ArticlePDF Available

Pulling Back the Curtain on Heritability Studies: Biosocial Criminology in the Postgenomic Era


Unfortunately, the nature versus nurture debate continues in criminology. Over the past five years there has been a surge of studies in criminology estimating the heritability of crime and related outcomes, which invariably report sizeable heritability estimates (~50%) and minimal to non-existent effects of the so-called shared environment. Reports of such high heritabilities for complex social behaviors such as crime are surprising, and findings indicating minimal shared environmental influences (usually interpreted to include parenting and community factors) seem implausible given decades of criminological research demonstrating their importance. Importantly, however, the models on which these estimates are based have fatal flaws for complex social behaviors such as crime. Moreover, the very goal of heritability studies—partitioning the effects of nature versus nurture—is misguided given the bidirectional, interactional relationship between genes, cells, organisms, and environments. The present study provides a critique of heritability study methods and assumptions to illuminate the dubious foundations of heritability estimates and questions the rationale and utility of partitioning genetic and environmental effects. After critiquing the major models, namely twin and adoption studies, from both the classical and recently emergent postgenomic paradigms, we call for an end to heritability studies given their flaws and their rather limited value for advancing knowledge on the etiology of crime. We then present what we perceive to be a more useful biosocial research agenda that is consonant with and informed by recent advances in our understanding of gene function and developmental plasticity. We conclude by noting that at the current state of knowledge social explanations of crime are not undermined by genetic or biological findings, but rather the more we learn about genes and biology, the more consequential the environment becomes.
School of Criminology and Criminal Justice, Arizona State University
KEYWORDS: behavioral genetics, heritability, twin study, epigenetics, life course,
Unfortunately, the nature-versus-nurture debate continues in criminology. Over the
past 5 years, the number of heritability studies in criminology has surged. These studies
invariably report sizeable heritability estimates (50 percent) and minimal effects of
the so-called shared environment for crime and related outcomes. Reports of such high
heritabilities for such complex social behaviors are surprising, and findings indicat-
ing negligible shared environmental influences (usually interpreted to include parent-
ing and community factors) seem implausible given extensive criminological research
demonstrating their significance. Importantly, however, the models on which these es-
timates are based have fatal flaws for complex social behaviors such as crime. More-
over, the goal of heritability studies—partitioning the effects of nature and nurture—is
misguided given the bidirectional, interactional relationship among genes, cells, organ-
isms, and environments. This study provides a critique of heritability study methods
and assumptions to illuminate the dubious foundations of heritability estimates and
questions the rationale and utility of partitioning genetic and environmental effects.
After critiquing the major models, we call for an end to heritability studies. We then
present what we perceive to be a more useful biosocial research agenda that is conso-
nant with and informed by recent advances in our understanding of gene function and
developmental plasticity.
Questions about nature versus nurture have been a perennial topic of debate in the
social sciences. Since the 1970s, these questions have been addressed by a field of study
known as behavioral genetics. A major focus of behavioral genetics research has been to
partition the variation in an outcome of interest into a proportion caused by genes (her-
itability) and a proportion caused by the environment (e.g., DiLalla, 2004; Plomin et al.,
2012). These heritability studies (also called quantitative genetics, nonmolecular genetics,
or biometrics) have compared phenotypes (observed characteristics of individuals) within
Additional supporting information can be found in the listing for this article in the
Wiley Online Library at
The authors would like to thank Steven Beach, Kara Hannula, Tanja Link, Travis Pratt, four
anonymous reviewers, and D. Wayne Osgood for valuable comments on earlier drafts of the ar-
ticle. The arguments presented in the article are entirely those of the authors and do not reflect
the views of those who provided feedback. Direct all correspondence to Callie H. Burt, School of
Criminology and Criminal Justice, Arizona State University, 411 N. Central Ave, Ste. 600, Phoenix,
AZ 85004 (e-mail:
C2014 American Society of Criminology doi: 10.1111/1745-9125.12036
CRIMINOLOGY Volume 52 Number 2 223–262 2014 223
and between families that vary in genetic relationships to estimate the influence of genes.
The general conclusion emerging from heritability studies is that variance in almost ev-
ery human characteristic—including preferences and social behaviors—is shaped signifi-
cantly by genetic influences (Plomin et al., 2012; Turkheimer, 2000). For example, studies
have shown substantial heritability (usually between 40 and 60 percent) of everything
from time spent watching television (Plomin et al., 1990), breastfeeding (Colodro-Conde,
Sanchez-Romera, and Ordonana, 2013), and breakfast eating patterns (Keski-Rahkonen
et al., 2004) to political ideology and party affiliation (Alford, Funk, and Hibbing, 2005)
and delinquency (Wright et al., 2008).
Since 2008, behavioral genetic studies by criminologists investigating the heritability of
criminal behavior and factors associated with crime have been published at a rapid and
seemingly increasing pace. Indeed, every few months new studies estimate the heritabil-
ity of some aspect of criminality, such as self-reported delinquency (e.g., Boisvert et al.,
2012; Wright et al., 2008), arrests (Beaver et al., 2012), self-control (e.g., Beaver et al.,
2008, 2009), and gang membership (Barnes, Boutwell, and Fox, 2012). Notably, behav-
ioral geneticists have been estimating the heritability of antisocial behavior, delinquency,
and criminal convictions for years prior to this recent surge (e.g., Mednick, Gabrielli, and
Hutchings, 1984; Rowe and Osgood, 1984; see Moffitt, 2005, for a review). These studies
have revealed ostensibly that genetic factors explain a substantial portion—between 30
and 90 percent of the variance—of every examined crime-related phenotype with shared
environmental factors playing a minimal role (often reported to be 0 percent).
We are surprised that these somewhat astonishing findings reported in recent studies,
such as the reports of more than 50 percent heritability for such complex social behav-
iors as crime and victimization, have not generated more critical attention in criminol-
ogy. We also are perplexed by the lack of response to the heritability study finding that
so-called shared environmental factors play a minor role in explaining variation in crime-
related phenotypes (e.g., Barnes, Boutwell, and Fox, 2012; Beaver et al., 2008; Beaver,
Ferguson, and Lynn-Whaley, 2010; Boisvert, Wright, et al., 2013). Indeed, the conclu-
sion from many of these heritability studies that little—if any—of the variance in criminal
behavior is due to shared environments, often interpreted to include parenting and com-
munity factors, contradicts a wealth of research conducted during the past century as well
as the major theories of crime. As the renowned psychiatrist and behavioral genetics prac-
titioner Michael Rutter (2006: 11) noted, “[The] sweeping assertions on the irrelevance of
the family environment are not supported by research evidence. It is quite striking that be-
havioral genetics reviews usually totally ignore the findings on environmental influences.
It is almost as if research by non-geneticists is irrelevant.”
The lack of critical attention to heritability studies in criminology is even more conspic-
uous given their known limitations. Since at least the early 1930s, scholars—including
prominent geneticists, neuroscientists, and molecular biologists—have been warning
about the fallibility of heritability studies in human populations, especially for complex so-
cial behaviors such as crime (e.g., Joseph, 2004; Kamin, 1974; Lewontin, Rose, and Kamin,
1984; Wahlsten, 1990; Wilson, 1934). As we discuss in this article, these methodological
problems are not merely a sampling or measurement issue that can be corrected with var-
ious model adjustments but are inherent to the method itself. Remarkably, as we discuss,
many of the recent criminology studies have failed to mention crucial assumptions and
technical limitations, and even fewer have discussed the implications of their violation,
especially for assumptions whose violations bias toward heritability.
It is not merely the techniques of heritability studies that are suspect, but also the ba-
sic conceptual framework of heritability studies is unsound (e.g., Charney, 2008; Crusio,
2012; Turkheimer, 2011). Heritability studies rest on a model of gene function that views
genetic effects as independent and separable from the environmental context in which
they operate. Genes, however, do not work this way. Genes are not a self-activating
code that can be understood apart from environmental inputs but are only one part of an
interactive, developmental biopsychosocial system. As we will discuss, genes and the en-
vironment are not separate analytical entities, and thus, it is biologically nonsensical to
attempt to partition genetic from environmental influences on phenotypes (e.g., Gottlieb,
2001; Greenberg, 2011).
Although the problems of quantitative genetic methods have been discussed elsewhere
(see, e.g., Charney, 2008, 2012; Joseph, 2004, 2006; Lewontin, Rose, and Kamin, 1984;
Rutter, 2006), and in criminology several decades ago or more (e.g., Walters and White,
1989), as yet there has been no critical response to the recent profusion of heritability
studies in criminology. Given the significant advances in our understanding of gene func-
tion and the profound theoretical and policy implications of this work, such a response is
sorely needed. It is our objective in this article to provide such a response. Notably, most
of the arguments in this article are not original but are those of prominent scientists, many
of whom we cite, whose criticisms have been largely unheeded by the criminological com-
munity in recent years. We hope to renew a dialogue in criminology about heritability
studies and stimulate what we view as a much-needed debate about the utility of heri-
tability studies for crime and related phenotypes.
The objectives of this article are twofold. First, we critique heritability study methods
and assumptions so that the criminological community can be more informed consumers
of these findings. Our critique is grounded in both methodological and conceptual issues.
Regarding the methodological critique, we argue that heritability studies are seriously
flawed and, thus, do not provide useful estimates of genetic influences on criminal pheno-
types. As we will show, these technical flaws have the effect of biasing estimates toward
inflating heritability and underestimating shared environmental effects. Second, we ar-
gue that, methodological problems aside, partitioning individual differences into genetic
versus environmental influences is a misguided endeavor in the first place. Drawing on
a variety of sources, including the arguments of prominent behavioral geneticists (e.g.,
Rutter, 2006; Turkheimer, 2011), we call for an end to heritability studies in criminology
and recognition of the problematic nature of existing heritability estimates for criminal
To be clear at the outset, it is not the position of this article that genes do not con-
tribute to individual differences in behavior. Criminal behavior, like all other behavior,
results from a combination of factors, including environmental and genetic ones. Any
claim to the contrary is patently false. We do not wish to move the field toward extreme
cultural or social determinism. Instead, we attempt to tackle these issues and put them
in a new perspective, not by denying the role of genes or other biological factors but by
recognizing the complexity of the biopsychosocial system. Heritability studies do not re-
solve the outworn nature-versus-nurture debate; they promote it. Successfully leveraging
the advances of the genomic era in the new era of postgenomics requires that we move
beyond heritability, transcending the outdated question of how much significance to at-
tribute to genetics versus the environment in the development of particular behaviors and
traits (Lickliter, 2009).
Thus, in the final sections of the article, we challenge scholars interested in biosocial
work in criminology to move beyond heritability toward research grounded in the reality
that genetic effects on human behavior are meaningful only when considered in com-
bination with environmental influences. Notably, many criminologists who are involved
in heritability studies also engage in biosocial research consistent with the postgenomic
biosocial agenda encouraged in this article. With this work, these scholars have advanced
biosocial criminology and our own understanding of biosocial mechanisms. This article is
not a criticism of scientists but a critique of a particular method.
Our argument unfolds in four sections. First, after a brief primer on genetics, we dis-
cuss the concept of heritability—a concept that is often and easily misunderstood—and
its quantified form, the heritability coefficient, as well as the foundational assumptions of
heritability studies. Next, we turn our attention to the various methods that behavioral
geneticists use to estimate heritability. We begin with the most common design, the twin
study. The fundamental assumptions of this approach are detailed, followed by a discus-
sion of the way that using this method to study criminal and antisocial behavior violates
crucial assumptions. Then, we discuss the other prominent method in quantitative genet-
ics used to estimate the heritability of criminal phenotypes: adoption studies. We also
briefly discuss the third heritability model, twins reared apart studies, which is less rel-
evant to criminology given its lack of use in this domain but is still cited as convergent
Although evidence from the different methods are used to “provide convergent find-
ings,” given that “each of the primary designs used by behavioral geneticists has its own
Achilles heel(s)” (Moffitt, 2005: 57), we show that all of these models are biased toward
inflating heritability and underestimating shared environmental influences. After this dis-
cussion of the methodological limitations in heritability studies, we turn our attention to
the misguided idea that we can partition genetic and environmental influences on com-
plex phenotypes. Finally, having presented what we believe to be compelling arguments
for abandoning heritability studies in criminology, we briefly present examples of bioso-
cial research that are consonant with and informed by the postgenomic paradigm that
views genes and the environment as inextricably linked and in a dynamic relationship.
Genes are segments of DNA coded for the production of RNA molecules and specific
proteins organized on chromosomes. Individuals possess two copies of each gene in all
cells, with a couple of exceptions; one copy comes from the egg and one from the sperm.
Each copy of the gene is known as an allele of that gene. For the overwhelming majority
of genes, only one allele exists. However, some genes vary across the population, such
that there are at least two “versions” of the allele (Beaver, 2008). Polymorphisms are
versions that occur in more than 1 percent of the population, whereas versions that occur
in less than 1 percent of the population are known as mutations. The term “genotype”
is commonly used to refer to whether an individual possesses a particular allele or the
alleles for a particular gene, and the term “genome” refers to an individual’s entire DNA
sequence (Charney, 2012; Schaffner, 2006).
Although in the mid-twentieth century it was thought that genes code for proteins in a
straightforward manner, we now know this is not the case. DNA is not a self-activating
code. Foreshadowing our later discussion, the process of gene activation, often referred
to as “gene expression,” is a multistep, multifactorial process. Genes are now understood
to operate more as catalysts; they exist in cells that have many components and “function
in a manner akin to chemicals in a test tube. Everything in the test tube affects everything
else in the test tube; so too, everything in the cell affects genes” (Greenberg, 2011: 177).
Importantly, the chemistry of the cell is influenced by environmental factors, such as a
person’s diet, exercise, exposure to various elements, and the like. Thus, environmental
influences are important determinants of gene activation or expression (Jablonka and
Lamb, 2006).
Moreover, evidence suggests that the genome itself is dynamic. DNA is composed of
transposable elements—deletions, insertions, and rearrangements—that can alter the ge-
netic sequence (the “genetic code”; Charney, 2012). Indeed, one of the more remarkable
findings from the Human Genome Project is that approximately 45 to 48 percent of the
genome is composed of such transposable elements (Lander et al., 2001). Evidence sug-
gests that changes in DNA sequencing continue throughout life, can be inherited, and
seem to be environmentally responsive (see Charney, 2012). Although there is much that
scientists still do not know, it is increasingly clear that partitioning genetic and environ-
mental influences is biologically nonsensical. Yet, attempts to do just that continue in
The heritability concept was originally designed for use in agricultural research to pre-
dict the outcomes of controlled breeding and thereby to assist animal and plant breeders
in selecting for desirable traits. Since the 1970s, this concept has been promoted as a
“nature–nurture ratio” of the relative influence of heredity (via genes) and environmen-
tal experiences on particular traits. However, this concept is frequently misunderstood
(Joseph, 2006).
Heritability was defined by Wahlsten (1990: 244) as “the proportion of variance in a
measure of behaviour or other phenotype in a breeding population that is attributable
to genetic variation” and by Plomin et al. (2012: 87) as “the proportion of phenotypic
variance that is accounted for by genetic differences among individuals.” The heritability
coefficient is the numeric index of heritability ranging from .0 (no genetic contribution)
to 1.0 (complete heritability). The estimation of this parameter involves a model based
on Mendelian principles as well as on several assumptions (discussed later).
Notably, heritability refers to the genetic contribution to trait variability in a popula-
tion, not to the import of genetic factors as they influence the individual (Joseph, 2004;
Plomin et al., 2012). Heritability estimates are often wrongly interpreted by laypersons
(and some scholars) as the proportion of a trait that is caused by genetic influence rather
than as the proportion of population variance in a trait. The methods of Mendelian genet-
ics are responsive only to the slight portion of genes that are polymorphic and make us dif-
ferent; they do not allow for conclusions about the role of heredity in general (Wahlsten,
1990). As Joseph (2004: 138–9) noted, “This leads to a paradoxical situation in which a
trait could be 100% inherited, yet have a heritability of zero—human beings having two
eyes for example.” Genetics, of course, explains humans’ “eyedness”; we have two eyes
because of our genetic endowment. Because everyone (or nearly everyone) who has fewer
than two eyes is that way as a result of life experiences, the heritability of “eyedness” is
zero (see also Lewontin, Rose, and Kamin, 1984).
Although in the past some behavioral geneticists have used genetic findings as
a basis for arguing that heritability estimates within groups can be used to ex-
plain differences between groups (e.g., Herrnstein and Murray, 1994; Jensen, 1969),
these arguments are now widely understood to be fallacious. Heritability estimates
are time and population specific. As such, findings of genetic influences on differ-
ences within populations do not extrapolate into explaining differences observed in
traits across populations (Plomin et al., 1997). This aspect of heritability is wonder-
fully illustrated by Lewontin, Rose, and Kamin (1984). Suppose one takes two hand-
fuls of heterogeneous corn seed and plants one handful on a field of nutrient-rich
soil and the other in a field of nutrient-poor soil. When the seeds have grown, we
can observe that there is variation within fields in plant height as well as variation
between fields, specifically with lower plant height for the poor-quality-soil field. Not
allowing any environmental variation within the respective fields (equal light, water,
etc.), differences in the resulting plant height within the fields are totally the result of
genetic factors (heritability =1.0). Differences in plant height between the two fields,
however, are completely the result of environmental factors, specifically, soil quality
(heritability =0). This example demonstrates that even for trait variance that is entirely
heritable within a population, the cross-population variance may be completely caused by
environmental factors (Joseph, 2004).
