Epigentics and Marriage: Implications In Divorce?

Here is a somewhat rough paper from the spring of 2009 where I needed to briefly describe an experiment that touched on social psychology.  I ended up writing about marriage and epigenetics. 
Sitting in class a couple weeks ago I heard the professor mention the popular adage of people growing similar in physical appearance over the course of a long term relationship, such as a marriage.  This set off a spark in my mind that set me on the path to this paper.  What might be the underlying biological mechanisms for this commonly observed phenomenon, if it is indeed a real occurrence and not just a perceptual trick having to do with familiarity of the pair?  Epigenetics is “defined as those modifications to DNA or the nucleosome that do not alter the sequence of nucleotides, but do modify the transcription of genes.” (Kevern, & Curley, 2008 p.399). These changes in gene transcription have direct effects on the phenotype of a cell or individual.  Richard Dawkins further describes the idea of the phenotype in his recent book, The Extended Phenotype, to describe all effects of the gene on the outside world (Dawkins, 1982). This would include both physical appearance as well as behavior. As a result of this definition it is reasonable to consider that the social observation of couples growing more similar in physical appearance might have something to do with epigenetics, or more specifically similar changes in gene transcription due to being exposed to extremely similar environments over an extended period of time. 

In a 2001 Neuroscience paper it is said that self-referent phenotype matching in mating is both widespread and adaptive (Hauber, & Sherman, 2001). Further, in another paper published in Evolution the authors discuss an alternative mating preference strategy, sexual imprinting.  Sexual imprinting is the process by which young animals (predominantly females) learn the characteristics of a desirable mate.  It is proposed that the three variations of sexual imprinting is as follows.

1)     Imprinting on the phenotype of the organisms mother.

2)     Imprinting on the phenotype of the organisms father.

3)     Imprinting on members of the general population.

These types of imprinting have been observed across a wide range of species, predominantly birds and mammals (VERZIJDEN, LACHLAN, & SERVEDIO, 2005).

What do this mean?  Basically, individuals often choose their mates based on a perceived similarity in gene sequencing based on phenotype expression.  However, due to mate selection by phenotype matching being based predominantly on physical appearance, either self-referencing or through sexual imprinting, overall gene sequence similarities should have more variability than might be expected from physical appearance alone.  Of particular interest would be genes associated with pro-social behaviors and more specifically behaviors that facilitate healthy long term relationships.

Because it is possible for gene expression to change over time I hypothesize that although phenotype may be a driving force in initial mate selection it is likely the underlying genotype similarities that facilitate the most successful long term cohabitations.  Even though couples may have drastically different gene expression at the beginning of a relationship, those that have more similar underlying genotypes should, over time through the effects of epigenetics, begin to have more and more similar gene expression throughout the course of their relationship.  As a result these individuals should, as the adage says, begin to look more like each other.  Continuing this line of though, it stands to reason that this effect should similarly impact gene expression that influences certain aspects of behavior.  Because of this, it is my hypothesis that there should be a positive relationship between the percentage of shared base genes and length of successful cohabitation of non-familial partners. Further, I also hypothesize that there will be another relationship between the synchronization of gene expression and length of successful cohabitation of non-familial partners.

According to a report issued in 2005 by the US census bureau, first marriages that end in divorce last about 8 years, on average (Kreider, 2005). It seems reasonable that couples with larger variation in base genes might experience a greater and more prolonged quantity of conflict in a relationship after the so called “honey moon period” due to less synchronized behavioral gene expression.  While it certainly seems likely that a lack of shared gene expression could contribute to conflict in a relationship, at what point this effect begins to add more stressors than more similar gene sequenced counterparts would need to be examined. 

