University of Southern California Dissertations and Theses

Nice by nature? A twin study of the development and physiology of prosocial personality

(USC Thesis Other)

Nice by nature? A twin study of the development and physiology of prosocial personality

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY 1
IN TWINS

Nice by Nature?: A Twin Study of the Development and Physiology
of Prosocial Personality

Leslie Berntsen

A dissertation presented to the Faculty of the USC Graduate School in partial fulfillment
of the requirements for the degree of Doctor of Philosophy in Psychology

December 2018

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

2
Dedication

For everyone—teachers and mentors, family and friends, colleagues and comrades—who
has ever taught me something. Thank you.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

3
Table of Contents
Nice by Nature?: A Twin Study of the Development and Physiology of Prosocial
Personality ............................................................................................................................1
Dedication ............................................................................................................................2
Table of Contents .................................................................................................................3
Dissertation Introduction .....................................................................................................6
References ......................................................................................................................10
Chapter 1: Genetic and Environmental Bases of Cooperativeness from Late
Childhood to Emerging Adulthood ................................................................................13
Abstract ..................................................................................................................13
Introduction ............................................................................................................14
Measuring Prosociality ..............................................................................15
Phenotypic Development of Prosociality ...................................................16
Genetic and Environmental Etiology of Prosociality ................................17
Cross-Sectional Studies .................................................................18
Longitudinal Studies ......................................................................19
Genetic and Environmental Etiology of Cooperativeness .............20
Gene-Environment Interplay .........................................................21
Methods..................................................................................................................23
Participants .................................................................................................23
Attrition ..........................................................................................24
Procedure ...................................................................................................24
Measures ....................................................................................................25
Statistical Analyses ....................................................................................26
Results ....................................................................................................................28
Psychometric Properties of the Cooperativeness Scale .............................28
Descriptive Statistics ..................................................................................29
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

4
Twin and Cross-Twin/Cross-Wave Correlations .......................................29
Exploratory Univariate Genetic Analyses .................................................30
Longitudinal Genetic Analyses ..................................................................31
Discussion ..............................................................................................................33
Phenotypic Development of Cooperativeness ..........................................33
Genetic and Environmental Etiology of Cooperativeness .........................34
Limited Evidence for Non-Additive Genetic Influences ...............34
Additive Genetic Influences Across Development ........................35
Phenotype-to-Environment Transmission in Female Twins ..........36
Concluding Remarks, Limitations, and Future Directions ........................37
References ..............................................................................................................41
Tables and Figures .................................................................................................52
Supplemental Tables .................................................................................60
Appendices .............................................................................................................80
Chapter 2: Genetic and Environmental Etiology of Autonomic Responses to Fear
and Sadness and their Relationship to Prosocial Personality ......................................83
Abstract ..................................................................................................................83
Introduction ............................................................................................................84
Autonomic Correlates of Fear and Sadness ...............................................84
Sex Differences ..........................................................................................85
Behavior Genetics of Affective Responses ................................................86
Personality-Based Modulations of Affective Responses ...........................87
Affective Responding and Prosociality .....................................................88
The Present Studies ....................................................................................89
Study 1 ...................................................................................................................90
Methods......................................................................................................90
Participants .....................................................................................90
` Exclusion Criteria ..............................................................91
Procedure .......................................................................................92
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

5
Materials ........................................................................................92
Psychophysiological Measures ......................................................93
Cardiovascular Activity .....................................................93
Electrodermal Activity .......................................................93
Data Acquisition and Processing .......................................93
Data Reduction ...............................................................................94
Statistical Analyses ........................................................................94
Results ........................................................................................................95
Manipulation Check .......................................................................95
Descriptive Statistics and Phenotypic Analyses ............................96
Twin Correlations ..........................................................................97
Univariate Genetic Analyses ..........................................................97
Study 2 ...................................................................................................................98
Methods......................................................................................................98
Participants .....................................................................................98
Procedure .......................................................................................98
Measures ........................................................................................99
Statistical Analyses ........................................................................99
Results ......................................................................................................100
Phenotypic Correlations ...............................................................100
Bivariate Genetic Analyses ..........................................................100
Discussion ............................................................................................................101
Nature of Fear and Sadness Responses ....................................................101
Etiology of Fear and Sadness Responses .................................................102
Autonomic Reactivity as a (Partial) Predictor of Cooperativeness .........104
Sex Differences ............................................................................105
Etiology of the Autonomic Reactivity/Cooperativeness Relationship ....106
Limitations and Concluding Remarks .....................................................106
References ............................................................................................................109
Tables and Figures ...............................................................................................119

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

6
Dissertation Introduction
Prosocial tendencies—values, traits and behaviors that benefit others—are of
great interest to social scientists across disciplines and integral to the functioning of
society. In the study of any complex trait, twins serve as a natural experiment of the
(often interactive) effects of nature and nurture, and longitudinal studies offer a unique
opportunity to examine how these influences may change across development. Here, we
analyze data from the Southern California Twin Study of Risk Factors for Antisocial
Behavior (RFAB; Baker et al., 2013), a longitudinal twin study of social and emotional
development from late childhood to emerging adulthood, to examine the genetic and
environmental bases of the development and physiology of prosocial personality.
At four of the five waves of RFAB data collection (Wave 1: 9-10 years, Wave 3:
14-15 years, Wave 4: 16-18 years, Wave 5: 19-20 years), twins provided self-reports of
their prosocial personality traits. Additionally, during Wave 1 (ages 9-10), twins provided
autonomic responses while passively viewing film clips that highlight characters’
negative circumstances (i.e., danger and loss). Through univariate, bivariate, and
longitudinal biometric modeling, we are able to quantify the genetic and environmental
contributions to both behavioral and biological measures of prosocial tendencies and
provide insights into the nature of the relationship between the two. Consisting of two
submission-ready empirical articles (i.e., chapters) addressing three specific aims, this
dissertation represents a comprehensive behavioral genetic examination of the
development and physiology of prosociality.
In Chapter 1, we investigate the phenotypic development (Aim 1a) and the
genetic and environmental etiology (Aim 1b) of prosocial personality traits (indexed by
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

7
the Cooperativeness scale of the Temperament and Character Inventory; Cloninger,
Przybeck, Svrakic, & Wetzel, 1994) from late childhood to emerging adulthood. Overall,
research on the development of prosocial tendencies across adolescence has produced
mixed results (Eisenberg, Spinrad, & Knafo-Noam, 2015). However, several research
teams have observed parabolic trends in prosocial behavior, with the lowest levels
occurring during mid-adolescence (Carlo, Crockett, Randall, & Roesch, 2007; Luengo
Kanacri, Pastorelli, Eisenberg, Zuffianò, & Caprara, 2013). Moreover, in both cross-
sectional and longitudinal twin studies, the heritability of prosociality appears to increase
across the lifespan, with shared environmental factors becoming less influential as people
age (Israel, Hasenfratz, & Knafo-Noam, 2015; Knafo-Noam, Vertsberger, & Israel, 2018).
Here, we replicate both the phenotypic and etiological developmental patterns
previously observed. In both sexes, Cooperativeness substantially decreases at ages 14-15
before gradually returning to previous levels by ages 19-20. Additionally, shared
environmental influences were significant at earlier waves, but ultimately gave rise to
significant genetic influences, although this shift occurs later for female twins (ages 16-
18) than male twins (ages 14-15). In two novel contributions to the behavior genetic
literature on prosociality, we identify a common genetic factor that influences
Cooperativeness from late childhood to emerging adulthood and also provide evidence of
phenotype-to-environment transmission (e.g., Beam & Turkheimer, 2013) of
Cooperativeness in female twins.
In Chapter 2, we shift from behavioral to biological measures and present the
results of two related studies. In Study 1, we examine the genetic and environmental
bases of autonomic responses to fear- and sadness-inducing film clips during late
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

8
childhood (Aim 2). Although a substantial number of studies describe the biological
correlates of specific affective responses (for reviews, see Kreibig, 2010; Wager et al.,
2008), significantly fewer employ twin and family designs to examine how genetic and
environmental factors influence these responses. Only one twin study has relied on film
clips as an emotion induction technique, but it focused on overt facial expressions as the
phenotypic outcome of interest (Kendler et al., 2008). Thus, the present work represents
the first genetically informative study of the autonomic responses to affectively laden
film clips.
After controlling for baseline physiological arousal, we identify significant
genetic influences on heart rate responses to fear and sadness in both sexes. For baseline-
corrected skin conductance levels, we identify significant genetic influences in male
twins, but significant shared environmental influences in female twins—a pattern that
held for both fear and sadness clips. We discuss these findings in light of the
oxytocinergic system’s role in facilitating social and emotional behavior, including
electrodermal responses to threatening stimuli (indirectly via the amygdala; Wood, Ver
Hoef, & Knight, 2014).
In Chapter 2, Study 2, we combine Aims 1 and 2 to investigate whether the
aforementioned autonomic responses to fear and sadness can predict (both concurrently
and longitudinally) Cooperativeness (Aim 3a) and whether shared genetic and/or
environmental influences underlie any observed relationships (Aim 3b). Over the past
two decades, researchers have identified genetic, neuroendocrine, physiological, and
neural markers associated with a wide variety of prosocial tendencies (Hastings, Miller,
Kahle, & Zahn-Waxler, 2013). In particular, autonomic responses to another person’s
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

9
negative circumstances (e.g., danger, pain, loss, etc.) have been found to predict
subjective feelings of empathy and helping behavior, even when such behavior occurs at
a cost to oneself (e.g., Hein, Lamm, Brodbeck, & Singer, 2011). However, it remains
unclear whether the relationship between physiology and prosociality is primarily
mediated by common genetic and/or environmental influences.
In this final set of analyses, we find that autonomic responses to fear and sadness
during childhood partially predict Cooperativeness throughout childhood and adolescence.
Notably, higher heart rate and skin conductance levels were associated with
Cooperativeness in male twins, while lower heart rate responses were associated with
Cooperativeness in female twins, which may reflect an alternative tend-and-befriend
stress response in women (Taylor, 2006; Taylor et al., 2000). Additionally, we find that
common genetic factors jointly influence the relationship between skin conductance
levels and Cooperativeness in male twins. This novel finding adds to the growing body of
literature attesting to genetic overlap between antisocial personality traits and their
biological correlates (e.g., Hicks et al., 2007; Isen, Iacono, Malone, & McGue, 2012;
Wang et al., 2015), and provides initial evidence that similar etiological dynamics are at
play with respect to their more prosocial counterparts.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

10
References
Baker, L. A., Tuvblad, C., Wang, P., Gomez, K., Bezdjian, S., Niv, S., & Raine, A.
(2013). The Southern California Twin Register at the University of Southern
California: III. Twin Research and Human Genetics, 16(1), 336-343.
Beam, C. R., & Turkheimer, E. (2013). Phenotype–environment correlations in
longitudinal twin models. Development and Psychopathology, 25(1), 7-16.
Beam, C. R., Turkheimer, E., Dickens, W. T., & Davis, D. W. (2015). Twin
differentiation of cognitive ability through phenotype to environment
transmission: The Louisville Twin Study. Behavior Genetics, 45(6), 622-634.
Carlo, G., Crockett, L. J., Randall, B. A., & Roesch, S. C. (2007). A latent growth curve
analysis of prosocial behavior among rural adolescents. Journal of Research on
Adolescence, 17(2), 301-324.
Cloninger, C. R., Przybeck, T. R., Svrakic, D. M., & Wetzel, R. D. (1994). The
Temperament and Character Inventory (TCI): A guide to its development and
use (pp. 19-28). St. Louis, MO: Center for Psychobiology of Personality,
Washington University.
Eisenberg, N., Spinrad, T. L., & Knafo-Noam, A. (2015). Prosocial development. In In R.
M. Lerner & M. E. Lamb (Eds.), Handbook of child psychology and
developmental science, Vol. 3: Socioemotional processes (7
th
ed., pp. 610-656).
Hoboken: John Wiley & Sons, Inc.
Hastings, P. D., Miller, J. G., Kahle, S., & Zahn-Waxler, C. (2013). The neurobiological
bases of empathic concern for others. In M. Killen & J. G. Smetana (Eds.),
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

11
Handbook of moral development, Vol. 2, (pp. 411-434). New York: Psychology
Press.
Hein, G., Lamm, C., Brodbeck, C., & Singer, T. (2011). Skin conductance response to the
pain of others predicts later costly helping. PLoS ONE, 6(8), e22759.
Hicks, B. M., Bernat, E., Malone, S. M., Iacono, W. G., Patrick, C. J., Krueger, R. F., &
McGue, M. (2007). Genes mediate the association between P3 amplitude and
externalizing disorders. Psychophysiology, 44(1), 98-105.
Isen, J. D., Iacono, W. G., Malone, S. M., & McGue, M. (2012). Examining
electrodermal hyporeactivity as a marker of externalizing psychopathology: A
twin study. Psychophysiology, 49(8), 1039-1048.
Israel, S., Hasenfratz, L., & Knafo-Noam, A. (2015). The genetics of morality and
prosociality. Current Opinion in Psychology, 6, 55-59.
Kendler, K. S., Halberstadt, L. J., Butera, F., Myers, J., Bouchard, T., & Ekman, P.
(2008). The similiarity of facial expressions in response to emotion-inducing
films in reared-apart twins. Psychological Medicine, 38(10), 1475-1483.
Knafo-Noam, A., Vertsberger, D., & Israel, S. (2018). Genetic and environmental
contributions to children's prosocial behavior: brief review and new evidence
from a reanalysis of experimental twin data. Current Opinion in Psychology, 20,
60-65.
Kreibig, S. D. (2010). Autonomic nervous system activity in emotion: A review.
Biological Psychology, 84(3), 394-421.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

12
Luengo Kanacri, B. P., Pastorelli, C., Eisenberg, N., Zuffianò, A., & Caprara, G. V.
(2013). The development of prosociality from adolescence to early adulthood:
The role of effortful control. Journal of Personality, 81(3), 302-312.
Wager, T. D., Barrett, L. F., Bliss-Moreau, E., Lindquist, K. A., Duncan, S., Kober, H.,
…Mize, J. (2008). The neuroimaging of emotion. In: M. Lewis, J. M. Haviland-
Jones, & L. F. Barrett (Eds.), Handbook of emotions (3
rd
ed., pp. 249-271). New
York: Guilford Press.
Wang, P., Gao, Y., Isen, J., Tuvblad, C., Raine, A., & Baker, L. A. (2015). Genetic
covariance between psychopathic traits and anticipatory skin conductance
responses to threat: Evidence for a potential endophenotype. Development and
Psychopathology, 27(4pt1), 1313-1322.
Wood, K. H., Ver Hoef, L. W., & Knight, D. C. (2014). The amygdala mediates the
emotional modulation of threat-elicited skin conductance response. Emotion,
14(4), 693-700.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

13
Chapter 1: Genetic and Environmental Bases of Cooperativeness from Late
Childhood to Emerging Adulthood
Abstract
Although fundamental to a healthy and just society, prosocial tendencies have
received significantly less empirical attention than their antisocial counterparts.
Developmental research has produced mixed findings regarding how prosociality may
ebb and flow during adolescence, while the few genetically informative studies focused
on this age group suggest that prosocial tendencies become increasingly heritable across
development. In the present study, we investigate the phenotypic development and
etiological bases of self-reported prosocial personality traits (assessed via the
Cooperativeness scale of the Temperament and Character Inventory; Cloninger, Przybeck,
Svrakic, & Wetzel, 1994) in an ethnically diverse community sample of twins at four
time points: ages 9-10, 14-15, 16-18, and 19-20. Consistent with previous behavior
genetic research, the influence of genetic factors on prosociality became more
pronounced with age, as the influence of shared environmental factors became less
pronounced. Moreover, the results of longitudinal multilevel structural equation modeling
(ML-SEM) yielded two novel contributions: a common genetic factor underlying
Cooperativeness across development in both males and females, as well as phenotype-to-
environment transmission (PxE) of Cooperativeness in females.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

14
Prosocial tendencies—emotions, attitudes, traits, and behaviors that benefit
others—are integral to the functioning of society. From the birth of Athenian democracy
(“What makes a good citizen?”) to the advent of evolutionary biology (“Can non-
adaptive altruism truly exist?”), there is no shortage of questions addressing when, how,
and why people think, feel, and act in ways that promote the welfare of those around
them. The psychological study of prosocial behavior became increasingly popular after
the 1964 murder of Kitty Genovese, when researchers began to identify the barriers that
could have prevented so many people from offering help despite their awareness of the
situation at hand. (Many details of the Genovese murder, namely the purported lack of
bystander response, have since been called into question; for a discussion, see Singal,
2016). In the decades that have followed, social scientists have broadened their study to
include a variety of phenomena often subsumed under the umbrella term “prosociality.”
(For detailed discussions of the multidimensionality of prosocial tendencies, see Padilla-
Walker & Carlo, 2014, and Eisenberg & Spinrad, 2014).
Penner and colleagues (2005) have distinguished between micro, meso, and
macro levels of analysis, which address, respectively, individual differences in prosocial
tendencies, the situational dynamics of prosocial interactions between two people, and
the nature of large-scale, collective prosocial action. Each of these levels can be explored
in further detail. For example, assisting with a goal-directed action (helping), providing a
necessary material good (sharing), responding to emotional needs (comforting),
providing useful knowledge (informing), and contributing to the pursuit of a
superordinate goal (cooperating) all constitute different forms of meso level prosocial
behavior (Dunfield, Kuhlmeier, O’Connell, & Kelly, 2011; Dunfield & Kuhlmeier, 2013).
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

15
Likewise, a variety of factors may motivate someone to act prosocially: pure concern for
another person’s well being (i.e., altruism), a reflexive instinct to provide assistance in
dire circumstances, a feeling of social obligation or a desire for social approval, potential
benefits to oneself, or perhaps a combination thereof (Carlo & Randall, 2002). Moreover,
a wide variety of micro level individual difference variables (e.g., personality traits,
perspective-taking abilities, positive or negative emotionality, etc.) can help explain
variability in meso or macro level prosocial behavior.
Measuring Prosociality
Several methods can be used to empirically assess prosocial tendencies. In
observational studies, subjects generally witness someone in need of assistance or support
while trained observers code their responses. Behavior observation is especially useful
when studying infants or young children who may not have the ability or insight to
accurately report their own thoughts or feelings, but is less common in studies involving
older children, adolescents, and adults (Eisenberg, Fabes, & Spinrad, 2006). For these
populations, researchers generally opt for self- or observer-report questionnaires that can
take a number of forms. Some (e.g., the Strengths and Difficulties Questionnaire [SDQ];
Goodman, 1997) are scenario-based, in which respondents indicate the likelihood that
they have behaved or would behave prosocially in response to certain situations. Others,
including the Cooperativeness scale of the Temperament and Character Inventory (TCI;
Cloninger, Przybeck, Svrakic, & Wetzel, 1994; Cloninger, Svrakic, & Przybeck, 1993)
employed in the current study and described in the Methods section) assess prosocial
values, affective responses, or personality traits that do not reference specific behaviors.
Although these questionnaires can streamline large amounts of data collection, the extent
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

16
to which survey responses can accurately predict real-world behavior remains an elusive
question for social scientists.
A limited number of studies investigate the longitudinal development of
prosociality across adolescence (for a review, see Padilla-Walker & Carlo, 2014), and
even fewer specifically analyze the genetic and environmental contributions to prosocial
tendencies during that period. As a result, the following sections will summarize findings
across the various forms of prosociality (i.e., traits, as well as reported and observed
behaviors), but will specify which form is germane to the particular study at hand.
Phenotypic Development of Prosociality
Different forms of prosociality emerge over the course of development. Although
the earliest forms of prosocial responding (e.g., crying in response to others’ crying)
reflect the limitations of our earliest capabilities, developing social, cognitive, and
linguistic abilities enable more advanced responses (e.g., offering anticipatory emotional
support) during childhood and adolescence. However, prosocial behavior does not
necessarily become more frequent as a linear function of sociocognitive development
(see Eisenberg, Spinrad, & Knafo-Noam, 2015 for a review). Indeed, as people age,
prosocial responses can become more targeted—influenced by the cost of the behavior,
social scripts that dictate who needs or deserves help, among other situational
contingencies (e.g., Eisenberg & Fabes, 1998; Hay & Cook, 2007). Although prosocial
behavior does generally become more common in adolescence (relative to childhood;
Eisenberg & Fabes, 1998), it is not entirely clear how prosociality (in its various forms)
may ebb and flow during this crucial developmental period.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

17
Using reports from multiple informants (i.e., mothers, teachers, and the children
themselves) across two different 10-15 year old samples, Nantel-Vivier and colleagues
(2009) discovered that the frequency of engaging in a wide variety of prosocial behaviors
does not always uniformly increase in frequency during childhood and adolescence. After
identifying several developmental trajectory groups, all but one were characterized by
relatively stable or, more frequently, declining levels of prosocial behavior, regardless of
informant. However, it is possible that these results constitute one portion of a larger
developmental pattern. In longitudinal studies tracking participants from late childhood to
emerging adulthood, initial decreases have been followed by subsequent increases in
prosocial tendencies. Among male and female Italian adolescents and emerging adults
(ages 13-21), there appears to be a parabolic trend in self-reported likelihood of engaging
in caring, sharing, and helping behaviors, with the lowest levels occurring at ages 16-17
(Luengo Kanacri, Pastorelli, Eisenberg, Zuffianò, & Caprara, 2013). Similar results have
been obtained in rural American adolescents: the self-reported frequency of recent
prosocial behavior declined from approximately age 12 through age 17, with a slight
increase observed during participants’ final year in high school (Carlo, Crockett, Randall,
& Roesch, 2007).
Genetic and Environmental Etiology of Prosociality
Behavior genetic studies seek to quantify the relative contributions of genetic and
environmental factors to a given phenotype by comparing the relative degree of similarity
between genetically related individuals. In the classical twin design, researchers compare
monozygotic (MZ) twins (who are assumed to share 100 percent of their genes) to
dizygotic (DZ) twins (who, like full, non-twin, siblings, share 50 percent of their genes,
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

18
on average). Given those genetic differences, if MZ twin pairs are more similar than DZ
twin pairs, any observed variance in that trait is likely due to genetic factors, broadly
construed. Conversely, if MZ twins (who have been raised together) are no more similar
to each other than DZ twins (who have been raised together), any observed variance is
likely due to shared environmental factors, broadly construed. More specifically, additive
genetic (A) effects are those that reflect the influence of a single allele on a given
phenotype, while non-additive (dominant; D) genetic effects reflect the combined,
interactive effects of multiple alleles. Likewise, shared environmental (C) effects are
those that are common to both twins in a pair (e.g., parenting style, childhood
neighborhood, etc.), while non-shared environmental (E) effects are those unique to each
twin (e.g., individual life experiences).
Cross-sectional studies. Genetic influences on prosocial tendencies appear early
in life and, with small exception, appear to persist throughout development. In a sample
of South Korean twins ranging from 2-9 years old, mother-reported prosocial behaviors
were 55% heritable for both males and females (Hur & Rushton, 2007). Likewise, Knafo-
Noam and colleagues (2015) found significant additive genetic influences in a variety of
mother-rated prosocial tendencies (i.e., empathy, kindness, sharing, helping and
comforting) in a sample of 3-7 year old Israeli twins, as well as evidence of a highly
heritable (69%) common factor that can account for the variance shared by those five
facets. Across a variety of measures, the heritability of prosociality appears to increase
throughout childhood and adolescence (Israel, Hasenfratz, & Knafo-Noam, 2015),
although there are several factors that moderate this trend, including the kind of
prosociality being measured, the method of measurement, and the target of the behavior
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

19
(for a review, see Fortuna & Knafo, 2014). In a sample of English twins ranging from 5
to 17 years old, genetic effects on parent-reported prosocial behavior were more
pronounced in twins older than 11 than those younger than 11. Although female twins, as
expected, were more prosocial than their male counterparts, distinct sex-specific genetic
and environmental influences were not responsible for these differences (Scourfield, John,
Martin, & McGuffin, 2004). Behavior genetic studies of prosociality in older adults are
rare, but one study conducted on Australian twins (mean age = 62 years) found
Cooperative character traits (the focus of the present study) to be 27% heritable, with the
remaining variance attributable to non-shared environmental factors (Gillespie, Cloninger,
Heath, & Martin, 2003).
Longitudinal studies. Although cross-sectional studies can provide an overview
of temporal dynamics, they cannot offer insight into the developmental processes that
occur as people age. However, the few genetically informative longitudinal studies of
prosociality do partially confirm the trend of shared environmental contributions giving
rise to more pronounced additive genetic contributions across development. But, despite
this overall shift, genetic influences are detectable very early in life. Beginning at 14
months, genetic factors influence mother-reported empathy and behavioral responses to
simulated pain, and continue to exert their influence through 20, 24, and 36 months
(Zahn-Waxler, Schiro, Robinson, Emde, & Schmitz, 2001). Using a similar paradigm on
a substantially larger sample, Knafo and colleagues (2008) found one common factor to
account for both cognitive empathy (attempting to understand another’s distress) and
affective empathy (responding to another’s distress) at 14, 20, 24, and 36 months. This
common factor could be explained by shared environmental (C) factors at 14 and 20
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

20
months, but by additive genetic (A) factors at 24 and 36 months. A similar shift from
shared environmental to additive genetic influences was also observed in parent-rated
prosocial behavior of twins from ages 2 to 7, with increasing heritability due largely to
new genetic effects emerging at later ages (Knafo & Plomin, 2006). Finally, self-reported
prosocial behavior could be attributed, in part, to significant additive genetic influences at
age 15, but only significant non-shared environmental influences at age 17 (Gregory,
Light‐Häusermann, Rijsdijk, & Eley, 2009). Taking note of these findings, there does
not appear to be a single age (or even a single age range) at which shared environmental
factors begin to give rise to additive genetic factors. Rather, in each of these studies,
shared environmental influences exert their effects at earlier waves of data collection,
while additive genetic influences exert their effects at later waves of data collection,
rather than any specific age(s), per se.
Genetic and environmental etiology of Cooperativeness. Cloninger’s
psychobiological model of personality (from which the Temperament and Character
Inventory was derived) consists of four temperament dimensions and three character
dimensions, including Cooperativeness. Briefly, those who score high on the
Cooperativeness scale are socially tolerant, empathic, compassionate, and ethical, while
those who score low are generally socially intolerant, critical, vengeful, and opportunistic
(Cloninger & Garcia, 2015). According to this model, temperament dimensions reflect
genetically based dispositions that arise early in life and are influenced by distinct neural
systems. In contrast, character dimensions (including Cooperativeness) are thought to
arise largely, if not entirely, from sociocultural learning that occurs later in life
(Cloninger et al., 1993).
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

