The possibility of genotype–environment interaction for memory performance and change was examined in 150 monozygotic (MZ) twin pairs from the Swedish Adoption Twin Study of Aging (SATSA). We used an MZ twin pair difference approach to examine the possibility that genotype was associated with intrapair variability and thus suggestive of genotype–nonshared environment interactions. Multiple ‘variability genes’ were found for longitudinal change in a semantic memory task including candidates coding for apolipoprotein E (APOE) and estrogen receptor alpha (ESR1) as well as serotonin candidates (HTR2A and 5HTT). One candidate also related to variability in change in episodic memory (5HTT). Of the significant associations observed, generally results indicated that MZ pairs who carry putative risk alleles were less variable than noncarriers, suggesting that noncarriers may be more sensitive to environmental contexts. We sought to ‘contextualize’ the possible nonshared environmental influences for found gene–environment (G × E) effects by considering intrapair differences in measured social and stress factors, including social support, life events and depressive symptoms. Results suggested that nonshared environmental influences associated with depressive symptoms may moderate the G × E relationship observed for ESR1 and APOE and longitudinal semantic memory change whereby noncarriers of putative risk alleles may be relatively more sensitive to depressionevoking environmental contexts than carriers of the risk allele. Thus, the contexts that facilitate or reduce depressive symptoms may affect semantic memory resiliency dependent on genotype. Further work ought to consider larger sample sizes as well as consider additional social and contextual factors.

Although social and intellectual engagement have been consistently associated with late-life functioning, rather than true causation, these associations may reflect the experiential choices of high functioning individuals (i.e., selection effects). We investigated the association of social activity with late-life physical functioning, cognitive functioning, and depression symptomatology using data from 1112 pairs of like-sex twins who participated in the Longitudinal Study of Aging Danish Twins. Consistent with previous research, we found that social activity was significantly correlated with overall level of physical functioning, cognitive functioning, and depression symptomatology. We also found that social activity was significantly and moderately heritable (estimate of .36), raising the possibility that its association with late-life functioning might reflect selection processes. Further, social activity did not predict change in functioning and in monozygotic twin pairs discordant on level of social activity, the more socially active twin was not less susceptible to age decreases in physical and cognitive functioning and increases in depression symptomatology than the less socially active twin. These results are interpreted in the context of the additional finding that nonshared environmental factors, although apparently not social activity, are the predominant determinant of changes in late-life functioning.

In previous studies we obtained evidence that variation in loneliness has a genetic component. Based on adult twin data, the heritability estimate for loneliness, which was assessed as an ordinal trait, was 48%. These analyses were done on loneliness scores averaged over items (‘I feel lonely’ and ‘Nobody loves me’) and over time points. In this article we present a longitudinal analysis of loneliness data assessed in 5 surveys (1991 through 2002) in Dutch twins (N = 8389) for the two separate items of the loneliness scale. From the longitudinal growth modeling it was found sufficient to have non-zero variance for the intercept only, while the other effects (linear, quadratic and cubic slope) had zero variance. For the item ‘I feel lonely’ we observed an increasing age trend up to age 30, followed by a decline to age 50. Heritability for individual differences in the intercept was estimated at 77%. For the item ‘Nobody loves me’ no significant trend over age was seen; the heritability of the intercept was estimated at 70%.

The physiological benefits of vigorous exercise are well established. The existence of psychological benefits is less clear, however, due to methodological limitations common to investigatory studies. Two of these limitations are the difficulty of establishing appropriate control groups and the large variety of highly specific measures of psychological function available for consideration of effects. To address these limitations, we identified 63 pairs of monozygotic twins from the National Survey of Midlife Development in the United States who were discordant for the amount of vigorous exercise in which they engaged regularly. The twins who regularly engaged in vigorous exercise experienced greater positive psychological functioning than their nonexercising co-twins as measured by the latent factor representing the variance common to 8 measures of mood, optimism, control over life, and interpersonal aspects of personality. The magnitude of the difference was in excess of .4 standard deviations.

