Research at the AFTA Annual Meeting by Martha Edwards, Ph.D.
Antisocial Behavior and Depression: The Roles of Environment and Genes A Report on the Presentation by David Reiss, M.D., for the Research Interest Group
Research over the past few decades on both adolescent antisocial behavior and adult depression has consistently found associations between the quality of the social environment and these patterns of adjustment. In particular, conflictual, hostile, undermining, and coercive parent-child relationships tend to be associated with adolescents' antisocial behavior; and marital dissatisfaction and conflict are associated with depression in wives. The ubiquity and consistency of these findings have led to confident conclusions about the importance of environmental processes on these aspects of behavior.
At the same time, however, behavioral geneticists have been conducting their own research and reaching different conclusions—that antisocial behavior, for example, is substantially influenced by genes. These seemingly contradictory conclusions have been the focus of study of Dr. David Reiss, our presenter for the Research Interest Group, for the past 10 years and will continue to be a focus for many years to come, given his most recent funding of two multi-site, longitudinal studies.
In his presentation, Reiss described two studies designed to explore the influence of both family processes and of genes on the adjustment of individuals (Reiss et al., 2000; Reiss et al., 2001a, 2001b). Furthermore, these studies were designed to illuminate the differential influence of shared (common to all family members) and non-shared (unique to family members) environmental factors.
Three conclusions drawn from these studies are: (1) genes "recruit" family interactions, i.e., the genetically influenced behavior of one family member evokes predictable reactions from other family members; (2) genes require certain interactional processes in order to become fully expressed; and (3) the expression of genetic influence on behavior waxes and wanes across time.
In twin studies, identical and fraternal twins are compared to determine if a particular characteristic is genetically influenced. If the identical twins are more similar than the fraternal twins, then one can conclude that genetics plays an important role. In a study of adolescents (Reiss et al., 2000) Reiss and his colleagues (including behavioral geneticist Robert Plomin and developmental psychologist Mavis Hetherington) creatively extended this analytic logic and designed a study involving both intact and blended families, with six groups of subjects and four different levels of genetic relatedness. (These included: identical twins, fraternal twins, and full siblings in intact families, and full siblings, half siblings and unrelated siblings in blended families.)
With the brief time available, Reiss could describe only a tiny piece of this huge study and take us through some of the design and analytic logic used to tease apart the environmental and genetic influences on adolescents' antisocial behavior. Specifically, he focused on examining the relationships among genetics, parent-child relationships, and antisocial behavior. In the context of this discussion, he also focused on three methodological aspects: (1) collecting multiple measures (e.g., from the adolescents, the parents, and observers), and combining them into measures that were more reliable than any single measure; (2) examining the "genetic cascade" to determine the degree of genetic influence on family processes or child adjustment; and (3) using cross-correlations to estimate the amount of overlap between genetic influences on family processes and genetic influences on child adjustment.
The data from this study did, indeed, replicate the findings from previous, non-genetic developmental studies of family processes. Specifically, conflict and negativity in the parent-child relationship were related to adolescent antisocial behavior, accounting for approximately 50% of the variance.
However, the data also replicated findings from behavioral genetics studies, demonstrating a strong genetic influence on antisocial behavior (approximately 65% of the variance).
If Reiss et al. had stopped here, they would not have been able to answer the question of the differential effect that environment and genes have on adolescent adjustment. Since they collected both family process and genetic data, they could look in detail at the relationships between genetics and family processes. If genes influence both family processes and adjustment, this might account for the relationship between adolescents' family environment and their antisocial behavior.
Indeed, they found substantial evidence for genetic influence on family relationships, specifically observing that parental negativity in the parent-child relationship was highly correlated in identical twins, less so in fraternal twins and full siblings, and less for half siblings and unrelated siblings (the genetic cascade). But it is possible that half of adolescents' antisocial behavior is explained by their genes and the other half is attributable to their family relationships. On the other hand, there could also be considerable overlap between the genetic and family influences on adjustment. Through statistical processes, Reiss and his team were able to calculate the degree of overlap, finding it quite high.
One might be tempted, therefore, to conclude that genetics explains the lion's share of adolescent antisocial behavior. But this influence can be conceived of in two ways—a passive mechanism focusing only on the genes themselves, and an active-evocative mechanism, involving the family environment.
The passive mechanism rests on the fact that adolescents share 50% of their parents' genes. Therefore, the same genes responsible for children's antisocial behavior may also be responsible for their parents' negativity. In the active-evocative mechanism, children's genetically driven behavior evokes predictable reactions in other family members. For example, children with difficult temperaments often evoke negative reactions in their parents. These reactions further exacerbate the child's difficult tendencies, leading to both persistent conflictual parent-child relationships as well as increasingly extreme negative behavior from the child, culminating in antisocial behavior. Since these social processes and outcomes were initiated by genes, one might describe this by saying that the genes recruit family responses and that these responses may be necessary for the full expression of these genes.