21
Few behavior genetic studies (all cross-sectional) have specifically assessed the
etiology of Cooperativeness. Lending slight support to Cloninger’s theory, shared
environmental factors (C) significantly influence Cooperativeness beginning at age 9
(Isen, Baker, Raine, & Bezdjian, 2009) and continuing, although decreasing in magnitude,
through age 15 (Garcia et al., 2013). At age 15, significant additive genetic influences
were also detected (Garcia et al., 2013). Although a follow-up study of those 15 year-old
twins (Lester et al., 2016) found evidence of shared environmental (C) influences in three
of the five subscales of the Cooperativeness scale, none of the estimates were statistically
significant. At some point between adolescence and early adulthood, shared
environmental influences seem to disappear, giving rise to additive genetic (A; 47%) and
non-shared environmental (E; 53%) effects (Ando et al., 2004). However, because the
subjects in this study ranged in age from 15 to 30 years old, these results may obscure
important age-related differences.
Taken together, this pattern of results mirrors the broader behavior genetic
literature on prosociality, as well as a wide variety of other phenotypes, including anxiety,
depression, cognitive ability, and IQ (Beam, Turkheimer, Dickens, & Davis, 2015;
Bergen, Gardner, & Kendler, 2007; Dickens & Flynn, 2001; Scourfield et al., 2003).
Across childhood and adolescence, shared environmental contributions become less
influential and genetic contributions become more influential. Our attention now turns to
potential explanations for this developmental pattern.
Gene-environment interplay. One leading theoretical explanation (Scourfield et
al., 2004; Israel et al., 2015; Knafo-Noam, Vertsberger, & Israel, 2018) for this shift
involves gene-environment correlation (rGE), which refers to the relationship between
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

22
individuals’ genetic endowments and the environments to which they are exposed.
Plomin and colleagues (1977) described three different kinds of rGE: passive (in which
people are placed into certain environments based on their genotype), active (in which
people select certain environments based on their genotype), and evocative (in which
people elicit certain environmental feedback based on their genotype). Developmentally,
passive rGE is more likely to occur early in life (when parents are most prone to provide
certain environments for their children), while active rGE becomes increasingly
influential as children age (and are more likely to select environments for themselves).
On the other hand, evocative rGE can occur throughout the lifespan (Scarr & McCartney,
1983).
Taken together, these reciprocal influences between genes and environment
contribute to an oft-cited “snowball effect,” whereby initially small genetically-based
phenotypic differences become increasingly more pronounced as people are exposed to
environments that amplify those differences (e.g., Scourfield et al., 2004). More recently,
Beam and Turkheimer (2013) have proposed that differences that occur within families
are one possible factor that can initiate the snowball effect, thus serving as one source of
rGE across development and accounting for the age-related increases in heritability that
have been observed in a variety of domains.
Evidence suggests that rGE does influence prosociality, as children who are
(genetically) more prosocial tend to elicit warmer, more prosocial responses from parents
(Avinun & Knafo, 2014) and unfamiliar peers (DiLalla, Bersted, & John, 2015).
However, the extent to which within-family phenotype-to-environment transmission (e.g.,
Beam & Turkheimer, 2013) can explain the observed gene-environment correlations
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

23
remains an open question (Knafo-Noam et al., 2018). The present study represents the
first study to test for phenotype-environment transmission of prosociality across
development using a within-family design, as well as the first longitudinal study of the
genetic and environmental bases of Cooperativeness.
Methods
Participants
Participants are drawn from the Southern California Twin Study of Risk Factors
for Antisocial Behavior (RFAB; Baker et al., 2013), a longitudinal study of the genetic
and environmental bases of social and emotional development from late childhood to
emerging adulthood. Behavioral, neuropsychological, and physiological data were
collected from a community sample of 783 sets of twins and triplets across five waves of
data collection: Wave 1: ages 9-10, Wave 2: ages 11-13, Wave 3: ages 14-15, Wave 4:
ages 16-18, and Wave 5: ages 19-20. While the majority of families were recruited prior
to Wave 1, additional families were recruited to participate beginning in Wave 3. The
sample reflects both the ethnic and socioeconomic diversity of the greater Los Angeles
area. Demographic statistics for the participants providing data for the current study (N =
1535; 97.6% of the full sample) are available in Table 1. For the current study, N = 288
twins participated in one wave, N = 351 participated in two waves, N = 513 participated
in three waves, and N = 383 participated in all four waves during which the appropriate
measure was administered. (Full wave-by-wave completion rates can be found in
Supplemental Table 1.)
DNA microsatellite analysis was used to determine the zygosity of same-sex twin
pairs. (By definition, opposite-sex twin pairs are dizygotic.) For a pair of twins to be
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

24
classified as monozygotic, they must have shared at least seven concordant DNA markers.
For a pair of twins to be classified as dizygotic, they must have shared one identifiable
discordant marker (Baker et al., 2013). For same-sex twin pairs whose zygosity could not
be determined using this method, zygosity was assigned according to responses on the
Twin Similarity Questionnaire (Lykken, 1978), which has been found to predict DNA
zygosity with 90% accuracy in the current sample (Baker et al., 2006).
Attrition. Attrition analyses were conducted to determine whether participants
who participated in multiple waves of data collection were more prosocial than those who
participated in fewer waves or dropped out of the study entirely after one wave of
participation. For those who began participating during the first wave of data collection,
twins scoring higher in Cooperativeness were more likely to continue participating after
initial recruitment (OR: = 4.17, 95% CI: 1.10-15.80). For those who began participating
during the third wave of data collection, twins scoring higher in Cooperativeness were no
more likely than their less Cooperative counterparts to continue participating after initial
recruitment (OR: = 14.55, 95% CI: 0.97-219.13). At each wave of data collection, twins
who participated more frequently were significantly more likely to have higher
Cooperativeness scores (all ps < .01). Therefore, the longitudinal analyses that follow
should be interpreted with caution since they rely on a sample that skews toward the
especially prosocial.
Procedure
During the first wave of data collection, twins participated in 4-6 hour-long
laboratory visits to the Southern California Twin Project. During these visits, they
completed a battery of behavioral surveys and neuropsychological measures, as well as
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

25
psychophysiological testing. In later waves, families were given the option to participate
remotely by completing the behavioral surveys online or requesting/returning them via
postal mail. Twins completed abbreviated versions of the Junior Temperament and
Character Inventory at ages 9-10 and 14-15 and the Temperament and Character
Inventory at ages 16-18 and 19-20. When participants were 9-10 years old, they provided
oral responses to the JTCI during an in-person interview. At later waves of data
collection, participants independently completed the applicable measure, either during
their lab visit or via mail/online surveys.
Measures
The Junior Temperament and Character Inventory (JTCI; 95 total items; Luby,
Svrakic, McCallum, Przybeck, & Cloninger, 1999) and Temperament and Character
Inventory (TCI; 125 total items; Cloninger et al., 1994; Cloninger et al., 1993) both
contain four temperament scales (Novelty Seeking, Harm Avoidance, Reward
Dependence, Persistence) and three character scales (Self-Directedness, Cooperativeness,
and Self-Transcendence), although the current study only focuses on the Cooperativeness
scale. In the TCI, the Cooperativeness scale can be further subdivided into five subscales:
Social Acceptance, Empathy, Helpfulness, Compassion, and Pure-Heartedness. Subscales
for the JTCI have not been formally published, but the same five Cooperativeness
subscales can be computed upon close inspection of the items (Isen, 2008, unpublished
analyses). All JTCI and TCI items are binary choice statements (coded as 0 = No, 1 =
Yes). Scores are computed by calculating the mean of the constituent items, such that
both overall scale and individual subscale scores can range from 0 to 1. The specific
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

26
items that comprise each Cooperativeness subscale in the JTCI and TCI can be found in
Appendix A and Appendix B, respectively.
Statistical Analyses
Several preliminary analyses were performed to assess the psychometric
properties of the Cooperativeness scale and the phenotypic development of
Cooperativeness in male and female twins. Twin correlations were also computed in
order to provide an initial sense of the genetic and environmental factors influencing
Cooperativeness at each wave. When correlations for monozygotic (r
MZ
) and dizygotic
(r
DZ
) twins are relatively similar in magnitude, shared environmental (C) influences are
likely at play. When r
MZ
is greater than r
DZ
, genetic influences are likely at play: non-
additive (D) genetic influences if r
MZ
> 2r
DZ
, and additive (A) genetic influences if r
DZ
<
r
MZ
< 2r
DZ
. Since Cooperativeness scores were highly negatively skewed, values were
squared in order to normalize the distribution prior to performing twin correlations and
genetic analyses. Furthermore, for longitudinal analyses, square-transformed
Cooperativeness scores at each wave were regressed on subjects’ age at that wave, such
that the unstandardized residual scores serve as the unit of analysis. (Psychometric and
descriptive statistics, as well as phenotypic analyses, are based on untransformed scores.)
Because twin correlations indicated the possibility of non-additive genetic (D)
influences, we employ exploratory non-normal structural equation modeling (nnSEM) to
simultaneously test for A, C, D, and E effects for twins that have been raised together
(e.g., Ozaki, Toyoda, Iwama, Kubo, & Ando, 2011). By making use of third-order
moments and non-normally distributed variables, nnSEM is able to overcome a primary
shortcoming of the classical twin design, which can only test models including three
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

27
parameters (ACE or ADE) to ensure non-negative degrees of freedom. Additionally, for
purposes of comparison, we also test the more traditional sex-limited ACE and ADE
models in Mx (Neale, Boker, Xie, & Maes, 2003).
Finally, to analyze our longitudinal data (and to test for phenotype-to-
environment transmission of Cooperativeness), we employ a series of multilevel
structural equation models (ML-SEM; e.g., Beam et al., 2015). In these models, genetic
and environmental effects on the observed variables are partitioned at both the within-
and between-family levels in a manner analogous to the traditional ACE model. The
combination of within-family (Aw) and between-family (Ab) genetic effects represent
additive (A) genetic influences; between-family environmental (Eb) effects represent
shared environmental (C) influences that lead to similarity between two twins in a pair;
within-family environmental (Ew) effects represent non-shared environmental (E)
influences that lead to dissimilarity between twins, as well as measurement error. As in
traditional twin modeling, parameters and path loadings can be dropped or constrained to
specific values in order to satisfy assumptions and identify the best-fitting, most
parsimonious model. Here, within-family genetic (Aw) effects are only modeled for
dizygotic twins, since monozygotic twins share 100% of their genes and, thus, have no
within-family genetic variation. Likewise, all path loadings on latent Aw and Ab
variables reflect the fact that monozygotic and dizygotic twins share 100% and 50% of
their genes, respectively. In the final model, phenotype-to-environment transmission is
modeled in the paths between an observed phenotype at time t and the latent Ew effects at
time t + 1, representing the influence of pre-existing Cooperativeness levels on future
individual variation in that trait. Lastly, because multilevel modeling assumes that two
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

28
members of a dyad cannot be meaningfully distinguished from one another and because
opposite-sex twins differ along a dimension (i.e., gender) that is theoretically relevant to
present study (and, thus, cannot justifiably be treated as indistinguishable), we restrict our
longitudinal analyses to same-sex monozygotic and dizygotic twin pairs. (For a more
detailed discussion of the impact of (in)distinguishability on the selection of dyadic
analysis techniques, see Ledermann & Kenny, 2017).
Results
Psychometric properties of the Cooperativeness scale
Overall, the Cooperativeness scale demonstrates acceptable reliability across all
four waves of data collection: at ages 9-10, α = .64; at ages 14-15, α = .73; at ages 16-18,
α = .78; and at ages 19-20, α = .78. However, alpha values are substantially lower for
each of its five subscales, with an average α = .42 across all subscales and waves. (Full
reliability statistics can be found in Supplemental Table 2.) To further investigate the
reliability of each of the subscales, percent agreement and Cohen’s kappa values were
computed for every possible item-by-item combination in each of the five subscales
across all four waves. (See Supplemental Tables 3-7 for complete results and
Supplemental Table 8 for summary statistics.) Due to the low alpha values, low kappa
values, and low item-by-item agreement for the subscales, as well as the limited amount
of variance to be observed in subscale scores comprised of three to six binary choice
items, all phenotypic and biometric analyses focus only on the overall Cooperativeness
scale (comprised of 20 items in the JTCI and 25 items in the TCI).
Descriptive Statistics
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

29
Mean levels of Cooperativeness scale scores stratified by sex are presented in
Table 2. At each of the four waves, female twins scored significantly higher than male
twins, all ps < .01. In a 4 (time) x 2 (sex) mixed ANOVA, a significant main effect of
time indicated that Cooperativeness scores significantly changed across development, F(3,
379) = 13.17, p < .01, while a non-significant time x sex interaction indicated that these
developmental changes in Cooperativeness were similar for males and females, F(3, 379)
= 0.64, p = .59. Pairwise comparisons revealed, for both sexes, a significant decrease in
Cooperativeness from ages 9-10 to ages 14-15 and a significant increase from ages 14-15
to ages 16-18, all ps < .04. The subsequent increase from ages 16-18 to ages 19-20 was
significant for females, p = .01, but not for males, p = .18. For each of the four waves,
Cooperativeness scale scores were not significantly related to individual participants’
ages at that wave, all ps > .16. In other words, participants at the higher end of the age
range at each wave were not necessarily more cooperative than those at the lower end of
each age range. Finally, for both sexes, Cooperativeness was highly stable across
development, for both sexes, all ps < .01 (see Supplemental Table 9 for full cross-wave
correlations).
Twin and cross-twin/cross-wave correlations
Twin correlations (each twin correlated with their co-twin at a single wave) are
presented along the diagonals of each of the five sections of Table 3. For male twins, the
twin correlations between MZ twin pairs and DZ twin pairs are relatively similar in
magnitude at ages 9-10. For female twins, the twin correlations between MZ twin pairs
and DZ twin pairs are relatively similar in magnitude at ages 9-10 and ages 14-15. As a
result, significant shared environmental (C) influences are likely at play at those time
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

30
points. Beginning at ages 14-15 for males and ages 16-18 for females, MZ twin
correlations become more than twice as large as DZ twin correlations, which indicate that
non-additive genetic (D) influences are likely at play.
Cross-twin/cross-wave correlations (one twin at one wave correlated with their
co-twin at a different wave) are presented on the respective off-diagonals of Table 10.
When MZ twins are more similar to each other at different time points compared to DZ
twins, developmental stability in that trait can likely be attributed to genetic factors. This
appears to be the case for Cooperativeness: for both sexes, cross-twin/cross-wave
correlations were substantially higher for MZ twins (average r = .27) than DZ twins
(average r = .10)
Exploratory univariate genetic analyses
Univariate ACDE model-fitting results derived from non-normal structural
equation modeling (nnSEM) are presented separately for males and females in
Supplemental Tables 10 and 11. As expected based on the twin correlations, significant C
effects were present for males at ages 9-10 and for females at ages 9-10 and 14-15.
Significant D effects emerged beginning at ages 14-15 for males and ages 16-18 for
females, and persisted through ages 19-20 for both sexes. Using Mx (Neale et al., 2003),
we also fit traditional univariate sex-limitation ACE and ADE models to the data. In ACE
models, a scalar model (which constrains male and female parameter estimates to
equality) provided the best fit at each wave: reduced to a CE model at ages 9-10 and
reduced to an AE model at each subsequent wave (see Supplemental Table 12). In ADE
models, scalar models likewise provided the best fit, and could be further reduced to an
AE model at each wave (see Supplemental Table 13).
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

31
We then compared the broad- and narrow-sense heritability estimates obtained
using all three methods: ACDE models with nnSEM, sex-limited ACE models, and sex-
limited ADE models (see Table 4). Because the heritability estimates between ACDE and
ACE models were relatively comparable to one another for both sexes at every wave of
data collection, this implies that most of the non-additive genetic effects detected in an
ACDE model would likely be recognized as additive genetic effects in a traditional ACE
model. By extension, fitting ADE models to the present data would likely lead to inflated
heritability estimates, which can be observed in the comparison between ACDE and ADE
models. Likewise, inclusion of non-additive genetic effects does not appear to be crucial,
as the D parameter could be dropped from each univariate ADE model without a
significant drop in fit. As a result, and because we are more theoretically interested in the
presence (and decline) of shared environmental influences across development than
distinguishing between additive and non-additive genetic effects, the subsequent
longitudinal analyses will employ ACE models.
Longitudinal genetic analyses
Full ML-SEM model-fitting results are presented in Table 5. For Models 2-10,
each row represents an additional constraint added to (i.e., parameters removed from) the
previous admissible model, such that a significant chi-square difference test (p < .05)
reflects reduced fit. For Models 11-14, each row represents the additional phenotype-to-
environment transmission parameter(s) added to the previous admissible model, such
that a significant chi-square difference test reflects improved fit. In all cases, lower AIC
and BIC values indicate better fitting models
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

32
To summarize the admissible model reductions: One common factor (in which
males and females were constrained to equality) could account for the between- (Ab) and
within-genetic (Aw) variance in Cooperativeness across development. Additionally, the
variance of the occasion-specific Ab and Aw effects for both male and female twins
could be constrained to zero. Likewise, the variances of the between-environmental (Eb)
effects could be constrained to equality between males and females and the covariances
between occasion-specific Eb effects could be constrained to zero without any significant
drops in model fit. (A graphic representation of this model can be found in Figure 1.)
Subsequently, four different models were fit to test for evidence of phenotype-to-
environment transmission (PxE). The first (Model 11) included separate PxE parameters
for males and females at each of three points of estimation (indicated with red arrows in
Figure 1). Additional models tested for PxE effects only in male twins (Model 12), only
in female twins (Model 13), and only in female twins with each of the three PxE
parameter estimates constrained to equality (Model 14). Of these, only Model 14
provided a significantly better fit to the data than a model that did not estimate PxE.
Parameter estimates for the PxE pathways included in Models 11-14 are displayed
in Table 6. Positive PxE parameters indicate that greater phenotypic differences between
two twins in a pair subsequently predict more distinct within-family (i.e., non-shared)
environmental experiences for those two twins (i.e., divergence over time). Negative PxE
parameters indicate that greater phenotypic differences between two twins in a pair
subsequently predict more similar within-family (i.e., non-shared) environmental
experiences for those two twins (i.e., convergence over time). Thus, the present results
provide evidence of phenotype-to-environment transmission in female twins, with the
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

33
effects driven primarily by convergence in within-family environmental exposure
occurring between the ages of 9-10 and 14-15.
Discussion
Phenotypic Development of Cooperativeness
Consistent with well-established gender differences in prosociality (e.g.,
Eisenberg & Fabes, 1998; Eisenberg et al., 2015), female twins reported higher levels of
Cooperativeness than male twins at every wave of data collection. However, we observed
a similar developmental trend in both sexes, wherein Cooperativeness decreased from
late childhood (ages 9-10) to mid-adolescence (ages 14-15) before returning to previous
levels by late adolescence (ages 16-18). Although longitudinal studies of the development
of prosociality during adolescence are relatively rare and measure a variety of related
constructs, similar non-linear changes have been observed in caring, sharing, helping, and
other related behaviors in Italian (Luengo Kanacri et al., 2013) and American adolescents
(Carlo et al., 2007). One potential theoretical explanation for this pattern may lie in the
fact that prosocial tendencies become increasingly selective during adolescence—more
frequently directed towards friends and more frequently withheld from neutral, unknown,
or antagonistic peers (e.g., Güroğlu, Bos, & Crone, 2014). In the years that follow, age-
related social and moral development (e.g., Eisenberg, Hofer, Sulik, & Liew, 2013) may
facilitate the return to pre-adolescent levels of prosocial behavior.
In order to preserve the psychometric integrity of our findings, we restricted the
present analyses to the overall Cooperativeness scale of the Temperament and Character
Inventory. However, it should be cautiously noted that the decline and subsequent
increase in Cooperativeness appears to be primarily driven by scores on the Compassion
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

34
subscale, which measures subjects’ propensity to respond forgivingly (vs. vengefully) to
those who have wronged them. (Supplemental analyses not presented here.) This
observation is consistent with the notion that different aspects of prosociality can follow
distinct trajectories across development (Eisenberg & Fabes, 1998; Eisenberg et al.,
2015) and with research investigating the development of forgiveness as a dispositional
trait. Specifically, a general willingness to forgive has been found to be higher among
adolescents aged 11-12 years relative to those aged 13-14 years (Chiaramello, Mesnil,
Muñoz Sastre, & Mullet, 2008) and higher among older adults (ages 61-92) than younger
adults (ages 18-23; Cheng & Yim, 2008).
Genetic and Environmental Etiology of Cooperativeness
Limited evidence for non-additive genetic influences. In wave-by-wave
univariate analyses, we find partial evidence for non-additive (modeled as dominant)
genetic influences on Cooperativeness during adolescence. Using non-normal structural
equation modeling (nnSEM; Ozaki et al., 2011) capable of simultaneously estimating A,
C, D, and E effects, significant dominance effects emerged at ages 14-15 in males and
ages 16-18 in females and persisted through ages 19-20 in both sexes. However, the
broad-sense (A+D) heritability estimates obtained through nnSEM and the narrow-sense
(A) heritability estimates obtained via a traditional ACE model were relatively
comparable. Moreover, in traditional ADE model fitting, the D parameter could be
dropped in favor of an AE model at each wave without a significant drop in fit.
Non-additive (dominant) genetic effects have been found to influence a variety of
personality traits, including Cloninger’s temperament dimensions among Australian
adults (Keller, Coventry, Heath, & Martin, 2005), prosocial behavior in American
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

35
preschoolers (DiLalla et al., 2015), as well as overall subjective well-being in zoo-kept
orangutans (James, King, & Weiss, 2012). An additional twin study of prosocial behavior
in South Korean children also provides likely evidence of non-additive genetic influences
(by virtue of MZ twin correlations more than double the magnitude of DZ twin
correlations), although the authors did not formally test for D effects via genetic
modeling (Hur & Rushton, 2007). However, artificially inflated MZ twin correlations can
arise if the equal environments assumption (EEA) is violated or if genetic and
environmental effects interact with one another to influence a given phenotype (discussed
further in the Limitations section). As a result, we have reason to believe that our
observed effects may not be entirely indicative of genetic dominance, per se.
Additive genetic influences across development. Consistent with the broader
behavior genetic literature on prosociality (Eisenberg et al., 2015), we observe increasing
heritability of Cooperativeness across development, largely at the expense of shared
environmental influences. Despite this shift, we find that a common factor can account
for the genetic variance in Cooperativeness across development (including the earlier
ages at which shared environmental influences are more pronounced) and that occasion-
specific genetic influences do not play an appreciable role at any wave of data collection.
Previous research has found a common genetic factor to underlie different components of
prosocial personality (Knafo-Noam et al., 2015), prosocial behavior (Lewis & Bates,
2011), and empathy (Knafo et al., 2008), as well as reports of twins’ empathy provided
by different informants (Knafo et al., 2008). To our knowledge, this is the first study to
identify a common genetic factor that influences any aspect of prosociality across
development.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

36
Molecular genetic research has begun to identify the specific polymorphisms
associated with prosociality, including those related to the dopaminergic (Reuter, Frenzel,
Walter, Markett, & Montag, 2011), oxytocinergic (Uzefovsky et al., 2015; Wu, Li, & Su,
2012), and vasopressinergic (Uzefovsky et al., 2015) systems. Given our findings, it
appears unlikely that these (or other yet discovered) genes implicated in prosocial
tendencies “turn on” at specific points in development (e.g., Whitelaw & Whitelaw,
2006). Rather, it seems more likely that “the role of genes becomes different as children’s
social environment changes” (Knafo-Noam et al., 2018, p. 62). In further support of the
interplay between genetic and environmental factors in the development of prosociality,
we now discuss the results of our phenotype-to-environment transmission analyses.
Phenotype-to-environment transmission in female twins. In the present work,
we present the first evidence of phenotype-to-environment transmission (PxE) of
prosociality. PxE processes have previously been observed with respect to anxiety (Dolan,
de Kort, van Beijsterveldt, Bartels, & Boomsma, 2014) and cognitive ability (Beam et al.,
2015) in a manner that suggests a “snowball effect” across development, such that initial
phenotypic differences between twins lead to the selection of increasingly tailored
environments that serve to both reinforce and amplify those differences. In contrast to
those findings, we find a significant inverse relationship between initial within-family
differences in Cooperativeness and the environmental exposure that immediately follows,
especially between the ages of 9-10 and 14-15. In other words, this appears to be a
developmental period during which girls are particularly sensitive to the influence of their
co-twin’s prosociality, especially as it pertains to the individual environments they
proceed to inhabit. By extension, it is likely that girls will be the most receptive to (and,
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

37
in the case of the less prosocial twin in each pair, in need of) targeted interventions that
occur during early adolescence.
At least one (school-based) intervention focused on this age group has effectively
promoted prosocial behavior six months later (Capara et al., 2014). However, it remains
to be seen whether an intervention (school-based or family-based) can have positive
effects on prosocial tendencies that persist throughout development. Additionally, it is
not entirely clear why the within-family dynamics that give rise to PxE apply only to
female twins in the present sample. One possibility is that girls (relative to boys) may be
more likely to turn to their same-sex co-twin for social support in the transition to
adolescence and, consequently, become more similar to each other in a variety of
domains. There is evidence to suggest that girls have more positive relationships with
their same-sex siblings than boys have with theirs (Buhrmester, 1992; Tucker, Barber, &
Eccles, 1997) and that higher quality sibling relationships can foster prosocial behavior
during late childhood/early adolescence (Smorti & Ponti, 2018), but additional research is
necessary to directly confirm this intuition.
Concluding Remarks, Limitations, and Future Directions
In sum, the present work makes several novel contributions, most notably: (1)
identification of a common genetic factor that contributes to variance in Cooperativeness
from late childhood to emerging adulthood and (2) evidence of phenotype-to-
environment transmission of Cooperativeness in female twins. However, several
limitations of the classical twin design merit further discussion with respect to the present
findings.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