From conception to death, helping relationships promote positive development and enable people to surmount challenges in their lives. Is it the case that the negative consequences of a genetic propensity for risky behaviors can be attenuated by helping relationships (a G × E)? But is it also the case that people with such a genetic propensity are less likely to have helping relationships compared to people without such a propensity (a rGE)? We illustrate this complex pattern of gene–environment interplay by drawing on the National Longitudinal Study of Adolescent Health and a combinatoric analytic strategy. We focus on a gene associated with dopamine receptor type 2 (DRD2 TaqIA), student–mentor relationships, and educational continuation beyond secondary school. Results reveal that, for both white and black males, DRD2 A1+ (A1A1 and A1A2 genotypes) is associated with a decreased likelihood of school continuation compared to their counterparts with DRD2 A1–; mentors who are teachers compensate for this negative association (a G × E); and youth with DRD2 A1+ are less likely to have a mentor who is a teacher than their counterparts with DRD2 A1– (a rGE).

Previous studies indicate that the emotional quality of marital relationships is mirrored in parent–child relationships. We explored the degree to which these associations are explained by genetic and environmental factors. Participants were drawn from the Twin and Offspring Study in Sweden (TOSS), and included 544 female twin pairs (258 monozygotic [MZ], 286 dizygotic [DZ]), and 311 male twin pairs (128 MZ, 183 DZ). The spouses and one adolescent child of each twin also participated in this study. The twins completed self-report measures that assessed their marital quality and their warmth and negativity towards their children. Observational ratings of marital warmth and negativity, and of maternal warmth and negativity were obtained for a subset of female twin pairs (150 MZ, 176 DZ). Selfreported marital satisfaction was associated with self-reported parental warmth and negativity for mothers (rs = .25, -.36) and fathers (rs = .25, -.44). For the observational measures, marital warmth was associated with maternal warmth (r = .42), while marital negativity was associated with maternal negativity (r = .34). On average genetic factors explained nearly half of the covariance between selfreported marital satisfaction and parenting for mothers (48%) and fathers (47%). Genetic factors explained 21% of the covariance between observed marital and maternal warmth, but did not contribute to associations between marital and maternal negativity. These findings indicate that parents' genetically influenced characteristics help shape the emotional climate of the family.

In contrast to many phenotypes that have been studied using twin designs, substance use shows considerable evidence of environmental influence. Accordingly, specifying the relevant environments and understanding the nature of their effects is an important research priority. Twin studies also have demonstrated that the importance of genetic and environmental influences varies across development for a variety of behavioral outcomes, including substance use. Here, we report analyses exploring moderating effects associated with parenting and peer characteristics on adolescent smoking and drinking, measured at ages 14 and 17. We find significant evidence of moderating effects associated with two dimensions of parenting (parental monitoring and time spent in activities with parents) on adolescent smoking, measured at two time points across development, but no moderating effects on adolescent drinking. Genetic influences on smoking increased, and common environmental effects decreased, as adolescents reported less parental monitoring and spending more time with their parents. Conversely, we find evidence that adolescent drinking is more strongly influenced by peer characteristics. The importance of genetic predispositions was increased among adolescents who reported more friends who used alcohol. These analyses illustrate the importance of incorporating measured aspects of the environment into genetically informative twin models to begin to understand how specific environments are related to various outcomes. Furthermore, they illustrate the importance of using a developmental perspective to understand how specific influences may vary across different ages, and across different phenotypes.

Evidence for a relation between neuroticism and religion is scarce and inconsistent. The aims of the present study were to determine the association of religious upbringing with adult neuroticism scores and to examine the effect of religious upbringing on the heritability of neuroticism. As part of a longitudinal survey of twin families from the Netherlands Twin Register, data were collected on neuroticism and religious upbringing. Restricting the sample to persons aged 25 and over resulted in a sample of 4369 twins and 1304 siblings from 2698 families. Religious upbringing was significantly associated with neuroticism; in both men and women neuroticism levels were lower in those who had received a religious upbringing. There were no sex or twinsibling differences in neuroticism variances and covariances. Structural equation modeling showed differences in heritability between those with and without religious upbringing. In the group with religious upbringing, variation in neuroticism was determined for 41% by additive genetic factors and for the remaining 59% by unique environmental factors. In the group who had not received a religious upbringing, variation in neuroticism was determined for 55% by genetic factors, with evidence for both additive and nonadditive factors, and for the remaining 45% by unique environmental influences. In conclusion, having received a religious upbringing is associated with lower neuroticism scores and a lower heritability in adulthood.

