to the baby you want to pick up.
White Doll Black Doll

 

 

Pick up, hold, and feed the baby.

 

 

FEED: L R B A

CTB C~TB HOAB

 

75

 

APPENDIX E

76

Example Handedness and Footedness Coding Form

 

 

 

 

 

 

 

 

TASK CODING
Add a monkey to the string of monkeys.
9 2 . 9
R B
L
Add another monkey to the string of monkeys.
R B
L E l j l R
Kick the ball to me.
i. .53 R B
L R
Kick the ball again to me.
)8. O.§ R B
L R
Throw the ball to me.
9 0 3
R B
L R
Throw the ball again to me.
90 o 9
R B
L R

 

 

77

 

REFERENCES

Achenbach, T. M., & Rescorla, L. A. (2001). Manual for ASEBA School-Age
Forms & Profiles. Burlington, VT: University of Vermont, Research Center for
Children, Youth, & Families.

Alley, T. R., & Kolker, J. I. (1989). Psychological gender, hand preference and
sex differences in book-carrying styles. Perceptual and Motor Skills, 66, 815-822.

Almerigi, J. B., Carbary, T. J., & Harris, L. J. (2002). Most adults show opposite-
side biases in the imagined holding of infants and objects. Brain & Cognition, 48, 25 8-
263.

Almerigi, J. B., & Harris, L. J. (2002). Preliminary evidence that infants aren’t
special: A leftward holding bias observed for imagined kittens. Poster presentation at
the 13th annual meeting of TENNET, Montreal, Canada, June 14-18.

Almerigi, J. B., Van Hooren, D., Bulrnan-Fleming, B., & Harris, L. J. (2002).
The leftward cradling bias and hemispheric lateralization of emotion. Poster presentation
at the 13th annual meeting of TENNET, Montreal, Canada, June 14-18.

Barnes, J. V. (2003). The evaluation of a community developed support system
for families with young children: The “Ready, Set, Grow!” Passport program.
Unpublished Dissertation.

Barnes, J. V., Determan, C. M., & Fitzgerald, H. E. (2002). Ready, Set Grow!
Passport: Report on Program Development and Evaluation. Poster presentation at the
Sixth National Head Start Research Conference, Washington, DC, June 2002.

Barry, H., 111., & Paxson, L. M. (1971). Infancy and early childhood: Cross-
cultural codes 2. Ethnology, 10, 466-508.

Best, C. T., Wonner, J. S., & Queen, H. F. (1994). Hemispheric asymmetries in
adults’ perception of infant emotional expressions. Journal of Experimental Psychology:
Human Perception and Performance, 20, 751-765.

Bogren, L. Y. (1984). Side preference in women and men when holding their
newborn child: Psychological background. Acta Psychiatrica Scandinavica, 69, 13-23.

Borod, J. C., Andelrnan, F., Obler, L. K., Tweedy, J. R., & Welkowitz, J. (1992).

Right hemisphere specialization for the perception of emotional words and sentences:
Evidence from stroke patients. Neuropsychologia, 30, 827-844.

Bowlby, J. (1958). The nature of the child's tie to his mother. International
Journal of Psychoanalysis, 36, 639-645.

78

Bruml, H. (1972). Age changes in preference and skill measures of handedness.
Perceptual and Motor Skills, 34, 3-14.

Briisser, E. (1981). Child Transport in Sri-Lanka. Current Anthropology, 22, 288-
290.

Buchanan, A. (1862). Mechanical theory of the predominance of the right hand
over the left, or, more generally, of the limbs of the right side over those of the left side of
the body. Proceedings of the Philosophical Society of Glasgow, 5, 142-167.

Bundy, R. S. (1979). Effects of infant head position on sides preference in adult
handling. Infant Behavior & Development, 2, 355-358.

Cohn, J. F., & Tronick, E. Z. (1988). Mother infant face-to-face interaction-
influence is bidirectional and unrelated to periodic cycles in either partners behavior.
Developmental Psychology, 24, 386-392.