38
First, the classical twin design assumes that mating occurs at random among
members of the population in question. Although not widely studied, there is initial
evidence that prosocial individuals do seek out comparably prosocial mates (Tognetti,
Berticat, Raymond, & Faurie, 2014), thus violating this specific assumption. However, in
the context of twin modeling, assortative (non-random) mating would lead to increased
similarity in DZ twins relative to MZ twins, which would appear as inflated estimates of
shared environmental (C) effects at the expense of additive genetic (A) effects. Given that
C effects were modest (when detectable at all), it is unlikely that assortative mating
substantively impacted the present findings. Second, the equal environments assumption
(EEA) posits that MZ twins and DZ twins, irrespective of zygosity, are treated similarly
to one another within their twin pairs. As a result of this assumption, any observed
dissimilarity between MZ and DZ twins can be attributed to genetic factors, thus leading
to higher MZ twin correlations and heritability estimates than would otherwise be
expected. Indeed, controlling for environmental similarity between MZ and DZ twins has
been found to reduce heritability estimates of prosocial obligation, but not significantly
so (Felson, 2014).
Despite these contributions and limitations, there remain promising avenues for
future research. As developmental researchers (e.g., Eisenberg & Fabes, 1998; Hay &
Cook, 2007) have argued, prosocial behaviors (at the within-subject level) have the
potential to become more selectively targeted as we age. However, it is important to
acknowledge that these behaviors are, at the between-subjects level, highly situation-
specific throughout the lifespan. For example, as both children and adults, people are
more likely to empathize with those similar to them or offer help when they are the sole
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

39
bystander (e.g., Hamlin, Mahajan, Liberman, & Wynn, 2013; Plötner, Over, Carpenter, &
Tomasello, 2015). Even so, most survey-based measures assess various aspects of a
broad prosocial disposition, with the implicit assumption that those traits and behaviors
are equally likely to occur across all targets and all circumstances.
Thus, it would be beneficial for both phenotypic and behavior genetic research to
account for person-by-situation contingencies (e.g., Mischel, Shoda, & Mendoza-Denton,
2002) that are likely to influence prosocial tendencies. In one such study, seven year-old
twins were given the opportunity to donate money they had just received (by virtue of
participating in a previous study) to UNICEF after viewing videos of children in poverty
(van IJzendoorn, Marinus, Bakermans-Kranenburg, Pannebakker, & Out, 2010).
Although the number of children who spontaneously donated was too small to allow for
genetic modeling, shared environmental factors were found to influence donation
behavior that followed light probing from an experimenter. These findings occur in stark
contrast to many other studies that establish the heritability of prosociality around this
age (Israel et al., 2015), and suggest that its etiology may differ depending on the specific
method of measurement.
An additional benefit of such a study design is that its findings more closely
describe the nature of prosocial behavior that is likely to naturally occur outside the
laboratory. Especially for socially desirable outcome measures, such attempts to increase
ecological validity are crucial. Although there is initial evidence that Cooperativeness can
predict charitable behavior towards an unknown stranger (Bereczkei, Birkas, Kerekes,
2007), increased use of well-controlled observational paradigms can only contribute
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

40
positively to the body of literature on prosocial tendencies, which, in turn, will aid in the
effective promotion of prosocial behavior to the benefit of society as a whole.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

41
References

Adams, M. J., King, J. E., & Weiss, A. (2012). The majority of genetic variation in
orangutan personality and subjective well-being is nonadditive. Behavior
Genetics, 42(4), 675-686.
Ando, J., Suzuki, A., Yamagata, S., Kijima, N., Maekawa, H., Ono, Y., & Jang, K. L.
(2004). Genetic and environmental structure of Cloninger's temperament and
character dimensions. Journal of Personality Disorders, 18(4), 379-393.
Avinun, R., & Knafo, A. (2014). Parenting as a reaction evoked by children’s genotype:
A meta-analysis of children-as-twins studies. Personality and Social Psychology
Review, 18, 87-102.
Baker, L. A., Barton, M., Lozano, D. I., Raine, A., & Fowler, J. H. (2006). The Southern
California Twin Register at the University of Southern California: II. Twin
Research and Human Genetics, 9(6), 933–940.
Baker, L. A., Tuvblad, C., Wang, P., Gomez, K., Bezdjian, S., Niv, S., & Raine, A.
(2013). The Southern California Twin Register at the University of Southern
California: III. Twin Research and Human Genetics, 16(1), 336-343.
Beam, C. R., & Turkheimer, E. (2013). Phenotype–environment correlations in
longitudinal twin models. Development and Psychopathology, 25(1), 7-16.
Beam, C. R., Turkheimer, E., Dickens, W. T., & Davis, D. W. (2015). Twin
differentiation of cognitive ability through phenotype to environment
transmission: The Louisville Twin Study. Behavior Genetics, 45(6), 622-634.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

42
Bereczkei, T., Birkas, B., & Kerekes, Z. (2007). Public charity offer as a proximate factor
of evolved reputation-building strategy: An experimental analysis of a real-life
situation. Evolution and Human Behavior, 28(4), 277-284.
Bergen, S. E., Gardner, C. O., & Kendler, K. S. (2007). Age-related changes in
heritability of behavioral phenotypes over adolescence and young adulthood: a
meta-analysis. Twin Research and Human Genetics, 10(3), 423-433.
Buhrmester, D. (1992). The developmental courses of sibling and peer relationships. In F.
Boer & J. Dunn (Eds.), Children’s sibling relationships: Developmental and
clinical issues (pp. 19-40). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Caprara, G. V., Kanacri, B. P. L., Gerbino, M., Zuffiano, A., Alessandri, G., Vecchio,
G., ... & Bridglall, B. (2014). Positive effects of promoting prosocial behavior in
early adolescence: Evidence from a school-based intervention. International
Journal of Behavioral Development, 38(4), 386-396.
Carlo, G., Crockett, L. J., Randall, B. A., & Roesch, S. C. (2007). A latent growth curve
analysis of prosocial behavior among rural adolescents. Journal of Research on
Adolescence, 17(2), 301-324.
Carlo, G., & Randall, B. A. (2002). The development of a measure of prosocial behaviors
for late adolescents. Journal of Youth and Adolescence, 31(1), 31-44.
Cheng, S. T., & Yim, Y. K. (2008). Age differences in forgiveness: The role of future
time perspective. Psychology and Aging, 23(3), 676-680.
Chiaramello, S., Mesnil, M., Muñoz Sastre, M. T., & Mullet, E. (2008). Dispositional
forgiveness among adolescents. European Journal of Developmental Psychology,
5(3), 326-337.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

43
Cloninger, C. R., & Garcia, D. (2015). The heritability and development of positive affect
and emotionality. In M. Pluess (Ed.), Genetics of psychological well-being: The
role of heritability and genetics in positive psychology (pp. 97-113). Oxford:
Oxford University Press.
Cloninger, C. R., Przybeck, T. R., Svrakic, D. M., & Wetzel, R. D. (1994). The
Temperament and Character Inventory (TCI): A guide to its development and
use (pp. 19-28). St. Louis, MO: Center for Psychobiology of Personality,
Washington University.
Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of
temperament and character. Archives of General Psychiatry, 50(12), 975-990.
Dickens, W. T., & Flynn, J. R. (2001). Heritability estimates versus large environmental
effects: The IQ paradox resolved. Psychological Review, 108(2), 346-369.
DiLalla, L. F., Bersted, K., & John, S. G. (2015). Evidence of reactive gene-environment
correlation in preschoolers’ prosocial play with unfamiliar peers. Developmental
Psychology, 51(10), 1464-1475.
Dolan, C. V., de Kort, J. M., van Beijsterveldt, T. C., Bartels, M., & Boomsma, D. I.
(2014). GE covariance through phenotype to environment transmission: An
assessment in longitudinal twin data and application to childhood anxiety.
Behavior Genetics, 44(3), 240-253.
Dunfield, K. A., & Kuhlmeier, V. A. (2013). Classifying prosocial behavior: Children's
responses to instrumental need, emotional distress, and material desire. Child
Development, 84(5), 1766-1776.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

44
Dunfield, K., Kuhlmeier, V. A., O’Connell, L., & Kelley, E. (2011). Examining the
diversity of prosocial behavior: Helping, sharing, and comforting in infancy.
Infancy, 16(3), 227-247.
Eisenberg, N., & Fabes, R. A. (1998). Prosocial development. In W. Damon & N.
Eisenberg (Eds.), Handbook of child psychology, Vol. 3: Social, emotional, and
personality development (5
th
ed., pp. 701–778). Hoboken, NJ: John Wiley & Sons,
Inc.
Eisenberg, N., Fabes, R. A., & Spinrad, T. L. (2006). Prosocial behavior. In N. Eisenberg
(Ed.), Social, emotional, and personality development. Handbook of child
psychology, Vol. 3: Social, emotional, and personality development (6th ed., pp.
646–718). Hoboken, NJ: John Wiley and Sons, Inc.
Eisenberg, N., Hofer, C., Sulik, M. J., & Liew, J. (2014). The development of prosocial
moral reasoning and a prosocial orientation in young adulthood: Concurrent and
longitudinal correlates. Developmental Psychology, 50(1), 58-70.
Eisenberg, N., & Spinrad, T. L. (2014). Multidimensionality of prosocial behavior:
Rethinking the conceptualization and development of prosocial behavior. In L. M.
Padilla-Walker & G. Carlo (Eds.), Prosocial behavior: A multidimensional
approach (pp. 17-39). Oxford: Oxford University Press.
Eisenberg, N., Spinrad, T. L., & Knafo-Noam, A. (2015). Prosocial development. In In R.
M. Lerner & M. E. Lamb (Eds.), Handbook of child psychology and
developmental science, Vol. 3: Socioemotional processes (7
th
ed., pp. 610-656).
Hoboken: John Wiley & Sons, Inc.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

45
Felson, J. (2014). What can we learn from twin studies? A comprehensive evaluation of
the equal environments assumption. Social Science Research, 43, 184-199.
Fortuna, K. M., & Knafo, A. (2014). Parental and genetic contributions to prosocial
behavior during childhood. In L. M. Padilla-Walker & G. Carlo (Eds.), Prosocial
behavior: A multidimensional approach (pp. 70-89). Oxford: Oxford University
Press.
Garcia, D., Lundström, S., Brändström, S., Råstam, M., Cloninger, C. R., Kerekes, N., ...
& Anckarsäter, H. (2013). Temperament and character in the Child and
Adolescent Twin Study in Sweden (CATSS): Comparison to the general
population, and genetic structure analysis. PLoS ONE, 8(8), e70475.
Gillespie, N. A., Cloninger, C. R., Heath, A. C., & Martin, N. G. (2003). The genetic and
environmental relationship between Cloninger's dimensions of temperament and
character. Personality and Individual Differences, 35(8), 1931-1946.
Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A research note.
Journal of Child Psychology and Psychiatry, 38(5), 581-586.
Gregory, A. M., Light‐Häusermann, J. H., Rijsdijk, F., & Eley, T. C. (2009). Behavioral
genetic analyses of prosocial behavior in adolescents. Developmental
Science, 12(1), 165-174.
Güroğlu, B., van den Bos, W., & Crone, E. A. (2014). Sharing and giving across
adolescence: An experimental study examining the development of prosocial
behavior. Frontiers in Psychology, 5, 291.
Hamlin, J. K., Mahajan, N., Liberman, Z., & Wynn, K. (2013). Not like me = bad: Infants
prefer those who harm dissimilar others. Psychological Science, 24(4), 589-594.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

46
Hay D. F., & Cook, K. V. (2007). The transformation of prosocial behavior from infancy
to childhood. In C. A. Brownell, & C. B. Kopp (Eds.), Socioemotional
development in the toddler years: Transitions and transformation (pp. 100–131).
New York: Guilford Press.
Hur, Y. M., & Rushton, J. P. (2007). Genetic and environmental contributions to
prosocial behaviour in 2-to 9-year-old South Korean twins. Biology Letters, 3(6),
664-666.
Isen, J. D., Baker, L. A., Raine, A., & Bezdjian, S. (2009). Genetic and environmental
influences on the Junior Temperament and Character Inventory in a preadolescent
twin sample. Behavior Genetics, 39(1), 36-47.
Isen, J. D. (2008). Subscale construction in the Junior Temperament and Character
Inventory. Unpublished analyses.
Israel, S., Hasenfratz, L., & Knafo-Noam, A. (2015). The genetics of morality and
prosociality. Current Opinion in Psychology, 6, 55-59.
Keller, M. C., Coventry, W. L., Heath, A. C., & Martin, N. G. (2005). Widespread
evidence for non-additive genetic variation in Cloninger’s and Eysenck’s
personality dimensions using a twin plus sibling design. Behavior Genetics, 35(6),
707.
Knafo, A., & Plomin, R. (2006). Prosocial behavior from early to middle childhood:
Genetic and environmental influences on stability and change. Developmental
Psychology, 42(5), 771-786.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

47
Knafo, A., Zahn-Waxler, C., Van Hulle, C., Robinson, J. L., & Rhee, S. H. (2008). The
developmental origins of a disposition toward empathy: Genetic and
environmental contributions. Emotion, 8(6), 737-752.
Knafo-Noam, A., Uzefovsky, F., Israel, S., Davidov, M., & Zahn-Waxler, C. (2015). The
prosocial personality and its facets: Genetic and environmental architecture of
mother-reported behavior of 7-year-old twins. Frontiers in Psychology, 6, 112.
Knafo-Noam, A., Vertsberger, D., & Israel, S. (2018). Genetic and environmental
contributions to children's prosocial behavior: Brief review and new evidence
from a reanalysis of experimental twin data. Current Opinion in Psychology, 20,
60-65.
Ledermann, T., & Kenny, D. A. (2017). Analyzing dyadic data with multilevel modeling
versus structural equation modeling: A tale of two methods. Journal of Family
Psychology, 31(4), 442-452.
Lester, N., Garcia, D., Lundström, S., Brändström, S., Råstam, M., Kerekes, N., ... &
Anckarsäter, H. (2016). The genetic and environmental structure of the character
sub-scales of the Temperament and Character Inventory in adolescence. Annals of
General Psychiatry, 15, 10.
Lewis, G. J., & Bates, T. C. (2011). A common heritable factor influences prosocial
obligations across multiple domains. Biology Letters, rsbl20101187.
Luby, J. L., Svrakic, D. M., McCallum, K., Przybeck, T. R., & Cloninger, C. R. (1999).
The Junior Temperament and Character Inventory: Preliminary validation of a
child self-report measure. Psychological Reports, 84, 1127-1138.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

48
Luengo Kanacri, B. P., Pastorelli, C., Eisenberg, N., Zuffianò, A., & Caprara, G. V.
(2013). The development of prosociality from adolescence to early adulthood:
The role of effortful control. Journal of Personality, 81(3), 302-312.
Lykken, D. T. (1978). The diagnosis of zygosity in twins. Behavior Genetics, 8, 437–473.
Mischel, W., Shoda, Y., & Mendoza-Denton, R. (2002). Situation-behavior profiles as a
locus of consistency in personality. Current Directions in Psychological Science,
11(2), 50-54.
Nantel-Vivier, A., Kokko, K., Caprara, G. V., Pastorelli, C., Gerbino, M. G., Paciello,
M., ... & Tremblay, R. E. (2009). Prosocial development from childhood to
adolescence: A multi‐informant perspective with Canadian and Italian
longitudinal studies. Journal of Child Psychology and Psychiatry, 50(5), 590-598.
Neale, M. C., Boker, S. M., Xie, G. & Maes, H. (2003). Mx: Statistical modeling.
Richmond, VA: Department of Psychiatry, Medical College of Virginia.
Ozaki, K., Toyoda, H., Iwama, N., Kubo, S., & Ando, J. (2011). Using non-normal SEM
to resolve the ACDE model in the classical twin design. Behavior Genetics, 41(2),
329-339.
Padilla-Walker, L. M., & Carlo, G. (2014). Prosocial behavior: Past, present, and future.
In L. M. Padilla-Walker & G. Carlo (Eds.), Prosocial behavior: A
multidimensional approach (pp. 3-16). Oxford: Oxford University Press.
Penner, L. A., Dovidio, J. F., Piliavin, J. A., & Schroeder, D. A. (2005). Prosocial
behavior: Multilevel perspectives. Annual Review of Psychology, 56, 365-392.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

49
Plomin, R., DeFries, J. C., & Loehlin, J. C. (1977). Genotype-environment interaction
and correlation in the analysis of human behavior. Psychological Bulletin, 84(2),
309-322.
Plötner, M., Over, H., Carpenter, M., & Tomasello, M. (2015). Young children show the
bystander effect in helping situations. Psychological Science, 26(4), 499-506.
Reuter, M., Frenzel, C., Walter, N. T., Markett, S., & Montag, C. (2010). Investigating
the genetic basis of altruism: the role of the COMT Val158Met polymorphism.
Social Cognitive and Affective Neuroscience, 6(5), 662-668.
Scarr, S., & McCartney, K. (1983). How people make their own environments: A theory
of genotype → environment effects. Child Development, 54, 424-435.
Scourfield, J., John, B., Martin, N., & McGuffin, P. (2004). The development of
prosocial behaviour in children and adolescents: A twin study. Journal of Child
Psychology and Psychiatry, 45(5), 927-935.
Scourfield, J., Rice, F., Thapar, A., Harold, G., Martin, N., & McGuffin, P. (2003).
Depressive symptoms in children and adolescents: Changing aetiological
influences with development. Journal of Child Psychiatry and Psychology, 44(7),
968-976.
Singal, J. (2016, April 5). How the false story of Kitty Genovese’s murder went viral.
New York Magazine. Retrieved from https://www.thecut.com/2016/04/how-the-
false-story-of-kitty-genovese-s-murder-went-viral.html
Smorti, M., & Ponti, L. (2018). How does sibling relationship affect children’s prosocial
behaviors and best friend relationship quality?. Journal of Family Issues, 39(8),
2413-2436.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

50
Tognetti, A., Berticat, C., Raymond, M., & Faurie, C. (2014). Assortative mating based
on cooperativeness and generosity. Journal of Evolutionary Biology, 27(5), 975-
981.
Tucker, C. J., Barber, B. L., & Eccles, J. S. (1997). Advice about life plans and personal
problems in late adolescent sibling relationships. Journal of Youth and
Adolescence, 26, 63-76.
Uzefovsky, F., Shalev, I., Israel, S., Edelman, S., Raz, Y., Mankuta, D., ... & Ebstein, R.
P. (2015). Oxytocin receptor and vasopressin receptor 1a genes are respectively
associated with emotional and cognitive empathy. Hormones and Behavior, 67,
60-65.
van IJzendoorn, M. H., Bakermans-Kranenburg, M. J., Pannebakker, F., & Out, D.
(2010). In defence of situational morality: Genetic, dispositional and situational
determinants of children’s donating to charity. Journal of Moral Education, 39(1),
1-20.
Whitelaw, N. C., & Whitelaw, E. (2006). How lifetimes shape epigenotype within and
across generations. Human Molecular Genetics, 15(suppl_2), R131-R137.
Wu, N., Li, Z., & Su, Y. (2012). The association between oxytocin receptor gene
polymorphism (OXTR) and trait empathy. Journal of Affective Disorders, 138(3),
468-472.
Zahn-Waxler, C., Schiro, K., Robinson, J. L., Emde, R. N., Schmitz, S. (2001). Empathy
and prosocial patterns in young MZ and DZ twins: Development, genetic, and
environmental influences. In R.N. Emde, & J. K. Hewitt (Eds.), Infancy to early
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

51
childhood: Genetic and environmental influences on developmental change (pp.
141-162). Oxford: Oxford University Press.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

52
Tables and Figures

Table 1: Demographic information

JTCI TCI
Any
Wave
Ages
9-10
Ages
14-15
Ages
16-18
Ages
19-20
All
Waves

Total
Participants
1535 1155 992 914 1000 383

Total Families 783 602 526 491 555 228

Total Complete
Pairs
752 553 466 423 445 155

Complete Pairs
by Zygosity

Complete MZM
Pairs
162 121 96 87 92 32
Complete MZF
Pairs
162 132 95 92 104 42
Complete DZM
Pairs
118 79 73 65 54 17
Complete DZF
Pairs
119 89 88 79 84 34
Complete DZOS
Pairs
190 132 113 100 111 30
Pairs with
Unknown
Zygosity
1 0 1 0 0 0

Sex/Gender
Male 767 558 481 434 452 155
Female 768 597 511 480 548 228

Race/Ethnicity
Asian 64 53 46 28 40 18
Black 202 155 118 100 121 48
Hispanic 529 426 316 307 325 117
Multiracial 274 200 174 173 183 69
Native American 1 1 0 0 0 0
White 455 320 328 302 327 131
Not Indicated 10 0 10 4 4 0

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

53
Mean Age 9.59 14.71 17.30 19.87

Table 2: Longitudinal Cooperativeness scale scores stratified by sex

Full Sample Males Females
Sex
Differences
Subsequent
Wave
Differences
Mean SD Mean SD Mean SD
Ages 9-10
(N = 1155)
0.79 0.14 0.76 0.14 0.81 0.13
t(1153) =
-5.81**
t(657) =
6.39**
Ages 14-15
(N = 992)
0.76 0.16 0.72 0.17 0.80 0.15
t(990) =
-7.95**
t(703) =
-6.56**
Ages 16-18
(N = 914)
0.80 0.16 0.76 0.16 0.84 0.14
t(912) =
-7.21**
t(657) =
-4.91**
Ages 19-20
(N = 1000)
0.83 0.15 0.80 0.16 0.86 0.14
t(998) =
-6.12**
---

Note: Cooperativeness scale scores can range from 0 to 1. Subsequent wave differences
reflect the pairwise comparison (for males and females combined) between one wave and
the wave immediately following. ** = p < .01.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

54
Table 3: Intraclass and cross-twin/cross-wave correlations

Wave X
Ages 9-10 Ages 14-15 Ages 16-18 Ages 19-20
Wave Y

MZ Males
Ages 9-10 .39** .43** .15 .19
Ages 14-15 .24* .38** .28* .23*
Ages 16-18 .08 .30* .58** .40**
Ages 19-20 .21 .36** .32** .49**

DZ Males
Ages 9-10 .36** .09 .23 .03
Ages 14-15 -.09 .07 -.01 .11
Ages 16-18 .00 .01 .08 -.02
Ages 19-20 .07 -.05 .20 .15

MZ Females
Ages 9-10 .43** .13 .33** .13
Ages 14-15 .08 .50** .40** .30**
Ages 16-18 .20 .45** .44** .33**
Ages 19-20 .10 .44** .38** .48**

DZ Females
Ages 9-10 .46** .20 .20 .21
Ages 14-15 .18 .42** .18 .10
Ages 16-18 -.07 .15 .19 .04
Ages 19-20 .15 .12 .14 .19

DZ Mixed Sex
Ages 9-10 .29** .27* -.03 .07
Ages 14-15 .09 .31** .14 .26*
Ages 16-18 .09 .13 .29** .21
Ages 19-20 .00 .14 .22* .25**

Notes: Intraclass correlations are presented in bold along the diagonals. Cross-twin cross-
wave correlations are presented off the diagonals; Wave X in Twin 1 correlated with
Wave Y in Twin 2 presented above the diagonal; Wave X in Twin 2 correlated with
Wave Y in Twin 1 presented below the diagonal. DZ = dizygotic; MZ = monozygotic. *p
< .05, **p < .01. Performed on square-transformed data
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

55
Table 4: Univariate heritability estimates derived from ACDE, ACE, and ADE models

ACDE Model Using
Non-Normal SEM
Common Effects
Sex-Limitation
ACE Model in Mx
Common Effects
Sex-Limitation
ADE Model in Mx
A + D estimates A estimates A + D estimates
Ages
9-10
Males 0.26 0.20 0.44
Females 0.00 0.01 0.48
Ages
14-15
Males 0.29 0.28 0.38
Females 0.14 0.19 0.53
Ages
16-18
Males 0.58 0.57 0.60
Females 0.36 0.43 0.44
Ages
19-20
Males 0.50 0.46 0.48
Females 0.42 0.46 0.47

Notes: Full model-fitting results are available in Supplemental Tables 10-13. Heritability
estimates are presented for common-effects models because they are the most constrained
models that provide separate estimates for males and females. Performed on square-
transformed data.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

56
Table 5: Longitudinal multi-level structural equation modeling (ML-SEM) results

Overall fit Chi-square difference test
Model AIC BIC RMSEA χ
2
df
Compared
to model
Δχ
2
df p
1: Baseline model -1228.35 -825.53 .05 27.44 16 -- -- -- --
2: C (Eb)
covariances = 0
-1234.90 -892.50 .05 44.90 28 1 17.46 12 .13
3: A common factor
for both sexes, males
≠ females
-1250.39 -1018.76 .04 73.41 50 2 28.51 22 .16
4. A variances = 0 -1257.02 -1065.68 .04 82.77 58 3 9.36 8 .31
5: A common factor
for both sexes, males
= females
-1258.34 -1072.03 .04 83.46 59 4 0.69 1 .41
6: C (Eb) variances
equal across sexes
-1260.60 -1094.44 .04 89.19 63 5 5.73 4 .22
7: Ew variances
stationary
-1257.89 -1121.94 .04 103.91 69 6 14.72 6 .02
8. EWcov = 0 in
males
-1235.38 -1099.42 .05 126.42 69 6 37.23 6 <.01
9. EWcov = 0 in
females
-1228.92 -1092.97 .06 132.87 69 6 43.68 6 <.01
10: E common factor
for both sexes
(separate EMVC &
EFVC estimates)
-1220.46 -1104.65 .06 149.33 73 6 60.14 10 <.01
11. Full phenotype-
to-environment
transmission (PxE),
-1257.29 -1060.91 .04 80.51 57 6 8.68 6 .19
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

57
males ≠ females,
three PxE parameters
per sex
12. PxE estimated in
males only
-1257.60 -1076.33 .04 86.19 60 6 3.00 3 .39
13. PxE estimated in
females only
-1260.37 -1079.10 .04 83.43 60 6 5.76 3 .12
14. PxE estimated in
females only, with
three PxE parameter
estimates constrained
to equality
-1262.68 -1091.48 .04 85.12 62 6 4.07 1 .04

Notes: AIC = Akaike’s information criterion, BIC = Bayesian information criterion, RMSEA = root mean square error of
approximation. χ
2
=difference in chi-square values between compared models, df = degrees of freedom. For ease of interpretation,
inadmissible models are presented with the fit statistics stricken through. Performed on square-transformed, age-regressed data.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

58
Table 6: Phenotype-to-environment transmission pathway parameter estimates

COOP
9-10
à Ew
14-15
COOP
14-15
à Ew
16-18
COOP
16-18
à Ew
19-20

Males Females Males Females Males Females
Model 11 0.18 -0.32* 0.13 -0.06 -0.08 -0.16
Model 12 0.19 -- 0.13 -- -0.08 --
Model 13 -- -0.33* -- -0.07 -- -0.16
Model 14 -- -0.17* -- -0.17* -- -0.17*