Previous studies in young and adolescent twins suggested substantial genetic contributions to the amplitude and latency of the P3 evoked by targets in an oddball paradigm. Here we examined whether these findings can be generalized to adult samples. A total of 651 twins and siblings from 292 families participated in a visual oddball task. In half of the subjects the age centered around 26 (young adult cohort), in the other half the age centered around 49 (middle-aged adult cohort). P3 peak amplitude and latency were scored for 3 midline leads Pz, Cz, and Fz. No cohort differences in heritability were found. P3 amplitude (∼50%) and latency (∼45%) were moderately heritable for the 3 leads. A single genetic factor influenced latency at all electrodes, suggesting a single P3 timing mechanism. Specific genetic factors influenced amplitude at each lead, suggesting local modulation of the P3 once triggered. Genetic analysis of the full event-related potential waveform showed that P3 heritability barely changes from about 100 ms before to 100 ms after the peak. Age differences are restricted to differences in means and variances, but the proportion of genetic variance as part of the total variance of midline P3 amplitude and latency does not change from young to middle-aged adulthood.

The effect of nonresponse on health and lifestyle measures has received extensive study, showing at most relatively modest effects. Nonresponse bias with respect to personality has been less thoroughly investigated. The present study uses data from responding individuals as a proxy for the missing data of their nonresponding family members to examine the presence of nonresponse bias for personality traits and disorders as well as health and lifestyle traits. We looked at the Big Five personality traits, borderline personality disorder (BPD) features, attention-deficit/hyperactivity disorder, Anger, and several measures of health (Body Mass Index, migraine) and lifestyle (smoking, alcohol use). In general, outcomes tend to be slightly more favorable for individuals from highly cooperative families compared to individuals from less cooperative families. The only significant difference was found for BPD features (p = .001). However, the absolute difference in mean scores is very small, less than 1 point for a scale ranging from 0 to 72. In conclusion, survey data on personality, health and lifestyle are relatively unbiased with respect to nonresponse.

Human growth is a complex and poorly understood process. We studied the effect of genetic and environmental factors on height and body mass index (BMI, kg/m2) based on maternal reports at 3, 4, 5, 7, 10 and 12 years of age in a large longitudinal cohort of Dutch twins (7755 complete twin pairs at age 3). Several multivariate variance component models for twins were fitted to the data using the Mx statistical package. The first-born twin was taller until age 10 and heavier until age 12 than the second-born co-twin. Heritability estimates were high for height (a2 = .58–.91) and BMI (a2 = .31–.82), but common and unshared environmental factors were also important. The phenotypic correlations across the ages for height and BMI were mainly explained by correlated additive genetic factors (ra = .77–.96 for height and .43–.92 for BMI), but common (rc = .40–.84 and .09–.78, respectively) and specific environmental correlations (re = .50–.81 and .42–.80, respectively) were also significant. Additive genetic factors decreased with increasing age difference for both height and BMI. However, the full Cholesky model, which does not make any assumptions regarding the underlying genetic structure, had the best fit. High genetic correlations across the ages, especially for height, may help further molecular genetic studies of human growth. Environmental factors affecting height and BMI during growth period are also important, and further studies are needed to identify these factors and test whether they interact with genetic factors.