One of the procedures that would be necessary for examining such phenomena would be genome sequencing.  The costs of such a procedure are falling rapidly. According to an article on Bloomberg.com the first human genome sequence cost $2.3 billion in 2003 while in the last  year one company, Knome, has reduced the cost of sequencing an individual genome from $350,000 to $99,500.  Further, a newer company, Complete Genomics, claims that in June it will begin offering human genome sequencing for $5,000/genome (Lauerman, 2009).  If Complete Genetics claims pan out it will be approaching the point at which, with appropriate funding, the sequencing of enough individuals for specific experiments begins to become feasible.  Also, if historic cost trends within the field continue it should only be 5-10 years at most before it becomes quite affordable for everyone to have their genome sequenced. There are a number of methods for analyzing gene expression ranging in both complexity and age.   This technology can be expected to improve along side of genome sequencing methods and the newest and best established method should be considered at the point when genome sequencing becomes reasonably priced. 

The basic idea for studying my hypothesis would require a longitudinal approach due to the nature of the topic.  One could begin by recruiting couples engaged to be married but have not yet cohabitated.  Length of relationship prior to marriage (the beginning of cohabitation) should be noted but unless there are an abundance of participants I do not think that it should be utilized as a criterion for selection.  Prior to the beginning of cohabitation a full genome sequence analysis should be performed as well as gene expression analysis of specific genes whose functions or associations have been well established. Particularly, genes that have been demonstrated to have strong associations with behaviors and personalities at this point.  Assessments should be completed on each individual evaluating some of the following

1)     Measures of temperament, character, and personality with a focus on interpersonal comparison aspects among family members. Specifically the Interpersonal Checklist and the Myers-Briggs Temperament Inventory.

2)     Techniques to assess levels of marital satisfaction, quality, or happiness(in the first evaluation it could look at expected values of these things)

3)     Family adaptation measures.

4)     Others measures that are appropriate to our knowledge of gene-behavior interaction at the time of data collection.

Once these measures and assessments have been collected the couples would be sent off to begin their lives together.  Periodically, perhaps once per year or every other year, pending costs, the couples would then be brought back in for re-examination of expression as well as retesting within the measures mentioned earlier.  Detailed history would be taken of all significant marital and family events.

This experiment should continue as long as the participants would be willing to continue coming in, even after any divorces, deaths, or other circumstances that end a relationship.  This experiment could possibly be done as an analysis of a subset of data from a larger genome and gene expression study.   My expectations for results 

1)     Expression of shared genes will become increasingly synchronized over time.

2)     There will be a strong positive correlation between shared genes and length of successful cohabitation.

3)     In the event of the termination of a relationship there will be a gradual shift away from synchronicity in shared genes that showed changes over the course of cohabitation.

            An experiment examining these issues could prove quite useful in marriage and family counseling settings.  It may help to identify key stressors on a particular relationship due to which genes, as well as which gene expressions, a couple has in common and which ones they do not.  Long term it could even provide evidence for pursuing possible gene therapy treatments for relationship issues, attempting to increase the rate of shared expression as a means of reducing conflict in a relationship.  


Dawkins, R. (1982). The Extended phenotype. Oxford: Oxford University Press.

Hauber, M.E., & Sherman, P.W. (2001). Self-referent phenotype matching: theoretical considerations and empirical evidence. TRENDS in Neurosciences, 24(10), 609-616.

Kevern, E.B., & Curley, J.P. (2008). Epigenetics, brain evolution and behaviour. Frontiers in Neuroendocrinology, 29, 398–412.

Kreider, R.M. (2005). Number, timing, and duration of marriages and divorces: 2001. Current Population Reports; Household Economic Studies, P70-97.

Lauerman, J. (2009, February 9). Complete genomics drives down cost of genome sequence to $5,000 . Retrieved from http://www.bloomberg.com/apps/news?pid=20601124&sid=aEUlnq6ltPpQ

VERZIJDEN, M.N, LACHLAN, R.F., & SERVEDIO, M.R. (2005). Female mate-choice behavior and sympatric speciation. Evolution, 59(10), 2097–2108.

Share on Reddit0Share on Google+0Share on Facebook0Tweet about this on TwitterShare on StumbleUpon0Digg thisShare on LinkedIn0