Importantly, heritability estimates do not speak to the responsiveness of a phenotype
to environmental intervention. Traits can be highly heritable and yet be drastically al-
tered or eliminated by changes in the environment (e.g., Lewontin, 1974; Plomin et al.,
2012). Moreover, the heritability of a trait can change as a result of changes in the envi-
ronment. Returning to the cornfield example, if one were to plant trees around the corn
field, thereby producing unequal sun exposure among the corn plants within a field, then
heritability would decrease as the resulting variation in sunlight exposure would account
in part for differences in plant height.
According to the prevailing behavioral genetics methodology, the heritability of a given
phenotype is determined by comparing concordances and discordances between subjects
relative to their presumed degree of genetic similarity. As noted, the index used to cap-
ture the relative strength of genetic influence on the population variance of a phenotype
is the heritability coefficient. Although the word “heritability” has been around since the
1800s, the concept as it is used today was introduced by Lush in 1936, who was interested
in facilitating agricultural breeding for economically desirable traits (Bell, 1977). Animal
and plant breeders needed a quantifiable method for predicting the results of programs
of selective breeding. For Lush (1949: 359), knowing “whether heritability is high or low
is important when making efficient breeding plans” (emphases added). In human popu-
lations, selective breeding is known as eugenics, and although some early behavioral ge-
neticists were interested in heritability for eugenics purposes (e.g., Fisher, 1918), currently
few to none advocate the use of heritability estimates for eugenics programs to prevent
or control crime or other behaviors.1Although some behavioral geneticists have argued
that the “only practical application of the heritability coefficient is to predict the results
of a program of selective breeding” (Wahlsten, 1990: 119), others obviously believe that
the estimates have other practical uses given the continued estimations of heritability on
Even if the heritability concept is appropriate for use in humans in nonselective breed-
ing programs, the methods used to derive these estimates are problematic in ways that
render the estimates from the model ambiguous at best (e.g., Charney, 2012; Goldberger,
1979; Joseph, 2004, 2006; Wahlsten, 1990). Because each of these methods has its own set
of problems, we discuss each in turn focusing in particular on the relevance of the meth-
ods’ assumptions for conclusions about the heritability of crime. Although the methods
are based on several different assumptions, in general, they are united by the fact that
all compare individual phenotypes across varying degrees of genetic relationships and
use these comparisons to estimate genetic and environmental influences without actually
measuring either.
The main “workhorse” used in behavioral genetics to estimate heritability is the classic
twin-based research design (or the “twin study”; Plomin et al., 2012). Identical (monozy-
gotic [MZ]) twins come from one fertilized egg and are assumed to share 100 percent
of their genetic material (genetic clones), whereas fraternal (dizygotic [DZ]) twins come
from two fertilized eggs and are presumed to share on average half of their genetic ma-
terial (the same amount as other nontwin biological siblings). Given this and several as-
sumptions, researchers have used MZ and DZ twins as a natural experiment to separate
genetic and environment influences on variation in phenotypes (for a review, see Plomin
et al., 2012).
The twin study separates phenotypic variation into three components: additive genetic
(h), shared environment (c), and unshared environment (e). The unshared environment
also includes model error. Notably, the terms “shared” and “unshared” environment do
not correspond directly to common sense interpretations. The so-called shared environ-
ment consists of all nongenetic influences that make twins similar to each other, whereas
“unshared” environmental influences consist of all nongenetic factors that make twins
different (Plomin, 2011; Suhay and Kalmoe, 2010). Whether “shared” and “unshared”
environments are actually shared is not at issue; instead, they refer to “‘effects’ rather
than ‘events’” that twins experience (Plomin, 2011: 582). Scholars frequently have failed
to describe clearly what is meant by these terms, and others have made inappropriate
conclusions about the insignificance of parental or community factors based on shared
environment estimates (Harris, 1998; Rowe, 1994). It is important to remember that, in
general, twin studies do not actually measure the shared or unshared environments; rather,
1. It is worth noting that it was not so long ago (as late as 1979 in Virginia) that compulsory ster-
ilization as a means of crime prevention and/or punishment was practiced in the United States.
More than two thirds of states adopted sterilization laws in the twentieth century, and probably
many more than the widely cited figure of 63,000 Americans were involuntarily sterilized from the
1890s through the 1970s (Largent, 2007). Isolated instances of “voluntary” sterilization for con-
victed criminals in exchange for a reduced sentence continue in the twenty-first century and not
only for sexual offenses (Largent, 2007).
these parameters are estimated based only on concordance rates or correlations between
MZ and DZ twins.
The basic logic of the twin study is to compare twin concordances for phenotypes and,
based on several assumptions, assign the greater phenotypic similarity of MZ relative to
DZ co-twins to their greater genetic similarity. Through the formula explained in more
detail in the subsequent discussion, heritability is usually estimated from twice the MZ–
DZ difference in correlations. Although in recent years the twin study model has become
more sophisticated, using latent variable models, the basic logic underlying these more
advanced models is identical to that of the earlier twin studies, in which the twins’ cor-
relations are inserted into a series of simple equations and the heritability coefficient is
calculated with elementary algebra (Suhay and Kalmoe, 2010). The basic assumptions
that underlie the twin-study method have remained largely unchanged since the 1920s
and are still central to these models given that neither genetic nor environmental influ-
ences are measured. Some of these assumptions are generally unproblematic, whereas
others are dubious. The assumptions are as follows (Charney, 2012; Joseph, 2006; Plomin
et al., 2012):
1. Researchers can reliably and accurately determine twin type (DZ vs. MZ).
2. The genes of MZ twins are 100 percent identical and are approximately 50 percent
identical for DZ twins.
3. The percentages of genes shared by different types of twin pairs remain the same
over the life course.
4. Phenotypic variation can be demarcated into genetic (G), shared environmental
(C), or unshared environmental (E) components.
5. The relevant genes exert effects additively.
6. The likelihood of receiving a diagnosis or label for a phenotype (e.g., criminal con-
viction) is the same among the twin and the nontwin population (generalizability).
7. The risk of receiving the diagnosis or label is the same among MZ and DZ co-
8. The phenotype (e.g., criminality or self-control) can be modeled as a quantitative
9. The environments of MZ co-twins are no more similar than that of DZ co-twins
(“equal environment assumption”).
In general, many recent twin studies in criminology often fail even to mention these
assumptions, much less discuss the adequacy of them. Although each assumption is po-
tentially subject to some qualification, the final assumption, the equal environment as-
sumption (EEA), has drawn the most attention among scholars criticizing the technical
problems of heritability studies. As discussed, this crucial assumption and its modified
form are clearly violated when the outcome of interest is criminal and related behaviors.
EEA Assumption
To observe the relevance of the EEA, it is useful to look at the basic equations un-
derlying the model (see Plomin et al., 2012; Suhay and Kalmoe, 2010). Note in the
equations that follow, the terms are squared because they represent the proportion
of variance explained.
·rMZ =h2+c2
represents trait correlations between MZ co-twins as a function of genes (h) and shared
environment (c).
·rDZ =h2/2+c2
represents trait correlations between DZ co-twins as a function of genes (h), half that of
MZ co-twins, and shared environment (c).
To calculate the heritability estimate, the DZ correlation is subtracted from the MZ
·rMZrDZ =(h2+c2)(h2/2+c2)
To move past this equation, the EEA assumption is necessary. By assuming that the
shared environment (c) is equivalent for MZ and DZ co-twins (the EEA assumption),
the equation can be simplified to the following equation:
·rMZ rDZ =h2h2/2
2(rMZ rDZ)=2(h2h2/2)
and finally
h2=2(rMZ rDZ)
The shared environmental influence is calculated by starting with the first equation and
solving for c2.
·c2=rMZ h2(or: c2=2rDZ rMZ)
Given the assumption (#4) that all effects are genetic, shared environmental, or un-
shared environmental effects, one can calculate the effects of the unshared environment
(e2) with the residual variance. Given that h2+c2+e2=1, then:
·e2=1h2c2(or,alternatively: e2=1rMZ)
As we can observe, these calculations are crucially dependent on the EEA assumption
that the environments of MZ co-twins are no more similar than that of DZ co-twins. Every
twin-based study is based on this EEA assumption to assign the greater concordance rates
among MZ co-twins to genetics. Without this assumption, the greater concordance rates
of MZ twins could be caused by more similar environments and/or genetics, and thus, h2
could not be calculated (e.g., Joseph, 2004).
As it happens, and likely not surprising to those who have experience with MZ and
DZ twins, this central assumption is flatly contradicted by both empirical evidence and
common sense. Research clearly demonstrates that MZ co-twins experience more sim-
ilar social environments than DZ co-twins. For instance, MZ twins are more likely to
be treated similarly by their parents (Evans and Martin, 2000), to have the same friends
(Cronk et al., 2002; Horowitz et al., 2003), to share the same classroom (Cronk et al.,
2002; Morris-Yates et al., 1990), to spend time together (and therefore experience the
same social environments more frequently; Horowitz et al., 2003; Rende et al., 2004), and
to go out together than DZ twins (Kendler and Gardner, 1998). Not surprisingly, MZ
co-twins also report greater closeness and identification with one another (Jackson, 1960;
LaBuda, Svikis, and Pickens, 1997; Segal, 2000) as well as mutual influence (Ainslie, 1997;
Sandbank, 1999). For example, studies reveal that MZ twins are more likely than DZ
twins to share bedrooms and clothes, and to share experiences like identity confusion (91
percent vs. 10 percent), being inseparable as children (73 percent vs. 19 percent), being
brought up as a unit (72 percent vs. 19 percent), and having a high level of closeness (65
percent vs. 19 percent) (Richardson, 2011). Perhaps at the most basic level, MZ twins are
treated more similarly and experience situations more similarly given their more similar
appearance (attractiveness, height, physicality, and the like; Horowitz et al., 2003; Joseph,
As persuasive evidence that the environments are more similar for MZ than DZ co-
twins began to mount in the 1960s, behavioral geneticists acknowledged that the EEA
was invalid (Joseph, 2004). Rather than concede that the twin study was ill suited for es-
timating heritability, they adopted a redefined “trait-relevant EEA,” which grants that
MZ co-twins might experience more similar social environments than DZ co-twins but
assumes that these differences are irrelevant for the trait being studied (e.g., Carey and
DiLalla, 1994; Kendler, 1983). For example, Kendler et al. (1993: 21) defined the equal-
trait relevant EEA as the assumption “that monozygotic (MZ) and dizygotic (DZ) twins
are equally correlated for their exposure to environmental influences that are of etiolog-
ical relevance to the trait under study.” Although this assumption may be reasonable for
some phenotypes such as diabetes or heart disease, it is clear that it is not tenable for phe-
notypes such as crime given evidence that the more similar environments of MZ co-twins
are trait relevant. Indeed, many of the shared environmental influences for which MZ co-
twins have been shown to be more similar than DZ co-twins (e.g., parenting, peers, and
leisure time together) represent some of the most potent predictors of the outcomes of
concern to criminologists.
Given these clear findings, among others, of more similar social environments for MZ
co-twins than DZ co-twins, many scholars have asserted that the EEA, including its trait-
relevant form, is invalid and that the more similar environments of MZ than DZ co-twins
bias heritability estimates upward to a significant degree (e.g., Beckwith and Morris, 2008;
Horowitz et al., 2003; Joseph, 2004; Lewontin, Rose, and Kamin, 1984; Richardson, 2011).
Even minor violations of the EEA can produce substantial overestimations of heritability
(and thus underestimates of the shared environment). For example, if the shared environ-
mental effect is .3 for MZ twins and .2 for DZ twins, then heritability will be inflated by
20 percent (Suhay and Kalmoe, 2010). Thus, it is a plausible interpretation of twin-study
findings that surprisingly high heritability and unexpectedly low shared environmental
estimates are caused in no small part by MZ co-twins having more similar environments
than DZ co-twins (e.g., Jackson, 1960; Joseph, 2006).
2. Although some behavioral geneticists argue that treatment similarity based on similar appearance
should be treated as a genetic effect (e.g., Bouchard et al., 1990), we disagree (see Joseph, 1998, for
a discussion).
Table 1 displays a list and description of 20 twin studies we identified that have been
published since 2008 examining crime and related phenotypes from a criminological per-
spective.3These studies were compiled from a search of the literature, primarily using
Google Scholar. Criteria for inclusion were that the twin-study method was used to esti-
mate the heritability of crime or a related phenotype (e.g., self-control, deviant peers, and
antisocial behavior) from a criminological perspective (indicated by the use of crimino-
logical terms and theories). No study was excluded for any reason other than the crimino-
logical criteria just mentioned. Our critique of recent criminological twin studies is based
on these 20 studies.
Given the centrality of the EEA, it is particularly surprising that most (19 of 20) of
the recent criminological twin studies displayed in table 1 fail even to mention the EEA
or its trait-relevant form at all, much less discuss its adequacy or the implications of its
violation. Thus, readers who are unfamiliar with the assumptions of the model are not
presented with this highly dubious assumption that is crucial for making genetic infer-
ences, is likely violated, and results in inflated heritability and underestimated shared
environmental estimates.
The EEA is even less reasonable for twin studies that combine same-sex and opposite-
sex twins into their models (e.g., Beaver et al., 2008; Boisvert et al., 2012; Boutwell et al.,
2013; Vaske, Boisvert, and Wright, 2012). As shown in table 1, most of the criminologi-
cal twin studies we identified since 2008 included opposite-sex DZ pairs, and nine of the
ten studies since 2012 did. (The percent of DZ co-twins in these studies that were op-
posite sex was approximately 45 percent.) In these cases, same-sex MZ twins are being
compared with same-sex and opposite-sex DZ twins, with the assumption that (same-
sex) MZ co-twins are treated no more similarly than opposite-sex DZ co-twins. Given
the voluminous research on sex/gender differences in experiences, this assumption seems
patently invalid. Evidence supporting this interpretation is found in a recent study by
Meier et al. (2011), who compared the correlation for childhood conduct disorder among
opposite-sex and same-sex DZ twins and found that the opposite-sex correlation was sig-
nificantly smaller (approximately half, r=.15) than that of the same-sex DZ twins (see
also Saudino, Ronald, and Plomin, 2005).
An even more surprising practice in a few recent criminology heritability studies is
the use of kinship pairs, which includes MZ twins, DZ twins, siblings (full and half), and
cousins (Barnes and Beaver, 2012; Barnes and Boutwell, 2012; Barnes, Boutwell, and Fox,
2012). [These extended twin-study designs are included in the category of “twin studies”
given that they use the same model and assumptions extended for different degrees of
genetic relatedness (Plomin et al., 2012)]. In these models, kinship pairs are compared
with twins based on their average genetic relatedness, ranging from 1.0 for MZ co-twins
to .125 for cousins. While these models often are not described clearly, by necessity they
rely on the EEA to infer genetic influence. Thus, heritability estimates in these studies
3. Given that we had to select some cutoff date, the year 2008 was chosen for three reasons. First, the
recent surge in heritability studies in criminology started in this year. Second, this is after the time at
which recent critiques of the method had been published (e.g., Horowitz et al., 2003; Joseph, 2004;
2006) and even prominent scholars (e.g., Rutter) in the behavioral genetics field had highlighted
the problems in the models and recommended that we should move beyond heritability. Finally,
we wanted to keep the list manageable so that the reader can get an idea of the characteristics of
the recent studies on which we focus without being overwhelmed.
Table 1. List of Identified Criminology Twin Studies and Kinship Studies Published Since 2008
Twins OS and Note Gene Mention Present Any
vs. SS DZ Mention Additivity Mention Shared CIs/ Envir.
Authors Year JournalaOutcome(s) Data Kinship Pairs EEA /Discuss G ×E “Effects”bSDs Measuredc
1. Viding et al. 2008 DS ASB Twins Early
Dev. Study
Twins SS No Yes/no No No Yes Yes
2. Ball et al. 2008 CP&P Bullying,
E-Risk Long.
Twin Study
Twins SS No Yes/no No Yes No Yes
3. Beaver et al. 2008 JCJ SC, SC stability Add Health Twins OS +SS No No/no No No No No
4. Wright et al. 2008 JQ SC, DPeers,
Add Health Twins OS +SS No Yes/no No No No Yes
5. Beaver, Schutt,
et al.
2009 CJ&B SC, DPeers Add Health Twins SS No No/no No No No Yes
6. Beaver,
Boutwell, et al.
