The key findings in the current study were that SCOR magnitude is heritable across development from ages 9 to 16 years. Genetic influences accounted for 56%, 83%, and 48% of the variance in SCOR magnitude at ages 9-10, 11-14 and 16-17 years, respectively, with the remaining variance due to non-shared environmental influences. SCOR magnitude was moderately stable across development, with phenotypic correlations ranging from .35 to .45. Longitudinal model fitting results showed that the majority of the variance in SCOR was explained by a common genetic factor. This common genetic factor explained 36%, 45% and 49% of the variance in SCOR magnitude across development. Age-specific genetic effects were found at ages 9-10 years and ages 11-13 years, explaining 18% and 35% of the variance, respectively. Further, the genetic correlations between the three time points were high (time 1 to time 2, .55; time 1 to time 3, .73; and time 2 to time 3, .71), indicating a large continuity of the same genetic influences across time. To the authors’ knowledge, this is the first behavioral genetic study examining the genetic and environmental etiology of SCOR magnitude from late childhood to adolescence.
We found that within each assessment SCOR magnitude was primarily explained by genetic and non-shared environmental influences. This finding is well in line with findings from previous research on SC using adult samples (Crider et al., 2004
; Lykken et al., 1988
). These studies have reported that genetic and non-shared environmental factors each explain about half of the variance in various indices of SC responses. In our study, genetic influences accounted for 55%, 82%, and 48% of the variance in SCOR magnitude at ages 9-10, 11-14 and 16-17 years, respectively. This finding that genetic effects vary across development has been found for other phenotypes as well, for example, heritability estimates of cognitive functioning increase from around 30% in preschool children to 80% in early adolescence (Polderman et al., 2006
). Further, our longitudinal analyses showed that age-specific genetic effects were important at ages 9-10 years and ages 11-13 years. In other words, not all genetic influences were accounted for by the common genetic factor across development. This indicates that there may be genetic effects that uniquely impact SCOR magnitude at these ages, over and above the genetic influence from the common genetic factor. It has been suggested that genes that are activated at puberty may be an important influence on traits or behaviors (Jacobson, Prescott & Kendler, 2002
). It is possible that the age-specific genetic effect found at ages 11-13 years in the current study are related to the onset of puberty.
Similar to previous research, we found no evidence of significant shared environmental influences in SCOR magnitude. The absence of shared environmental influences does not imply that environmental influences are of no importance. Rather, this finding indicates that those environmental experiences unique to each twin are especially important in explaining individual differences in SCOR magnitude. These non-shared environmental experiences may for example include pre- and postnatal experiences, differential treatment by parents, and head trauma.
We found no evidence for sex specific effects at any of the visits, and genetic and environmental components could be equated across sexes. This indicates that the relative magnitude of genetic and non-shared environmental factors in SCOR magnitude is the same in boys and girls.
Our longitudinal analyses showed that a common genetic factor explained the variance in SCOR magnitude across development. Non-shared environmental influences (including measurement error) were age-specific. The longitudinal analyses further demonstrated that genetic influences were stable across development. The genetic correlations between each pair of time points were rather high, indicating a large continuity of the same genetic influences on SCOR magnitude across the three time points. The importance of stable genetic influences has previously been reported in studies on child psychopathology (Eley, Lichtenstein & Moffitt, 2003
; Haberstick et al., 2005
; O’Connor et al., 1998
The finding that genetic factors were the most important influence explaining the variance in SCOR magnitude, cross-sectionally as well as longitudinally, has important implications. It is possible that a genetic predisposition for a psychopathological phenotype is partly expressed through biological correlates, e.g., a psychophysiological characteristic, such as abnormal autonomic orienting to various stimuli in one’s environment (Raine, 2002
). An important next step would therefore be to examine the genetic covariation between SCOR magnitude and those psychopathological behaviors that have been found to be correlated with SCOR magnitude.
The results of this study are subject to several potential limitations. First, the sample size at the second visit was relatively small; this could have affected the outcome. For example, with a larger sample we might have been able to detect significant shared environmental influences. Therefore it will be important to replicate our findings in larger samples with more power.
Second, our method to assess SCOR does not allow the differentiation of the influences of central and peripheral nervous system. Peripheral factors such as number of sweat glands are associated with SC, especially the SC levels, a measure of arousal (Freedman et al., 1994
), but firing up those glands in the orienting paradigm is in a way over and above those peripheral influences (and over and above tonic SC levels). Specifically, SCOR has been found to be sensitive to a variety of stimuli, including stimulus novelty, intensity, arousal content, and significance, and it has been widely used as a psychophysiological index of information processing capability (Dawson et al., 2007
). Among all forms of autonomic nervous system activity, individual differences in SC responses appear to be most reliably associated with psychopathological states, again suggesting that it is not all noise. The psychological implications of SCOR are further manifested by the reported significant relationship between psychopathic personality and SCOR magnitude in 9-10-year olds (Isen et al., 2010
), and the fact that this personality trait continues to be linked to SCOR magnitude in our male participants during adolescence. Despite these findings, we cannot tease out the possibility that the heritability of SCOR magnitude found in our study mainly reflects the characteristics of underlying physiological rather than psychological factors.
In conclusion, our results showed that SCOR magnitude was roughly equally explained by genetic and non-shared environmental factors within each measurement time point. A common genetic factor explained the majority of the variance in SCOR magnitude across development. These findings increase our understanding of SCOR by demonstrating the importance of genetic influences for individual differences in SCOR magnitude.