Notes: Significant (p < .05) parameter estimates are presented in bold and marked with asterisks.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

59
Figure 1: Multilevel structural equation model with a common genetic factor and phenotype-to-
environment transmission

Notes: Phenotype-to-environment transmission pathways highlighted in red. For ease of
interpretation, covariance paths (between each of the occasion-specific latent variables of each
category—Ab, Eb, Aw, Ew) have been omitted from the diagram. Unless otherwise specified in
model fitting results, covariance paths were freely estimated.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

60
Supplemental Tables

Supplemental Table 1: Wave-by-Wave (J)TCI Completion Rates

One
Wave
Two
Waves
Three
Waves
Four
Waves
Total
Ages 9-10 232 234 306 383 1155
Ages 14-15 37 175 397 383 992
Ages 16-18 4 131 396 383 914
Ages 19-20 15 162 440 383 1000
Total 288 351 513 383 1535

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

61
Supplemental Table 2: Reliability (Cronbach’s alpha) statistics for the Cooperativeness scale and
subscales

(Sub)scale name
(Number of JTCI items, number of
TCI items)
JTCI TCI
Ages 9-10 Ages 14-15 Ages 16-18 Ages 19-20
Cooperativeness (20, 25) .64 .73 .78 .78
Social Acceptance (4, 5) .27 .27 .55 .48
Empathy (3,5) .15 .39 .44 .51
Helpfulness (5, 5) .39 .55 .41 .41
Compassion (4, 5) .38 .61 .72 .68
Pure-Hearted Conscience (4, 5) .20 .31 .39 .32

Note: Cronbach’s alpha values above .70 are generally regarded as acceptable.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

62
Supplemental Table 3: Inter-item agreement statistics: Social acceptance subscale

JTCI Items TCI Items
2 55 61 68 4 12 28 93 123
JTCI
2 --
71.5%
.11
71.6%
.03
68.5%
.11
-- -- -- -- --
55
77.6%
.12
--
77.9%
.06
72.6%
.13
-- -- -- -- --
61
76.6%
.08
77.7%
.03
--
72.1%
.03
-- -- -- -- --
68
73.9%
.12
75.1%
.09
73.8%
.04
-- -- -- -- -- --
TCI
4 -- -- -- -- --
89.3%
.15
80.6%
.06
93.0%
.31
81.4%
.05
12 -- -- -- --
88.0%
.17
--
84.2%
.29
88.9%
.13
81.9%
.18
28 -- -- -- --
77.6%
.06
82.1%
.26
--
81.2%
.08
79.0%
.24
93 -- -- -- --
89.6%
.18
87.5%
.10
79.1%
.11
--
81.8%
.07
123 -- -- -- --
82.1%
.13
85.4%
.31
77.9%
.24
83.6%
.19
--

Note: The top value in each cell indicates percent agreement between the two items; the bottom
value indicates the kappa statistic. Kappa values below .40 are generally regarded as poor. Items
in italics are reverse-scored. For the JTCI items, data for ages 9-10 are listed below the diagonal
and data for ages 14-15 are listed above the diagonal. For the TCI items, data for ages 16-18 are
listed below the diagonal and data for ages 19-20 are listed above the diagonal. The text for each
item can be found in Appendices A and B.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

63
Supplemental Table 4: Inter-item agreement statistics: Empathy subscale

JTCI Items TCI Items
28 72 87 18 41 74 89 101
JTCI
28 --
67.3%
.13
63.5%
.19
-- -- -- -- --
72
49.0%
-.02
--
62.2%
.18
-- -- -- -- --
87
46.4%
.05
55.4%
.11
-- -- -- -- -- --
TCI
18 -- -- -- --
79.4%
.25
80.2%
.32
77.9%
.29
60.7%
.04
41 -- -- --
73.4%
.13
--
79.3%
.26
78.8%
.30
59.2%
.01
74 -- -- --
75.3%
.27
72.0%
.14
--
82.2%
.43
58.5%
.00
89 -- -- --
74.3%
.25
71.5%
.13
76.8%
.34
--
56.4%
-.03
101 -- -- --
59.2%
.04
59.7%
.05
57.5%
.03
56.7%
.01
--

Note: The top value in each cell indicates percent agreement between the two items; the bottom
value indicates the kappa statistic. Kappa values below .40 are generally regarded as poor. Items
in italics are reverse-scored. For the JTCI items, data for ages 9-10 are listed below the diagonal
and data for ages 14-15 are listed above the diagonal. For the TCI items, data for ages 16-18 are
listed below the diagonal and data for ages 19-20 are listed above the diagonal. The text for each
item can be found in Appendices A and B.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

64
Supplemental Table 5: Inter-item agreement statistics: Helpfulness subscale

JTCI Items TCI Items
10 34 49 76 91 7 27 50 84 95
JTCI
10 --
84.4%
.36
74.9%
.20
78.2%
.19
82.9%
.12
-- -- -- -- --
34
86.6%
.24
--
76.9%
.26
80.2%
.26
83.9%
.17
-- -- -- -- --
49
80.9%
.22
86.0%
.21
--
71.9%
.15
74.1%
.06
-- -- -- -- --
76
82.7%
.12
90.5%
.17
82.6%
.10
--
80.3%
.13
-- -- -- -- --
91
75.0%
.07
80.6%
.07
72.4%
-.03
78.6%
.05
-- -- -- -- -- --
TCI
7 -- -- -- -- -- --
70.8%
.07
88.9%
.36
76.3%
.01
83.2%
.11
27 -- -- -- -- --
72.2%
.17
--
68.7%
.04
71.7%
.21
72.9%
.18
50 -- -- -- -- --
80.7%
.15
69.5%
.07
--
73.7%
-.04
80.4%
.04
84 -- -- -- -- --
70.3%
.02
67.6%
.14
70.9%
02
--
78.3%
.17
95 -- -- -- -- --
76.4%
.09
71.7
.19
76.6
.08
74.8
.21
--

Note: The top value in each cell indicates percent agreement between the two items; the bottom
value indicates the kappa statistic. Kappa values below .40 are generally regarded as poor. Items
in italics are reverse-scored. For the JTCI items, data for ages 9-10 are listed below the diagonal
and data for ages 14-15 are listed above the diagonal. For the TCI items, data for ages 16-18 are
listed below the diagonal and data for ages 19-20 are listed above the diagonal. The text for each
item can be found in Appendices A and B.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

65
Supplemental Table 6: Inter-item agreement statistics: Compassion subscale

JTCI Items TCI Items
5 39 43 97 5 33 67 80 118
JTCI
5 --
61.7%
.22
63.3%
.25
58.0%
.18
-- -- -- -- --
39
56.7%
.02
--
67.0%
.31
63.7%
.30
-- -- -- -- --
43
62.9%
.11
71.8%
.18
--
62.7%
.31
-- -- -- -- --
97
60.2%
.10
65.5%
.12
76.0%
.30
-- -- -- -- -- --
TCI
5 -- -- -- -- --
89.8%
.62
76.8%
.33
86.6%
.42
81.8%
.14
33 -- -- -- --
82.8%
.52
--
76.4%
.33
84.6%
.35
80.8%
.13
67 -- -- -- --
76.5%
.38
75.9%
.40
--
71.9%
.14
74.6%
.20
80 -- -- -- --
82.0%
.38
79.2%
.35
72.5%
.22
--
86.2%
.20
118 -- -- -- --
78.9%
.24
74.6%
.18
74.1%
.23
82.9%
.21
--

Note: The top value in each cell indicates percent agreement between the two items; the bottom
value indicates the kappa statistic. Kappa values below .40 are generally regarded as poor. Items
in italics are reverse-scored. For the JTCI items, data for ages 9-10 are listed below the diagonal
and data for ages 14-15 are listed above the diagonal. For the TCI items, data for ages 16-18 are
listed below the diagonal and data for ages 19-20 are listed above the diagonal. The text for each
item can be found in Appendices A and B.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

66
Supplemental Table 7: Inter-item agreement statistics: Pure-hearted conscience subscale

JTCI Items TCI Items
15 21 80 94 13 40 75 88 102
JTCI
15 --
77.5%
.04
79.4%
.05
84.3%
.09
-- -- -- -- --
21
79.9%
.03
--
71.1%
.12
75.1%
.18
-- -- -- -- --
80
92.4%
.04
78.2%
.07
--
72.8%
.05
-- -- -- -- --
94
74.2%
.04
67.4%
.08
71.9%
.03
-- -- -- -- -- --
TCI
13 -- -- -- -- --
69.0%
.06
74.8%
.16
70.9%
.18
80.2%
.11
40 -- -- -- --
66.2%
.07
--
61.9%
-.01
59.4%
.03
74.6%
.24
75 -- -- -- --
75.0%
.14
63.9%
.11
--
65.6%
.13
68.8%
-.04
88 -- -- -- --
61.8%
.06
58.4%
.09
63.8%
.18
--
64.7%
.01
102 -- -- -- --
80.8%
.09
69.9%
.19
71.4%
.08
59.9%
.03
--

Note: The top value in each cell indicates percent agreement between the two items; the bottom
value indicates the kappa statistic. Kappa values below .40 are generally regarded as poor. Items
in italics are reverse-scored. For the JTCI items, data for ages 9-10 are listed below the diagonal
and data for ages 14-15 are listed above the diagonal. For the TCI items, data for ages 16-18 are
listed below the diagonal and data for ages 19-20 are listed above the diagonal. The text for each
item can be found in Appendices A and B.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

67
Supplemental Table 8: Summary of inter-item agreement statistics across subscales and waves

JTCI TCI Average
Across Waves Ages 9-10 Ages 14-15 Ages 16-18 Ages 19-20
Social Acceptance
75.8%
.08
72.4%
.08
83.3%
.17
84.1%
.16
80.1%
.13
Empathy
50.3%
.05
64.3%
.17
67.6%
.14
71.3%
.19
66.7%
.15
Helpfulness
81.6%
.12
78.8%
.19
73.1%
.11
76.5%
.11
77.5%
.13
Compassion
65.5%
.14
62.8%
.26
77.9%
.31
81.0%
.29
73.7%
.26
Pure-Hearted
Conscience
77.3%
.05
76.7%
.09
67.1%
.10
69.0%
.09
71.4%
.09
Average Across
Subscales
73.5%
.10
72.6%
.16
73.8%
.17
76.5%
.17
74.0%
.15

Note: The top value in each cell indicates percent agreement; the bottom value indicates the
kappa statistic. Kappa values below .40 are generally regarded as poor. The values in each cell
represent the average of the item-by-item values for that subscale and wave (see Supplemental
Tables 3-7). For example, 75.8% (social acceptance; ages 9-10) is the average of the six percent
agreement statistics provided below the diagonal in the top half of Supplemental Table 3.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

68
Supplemental Table 9: Longitudinal stability of Cooperativeness

Ages 9-10 Ages 14-15 Ages 16-18 Ages 19-20
Ages 9-10 --- .23** .27** .24**
Ages 14-15 .38** --- .42** .40**
Ages 16-18 .21** .42** --- .58**
Ages 19-20 .24** .28** .58** ---

Note: Cross-wave correlations for male twins are presented below the diagonal; data for female
twins are presented above the diagonal. * = p < .05; ** = p < .01.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

69

Supplemental Table 10: Univariate ACDE model-fitting results and parameter estimates for male twins

Components of variance estimates Goodness-of-fit indices
a
2
(95% CI) c
2
(95% CI) d
2
(95% CI) e
2
(95% CI) H
2
(A+D) AIC BIC RMSEA
Ages 9-10
Model 1
nn C & E
0.26
(-0.01-0.53)
0.25*
(0.03-0.47)
0.00
(0.00-0.00)
0.49*
(0.39-0.59)
0.26 -6.89 -26.68 0.00
Model 2
nn D & E
0.00
(0.00-0.00)
0.34
(0.18-0.50)
0.17
(-0.03-0.36)
0.50
(0.39-0.60)
0.17 -5.14 -21.63 0.00
Model 3
nn C, D, E
0.24
(-0.02-0.51)
0.26
(0.04-0.48)
0.00
(0.00-0.00)
0.50
(0.40-0.60)
0.24 -5.16 -21.65 0.00
Ages 14-15
Model 1
nn C & E
0.00
(0.00-0.00)
0.04
(-0.18-0.25)
0.29*
(0.02-0.57)
0.67*
(0.56-0.79)
0.29 -3.90 -22.68 0.05
Model 2
nn D & E
0.00
(0.00-0.00)
0.01
(-0.21-0.23)
0.36
(0.06-0.66)
0.64
(0.51-0.77)
0.36 -3.10 -18.74 0.05
Model 3
nn C, D, E
0.00
(0.00-0.00)
0.02
(-0.25-0.28)
0.35
(0.00-0.69)
0.64
(0.51-0.77)
0.35 -3.10 -18.75 0.05
Ages 16-18
Model 1
nn C & E
0.00
(0.00-0.00)
0.24
(0.08-0.41)
0.29
(0.10-0.48)
0.47
(0.37-0.57)
0.29 0.97 -17.18 0.09
Model 2
nn D & E
0.00
(0.00-0.00)
0.00
(0.00-0.00)
0.58*
(0.46-0.70)
0.42*
(0.32-0.51)
0.58 -6.28 -21.40 0.00
Model 3
nn C, D, E
0.00
(0.00-0.00)
0.00
(0.00-0.00)
0.58*
(0.46-0.70)
0.42*
(0.32-0.51)
0.58 -6.28 -21.40 0.00
Ages 19-20
Model 1
nn C & E
0.00
(0.00-0.00)
0.00
(0.00-0.00)
0.50*
(0.38-0.61)
0.51*
(0.41-0.60)
0.50 -2.94 -20.84 0.06
Model 2
nn D & E
0.21
(-0.17-0.58)
0.00
(0.00-0.00)
0.30
(-0.08-0.67)
0.50
(0.41-0.59)
0.51 -1.08 -15.99 0.07
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

70
Model 3
nn C, D, E
0.21
(-0.17-0.58)
0.00
(0.00-0.00)
0.30
(-0.08-0.67)
0.50
(0.41-0.59)
0.51 -1.08 -15.99 0.07

Notes: nn = non-normally distributed. a
2
= additive genetic influences, d
2
= non-additive (dominant) genetic influences, c
2
= shared
environmental influences, e
2
= non-shared environmental influences, H
2
= broad-sense heritability. AIC = Akaike’s information criterion,
BIC = Bayesian information criterion, RMSEA = root mean square error of approximation. The best-fitting model for each wave is
highlighted in bold. Best-fitting models were those with the lowest RMSEA values and most negative AIC values. Significant parameter
estimates for each best-fitting model are marked with an asterisk.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

71
Supplemental Table 11: Univariate ACDE model-fitting results and parameter estimates for female twins

Components of variance estimates Goodness-of-fit indices
a
2
(95% CI) c
2
(95% CI) d
2
(95% CI) e
2
(95% CI) G (A+D) AIC BIC RMSEA
Ages 9-10
Model 1
nn C & E
0.05
(-0.18-0.27)
0.40
(0.21-0.60)
0.00
(0.00-0.00)
0.55
(0.47-0.63)
0.05 -2.69 -23.08 0.05
Model 2
nn D & E
0.00
(0.00-0.00)
0.43
(0.35-0.52)
0.00
(0.00-0.00)
0.57
(0.50-0.63)
0.00 -4.90 -21.89 0.01
Model 3
nn C, D, E
0.00
(-0.04-0.04)
0.45*
(0.36-0.55)
0.00
(0.00-0.00)
0.55*
(0.48-0.61)
0.00 -6.46 -23.46 0.00
Ages 14-15
Model 1
nn C & E
0.25
(0.05-0.44)
0.30
(0.14-0.47)
0.00
(0.00-0.00)
0.45
(0.36-0.54)
0.25 -4.99 -24.25 0.03
Model 2
nn D & E
0.00
(0.00-0.00)
0.39*
(0.25-0.53)
0.14
(-0.05-0.33)
0.47*
(0.36-0.58)
0.14 -7.93 -23.98 0.00
Model 3
nn C, D, E
0.26
(0.06-0.46)
0.29
(0.11-0.46)
0.00
(0.00-0.00)
0.45
(0.36-0.54)
0.26 -7.82 -23.87 0.00
Ages 16-18
Model 1
nn C & E
0.22
(0.07-0.36)
0.15
(-0.01-0.31)
0.00
(0.00-0.00)
0.64
(0.53-0.75)
0.22 31.21 12.36 0.19
Model 2
nn D & E
0.03
(-0.37-0.43)
0.08
(-0.25-0.40)
0.33*
(0.14-0.53)
0.56*
(0.43-0.69)
0.36 -7.64 -23.35 0.00
Model 3
nn C, D, E
0.06
(-0.18-0.31)
0.07
(-0.20-0.34)
0.31
(0.08-0.54)
0.56
(0.43-0.69)
0.37 -7.64 -23.34 0.00
Ages 19-20
Model 1
nn C & E
0.00
(0.00-0.00)
0.31
(0.17-0.45)
0.09
(-0.03-0.20)
0.61
(0.48-0.73)
0.09 -5.14 -24.56 0.03
Model 2
nn D & E
0.00
(0.00-0.00)
0.07
(-0.04-0.17)
0.42*
(0.23-0.60)
0.52*
(0.38-0.66)
0.42 -5.82 -22.00 0.00
Model 3
nn C, D, E
0.00
(0.00-0.00)
0.09
(-0.07-0.25)
0.39
(0.16-0.62)
0.53
(0.39-0.66)
0.39 9.87 -6.31 0.13
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

72
Notes: nn = non-normally distributed. a
2
= additive genetic influences, d
2
= non-additive (dominant) genetic influences, c
2
= shared
environmental influences, e
2
= non-shared environmental influences, H
2
= broad-sense heritability. AIC = Akaike’s information criterion,
BIC = Bayesian information criterion, RMSEA = root mean square error of approximation. The best-fitting model for each wave is
highlighted in bold. Best-fitting models were those with the lowest RMSEA values and most negative AIC values. Significant parameter
estimates for each best-fitting model are marked with an asterisk.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

73
Supplemental Table 12: Univariate ACE sex-limitation model fitting results and parameter estimates

Overall fit
Chi-square
difference test
Parameter estimates
(95% CI)
-2LL df AIC BIC χ
2
df p Δχ
2
df p A C E
Ages 9-10
1. Saturated
model
-540.78 1130 -2800.78 -3878.98 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
-535.00 1138 -2811.00 -3901.63 5.79 8 .67 -- -- --
M: 0.19
(0.00-0.47)
F: 0.01
(0.00-0.33)
M: 0.22
(0.02-0.44)
F: 0.45
(0.15-0.55)
M: 0.45
(0.15-0.55)
F: 0.55
(0.44-0.66)
2b. Full sex-
limitation
model (rcOS
free)
-535.09 1138 -2811.09 -3901.68 5.70 8 .68 -- -- --
M: 0.14
(0.00-0.53)
F: 0.00
(0.00-0.33)
M: 0.26
(0.00-0.48)
F: 0.45
(0.15-0.55)
M: 0.59
(0.46-0.74)
F: 0.55
(0.44-0.66)
3. Common
effects model
-535.00 1139 -2813.00 -3904.83 5.79 9 .76 0.09 1 .76
M: 0.20
(0.00-0.47)
F: 0.01
(0.00-0.32)
M: 0.22
(0.02-0.44)
F: 0.45
(0.15-0.55)
M: 0.58
(0.46-0.74)
F: 0.54
(0.44-0.66)
4. Scalar
model (ACE)
-531.38 1142 -2815.38 -3912.60 9.41 12 .67 3.62 3 .31
0.13
(0.00-0.40)
0.29
(0.07-0.44)
0.58
(0.49-0.68)
4a. Scalar
model (AE)
-524.91 1143 2810.92 -3912.56 15.87 13 .26 6.46 1 .01
0.46
(0.38-0.54)
--
0.54
(0.46-0.62)
4b. Scalar
model (CE)
-530.45 1143 -2816.45 -3915.32 10.34 13 .67 0.93 1 .33 --
0.38
(0.31-0.45)
0.62
(0.55-0.69)
4c. Scalar
model (E)
-445.77 1144 -2733.77 -3876.18 95.02 14 <.01 85.61 2 <.01 -- -- 1.00
5. Null model -502.99 1143 -2788.99 -3901.60 37.79 13 <.01 28.38 1 <.01
0.16
(0.00-0.43)
0.27
(0.06-0.44)
0.56
(0.47-0.67)
Ages 14-15
1. Saturated
model
-213.83 967 -2147.83 -3129.71 -- -- -- -- -- -- -- -- --
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

74
2a. Full sex-
limitation
model (rgOS
free)
-209.83 975 -2159.83 -3152.72 3.99 8 .91 -- -- --
M: 0.28
(0.00-0.49)
F: 0.19
(0.00-0.58)
M: 0.07
(0.00-0.36)
F: 0.31
(0.00-0.54)
M: 0.65
(0.50-0.82)
F: 0.50
(0.39-0.64)
2b. Full sex-
limitation
model (rcOS
free)
-210.45 975 -2160.45 -3153.03 3.37 8 .86 -- -- --
M: 0.35
(0.35-0.35)
F: 0.16
(0.16-0.16)
M: 0.00
(0.00-0.00)
F: 0.33
(0.33-0.33)
M: 0.64
(0.64-0.64)
F: 0.51
(0.51-0.51)
3. Common
effects model
-209.83 976 -2161.83 -3155.85 3.99 9 .91 0.62 1 0.43
M: 0.28
(0.00-0.49)
F: 0.19
(0.00-0.58)
M: 0.07
(0.00-0.36)
F: 0.31
(0.00-0.54)
M: 0.65
(0.50-0.82)
F: 0.50
(0.39-0.64)
4. Scalar
model (ACE)
-202.97 979 -2160.97 -3161.79 10.85 12 .54 6.86 3 0.08
0.29
(0.00-0.52)
0.14
(0.00-0.37)
0.57
(0.47-0.69)
4a. Scalar
model (AE)
-201.61 980 -2161.61 -3164.24 12.22 13 .51 1.37 1 0.24
0.45
(0.35-0.54)
--
0.55
(0.46-0.65)
4b. Scalar
model (CE)
-199.49 980 -2159.49 -3163.18 14.34 13 .35 3.49 1 0.06 --
0.34
(0.26-0.42)
0.66
(0.58-0.74)
4c. Scalar
model (E)
-139.39 981 -2101.39 -3136.25 74.44 14 <.01 63.59 2 <.01 -- -- 1.00
5. Null model -152.55 980 -2112.55 -3139.71 61.27 13 <.01 50.42 1 <.01
0.32
(0.03-0.55)
0.15
(0.00-0.37)
0.53
(0.44-0.64)
Ages 16-18
1. Saturated
model
-232.49 893 -2018.49 -2879.30 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
-224.36 901 -2026.36 -2899.99 8.13 8 .42 -- -- --
M: 0.57
(0.33-0.69)
F: 0.43
(0.31-0.59)
M: 0.00
(0.00-0.17)
F: 0.00
(0.00-0.32)
M: 0.43
(0.31-0.59)
F: 0.57
(0.44-0.72)
2b. Full sex-
limitation
model (rcOS
free)
-224.36 901 -2026.36 -2899.99 8.13 8 .42 -- -- --
M: 0.57
(0.32-0.69)
F: 0.43
(0.31-0.61)
M: 0.00
(0.00-0.17)
F: 0.00
(0.00-0.40)
M: 0.43
(0.31-0.61)
F: 0.57
(0.44-0.76)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

75
3. Common
effects model
(ACE)
-224.36 902 -2028.36 -2903.08 8.13 9 .52 0.00 1 1.00
M: 0.57
(0.33-0.69)
F: 0.43
(0.06-0.56)
M: 0.00
(0.00-0.17)
F: 0.00
(0.00-0.32)
M: 0.43
(0.31-0.59)
F: 0.57
(0.44-0.72)
4. Scalar
model (ACE)
-219.08 905 -2029.08 -2909.73 13.40 12 .34 5.27 3 0.15
0.49
(0.29-0.58)
0.00
(0.00-0.15)
0.51
(0.41-0.61)
4a. Scalar
model (AE)
-219.08 906 -2031.08 -2912.83 13.40 13 .42 0.00 1 1.00
0.49
(0.39-0.59)
--
0.51
(0.41-0.61)
4b. Scalar
model (CE)
-204.81 906 -2016.81 -2905.69 27.68 13 .01 14.28 1 <.01 --
0.33
(0.24-0.41)
0.67
(0.59-0.76)
4c. Scalar
model (E)
-157.40 907 -1971.40 -2885.08 75.09 14 <.01 61.69 2 <.01 -- -- 1.00
5. Null model -171.66 906 -1983.66 -2889.11 60.83 13 <.01 47.43 1 <.01
0.52
(0.34-0.61)
0.00
(0.00-0.13)
0.48
(0.39-0.59)
Ages 19-20
1. Saturated
model
-269.97 979 -2227.97 -3224.58 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
-263.38 987 -2237.38 -3246.53 6.58 8 .58 -- -- --
M: 0.46
(0.00-0.60)
F: 0.46
(0.09-0.58)
M: 0.01
(0.00-0.48)
F: 0.00
(0.00-0.30)
M: 0.53
(0.40-0.68)
F: 0.53
(0.42-0.68)
2b. Full sex-
limitation
model (rcOS
free)
-263.38 987 -2237.38 -3246.53 6.58 8 .58 -- -- --
M: 0.46
(0.00-0.60)
F: 0.46
(0.10-0.58)
M: 0.01
(0.00-0.50)
F: 0.00
(0.00-0.30)
M: 0.53
(0.40-0.69)
F: 0.54
(0.42-0.68)
3. Common
effects model
(ACE)
-263.38 988 -2239.38 -3249.69 6.59 9 .68 0.01 1 .92
M: 0.46
(0.00-0.60)
F: 0.46
(0.09-0.58)
M: 0.01
(0.00-0.48)
F: 0.00
(0.00-0.30)
M: 0.53
(0.40-0.68)
F: 0.54
(0.42-0.68)
4. Scalar
model (ACE)
-258.74 991 -2240.74 -3256.84 11.22 12 .51 4.63 3 .20
0.47
(0.20-0.55)
0.00
(0.00-0.21)
0.53
(0.45-0.63)
4a. Scalar
model (AE)
-258.74 992 -2242.74 -3259.99 11.22 13 .59 0.00 1 1.00
0.47
(0.37-0.55)
--
0.53
(0.45-0.63)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