2009 YV&JJ Victimization Add Health Twins SS No No/no No No No Yes
7. Beaver,
Ferguson, and
2010 CJ&B SC Add Health Twins SS No No/no No No No Yes
8. Meier et al. 2011 JAP Conduct disorder, ASB Australian
Twin Reg
Twins OS vs. SSeYes Yes/no Yes No Yes Yes
9. Beaver 2011b JQC Parenting, SC,
Add Health Twins SS No No/no Yes No Yes Yes
10. Beaver, Gibson,
et al.
2011 C&D DPeers, DPeers
Add Health Twins SS No No/no No No No No
11. Barnes and
2012 JIV Victim–offender
Add Health Kinship OS +SS Yes No/no Yes No Yes No
12. Barnes,
Boutwell, and
2012 YV&JJ Victimization, gang
membership, overlap
Add Health Kinship OS +SS No No/no No No Yes Yes
13. Barnes and
2012 JCJ Delinquency, stability Add Health Kinship OS +SS No No/no YesgYes Yes No
14. Boisvert et al. 2012 JQC SC, delinquency,
Add Health Twins OS +SS No No/no No No Yes No
Table 1. Continued
Twins OS and Note Gene Mention Present Any
vs. SS DZ Mention Additivity Mention Shared CIs/ Envir.
Authors Year JournalaOutcome(s) Data Kinship Pairs EEA /Discuss G ×E “Effects”bSDs Measuredc
15. Vaske,
Boisvert, and
2012 JIV Victim, delinquency,
Add Health Twins OS +SS No No/no Yes No Yes No
16. Boisvert,
Wright, et al
2013 JQ SC (sex differences) Add Health Twins SS No Yes/no No Yes Yes No
17. Boisvert et al. 2014 JCJ SC, subuse, overlap Add Health Twins OS +SS No Yes/no YeshYes Yes No
18. Boutwell et al. 2013 JA SC, victim, overlap Add Health SiblingdOS +SS No No/no No No Yes No
19. Boisvert et al. 2013 CJ&B DPeers, delinquency,
Add Health Sibling OS +SS No Yes/no No Yes Yes No
20. Beaver,
Boutwell, and
2013 CJ&B Social support, SC Add Health Twins OS +SSfNo Yes/no No No Yes Yes
ABBREVIATIONS: ASB =antisocial behavior; CI =confidence interval; DPeers =delinquent peers; OS =opposite sex; SC =self-control; SD =standard
deviation; SS =same sex.
aJournal abbreviations/acronyms are as follows (in order of appearance): DS: Developmental Science; CP&P: Journal of Child Psychology and Psychiatry; JCJ:
Journal of Criminal Justice; JQ: Justice Quarterly; CJ&B: Criminal Justice and Behavior; YV&JJ: Youth Violence and Juvenile Justice; JAP: Journal of Abnormal
Psychology; JQC: Journal of Quantitative Criminology; C&D: Crime and Delinquency; JIV: Journal of Interpersonal Violence; JA: Journal of Adolescence.
bThis column indicates whether the study mentioned that “shared environmental estimates” actually capture shared environmental “effects.” No study goes beyond
a brief mention.
cWhether an environmental factor is measured is determined broadly, such that factors that could conceivably include environmental effects are considered
“environmental” here, such as sex and age.
dCousins were excluded. The authors noted: “[P]arameter estimates were substantively unchanged when only MZ and DZ twins were analyzed” (p. 660).
eOS twins were not combined but assessed separately for comparison.
fDoes not specify OS, but given the reported sample size, we determined that OS twins were included.
gThe authors mentioned in a footnote.
hExplained to the reader how interactions were treated in the model.
are based on the assumption that the environments of pairs of opposite-sex cousins are
no less similar than that of identical twins for the outcomes under study. Quite simply,
the EEA seems preposterous for studies of kinship pairs. That such studies report high
heritabilities and nil shared environmental effects is not surprising.
Statements that ignore the EEA and its potential violation, such as “the only reason
MZ twins should be more similar than DZ twin pairs is because they share twice as much
genetic material,” are found in recent heritability studies (e.g., Beaver, 2011b: 86; Beaver,
Ferguson, and Lynn-Whaley, 2010). As we have discussed, MZ co-twins may be more
similar than DZ co-twins for criminal phenotypes for two reasons: genetics and/or more
similar shared environments. Given the evidence of more similar treatment for MZ co-
twins as opposed to DZ co-twins (and especially in the case of opposite-sex DZ twins
and kinship pairs), we think it is unquestionably the case that violations of the EEA are
inflating heritability and decreasing shared environmental effects to a substantial degree.
Genetic Additivity Assumption
The additivity assumption (assumption #5) also is implicated in the accuracy of her-
itability estimates. Although the evidence of nonadditivity of genetic and environmen-
tal influences is discussed more below, this assumption refers to the additivity of genetic
influences on a phenotype. Genetic variance can be separated into additive and nonad-
ditive components.4Additive genetic variance, which is that assumed by twin studies,
is a model of gene combinatorial effects where many genes each contribute small indi-
vidual effects that add up to shape the phenotype. Nonadditive genetic variance is that
which arises because of the interactions between genes such that the resulting quanti-
tative phenotype is significantly different from the sum of the individual genetic effects
(Stoolmiller, 1999). Nonadditive variance can be of two kinds: dominance and epistasis.
Dominance is that occurring when the alleles at a given locus (one from each parent) in-
teract to produce a phenotype. This interaction occurs among genes that operate with a
strict dominance-recessive mode of inheritance. Epistasis occurs when several genes (al-
leles at different loci) interact to produce a behavior. Although the extent of nonadditive
genetic variance involved in criminal phenotypes is not known, there is good reason to
believe it is operative for all complex traits and behaviors, including crime (Plomin et al.,
Violations of the additivity assumption have consequences for heritability estimates.
Several decades ago, Grayson (1989) showed that in twin studies, undetected nonadditive
genetic variance inflates heritability and deflates shared environmental influences. The
population genetic correlation between first-degree relatives, such as DZ co-twins, on a
trait where all genetic influences are additive is .50. Conversely, nonadditive genetic influ-
ences can, at most, produce a correlation of .25 if the nonadditivity is caused entirely by
dominance interactions. As the nonadditivity shifts to epistasis, the expected genetic cor-
relation declines toward zero (Grayson, 1989; Stoolmiller, 1999). Genetic nonadditivity
will thus have the effect of reducing the genetic correlation for DZ co-twins but not for
4. Given the rarity of monogenic traits, there is consensus that in nearly all cases, genes combine to
influence phenotypes. For example, eye color, which was once thought to be determined by around
three genes or less, has been shown to be influenced by at least 20 and possibly hundreds of genes
(Liu et al., 2010).
MZ co-twins (because MZ co-twins have all of the same interacting alleles), which will
lead to a reduction in the overall correlation for DZ twins but not for MZ twins. Twice
the MZ–DZ difference in correlations (the estimate of heritability) is thus an overesti-
mate, and twice the DZ correlation minus the MZ correlation (the estimate of shared
environment) is thus an underestimate (Grayson, 1989).
Thus, similar to violations of the EEA, violations of the additivity assumption serve to
“maximize the potential role of genetic influences and minimize the potential contribution
of shared/familial environmental influences” (Grayson, 1989: 594). Although dominance
interactions can be modeled in twin-study designs, we have only observed this attempted
once in criminology (Boisvert, Boutwell, et al., 2013), and doing so has its own set of
problems (for example, shared environmental effects are ignored because they cannot
be estimated simultaneously; Neale and Cardon, 1992). Moreover, modeling epistatic in-
teractions with human kinship data is generally considered impracticable (Eaves, 1988).
Regarding nonadditivity, Plomin et al. (2012: 401) stated: “These types of effects compli-
cate model fitting because there are many forms in which they could occur. Normal twin
study designs do not offer much hope for identifying them.”
Recent published twin studies in criminology often, but not always, have noted that
additive genetic and environmental influences were being estimated; however, as shown
in table 1, these studies invariably have not explained what this means or the implications
of the likely violation of the assumption of additive effects (deflated shared environmen-
tal estimates and inflated heritability). As such, the resulting biases caused by the likely
violation of the additivity assumption have not been made clear.
Additional Issues
Although not technical limitations of the twin-study model, we believe that three ad-
ditional characteristics of recent criminology twin studies are worth noting. As can be
observed in table 1, of the identified 20 criminological twin studies published since 2008,
17 used the Add Health data. We do not argue that the genetic twin sample in the Add
Health is deficient; indeed, the quality of the data seems to be extraordinary (Harris et al.,
2006). We do believe, however, that reproducing findings of similar heritabilities for var-
ious criminal-related traits on the same set of 289 MZ and 452 DZ twin pairs is prob-
lematic. Moreover, this means that most recent heritability estimates in criminology have
been based on the same imperfect measures (self-control, delinquent peers, delinquency,
and victimization) that are available in the Add Health data. The measure of delinquent
peers, for example, only includes respondents’ perceptions of their three closest friends’
use of three substances (e.g., Beaver, Schutt, et al., 2009; Beaver et al., 2011; Boisvert,
Boutwell, et al., 2013), and the measure of victimization only includes experiences with
physical violence (Boisvert et al., 2014; Vaske, Boisvert, and Wright, 2012). To be sure,
we are not encouraging replication of these findings in other data sets as we believe the
twin-study method has too many problems to provide valid or reliable estimates. We do
wish to draw the reader’s attention to the fact that similar twin-study findings in criminol-
ogy, at least over the past 6 years, have been based largely on the same group of twins
(and kinship pairs) with generally the same measures taken at the same wave.
An additional reason to view twin-study heritability estimates with caution is that they
tend to have large confidence intervals (DiLalla, 2002). As shown in table 1, several recent
studies in criminology have not reported the confidence intervals or the standard errors
for the estimates, which serves to reify an inherently imprecise estimate.
A final caution about twin-study findings has to do with interpretations of the results.
Several recent criminological twin studies draw unjustified conclusions from insignificant
estimates of shared environmental effects (e.g., Beaver, Boutwell, et al., 2009; Beaver,
Schutt, et al., 2009). For example, Beaver, Ferguson, and Lynn-Whalley (2010: 1060) con-
cluded: “The null results for the shared environment across various measures of parenting
indicate that parents do not have a causal effect on shaping and molding their offspring’s
level of self-control.” However, a basic principle of hypothesis testing is that failing to
reject the null hypothesis does not mean that the null hypothesis is true. As such, failing
to reject the null hypothesis that the shared environmental effect equals zero is not ev-
idence that the effect is, in fact, zero. All one can say from such null findings is that we
do not have enough evidence to say that the shared environmental effect is not different
from zero. As such, the evidence does not show that shared environmental effects, such
as parenting, do not have a causal effect on variation in various phenotypes.
Adoption studies have been deemed a powerful design for investigating genetic and
environmental influences. Adoption studies investigate correlations for phenotypes be-
tween adoptees and their biological parents and genetically unrelated adoptive parents.
Several different adoption designs exist (see Joseph, 2004, 2010; Rutter et al., 1990, for
overviews), but all make use of this “genetic parent”–“social parent” contrast for infer-
ring genetic influences.
In theory, correlations between adoptees and biological parents indicate the influence
of genes, whereas those between adoptees and social parents represent the influence of
environmental factors. As such, adoption studies have been promoted as “natural ex-
periments in which the effects of genetics and rearing influences may be separated to a
high degree” (Mednick and Kandel, 1988: 103). Many scientists unswayed by the findings
from twin studies have pointed to adoption studies as powerful confirming evidence, and
scholars have pointed to the salient role of adoption study findings in shifting the tide
toward genetic explanations (Joseph, 2004; Rowe and Jacobson, 1999). Like twin stud-
ies, however, adoption studies suffer from several limitations that render their estimates
questionable and seem to distort estimates toward inflating heritability and underesti-
mating environmental influences, especially the shared environment (e.g., Joseph, 2004;
Rutter, 2006; Stoolmiller, 1999). Included in these limitations are late separation, non-
representativeness and range restriction, selective placement, and prenatal environment
Late Separation
Late (nonbirth and beyond) separation of the child from his or her biological parent is
an obvious environmental confound, for two reasons. First, late separation confounds the
allegedly pure genetic influence of the biological parent (e.g., Faraone et al., 1999), espe-
cially given the large body of research evincing that early childhood is a crucial period for
development. In addition, late separation itself alters the natural experiment of adoptions.
Adoptions can have a significant disruptive effect on child development when separations
occur after the formation of an attachment relationship (e.g., Baumrind, 1993). Moreover,
in addition to the stress of adoption, the status of being an adoptive child is itself a source
of strain (Brodzinsky, Singer, and Braff, 1984; Steinhauer, 1983).
Nonrepresentativeness and Range Restriction
Perhaps the most serious problem with adoption studies is their nonrepresentativeness.
Adoptive parents severely underrepresent high-risk social environments as they tend to
be more affluent, more highly educated, and live in better communities than the general
population (e.g., Lewontin, Rose, and Kamin, 1984; Stoolmiller, 1999). Thus, the range
of environmental (especially familial) quality in adoption studies is both well above aver-
age and more restricted in range than in the general population (Kamin, 1981; Miles and
Carey, 1997). As Stoolmiller (1998) noted, this restricted range results from at least three
different selection mechanisms: 1) Adoptive parents are carefully screened by adoption
agencies before they are allowed to adopt, 2) adoptive families always choose to have a
child, and 3) adoptive families volunteer to participate in the study (presumably adop-
tive families who provide a poor family environment would not agree to participate in an
adoption study). It seems clear that “powerful selection forces are at work in determin-
ing what kind of families will be present in an adoption study” (Stoolmiller, 1999: 395).
Indeed, evidence suggests that at least the bottom 60 percent of the general population’s
range of family environmental quality is excluded from adoption studies (Stoolmiller,
Given that range restriction reduces correlations, the effect of this significant restriction
of range in family environment quality in adoptive samples will be to attenuate the corre-
lation between adoptive parents and children, thereby decreasing shared environmental
estimates and inflating genetic estimates (Lewontin, Rose, and Kamin, 1984; Miles and
Carey, 1997). Applying a correction for range truncation to analyses of the Texas Adop-
tion Project, Stoolmiller (1998) estimated that shared family environment accounted for
55 percent of the variance in child IQ scores, noting that conventional estimates are usu-
ally less than half that amount. Based on his research, Stoolmiller (1999: 393) concluded:
“This pattern of differential attenuation and inflation is the reason for the apparent lack
of relative importance of SE [shared environment] in behavior-genetic adoption stud-
ies.” Moreover, range restriction may account for the ostensibly contradictory findings
between individual and group studies of adoption (Stoolmiller, 1999). In contrast to in-
dividual studies, group studies of adoption clearly indicate a beneficial effect of being
adopted from higher risk environments to lower risk environments (e.g., lower class to
middle class) on IQ and deviance, which is consistent with a strong effect of the shared
environment (Turkheimer, 1991).
Selective Placement
Another problematic issue with adoption studies is the “no selective placement” as-
sumption, which is the assumption that there is no association between characteristics
of the biological parents and the adoptive family (Joseph, 2010; Kamin, 1985; Lewontin,
Rose, and Kamin, 1984). In regard to criminality, selective placement could bias the sepa-
ration of genetic and environmental effects if the adopted children of the least criminally
inclined biological parents were adopted by the families who provided an environment
least conducive to criminality (and vice versa). If selective placement occurs, then genetic
effects will be inflated. Selective placement can result from adoption agencies “fitting the
home to the child” (attempting to maximize similarities between biological and adoptive
families to increase the child’s chance of fitting in; Munsinger, 1975: 627) or children of
criminal parents being considered less desirable adoptees and therefore ending up in less
desirable adoptive families (Joseph, 2004; Munsinger, 1975).
Evidence suggests that selective placement related to criminal characteristics is a fac-
tor in adoption studies (see the review in Joseph, 2004). For example, evidence shows
that children whose biological parents had a history of criminal behavior or mental dis-
orders were more likely to be placed in inferior adoptive family environments (within
the restricted range) compared with adoptees without convictions or disorders among
their biological parents (Joseph, 2004; Kamin, 1985). In addition to selective placement,
it seems that selective late placement is also a biasing factor. For example, Bohman (1978:
275), an adoption study proponent, found that adoptees with criminal or alcoholic biolog-
ical parents were placed on average 2 to 3 months later than controls, noting that “later
placement is associated with selective factors that contributed independently to poorer
social adjustment later in life and an increased risk of appearance in the [criminal] regis-
ters.” In sum, it seems that selective placement is operating in adoptive studies, especially
in regard to factors such as criminality, in violation of random assignment, which valid
adoption studies require (Joseph, 2004).