A typical assumption about genetic influence is that genes exert a stable influence over the course of a person's life. Since the families in this study participated at two points in time, separated by three years, the stability of genetic influence across time could be assessed. First, antisocial behavior, in and of itself, was relatively stable from time 1 to time 2 (r=.61). The influence of genes on this stability was assessed by using "cross-correlations," i.e., predicting the score of one sibling at time 2 by the other sibling's score at time 1. If the cross-correlation is high for genetically identical siblings and lower for siblings who share fewer genes, then one can conclude that genetic factors play a role in the stability of antisocial behavior. The cross-correlations show that genetic influences play a large role (66%) in the stability of antisocial behavior over time. At the same time, however, the genetic influences on the change in antisocial behavior are also quite high (75%). There are several possible explanations for this finding. One is that particular genes may switch on and off in the course of development across adolescence. Genes operating in early adolescence may actually switch off, to be replaced by newly activated genes later in adolescence.
The next study Reiss undertook focused on the other side of the parent-child relationship, examining the genetic and social influences on the psychological adjustment of twin mothers of adolescents (Reiss et al., 2001a). The initial results of this study are just being published (Reiss et al., 2001b). In this study, there were notable genetic influences on marital satisfaction as reported by the twin moms and by their husbands and as directly observed. But these genetic influences were uncorrelated with genetic influences on mother's depression. The associations between marital satisfaction and depression were due to common environmental influences, not common genetic influences.
Taken together, the results of these two studies suggest that when genetic influences are newly expressed (as in adolescence), relationships appear to be the medium for their expression. When genetic influences are no longer new (as is the case more in adulthood), relationships shape individual adjustment more independently of genetic influences.
Reiss's next endeavor to further explore these extremely complex relationships is an experimental study of adopted toddlers. Since toddlerhood appears to be the other developmental period where genetic influences both wane and wax, this period provides an opportunity to look at these processes and the effect of the environment on gene expression. Furthermore, correspondences between the birth mother and the child will provide information on the genetic influences transmitted across time. Ultimately, this study will also involve an intervention designed to influence parenting processes to test whether changes in these processes will influence changes in child outcomes.
This article is a very modest attempt to report on the monumental work that Reiss and his colleagues have done to unravel the complex threads of the nature-nurture tapestry. But a little knowledge may, in this case, be a dangerous thing. I urge you to read the original sources to gain the complete picture of what they undertook and the complex conclusions they drew.
Reiss's final message was concerned with the importance of collaborative endeavors. Genetically informed research designs that allow us to tease apart the influence of nature and nurture are critical for the field of family therapy. But equally important are systemically informed research designs that will enable the adequate exploration of social processes on development and functioning. Family therapists need to "be at the table" with behavioral geneticists and developmental psychologists and others in the design of these studies and in the interpretative discussions of their results.
References
Reiss, D., with Neiderhiser, J.M., Hetherington, E.M., and Plomin, R. (2000). The Relationship Code: Deciphering genetic and social influences on adolescent development. Cambridge, MA: Harvard University Press.
Reiss, D., Cederblad, M., Pedersen, N.L., Lichtenstein, P., Elthammar, O., Neiderhiser, J.M., & Hansson, K. (2001). Genetic probes of three theories of maternal adjustment: II. Genetic and environmental influences. Family Process, 40:261-272.
Reiss, D., Pedersen, N.L., Cederblad, M., Lichtenstein, P., Hansson, K., Neiderhiser, J.M. & Elthammar, O. (2001). Genetic probes of three theories of maternal adjustment: I. Recent evidence and a model. Family Process, 40:247-260.
Poster Session and Research Consultations
A new twist on the poster session was added this year by the collaboration between the Research and Larger Systems Committees as co-sponsors of the Poster Session. Innovative research and interventions in larger systems were presented by a number of AFTA members. Two presentations focused on accessing and using client resources in the therapy process (Adena Rene Vanderwielen & Edith (Winx) Lawrence, and Silvia Echevarria-Doan). Others examined the impact of reflecting teams in family therapy (Anne Fishel) and the use of computer simulation programs in training family therapists (Carmen Guanipa and Grace Chao). Several researchers used qualitative methods to explore areas such as parenting (Carolyn Tubbs) and homelessness (Peter Fraenkel). The larger systems posters illustrated the many and creative ways that family therapy ideas are being disseminated beyond the therapy office. These included training programs in marriage and family therapy that focus on larger systems (Linda Terry) and becoming community catalysts (W. Robert Beavers), as well as community based training programs for tackling difficult issues such as incest and sexual abuse (Fiona True). Three posters described multiple family discussion groups: for children with learning disabilities (Marcia Stern); for families with infants and toddlers to promote school readiness and infant mental health (Martha Edwards); and for families making the transition from welfare to work (Peter Fraenkel and Thomas Hameline). We even had a poster session about AFTA as an organization (Gonzalo Bacigalupe). The Research Committee continued to provide research consultations to AFTA members. Committee members Silvia Echheverria-Doan, Peter Fraenkel, Volker Thomas, Betsy Wood, and Martha Edwards (Chair) were joined by David Reiss to provide assistance and to engage in a lively conversation about the research process.
Martha Edwards, Ph.D., is a member of AFTA's Board of Directors and incoming Chair of the Research Committee. She is a faculty member and Director of the Early Prevention and Enrichment Project at the Ackerman Institute for the Family. She is also a Clinical Assistant Professor of Psychiatry at the NYU Child Study Center, working as Co-Principal Investigator on a project to develop family-focused treatment for bipolar disorder in children and supervising child psychiatry fellows in family therapy.