76
4b. Scalar
model (CE)
-248.12 992 -2232.12 -3254.68 21.84 13 .06 10.62 1 <.01 --
0.34
(0.26-0.42)
0.66
(0.58-0.74)
4c. Scalar
model (E)
-192.34 993 -2178.34 -3229.95 77.62 14 <.01 66.40 2 <.01 -- -- 1.00
5. Null model -229.85 992 -2213.84 -3245.54 40.12 13 <.01 28.90 1 <.01
0.49
(0.26-0.57)
0.00
(0.00-0.18)
0.51
(0.43-0.61)

Notes: -2LL = -2*log-likelihood, df = degrees of freedom, AIC = Akaike’s information criterion, BIC=Bayesian information criterion,
χ
2
=difference in chi-square values between compared models, A = additive genetic influences, C = shared environmental influences, E =
non-shared environmental influences, CI = confidence interval, rgOS = genetic correlation for opposite-sex twins, rcOS = shared
environmental correlation for opposite-sex twins. Best-fitting models for each wave (as identified by lowest AIC/BIC values and highest p-
values in the chi-square difference test) are highlighted in bold.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

77
Supplemental Table 13: Univariate ADE sex-limitation model fitting results and parameter estimates

Overall fit
Chi-square
difference test
Parameter estimates
(95% CI)
-2LL df AIC BIC χ
2
df p Δχ
2
df p A D E
Ages 9-10
1. Saturated
model
-540.78 1130 -2800.78 -3878.98 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model
-526.26 1137 -2800.26 -3894.07 14.52 7 .04 -- -- --
M: 0.44
(0.00-0.55)
F: 0.47
(0.17-0.58)
M: 0.00
(0.00-0.45)
F: 0.00
(0.00-0.30)
M: 0.56
(0.45-0.71)
F: 0.53
(0.42-0.65)
3. Common
effects model
-525.76 1139 -2803.76 -3900.21 15.03 9 .09 0.51 2 .77
M: 0.44
(0.19-0.56)
F: 0.48
(0.26-0.58)
M: 0.00
(0.00-0.27)
F: 0.00
(0.00-0.22)
M: 0.56
(0.44-0.69)
F: 0.52
(0.42-0.64)
4. Scalar
model (ADE)
-524.92 1142 -2808.92 -3909.37 15.87 12 .20 0.84 3 .84
0.46
(0.30-0.54)
0.00
(0.00-0.16)
0.54
(0.46-0.62)
4a. Scalar
model (AE)
-524.92 1143 -2810.92 -3912.56 15.87 13 .26 0.00 1 1.00
0.46
(0.38-0.54)
--
0.54
(0.46-0.62)
4b. Scalar
model (E)
-445.77 1144 -2733.77 -3876.18 95.02 14 <.01 79.15 2 <.01 -- -- 1.00
5. Null model -497.03 1143 -2783.02 -3898.62 43.76 13 <.01 27.89 1 <.01
0.48
(0.31-0.55)
0.00
(0.00-0.16)
0.52
(0.45-0.61)
Ages 14-15
1. Saturated
model
-213.83 967 -2147.83 -3129.71 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model
-208.14 974 -2156.13 -3148.75 5.69 7 .58 -- -- --
M: 0.00
(0.00-0.49)
F: 0.52
(0.13-0.63)
M: 0.37
(0.00-0.52)
F: 0.00
(0.00-0.39)
M: 0.63
(0.48-0.81)
F: 0.48
(0.37-0.61)
3. Common
effects model
-206.79 976 -2158.79 -3154.32 7.04 9 .63 1.35 2 .51
M: 0.38
(0.05-0.51)
M: 0.00
(0.00-0.37)
M: 0.62
(0.49-0.78)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

78
F: 0.53
(0.24-0.63)
F: 0.00
(0.00-0.29)
F: 0.47
(0.37-0.60)
4. Scalar
model (ADE)
-201.61 979 -2159.61 -3161.11 12.22 12 .43 5.18 3 .16
0.45
(0.18-0.54)
0.00
(0.00-0.28)
0.55
(0.46-0.65)
4a. Scalar
model (AE)
-201.61 980 -2161.61 -3164.24 12.22 13 .51 0.00 1 1.00
0.45
(0.35-0.54)
--
0.55
(0.46-0.65)
4b. Scalar
model (E)
-139.39 981 -2101.39 -3136.25 74.44 14 <.01 62.22 1 <.01 -- -- 1.00
5. Null model -150.95 980 -2110.95 -3138.91 62.88 13 <.01 50.66 1 <.01
0.49
(0.23-0.57)
0.00
(0.00-0.27)
0.51
(0.43-0.60)
Ages 16-18
1. Saturated
model
-232.49 893 -2018.49 -2879.30 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model
-228.19 900 -2028.19 -2898.81 4.30 7 .75 -- -- --
M: 0.00
(0.00-0.61)
F: 0.39
(0.00-0.55)
M: 0.60
(0.00-0.71)
F: 0.03
(0.00-0.55)
M: 0.40
(0.29-0.55)
F: 0.58
(0.44-0.74)
3. Common
effects model
-225.88 902 -2029.88 -2903.85 6.61 9 .68 2.31 2 .32
M: 0.23
(0.00-0.67)
F: 0.33
(0.00-0.55)
M: 0.37
(0.00-0.69)
F: 0.11
(0.00-0.55)
M: 0.40
(0.29-0.56)
F: 0.56
(0.43-0.71)
4. Scalar
model (ADE)
-219.95 905 -2029.95 -2910.17 12.53 12 .40 5.92 3 .12
0.28
(0.00-0.58)
0.24
(0.00-0.60)
0.49
(0.39-0.60)
4a. Scalar
model (AE)
-219.08 906 -2031.08 -2912.83 13.40 13 .42 0.87 1 .35
0.49
(0.39-0.59)
--
0.51
(0.41-0.61
4b. Scalar
model (E)
-157.40 907 -1971.40 -2885.08 75.09 14 <.01 62.56 1 <.01 -- -- 1.00
5. Null model -173.16 906 -1985.16 -2889.86 59.33 13 <.01 46.80 1 <.01
0.23
(0.00-0.59)
0.31
(0.00-0.62)
0.46
(0.37-0.57)
Ages 19-20
1. Saturated
model
-269.97 979 -2227.97 -3224.58 -- -- -- -- -- -- -- -- --
2a. Full sex- -263.69 986 -2235.69 -3243.53 6.27 7 .51 M: 0.47 M: 0.00 M: 0.53
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

79
limitation
model
(0.00-0.60)
F: 0.23
(0.00-0.58)
(0.00-0.59)
F: 0.25
(0.00-0.59)
(0.40-0.68)
F: 0.52
(0.40-0.68)
3. Common
effects model
-263.54 988 -2239.54 -3249.77 6.42 9 .70 0.15 2 .93
M: 0.45
(0.00-0.60)
F: 0.35
(0.00-0.58)
M: 0.03
(0.00-0.56)
F: 0.12
(0.00-0.56)
M: 0.52
(0.40-0.67)
F: 0.53
(0.40-0.68)
4. Scalar
model (ADE)
-258.84 991 -2240.84 -3256.89 11.13 12 .52 4.71 3 .19
0.39
(0.00-0.55)
0.08
(0.00-0.54)
0.53
(0.44-0.63)
4a. Scalar
model (AE)
-258.74 992 -2242.74 -3259.99 11.22 13 .59 0.09 1 .76
0.47
(0.37-0.55)
--
0.53
(0.45-0.63)
4b. Scalar
model (E)
-192.34 993 -2178.34 -3229.95 77.62 14 <.01 66.49 1 <.01 -- -- 1.00
5. Null model -230.21 992 -2214.21 -3245.73 39.76 13 <.01 28.63 1 <.01
0.23
(0.00-0.59)
0.31
(0.00-0.62)
0.46
(0.37-0.56)

Notes: -2LL = -2*log-likelihood, df = degrees of freedom, AIC = Akaike’s information criterion, BIC=Bayesian information criterion,
χ
2
=difference in chi-square values between compared models, A = additive genetic influences, D = non-additive (dominant) genetic
influences, E = non-shared environmental influences, CI = confidence interval. Best-fitting models for each wave (as identified by lowest
AIC/BIC values and highest p-values in the chi-square difference test) are highlighted in bold.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

80
Appendix A: JTCI Cooperativeness Subscales and Items

Social Acceptance vs. Social Intolerance

• I usually like other kids even when they are very different from me. (2)
• I can learn a lot from other children. (55)
• Other kids have to learn how to do things my way. (61)
• I don't make fun of kids who are different than I am. (68)

Empathy vs. Social Disinterest

• I understand other kids' feelings. (28)
• I don't understand why people have the feelings that they do. (72)
• I don't like to be bothered by other children's problems. (87)

Helpfulness vs. Unhelpfulness

• I usually help to make things turn out all right so that everyone is happy. (10)
• I really like to help others. (34)
• I like to share what I have learned with other kids. (49)
• Everybody wins when people help each other. (76)
• It is usually bad for me if I help other kids. (91)

Compassion vs. Vengefulness

• I usually try to get even when someone hurts me. (5)
• It bothers me for a long time when I treat other kids badly, even if they've been
mean to me. (39)
• I am usually nice to other kids even if they've been mean to me in the past. (43)
• I enjoy helping others even if they treat me badly. (97)

Pure-Hearted Conscience vs. Self-Serving Advantage

• I would do nasty things to become popular. (15)
• You don't have to cheat or lie to get what you want or to get ahead. (21)
• Everyone should be treated with kindness and respect no matter how unimportant
or bad they are. (80)
• Being fair and telling the truth often do not matter to me. (94)

Notes: Reverse-coded items are italicized. Item numbers (referenced in Supplemental
Tables 3-7) appear in parentheses.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

81
Appendix B: TCI Cooperativeness Subscales and Items

Social Acceptance vs. Social Intolerance

• I can usually accept other people as they are, even when they are very different
from me. (4)
• I generally don't like people who have different ideas from me. (12)
• I have no patience with people who don't accept my views. (28)
• I usually respect the opinions of others. (93)
• People involved with me have to learn how to do things my way. (123)

Empathy vs. Social Disinterest

• I often consider another person's feelings as much as my own. (18)
• People will usually tell me how they feel. (41)
• I usually try to imagine myself in other people's shoes, so I can really understand
them. (74)
• I often try to put aside my own judgments so that I can better understand what
other people are experiencing. (89)
• I wish other people didn't talk as much as they do. (101)

Helpfulness vs. Unhelpfulness

• I like to help find a solution to problems so that everyone comes out ahead. (7)
• I usually try to get just what I want for myself because it is not possible to satisfy
everyone anyway. (27)
• I like to share what I have learned with other people. (50)
• Members of a team rarely get their fair share. (84)
• It is usually foolish to promote the success of other people. (95)

Compassion vs. Vengefulness

• I enjoy getting revenge on people who hurt me. (5)
• When someone hurts me in any way, I usually try to get even. (33)
• I would rather be kind than to get revenge when someone hurts me. (67)
• I like to imagine my enemies suffering. (80)
• I hate to see anyone suffer. (118)

Pure-Hearted Conscience vs. Self-Serving Advantage

• I would do almost anything legal in order to become rich and famous, even if I
would lose the trust of many old friends. (13)
• I cannot have any peace of mind if I treat other people unfairly, even if they are
unfair to me. (40)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

82
• Principles like fairness and honesty have little role in some aspects of my life.
(75)
• I don't think that religious or ethical principles about what is right and wrong
should have much influence in business decisions. (88)
• Everyone should be treated with dignity and respect, even if they seem to be
unimportant or bad. (102)

Notes: Reverse-coded items are italicized. Item numbers (referenced in Supplemental
Tables 3-7) appear in parentheses.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

83
Chapter 2: Genetic and Environmental Etiology of Autonomic Responses to Fear
and Sadness and their Relationship to Prosocial Personality
Abstract
Through their subjective and biological components, emotions guide our attitudes,
cognitions, and behaviors. Although a great deal of research has focused on identifying
and describing the biological correlates of emotional experience, the etiological bases of
specific affective responses remain unclear. Similarly, despite a well established link
between autonomic functioning and prosociality, it is not yet known how genetic and
environmental factors influence this relationship. In a large, ethnically diverse twin
sample, heart rate and skin conductance levels were continuously recorded while
participants (ages 9-10) watched short clips from feature films intended to elicit feelings
of fear and sadness. Self-report measures of prosocial personality traits were also
administered at four time points (ages 9-10, 14-15, 16-18, and 19-20) throughout data
collection. Biometric model fitting indicated that cardiac responses to fear and sadness
were moderately heritable in both sexes, while electrodermal responses to fear and
sadness were primarily mediated by genetic factors in male twins, but shared
environmental factors in female twins. Psychophysiological responses to fear and sadness
during childhood partially predicted prosociality across childhood and adolescence in a
sex-specific manner, and these relationships were best explained by genetic influences
common to both phenotypes.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

84
Emotions are affective responses that have the potential to guide attitudes,
cognitions, and behaviors. Although they generally occur in response to everyday
situations and events, there are a number of methods that can be used to elicit emotional
responses in laboratory settings, including static images, sounds, film clips, short written
vignettes, and personal recollection. Of these techniques, film clips provide several
advantages in that they are standardized stimuli capable of ethically evoking intense
responses while maintaining ecological validity (Rottenberg, Ray, & Gross, 2007). By
measuring participants’ subjective, behavioral, psychophysiological, and/or neural
responses, researchers can assess the effectiveness of the induction technique, identify
potential downstream consequences of certain emotions, as well as describe their
biological correlates. Whether discrete emotions can be meaningfully distinguished from
one another on the basis of biological signatures is a point of spirited contention among
affective scientists (e.g., Barrett, 2006; Barrett et al., 2007; Ekman, Levenson, & Friesen,
1983; Izard, 2007; Lench, Flores, & Bench, 2007; Levenson, 2009; Lindquist, Siegel,
Quigley, & Barrett, 2013; Panksepp, 2000). Although commentary on the fundamental
nature of emotion is outside of the scope of the studies presented here, we begin by
summarizing the physiological responses known to co-occur with the experience of fear
and sadness.
Autonomic Correlates of Fear and Sadness
Fear occurs in response to the threat of impending physical or psychological harm.
On the other hand, sadness is often associated with loss, which may include separation
from/death of a loved one or the failure to achieve a desired goal (Power & Dalgleish,
1997). According to one particularly comprehensive review (Kreibig, 2010), fear
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

85
generally yields reliable activation of the sympathetic nervous system (e.g., increases in
cardiovascular and electrodermal activity) across induction techniques. Sadness tends to
yield more mixed results, which likely hinge on a variety of confounding variables. More
specifically, crying sadness tends to be characterized by sympathetic activation, whereas
non-crying sadness tends to be characterized by sympathetic-parasympathetic withdrawal.
However, the vast majority of studies that employ sad film clips find a deactivating
withdrawal response, although few specifically investigated subjects’ cry-status as a
possible mediator (Kreibig, 2010).
More recent film clip studies have obtained a similar pattern of results,
implicating sympathetic activity in fear, but providing little clarification with regard to
sadness. Fear has been found to yield significant increases in heart rate (HR) and skin
conductance levels (SCL) relative to a neutral emotional state, while sadness produced no
such change (Fernández et al., 2012). However, in a study involving adolescent males
with disruptive behavior disorder, healthy control subjects exhibited a significant
decrease in HR in response to sad clips relative to baseline (de Wied, van Boxtel,
Matthys, & Meeus, 2012). (Neither fear clips nor other autonomic measures were
employed in this study.)
Sex differences. Women are largely, although not universally, considered to be
more behaviorally, autonomically, and neurally responsive to emotional stimuli. In the
most recent study of affective film clips, women (compared to men) displayed higher
heart rates in response to film clips that depict a threat of physical harm, as well as clips
that depict the loss of a loved one. For skin conductance levels, women were more
reactive than men in response to threat clips, but not loss clips (Wilhelm et al., 2017).
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

86
Interpreting their results in the framework of the defense cascade model (Lang, Bradley,
& Cuthbert, 1997), the authors conclude that the differentiated reactivity patterns
correspond to different stages of the overall threat response. More specifically, men
exhibited a joint sympathetic/parasympathetic orienting response characterized by
increased electrodermal activity and heart rate deceleration, while women exhibited a
sympathetic defensive response characterized by increased electrodermal activity and
heart rate acceleration.
That women in the previous study (Wilhelm et al., 2017) responded in manner
that resembles a typical fight-or-flight response seems to contradict the growing body of
literature on an alternative “tend-and-befriend” response. According to Taylor and
colleagues (2000, 2006), the female stress response also likely involves an affiliative
component—protecting and caring for conspecifics in the face of threat—theoretically
facilitated by dampened sympathetic nervous system activity. Given this framework (and,
crucially, to the extent that film clips serve as a valid analog of physical or social threat),
it may be reasonable to expect women to respond with lower heart rate and skin
conductance levels than men.
Behavior Genetics of Affective Responses
By comparing relatives with varying degrees of genetic similarity, twin and
family studies allow researchers to investigate the extent to which genetic and
environmental factors play a role in shaping any trait, behavior, or biological response of
interest. Briefly, when monozygotic (MZ) twins (who are assumed to share 100 percent
of their genes) are more similar to each other than dizygotic (DZ) twins (who share, on
average, 50 percent of their genes), genetic effects are likely at play. The genetic effects
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

87
are presumed to be additive (A) if the correlation between MZ twins is less than twice as
large as the correlation between DZ twins, and non-additive (D) if the MZ twin
correlation is more than twice as large as the DZ twin correlation. When MZ twins and
DZ twins are relatively similar to each other, shared environmental (C) effects (e.g.,
parenting style, childhood neighborhood, etc.) are likely at play. Finally, non-shared
environmental (E) effects (e.g., individual life experiences) are those that distinguish one
twin in a pair from their co-twin.
Although studies of the genetic and environmental bases of affective responses
are relatively limited, they have employed a wide variety of induction techniques and
behavioral and biological measures. Significant genetic effects have been found to
underlie the P300 response to unpleasant pictures depicting physical injury or threat
(Weinberg, Venables, Hajcak Proudfit, & Patrick, 2015), the N240 and P300 responses to
fearful facial expressions (Anokhin, Golosheykin, & Heath, 2010), as well as the
frequency and duration of twins’ facial expressions during emotional film clips (Kendler
et al., 2008). However, there appears to be no genetic basis to the affect-modulated startle
reflex, despite substantial heritability of the overall startle response (Anokhin,
Golosheykin, & Heath, 2006; Dhamija, Tuvblad, Dawson, Raine, & Baker, 2017;
Vaidyanathan, Malone, Miller, McGue, & Iacono, 2014). Taken together, these results
suggest that, while common, the heritability of any given affective response is not
universal across induction methods and, perhaps, may not be evident in response
measures with very low temporal resolutions.
Personality-Based Modulations of Affective Responses
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

88
The growing field of personality neuroscience seeks to identify relationships
between nervous system functions, traits, and temperaments (DeYoung & Gray, 2009).
However, the idea that there is a biological basis to personality is not new. According to
Cloninger and colleagues’ (1993) psychobiological model, personality consists of both
temperament and character dimensions. While temperaments are thought to reflect
genetically based dispositions that arise early in life and are influenced by distinct neural
systems, character dimensions (including Cooperativeness—the focus of Study 2) are
thought to arise largely, if not entirely, from sociocultural learning that occurs later in life.
However, several findings appear to call this distinction into question (for a review, see
Paris, 2005), including comparable heritability estimates between temperament and
character dimensions (Ando et al., 2004) and identifiable neuroanatomical bases of
Cooperativeness (Yamasue et al., 2008).
Nevertheless, a great deal of evidence suggests that affective responses can be
moderated by individual differences that appear to be both genetic and environmental in
nature. For example, when exposed to fear-inducing film clips, temperamentally more
fearful 3-4 year-old children with lower quality parent-child relationships exhibited
higher skin conductance levels relative to children who were less fearful or had higher
quality relationships (Gilissen, Koolstra, van Ijzendoorn, Bakermans-Kranenburg, & van
der Veer, 2007). Likewise, participants from lower social classes reported increased
levels of compassion and experienced greater heart rate deceleration in response to videos
of pediatric cancer patients (Stellar, Manzo, Kraus, & Deltner, 2012).
Affective Responding and Prosociality
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

89
Prosocial tendencies, broadly construed, are reliably linked to genetic,
neuroendocrine, physiological, and neural markers (for a review, see Hastings, Miller,
Kahle, & Zahn-Waxler, 2013). For example, moderate resting vagal tone (an index of
parasympathetic regulation) has been found to concurrently and prospectively predict
children’s prosocial tendencies, as well as their prosocial behaviors during a donation
game and a simulated injury paradigm (Miller, Kahle, & Hastings, 2017). Likewise,
participants who could more effectively upregulate their parasympathetic nervous system
activity (measured by high-frequency heart rate variability) during a biofeedback task
were more likely to engage in altruistically-motivated prosocial behaviors across a
variety of decision-making games (Bornemann, Kok, Böckler, & Singer, 2016).
In addition to baseline measures, it appears that responses to affectively laden
stimuli might similarly predict prosociality. Young children’s dynamic changes in
respiratory sinus arrhythmia in response to sadness-inducing vignettes longitudinally
predicted prosocial responses to an adult’s simulated injury during middle childhood.
(Miller, Nuselovici, & Hastings, 2016). (Only children’s subjective empathic responses to
the vignettes concurrently predicted prosocial responses.) Furthermore, physiological
responses to others’ distress also appear to predict self-sacrificial prosocial behaviors.
After witnessing a confederate purportedly receive electrical shocks, those participants
who exhibited stronger skin conductance responses were also more likely to later choose
to receive the shocks themselves, thus sparing the confederate (Hein, Lamm, Brodbeck,
& Singer, 2011).
The Present Studies
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

90
To date, there have been no genetically informative studies involving autonomic
responses to affective film clips and their relationship to prosociality. In the present
studies, we investigate the genetic and environmental etiology of heart rate and skin
conductance levels in response to fear- and sadness-inducing clips (Study 1), as well as
the nature of the relationship between these responses and trait measures of prosocial
personality across childhood and adolescence (Study 2). In Study 1, we expect to find
significant genetic influences on autonomic responses to film clips. The heritability of
heart rate and skin conductance is well documented (e.g., Singh et al., 1999; Tuvblad et
al., 2012). Likewise, although the only genetically informative study to employ film clips
measured subjects’ facial expressions (Kendler et al., 2008), facial expressions and
specific autonomic responses can be expected to co-occur during emotional experience
(Levenson, Ekman, & Friesen, 1990). In Study 2, we expect a positive relationship
between prosocial personality traits and psychophysiological responses that is mediated
by genetic factors, given the moderate heritability of prosocial tendencies (Conway &
Slavich, 2017; Knafo-Noam, Vertsberger, & Israel, 2018) and the hypothesized
heritability of the autonomic responses investigated in Study 1.
Study 1
Methods
Participants. Participants are drawn from the first wave of the Southern
California Twin Study of Risk Factors for Antisocial Behavior (RFAB; Baker et al.,
2013), conducted between the years of 2001-2004, when twins were approximately 9-10
years old. RFAB is a longitudinal study of the genetic and environmental bases of social
and emotional development. Film clip data were available for a total of 1090 twins. Of
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

91
those, 937 were included in heart rate analyses and 910 were included in skin
conductance analyses. (Exclusion criteria are discussed in the following paragraphs and
demographic statistics for the present sample can be found in Table 1.)
The majority of same-sex twin pairs in the sample had their zygosity determined
via DNA microsatellite analysis. Monozygosity required at least seven concordant DNA
markers shared between twins, while dizygosity required one discordant marker (Baker et
al., 2013). In cases where DNA results were either inconclusive or unavailable, zygosity
was determined based on twins’ scores on the Twin Similarity Questionnaire (Lykken,
1978), a reliable predictor of DNA zygosity among RFAB participants (Baker et al.,
2006).
Exclusion criteria. Participants were automatically excluded from all analyses if
they withdrew consent mid-task, if they did not watch all four clips (at their own or their
parents’ request), if their autonomic data were rendered unusable during data collection
or initial data processing (e.g., due to equipment malfunction, excessive movement, etc.),
or if they did not provide valid data for the duration of each clip’s target response period
(described under “data reduction” heading) or the three-minute baseline rest period at the
beginning of testing.
For heart rate analyses only, subjects were automatically excluded if processing
software (described below) could not detect their R-waves with at least 80% confidence,
or if the metric was otherwise unavailable or incomputable. For skin conductance
analyses only, participants were automatically excluded if room temperature data were
not available for their testing session or if they provided negative skin conductance levels
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

92
(due to recording errors) or skin conductance levels at the recording equipment’s upper
limit (25 microsiemens) during any of the clips’ target periods.
Additionally, due to the large sample size, several metrics
12
were established to
systematically identify subjects with potentially unusable data (e.g., due to equipment
malfunction, excessive movement, etc.) that may not have been noted during initial data
collection and processing. Flagged cases were only excluded once visual inspection of
their second-by-second data confirmed the presence of artifacts and/or noise.
Procedure. The affective film clips paradigm was one of several tasks included in
an approximately one-hour battery of psychophysiological tasks, which began with a
three-minute rest period to collect baseline data. After viewing all of the film clips,
participants were asked whether they had previously seen any of the movies and were
also asked to rate (on a five-point scale, 1 = Not at all, 5 = Very) how sad, happy, scared,
and amused each of the clips made them feel.
Materials. During the task, participants viewed short (1-3 minute), emotionally
evocative clips of the following feature films (in order): A Little Princess (Johnson &
Cuarón, 1995), Jurassic Park (Kennedy, Molen, & Spielberg, 1993), Poltergeist
(Marshall, Spielberg, & Hooper, 1982), and The Champ (Lovell & Zeffirelli, 1979). Clips
from Jurassic Park and Poltergeist were intended to induce feelings of fear, while clips
from A Little Princess and The Champ were intended to induce feelings of sadness. A

1
For heart rate, subjects were flagged for visual inspection if their minimum heart rate
value during any of the target periods fell in the bottom 2% of the sample; any of the
following fell in the top 2% of the sample: (2) their maximum heart rate value during any
of the target periods, (3) their total variance (across the 971 seconds of the task), or (4)
their mean value of the absolute differences between adjacent seconds of the task.