Prenatal Influences
A final notable limitation in adoption studies concerns prenatal influences. Birth moth-
ers in adoptive studies provide not only genes but also a prenatal environment; thus,
similarity between biological mothers and children—even those adopted at the moment
of birth—can be caused by both genetic and (prenatal) environmental influences (e.g.,
Plomin et al., 2012). Importantly, poor obstetric care, exposure to toxins, stress, and poor
nutrition during pregnancy are not randomly distributed among the population of preg-
nant women. Instead, some of the same social factors associated with criminality and/or
consequences of criminality (e.g., poverty, neighborhood disadvantages, and substance
use) are associated with factors known to have an influence on fetal development and
postnatal cognitive and behavioral developmental (e.g., lead, illicit drugs, nicotine, and
endocrine disrupters). Inasmuch as prenatal exposure to these factors increases the like-
lihood of crime (e.g., Raine, 2002a, 2002b), children of criminal parents would be at
a higher risk for crime for nongenetic reasons. Such shared environmental factors be-
tween mother and child (in utero and after birth in the case of late placement) will be
included in the genetic estimate in adoption studies. Given this evidence, Conley (2011)
[W]e know, ipso facto, that families who adopt are a distinct social group on
unobservables—as are the adoptees themselves ...The only adoption study that
would avoid such [problems] would be one in which adoptees were randomly se-
lected from the new-born population and then randomly assigned to parents, with
both groups blind to the treatment (i.e., not knowing whether they were adopted
or not)—all while prenatal environment was held constant. In other words, it is an
impossibility to reliably estimate genetic heritability using [the adoption method].
(p. 597)
Adoption Studies of Crime-Related Phenotypes
Heritability estimates of criminality and related phenotypes (e.g., aggression and anti-
social behavior) from adoption studies are lower than those from twin studies (Joseph,
2004; Moffitt, 2005), and not all find evidence of the heritability of crime (Bohman,
1978).5In general, all of the adoption studies contain various invalidating flaws, includ-
ing but not limited to those mentioned previously. For example, the most frequently
cited adoption study in support of genetic effects on crime by Mednick, Gabrielli, and
Hutchings (1984) suffers from selective placement, late placement, and nonrepresenta-
tiveness, among others (see Joseph, 2004; Kamin, 1985). Criticisms of particular features
of all but the most recent studies have been detailed elsewhere (see Gottfredson and
Hirschi, 1990; Joseph, 2001, 2004; Walters and White, 1989). We have not yet found crit-
ical attention to the latest adoption studies. Thus, we briefly comment on two recent
Beaver (2011a: 282) investigated “genetic influences on being processed through the
criminal justice system” using the subsample of adoptees included in the Add Health
Study. Although Beaver (2011a) found that adoptees whose biological parents “had ever
spent time in jail or prison” were significantly more likely to have contact with the crim-
inal justice system, this finding is vitiated by several serious limitations, including those
mentioned. Perhaps the most significant of these is the study sample. The only require-
ment for inclusion in the study sample was that the respondent indicated that he or she
was adopted sometime before the survey (which took place when youth were in grades
7 through 12) and did not currently live with a biological parent. Because of data limi-
tations, Beaver (2011a) could not ascertain the age at which the children were adopted
and did not control for contact with biological parents. Additionally, information about
the biological parents’ incarceration or lack thereof came from the adoptees themselves,
and only respondents who were aware of their biological parents’ jail or prison ex-
periences were included in the analyses. (Adoptees who answered “I don’t know” to
the question of biological parents’ prison or jail experience were excluded.) As such,
those respondents who had no knowledge about their biological parents’ jail or prison
status—almost certainly those who had the least contact with their biological parents
(and could not be influenced by potential labeling processes involved in having a crim-
inal parent)—were not included in the analyses. This same Add Health adoption subsam-
ple and model also was used to “estimate genetic influences on victimization” (Beaver
et al., 2013: 149).
In sum, the adoption method was promoted as a powerful model for separating genetic
and environmental influences that avoided limitations of twin studies by “more cleanly
[separating] genetic and environmental influences” (Raine, 1993: 60; also Mednick and
Kandel, 1988; Plomin and DeFries, 1985) and as such has played a crucial role in bol-
stering findings from twin studies. It is clear, however, that the adoption method suffers
from several of its own invalidating flaws, which—like twin studies—seem to bias esti-
mates systematically toward genetic influences and against shared environmental ones
(e.g., Joseph, 2004; Stoolmiller, 1999).
5. For a more thorough critical review of adoption studies of crime, see Joseph (2004), and for a list
of heritability estimates and brief methodological features of the studies prior to 2004, see Moffitt
The third heritability model, which has been used less frequently than the other two,
largely because of the rarity of the situation, is known as the twins reared apart (TRA)
study. TRA studies compare the phenotypes of reared-apart MZ pairs (MZAs) with a
control group of MZ pairs raised together. The rationale for these studies is in large part
to avoid the problems generated by the EEA (Joseph, 2010). Despite the paucity of stud-
ies, TRA findings often are cited as providing crucial support for claims that dismiss the
import of shared environmental factors on psychosocial development and in support of
the EEA (e.g., Alford, Funk, and Hibbing, 2005; Harris, 1998; Pinker, 2002). Although
only one systematic study of TRAs has investigated antisocial behavior (Grove et al.,
1990), this study often is cited in criminological work as support for the findings obtained
from classic twin studies regarding the heritability of crime/antisocial behavior (e.g., Mof-
fitt, 2005; Raine, 2002b). In general, TRA studies have received considerable attention
from both the scientific community and the media over the past several decades, partly
because of the interest in the compelling stories of eerie similarities among reared-apart
twins (see Joseph, 2004).
Despite their intuitive appeal and publicity, however, TRA studies suffer from a host of
limitations that render their heritability estimates highly problematic at best (e.g., Joseph,
2004; Rutter, 2006). The limitations include the following: 1) Many common environmen-
tal factors shared by MZAs could lead to greater concordance; 2) many if not most twins
were only reared “partially apart”; and 3) biases in sampling, which favor recruitment of
MZA pairs that are more similar than the MZA population (e.g., Charney, 2008; Farber,
1981; Joseph, 2004). As a result of the dearth of TRA studies of criminal behavior, a de-
tailed discussion of these limitations is not possible in this article. However, given many
behavioral geneticists’ assertions that TRAs provide crucial support for twin-study find-
ings, we provide a discussion of this method and its limitations in the online supporting
information.6There, we provide evidence supporting our position that TRA studies also
suffer from serious problems that render their results highly dubious and biased toward
inflated heritability.
Aside from their methodological pitfalls, an equally serious problem with heritabil-
ity studies is the notion that genetic and environmental effects can be partitioned into
separate additive influences in the first place (assumption #4). Obviously, an estimation
of heritability requires that one can in fact separate genetic from environmental influ-
ences on behavior. Reality is not so simple; next, we attempt to show that quantita-
tive genetics uses an outdated model of genetic structure and function that is explicitly
rejected by the scientists who study the actions of genes directly (e.g., Gottlieb, 1992;
Meaney, 2010).
From a classical genetic perspective, two contributions to phenotypic variance preclude
the separation of G and E: genetic–environmental covariance and genetic–environment
6. Additional supporting information can be found in the listing for this article in the Wiley Online
Library at
interaction. Genetic–environmental covariance occurs when certain genotypes are associ-
ated with particular environments (Plomin et al., 2012). Scholars have used the example of
children’s genetically influenced behavioral problems evoking harsh parental discipline,
which, in turn, could increase behavior problems. It is an ongoing matter of debate as
to how to classify this covariance in the calculation of heritability; certainly many, if not
most, scholars would agree that it fits neatly in neither the G nor the E category. Sev-
eral behavioral geneticists have argued that such gene–environment correlations should
be classified as genetic effects (e.g., Fowler, Baker, and Dawes, 2008; Segal and Johnson,
2009); however, as Rutter (2002: 4) noted, “it is misleading to suppose that just because
genetic factors influence the occurrence of an environmental risk factor, this must mean
that the risk process is genetically mediated. This assumption does not follow because
there is no necessary connection between the causes of the origin of a risk factor and its
mode of risk mediation.” This can be illustrated with real-world examples such as skin
pigmentation (Billings, Beckwith, and Alper, 1992; Joseph, 2004).
In the United States, a person genetically coded to have darker skin will experience a
different social environment, on average, than one with lighter skin. Darker skin pigmen-
tation is associated with exposure to social criminogenic risk factors, such as racial dis-
crimination, lower socioeconomic status (SES), and disadvantaged communities, among
others (e.g., Burt, Simons, and Gibbons, 2012; Sampson and Wilson, 1995). Surely, most
scientists believe that classifying offending that results from racial discrimination because
of skin pigmentation as solely “genetic” is preposterous. But even in less manifestly genet-
ically spurious cases, the classification is still not clear and rests on algebra and statistics,
which are not up to the task of classifying interactional biopsychosocial relationships (e.g.,
Spencer and Harpalani, 2004; Wahlsten, 1990).
In addition, the idea that the effects of nature and nurture can be partitioned into per-
centages requires one to assume that genetic and environmental influences are compet-
ing and noninteractional or at least that interaction effects are trivial (Wahlsten, 1990).
Mounting evidence suggests that this not the case. Rather than the rare exception, gene–
environment (G ×E) interactions seem to be the rule (e.g., Bagot and Meaney, 2010;
Charney, 2008; Rutter, 2007). As with many of the other problems we have discussed
in quantitative genetics, this criticism of heritability studies is not new. Indeed, Hogben
(1932: 201) first identified the problem that G ×E posed for efforts to estimate heri-
tability more than 80 years ago, criticizing the “false antithesis of heredity and environ-
ment,” while demonstrating the effects of G ×E on heritability estimates (for a historical
overview, see Tabery, 2007).
Although the debate over the problems posed by G ×E interactions for heritability
studies continued for years after Hogben’s initial critique, perhaps reaching its height dur-
ing the IQ controversies (e.g., Lewontin, 1974; Wahlsten, 1979), this matter is much less
controversial currently, as the idea that heritability studies attempt the impossible is more
widely accepted among scholars, including prominent behavioral geneticists who had pre-
viously been proponents of heritability studies (e.g., Rutter, 2006; Turkheimer, 2011). Im-
portantly, interactions between shared environmental effects and genetic influences are
captured in the heritability estimate (Alper and Beckwith, 1993; Miles and Carey, 1997;
Moffitt, 2005). Such G ×E interactions are widespread, especially for complex socially
mediated phenotypes such as crime, and they result in substantial underestimates of the
shared environment in twin studies (Rutter, 2002). In sum, from the classical genetics
perspective, strong evidence invalidates the rationale of heritability studies.
Perhaps the most decisive evidence against the validity of partitioning G and E influ-
ences comes from recent advances in molecular genetics, which demonstrate among other
things that the human genome is dynamic and environmentally responsive. A wealth of
evidence has accumulated in recent years supporting an interactional, bidirectional model
of gene and environmental function (e.g., Charney, 2012; Jablonka and Lamb, 2006).
These developments in molecular genetics have altered the scientific understanding of
heredity, the role of the gene, and the relationship between genotype and phenotype and
mark a shift from the classical “gene-centric” paradigm, on which existing behavioral ge-
netics methodologies are grounded, to a “postgenomic” view.
A key theme of the postgenomic view is the idea that genes do not stand outside the
developmental system of which they are a part. Humans develop through a process of
dynamic relations involving factors from the biological to the sociocultural levels of orga-
nization. Influences from all levels contribute integratively to the structure and function
of human development (Gottlieb, 2001; Greenberg, 2011; Lickliter, 2009). It is not merely
that genes and environments are independent and interact to influence a phenotype (as
in G ×E) but that genes and environments are not separate analytical entities in the first
place. Perhaps the most compelling evidence for the interactional relationship between
genes and environments in cellular function emerges from the relatively recent field of
study known as epigenetics. Research from this burgeoning field exemplifies the blurred
boundaries between environments and genes and illustrates the interactional relation-
ship among genes, biology, and environments even at the level of cellular molecules (e.g.,
Bagot and Meaney, 2010; Charney, 2012).
A Primer on Epigenetics
Our environment influences us by regulating our genetic activity. As noted, DNA is
not self-activating. DNA has to be transcribed to produce RNA and proteins, but before
DNA can be transcribed, it must be activated by the epigenome—the complex, biochem-
ical regulatory system that can turn on, silence (leave off), or change the transcriptional
activity of genes (Charney, 2012; Martiensen, Riggs, and Russo, 1996). As such, the mere
presence of a gene does not ensure that it is going to be used (activated by the cell),
and changes in the epigenome can alter the phenotype without any underlying change
to the genome (Bernstein, Meissner, and Lander, 2007). In this way, the epigenome reg-
ulates gene expression. As the epigenome is responsive to environmental input (both
internal and external to the cell), the environment influences gene expression through the
epigenome (e.g., Charney, 2012; Jablonka and Lamb, 2006).
Most gene regulation is a response to the immediate demands of the environment,
takes place in time spans ranging from minutes to weeks, and differs between specialized
cells, which contain identical DNA (Francis, 2011; Plomin et al., 2012). For example,
our liver cells will react one way to food poisoning, our intestinal cells will react another
way, and many cell types will not react at all. In recent years, however, researchers
have devoted much attention to a type of gene regulation that takes place over much
longer intervals. Epigenetics focuses on the mechanisms of gene regulation implicated
in changes in gene expression and phenotype, which can last for months, years, or across
the life span and can even be transmitted onto future generations (Bagot and Meaney,
2010; Bollati and Baccarelli, 2010; Charney, 2012).
Although new epigenetic mechanisms are still being uncovered, perhaps the best
understood fall into a class known as chromatin markers, often referred to as “epigenetic
markers.” Two of the most well known are histone modification and methylation.7These
markers affect cellular activity and phenotypes by influencing the accessibility of DNA
to transcription factors, thereby shaping gene expression. Evidence is accumulating for
the effects of environmental factors on these epigenetic markers, thereby illustrating
ways in which gene expression is influenced by the environment (Bagot and Meaney,
2010; Francis, 2011). Perhaps most important for our purposes, evidence for the effects
of epigenetic factors in neural development is accumulating rapidly. Studies have shown,
for example, that epigenetic modification influences perception, emotion, memory,
cognition and learning, and neural and behavioral plasticity (e.g., Allen, 2008; Crews,
2008). Indeed, gene expression is an essential step in the numerous cellular processes
involved in learning and memory and in altering the growth and organization of nerve
and especially neural cells (Kaplan and Rodgers, 2003; Molfese, 2011).
Neogenome and Heritability
What relevance do these recent postgenomic advances have for heritability studies?
The implications of these phenomena, which Charney (2012) collectively referred to as
the neogenome, are several. First, the research on epigenetics provides powerful evidence
that differences in phenotypes cannot be separated into genetic versus environmental
influences. The environment influences gene function, and the neogenome behaves like
neither the G nor E category (Charney, 2012). Instead of genome/neogenome versus en-
vironment, research evinces that the system is bidirectional and interactive. As Kaplan
and Rodgers (2003: 5) noted, “development is not merely a process involving a battle be-
tween nature (genes) and nurture (experience) but the interweaving of dynamic processes
within a system that is inseparably both the organism and its environment.” Epigenetics
illuminates the flexibility of the biological organism and the complexity of the relationship
between genes and the environment during development.
Grounded in the now outdated (oversimplified and incorrect) paradigm pitting nature
versus nurture and emphasizing differences in the genetic code as the way in which nature
shapes phenotypic variance, heritability studies have no place in a postgenomic paradigm.
As research in epigenetics makes clear, the mere presence of a gene as part of a geno-
type does not ensure that it is going to be activated; rather, changes in the environment
can alter the activation of genes and, in turn, shape or change a phenotype without any
underlying change to the genome (Bernstein, Meissner, and Lander, 2007). Moreover,
although evidence suggests that environmental events that occur early in life (e.g., pre-
natal factors or child abuse) tend to produce more pronounced epigenetic effects than
those that occur later (Bradley et al., 2008; Trembley, 2010), methylation and other epi-
genetic processes continue throughout the life span (Francis, 2011; Weaver, Meaney, and
Szyf, 2006). In sum, these recent advances provide clear evidence that the genome is im-
mensely flexible in its expression, and this expression is responsive to context, experience,
and developmental history. Several assumptions are necessary for heritability studies to
be meaningful, but the most crucial of these is the model of gene function as separable and
7. For a friendly introduction to epigenetics, see Francis (2011), and for a more technical overview,
see Charney (2012).
independent from the environment (Charney, 2012). Evidence is clear that genes do not
work in this way.
Heritability Studies: Concluding Remarks
Research has evinced that human behavior is a function of the interplay of biology and
the environment. As we have argued, we believe that this evidence clearly demonstrates
that quantitative genetics is a misguided endeavor that asks the wrong questions and uses
flawed methods to try to answer them.8What remains unclear is why this enterprise con-
tinues. The question of nature versus nurture no longer makes any sense whatsoever in the
context of modern genetics. These recent advances in molecular genetics “really should
be the final nail in heritability’s coffin” (Crusio, 2012: 362).
Moreover, it is now widely accepted—even by prominent behavioral geneticists—that
heritability estimates are of little practical relevance (Rutter, 1997; Turkheimer, 2011).
Even if estimated accurately, which is impossible for the reasons we mention, they do not
predict the likely developmental endpoints for individuals or groups, the consequences
of interventions, or the causal processes or mechanisms involved in phenotypic variations
(Rutter, 1997). Given all of these limitations, we recommend an end to heritability stud-
ies in criminology. In addition, we urge scholars to recognize that existing heritability
estimates are the result of models biased toward inflating genetic influences and under-
estimating shared environmental ones, and that using these rough and biased heritability
estimates to undergird specious debates about the irrelevance of shared environmental
factors, such as the family, neighborhoods, and SES (e.g., Harris, 1998; Rowe, 1994), does
a disservice to both scientific and public knowledge.