Comwell, K. S., Barnes, C. L., Fitzgerald, H. E., & Harris, L. J. (1985).
Neurobehavioral reorganization in early infancy: Patterns of head orientation following
lateral and midline holds. Infant Mental Health Journal, 6, 126-136.

Cosmedes, L., & Tooby, J. (1994). Origins of domain specificity: The evolution
of functional organization. In Hirschfeld and Gelsman (Eds), Mapping the Mind:
Domain Specificity in Cognition and Culture (pp. 85-116). Cambridge: Cambridge
University Press.

Dagenbach, D., Harris, L. J., & Fitzgerald, H. E. (1988). A longitudinal study of
lateral biases in parents' cradling and holding of infants. Infant Mental Health Journal, 9,
218-234.

De Casper, A. J ., & Fifer, W. P. (1980). On hmnan bonding: Newborns prefer
their mothers’ voices. Science, 208, 1174-1176.

de Chateau, P. (1991). Clinical application of asymmetrical behavioral
communication in parent-infant interaction. In H. E. Fitzgerald, B. M. Lester & et al.
(Eds), Theory and research in behavioral pediatrics (V 01. 5, pp. 185-208). New York,
NY: Plenum Press.

de Chateau, P., & Andersson, Y. (1976). Left-side preference for holding and
carrying newborn infants 11: Doll-holding and carrying from 2 to 16 years.
Developmental Medicine & Child Neurology, 18, 738-744.

De Renzi, E., Faglioni, P., & Spinnler, H. (1968). The performance of patients
with unilateral brain damage on face recognition tasks. Cortex, 4, 17-34.

79

Erber, N. L., Almerigi, J. B., Carbary, T. J., & Harris, L. J. (2002). The
contribution of postural bias to lateral preferences for holding human infants. Brain &
Cognition, 48, 352-356.

Fagan, J. F. (1972). Infants’ recognition memory for faces. Journal of
Experimental Child Psychology, 14, 453-476.

F eldman, R. (2003). Infant-mother and infant-father synchrony: The
coregulation of positive arousal. Infant Mental Health Journal, 24, 1-23.

Fennel], E., Satz, P., & Morris, P. (1983). The development of handedness and
dichotic ear listening asymmetries in relation to school achievement: A longitudinal
study. Journal of Experimental Psychology, 35, 248-262.

Field, T. (2001). Touch. Cambridge: MIT Press.

Fitzgerald, H. E., Harris, L. J ., Barnes, C. L., Wang, X., Comwell, K. S.,
Karnptner, N. L., et al. (1991). The organization of lateralized behavior during infancy. In
H. E. Fitzgerald & B. M. Lester (Eds.), Theory and research in behavioral pediatrics,
Vol. 5 (pp. 155-184). New York, NY: Plenum Press.

Fodor, J. A. (1983). The Modularity of Mind. Cambridge: MIT Press.

Fogel, A. (2001). Infancy: Infant, Family, and Society, (4th Ed.). Belmont,
Califomia: Wadsworth/Thomson Learning.

Fogel, A., Messinger, D. S., Dickson, K. L., & Hsu, H-C. (1999). Posture and
gaze in early mother-infant communication: Synchronization of developmental
trajectories. Developmental Science, 2, 325-332.

Ginsburg, H. J ., Fling, S., Hope, M. L, Musgrove, D., & Andrews, C. (1979).
Maternal holding preferences: A consequence of newborn head-turning response. Child
Development, 50, 280-281.

Grfisser, O. J. (1983). Mother-Child Holding Patterns in Western Art - a
Developmental- Study. Ethology and Sociobiology, 4, 89-94.

Gunnar, M. R., & Donahue, M. (1980). Sex-differences in social responsiveness
between 6 months and 12 months. Child Development, 51, 262-265.

Gunnar, M. R., & Stone, C. (1984). The effects of positive maternal affect on
infant responses to pleasant, ambiguous, and fear-provoking toys. Child Development,
55, 1231-1236.

Hall, J. A. (1984). Nonverbal Sex Diflerences: Communication Accuracy and
Expressive Style. Baltimore: Johns Hopkins University Press.

80

Hambleton, R. K., Swaminathan, H., & Rogers, H. J. (2001). Fundamentals of
Item Response Theory. Newbury Park, CA: Sage.