2
For skin conductance, subjects were flagged if their range during any of the target
periods fell in the top or bottom 2% of the sample.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

93
black screen and a neutral nature scene (approximately 60 seconds each) were presented
in between clips.
Psychophysiological measures.
Cardiovascular activity. Heart rate was recorded via disposable adhesive
electrodes placed between each participant’s first and second rib on both sides of the
chest. Prior to electrode placement, electrode sites were cleaned with a cotton pad and
isopropyl rubbing alcohol. Alligator clips were attached to tabs on the electrodes and
connected to the recording device via a 48-inch lead wire.
Electrodermal activity. Skin conductance levels were recorded via reusable
silver-silver chloride electrodes placed on the distal phalanges of the index and middle
fingers of each participant’s non-dominant hand, as identified based on Edinburgh
Handedness Inventory scores (Oldfield, 1971). Electrode cavities were filled with a
water-soluble lubricant (i.e., KY jelly) in order to facilitate electrical conduction.
Electrodes were secured to participants’ fingers with waterproof tape and participants
were instructed to position their hand palm-side up on the arm of the chair and keep it as
still as possible.
Data acquisition and processing. Heart rate and skin conductance data were
recorded with equipment and software from the James Long Company (1999; Caroga
Lake, New York) and a 31-channel Isolated Bioelectric amplifier with a sampling rate of
512 Hz. Heart rate data were processed using the James Long Company’s interbeat
interval (IBI) analysis program, in which R-waves were detected offline using a four-pass
self-scaling peak-detection algorithm. This algorithm produced a data file containing the
onset time for each R-wave detected during the testing session. For purposes of artifact
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

94
detection, a diagram displaying each subject’s detected R-waves was visually compared
to the sampled ECG signal and edited accordingly. Each edited R-wave series was then
converted to a prorated heart period series with a sampling interval of 250 ms.
Data reduction. In order to effectively examine responses to the emotional film
clips, each second of raw HR and SCL data was corrected by subtracting each
participant’s mean arousal level during of the three-minute rest period at the beginning of
testing. Following de Wied and colleagues (2012), we then identified target periods
during each clip that will serve as the primary unit of analysis. These target periods
include the closing scene of each clip as well as the first 30 seconds
3
of each black screen
that immediately followed each clip, and ranged from 65 to 86 seconds in duration.
During each closing scene, characters displayed the target emotion by mourning the loss
of a loved one (sadness) or facing imminent physical danger (fear). Thus, each subject’s
response to each clip constitutes the mean of each second of baseline-corrected data
during the clip’s target period. Additionally, baseline-corrected skin conductance scores
have been further adjusted by regressing out the effects of ambient room temperature
recorded at the beginning and end of each testing session.
Statistical analyses. First, a series of paired sample t-tests were conducted to
determine whether the specific film clips successfully evoked the desired emotions. Next,
descriptive and phenotypic analyses were performed to assess mean differences and
bivariate correlations between film clip responses, as well as sex differences in

3
Because a substantial number of participants did not provide data for the full 60 seconds
after the final film clip, we utilize the first 30 seconds of each post-clip black screen in
order to maximize the amount of data included in the final analyses and ensure that
reactivity scores are comparable between the four clips.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

95
physiological reactivity. Twin correlations (between the first and second twin in each
pair) were also computed prior to biometric model fitting.
Finally, a series of univariate sex-limitation models were fit to the data to examine
sex differences in the genetic and environmental etiology of autonomic responses to fear
and sadness. A full sex-limitation model allows either the genetic correlation (rgOS) or
the shared environmental correlation (rcOS) for opposite-sex dizygotic twins to be freely
estimated—potentially below the values assigned to same-sex dizygotic twins. (A model
cannot simultaneously estimate rgOS and rcOS while remaining identifiable). Next, a
common effects model constrains both rgOS and rcOS to the values expected of same-
sex dizygotic twins (0.5 and 1.0, respectively). A significant drop in fit (as assessed by a
chi-square difference test) between a common effects model and a full sex-limitation
model indicates qualitative sex differences—different genes or shared environments
influencing the same phenotype in males and females. A scalar model likewise constrains
rgOS and rcOS while also constraining male and female ACE parameter estimates to be
equal. A significant drop in fit between a scalar model and a common effects model
indicates quantitative sex differences—differences in genetic/environmental influences
between males and females caused by the same genes or shared environments. Lastly, a
null model assumes no sex differences by further constraining the phenotypic variances
between male and female twins to be equal. Reduced models (e.g., AE, CE, E) were then
fit nested within each best-fitting common effects, scalar, or null model.
Results
Manipulation check. As intended, participants reported feeling significantly
sadder when viewing both A Little Princess and The Champ (relative to both Jurassic
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

96
Park and Poltergeist), all ps < .001. Likewise, they reported feeling significantly more
scared when viewing both Jurassic Park and Poltergeist (relative to both A Little
Princess and The Champ), all ps < .001. Finally, participants reported feeling the
intended negative emotion during each of the four clips (e.g., sadness during The Champ)
significantly more strongly than the unintended negative emotion (e.g., fear during The
Champ), all ps < .001 (see Table 2).
Descriptive statistics and phenotypic analyses. For ease of interpretation,
uncorrected raw scores are displayed in Table 3 and Figures 1 and 2, but all analyses are
performed on data that have been corrected as previously described. A 4 (clip) x 2 (sex)
mixed ANOVA on heart rate responses revealed a significant main effect of clip,
F(3,933) = 80.48, p < .01, and a significant interaction between clip and sex, F(3,933) =
2.88, p = .04. Pairwise comparisons indicated that, for both sexes, responses to A Little
Princess (the first clip) were significantly lower in magnitude than responses to the three
remaining clips, all ps < .01. For female twins, the response to Jurassic Park was
significantly higher than the response to Poltergeist, but not The Champ, and the
response to The Champ was significantly higher than the response to Poltergeist. For
male twins, the response to Jurassic Park was higher than the response to both
Poltergeist and The Champ, but there was no significant change in response magnitude
between Poltergeist and The Champ. Finally, female twins were significantly more
reactive than males only during The Champ, but not during the previous three clips. A 4
(clip) x 2 (sex) mixed ANOVA on skin conductance levels revealed no effect of clip,
F(3,906) = 0.14, p = .94, and a marginally significant interaction between clip and sex,
F(3,906) = 2.56, p = .05, such that male twins had significantly higher skin conductance
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

97
levels than female twins for each clip, all ps < .01, with no significant differences
between clips for either sex. (Mean levels for each clip’s target period are marked with
red horizontal bars in Figures 1 and 2.) For both heart rate and skin conductance levels,
the response to each clip was significantly positively correlated with the response to each
remaining clip for both males and females, all ps < .01, (see Table 4). As a result, further
analyses are conducted on the average of both fear clip responses and the average of both
sad clip responses.
Twin correlations. Twin correlations are presented in Table 5. With the
exception of sadness HR, all MZ correlations were significant and at least as high in
magnitude as DZ correlations. For fear HR, MZ correlations were notably higher than DZ
correlations in both sexes, suggesting genetic influences. For fear and sadness SCL, the
patterns of MZ and DZ correlations differed by sex, with genetic influences appearing
likely for male twins and shared environmental influences appearing likely for female
twins (due to comparable MZ/DZ correlations). For both emotions and physiological
measures, correlations for opposite-sex DZ twins were substantially lower than those of
their same-sex DZ counterparts, indicating potential sex differences in the etiology of
these phenotypes.
Univariate genetic analyses. Model fitting results are presented in Tables 6 and
7. For heart rate (fear, sadness, and both emotions combined), full sex-limitation models
could be reduced to null models (assuming no sex differences) without a significant drop
in fit, all ps > .10. Each null ACE model could be further reduced to an AE model, all ps
> .60, that yielded significant genetic influences on cardiac responses to fear (h
2
= 0.29),
sadness (h
2
= 0.23), and both emotions combined (h
2
= 0.30). Each null AE model also
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

98
had the lowest AIC and BIC values of all of the models fit for each category of heart rate
data. For skin conductance level (fear, sadness, and both emotions combined), full sex-
limitation models could be reduced to common effects models, all ps > .85. When
compared to the common effects models, scalar and null models led to a significant drop
in model fit according to chi-square difference tests, all ps < .02, as well as increases in
AIC and BIC values, indicating quantitative sex differences between male and female
twins that are influenced by similar genes and shared environments. Each common
effects ACE model could be further reduced to a model that retained AE effects for males
and CE effects for females, all ps > .69, which yielded significant genetic effects for male
twins (fear h
2
= 0.35, sadness h
2
= 0.36, combined h
2
= 0.36) and significant shared
environmental effects for female twins (fear c
2
= 0.34, sadness c
2
= 0.28, combined c
2
=
0.31).
Study 2
Methods
Participants. Participants include all RFAB twins included in Study 1 who also
provided (Junior) Temperament and Character Inventory data during any of the waves
during which the measure was administered: 921 for heart rate analyses and 896 for skin
conductance analyses.
Procedure. At each wave of RFAB data collection, twins completed a battery of
personality surveys and neuropsychological tests in addition to psychophysiological tasks.
During the first and third waves, twins (aged 9-10 and 14-15) completed an abbreviated
version of the Junior Temperament and Character Inventory (JTCI; Luby, Svrakic,
McCallum, Przybeck, & Cloninger, 1999). During the fourth and fifth waves, twins (aged
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

99
16-18 and 19-20) completed an abbreviated version of the Temperament and Character
Inventory (TCI; Cloninger, Przybeck, Svrakic, & Wetzel, 1994; Cloninger, Svrakic, &
Przybeck, 1993). During the first wave, twins participated in an oral interview to provide
JTCI responses. During all other waves, participants independently completed the (J)TCI
either during their in-person lab visit or via mail/online surveys. Details of
psychophysiological data collection are as described in Study 1.
Measures. The TCI Cooperativeness scale assesses subjects’ levels of social
acceptance, empathy, helpfulness, compassion, and pure-heartedness via 25 binary-
choice items (0 = No, 1 = Yes). The JTCI contains 20 Cooperativeness items that are
largely comparable to the TCI items, but slightly revised to be more accessible for
younger respondents. Scale scores are computed by averaging responses to all of the
items in the scale, with higher scores (out of 1) indicating higher levels of
Cooperativeness.
Statistical analyses. First, a series of bivariate correlations were performed to
identify potential phenotypic relationships between film clip responses during childhood
and Cooperativeness throughout childhood and adolescence. Next, a bivariate Cholesky
decomposition was conducted for each significant Cooperativeness/film clip relationship.
By partitioning the individual variances of two phenotypes as well as their covariance,
the Cholesky decomposition can identify the independent and overlapping genetic and
environmental influences underlying those phenotypes. Each decomposition can take the
form of an ACE model or an ADE model—identifying additive genetic (A) and non-
shared environmental (E) influences and either shared environmental (C) or non-additive
genetic (D) influences on both phenotypes. Genetic (r
g
/r
d
) and environmental (r
c
/r
e
)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

100
correlations each indicate the extent to which effects on one measure overlap with the
same type of effects on another measure. These values can range from -1.0 and +1.0, and
are not related to the magnitudes of independent genetic or environmental influences on
each set of measures (Posthuma et al., 2003). For the present analyses, ADE Cholesky
decompositions were performed because shared environmental (C) effects did not
influence the psychophysiological measures of interest (Study 1) and because non-
additive genetic (D) effects were found to influence Cooperativeness more so than shared
environmental (C) effects (Berntsen, Beam, Tuvblad, Raine, & Baker, submitted). All
Cooperativeness and psychophysiological reactivity scores were standardized (via z-
scores) prior to genetic analyses.
Results
Phenotypic correlations. In male twins, fear HR positively predicted
Cooperativeness levels at ages 14-15, r(219)= .17, p = .01, and ages 19-20, r(238) = .16,
p = .02, while fear SCL, r(223) = .17, p = .01, and sadness SCL, r(223) = .14, p = .04,
positively predicted Cooperativeness at ages 14-15. In female twins, sadness HR
negatively predicted Cooperativeness at ages 16-18, r(279) = -.15, p = .02. In each of
these cases, the significant relationship between Cooperativeness and autonomic
responses held when averaging the responses to fear and sadness clips into a single score
(see Table 8).
Bivariate genetic analyses. Full model-fitting results are presented in Table 9.
For all Cooperativeness/autonomic response pairings, an AE Cholesky decomposition
with the non-shared environmental correlation (r
e
) constrained to zero provided the best
fit, based on lowest AIC and BIC values. For autonomic responses, independent genetic
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

101
effects accounted for 24% to 39% of the total variance, while independent non-shared
environmental effects accounted for 61% to 76% of the total variance. For
Cooperativeness independent genetic effects accounted for 22% to 39% of the total
variance, while independent non-shared environmental effects accounted for 61% to 78%
of the total variance. Although modest, heritability estimates for both phenotypes were
significant (with 95% confidence intervals excluding zero) in each best-fitting model.
In each case, the genetic correlation (r
g
) was substantial, but not necessarily
statistically significant. For male twins, the relationship between skin conductance levels
during both fear- and sadness-inducing film clips at ages 9-10 and Cooperativeness at
ages 14-15 could be explained by genetic factors; for fear, r
g
= 0.82 (95% CI: 0.41-1.00),
for sadness, r
g
= 0.71 (95% CI: 0.29-1.00). Although the genetic correlations between
heart rate responses and Cooperativeness were non-negligible in magnitude (all r
g
values
more extreme than +/- 0.28), their 95% confidence intervals did include zero. As a result,
we can surmise, but not definitively conclude, that common genetic factors are the most
likely source of the relationship between cardiac responses and Cooperativeness.
Discussion
Nature of Fear and Sadness Responses
In male and female twins, both heart rate and skin conductance levels steadily
increased during the target periods of fear and sadness clips (see Figures 1 and 2).
Although consistent with the well-documented sympathetic fear response, these findings
appear to conflict with the withdrawal response typically observed during sadness-
inducing film clips (Kreibig, 2010). Because the manipulation check confirmed that both
A Little Princess and The Champ successfully induced feelings of sadness, one potential
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

102
explanation for these results may lie in the diversity of techniques used to handle and
reduce psychophysiological data collected during film clip tasks. The most popular
approach defines a subject’s response as the mean or median value of every second
across the entire duration of a clip (e.g., Gilissen et al., 2007; Gomez, Zimmermann,
Guttormsen-Schär, & Danuser, 2005; Kreibig, Wilhelm, Roth, & Gross, 2007; Kreibig,
Samson, & Gross, 2013; Stellar et al., 2012). Some average across multiple epochs of
predetermined length (e.g., 60 seconds) occurring during each clip (e.g., Codispoti,
Surcinelli, & Baldaro, 2008; Palomba, Sarlo, Angrilli, Mini, & Stegagno, 2000), while
others rely on the single highest value recorded during a clip (e.g., Fernández et al., 2012).
Moreover, each of these studies assessed whether arousal levels during specific clips
increased or decreased relative to a baseline period, rather than examining second-by-
second responses to the emotionally evocative events in each clip. As a result, it may be
possible that sadness is a multi-component response (e.g., Miller et al., 2016): partially
characterized by autonomic arousal in direct response to grief or loss, but not to the
extent that would yield a significant overall increase relative to a neutral baseline period.
However, because the experimental design of the film clips task precludes us from
drawing additional descriptive conclusions about the nature of fear and sadness responses
(discussed further in the Limitations section), we now focus on one of the main goals of
the present studies.
Etiology of Fear and Sadness Responses
To our knowledge, this is the first study to examine the genetic and environmental
etiology of autonomic responses to affective film clips. In male and female twins, heart
rate responses to both fear and sadness were significantly influenced by additive genetic
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

103
(A) influences, with the remaining variance attributable to non-shared environmental (E)
factors. This finding is consistent with our hypothesis and with other studies that have
established the heritability of affective responses, including the magnitude of the N240
and P300 event-related potentials evoked by negative stimuli (Anokhin et al., 2010;
Weinberg et al., 2015), as well as heart rate levels, broadly construed (Singh et al., 1999).
The etiology of electrodermal reactivity to fear and sadness, however, is more
complex. Although broadly influenced by the same genes/environments, the magnitude
of genetic and environmental influences significantly differed between males and females
(evident in the drop in fit between the common effects and scalar models). Moreover, a
common effects ACE model could be further reduced such that additive genetic (A)
influences could explain significant amount of the total variance in male (but not female)
twins while shared environmental (C) influences could explain a significant amount of
the total variance in female (but not male) twins—a pattern that held for both fear and
sadness clips. Multivariate genetic analyses conducted on the present sample have
revealed comparable genetic (A) and shared environmental (C) influences on resting
sympathetic nervous system activity (including both HR and SCL) in both males and
females (Tuvblad et al., 2010). Thus, it is likely that the observed sex-specific etiological
bases are driven specifically by the fear and sadness manipulation. Furthermore, because
a common effects model implies that the same genes and environments influence both
sexes (albeit to different extents), this suggests two related conclusions: (1) that female
twins may differ from each other due to shared environments that they are more likely to
encounter than male twins and (2) male twins may differ from each other due to
biological predispositions that are more likely to occur in females than males.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

104
The first conclusion is widely supported; from an early age, women are exposed
to socialization practices that encourage them to be more attuned and responsive to the
emotional needs of others (e.g., Eccles, Jacobs, & Harold, 1990; Kennedy Root &
Denham, 2010; Eisenberg, Spinrad, & Knafo-Noam, 2015). (While it is certainly true that
these socialization practices may interact with an evolutionary tendency towards
nurturance, it is outside of the scope of the present work to comment on the relative
importance of those two factors.) While our second conclusion is more speculative in
nature, the existing evidence points to likely involvement of the oxytocinergic system—
implicated in attachment and caregiving, among other social behaviors (for a review, see
MacDonald & MacDonald, 2010). Human oxytocin receptors are at least partially
localized to the amygdala (Boccia, Petrusz, Suzuki, Marson, & Pedersen, 2013), which
has been found to mediate skin conductance responses to threatening stimuli (Wood, Ver
Hoef, & Knight, 2014). Molecular genetic research has also established the link between
naturally occurring variants of the oxytocin receptor gene and trait levels of emotional
empathy (Uzefovsky et al., 2015; Wu, Li, & Su, 2012), and intranasal administration of
oxytocin has been found to increase self-reported emotional empathy in men (Hurlemann
et al., 2010).
Autonomic Reactivity as a (Partial) Predictor of Cooperativeness
Cardiac and electrodermal responses to affective film clips selectively predicted
Cooperativeness throughout development. For male twins, elevated heart rate during fear
clips was associated with increased Cooperativeness at ages 14-15 and ages 19-20, while
elevated skin conductance levels during both fear and sadness clips were associated with
increased Cooperativeness at ages 14-15 alone. For female twins, lower heart rate during
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

105
sad clips was associated with increased Cooperativeness at ages 16-18. Contrary to our
expectations, neither measure of autonomic reactivity was concurrently associated with
Cooperativeness levels at ages 9-10 in either sex. However, during the first wave of data
collection, the JTCI was administered during an oral interview, which may have
implicitly pressured participants to provide more socially desirable (but potentially less
valid) responses than they might while completing an anonymous survey. An explanation
for the partial prediction of Cooperativeness throughout adolescence is less clear, but
may lie in the nature of the scale itself. As the (J)TCI only consists of binary choice items
and scores were highly negatively skewed during all four waves, the measure may not
have effectively discriminated among the majority of participants with more prosocial
tendencies. However, the relationship between autonomic reactivity and Cooperativeness
at any time point is notable in that it suggests that there are biological bases to a character
dimension of Cloninger’s psychobiological model, contrary to its initial conceptualization
(Cloninger et al., 1993).
Sex differences. An additional notable result is that higher levels of
Cooperativeness were associated with increased autonomic arousal (both in terms of HR
and SCL) in males, but lower HR reactivity in females, but only in response to sadness-
inducing clips. Here, it may be that higher levels of Cooperativeness in female twins
facilitated the tend-or-befriend response, in which women respond to threat by protecting
or offering social support to those in need, rather than confronting or avoiding the threat
itself (Taylor et al., 2000; Taylor, 2006). Although there is limited research on the real-
time autonomic profile of the tend-and-befriend response per se, this finding is broadly
consistent with research implicating the parasympathetic nervous system in the
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

106
experience of compassion (Stellar, Cohen, Oveis, & Keltner, 2015; Stellar & Keltner,
2017)
Etiology of the Autonomic Reactivity/Cooperativeness Relationship
To our knowledge, this is also the first study to investigate the shared etiological
bases of affectively modulated autonomic responses and prosocial personality traits. We
found common genetic factors could explain the relationship observed in male twins
between fear and sadness SCL at ages 9-10 and Cooperativeness at ages 14-15. This
finding adds to a growing body of literature establishing genetic overlap between
biological responses and antisocial personality traits (e.g., Hicks et al., 2007; Isen, Iacono,
Malone, & McGue, 2012; Wang et al., 2015) and provides initial evidence that similar
etiological dynamics are at play with respect to their more prosocial counterparts.
For each bivariate model, the overall fit improved and 95% confidence intervals
for r
g
became increasingly narrow as additional parameters were dropped. However,
confidence intervals for each best-fitting model only excluded zero in cases where skin
conductance level—but not heart rate—was the physiological variable of interest.
However, heart rate in both sexes (like skin conductance in male twins) was significantly
heritable in univariate analyses and the same bivariate model (AE, with r
e
dropped) was
best fitting for each phenotypic relationship. As a result, it appears that shared genetic
influences are the most likely source of covariation between heart rate responses and
Cooperativeness, but future research would be necessary to confirm this assumption.
Limitations and Concluding Remarks
There are limitations to the present study, the majority of which are due to the fact
that the affective film clips paradigm was designed and implemented nearly two decades
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

107
ago—prior to the development and widespread dissemination of best practices for these
kinds of tasks. First, the film clips paradigm was embedded within an approximately one-
hour battery that did not originally feature dedicated recovery periods before and after
each task. As a result, our baseline data is derived from a three-minute rest period that
occurred at the start of psychophysiological data collection. While this approach does
allow us to control for participants’ trait arousal levels (and isolate responses to the film
clips, per se), it does not allow us to draw meaningful conclusions about the nature of
change that occurs between baseline and any of the clips.
4
Likewise, because the neutral
nature scenes featured in between emotional clips were shorter than the three minutes
recommended to effectively minimize response spillover effects (e.g., Fernández et al.,
2012; Kreibig et al., 2007), we are similarly unable to draw meaningful conclusions
regarding the nature of change occurring from clip to clip.
5
Additionally, our findings
must be interpreted in light of limitations inherent to the classical twin design (including
the equal environments assumption), which are discussed in greater detail elsewhere
(Tuvblad & Baker, 2011).
Limitations notwithstanding, the current research makes several novel
contributions. Using a large, ethnically diverse community sample, we provide evidence
of the heritability of cardiovascular responses to fear and sadness-inducing film clips, as

4
Participants’ mean arousal levels were significantly higher during the initial rest period
than during any of the film clips’ target periods. As the rest period occurred immediately
after our nine-year old participants were fit with psychophysiological recording
equipment for the first time, it is not entirely unexpected that they were in a more
heightened state of arousal at that point in time relative to an hour later.