As we have noted, however, we are not trying to push the criminology community away
from biosocial theory and research. Instead, we recommend a different research agenda
for criminologists interested in doing biosocial research, one that is consistent with de-
velopments in the postgenomic paradigm, with its emphasis on the interactional, bidirec-
tional, and multifactorial relationships among genes, biology, environment, and behavior.
Our recommendations have two main underpinnings. First, we believe research must go
beyond simply recognizing or paying lip service to the fact that behavior, including crime,
is a function of the interplay of genetic and environmental influences. Too often, this in-
terplay has been wrongly translated into the misleading notion that “bad genes” (or a
“bad brain” in the case of neuroscience) combined with a “bad environment” produce
antisocial behavior, with the assumption of a causal influence from genes to brain to be-
havior. This notion of “bad genes” and a unidirectional effect from genes to brain and
behavior is misguided, as research suggests a bidirectional relationship from behavior
and environment to the brain through the epigenome. Furthermore, there is the usu-
ally unstated implication that genetic and environmental influences are equally important
(because heritability has been estimated at 50 percent) and that both exert main effects
in addition to their interaction. Accumulating molecular genetics research suggests, how-
ever, that environmental influences are more salient in explaining individual differences
8. Had we unlimited space we would discuss other problems with heritability studies, such as pheno-
type ambiguity and classification issues. Given space constraints, we point the reader elsewhere to
excellent discussions of these and other limitations (e.g., Charney, 2008; Duster, 1990, 2006; Joseph,
2006; Lewontin, Rose, and Kamin, 1984).
in complex phenotypes such as criminal behavior, with genetic influences usually being
limited to moderating the effect of the environment (e.g., Moffitt, Caspi, and Rutter, 2005;
Simons, Beach, and Barr, 2012).9
For example, candidate G ×E studies, which investigate the interaction between ge-
netic polymorphisms and environmental risk factors, invariably find that the genetic influ-
ence on criminal or related phenotypes is limited to its moderation of the environmental
effect. Caspi et al.’s (2002) now famous Science article, which marked the beginning of the
wave of G ×E studies, examined the effects of childhood mistreatment and the MAOA
polymorphism on antisocial behavior among a sample of White males. The MAOA risk
allele, which was found in 37 percent of the sample, had no direct effect on antisocial be-
havior but rather served to augment the likelihood of antisocial behavior in the presence
of childhood maltreatment. Importantly, regardless of MAOA status, childhood maltreat-
ment increased the likelihood of antisocial behavior. Subsequently, scores of recent can-
didate G ×E studies have shown that polymorphisms associated with various genes (e.g.,
DRD2, DRD4, MAOA, GABRA2, and 5-HTT) interact with environmental risk factors
to increase the likelihood of various types of internalizing and externalizing problems (see
Belsky and Pluess, 2009; Simons, Beach, and Barr, 2012, for reviews).10
In addition to research highlighting the interplay of biology and the environment, we
believe that work that seeks to elucidate processes or pathways of influence should be
given priority among biosocial criminologists. As Rutter (2002: 6) has noted, although G
×E studies are a useful first step toward this end, they are “the relatively easy part”; the
challenge is understanding “what these genes do” (i.e., their effects on proteins and the
process through which these proteins lead to various cellular and biological activities) and
understanding how these various biological activities shape responses to environmental
factors. An integrated, interdisciplinary research approach is required for a clear under-
standing of these mechanisms. Moreover, such work should be informed by evidence that
the neogenome is itself not a fixed or static entity but that it changes in response to envi-
ronmental conditions (Charney, 2012). In other words, biosocial researchers in criminol-
ogy need to incorporate the effects of the environment on changes in biological and/or
neogenomic factors into their models.
Several ongoing lines of biosocial research are consistent with this research agenda
(e.g., Walsh and Beaver, 2009). We wish to draw attention to epigenetics and social
neuroscience as promising avenues of future research in biosocial criminology. Work in
these two overlapping areas is rooted in the postgenomic paradigm and has the potential
to elucidate the complex biological pathways linking environmental factors to criminal
9. Although a few studies have shown that candidate genes such as MAOA have direct effects on
antisocial outcomes (e.g., Beaver, Delisi et al., 2010; Beaver et al., 2012), these models have sev-
eral limitations that vitiate the implications of their main effects findings, not the least of which is
omitted variable bias (e.g., Charney and English, 2013). Few to no environmental risk factors are
included in the models in these studies; therefore, both the main effect for the environment and
any G ×E effects are omitted from the analyses, and the model is misspecified.
10. Although out of the scope of this article, see Charney and English (2012, 2013) for an excellent
critique of G ×E studies, including their lack of replicated findings. Notably, G ×E studies are
based in the classical genetic paradigm, whereby a snapshot of DNA (not the epigenome) is ex-
amined and interacted with measures of the environment. Recent advances suggest the wisdom
of moving beyond this more static approach and attending to epigene–environment interplay and
interactions (e.g., Bollati and Baccarelli, 2010; Charney, 2012).
behavior. (To be sure, much of this work is preliminary and should be subject to various
critiques and to the caveat that the findings require replication before being accepted.)
In the final section of the article, we briefly discuss how these endeavors might be incor-
porated into criminological research in a manner that complements current work being
done in the field.
The postgenomic paradigm, with its emphasis on gene regulation and epigenetics, con-
tributed to a paradigm shift in neuroscience. Prior to the new century, conventional wis-
dom in neuroscience held that the adult mammalian brain is fixed in two respects: No new
neurons are created (as neurons die there is no replacement), and the fundamental struc-
ture of an individual’s brain is dictated by genes and is largely immutable (Doidge, 2007).
A wealth of recent evidence suggests that these conventional dogmas are incorrect and
informs the current neuroplasticity paradigm (Adophs, 2010). This new view holds that
neurogenesis—the manufacture of new neurons within the brain—takes place well into
old age (Kempermann, 2011). Furthermore, this view asserts that repeated experiences,
activities, and thoughts alter gene expression, which influences the wiring of the brain
(Adolphs, 2010; Davidson and McEwen, 2012). This process involves growth in cortical
(brain) space devoted to processes and functions that are used more frequently and a cor-
responding decrease in cortical space devoted to rarely performed processes. Thus, “the
very structure of our brain—the relative size of different regions, the strength of connec-
tions between one area and another—reflects the lives we have led” (Begley, 2007: 8–9).
There is strong evidence for neuroplasticity during childhood and some indication that
the types of childhood environments and psychological characteristics that criminologists
have linked to crime are associated with distinct neurological patterns (Davidson and
McEwen, 2012; Walsh and Bolen, 2012). Magnetic resonance imaging (MRI) and func-
tional magnetic resonance imaging (fMRI) studies indicate, for example, that exposure to
harsh and unpredictable childhood conditions (e.g., parental neglect) is associated with
greater volume and reactivity of the amygdala, a portion of the brain that is responsible
for vigilance and emotional responsiveness to threat (e.g., Mehta et al., 2009), and alter-
ation of the prefrontal cortex, the area responsible for executive control (Hanson et al.,
2010; Wilson, Hansen, and Li, 2011). The amygdala and prefrontal cortex, as well as their
interconnection, are implicated in emotional regulation (e.g., Wager et al., 2008), impul-
sivity (e.g., Kim and Lee, 2011; Raine, 2002b), and reactive aggression (Crowe and Blair,
2008). This research might be viewed as identifying the neurological underpinnings of
criminological research showing that childhood adversity decreases self-control.
Other neuroscientific studies have linked callused instrumental aggression to decreased
amygdala volume and low fear response (e.g., Fairchild et al., 2011; Raine, 2013). Impor-
tantly, evidence shows that this pattern of underresponsiveness may develop in response
to severe adversity (Del Guidice et al., 2012). Rutter (2012) labeled this “experience-
adaptive programming.” For example, surviving in extremely harsh environments re-
quires that one can act calmly even when danger is high (Del Guidice, Ellis, and Shirtcliff,
2011). Evidence suggests that methylation of the oxytocin receptor gene is associated
with behavior deemed callous and unemotional (Kumsta et al., in press). Thus, it seems
that severe adversity calibrates the biological system and molds the brain so that it be-
comes better adapted to deal with the dangers of a highly unpredictable and threatening
environment. Unfortunately, these adaptations also might increase the chances of aggres-
sion and crime as a response to perceived dangers and exploitation.
Moreover, evidence suggests not only that neuroplasticity takes place in childhood
and is influenced by environmental variables identified as important in traditional crim-
inology theory and research but also that it continues in adulthood (Bloss et al., 2010;
Davidson and McEwen, 2012). Studies, for example, reveal the growth of areas in the
brain associated with context and space among taxicab drivers (Maguire, Woolett, and
Spiers, 2006) and areas associated with finger movements in virtuoso violinists (Ebert
et al., 1995). Indeed, simply imagining oneself playing a simple five-note sequence re-
peatedly on the piano has been shown to increase the space in the motor cortex devoted
to the fingers (Pascual-Leone et al., 2005). Importantly, given our concern with deviant
behavior, hundreds of studies have shown that cognitive behavior therapy (CBT) can ef-
fectively change the (disturbed) thought processes and behaviors of adults with various
types of psychopathology (Hofmann et al., 2012). Moreover, strong evidence suggests
that these cognitive and behavioral changes are associated with neurological changes as-
sessed with MRI and fMRI (see Jokic-Begic, 2010). CBT also has been shown to be effec-
tive in changing cognitive and behavioral patterns of antisocial persons, including prison-
ers (Cullen and Jonson, 2011; Lipsey, Landenberger, and Wilson, 2007), and presumably
these changes brought about by CBT are mediated by changes in the brain.
Similarly, both mindfulness and compassion-focused meditation have been shown to
be effective in reducing anxiety, depression, and anger while enhancing emotion regula-
tion, empathy, and psychological well-being (e.g., Eberth and Sedimeier, 2012; Hofmann,
Grossman, and Hinton, 2011). These changes have been linked to changes in gene expres-
sion (e.g., Kaliman et al., 2014; Sharma et al., 2008) and in brain function and structure
(Hofmann, Grossman, and Hinton, 2011; Izel et al., 2013). As with CBT, evidence shows
that meditation promotes improved mood and behavior among prison inmates (Himel-
stein, 2011; Perelman, Miller, and Clements, 2012). Based on the findings with nonpris-
oner populations, these emotional and behavioral changes are likely mediated by interre-
lated changes in inmates’ thinking, gene expression, and neurological patterns.
In recent years, neuroscientists have published several studies of conduct-disordered
youth and a few of antisocial adults showing that these individuals manifest minor dif-
ferences in brain structure or function compared with more conventional individuals (see
Portnoy et al., 2013; Raine, 2013). This research often is viewed as contradicting criminol-
ogy’s emphasis on the importance of social environmental influences insofar as these neu-
rological differences are considered to be genetically determined. However, as we have
noted, genetic determinism is a flawed and outdated perspective. The postgenomic model,
with its emphasis on gene–environment interplay, argues for a social and developmental
neuroscience that recognizes the effect of environmental influences on brain structure and
function. It highlights the lifelong plasticity of the human developmental system and calls
for research that examines the manner in which environmental (social and psychological)
factors linked to crime are associated with epigenetic and neurological changes.
Summarizing, we believe that research in this postgenomic era suggests the following
biosocial paradigm: The social environment, especially during the critical periods of child-
hood and adolescence, becomes biologically embedded. Through processes like gene ex-
pression, biological systems such as the sympathetic nervous system and hypothalamic-
pituitary-adrenal axis are calibrated and the brain sculpted in a manner that prepares the
individual to function and survive in existing environmental conditions. These biological
systems, then, influence the way that an individual responds to subsequent situations and
Importantly, new environments and experiences can change gene expression and al-
ter biological systems. In other words, systems can be recalibrated and the brain can be
resculpted. As such, the new paradigm is a biosocial life-course perspective. Under this
paradigm, the challenge for biosocial criminologists entails: 1) identifying how adverse
environments sculpt an individual’s physiology, especially the brain, to respond to envi-
ronmental events with aggression, coercion, or violence, and 2) ascertaining how environ-
ments or experiences (whether naturally occurring changes or interventions) can change
the person’s biological systems and, in turn, their response to situations. Importantly, we
believe that not only are such integrative research efforts potentially valuable for under-
standing the biological processes through which the environment influences development
and behavior, but also such knowledge has the potential to inform social interventions
to prevent crime and humanely reform offenders (e.g., meditation, CBT, exercise, and
This marriage of social–behavioral science, epigenetics, and neuroscience offers a more
thorough multilevel and processual understanding of the etiology of offending, as well as
its onset, persistence, and desistence. For example, several criminological theories empha-
size the way that adverse childhood experiences give rise to cognitive/psychological traits
that increase the likelihood of offending (e.g., Moffitt, 1993; Simons and Burt, 2011). Re-
casting these theories within a postgenomic biosocial model would involve investigating
the biological pathways, including epigenetic and neurological changes, through which
such adverse experiences shape changes in cognitions and behavior. Such a model would
explain why criminogenic traits are difficult to change as they have epigenetic and neuro-
logical concomitants or underpinnings. But that is only part of the story.
Although the new paradigm focuses on the biological embedding of experiences, it also
emphasizes the possibility of change. As noted, rehabilitative programs have been shown
to modify “criminal ways of thinking” and such cognitive changes tend to be associated
with changes in neurological patterns. Consonant with these findings, criminological work
has shown that changes in crime are associated with changes in cognitions, including traits
and schemas. Giordano, Cernkovich, and Rudolph (2002) presented evidence that de-
sistence from crime is associated with a change in cognitive style, and Simons and Barr (in
press) recently reported that romantic relationships lead to desistance because they foster
a change in psychological traits and schemas. The new biosocial paradigm would suggest
that these cognitive changes are likely associated with epigenetic reprogramming and res-
culpting of brain circuitry. In highlighting the lifelong plasticity of the biopsychosocial
system, this approach emphasizes the ability of interventions to enhance the life course
for all individuals.
We have argued that there is compelling evidence that heritability studies are method-
ologically flawed, especially for complex social behaviors such as crime. We have argued
also that heritability studies are based on an oversimplified and incorrect model of gene
function and that the goal of partitioning genetic versus environmental influences on vari-
ance in phenotypes is biologically unsound. We therefore recommend an end to heritabil-
ity studies in criminology. Moreover, given the many flaws in heritability studies, we also
call for an end to the use of the oft-repeated version of the phrase: “We know from a
wealth of behavioral genetic studies that the heritability of [insert crime or related phe-
notype] is roughly 50 percent.” Based on the arguments and research discussed in this
article, it is apparent that we unequivocally do not know this to be the case. Furthermore,
no amount of quantitative genetic research can establish the validity of such heritability
estimates or their putative support for the irrelevance of shared environmental factors.
Technically flawed and conceptually unsound models—no matter how often published or
repeated—do not by virtue of their numbers make for sound evidence.
Notwithstanding our strong objections to heritability studies, we are enthusiastic about
truly biosocial research programs that focus on the interactional, bidirectional relation-
ship between social and biological factors and do not rely on an outdated gene-centric
paradigm. The postgenomic paradigm shift has brought about changes in genetics that
recognize the inextricable interplay between the environment and cellular processes in
the organism, as evidenced in research in epigenetics and neuroplasticity. The challenge
becomes harnessing the rapid advances in these areas to enhance our understanding of the
etiology of crime. As such, social scientists, geneticists, and neuroscientists are confronted
with the opportunity to take steps toward forging a broader interdisciplinary understand-
ing of how the concepts central to their disciplines influence human behavior. To achieve
this deeper, integrated understanding, social scientists need the assistance of geneticists
and neuroscientists, and vice versa. Transcending disciplinary silos to form such interdisci-
plinary alliances will enable criminologists to develop more comprehensive explanations
of criminality.
At the same time, we wish to underscore that criminologists who are uninterested in
biological influences or pathways and who prefer to focus on social influences should
understand that their work and the importance of social factors are not undermined by
biological findings at the current state of knowledge. Although some scholars have argued
that “the sociological conceptualization of environmental influences, which is the sine qua
non of sociological criminology, are both incorrect and not very useful” (Cleveland, Beek-
man, and Zheng, 2011: 249), we do not believe the evidence even faintly supports such an
argument. As we have noted, the identified effects of genetic and biological factors on
crime and related behaviors are consistently limited to the role of mediating or moderat-
ing the effects of environmental factors. Furthermore, it seems unlikely that social models
will be undermined in the future by biological research, as the more we learn about bio-
logical and genetic influences and mechanisms, the more consequential and intertwined
social influences become. However, as we point out, incorporating biological influences
can make our explanations more precise and more comprehensive; thus, as a discipline,
we ignore them at our own peril.
Asterisks () placed in front of references indicate studies included in table 1.
Adolphs, Ralph. 2010. Conceptual challenges and directions for social neuroscience. Neu-
ron 65:752–67.