Harris, L. J. (2002). Lateral biases for holding infants: Early opinions,
observations, and explanations with some possible lessons for theory and research today.
Brain & Cognition, 48, 392-394.

Harris, L. J. (2003). What to do about your child's handedness? Advice from five
eighteenth-century authors, and some questions for today. Laterality, 8, 99-120.

Harris, L. J ., & Almerigi, J. B. (2001). Lateralization. In A. S. Honig, H. E.
Fitzgerald, & H. E. Brophy-Herb (Eds.), Infancy in America (Vol. 2) (pp. 408-421).
Santa Barbara: ABC Clio.

Harris, L. J ., Almerigi, J. B., Carbary, T. J ., & Fogel, T. G. (2001). Left-side
infant holding: A test of the hemispheric arousal-attentional hypothesis. Brain &
Cognition, 46, 159-165.

Harris, L. J ., Kang, Y., & Ahnerigi, J. B. (2003). Imagined infant-holding in a
Korean sample: Do restrictions against left-hand use affect the left-side holding bias?
Poster presentation at the 14th annual meeting of TENNET, Montreal, Canada, June 19-
21.

Harris, L. J ., Almerigi, J. B., & Kirsch, E. A. (2000). Side preference in adults for
holding infants: Contributions of sex and handedness in a test of imagination. Brain &
Cognition, 43, 246-252.

Harris, L. J ., & Fitzgerald, H. E. (1983). Postural orientation in human infants:
changes from birth to three months. In G. Young, S. J. Segalowitz, C. M. Corter, & S. E.
Trehub (Eds) Manual Specialization and the Developing Brain (pp. 285-306). New
York, NY: Academic Press.

Harris, L. J ., & Fitzgerald, H. E. (1985). Lateral cradling preferences in men and

women: Results fiom a photographic study. Journal of General Psychology, 112, 185-
189.

Harris, L. J ., Jentoft, C., & Almerigi, J. B. (2002). Do biases against left-hand use
affect side-preferences for holding infants? Evidence from photographs in Bateson and
Mead’s (1942) and Mead and Macgregor’s (1951) studies of the people of Bajoeng Gedé,
Bali. Poster presentation at the annual meeting of the Midwestern Psychological
Association, Chicago, May 2-4.

Hatta, T., & Koike, M. (1991). Left-hand preference in fiightened mother
monkeys in taking up their babies. Neuropsychologia, 29, 207-209.

81

Haviland, J. M., & Malatesta, C. (1981). A description of the development of sex
differences in nonverbal signals: Fantasies, fallacies, and facts. In C. Mayo & N. Henley
(Eds.), Gender and Nonverbal Behaviors, New York: Springer-Verlag.

Kanwisher, N., McDermott, J ., & Chun, M. M. (1997). The fusiforrn face area:
A module in human extrastriate cortex specialized for face perception. The Journal of
Neuroscience, 1 7, 4302-431 1.

Lever, J. (197 8). Sex differences in the complexity of children’s play and games.
American Sociological Review, 43, 471-483.

Little, R. J. A. & Rubin, D. B. (1987). Statistical Analysis with Missing Data.
New York: Wiley.

Lockard, J. S., Daley, P. C., & Gunderson, V. M. (1979). Maternal and Paternal
Differences in Infant Carry - United States and African Data. American Naturalist, 113,
235-246.

Lonstein, J. S., & Stern, J. M. (1999). Unilateral suckling maintains bilaterally-
symmetrical nursing postures and c-fos activity in the midbrain periaqueductal gray of
lactating rats. Developmental Psychobiology, 35 , 264-275.

Lorenz, K. (1964). Comparative studies of behavior. In C. H. Schiller (trans. and
ed.), Instinctive Behavior: The Development of a Modern Concept (pp. 239-263). New
York, NY: International Universities Press.

Maccoby, E. E. (1990). Gender and relationships. American Psychologist, 45,
513-520.

Maccoby, E. E., & J acklin, C. (1974). The Psychology of Sex Differences.
Stanford: Stanford University Press.