We investigated whether women show larger heart rate variability (HRV) than men after controlling for a large number of health-related covariates, using two indices of HRV, namely respiratory sinus arrhythmia (RSA) and approximate entropy (ApEn). In a twin design, the heritability of both indices was examined. The covariation between RSA and ApEn, a measure of heart rate dynamics derived from nonlinear dynamical systems theory, was decomposed into genetic and environmental components. Subjects were 196 male and 210 female middle-aged twins. Females showed larger HRV than men before (ApEn: p < .001; RSA: p = .052) and after adjustment for covariates (ApEn: p < .001; RSA: p = .015). This sex difference was confirmed by significant intrapair differences in the opposite-sex twin pairs for both ApEn (p < .001) and RSA (p = .03). In addition to sex, only heart period and age (both p < .001) were found to be independent predictors of ApEn, whereas RSA was also influenced by respiration rate and smoking (both p < .001). Age explained 16% and 6% of the variance in RSA and ApEn, respectively. Oral contraceptive use and menopausal status had no effect on HRV. Genetic model fitting yielded moderate heritability estimates for RSA (30%) and ApEn (40%) for both males and females. The correlation between RSA and ApEn (r = .60) could be attributed to genetic factors (48%), environmental factors (36%) and age (16%). The present study found support for a gender difference in HRV with women having greater HRV than men even after controlling for a large number of potential confounders. Indices of heart rate dynamics derived from nonlinear dynamical systems theory are moderately heritable and may be more sensitive than traditional indices of HRV to reveal subtle sex differences with important implications for health and disease.

While evidence supporting for nonadditive genetic influences on personality traits in Caucasian populations has been growing in recent years, twin studies that explored the existence of genetic nonadditivity in personality variation in Asian populations are still lacking. Seven hundred and sixty-five pairs of adolescent and young adult twins registered with the South Korean Twin Registry completed the 7 scales of the Eysenck Personality Scales through a mail survey. Maximum likelihood twin correlations were computed and model-fitting analyses were conducted. Monozygotic twin correlations were consistently higher than twice the dizygotic twin correlations for all 7 scales, suggesting pervasive influences of nonadditive genetic effects on personality traits in the South Korean population. Model-fitting analyses indicated that genetic nonadditivity is particularly important for the variation of Impulsivity, Venturesomeness, Empathy, Lie, and Psychoticism. According to the best fitting models, nonadditive genetic effects ranged from 34 to 49% for these scales. For Neuroticism and Extraversion, models that included an additive genetic component fit better than those including a nonadditive genetic variance component.

This study was performed using population-based data to analyze whether motor development in early life is different between singletons and twins in Japan. For better comparison and investigation, we divided the 2 groups into a group with a birthweight of 2.5 kg or greater (subgroup A) and a group with a birthweight of less than 2.5 kg (subgroup B), respectively. We analyzed the database of medical check-ups for children aged 3 years between April 2001 and July 2004. They received medical checkups at 4 months, 1.5, and 3 years of age. Children who were suspected of having neurological abnormality or disability were referred to specialists and excluded from the database. The data of 14,132 children were analyzed. Among these, 13,040 (92.3%) children were singletons in subgroup A, 75 (0.5%) were twins in subgroup A, 866 (6.1%) were singletons in subgroup B and 151 (1.1%) were twins in subgroup B. The mean age at achieving milestones was slower in twins of subgroup A for each developmental outcome than singletons of subgroup A, and the difference between twins and singletons was significant after adjustment for rolling over. On the contrary, after adjusting for a confounding factor (gestational age), singletons of subgroup B attained motor development facilitating walking independently slower than twins of subgroup B. There were different tendencies in the results regarding the motor development of subjects of subgroup A and that of subjects of subgroup B.