Ainslie, Ricardo C. 1997. The Psychology of Twinship. New York: Rowman & Littlefield.
Alford, John R., Carolyn L. Funk, and John R. Hibbing. 2005. Are political orientations
genetically transmitted? American Political Science Review 99:153–67.
Allen, Nicholas D. 2008. Temporal and epigenetic regulation of neurodevelopmental
plasticity. Philosophical Transactions of the Royal Society B 363:23–38.
Alper, Joseph S., and Jonathan Beckwith. 1993. Genetic fatalism and social policy: The
implications of behavior genetics research. Yale Journal of Biology and Medicine
Bagot, Rosemary C., and Michael J. Meaney. 2010. Epigenetics and the biological basis
of gene x environment interactions. Journal of the American Academy of Child and
Adolescent Psychiatry 49:752–71.
Ball, Harriet A., Louise Arnseneault, Alan Taylor, Barbara Maughan, Avshalom Caspi,
and Terrie Moffitt. 2008. Genetic and environmental influences on victims, bullies,
and bully-victims in childhood. Journal of Child Psychology and Psychiatry 49:104–
Barnes, J. C., and Kevin Beaver. 2012. Extending research on the victim-offender over-
lap: Evidence from a genetically informed analysis. Journal of Interpersonal Violence
Barnes, J. C., and Brian B. Boutwell. 2012. On the relationship of past to future involve-
ment in crime and delinquency: A behavior genetic analysis. Journal of Criminal Jus-
tice 40:94–102.
Barnes, J. C., Brian B. Boutwell, and Kathleen A. Fox. 2012. The effect of gang member-
ship on victimization: A behavioral genetic explanation. Youth Violence and Juvenile
Justice 10:227–44.
Baumrind, Diane. 1993. The average expectable environment is not good enough: A re-
sponse to Scarr. Child Development 64:1299–317.
Beaver, Kevin M. 2008. The Nature and Nurture of Antisocial Outcomes. El Paso, TX:
Beaver, Kevin M. 2011a. Genetic influences on being processed through the criminal jus-
tice system: Results from a sample of adoptees. Biological Psychiatry 69:282–7.
Beaver, Kevin M. 2011b. The effects of genetics, the environment, and low self-control
on perceived maternal and paternal socialization: Results from a longitudinal sample
of twins. Journal of Quantitative Criminology 27:85–105.
Beaver, Kevin M., Brian B. Boutwell, and J. C. Barnes. 2013. Social support
or biosocial support? A genetically informed analysis of social support and its
relation to self-control. Criminal Justice and Behavior. Epub ahead of print.
Beaver, Kevin M., Brian B. Boutwell, J. C. Barnes, and Jonathon A. Cooper. 2009. The
biosocial underpinnings to adolescent victimization: Results from a longitudinal sam-
ple of twins. Youth Violence and Juvenile Justice 7:350–60.
Beaver, Kevin M., Brian B. Boutwell, J. C. Barnes, Matt DeLisi, and Michael G. Vaughn.
2013. Exploring the genetic origins of adolescent victimization in a longitudinal sam-
ple of adoptees. Victims & Offenders: An International Journal of Policy and Practice
Beaver, Kevin M., Matt DeLisi, Michael G. Vaughn, and J. C. Barnes. 2010. Monoamine
oxidase A genotype is associated with gang membership and weapon use. Compre-
hensive Psychiatry 51:130–4.
Beaver, Kevin M., Christopher J. Ferguson, and Jennifer Lynn-Whaley. 2010. The asso-
ciation between parenting and levels of self-control: A genetically informed analysis.
Criminal Justice and Behavior 37:1045–65.
Beaver, Kevin M., Chris L. Gibson, Michael G. Turner, Matt DeLisi, Michael G.
Vaughn, and Ashleigh Holand. 2011. Stability of delinquency peer associations: A
biosocial test of Warr’s sticky-friends hypothesis. Crime & Delinquency 57:907–27.
Beaver, Kevin M., J. Eagle Schutt, Brian B. Boutwell, Marie Ratchford, Kathleen
Roberts, and J. C. Barnes. 2009. Genetic and environmental influences on levels of
self-control and delinquent peer affiliation: Results from a longitudinal sample of ado-
lescent twins. Criminal Justice and Behavior 36:41–60.
Beaver, Kevin M., John Paul Wright, Brian B. Boutwell, J. C. Barnes, Matt DeLisi, and
Michael G. Vaughn. 2012. Exploring the association between the 2-repete allele of
the MAOA gene promoter polymorphism and psychopathic personality traits, arrests,
incarceration, and lifetime antisocial behavior. Personality and Individual Differences
Beaver, Kevin M., John Paul Wright, Matt DeLisi, and Michael G. Vaughn. 2008. Ge-
netic influences on the stability of low self-control: Results from a longitudinal sample
of twins. Journal of Criminal Justice 36:478–85.
Beckwith, Jon, and Corey A. Morris. 2008. Twin studies of political behavior: Untenable
assumptions? Perspectives on Politics 8:785–91.
Begley, Sharon. 2007. Train Your Mind, Change Your Brain: How a New Science Reveals
Our Extraordinary Potential to Transform Ourselves. New York: Ballatine.
Bell, A. Earl. 1977. Heritability in retrospect. Journal of Heredity 68:297–300.
Belsky, Jay, and Michael Pluess. 2009. Beyond diathesis stress: A differential susceptibil-
ity to environmental influences. Psychological Bulletin 135:885–908.
Bernstein, Bradley E., Alexander Meissner, and Eric S. Lander. 2007. The mammalian
epigenome. Cell 128:669–81.
Billings, Paul R., Jonathan Beckwith, and Joseph Alper. 1992. The genetic analysis of
human behavior: A new era? Social Science and Medicine 35:227–38.
Bloss, Erik B., William G. Janssen, Bruce S. McEwen, and John H. Morrison. 2010. Jour-
nal of Neuroscience 30:6726–31.
Bohman, Michael C. 1978. Some genetic aspects of alcoholism and crime: A population
of adoptees. Archives of General Psychiatry 35:269–76.
Boisvert, Danielle, Brian B. Boutwell, J. C. Barnes, and Jamie Vaske. 2013. Genetic and
environmental influences underlying the relationship between low self-control and
substance use. Journal of Criminal Justice 41:262–72.
Boisvert, Danielle, Brian B. Boutwell, Jamie Vaske, and Jamie Newsome. 2014. Genetic
and environmental overlap between delinquent peer association and delinquency in
adolescence. Criminal Justice and Behavior 41:58–74.
Boisvert, Danielle, John Paul Wright, Valerie Knopik, and Jamie Vaske. 2012. Ge-
netic and environmental overlap between low self-control and delinquency. Journal
of Quantitative Criminology 28:477–507.
Boisvert, Danielle, John Paul Wright, Valerie Knopik, and Jamie Vaske. 2013. A twin
study of sex differences in self-control. Justice Quarterly 30:529–59.
Bollati, Valentina, and Andrea Baccarelli. 2010. Environmental epigenetics. Heredity
Bouchard, Thomas J., David T. Lykken, Matthew McGue, Nancy L. Segal, and Auke
Tellegen. 1990. Sources of human psychological differences: The Minnesota Study of
Twins Reared Apart. Science 250:223–8.
Boutwell, Brian B., Cortney A. Franklin, J. C. Barnes, Amanda K. Tamplin, Kevin M.
Beaver, and Melissa Petkovsek. 2013. Unraveling the covariation of low self-control
and victimization: A behavior genetic approach. Journal of Adolescence 36:657–66.
Bradley, Rebekah G., Elisabeth B. Binder, Michael P. Epstein, Yilang Tang, Hemu P.
Nair, Wei Liu, Charles F. Gillespie, Tiina Berg, Mark Evces, D. Jeffrey Newport,
Zachary N. Stowe, Christine M. Heim, Charles B. Nemeroff, Ann Schwartz, Joseph
F. Cubells, and Kerry J. Ressler. 2008. Influence of child abuse on adult depression:
Moderation by the corticotrophin-releasing hormone receptor gene. Archives of Gen-
eral Psychiatry 65:190–200.
Brodzinsky, David M., Leslie M. Singer, and Anne M. Braff. 1984. Children’s understand-
ing of adoption. Child Development 55:869–78.
Burt, Callie H., Ronald L. Simons, and Frederick X. Gibbons. 2012. Racial discrimina-
tion, ethnic-racial socialization, and crime: A micro-sociological model of risk and
resilience. American Sociological Review 77:648–77.
Carey, Gregory, and David L. DiLalla. 1994. Personality and psychopathy: Genetic per-
spectives. Journal of Abnormal Psychology 103:32–43.
Caspi, Avshalom, Joseph McClay, Terrie E. Moffitt, Jonathan Mill, Judy Martin, Ian W.
Craig, Alan Taylor, and Richie Poulton. 2002. Role of genotype in the cycle of vio-
lence in maltreated children. Science 297:851–54.
Charney, Evan. 2008. Genes and ideologies. Perspectives on Politics 6:299–319.
Charney, Evan. 2012. Behavior genetics and postgenomics. Behavioral and Brain Sciences
Charney, Evan, and William English. 2012. Candidate genes and political behavior.
American Political Science Review 106:1–34.
Charney, Evan, and William English. 2013. Genopolitics and the science of genetics.
American Political Science Review 107:382–95.
Cleveland, H. Harrington, Charles Beekman, and Yao Zheng. 2011. The independence
of criminological “predictor” variables: A good deal of concerns and some answers
from behavioral genetic research. In The Ashgate Research Companion to Biosocial
Theories of Crime, eds. K. Beaver and A. Walsh. Burlington, VT: Ashgate.
Colodro-Conde, Lucia, Juan F. Sanchez-Romera, and Juan R. Ordonana. 2013. Heritabil-
ity of initiation and duration of breastfeeding behavior. Twin Research and Human
Genetics 16:575–80.
Conley, Dalton. 2011. Commentary: Reading Plomin and Daniels in the post-genomic
age. International Journal of Epidemiology 40:596–98.
Crews, David. 2008. Epigenetics and its implications for behavioral neuroendocrinology.
Frontiers in Neuroendocrinology 29:344–57.
Cronk, Nikole J., Wendy S. Slutske, Pamela A. F. Madden, Kathleen K. Bucholz, Wendy
Reich, and Andrew C. Heath. 2002. Emotional and behavioral problems among fe-
male twins: An evaluation of the equal environments assumption. Journal of the
American Academy of Child and Adolescent Psychiatry 41:829–37.
Crowe, Samantha L., and Robert James R. Blair. 2008. The development of antisocial
behavior: What can we learn from functional neuroimaging studies? Development and
Psychopathology 20:1145–59.
Crusio, Wim E. 2012. Heritability estimates in behavior genetics: Wasn’t that station
passed long ago? Behavioral and Brain Sciences 35:361–62.
Cullen, Francis T., and Cheryl Lero Jonson. 2011. Rehabilitation and treatment programs.
In Crime and Public Policy, eds. James Q. Wilson and Joan Petersilia. New York:
Oxford University Press.
Davidson, Richard J., and Bruce McEwen. 2012. Social influences on neuroplasticity:
Stress and interventions to promote well-being. Nature Neuroscience 15:689–95.
Del Guidice, Marco, Bruce J. Ellis, and Elizabeth A. Shirtcliff. 2011. The adaptive calibra-
tion model of stress responsivity. Neuroscience and Biobehavioral Reviews 35:1562–
Del Guidice, Marco, Benjamin J. Hinnant, Bruce J. Ellis, and Mona El-Sheikh. 2012.
Adaptive patterns of stress responsivity: A preliminary investigation. Developmental
Psychology 48:775–90.
DiLalla, Lisabeth F. 2002. Behavior genetics of aggression in children: Review and future
directions. Developmental Review 22:593–622.
DiLalla, Lisabeth F. 2004. Behavioral genetics: Background current research, and goals
for the future. In Behavior Genetics Principles: Perspectives in Development, Person-
ality, and Psychopathy, ed. Lisabeth F. DiLalla. Washington, DC: American Psycho-
logical Association.
Doidge, Norman. 2007. The Brain that Changes Itself. New York: Penguin Books.
Duster, Troy. 1990. Backdoor to Eugenics. New York: Routledge.
Duster, Troy. 2006. Behavioral genetics and crime, violence, and race. In Wrestling with
Behavioral Genetics: Science, Ethics, and Public Conversation, eds. Nancy Press, Au-
drey R. Chapman, and Erik Parens. Baltimore, MD: Johns Hopkins Press.
Eaves, L. J. 1988. Dominance alone is not enough. Behavior Genetics 18:27–33.
Ebert, Thomas, Christo Pantev, Christian Wienbruch, Brigitte Rockstroh, and Edward
Taub. 1995. Increased cortical representation of the fingers of the left hand in string
players. Science 270:305–7.
Eberth, Juilan, and Peter Sedimeier. 2012. The effects of mindfulness meditation: A meta-
analysis. Mindfulness 3:174–89.
Evans, David M., and Nicholas G. Martin. 2000. The validity of twin studies. GeneScreen
Fairchild, Graeme, Luca Passamonti, Georgina Hurford, Cindy C. Hagan, Elisabeth A.
H. von dem Hagen, Stephanie H. M. van Goozen, Ian M. Goodyer, and Andrew J.
Calder. 2011. Brain structure abnormalities in early-onset and adolescent-onset con-
duct disorder. American Journal of Psychiatry 168:624–33.
Faraone, Stephen V., Joseph Biederman, Barbara Weiffenbach, Tim Keith, Monica P.
Chu, Alix Weaver, Thomas J. Spencer, Timothy E. Wilens, Jean Frazier, Mario
Cleves, and Jun Sakai. 1999. Dopamine D4 gene 7-repeat allele and attention deficit
hyperactivity disorder. American Journal of Psychiatry 156:768–70.
Farber, Susan L. 1981. Identical Twins Raised Apart: A Reanalysis.NewYork:Basic
Fisher, R. A. 1918. The correlation between relatives on the supposition of Mendelian
inheritance. Transactions of the Royal Society of Edinburgh 52:399–433.
Fowler, James H., Laura A. Baker, and Christopher T. Dawes. 2008. Genetic variation in
political participation. American Political Science Review 102:233–48.
Francis, Richard C. 2011. Epigenetics: The Ultimate Mystery of Inheritance.NewYork:
Giordano, Peggy C., Stephen A. Cernkovich, and Jennifer L. Rudolph. 2002. Gender,
crime, and desistance: Toward a theory of cognitive transformation. American Journal
of Sociology 107:990–1064.
Goldberger, Arthur S. 1979. Heritability Economica 46:327–47.
Gottfredson, Michael R., and Travis Hirschi. 1990. A General Theory of Crime. Palo Alto,
CA: Stanford University Press.
Gottlieb, Gilbert. 1992. Individual Development and Evolution: The Genesis of Novel Be-
havior. New York: Oxford University Press.
Gottlieb, Gilbert. 2001. A developmental psychobiological system view: Early formula-
tion and current status. In Cycles of Contingency: Developmental Systems and Evo-
lution, eds. S. Oyama, Paul E. Griffiths, and Russell D. Gray. Cambridge, MA: MIT
Grayson, D. A. 1989. Twins reared together: Minimizing shared environmental effects.
Behavior Genetics 19:593–604.
Greenberg, Gary. 2011. The failure of biogenic analysis in psychology: Why psychology is
not a biological science. Research in Human Development 8:173–91.
Grove, William M., Elke D. Eckert, Leonard Heston, Thomas J. Bouchard, Nancy Segal,
and David T. Lykken. 1990. Heritability of substance abuse and antisocial behavior:
A study of monozygotic twins reared apart. Biological Psychiatry 27:1293–304.
Hanson, Jamie L., Moo K. Chung, Brian B. Avants, Elizabeth A. Shirtcliff, James C.
Gee, Richard J. Davidson, and Seth D. Pollak. 2010. Early stress is associated with
alterations in the orbitofrontal cortex: A tensor-based morphometry investigation of
brain structure and behavioral risk. The Journal of Neuroscience 30:7466–72.
Harris, Judith Rich. 1998. The Nurture Assumption: Why Children Turn out the Way They
Do. New York: Free Press.
Harris, Kathleen Mullan, Carolyn Tucker Halpern, Andrew Smolen, and Brett C. Haber-
stick. 2006. The national longitudinal study of adolescent health (Add Health) twin
data. Twin Research and Human Genetics 9:988–97.
Herrnstein, Richard J., and Charles Murray. 1994. The Bell Curve. New York: Free Press.
Himelstein, Samuel. 2011. Meditation research: The state of the art in correctional
settings. International Research in Offender Therapy and Comparative Criminology
Hofmann, Stefan G., Anu Asnaani, Imke Vonk, Alice Sawyer, and Angela Fang. 2012.
The efficacy of cognitive behavior therapy: A review of meta-analyses. Cognitive
Therapy and Research 36:427–40.
Hofmann, Stefan G., Paul Grossman, and Devon E. Hinton. 2011. Loving-kindness and
compassion meditation: Potential for psychological interventions. Clinical Psychology
Review 31:1126–32.