Manning, J. T. (1991). Sex differences in left-side infant holding: Results from
"family album" photographs. Ethology & Sociobiology, 12, 337-343.

Manning, J. T., & Chamberlain, A. T. (1990). The left-side cradling preference in
great apes. Animal Behaviour, 39, 1224-1227.

Manning, J. T., & Chamberlain, A. T. (1991). Left-side cradling and brain
lateralization. Ethology and Sociobiology, 12, 237-244.

McManus, I. C., Sik, G., Cole, D. R., Mellon, A. F ., Wong, J ., & Kloss, J.

(1988). The development of handedness in children. British Journal of Developmental
Psychology, 6, 257-273.

82

McGuinness, D., and Pribram, K. (1980). The neuropsychology of attention:
Emotional and motivational controls. In Wittrock, M. C. (ed.), The Brain and
Psychology, Academic Press, New York, pp. 95-139.

Mead, M. (1942). Balinese Character. In G. Bateson and M. Mead (1942).
Balinese Character: A Photographic Analysis (pp. 1-48). New York: Special
Publications of the New York Academy of Sciences, Vol. II.

Meyer—Bahlburg, H. F ., Sandberg, D. E., Dolezal, C. L., & Yager, T. J. (1994).
Gender-related assessment of childhood play. Journal of Abnormal Child Psychology,
22, 643-660.

Nakarnichi, M., & Takeda, S. (1995). A child-holding thought experiment:
Students prefer to imagine holding an infant on the left side of the body. Perceptual &
Motor Skills, 80, 687-690.

Natale, M., Gur, R. E., & Gur, R. C. (1983). Hemispheric asymmetries in
processing emotional expressions. Neuropsychologia, 21, 555-565.

Nicolson, N. (1987). Infants, mothers, and other females. In B. B. Smuts, D. L.
Cheney, R. M. Seyfarth, R. W. Wrangham, & T. T. Struthsaker (Eds.), Primate Societies
(pp. 330-342). Chicago, IL: University of Chicago Press.

Nightingale, F. (1860). Notes on nursing: What it is, and what it is not. Journal
of Mental Science, 6, 481-495.

Panksepp, J. (1998). Aflective Neuroscience: The Foundations of Human and
Animal Emotions. New York, NY: Oxford University Press.

Papousek, H., & Papousek, M. (1986). Structure and dynamics of human
communication at the beginning of life. European Archives of Psychiatry and
Neurological Sciences, 236, 21-25.

Papousek, H., & Papousek, M. (1989). Forms and functions of vocal matching in
interactions between mothers and their precanonical infants. First Language, 9, 137-158.

Peters, M. & Servos, P. (1989). Performance of subgroups of left-handers and
right-handers Canadian Journal of Psychology, 43, 341-358.

Pitcher, E. G., & Schultz, L. H. (1983). Boys and Girls at Play: The Development
of Sex Roles. New York: Praeger.

Pryce, C. R. (1993). The regulation of maternal behaviour in marmosets and
tamarins. Behavior Processes, 30, 201-224.

83

Renter-Lorenz, P. A., Kinsboume, M., & Moscovitch, M. (1990). Hemispheric
control of spatial attention. Brain and Cognition, 12, 240-266.

Rheingold, H. L., & Keene, G. C. (1965). Transport of the Human Young. In B.
M. Foss (Ed.), Determinants of Infant's Behaviour (V 01. 3, pp. 87-110). London:
Methuen.

Robson, K. S. (1967). The role of eye-to-eye contact in matemal-infant
attachment. Journal of Child Psychology and Psychiatry and Allied Disciplines, 8, 13-25.

Saling, M., & Bonert, R. (1983). Lateral cradling preferences in female
preschoolers. Journal of Genetic Psychology, 142, 149-150.

Saling, M. M., & Cooke, W.-L. (1984). Cradling and transport of infants by South
African mothers: A cross-cultural study. Current Anthropology, 25, 333-335.

Saling, M., & Tyson, G. (1981). Lateral cradling preferences in nulliparous
females. Journal of Genetic Psychology, 139, 309-310.