5
Exploratory analyses not presented here confirmed that uncorrected mean arousal levels
were significantly higher during the neutral nature scenes than during any of the clips. (In
Figures 1 and 2, the (unlabeled) neutral nature scenes are represented by the sharp
deceleration occurring after each clip’s (labeled) target period.)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

108
well as the sex-specific etiological bases of electrodermal responses. Additionally, we
find that the relationship between electrodermal responses to fear and sadness and
prosocial personality traits during adolescence can be accounted for by common genetic
influences.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

109
References
Ando, J., Suzuki, A., Yamagata, S., Kijima, N., Maekawa, H., Ono, Y., & Jang, K. L.
(2004). Genetic and environmental structure of Cloninger's temperament and
character dimensions. Journal of Personality Disorders, 18(4), 379-393.
Anokhin, A. P., Golosheykin, S., & Heath, A. C. (2007). Genetic and environmental
influences on emotion-modulated startle reflex: A twin study. Psychophysiology,
44(1), 106-112.
Anokhin, A. P., Golosheykin, S., & Heath, A. C. (2010). Heritability of individual
differences in cortical processing of facial affect. Behavior Genetics, 40(2), 178-
185.
Baker, L. A., Barton, M., Lozano, D. I., Raine, A., & Fowler, J. H. (2006). The Southern
California Twin Register at the University of Southern California: II. Twin
Research and Human Genetics, 9(6), 933-940.
Baker, L. A., Tuvblad, C., Wang, P., Gomez, K., Bezdjian, S., Niv, S., & Raine, A.
(2013). The Southern California Twin Register at the University of Southern
California: III. Twin Research and Human Genetics, 16(1), 336-343.
Barrett, L. F. (2006). Are emotions natural kinds?. Perspectives on Psychological Science,
1(1), 28-58.
Barrett, L. F., Lindquist, K. A., Bliss-Moreau, E., Duncan, S., Gendron, M., Mize, J., &
Brennan, L. (2007). Of mice and men: Natural kinds of emotions in the
mammalian brain? A response to Panksepp and Izard. Perspectives on
Psychological Science, 2(3), 297-312.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

110
Berntsen, L., Beam, C. R., Tuvblad, C., Raine, A., & Baker, L. A. (submitted). Genetic
and environmental bases of Cooperativeness from late childhood to emerging
adulthood.
Boccia, M. L., Petrusz, P., Suzuki, K., Marson, L., & Pedersen, C. A. (2013).
Immunohistochemical localization of oxytocin receptors in human brain.
Neuroscience, 253, 155-164.
Bornemann, B., Kok, B. E., Böckler, A., & Singer, T. (2016). Helping from the heart:
Voluntary upregulation of heart rate variability predicts altruistic behavior.
Biological Psychology, 119, 54-63.
Cloninger, C. R., Przybeck, T. R., Svrakic, D. M., & Wetzel, R. D. (1994). The
Temperament and Character Inventory (TCI): A guide to its development and
use (pp. 19-28). St. Louis, MO: Center for Psychobiology of Personality,
Washington University.
Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of
temperament and character. Archives of General Psychiatry, 50(12), 975-990.
Codispoti, M., Surcinelli, P., & Baldaro, B. (2008). Watching emotional movies:
Affective reactions and gender differences. International Journal of
Psychophysiology, 69(2), 90-95.
Conway, C. C., & Slavich, G. M. (2017). Behavior genetics of prosocial behavior. In P.
Gilbert (Ed.), Compassion: Concepts, research and applications (pp. 151-170).
London: Routledge.
de Wied, M., van Boxtel, A., Matthys, W., & Meeus, W. (2012). Verbal, facial and
autonomic responses to empathy-eliciting film clips by disruptive male
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

111
adolescents with high versus low callous-unemotional traits. Journal of Abnormal
Child Psychology, 40(2), 211-223.
DeYoung, C. G., & Gray, J. R. (2009). Personality neuroscience: Explaining individual
differences in affect, behavior, and cognition. In P. J. Corrs & G. Matthews (Eds.),
The Cambridge handbook of personality psychology (pp. 323-346). Cambridge:
Cambridge University Press.
Dhamija, D., Tuvblad, C., Dawson, M. E., Raine, A., & Baker, L. A. (2017). Heritability
of startle reactivity and affect modified startle. International Journal of
Psychophysiology, 115, 57-64.
Eccles, J. S., Jacobs, J. E., & Harold, R. D. (1990). Gender role stereotypes, expectancy
effects, and parents' socialization of gender differences. Journal of Social Issues,
46(2), 183-201.
Eisenberg, N., Spinrad, T. L., & Knafo-Noam, A. (2015). Prosocial development. In R.
M. Lerner & M. E. Lamb (Eds.), Handbook of child psychology and
developmental science, Vol. 3: Socioemotional processes (7
th
ed., pp. 610-656).
Hoboken: John Wiley & Sons, Inc.
Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic nervous system
activity distinguishes among emotions. Science, 221, 1208–1210.
Fernández, C., Pascual, J. C., Soler, J., Elices, M., Portella, M. J., & Fernández-Abascal,
E. (2012). Physiological responses induced by emotion-eliciting films. Applied
Psychophysiology and Biofeedback, 37(2), 73-79.
Gilissen, R., Koolstra, C. M., van Ijzendoorn, M. H., Bakermans-Kranenburg, M. J., &
van der Veer, R. (2007). Physiological reactions of preschoolers to fear-inducing
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

112
film clips: Effects of temperamental fearfulness and quality of the parent–child
relationship. Developmental Psychobiology, 49(2), 187-195.
Gomez, P., Zimmermann, P., Guttormsen-Schär, S., & Danuser, B. (2005). Respiratory
responses associated with affective processing of film stimuli. Biological
Psychology, 68(3), 223-235.
Hastings, P. D., Miller, J. G., Kahle, S., & Zahn-Waxler, C. (2013). The neurobiological
bases of empathic concern for others. In M. Killen & J. G. Smetana (Eds.),
Handbook of moral development, Vol. 2, (pp. 411-434). New York: Psychology
Press.
Hein, G., Lamm, C., Brodbeck, C., & Singer, T. (2011). Skin conductance response to the
pain of others predicts later costly helping., 6(8), e22759.
Hicks, B. M., Bernat, E., Malone, S. M., Iacono, W. G., Patrick, C. J., Krueger, R. F., &
McGue, M. (2007). Genes mediate the association between P3 amplitude and
externalizing disorders. Psychophysiology, 44(1), 98-105.
Hurlemann, R., Patin, A., Onur, O. A., Cohen, M. X., Baumgartner, T., Metzler, S., ... &
Kendrick, K. M. (2010). Oxytocin enhances amygdala-dependent, socially
reinforced learning and emotional empathy in humans. Journal of Neuroscience,
30(14), 4999-5007.
Isen, J. D., Iacono, W. G., Malone, S. M., & McGue, M. (2012). Examining
electrodermal hyporeactivity as a marker of externalizing psychopathology: A
twin study. Psychophysiology, 49(8), 1039-1048.
Izard, C. (2007) Basic emotions, natural kinds, emotion schemas, and a new paradigm.
Perspectives in Psychological Science, 2, 260–280.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

113
Johnson, M. (Producer), & Cuarón, A. (Director). (1995). A little princess [Motion
Picture]. United States: Warner Bros Pictures.
Kendler, K. S., Halberstadt, L. J., Butera, F., Myers, J., Bouchard, T., & Ekman, P.
(2008). The similiarity of facial expressions in response to emotion-inducing
films in reared-apart twins. Psychological Medicine, 38(10), 1475-1483.
Kennedy, K. & Molen, G. R. (Producers), & Spielberg, S. (Director). (1993). Jurassic
park [Motion Picture]. United States: Universal Pictures.
Kennedy Root, A., & Denham, S. A. (2010). The role of gender in the socialization of
emotion: Key concepts and critical issues. In A. Kennedy Root & S. Denham
(Eds.), The role of gender in the socialization of emotion: Key concepts and
critical issues. New Directions for Child and Adolescent Development, 128, 1–9.
San Francisco: Jossey-Bass.
Knafo-Noam, A., Vertsberger, D., & Israel, S. (2018). Genetic and environmental
contributions to children's prosocial behavior: Brief review and new evidence
from a reanalysis of experimental twin data. Current Opinion in Psychology, 20,
60-65.
Kreibig, S. D. (2010). Autonomic nervous system activity in emotion: A review.
Biological Psychology, 84(3), 394-421.
Kreibig, S. D., Samson, A. C., & Gross, J. J. (2013). The psychophysiology of mixed
emotional states. Psychophysiology, 50(8), 799-811.
Kreibig, S. D., Wilhelm, F. H., Roth, W. T., & Gross, J. J. (2007). Cardiovascular,
electrodermal, and respiratory response patterns to fear‐and sadness‐inducing
films. Psychophysiology, 44(5), 787-806.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

114
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1997). Motivated attention: Affect,
activation, and action. In P. J. Lang, R. F. Simons, & M. Balaban (Eds.). Attention
and orienting: Sensory and motivational processes (pp. 97–135). New Jersey:
Lawrence Erlbaum Associates.
Lench, H. C., Flores, S. A., & Bench, S. W. (2011). Discrete emotions predict changes in
cognition, judgment, experience, behavior, and physiology: A meta-analysis of
experimental emotion elicitations. Psychological Bulletin, 137(5), 834-855.
Levenson, R.W. (2009). Autonomic specificity and emotion. In: R. J. Davidson, K. R.
Scherer, & H. H. Goldsmith (Eds.), Handbook of affective sciences, (pp. 212-224).
Oxford: Oxford University Press.
Levenson, R. W., Ekman, P., & Friesen, W. V. (1990). Voluntary facial action generates
emotion-specific autonomic nervous system activity. Psychophysiology, 27(4),
363-384.
Lindquist, K. A., Siegel, E. H., Quigley, K. S., & Barrett, L. F. (2013). The hundred-year
emotion war: Are emotions natural kinds or psychological constructions?
Comment on Lench, Flores, and Bench (2011). Psychological Bulletin, 139(1),
255-263.
Lovell, D. (Producer), & Zeffirelli, F. (Director). (1979). The champ [Motion Picture].
United States: Metro-Goldwyn-Mayer.
Luby, J. L., Svrakic, D. M., McCallum, K., Przybeck, T. R., & Cloninger, C. R. (1999).
The Junior Temperament and Character Inventory: Preliminary validation of a
child self-report measure. Psychological Reports, 84, 1127-1138.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

115
Lykken, D. T. (1978). The diagnosis of zygosity in twins. Behavior Genetics, 8(5), 437-
473.
MacDonald, K., & MacDonald, T. M. (2010). The peptide that binds: A systematic
review of oxytocin and its prosocial effects in humans. Harvard Review of
Psychiatry, 18(1), 1-21.
Marshall, F. & Spielberg, S. (Producers), & Hooper, T. (Director). (1982). Poltergeist
[Motion Picture]. United States: MGM/UA Entertainment Co.
Miller, J. G., Kahle, S., & Hastings, P. D. (2017). Moderate baseline vagal tone predicts
greater prosociality in children. Developmental Psychology, 53(2), 274-289.
Miller, J. G., Nuselovici, J. N., & Hastings, P. D. (2016). Nonrandom acts of kindness:
Parasympathetic and subjective empathic responses to sadness predict children's
prosociality. Child Development, 87(6), 1679-1690.
Oldfield, R. C (1971). The assessment and analysis of handedness: The Edinburgh
Inventory. Neuropsychologia, 9, 97-113.
Palomba, D., Sarlo, M., Angrilli, A., Mini, A., & Stegagno, L. (2000). Cardiac responses
associated with affective processing of unpleasant film stimuli. International
Journal of Psychophysiology, 36(1), 45-57.
Panksepp, J. (2000). Emotions as natural kinds within the mammalian brain. In: M. Lewis
& J. Haviland (Eds.), Handbook of emotions, Vol. 2 (pp. 87–107). New York:
Guilford Press.
Paris, J. (2005). Neurobiological dimensional models of personality: A review of the
models of Cloninger, Depue, and Siever. Journal of Personality Disorders, 19(2),
156-170.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

116
Posthuma, D., Beem, A. L., De Geus, E. J., Van Baal, G. C. M., von Hjelmborg, J. B.,
Iachine, I., & Boomsma, D. I. (2003). Theory and practice in quantitative genetics.
Twin Research and Human Genetics, 6(5), 361-376.
Power, M. J., & Dalgleish, T. (1997). Cognition and emotion: From order to disorder.
Hove, UK: Psychology Press.
Rottenberg, J., Ray, R. D., & Gross, J. J. (2007). Emotion elicitation using films. In J. A.
Coan & J. J. B. Allen (Eds.), Handbook of emotion elicitation and assessment (pp.
9-28). Oxford: Oxford University Press.
Singh, J. P., Larson, M. G., O’Donnell, C. J., Tsuji, H., Evans, J. C., & Levy, D. (1999).
Heritability of heart rate variability: The Framingham Heart Study.
Circulation, 99(17), 2251-2254.
Stellar, J. E., Cohen, A., Oveis, C., & Keltner, D. (2015). Affective and physiological
responses to the suffering of others: Compassion and vagal activity. Journal of
Personality and Social Psychology, 108(4), 572-585.
Stellar, J. E., & Keltner, D. (2017). Compassion in the autonomic nervous system: The
role of the vagus nerve. In P. Gilbert (Ed.), Compassion: Concepts, research and
applications (pp. 120-134). London: Routledge.
Stellar, J. E., Manzo, V. M., Kraus, M. W., & Keltner, D. (2012). Class and compassion:
Socioeconomic factors predict responses to suffering. Emotion, 12(3), 449-459.
Taylor, S. E. (2006). Tend and befriend: Biobehavioral bases of affiliation under stress.
Current Directions in Psychological Science, 15(6), 273-277.
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

117
Taylor, S. E., Klein, L. C., Lewis, B. P., Gruenewald, T. L., Gurung, R. A., & Updegraff,
J. A. (2000). Biobehavioral responses to stress in females: Tend-and-befriend, not
fight-or-flight. Psychological Review, 107(3), 411-429.
Tuvblad, C., & Baker, L. A. (2011). Human aggression across the lifespan: Genetic
propensities and environmental moderators. In Advances in genetics, Vol. 75, (pp.
171-214). Academic Press.
Tuvblad, C., Gao, Y., Isen, J., Botwick, T., Raine, A., & Baker, L. A. (2012). The
heritability of the skin conductance orienting response: A longitudinal twin study.
Biological Psychology, 89(1), 47-53.
Tuvblad, C., Isen, J., Baker, L. A., Raine, A., Lozano, D. I., & Jacobson, K. C. (2010).
The genetic and environmental etiology of sympathetic and parasympathetic
activity in children. Behavior Genetics, 40(4), 452-466.
Uzefovsky, F., Shalev, I., Israel, S., Edelman, S., Raz, Y., Mankuta, D., ... & Ebstein, R.
P. (2015). Oxytocin receptor and vasopressin receptor 1a genes are respectively
associated with emotional and cognitive empathy. Hormones and Behavior, 67,
60-65.
Vaidyanathan, U., Malone, S. M., Miller, M. B., McGue, M., & Iacono, W. G. (2014).
Heritability and molecular genetic basis of acoustic startle eye blink and
affectively modulated startle response: A genome-wide association study.
Psychophysiology, 51(12), 1285-1299.
Wang, P., Gao, Y., Isen, J., Tuvblad, C., Raine, A., & Baker, L. A. (2015). Genetic
covariance between psychopathic traits and anticipatory skin conductance
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

118
responses to threat: Evidence for a potential endophenotype. Development and
Psychopathology, 27(4pt1), 1313-1322.
Weinberg, A., Venables, N. C., Proudfit, G. H., & Patrick, C. J. (2015). Heritability of the
neural response to emotional pictures: Evidence from ERPs in an adult twin
sample. Social Cognitive and Affective Neuroscience, 10(3), 424-434.
Wilhelm, F. H., Rattel, J. A., Wegerer, M., Liedlgruber, M., Schweighofer, S., Kreibig, S.
D., ... & Blechert, J. (2017). Attend or defend? Sex differences in behavioral,
autonomic, and respiratory response patterns to emotion–eliciting films.
Biological Psychology, 130, 30-40.
Wood, K. H., Ver Hoef, L. W., & Knight, D. C. (2014). The amygdala mediates the
emotional modulation of threat-elicited skin conductance response. Emotion,
14(4), 693-700.
Wu, N., Li, Z., & Su, Y. (2012). The association between oxytocin receptor gene
polymorphism (OXTR) and trait empathy. Journal of Affective Disorders, 138(3),
468-472.
Yamasue, H., Abe, O., Suga, M., Yamada, H., Rogers, M. A., Aoki, S., ... & Kasai, K.
(2008). Sex-linked neuroanatomical basis of human altruistic cooperativeness.
Cerebral Cortex, 18(10), 2331-2340.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

119
Tables and Figures

Table 1: Demographic statistics

Either
Measure
Heart Rate
Skin
Conductance
Both
Measures

Total Participants 1030 937 910 817

Total Families 564 546 525 495

Total Complete Pairs 466 391 385 322

Complete Pairs
by Zygosity

Complete MZM Pairs 103 88 89 75
Complete MZF Pairs 115 96 95 79
Complete DZM Pairs 68 56 57 48
Complete DZF Pairs 70 58 57 48
Complete DZOS Pairs 110 93 87 72

Sex/Gender
Male 497 451 439 392
Female 533 486 471 424

Race/Ethnicity
Asian 45 40 41 36
Black 140 123 114 97
Hispanic 394 360 361 327
Multiracial 182 170 159 147
Native American 2 2 2 2
White 267 242 233 207
Not Indicated

Mean Age 9.59 9.60 9.58 9.59

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

120
Table 2: Emotion induction ratings by film clip

Intended
Emotion
Film Clip Sadness Rating Fear Rating
t-statistic
(intended emotion –
unintended emotion)
Fear
Jurassic Park 2.23 3.61 t(1029) = 27.68, p < .001
Poltergeist 2.89 3.61 t(1029) = 6.76, p < .001
Sadness
A Little Princess 3.27 1.99 t(1029) = 12.21 , p < .001
The Champ 4.28 2.45 t(1029) = 10.78, p < .001

Notes: Emotion ratings could range from 1 (“Not at all”) to 5 (“ A lot”). * = p < .05; ** =
p < .01.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

121
Table 3: Descriptive statistics

Heart Rate
(bpm)
Sex Mean SD Min Max
Rest
Male 84.44 10.69 57.93 125.02
Female 87.64 10.27 55.08 123.18
A Little
Princess
Male 78.92 10.75 52.83 128.69
Female 82.12 10.14 56.21 116.17
Jurassic
Park
Male 80.61 11.02 57.40 129.87
Female 84.37 10.57 54.69 119.87
Poltergeist
Male 79.52 11.08 56.41 129.78
Female 83.22 10.49 57.95 138.69
The Champ
Male 80.16 10.83 55.17 129.23
Female 84.20 10.72 56.89 127.05
Skin Conductance
Level (µs)

Rest
Male 7.64 3.63 1.62 20.92
Female 7.50 3.42 0.35 21.54
A Little
Princess
Male 10.42 4.21 2.41 22.54
Female 9.37 3.83 2.78 23.15
Jurassic Park
Male 10.24 4.19 2.04 21.91
Female 9.30 3.88 2.61 22.19
Poltergeist
Male 10.35 4.14 1.89 21.77
Female 9.44 3.89 2.66 22.29
The Champ
Male 10.47 4.17 2.28 21.41
Female 9.56 3.96 2.82 22.49

Note: Descriptive statistics performed on raw, uncorrected data for ease of interpretation.
All phenotypic and genetic analyses for each clip performed on rest-corrected data, with
skin conductance data further corrected for room temperature. bpm = beats per minute; µs
= microsiemens; SD = standard deviation; min = minimum value; max = maximum value

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

122
Table 4: Bivariate phenotypic correlations between film clip responses stratified by
psychophysiological measure

Fear Sadness
Jurassic
Park
Poltergeist
A Little
Princess
The
Champ
Fear
Jurassic
Park
-- .95** .95** .92**
Poltergeist .72** -- .93** .94**
Sadness
A Little
Princess
.69** .62** -- .92**
The Champ .66** .69** .64** --

Note: Correlations for heart rate presented below the diagonal; correlations for skin
conductance level presented above the diagonal. Skin conductance scores are corrected
for room temperature. Analyses are collapsed across sex as results hold for both male and
female twins. * = p < .05; ** = p < .01.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

123
Table 5: Twin correlations for film clip responses

MZ
Males
DZ
Males
MZ
Females
DZ
Females
DZ
Mixed-Sex

Heart Rate
Fear .25* .18 .37** .22 .04
Sadness .14 .30* .24* .30* .05
Both .21* .29* .33** .30* .05

Skin
Conductance
Level
Fear .38** .18 .34** .31* .09
Sadness .41** .22 .30** .24 .06
Both .40** .21 .32** .28* .07

Notes: MZ = monozygotic; DZ = dizygotic. * = p < .05; ** = p < .01.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

124
Table 6: Univariate sex-limitation model-fitting results for heart rate (HR)

Overall fit
Chi-square
difference test
Parameter estimates
(95% CI)
-2LL df AIC BIC χ
2
df p Δχ
2
df p A C E
Fear HR
1. Saturated
model
5701.56 920 3861.56 -48.43 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
5714.94 928 3858.94 -66.94 25.00 16 .07 -- -- --
M: 0.09
(0.00-0.44)
F: 0.26
(0.00-0.48)
M: 0.17
(0.00-0.40)
F: 0.08
(0.00-0.41)
M: 0.74
(0.56-0.93)
F: 0.66
(0.52-0.82)
2b. Full sex-
limitation
model (rcOS
free)
5714.91 928 3858.91 -66.96 24.96 16 .07 -- -- --
M: 0.04
(0.00-0.44)
F: 0.22
(0.00-0.48)
M: 0.21
(0.00-0.40)
F: 0.12
(0.00-0.42)
M: 0.75
(0.56-0.94)
F: 0.66
(0.52-0.83)
3. Common
effects model
5714.94 929 3856.94 -70.09 25.00 17 .10 0.04 1 .84
M: 0.09
(0.00-0.42)
F: 0.26
(0.00-0.48)
M: 0.17
(0.00-0.40)
F: 0.08
(0.00-0.41)
M: 0.74
(0.57-0.93)
F: 0.66
(0.52-0.82)
4. Scalar
model
5717.68 932 3853.68 -78.18 27.73 20 .12 2.73 3 .44
0.29
(0.00-0.40)
0.00
(0.00-0.23)
0.71
(0.60-0.83)
5. Null model
(ACE)
5720.20 933 3854.20 -80.07 30.25 21 .09 2.52 1 .11
0.29
(0.00-0.40)
0.00
(0.00-0.23)
0.71
(0.60-0.83)
5a. Null
model (AE)
5720.20 934 3852.20 -83.23 30.25 22 .11 0.00 1 1.00
0.29
(0.17-0.40)
--
0.71
(0.60-0.83)
5b. Null model
(CE)
5723.83 934 3855.83 -81.41 33.88 22 .05 3.63 1 .06 --
0.21
(0.11-0.30)
0.79
(0.70-0.89)
5c. Null model 5741.85 935 3871.85 -75.55 51.90 23 <.01 21.65 2 <.01 -- -- 1.00
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

125
(E)
Sad HR
1. Saturated
sex-limitation
model
5586.70 920 3746.70 -105.85 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
5613.56 928 3757.56 -117.64 33.07 16 .01 -- -- --
M: 0.04
(0.00-0.40)
F: 0.22
(0.00-0.43)
M: 0.19
(0.00-0.39)
F: 0.05
(0.00-0.38)
M: 0.77
(0.59-0.99)
F: 0.73
(0.57-0.91)
2b. Full sex-
limitation
model (rcOS
free)
5612.59 929 3756.59 -118.12 32.10 16 .01 -- -- --
M: 0.00
(0.00-0.37)
F: 0.00
(0.00-0.41)
M: 0.23
(0.00-0.39)
F: 0.25
(0.00-0.39)
M: 0.77
(0.60-0.96)
F: 0.75
(0.58-0.91)
3. Common
effects model
5613.56 929 3755.56 -120.79 33.07 17 .01 0.97 1 .32
M: 0.03
(0.00-0.40)
F: 0.22
(0.00-0.43)
M: 0.19
(0.00-0.39)
F: 0.05
(0.00-0.38)
M: 0.78
(0.60-0.99)
F: 0.73
(0.57-0.91)
4. Scalar
model
5616.74 932 3752.74 -128.65 36.25 20 .01 3.18 3 .36
0.15
(0.00-0.34)
0.06
(0.00-0.25)
0.79
(0.66-0.92)
5. Null model
(ACE)
5618.13 933 3752.13 -131.11 37.65 21 .01 1.40 1 .24
0.14
(0.00-0.35)
0.07
(0.00-0.26)
0.79
(0.65-0.92)
5a. Null
model (AE)
5618.35 934 3750.35 -134.15 37.86 22 .02 0.21 1 .65
0.23
(0.10-0.35)
--
0.77
(0.65-0.90)
5b. Null model
(CE)
5618.68 934 3750.68 -133.98 38.19 22 .02 0.54 1 .46 --
0.17
(0.07-0.26)
0.83
(0.74-0.93)
5c. Null model
(E)
5630.12 935 3760.12 -131.42 49.63 23 <.01 11.98 2 <.01 -- -- 1.00
Collapsed HR
1. Saturated
sex-limitation
5535.37 920 3695.37 -131.52 -- -- -- -- -- -- -- -- --
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

126
model
2a. Full sex-
limitation
model (rgOS
free)
5555.46 928 3699.46 -146.69 26.91 16 .04 -- -- --
M: 0.04
(0.00-0.43)
F: 0.28
(0.00-0.49)
M: 0.23
(0.00-0.43)
F: 0.06
(0.00-0.43)
M: 0.72
(0.55-0.91)
F: 0.65
(0.51-0.82)
2b. Full sex-
limitation
model (rcOS
free)
5554.64 928 3698.64 -147.10 26.09 16 .05 -- -- --
M: 0.00
(0.00-0.41)
F: 0.06
(0.00-0.48)
M: 0.28
(0.00-0.43)
F: 0.26
(0.00-0.44)
M: 0.72
(0.56-0.91)
F: 0.67
(0.52-0.83)
3. Common
effects model
5555.46 929 3697.46 -149.84 26.91 17 .06 0.82 1 .37
M: 0.04
(0.00-0.43)
F: 0.28
(0.00-0.49)
M: 0.23
(0.00-0.43)
F: 0.06
(0.00-0.43)
M: 0.72
(0.55-0.91)
F: 0.65
(0.51-0.82)
4. Scalar
model
5559.67 932 3695.67 -157.19 31.12 20 .05 4.21 3 .24
0.27
(0.00-0.40)
0.02
(0.00-0.27)
0.71
(0.60-0.85)
5. Null model
(ACE)
5561.78 933 3695.78 -159.28 33.23 21 .04 2.11 1 .15
0.26
(0.00-0.41)
0.03
(0.00-0.28)
0.71
(0.59-0.85)
5a. Null
model (AE)
5561.81 934 3693.81 -162.42 33.26 22 .06 0.03 1 .86
0.30
(0.17-0.41)
--
0.70
(0.59-0.83)
5b. Null model
(CE)
5563.77 934 3695.77 -161.44 35.22 22 .04 1.99 1 .16 --
0.21
(0.12-0.31)
0.79
(0.69-0.88)
5c. Null model
(E)
5582.47 935 3712.47 -155.24 53.92 23 <.01 20.69 2 <.01 -- -- 1.00

Notes: -2LL = -2*log-likelihood, df = degrees of freedom, AIC = Akaike’s information criterion, BIC=Bayesian information criterion,
χ
2
=difference in chi-square values between compared models, A = additive genetic influences, C = shared environmental influences,
E = non-shared environmental influences, CI = confidence interval, rgOS = genetic correlation for opposite-sex twins, rcOS = shared
environmental correlation for opposite-sex twins. Best-fitting models for each wave (as identified by lowest AIC/BIC values and
highest p-values in the chi-square difference test) are highlighted in bold.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

127
Table 7: Univariate sex-limitation model-fitting results for skin conductance levels (SCL)

Overall fit
Chi-square
difference test
Parameter estimates
(95% CI)
-2LL df AIC BIC χ
2
df p Δχ
2
df p A C E
Fear SCL
1. Saturated
model
3911.85 893 2125.85 -840.68 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
3926.17 901 2124.17 -858.58 25.98 16 .05 -- -- --
M: 0.33
(0.00-0.50)
F: 0.01
(0.00-0.49)
M: 0.01
(0.00-0.34)
F: 0.33
(0.00-0.48)
M: 0.66
(0.50-0.84)
F: 0.66
(0.50-0.81)
2b. Full sex-
limitation
model (rcOS
free)
3926.15 901 2124.15 -858.58 25.96 16 .06 -- -- --
M: 0.35
(0.17-0.50)
F: 0.00
(0.00-0.49)
M: 0.00
(0.00-0.00)
F: 0.34
(0.00-0.48)
M: 0.65
(0.50-0.83)
F: 0.66
(0.50-0.81)
3. Common
effects model
(ACE)
3926.17 902 2122.17 -861.71 25.98 17 .08 0.02 1 .89
M: 0.34
(0.00-0.50)
F: 0.01
(0.00-0.47)
M: 0.01
(0.00-0.34)
F: 0.33
(0.00-0.48)
M: 0.66
(0.50-0.84)
F: 0.66
(0.50-0.81)
3a. Common
effects model
(AE)
3928.65 904 2120.65 -866.73 28.46 19 .08 2.48 2 .29
M: 0.33
(0.15-0.48)
F: 0.36
(0.18-0.51)
--
M: 0.67
(0.52-0.85)
F: 0.64
(0.49-0.82)
3b. Common
effects model
(CE)
3931.81 904 2123.81 -865.15 31.62 19 .03 5.64 2 .06 --
M: 0.21
(0.05-0.36)
F: 0.30
(0.13-0.45)
M: 0.79
(0.64-0.95)
F: 0.70
(0.55-0.87)
3c. Common
effects model
3926.87 904 2118.87 -867.62 26.68 19 .11 0.70 2 .70
M: 0.35
(0.17-0.50)
F: 0.34
(0.19-0.48)
M: 0.65
(0.50-0.83)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