Hogben, Lancelot. 1932. Genetic Principles in Medicine and Social Science.NewYork:
Horowitz, Allan V., Tami M. Videon, Mark F. Schmitz, and Diane Davis. 2003. Rethink-
ing twins and environments: Possible social sources for assumed genetic influences in
twin research. Journal of Health and Social Behavior 44:111–29.
Izel, Britta K. H., Sara W. Lazar, Tim Gard, Zev Schuman-Olivier, David R. Vago, and
Ulrich Ott. 2013. How does mindfulness meditation work? Proposing mechanisms
of action from a conceptual and neural perspective. Perspectives on Psychological
Science 6:537–59.
Jablonka, Eva, and Marion J. Lamb. 2006. Evolution in Four Dimensions. Genetic, Epi-
genetic, Behavioral, and Symbolic Variation in the History of Life. Cambridge, MA:
MIT Press.
Jackson, Don D. 1960. A critique of the literature on the genetics of schizophrenia. In
The Etiology of Schizophrenia, ed. Don D. Jackson. New York: Basic Books.
Jensen, Arthur R. 1969. How much can we boost IQ and scholastic achievement? Harvard
Educational Review 39:1–123.
Jokic-Begic, Natasa. 2010. Cognitive-behavioral therapy and neuroscience: Towards
closer integration. Psychological Topics 19:235–54.
Joseph, Jay. 1998. The equal environments assumption of the classical twin method: A
critical analysis. The Journal of Mind and Behavior 19:325–58.
Joseph, Jay. 2001. Separated twins and the genetics of personality differences: A critique.
The American Journal of Psychology 114:1–30.
Joseph, Jay. 2004. The Gene Illusion: Genetic Research in Psychiatry and Psychology Un-
der the Microscope. New York: Algora.
Joseph, Jay. 2006. The Missing Gene: Psychiatry, Heredity, and the Fruitless Search for
Genes. New York: Algora.
Joseph, Jay 2010. Genetic research in psychiatry and psychology: A critical overview. In
Handbook of Developmental Science, Behavior, and Genetics, eds. Kathryn E. Hood,
Carolyn T. Halpern, Gary Greenberg, and Richard M. Learner. Malden, MA: Wiley-
Kaliman, Perla, Maria Jesus Alvarez-Lopez, Marta Cosin-Tomas, Melissa A. Rosenkranz,
Antoine Lutz, and Richard J. Davidson. 2014. Rapid changes in histone deacetylases
and inflammatory gene expression in expert meditators. Psychoneuroendocrinology
Kamin, Leon J. 1974. The Science and Politics of IQ. Potomac, MD: Lawrence Erlbaum.
Kamin, Leon J. 1981. Intelligence: The Battle for the Mind (H.J. Eysenck versus Leon
Kamin). London, U.K.: MacMillan.
Kamin, Leon J. 1985. Criminality and adoption [Letter to the editor]. Science 227:983.
Kaplan, Gisela, and Leslie J. Rodgers. 2003. Gene Worship: Moving Beyond the Na-
ture/Nurture Debate over Genes, Brain, and Gender. New York: Other Press.
Kempermann, Gerd. 2011. The pessimist’s and optimist’s view of adult neurogenesis. Cell
Kendler, Kenneth S. 1983. Overview: A current perspective on twin studies of schizophre-
nia. American Journal of Psychiatry 140:1413–25.
Kendler, Kenneth S., and Charles O. Gardner. 1998. Twin studies of adult psychiatric and
substance dependence disorders: Are they biased by differences in the environmental
experiences of monozygotic and dizygotic twins in childhood and adolescence? Psy-
chological Medicine 28:625–33.
Kendler, Kenneth S., Michael C. Neale, Ronald C. Kessler, Andrew C. Heath, and Lin-
don J. Eaves. 1993. A test of the equal-environment assumption in twin studies of
psychiatric illness. Behavior Genetics 23:21–7.
Keski-Rahkonen, Anna, Richard J. Viken, Jaakka Kaprio, Aila Rissanen, and Richard J.
Rose. 2004. Genetic and environmental factors in breakfast eating patterns. Behavior
Genetics 34:503–14.
Kim, Soyoun, and Daeyeol Lee. 2011. Prefrontal cortex and impulsive decision making.
Biological Psychiatry 69:1140–6.
Kumsta, Robert, Elisabeth Hummel, Frances S. Chen, and Markus Heinrichs. In press.
Epigenetic regulation of the oxytocin receptor gene: Implications for behavioral neu-
roscience. Frontiers in Neuroscience.
LaBuda, Michele C., Dace S. Svikis, and Roy W. Pickens. 1997. Twin closeness and co-
twin risk for substance use disorders: Assessing the impact of the equal environment
assumption. Psychiatric Research 70:155–64.
Lander, Eric S., Lauren M. Linton, Bruce Birren, Chad Nusbaum, Michael C. Zody,
Jennifer Baldwin, et al. 2001. Initial sequencing and analysis of the human genome.
Nature 409:860–921.
Largent, Mark A. 2007. Breeding Contempt: The History of Coerced Sterilization in the
United States. New Brunswick, NJ: Rutgers University Press.
Lewontin, Richard C. 1974. The analysis of variance and the analysis of causes. American
Journal of Human Genetics 26:400–11.
Lewontin, Richard C., Steven Rose, and Leon J. Kamin. 1984. Not in Our Genes: Biology,
Ideology, and Human Nature. New York: Pantheon Books.
Lickliter, Robert. 2009. The fallacy of partitioning: Epigenetics’ validation of the
organism-environment system. Ecological Psychology 21:138–46.
Lipsey, Mark W., Nana A. Landenberger, and Sandra J. Wilson. 2007. The effects of
cognitive-behavioral programs for criminal offenders. Campbell Systematic Reviews
Liu, Fan, Andreas Wollstein, Pirro G. Hysi, Georgina A. Ankra-Badu, Timothy D. Spec-
tor, Dandile Park, et al. 2010. Digital quantification of human eye color highlights
genetic association of three new loci. PLoS Genetics 6:e1000934.
Lush, Jay L. 1949. Heritability of quantitative characteristics in farm animals. Hereditas,
Maguire, Eleanor A., Katherine Woollett, and Hugo J. Spiers. 2006. London taxi drivers
and bus drivers: A structural MRI and neuropsychological analysis. Hippocampus
Martiensen, Robert A., Arthur D. Riggs, and Vincenzo E. A. Russo. 1996. Epigenetic
Mechanisms of Gene Regulation. Cold Spring Harbor, NY: Cold Springs Harbor Lab-
oratory Press.
Meaney, Michael J. 2010. Epigenetics and the biological definition of gene x environment
interactions. Child Development 81:41–79.
Mednick, Sarnoff A., William H. Gabrielli, and Barry Hutchings. 1984. Genetic influences
on criminal convictions: Evidence from an adoption cohort. Science 224:891–4.
Mednick, Sarnoff A., and Elizabeth S. Kandel. 1988. Congenital determinants of violence.
Bulletin of the American Academy of Psychiatry and the Law 16:101–9.
Mehta, Mitul A., Micole I. Golembo, Chiara Nosarti, Emma Colvert, Ashley Mota,
Steven C. R. Williams, Michael Rutter, and Edmund J. S. Sonuga-Barke. 2009. Amyg-
dala, hippocampal and corpus callosum size following severe early institutional depri-
vation: The English and Romanian Adoptees Study Pilot. Journal of Child Psychology
and Psychiatry 50:943–51.
Meier, Madeline H., Wendy S. Slutske, Andrew C. Heath, and Nicholas G. Martin. 2011.
Sex differences in the genetic and environmental influences on childhood conduct
disorder and adult antisocial behavior. Journal of Abnormal Psychology 120:377–88.
Miles, Donna R., and Gregory Carey. 1997. Genetic and environmental architecture of
human aggression. Journal of Personality and Social Psychology 72:207–17.
Moffitt, Terrie E. 1993. Adolescent-limited and life-course-persistent antisocial behavior:
A developmental taxonomy. Psychological Review 100:674–701.
Moffitt, Terrie E. 2005. Genetic and environmental influences on antisocial behaviors:
Evidence from behavioral-genetic research. Advances in Genetics 55:41–103.
Moffitt, Terrie E., Avshalom Caspi, and Michael Rutter. 2005. Strategy for investigating
interactions between measured genes and measured environments. Archives of Gen-
eral Psychiatry 62:473–81.
Molfese, David L. 2011. Advancing neuroscience through epigenetics: Molecular mecha-
nisms of learning and memory. Developmental Neuropsychology 36:810–827.
Morris-Yates, Allen, Gavin Andrews, P. Howie, and S. Henderson. 1990. Twins: A test of
the equal environments assumption. Acta Psychiatrica Scandinavica 81:322–6.
Munsinger, Harry. 1975. The adopted child’s IQ: A critical review. Psychological Bulletin
Neale, Michael C., and Lon R. Cardon. Methodology for Genetic Studies of Twins and
Families. Boston, MA: Kluwer Academic.
Pascual-Leone, Alvaro, Amir Amedi, Felipe Fregni, and Lotfi B. Merabet. 2005. The plas-
tic human brain cortex. Annual Review of Neuroscience 28:377–401.
Perelman, Abigayl, Sarah L. Miller, and Carl B. Clements. 2012. Meditation in a Deep
South prison: A longitudinal study of the effects of Vipassana. Journal of Offender
Rehabilitation 51:176–98.
Pinker, Steven. 2002. The Blank Slate: The Modern Denial of Human Nature.NewYork:
Plomin, Robert. 2011. Commentary: Why are children in the same family so different?
Non-shared environment three decades later. International Journal of Epidemiology
Plomin, Robert, and John C. DeFries. 1985. Origins of Individual Differences in Infancy:
The Colorado Adoption Project. Orlando, FL: Academic Press.
Plomin, Robert, John C. DeFries, Valerie S. Knopik, and Jenae M. Neiderhiser. 2012.
Behavioral Genetics, 6th ed. New York: Worth.
Plomin, Richard, John C. DeFries, Gerald E. McClearn, and Michael Rutter. 1997. Be-
havioral Genetics, 3rd ed. New York: Freeman.
Plomin, Robert, Robin Corley, John C. DeFries, and David W. Fulker. 1990. Individual
differences in television viewing in early childhood: Nature as well as nurture. Psy-
chological Science 1:371–7.
Portnoy, Jill, Yu Gao, Andrea L. Glenn, Sharon Niv, Melissa Peskin, Anna Rudo-Hunt,
Robert A. Schug, Yaling Yang, and Adrian Raine. 2013. The biology of childhood
crime and antisocial behavior. In Handbook of Life Course Criminology: Emerg-
ing Trends and Directions for Future Research, eds. Chris L. Gibson and Marvin D.
Krohn. New York: Springer.
Raine, Adrian. 1993. The Psychopathy of Crime: Criminal Behavior as a Clinical Disorder.
San Diego, CA: Academic Press.
Raine, Adrian. 2002a. Biosocial studies of antisocial behavior in children and adults: A
review. Journal of Abnormal Child Psychology 30:311–26.
Raine, Adrian. 2002b. The biological basis of crime. In Crime: Public Policies for Crime
Control eds. James Q. Wilson and Joan Petersilia. Oakland, CA: ICS Press.
Raine, Adrian. 2013. The Anatomy of Violence: The Biological Roots of Crime.NewYork:
Rende, Richard, Cheryl Slomkowski, Elizabeth Lloyd-Richardson, and Raymond Niaura.
2004. Sibling effects on substance use in adolescence: Social contagion and genetic
relatedness. Journal of Family Psychology 19:611–8.
Richardson, Ken. 2011. Wising up to the heritability of intelligence. GeneWatch 24:15–8.
Rowe, David C. 1994. On the Limits of Family Influence: Genes, Experience, and Behav-
ior. New York: Guilford Press.
Rowe, David C., and Kristen C. Jacobson. 1999. In the mainstream. In Behavioral Ge-
netics: The Clash of Culture and Biology, eds. Ronald Carson and Mark Rothstein.
Baltimore, MD: The John Hopkins University Press.
Rowe, David C., and D. Wayne Osgood. 1984. Heredity and sociological theories of delin-
quency: A reconsideration. American Sociological Review 49:526–40.
Rutter, Michael L. 1997. Nature-nurture integration: The example of antisocial behavior.
American Psychologist 52:390–8.
Rutter, Michael L. 2002. Nature, nurture, and development: From evangelism through
science toward policy and practice. Child Development 73:1–21.
Rutter, Michael L. 2006. Genes and Behavior: Nature-Nurture Interplay Explained.
Malden, MA: Blackwell.
Rutter, Michael L. 2007. Gene-environment interdependence. Developmental Science
Rutter, Michael L. 2012. Achievements and challenges in the biology of environmental
effects. Proceedings of the National Academy of Science 109:17149–53.
Rutter, Michael, Patrick Bolton, Richard Harrington, Ann Le Couteur, Hope Macdonald,
and Emily Simonoff. 1990. Genetic factors in child psychiatric disorders—I. A review
of research strategies. Journal of Child Psychology and Psychiatry 31:3–37.
Sampson, Robert J., and William Julius Wilson. 1995. Toward a theory of race, crime, and
urban inequality. In Crime and Inequality, eds. John Hagan and Ruth D. Peterson.
Stanford, CA: Stanford University Press.
Sandbank, Audrey. 1999. Twin and Triplet Psychology: A Professional Guide to Working
with Multiples. New York: Routledge.
Saudino, Kimberly J., Angelica Ronald, and Robert Plomin. 2005. The etiology of behav-
ior problems in 7-year-old twins: Substantial genetic influence and negligible shared
environmental influence for parent ratings and ratings by same and different teachers.
Journal of Abnormal Child Psychology 33:113–30.
Schaffner, Kenneth F. 2006. Behavior: Its nature and nurture, part 1. In Wrestling with Be-
havioral Genetics: Science, Ethics, and Public Conversation, eds. Nancy Press, Audrey
R. Chapman, and Erik Parens. Baltimore, MD: The Johns Hopkins Press.
Segal, Nancy L. 2000. Entwined Lives: Twins and What They Tell Us about Human Be-
Segal, Nancy L., and Wendy Johnson. 2009. Twin studies of general mental ability. In
Handbook of Behavior Genetics, ed. Y. Kim. London, U.K.: Springer.
Sharma, Himani, Palika Datta, Archna Singh, Sudip Sen, Narendra K. Bhardwaj, Vinod
Kochpuillai, and Neeta Singh. 2008. Gene expression in profiling in practitioners of
Sudarshan Kriya. Journal of Psychosomatic Research 64:213–8.
Simons, Ronald L., and Ashley Barr. In press. Shifting perspectives: Cognitive changes
mediate the impact of romantic relationships on desistance from crime. Justice
Simons, Ronald L., Steven R. H. Beach, and Ashley B. Barr. 2012. Differential suscepti-
bility to context: A promising model of the interplay of genes and the social environ-
ment. Advances in Group Processes 29:139–63.
Simons, Ronald L., and Callie Harbin Burt. 2011. Learning to be bad: Adverse conditions,
social schemas, and crime. Criminology 49:553–98.
Spencer, Margaret Beale, and Vinay Harpalani. 2004. Nature, nurture, and the question
of “how?” A phenomenological variant of ecological systems theory. In Nature and
Nurture, eds. Cynthia G. Coll, Elaine L. Bearer, and Richard M. Lerner. Mahwah, NJ:
Lawrence Erlbaum.
Steinhauer, Paul D. 1983. Issues of attachment and separation: Foster care and adoption.
In Psychological Problems of the Child in the Family, 2nd ed., eds. Paul Steinhaur and
Quentin Rae-Grant. New York: Basic Books.
Stoolmiller, Mike. 1998. Correcting estimates of shared environmental variance for range
restriction in adoption studies using a truncated multivariate normal model. Behavior
Genetics 28:429–41.
Stoolmiller, Mike. 1999. Implications of the restricted range of family environments for
estimates of heritability and nonshared environment in behavior-genetic adoption
studies. Psychological Bulletin 125:392–409.
Suhay, Elizabeth, and Nathan P. Kalmoe. 2010. The equal environment assumption in
twin studies of political traits: Social confounds and suggested remedies. Unpublished
Tabery, James. 2007. Biometric and developmental gene-environment interactions:
Looking back, moving forward. Development and Psychopathology 19:961–76.
Trembley, Richard E. 2010. Developmental origins of disruptive behaviour problems:
The “original sin” hypothesis, epigenetics and their consequences for prevention. The
Journal of Child Psychology and Psychiatry 51:341–67.
Turkheimer, Eric. 1991. Individual and group differences in adoption studies of IQ. Psy-
chological Bulletin 110:392–405.
Turkheimer, Eric. 2000. Three laws of behavior genetics and what they mean. Current
Directions in Psychological Science 9:160–4.
Turkheimer, Eric. 2011. Commentary: Variation and causation in the environment and
genome. International Journal of Epidemiology 40:598–601.