Salk, L. (1960). The Effects of the Normal Heartbeat Sound on the Behavior of
the New-bom Infant: Implications for Mental Health. World Mental Health, 12, 168-175.

Salk, L. (1961). The Importance of heart-beat rhythm to human nature:
Theoretical, Clinical, and experimental observations. In Proceedings of the 3rd World
Congress of Psychiatry (pp. 740-760). Toronto: University of Toronto Press.

Salk, L. (1973). The role of the heartbeat in the relations between mother and
infant. Scientific American, 228, 24-29.

Scheman, J. D., Lockard, J. S., & Mehler, B. L. (1978). Influences of anatomical
differences on gender-specific book-carrying behavior. Bulletin of the Psychonomic
Society, 11, 17-20.

Schuster, C., & von Eye, A. (1998). Modeling turn-over tables using the GSK
approach. Methods of Psychological Research, 3, 39-53.

Sergent, J ., Ohta, S., & MacDonald, B. (1992). Functional neuroanatomy of face
and object processing. Brain, 115, 15-36.

Sieratzki, J. S., Roy, R, & Woll, B. (2002). Left cradling and the left ear
advantage for emotional speech: Listen to the other side too. Laterality, 7, 343-345.

Souza-Godeli, M. R. C. (1996). Lateral cradling preferences in children.
Perceptual and Motor Skills, 83(3, Pt 2), 1421-1422.

84

Sperber, Z., & Weiland, I. H. (1973). Anxiety as a determinant of parent-infant
contact patterns. Psychosomatic Medicine, 35, 472-483.

Stables, D., & Hewitt, G. (1995). The Effect of Lateral Asymmetries on Breast-
Feeding Skills - Can Midwives Holding Interventions Overcome Unilateral Breast-
Feeding Problems. Midwifery, 11, 28-36.

Trevathan, W. R. (1982). Maternal lateral preference at first contact with her
newborn infant. Birth: Issues in Perinatal Care & Education, 9(2), 85-89.

Turkewitz, G., & Birch, H.G. (1971). Neurobehaviorla organization of the human
newborn. In J. Hellrnuth (Ed.), Exceptional Infant (V 01. 2, pp. 24-40). New York, NY:
Brunner/Mazel.

Turkewitz, G., Gordon, B. W., & Birch, H. G. (1965). Head turning in the human
neonate: Spontaneous patterns. The Journal of Genetic Psychology, 107, 143-148.

Tumbull, O. H. (2002). The leftward cradling bias and audition: Cross-modal
confusion? A response to Sieratzki, Roy, and W011. Laterality, 7, 355-357.

Tumbull, O. H., & Bryson, H. E. (2001). The leftward cradling bias and
hemispheric asymmetry for speech prosody. Laterality, 6, 21-28.

Tumbull, O. H., & Lucas, M. D. (2000). "Tell Me, Where is [this] Fancy Bred?":
The Cardiac and Cerebral Accounts of the Lateral Cradling Bias. In M. K. Mandal, M. B.
Bulrnan-Fleming & G. Tiwari (Eds.), (pp. 267-287). Dordrecht: Kluwer Academic.

Vauclair, J ., & Donnot, J. (2003). Unpublished data.

von Eye, A. & Clogg, C. C. (Eds). (1994). Latent variables analysis —
Applications for developmental research. Newbury Park, CA: Sage.

Weatherill, R. P., Almerigi, J. B., Levendosky, A., Bogat, A., & Harris, L. J.
(2002). Is Maternal Depression Related to Side of Infant Holding? Poster presentation at
the 13th annual meeting of TENNET, Montreal, Canada, June 14-18.

Weiland, I. H. (1964). Heartbeat Rhythm and Maternal Behavior. Journal of The
American Academy of Child Psychiatry, 3, 161-164.

Weiland, I. H., 81. Sperber, Z. (1970). Patterns of mother-infant contact: The
significance of lateral preference. Journal of Genetic Psychology, 11 7, 157-165.

Whiting, B. B., & Edwards, C. P. (1988). Children of diflerent worlds: The
formation of social behavior. Cambridge, MA: Harvard University Press.

85

   

\lllllillijijllljllimit