There are conflicting studies associating twin pregnancies derived from assisted reproductive technology (ART) with preterm birth, low birthweight, and other negative outcomes. This work investigates whether ART is linked with any placental pathology, given that placentation significantly influences fetal development. A 5-year, retrospective cohort study was conducted on placentas from twin pregnancies. The placental information from 417 patients was divided into two groups: placentas derived from ART and placentas derived from spontaneous pregnancies (non-ART). Available clinical information and pathologic findings from both groups then were compared. There was no statistical difference in the prevalence of placental pathology between the non-ART and ART cohorts (i.e., cord insertion, single umbilical artery, cord knot, retroplacental hemorrhage, infarction, vasculopathy, vascular anastomoses, chorangiosis, villitis, deciduitis, chorioamnionitis, meconium staining). However, 8% of ART multiple pregnancies were monochorionic. While monochorionicity is a known risk factor for adverse obstetric and neonatal outcomes, the rate of monochorionic placentation did not increase as a result of ART. Nevertheless, it is interesting to note that this small percentage of monochorionic placentation occurred in the ART cohort despite the implantation of individual embryos. Overall, the data suggests that ART does not have a role in the pathologic placentation of twin pregnancies.

Available hypotheses proposed to explain the mechanism of zygotic splitting fail to explain why monozygotic twins are more prevalent after all methods of assisted reproduction and which structure is likely to control this phenomenon. Arguably, a small proportion of oocytes might have an inborn propensity to undergo splitting upon fertilization leading to the constant prevalence of spontaneous monozygotic conceptions among different populations. Ovarian stimulation would then predictably increase the number of available splitting-prone oocytes and consequently would increase the chance for such oocytes to develop into monozygotic twins, leading to a ‘dose’-dependent relationship between monozygosity rates and the combined effect of infertility treatment. Embryonic division into 2 distinct cell lines begins and accommodates within an intact zona pellucida that controls the process by preventing ill-timed hatching. Human fertilized oocytes are able to undergo 2 binary fissions, just as is the case for the 9-banded armadillo (the only other mammal that produces monozygotic quadruplets) and to give rise to a variety of combinations of monozygotic pregnancies. This hypothetical explanation does not negate the already existing and genetically sound hypotheses, but places them into a broader perspective that respects recent observations from modern infertility treatment.

This study aimed to examine differences in multiple maternities by neighborhood-income levels in Toronto, Canada. Hospital records were used to perform secondary analysis of 144,731 maternities resulting in single or multiple infants live-born to mothers residing in the City of Toronto 1996 to 2001. The independent variable was neighborhood income, defined as mean household neighborhoodincome quintiles. Multiple logistic regression analysis was used to compute adjusted odds ratios (AORs) and 95% confidence intervals (CIs). Differences by income levels were found in twin maternities but not in higher order maternities. Twin maternities were more likely to occur in the richest neighborhood-income quintile compared to the rest of the population (AOR: 1.25, 95% CI: 1.10-1.41), after adjustment for potential confounders. The positive association between high neighborhood income and twin maternities found in this study suggests that the richest neighborhoods select families whose characteristics pose them at increased risk of having twins. Further studies are needed to clarify the underlying mechanisms leading to socioeconomic differences in multiple births.

The aim of the study was to assess, using population-based data, trends and regional variations in multiple births during the period of increasing use and changes in practice patterns for infertility treatments. National data for 24,554,977 births (live births and stillbirths) were used, including 569,423 twins during the period 1972 to 2003, and 14,599 triplets for 1984 to 2003. Statistical analyses included age-adjusted hierarchical logistic regression models for twin births and separate analyses for triple, same-sex, and different-sex twin births. Due to confidentiality considerations, the only variable available for adjustment was maternal age. Regionallevel variations were estimated using median odds ratios based on random-intercept hierarchical logistic regression models. Overall, twin births increased from 18.1 per 1000 births (95% confidence interval [CI] 17.9–18.2) in 1972 to 1975 to 29.9 per 1000 (95% CI 29.7–30.1) in 2000 to 2003. Twin births increased progressively across all regions, whereas triple births reached a peak in the early 1990s and decreased thereafter. Trends for both twin and triple births varied significantly across regions. Both trends and regional variations were greater for different-sex as compared with same-sex twin births. Regional variations in the proportion of multiple births increased in the case of twin births and decreased for triple births. Differences in multiple births at the regional level in France were comparable to country-level differences observed across several western and northern European countries. Regional differences in multiple births need to be monitored and used to inform policies aimed at regulating the use of infertility treatments.