Vaske, Jamie, Danielle Boisvert, and John Paul Wright. 2012. Genetic and environmen-
tal contributions to the relationship between violent victimization and criminal behav-
ior. Journal of Interpersonal Violence 27:3213–35.
Viding, Essi, Alice P. Jones, Paul J. Frick, Terrie E. Moffitt, and Robert Plomin. 2008.
Heritability of antisocial behaviour at age 9: Do callous-unemotional traits matter?
Developmental Science 11:17–22.
Wager, Tor D., Matthew L. Davidson, Brent L. Hughes, Martin A. Lindquist, and Kevin
N. Ochsner. 2008. Prefrontal-subcortical pathways mediating successful emotion reg-
ulation. Neuron 59:1037–50.
Wahlsten, Douglas 1979. A critique of the concepts of heritability and heredity in behav-
ioral genetics. In Theoretical Advances in Behavioral Genetics, eds. Joseph R. Roye
and Leendert P. Mos. Alphen aan den Rijn, the Netherlands: Sijthoff & Noordhoff.
Wahlsten, Douglas. 1990. Insensitivity of the analysis of variance to heredity-environment
interaction. Behavioral and Brain Sciences 13:109–20.
Walsh, Anthony, and Kevin M. Beaver. 2009. Biosocial Criminology: New Directions in
Theory and Research. New York: Routledge.
Walsh, Anthony, and Jonathan D. Bolen. 2012. The Neurobiology of Criminal Behavior:
Gene-Brain-Culture Interaction. Burlington, VT: Ashgate.
Walters, Glenn D., and Thomas W. White. 1989. Heredity and crime: Bad genes or bad
research? Criminology 27:455–86.
Weaver, Ian C. G., Michael J. Meaney, and Moshe Szyf. 2006. Maternal care effects on
the hippocampal transcriptome and anxiety-mediated behaviors in the offspring that
are reversible in adulthood. Proceedings of the National Academy of Sciences of the
United States of America 103:3480–5.
Wilson, Paul T. 1934. A study of twins with special reference to heredity as a factor deter-
mining differences in environment. Human Biology 6:324–54.
Wilson, Kathryn R., David J. Hansen, and Ming Li. 2011. The traumatic stress response in
child maltreatment and resultant neuropsychological effects. Aggression and Violent
Behavior 16:87–97.
Wright, John, Kevin Beaver, Matt DeLisi, and Michael Vaughn. 2008. Evidence of neg-
ligible parenting influences on self-control, delinquent peers, and delinquency in a
sample of twins. Justice Quarterly 25:544–69.
Callie H. Burt is an assistant professor in the School of Criminology and Criminal Jus-
tice at Arizona State University. Her research focuses on criminological theories, with
particular emphasis on elucidating the social, psychological, and biological mechanisms
through which social factors, such as racial discrimination, community crime, parenting
practices, and peers, influence criminal offending across the life course. Her recent work
also has appeared in American Journal of Sociology,American Sociological Review, and
Justice Quarterly.
Ronald L. Simons is a foundation professor in the School of Criminology and Criminal
Justice at Arizona State University and a fellow in the Institute for Behavioral Research
at the University of Georgia. His research has focused on onset, amplification, and desis-
tance from various externalizing and internalizing problems. Recently, he has expanded
his research program to include genes and other biological variables.
Additional Supporting Information may be found in the online version of this article at
the publisher’s web site:
... In the end, this inability to compare heritability estimates across populations is no great loss. Trying to refine, as we and others have argued, a conceptually and methodologically problematic and ultimately not very useful (outside of controlled breeding) heritability estimate is a wasteful distraction (Burt & Simons, 2014;Turkheimer, 2011). As Turkheimer (2011) averred more than a decade ago: "In the real world of humans, in a given context everything is heritable to some extent and environmental to some other extent, but the magnitudes of the proportions are variable from situation to situation, and have nothing whatsoever to do with the causal properties of genes and environments for the trait in question, unless one is interested in the pointless null hypothesis that one of the components is zero" (p. ...
Uchiyama et al. rightly consider how cultural variation may influence estimates of heritability by contributing to environmental sources of variation. We disagree, however, with the idea that generalisable estimates of heritability are ever a plausible aim. Heritability estimates are always context-specific, and to suggest otherwise is to misunderstand what heritability can and cannot tell us.
... In the end, this inability to compare heritability estimates across populations is no great loss. Trying to refine, as we and others have argued, a conceptually and methodologically problematic and ultimately not very useful (outside of controlled breeding) heritability estimate is a wasteful distraction (Burt & Simons, 2014;Turkheimer, 2011). As Turkheimer (2011) averred more than a decade ago: "In the real world of humans, in a given context everything is heritable to some extent and environmental to some other extent, but the magnitudes of the proportions are variable from situation to situation, and have nothing whatsoever to do with the causal properties of genes and environments for the trait in question, unless one is interested in the pointless null hypothesis that one of the components is zero" (p. ...
We need better understanding of functional differences of behavioral phenotypes across cultures because cultural evolution (e.g., temporal changes in innovation within populations) is less important than culturally molded phenotypes (e.g., differences across populations) for understanding gene effects. Furthermore, changes in one behavioral domain likely have complex downstream effects in other domains, requiring careful parsing of phenotypic variability and functions.
... These structures also criminalize some harmful actions and not others (e.g., Quinney, 1975;Schur, 1969). For example, genetically influenced traits such as skin pigmentation impose significant constraints on development (and the likelihood of crime) for reasons that are neither caused by nor reducible to genes but significantly shaped by social relations that can only be understood in a sociohistorical context (Burt and Simons, 2014). ...
Full-text available
Professing interactionist bio + social terminology, contemporary biocriminology asserts a break from its biologically essentialist past. Assurances notwithstanding, whether biocriminology has undergone a decisive paradigm shift rejecting notions of biological criminals and bad brains remains uncertain. Unfortunately, discussions of biocriminology's assumptions are mired in politics, obscuring important scientific issues. Motivated to clarify misunderstanding, I address the ontoepistemology of biocriminology from a scientific realist perspective. Drawing on familiar notions of crime as a social construction, I explain how and why biocriminology's ontoepistemology is inconsistent with the social reality of crime for scientific not ideological reasons. I explain that recognizing crime as a social construction does not imply that crime is not real or objective and cannot be studied scientifically. On the contrary, the irreducibly social nature of crime requires that scientific realists reject assumptions of “biological crime” as well as the biologically reductionist epistemology on which biocriminology depends.
... Since the 1960s, empirical evidence has accumulated that monozygotic twins live in more similar social environments than dizygotic twins. For example, they are more likely to be treated the same by their parents, have the same friends, be in the same class, spend time together, be more attached to each other through their whole life, etc. (Joseph 2013;Burt and Simons 2014). Furthermore, the prenatal (intrauterine) environment of monozygotic and dizygotic twins is different: the prenatal environments of MZ twins (who often share the same placenta) are more similar than those of DZ twins (who never share the same placenta). ...
Full-text available
In this paper, we explain the concept of heritability and describe the different methods and the genotype–phenotype correspondences used to estimate heritability in the specific field of human genetics. Heritability studies are conducted on extremely diverse human traits: quantitative traits (physical, biological, but also cognitive and behavioral measurements) and binary traits (as is the case of most human diseases). Instead of variables such as education and socio-economic status as covariates in genetic studies, they are now the direct object of genetic analysis. We make a review of the different assumptions underlying heritability estimates and dispute the validity of most of them. Moreover, and maybe more importantly, we show that they are very often misinterpreted. These erroneous interpretations lead to a vision of a genetic determinism of human traits. This vision is currently being widely disseminated not only by the mass media and the mainstream press, but also by the scientific press. We caution against the dangerous implication it has both medically and socially. Contrarily to the field of animal and plant genetics for which the polygenic model and the concept of heritability revolutionized selection methods, we explain why it does not provide answer in human genetics.
... The additional strength of the study comes from the very traits for which the components of variance were estimated. Namely, critics of twin models state that the assumption of equal environment is not valid, because MZ twins receive a more similar treatment, and that heritability estimates are therefore not trustworthy (e.g., Burt & Simons, 2014;Willems et al., 2019). Here, we want to emphasize that dermatoglyphs are the kind of traits that are formed during the fetal phase of development; at birth they are completely formed and remain stable throughout life. ...
Full-text available
Dermatoglyphs are epidermal ridge configurations on the fingers, palms and soles that are formed during fetal development, and therefore only the intrauterine environment can have any influence on their formation. This study aims at investigating the genetic and environmental contribution in determining quantitative dermatoglyphic traits in 32 monozygotic (MZ) and 35 dizygotic (DZ) same-sex twins from the Albanian population of Kosovo. All genetic analyses were run in the statistical program Mx. After assumptions testing, based on the pattern of MZ–DZ correlations, univariate models were fitted to the data in order to estimate additive genetic (A), common (C) and individual (E) environmental influences for all variables. The exception was the atd-angle for which a model with nonadditive genetic (D) influences was tested, since DZ correlations were less than half of MZ correlations. Goodness of fit of the full ACE or ADE model was compared to the saturated model. The fit of nested models (AE, CE, DE or E) was compared to the full models (ACE or ADE). Our results indicate that additive genetic component strongly contributes to individual differences in finger ridge counts (49−81%), and weakly (0−50%) on the formation of the palmar ridge counts between the palmar triradii a, b, c, and d. The specific pattern found for the atd-angle implies the impact of a nonadditive genetic component, possibly the effect of a major gene. Further, more powered studies are needed to confirm this pattern, especially for resolving the issue of the huge difference in MZ and DZ twin similarity for the atd-angle palmar trait.
... Biosocial criminology views crime and antisocial behaviors as biological and social phenomena, since biology affects behavior and the environment affects biology (Rocque & Posick, 2017;Walsh, 2009;Walsh, 2014). Over the years, many scholars have discussed the future of biosocial criminology and how genetics could also influence delinquent and criminal behavior (Beaver et al., 2009;Beaver et al., 2015;Burt & Simons, 2014;Carrier & Walby, 2014;Carrier & Walby, 2015a;Carrier & Walky, 2015b;Gajos et al., 2016;Heylen et al., 2015;Posick et al., 2015;Rafter, 1997;Rafter, 2004;Rafter, 2006;Rafter, 2008a;Rafter, 2008b;Rafter et al., 2016;Raine, 2002;Rocque et al., 2012;Rocque et al., 2014;Vaughn, 2016;Walsh, 2009;Walsh & Wright, 2015;Wright & Boisvert, 2009). Social learning theory and low self-control theory are two of the most empirically supported criminological theories (Pratt & Cullen, 2000). ...
The prevalence of cyber-dependent crimes and illegal activities that can only be performed using a computer, computer networks, or other forms of information communication technology has significantly increased during the last two decades in the USA and worldwide. As a result, cybersecurity scholars and practitioners have developed various tools and policies to reduce individuals' and organizations' risk of experiencing cyber-dependent crimes. However, although cybersecurity research and tools production efforts have increased substantially, very little attention has been devoted to identifying potential comprehensive interventions that consider both human and technical aspects of the local ecology within which these crimes emerge and persist. Moreover, it appears that rigorous scientific assessments of these technologies and policies "in the wild" have been dismissed during the process of encouraging innovation and marketing. Consequently, governmental organizations, public and private companies allocate a considerable portion of their operations budgets to protecting their computer and internet infrastructures without understanding the effectiveness of various tools and policies in reducing the myriad of risks they face. Unfortunately, this practice may complicate organizational workflows and increase costs for government entities, businesses, and consumers. The success of the evidence-based approach in improving the performances of a wide range of professions (for example, medicine, policing, and education) leads us to believe that an evidence-based cybersecurity approach is critical for improving cybersecurity efforts. This book seeks to explain the foundation of the evidence-based cybersecurity approach, reviews its relevance in the context of existing security tools and policies, and the authors provide concrete examples of how adopting this approach could improve cybersecurity operations and guide policymakers' decision-making process. The evidence-based cybersecurity approach explained aims to support security professionals', policymakers', and individual computer users' decision-making processes regarding the deployment of security policies and tools by calling for rigorous scientific investigations of the effectiveness of these policies and mechanisms in achieving their goals in protecting critical assets. This book illustrates how this approach provides an ideal framework for conceptualizing an interdisciplinary problem like cybersecurity because it stresses moving beyond decision-makers political, financial, social backgrounds, and personal experiences when adopting cybersecurity tools and policies. This approach is also a model in which policy decisions are made based on scientific research findings.
Uchiyama et al. propose a unified model linking cultural evolutionary theory to behavior genetics (BG) to enhance generalizability, enrich explanation, and predict how social factors shape heritability estimates. A consideration of culture evolution is beneficial but insufficient for purpose. I submit that their proposed model is underdeveloped and their emphasis on heritability estimates misguided. I discuss their ambiguous conception of culture, neglect of social structure, and the lack of a general theory in BG.
Psychopathology is the result of a complex interaction between one's genetics and environmental experiences that unfolds over the course of the lifespan, varying with context (family, neighborhood) and developmental stage. This entry will focus on how genetic influences and the parent-child relationship contribute to the development of youth psychopathology, and will consider future directions for the broader neighborhood context.
Twin and adoption studies compare the similarities of people with differing degrees of relatedness to estimate genetic and environmental contributions to trait population variance. The analytic workhorse of these kinds of variance-focused designs is the intraclass correlation, which estimates similarity between pairs of individuals. Group means, by contrast, play no overt role in estimating genetic and environmental influences. Although this focus on variance has made very important contributions to understanding psychological characteristics, we contend that the exclusion of mean effects from behavioral genetic designs may have obscured key environmental influences and impeded full appreciation of the ubiquity and nature of gene–environment interplay in human outcomes. We provide empirical examples already in the literature and a theoretical framework for thinking through the incorporation of mean effects using largely forgotten, non-Mendelian theory regarding how genes influence human outcomes. We conclude that the field needs to develop models capable of fully incorporating mean effects into twin and adoption studies.
Genetic research in the area of child and adolescent psychiatry has increased greatly in the past two decades. Through the establishment of large-scale twin studies, which combine longitudinal data collection from early childhood through to adolescence with detailed phenotypic assessments, evidence from twin research has substantially changed our understanding of the etiology and development of childhood psychiatric disorders and related traits. Based on more than 1000 published research articles, it is now clear that genetic factors play an important role in all psychiatric disorders. Yet, the development of child psychiatric problems is the result of a complex interplay between genetic sensitivity and environmental risk-factors. Twin studies have not only established the relative influence of genetic and environmental factors in the etiology of child psychiatric disorders and traits but also clarified developmental stability and change, causes of comorbidity, and begun to unravel complex gene–environment interplay. In this chapter, we provide an overview of what can be concluded about the etiology and development of childhood psychiatric disorders and traits from over two decades of twin research and discuss future direction for the field.
Genetic approaches have improved our understanding of the neurobiological basis of social behavior and cognition. For instance, common polymorphisms of genes involved in oxytocin signaling have been associated with sociobehavioral phenotypes in healthy samples as well as in subjects with mental disorders. More recently, attention has been drawn to epigenetic mechanisms, which regulate genetic function and expression without changes to the underlying DNA sequence. We provide an overview of the functional importance of oxytocin receptor gene (OXTR) promoter methylation and summarize studies that have investigated the role of OXTR methylation in behavioral phenotypes. There is first evidence that OXTR methylation is associated with autism, high callous-unemotional traits, and differential activation of brain regions involved in social perception. Furthermore, psychosocial stress exposure might dynamically regulate OXTR. Given evidence that epigenetic states of genes can be modified by experiences, especially those occurring in sensitive periods early in development, we conclude with a discussion on the effects of traumatic experience on the developing oxytocin system. Epigenetic modification of genes involved in oxytocin signaling might be involved in the mechanisms mediating the long-term influence of early adverse experiences on socio-behavioral outcomes.
![Figure][1] In this book Mike Rutter sets out to explain how genes might influence behaviour and how this might be important in understanding the causal pathways leading to various behavioural traits and psychiatric disorders. This is an ambitious and challenging project, not just because
Preface. List of Figures. List of Tables. 1. The Scope of Genetic Analyses. 2. Data Summary. 3. Biometrical Genetics. 4. Matrix Algebra. 5. Path Analysis and Structural Equations. 6. LISREL Models and Methods. 7. Model Fitting Functions and Optimization. 8. Univariate Analysis. 9. Power and Sample Size. 10. Social Interaction. 11. Sex Limitation and GE Interaction. 12. Multivariate Analysis. 13. Direction of Causation. 14. Repeated Measures. 15. Longitudinal Mean Trends. 16. Observer Ratings. 17. Assortment and Cultural Transmission. 18. Future Directions. Appendices: A. List of Participants. B. The Greek Alphabet. C. LISREL Scripts for Univariate Models. D. LISREL Script for Power Calculation. E. LISREL Scripts for Multivariate Models. F. LISREL Script for Sibling Interaction Model. G. LISREL Scripts for Sex and GE Interaction. H. LISREL Script for Rater Bias Model. I. LISREL Scripts for Direction of Causation. J. LISREL Script and Data for Simplex Model. K. LISREL Scripts for Assortment Models. Bibliography. Index.