128
(AE in males,
CE in
females)
F: 0.66
(0.52-0.81)
3d. Common
effects model
(E)
3956.76 906 2144.76 -858.94 56.57 21 <.01 30.59 4 <.01 -- -- 1.00
4. Scalar
model
3941.37 905 2131.37 -863.50 41.19 20 <.01 15.21 3 <.01
0.34
(0.01-0.45)
0.00
(0.00-0.26)
0.66
(0.55-0.79)
5. Null model 3967.06 906 2155.06 -853.79 66.87 21 <.01 25.68 1 <.01
0.36
(0.07-0.47)
0.00
(0.00-0.23)
0.64
(0.53-0.76)
Sad SCL
1. Saturated
sex-limitation
model
3935.07 893 2149.07 -829.07 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
3950.84 901 2148.84 -846.24 26.93 16 .04 -- -- --
M: 0.29
(0.00-0.51)
F: 0.05
(0.00-0.45)
M: 0.06
(0.00-0.41)
F: 0.24
(0.00-0.42)
M: 0.64
(0.49-0.82)
F: 0.71
(0.55-0.88)
2b. Full sex-
limitation
model (rcOS
free)
3950.84 901 2148.84 -846.24 26.93 16 .04 -- -- --
M: 0.28
(0.00-0.51)
F: 0.04
(0.00-0.45)
M: 0.07
(0.00-0.42)
F: 0.25
(0.00-0.42)
M: 0.65
(0.49-0.83)
F: 0.72
(0.55-0.88)
3. Common
effects model
(ACE)
3950.84 902 2146.84 -849.37 26.93 17 .06 0.00 1 1.00
M: 0.29
(0.00-0.51)
F: 0.05
(0.00-0.44)
M: 0.06
(0.00-0.41)
F: 0.24
(0.00-0.42)
M: 0.64
(0.49-0.82)
F: 0.71
(0.55-0.88)
3a. Common
effects model
(AE)
3952.58 904 2144.58 -854.77 28.67 19 .07 1.74 2 .42
M: 0.35
(0.17-0.50)
F: 0.29
(0.11-0.45)
--
M: 0.65
(0.50-0.83)
F: 0.71
(0.55-0.89)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

129
3b. Common
effects model
(CE)
3955.70 904 2147.70 -853.21 31.79 19 .03 4.86 2 .09 --
M: 0.26
(0.10-0.41)
F: 0.22
(0.05-0.37)
M: 0.74
(0.59-0.90)
F: 0.78
(0.63-0.95)
3c. Common
effects model
(AE in males,
CE in
females)
3951.27 904 2143.27 -855.42 27.36 19 .10 0.43 2 .81
M: 0.36
(0.19-0.51)
F: 0.28
(0.12-0.42)
M: 0.64
(0.49-0.81)
F: 0.72
(0.58-0.88)
3d. Common
effects model
(E)
3977.34 906 2165.34 -848.65 53.42 21 <.01 26.49 4 <.01 -- -- 1.00
4. Scalar
model
3964.34 905 2154.34 -852.02 40.42 20 <.01 13.49 3 <.01
0.32
(0.00-0.43)
0.00
(0.00-0.27)
0.68
(0.57-0.81)
5. Null model 3992.78 906 2180.78 -840.93 68.87 21 <.01 28.45 1 <.01
0.35
(0.04-0.45)
0.00
(0.00-0.24)
0.65
(0.55-0.78)
Collapsed
SCL

1. Saturated
sex-limitation
model
3906.28 893 2120.28 -843.47 -- -- -- -- -- -- -- -- --
2a. Full sex-
limitation
model (rgOS
free)
3921.90 901 2119.90 -860.71 27.13 16 .04 -- -- --
M: 0.34
(0.00-0.51)
F: 0.00
(0.00-0.47)
M: 0.02
(0.00-0.38)
F: 0.31
(0.00-0.45)
M: 0.64
(0.49-0.82)
F: 0.69
(0.52-0.84)
2b. Full sex-
limitation
model (rcOS
free)
3921.90 901 2119.90 -860.71 27.13 16 .04
M: 0.33
(0.00-0.51)
F: 0.00
(0.00-0.47)
M: 0.02
(0.00-0.41)
F: 0.31
(0.00-0.45)
M: 0.64
(0.49-0.83)
F: 0.69
(0.52-0.84)
3. Common 3921.90 902 2117.90 -863.84 27.13 17 .06 0.00 1 1.00 M: 0.33 M: 0.02 M: 0.64
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

130
effects model
(ACE)
(0.00-0.51)
F: 0.00
(0.00-0.46)
(0.00-0.38)
F: 0.31
(0.00-0.45)
(0.49-0.82)
F: 0.69
(0.53-0.84)
3a. Common
effects model
(AE)
3924.09 904 2116.09 -869.01 29.32 19 .06 2.19 2 .33
M: 0.34
(0.16-0.50)
F: 0.33
(0.14-0.48)
--
M: 0.66
(0.50-0.84)
F: 0.67
(0.52-0.86)
3b. Common
effects model
(CE)
3927.38 904 2119.38 -867.36 32.62 19 .03 5.49 2 .06 --
M: 0.23
(0.08-0.39)
F: 0.26
(0.09-0.41)
M: 0.77
(0.61-0.92)
F: 0.74
(0.59-0.91)
3c. Common
effects model
(AE in males,
CE in
females)
3922.41 904 2114.41 -869.85 27.65 19 .09 0.52 2 .77
M: 0.36
(0.18-0.51)
F: 0.31
(0.16-0.45)
M: 0.64
(0.49-0.82)
F: 0.69
(0.55-0.84)
3d. Common
effects model
(E)
3950.95 906 2138.95 -861.84 56.19 21 <.01 29.06 4 <.01 -- -- 1.00
4. Scalar
model
3936.10 905 2126.10 -866.14 41.33 20 <.01 14.20 3 .01
0.33
(0.00-0.44)
0.00
(0.00-0.27)
0.67
(0.56-0.79)
5. Null model 3963.64 906 2151.64 -855.50 68.88 21 <.01 27.55 1 <.01
0.36
(0.06-0.46)
0.00
(0.00-0.23)
0.64
(0.54-0.76)

Notes: -2LL = -2*log-likelihood, df = degrees of freedom, AIC = Akaike’s information criterion, BIC=Bayesian information criterion,
χ
2
=difference in chi-square values between compared models, A = additive genetic influences, C = shared environmental influences,
E = non-shared environmental influences, CI = confidence interval, rgOS = genetic correlation for opposite-sex twins, rcOS = shared
environmental correlation for opposite-sex twins. Best-fitting models for each wave (as identified by lowest AIC/BIC values and
highest p-values in the chi-square difference test) are highlighted in bold.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

131
Table 8: Bivariate phenotypic correlations between Cooperativeness scale scores and film
clip responses

Cooperativeness Scale Scores
Ages
9-10
Ages
14-15
Ages
16-18
Ages
19-20
Mean
(All Ages)
Male Twins
Heart
Rate
Fear .09 .17* .09 .16* .15**
Sadness .04 .11 .03 .10 .09
Both .07 .15* .07 .14* .13**
Skin
Conductance
Level
Fear .00 .17* .11 .08 .05
Sadness -.01 .14* .08 .07 .04
Both -.01 .15* .10 .08 .04
Female Twins
Heart
Rate
Fear -.01 .04 -.08 -.01 .00
Sadness -.06 .00 -.15* -.07 -.05
Both -.04 .02 -.12* -.04 -.03
Skin
Conductance
Level
Fear .03 -.07 .06 -.05 .02
Sadness .03 -.09 .05 -.03 .01
Both .03 -.08 .06 -.04 .02

Notes: Cooperativeness scale scores are untransformed. Film clip responses have been
corrected as previously described, with all skin conductance scores further corrected for
room temperature. * = p < .05; ** = p < .01.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

132
Table 9: Bivariate model-fitting results between film clip responses and Cooperativeness

Overall fit
Chi-square
difference test
Parameter estimates
(95% CI)
-2LL df AIC BIC χ
2
df p Δχ
2
df p A D E
COOP
AGES 14-15

Fear HR
in Males

Saturated 1313.27 486 341.27 -628.42 -- -- -- -- -- -- -- -- --
Bivariate ADE 1331.99 497 337.99 -648.14 18.73 11 .07 -- -- --
HR: 0.24
(0.00-0.41)
COOP: 0.03
(0.00-0.45)
r
g
: 1.00
(-1.00-1.00)
HR: 0.00
(0.00-0.40)
COOP: 0.26
(0.00-0.53)
r
d
: 1.00
(-1.00-1.00)
HR: 0.76
(0.59-0.95)
COOP:
0.71
(0.46-1.00)
r
e
: 0.06
(-0.22-0.33)
Bivariate ADE
(drop D
HR
)
1332.00 499 334.00 -653.42 18.74 13 .13 0.01 2 1.00
HR: 0.24
(0.06-0.41)
COOP: 0.04
(0.00-0.45)
r
g
: 1.00
(-1.00-1.00)
HR: --
--
COOP: 0.25
(0.00-0.52)
r
d
: --
--
HR: 0.76
(0.59-0.94)
COOP:
0.71
(0.46-1.00)
r
e
: 0.06
(-0.21-0.32)
Bivariate ADE
(drop D
HR,
r
e
)
1332.19 500 332.19 -655.97 18.92 14 .17 0.17 3 .98
HR: 0.24
(0.06-0.41)
COOP: 0.06
(0.00-0.45)
r
g
: 1.00
(-1.00-1.00)
HR: --
--
COOP: 0.23
(0.00-0.50)
r
d
: --
--
HR: 0.76
(0.59-0.94)
COOP:
0.71
(0.46-0.98)
r
e
: --
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

133
--
Bivariate AE 1333.10 500 333.10 -655.52 19.84 14 .14 1.11 3 .77
HR: 0.24
(0.05-0.41)
COOP: 0.22
(0.00-0.49)
r
g
: 0.51
(-1.00-1.00)
HR: --
--
COOP: --
--
r
d
: --
--
HR: 0.76
(0.59-0.95)
COOP:
0.78
(0.51-1.00)
r
e
: 0.03
(-0.24-0.31)
Bivariate AE
(drop r
e
)
1333.16 501 331.16 -658.13 19.89 15 .18 1.26 4 .87
HR: 0.24
(0.05-0.41)
COOP:
0.22
(0.01-0.49)
r
g
: 0.59
(-0.02-1.00)
HR: --
--
COOP: --
--
r
d
: --
--
HR: 0.76
(0.59-0.95)
COOP:
0.78
(0.51-0.99)
r
e
: --
--
Bivariate E 1342.00 503 336.00 -659.00 28.73 17 .04 8.89 3 .03
HR: --
--
COOP: --
--
r
g
: --
--
HR: --
--
COOP: --
--
r
d
: --
--
HR: 1.00
(1.00-1.00)
COOP:
1.00
(1.00-1.00)
r
e
: 0.15
(0.00-0.30)
Fear SCL
in Males

Saturated 1370.10 486 398.09 -592.52 -- -- -- -- -- -- -- -- --
Bivariate ADE 1392.60 497 398.60 -610.19 22.50 11 .02 -- -- --
SCL: 0.24
(0.00-0.51)
COOP: 0.00
(0.00-0.42)
r
g
: 1.00
(-1.00-1.00)
SCL: 0.13
(0.00-0.52)
COOP: 0.30
(0.00-0.52)
r
d
: 1.00
(-0.44-1.00)
SCL: 0.63
(0.48-0.80)
COOP:
0.70
(0.48-0.96)
r
e
: 0.14
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

134
(-0.13-0.39)
Bivariate ADE
(drop D
SCL
)
1393.24 499 395.24 -615.13 23.14 13 .04 0.64 2 .73
SCL: 0.37
(0.19-0.52)
COOP: 0.09
(0.00-0.44)
r
g
: 1.00
(0.00-1.00)
SCL: --
--
COOP: 0.20
(0.00-0.41)
r
d
: --
--
SCL: 0.63
(0.48-0.81)
COOP:
0.72
(0.50-0.98)
r
e
: 0.18
(-0.09-0.41)
Bivariate ADE
(drop D
SCL,
r
e
)
1395.00 500 395.00 -616.88 24.90 14 .04 2.40 3 .49
SCL: 0.38
(0.21-0.53)
COOP: 0.17
(0.04-0.47)
r
g
: 1.00
(-1.00-1.00)
SCL: --
--
COOP: 0.14
(0.00-0.38)
r
d
: --
--
SCL: 0.62
(0.47-0.79)
COOP:
0.70
(0.48-0.93)
r
e
: --
--
Bivariate AE 1394.24 500 394.24 -617.25 24.15 14 .04 1.65 3 .65
SCL: 0.37
(0.19-0.52)
COOP: 0.24
(0.00-0.47)
r
g
: 0.62
(-0.01-1.00)
SCL: --
--
COOP: --
--
r
d
: --
--
SCL: 0.63
(0.48-0.81)
COOP:
0.76
(0.53-1.00)
r
e
: 0.16
(-0.11-0.41)
Bivariate AE
(drop r
e
)
1395.67 501 393.67 -619.17 25.58 15 .04 3.08 4 .54
SCL: 0.38
(0.20-0.52)
COOP:
0.27
(0.06-0.49)
r
g
: 0.82
(0.41-1.00)
SCL: --
--
COOP: --
--
r
d
: --
--
SCL: 0.62
(0.48-0.80)
COOP:
0.73
(0.51-0.94)
r
e
: --
--
Bivariate E 1411.49 503 405.49 -616.51 41.40 17 <.01 18.90 6 <.01
SCL: --
--
SCL: --
--
SCL: 1.00
(1.00-1.00)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

135
COOP: --
--
r
g
: --
--
COOP: --
--
r
d
: --
--
COOP:
1.00
(1.00-1.00)
r
e
: 1.00
(1.00-1.00)
Sad SCL
in Males

Saturated 1370.80 486 398.80 -592.17 -- -- -- -- -- -- -- -- --
Bivariate ADE 1393.60 497 399.60 -609.69 22.81 11 .02 -- -- --
SCL: 0.33
(0.00-0.52)
COOP: 0.00
(0.00-0.42)
r
g
: 1.00
(-1.00-1.00)
SCL: 0.06
(0.00-0.52)
COOP: 0.29
(0.00-0.51)
r
d
: 1.00
(-1.00-1.00)
SCL: 0.61
(0.47-0.79)
COOP:
0.71
(0.49-0.98)
r
e
: 0.19
(-0.07-0.42)
Bivariate ADE
(drop D
SCL
)
1393.90 499 395.90 -614.80 23.10 13 .04 0.29 2 .87
SCL: 0.38
(0.21-0.53)
COOP: 0.05
(0.00-0.43)
r
g
: 1.00
(-1.00-1.00)
SCL: --
--
COOP: 0.23
(0.00-0.44)
r
d
: --
--
SCL: 0.62
(0.47-0.79)
COOP:
0.72
(0.50-0.98)
r
e
: 0.21
(-0.04-0.43)
Bivariate ADE
(drop D
SCL,
r
e
)
1396.52 500 396.52 -616.11 25.73 14 .03 2.92 3 .40
SCL: 0.40
(0.23-0.54)
COOP: 0.12
(0.02-0.46)
r
g
: 1.00
(0.34-1.00)
SCL: --
--
COOP: 0.18
(0.00-0.41)
r
d
: --
--
SCL: 0.60
(0.46-0.77)
COOP:
0.70
(0.49-0.95)
r
e
: --
--
Bivariate AE 1395.07 500 395.07 -616.84 24.27 14 .04 1.46 3 .69
SCL: 0.38
(0.21-0.53)
SCL: --
--
SCL: 0.62
(0.47-0.79)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

136
COOP: 0.23
(0.00-0.47)
r
g
: 0.46
(-1.00-1.00)
COOP: --
--
r
d
: --
--
COOP:
0.77
(0.53-1.00)
r
e
: 0.20
(-0.06-0.44)
Bivariate AE
(drop r
e
)
1397.41 501 395.41 -618.30 26.61 15 .03 3.80 4 .43
SCL: 0.39
(0.23-0.54)
COOP:
0.26
(0.04-0.48)
r
g
: 0.71
(0.29-1.00)
SCL: --
--
COOP: --
--
r
d
: --
--
SCL: 0.61
(0.46-0.77)
COOP:
0.74
(0.52-0.96)
r
e
: --
--
Bivariate E 1414.62 503 408.62 -614.95 43.82 17 <.01 21.01 6 <.01
SCL: --
--
COOP: --
--
r
g
: --
--
SCL: --
--
COOP: --
--
r
d
: --
--
SCL: 1.00
(1.00-1.00)
COOP:
1.00
(1.00-1.00)
r
e
:
()
COOP.
AGES 16-18

Sad HR
in Females

Saturated 1538.03 563 412.03 -741.51 -- -- -- -- -- -- -- -- --
Bivariate ADE 1548.08 574 400.08 -766.00 10.05 11 .53 -- -- --
HR: 0.25
(0.00-0.41)
COOP: 0.30
(0.00-0.51)
r
g
: -0.26
(-1.00-1.00)
HR: 0.00
(0.00-0.38)
COOP: 0.00
(0.00-0.48)
r
d
: -0.75
(-1.00-1.00)
HR: 0.75
(0.59-0.92)
COOP:
0.70
(0.49-0.95)
r
e
: -0.09
(-0.29-0.13)
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

137
Bivariate ADE
(drop D
HR
)
1548.08 576 396.07 -771.36 10.05 13 .69 0.00 2 1.00
HR: 0.25
(0.08-0.41)
COOP: 0.30
(0.00-0.51)
r
g
: -0.26
(-1.00-1.00)
HR: --
--
COOP: 0.00
(0.00-0.47)
r
d
: --
--
HR: 0.75
(0.59-0.92)
COOP:
0.70
(0.49-0.95)
r
e
: -0.09
(-0.29-0.13)
Bivariate ADE
(drop D
HR,
r
e
)
1548.69 577 394.69 -773.74 10.66 14 .71 0.61 3 .89
HR: 0.25
(0.09-0.41)
COOP: 0.30
(0.00-0.51)
r
g
: -0.41
(-1.00-0.01)
HR: --
--
COOP: 0.00
(0.00-0.45)
r
d
: --
--
HR: 0.75
(0.59-0.91)
COOP:
0.70
(0.49-0.95)
r
e
: --
--
Bivariate AE 1548.08 577 394.08 -774.05 10.05 14 .76 0.00 3 1.00
HR: 0.25
(0.08-0.41)
COOP: 0.30
(0.05-0.51)
r
g
: -0.26
(-0.96-0.37)
HR: --
--
COOP: --
--
r
d
: --
--
HR: 0.75
(0.59-0.92)
COOP:
0.70
(0.49-0.95)
r
e
: -0.09
(-0.29-0.13)
Bivariate AE
(drop r
e
)
1548.69 578 392.69 -776.42 10.66 15 .78 0.61 4 .96
HR: 0.25
(0.09-0.41)
COOP:
0.30
(0.05-0.51)
r
g
: -0.41
(-1.00-0.01)
HR: --
--
COOP: --
--
r
d
: --
--
HR: 0.75
(0.59-0.91)
COOP:
0.70
(0.49-0.95)
r
e
: --
--
Bivariate E 1561.97 580 401.97 -775.15 23.94 17 .12 13.89 3 <.01
HR: --
--
COOP: --
HR: --
--
COOP: --
HR: 1.00
(1.00-1.00)
COOP:
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

138
--
r
g
: --
--
--
r
d
: --
--
1.00
(1.00-1.00)
r
e
:
()
COOP.
AGES 19-20

Fear HR
in Males

Saturated 1387.59 506 375.59 -644.14 -- -- -- -- -- -- -- -- --
Bivariate ADE 1394.97 517 360.97 -669.53 7.39 11 .77 -- -- --
HR: 0.14
(0.00-0.41)
COOP: 0.18
(0.00-0.59)
r
g
: -0.51
(-1.00-1.00)
HR: 0.11
(0.00-0.40)
COOP: 0.22
(0.00-0.59)
r
d
: 1.00
(-1.00-1.00)
HR: 0.75
(0.58-0.94)
COOP:
0.60
(0.40-0.87)
r
e
: 0.05
(-0.22-0.31)
Bivariate ADE
(drop D
HR
)
1395.08 519 357.08 -674.76 7.49 13 .88 0.10 2 .95
HR: 0.25
(0.06-0.41)
COOP: 0.39
(0.00-0.60)
r
g
: 0.20
(-1.00-1.00)
HR: --
--
COOP: 0.00
(0.00-0.58)
r
d
: --
--
HR: 0.75
(0.59-0.94)
COOP:
0.61
(0.40-0.87)
r
e
: 0.05
(-0.21-0.31)
Bivariate ADE
(drop D
HR,
r
e
)
1395.23 520 355.23 -677.34 7.64 14 .91 0.25 3 .97
HR: 0.25
(0.06-0.41)
COOP: 0.39
(0.00-0.59)
r
g
: 0.28
(-1.00-1.00)
HR: --
--
COOP: 0.00
(0.00-0.57)
r
d
: --
--
HR: 0.75
(0.59-0.94)
COOP:
0.61
(0.40-0.87)
r
e
: --
--
DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS

139
Bivariate AE 1395.08 520 355.08 -677.41 7.49 14 .91 0.10 3 .99
HR: 0.25
(0.06-0.41)
COOP: 0.39
(0.13-0.60)
r
g
: 0.20
(-0.44-0.92)
HR: --
--
COOP: --
--
r
d
: --
--
HR: 0.75
(0.59-0.94)
COOP:
0.61
(0.40-0.87)
r
e
: 0.05
(-0.21-0.31)
Bivariate AE
(drop r
e
)
1395.23 521 353.23 -679.98 7.64 15 .94 0.25 4 .99.
HR: 0.25
(0.06-0.41)
COOP:
0.39
(0.12-0.59)
r
g
: 0.28
(-0.15-0.80)
HR: --
--
COOP: --
--
r
d
: --
--
HR: 0.75
(0.59-0.94)
COOP:
0.61
(0.40-0.88)
r
e
: --
--
Bivariate E 1409.33 523 363.33 -678.22 21.75 17 .20 14.36 6 .03
HR: --
--
COOP: --
--
r
g
: --
--
HR: --
--
COOP: --
--
r
d
: --
--
HR: 1.00
(1.00-1.00)
COOP:
1.00
(1.00-1.00)
r
e
:
()

Notes: HR = heart rate, SCL = skin conductance level, COOP = Cooperativeness. -2LL = -2*log-likelihood, df = degrees of freedom, AIC =
Akaike’s information criterion, BIC=Bayesian information criterion, χ
2
=difference in chi-square values between compared models, A = additive
genetic influences, D = non-additive (dominant) genetic influences, E = non-shared environmental influences, CI = confidence interval, r
g
=
additive genetic correlation, r
d
= non-additive (dominant) genetic correlation, r
e
= non-shared environmental correlation. Best-fitting models for
each Cooperativeness/physiology pair (as identified by lowest AIC/BIC values and highest p-values in the chi-square difference test) are
highlighted in bold.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS 140

Figure 1: Time-course of (uncorrected) cardiac responses during the film clips task

Note: LP = A Little Princess, JP = Jurassic Park, PG = Poltergeist, CH = The Champ.
Vertical gray bars indicate the beginning and end of the target period for each clip.
Horizontal red bars indicate the mean response values for males and females for each
target period.

DEVELOPMENT AND PHYSIOLOGY OF PROSOCIALITY IN TWINS 141

Figure 2: Time-course of (uncorrected) electrodermal responses during the film clips task

Note: LP = A Little Princess, JP = Jurassic Park, PG = Poltergeist, CH = The Champ.
Vertical gray bars indicate the beginning and end of the target period for each clip.
Horizontal red bars indicate the mean response values for males and females for each
target period.

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Psychophysiological reactivity to stress, psychopathic traits, and developmental trajectories of antisocial behavior: a prospective, longitudinal study

PDF

Developmental etiology of reactive and proactive aggression, electrodermal reactivity and their relationships: a twin study

PDF

Biometric models of psychopathic traits in adolescence: a comparison of item-level and sum-score approaches

PDF

Patterns of EEG spectral power in 9-10 year old twins and their relationships with aggressive and nonaggressive antisocial behavior in childhood and adolescence

PDF

Different genetic and environmental structures for the overlap of three antisocial behavior factors with alcohol initiation

PDF

The effects of acceptance and commitment therapy-based exercises on eating behaviors in a laboratory setting

PDF

Psychometric and biometric evaluation of psychopathic personality traits in 9 and 10 year old twins

PDF

Multivariate biometric modeling among multiple traits across different raters in child externalizing behavior studies

PDF

A comparison of standard and alternative measurement models for dealing with skewed data with applications to longitudinal data on the child psychopathy scale

PDF

Lifestyle-related exposures and diseases in twins

PDF

Are life events differentially associated with dementia risk by gender? A twin study

PDF

On second thought, people choose to be prosocial

PDF

Evaluating the effects of virtual reality and acceptance and commitment therapy on music performance anxiety

PDF

The development of electrodermal fear conditioning from ages 3 to 8 years and relationships with antisocial behavior at age 8 years

PDF

Clinical decision analysis for effective supervision: a review of the supervisor-supervisee relationship

PDF

Family aggression, prosocial friends, and the risk of dating and friend victimization in late adolescence

PDF

Let’s do something different: evaluation of lag schedules of reinforcement to increase flexibility and promote generalization in children with autism

PDF

K-12 college and career readiness guidance: an improvement study

PDF

Using acceptance and commitment training to enhance the effectiveness of behavioral skills training

PDF

Heritability and longitudinal stability of impulsivity in adolescence

Asset Metadata

Creator Berntsen, Leslie (author)

Core Title Nice by nature? A twin study of the development and physiology of prosocial personality

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 10/18/2018

Defense Date 08/22/2018

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag behavior genetics,OAI-PMH Harvest,personality development,prosocial personality,psychophysiology

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Baker, Laura (committee chair), Beam, Christopher (committee member), Dawson, Michael (committee member), Yang, Yaling (committee member)

Creator Email leslie.berntsen@gmail.com,leslie.berntsen@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-84508

Unique identifier UC11675710

Identifier etd-BerntsenLe-6882.pdf (filename),usctheses-c89-84508 (legacy record id)

Legacy Identifier etd-BerntsenLe-6882.pdf

Dmrecord 84508

Document Type Dissertation

Format application/pdf (imt)

Rights Berntsen, Leslie

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA