Abstract: Male-female courtship interactions inDrosophila melanogaster are mediated in part by chemical cues. A comparison of the courtship behaviors of normal and genetically olfaction-deficient (olfD) flies lends strong support to this hypothesis and leads to the following inferences regarding the importance of chemical interchange during courtship. Virgin females respond to male courtship by slowing and finally stopping their movements, which appears to enhance the probability of copulation; (olfD) females are defective in this stopping response and in their receptivity to mating. Males can be stimulated to court by airborne cues from virgin females which are effective over a distance of ca. 5 mm. By comparison, the chemical courtship-stimulating cues from immature flies have little airborne efficacy; immature males appear to stimulate courtship only by contact chemoreception. The genetic upset or removal from courtship interactions of olfactory, visual, and auditory cues (thus leaving contact chemoreception as the probable major means of male-female interaction) results in virtual behavioral sterility.

Abstract: A biometrical model-fitting approach was applied to data from a full adoption design to study phenotypic variation and covariation among multiple discontinuous traits. Using statistical methods available for factor analyses of dichotomous-item data, generalized least-squares estimates were obtained for parameters of additive polygenic and environmental influences on criminal convictions and psychiatric-hospital diagnoses in 2532 Danish male adoptees and their family members. The procedure also provided estimates of genetic correlations, environmental correlations, and genotype-environment correlations among the measures while taking into account the effects of selective placement, assortative mating, and cultural transmission from parental phenotypes to off-spring environments. Significant heritable influences were found for property crime and for a composite psychiatric variable pertaining to hospital admissions for alcohol abuse, drug abuse, and personality disorders. Both correlated environmental factors and correlated genetic factors important to the different variables accounted for the observed phenotypic relationship between property crime and these antisocial disorders. No cultural transmission was present for the adoptees' antisocial disorders, although significant paternal cultural transmission was found for property crimes. Patterns of assortative mating differed for the biological versus adoptive parents, and selective placement was found to be negligible.

Abstract: To honor Steven G. Vandenberg on the occasion of his retirement from the University of Colorado, a symposium on multivariate behavioral genetics and development was held on May 2–4, 1985. The focus of this symposium was recent advances in multivariate behavioral genetics and their importance for the analysis of developmental phenomena. Because of Vandenberg's role in foundingBehavior Genetics, as well as the subject's relevance, the organizers of this symposium elected to publish its proceedings as a special issue of this journal. The history of multivariate behavioral genetics is outlined in this introductory paper, and a synopsis of the other papers in the issue is presented.

Abstract: The 47, XXY karyotype has been associated with diminished language skills in the presence of average intelligence, but specific deficits in cognitive abilities have not been clearly and consistently established. The present study examined cognitive skills in 14 unselected 47,XXY boys (ages 8 to 18 years) and 14 euploid control boys matched for age and socioeconomic status. No significant differences were found on measures of intelligence, perception, and language, although mean score trends suggested mildly diminished skills in these areas. The 47,XXY group demonstrated significantly poorer auditory short-term memory and reading skills. One-half of the propositi exhibited a common constellation of deficits suggesting that their karyotype may represent a risk factor for a language-based dyslexia with primary features including deficits in memory and speed of processing. Small sample size necessitates caution in making generalizations which await verification from other studies of unselected 47,XXY boys. Possible biological mechanisms responsible for neurocognitive variations in children with sex chromosome anomalies have been hypothesized, but evidence for their existence remains elusive.

Abstract: Psychological constructs involving genetic characters can be of two types. The first type involves behavioral characters in which a score in a given direction should consistently confer a selective advantage on an organism across environments and age. Behavioral measurements taken in different situations should show low additive genetic correlations but similar patterns of directional dominance. A composite score summed across many measures representing the construct should show a high proportion of nonadditive genetic variance. Type 2 constructs involve behavior patterns in which high scores confer a selective advantage in some circumstances and a disadvantage in others. Behaviors representing such constructs should be characterized by high additive genetic correlations, suggesting a common set of genes influencing the behaviors at different ages or in different environments. The direction of nonadditive genetic effects should differ in accordance with optimal behavior levels in each test situation. Activity levels of housemice are shown to fit the model of a type 2 construct. Most measures used in human behavior genetics probably also reflect type 2 constructs.

Abstract: Longitudinal behavioral genetic studies provide an opportunity to analyze both developmental change and continuity, the two essential issues addressed by developmental behavioral genetics Longitudinal analysis can be viewed as multivariate analysis, thus profiting from recent advances made in that area Moreover, the major multivariate concepts of genetic correlation and bivariate heritability assume added significance in longitudinal analyses: They connote age-to-age genetic continuity and change In this paper, the relationship between multivariate and longitudinal analyses is discussed, the meaning of age-to-age genetic change is examined at the level of molecular genetics as well as quantitative genetics, and the implications of age-to-age genetic change for familial resemblance are considered in the context of parent—offspring data from the Colorado Adoption Project, which can be viewed as an “instant” longitudinal study from infancy to adulthood

Abstract: Data from 15 tests of cognitive ability obtained from the Hawaii Family Study of Cognition were examined for relationships with 18 blood polymorphisms. The number of significant associations did not exceed the number expected by chance alone. Significant regression of pooled verbal and spatial test scores on a constructed zygosity index demonstrated that increasing homozygosity was associated with improved scores. The effect was consistent in both parents and offspring, probably in both sexes, and for subjects of both Japanese and European ancestry. However, these blood polymorphisms contributed very little to total variation in test scores.

Abstract: This paper focuses on continuity and change in cognitive ability profile, as represented by the pattern of subtest scores obtained on the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) and Wechsler Intelligence Scale for Children (WISC) during middle childhood. A large number of twins have been tested annually in a longitudinal study, and their scores were evaluated for age-to-age consistency and for withinpair concordance. The following questions were addressed: (1) Does each child display a consistent pattern of subtest scores from age to age? (2) How can genuine transformations in the cognitive profile be separated from measurement error? (3) Do twins match one another for cognitive profile and for systematic changes over ages? The results showed that Verbal IQ scores and Performance IQ scores were more reliable than subtest scores, and the systematic effects of age-to-age consistency and twin concordance were more strongly represented in the V-IQ and P-IQ scores. Within the limits of subtest reliability, however, the cognitive profile was consistently reproduced over ages and was displayed in common by monozygotic (MZ) twins. Some systematic changes in the cognitive profile were detected over longer intervals from childhood to adolescence. Genetic influences were apparent for both continuity and change, although continuity played the more prominent role. The results also confirmed that a high test-retest reliability was a prime requisite for extracting the systematic variance attributable to continuity and change in cognitive ability profile.

Abstract: Following a recent change in location and shuttle-box testing conditions, RHA/Verh and RLA/Verh rats were found to have maintained their characteristic patterns of avoidance (RHA/Verh) vs nonavoidance (RLA/Verh) behavior, regardless of the time of day tested. A second experiment demonstrated dose-related differences in the effects of amphetamine in both lines. In a third experiment it was seen that, whereas 2 mg/kg amphetamine was capable of increasing two-way avoidance acquisition in naive RLA/Verh rats to the levels of nondrugged RHA/Verh rats, this improvement was accompanied by substantial increases in spontaneous activity (intertrial responses). That “learning” was not genuinely enhanced by amphetamine was further indicated by the observation that the avoidance improvement disappeared when the drug was abruptly, or gradually, discontinued. The return toward control levels of avoidance and freezing responses was accompanied by a similar normalization of activity levels.

Abstract: A multivariate path model of genetic and environmental transmission was fitted to specific cognitive abilities data (verbal, spatial, perceptual speed, and visual memory) and evaluated using a maximum-likelihood estimation procedure. In this first multivariate behavioral genetic analysis of adoption data, cultural transmission was modeled via parental phenotypes, each parental phenotype being allowed to affect all traits in the offspring, and assortative mating and cross-assortative mating were accommodated. Results of a preliminary analysis indicated that selective placement is absent in this study. When the full model was fitted to the data, it was found that assortative mating is more important than cross-assortative mating but is substantial only for verbal ability. Genetic transmission parameters are modest in size, whereas cultural transmission is both small and nonsignificant. A simplified model in which cultural transmission and genotype-environment correlation are constrained to be zero provided a good fit to the data. Analyses of the structures of genetic and environmental correlation matrices indicated a strong genetic general factor and a similar, but weaker, environmental factor. Inspection of the genetic transmission parameters suggests that genetic continuity between early childhood and adulthood may be substantial for verbal ability, spatial ability, and perceptual speed.

Abstract: Schizophrenia is a relatively common familial disorder that has been the subject of numerous family, twin, and adoption studies. Despite the large amount of family data available, there is no consensus among geneticists about the mode of inheritance for the disorder. The methods that have been used in the analysis of schizophrenia family data are described and their results summarized. It is concluded that both genetic and environmental factors contribute to the liability for developing schizophrenia. Environmental factors are more likely to be nonfamilial than familial. Genetic effects upon schizophrenia liability are consistent with either a multifactorial or a mixed multifactorial single-gene model of transmission; a single-major-gene model, however, is not consistent with the family data. Although major-gene effects upon schizophrenia are possible, it appears that the identification of such effects will be difficult unless a multivariate approach to segregation analysis is taken.

Abstract: The Wechsler Adult Interlligence Scale (WAIS) subtest scores of 40 pairs of monozygotic (MZ) and 40 pairs of like-sexed dizygotic (DZ) twins were analyzed. Estimates were obtained with respect to the relative influence of genetic and environmental sources of variance common for the subtests of WAIS and sources specific for each subtest. A model was also specified to detect variance common only for the subtests usually reported to load on the same Cohen factor. The major part of the covariance between subtests seemed to be due to common genes. Freedom from Distractibility seemed to be influenced by specific genes, but the other two “Cohen factors,” Verbal Comprehension and Perceptual Organization did not. Environmental sources acting between families seemed to be mostly common, but except for the Perceptual Organization and the Digit Symbol subtests, the relative contribution from these sources was modest.

Abstract: Correlations on scales of the Thurstone Temperament Schedule are examined in two twin studies (Michigan and Veterans twin samples), an adoption study (Texas), and an unpublished twin-family study. It is concluded that the joint evidence suggests (1) an appreciable effect of additive genes, differing across scales; (2) a negligible influence of shared environments, except possibly for monozygotic (MZ) twins; (3) a possible contrast effect among dizygotic (DZ) twins on temperament and personality traits, but shared interests; and (4) a causally ambiguous elevation of MZ twin correlations, which could be due either to nonadditive genetic effects or to a special sharing of environments (or self-concepts) in this group.

Abstract: SWR/J mice avoid sucrose octaacetate (SOA) solutions at concentrations which other inbred strains do not. This phenotypic difference has been hypothesized to result from variation at a single autosomal locus with two alleles, one dominant (Soaa, aversion) and one recessive (Soaa, blind). Data from reciprocal F1 and F2 crosses of SWR/J (taster) and C57BL/6J (nontaster) mice and from four generations of selective lineal backcrossing to the C57BL/6J strain, in two-bottle preference tests with 10−5M SOA, were used to test this monogenic model against two polygenic models. The phenotypic ratios expected in the segregating generations according to the single-locus model were consistent with the observed ratios. The ratios expected with either two-locus model were inconsistent with those found. A strain distribution pattern, also consistent with monogenic variation, was found when a set of recombinant inbred strains (SWXL/Ty) derived from SWR/J and C57L/J (nontaster) mice was similarly tested. Outbred CFW mice (inbred substrains of which had been reported by separate laboratories to be both SOA tasters and SOA nontasters) were found to be polymorphic for SOA tasting. An allele identical by descent to that in the SWR/J strain may be segregating in this (distantly) related line.

Abstract: The procedure for classifying items from the California Psychological Inventory into “genetic” and “environmental” groups, as carried out by Horn, Plomin, and Rosenman (Behav. Genet. 6:17–30, 1976) with the Veterans twin sample, was repeated with male pairs from the National Merit twin sample. There was no significant consistency in the items selected into the two categories in the two samples. It was concluded that individual personality questionnaire items may be inherently too ambiguous for profitable behavior genetic analysis.

Abstract: Passive dark avoidance conditioning was studied in four inbred strains of paradise fish in the presence and absence of a fish-like dummy during the training process. Strain differences were found in exploratory activity during habituation trials and also in the sensitivity to the mild electric shock punishment. The impact of the fish-like dummy also depended on the genotype.

Abstract: A comparison of pupation height in light and dark was made using 12 species ofDrosophila, representing four species groups and four different ecological backgrounds (temperate-montane forest,virilis group desert,replate group; cosmopolitanmelanogaster group; tropical forest,willistoni group). Light condition has a significant effect on pupation height in only two of the species. In the light,D. montana stays close to the food surface, whileD. melanogaster pupates higher in light than in dark. Light-dependent patterns of pupation response do not correspond to those previously reported for the light-dependent mating response. Considerable interspecific variation exists for pupation height in each species triad, some of which could provide a basis for larval niche separation. Patterns of species differences in the desertrepleta triad are the same in light and in darkness.

Abstract: Recombinant inbred (RI) lines have been constructed by crossing the Bristol and Bergerac BO strains of the nematodeCaenorhabditis elegans. The F1 hermaphroditic progeny are allowed to self-fertilize for 20 generations, yielding the RI lines. Heritability estimates for several behavioral traits and for life span, as well as both phenotypic and genetic covariance estimates for these traits, have been obtained. Significant heritability is detected only for forward movement. Large genetic covariances are detected between different behavioral components, corroborating earlier studies based on observations of stereotypical behaviors in nematodes. A strain distribution pattern (SDP) has been generated using genetic loci defined both by mutation and by restriction fragment length polymorphisms. Estimates of the number of loci affecting the life span range from three to six; these estimates were obtained both by classic quantitative techniques and by mapping, using the SDP for these RI lines. Thus, loci specifying life span are distributed throughout the genome rather than localized to any one region. EMS-induced, long-lived (Age) mutants were also examined. In contrast to an earlier report, these mutants ingest normal amounts of food. Theage locus of one strain, MK546, segregates independently of a behavioral alteration which was coinduced in the original mutant hunt.

Abstract: In two-bottle preference tests aversion to phenylthiocarbamide (PTC) develops over a period of days. Thus, as previously reported, following experience with appropriate concentrations of PTC, mice of the BALB inbred strain display an aversion in contrast to C57BL inbred mice. It is suggested that differential learning in a conditioned taste-aversion paradigm might be responsible for the phenotypic strain contrast. The difference in PTC aversion phenotype among mice could be due to differences in any mechanism contributing to differential flavor toxicosis conditioning instead of, or in addition to, strain differences in sensitivity to the sensory attributes of PTC.

Abstract: Four sets of adult mice (Mus musculus), each comprised of individuals from two inbred strains and both reciprocal F1 crosses, were tested during male-female and female-female dyadic encounters for their emission of 70-kHz ultrasonic vocalizations. For each sex-dyad type of each set, a single progeny mean representing both reciprocal F1 groups was calculated and compared to (1) the average value from the inbred parents and (2) the higher mean of the within-set progenitor inbred strains. In addition to demonstrating strain-and sex-influenced ultrasonic vocalizing levels, the results indicated that for each set examined, the F1-progeny mean amount of ultrasonic vocalizing significantly exceeded the average inbred parent value. This was true for both dyad types, providing strong evidence that ultrasonic vocalizing displlays a directional dominance mode of inheritance in both female and male mice. Moreover, for female-female dyads of all four sets and for male-female dyads of three of four sets, the F1-progeny mean amount of ultrasonic vocalizing significantly exceeded that of the highest progenitor inbred strain. Analyses of ultrasonic vocalizing latencies yielded similar hybrid-inbred differences. Collectively, these findings are interpreted as being consistent with the notion that, for both sexes of mice, ultrasonic vocalizing is a phenotypically heterotic behavioral trait.

Abstract: We find that the two reviews of dipteran learning in the September 1984 issue ofBehavior Genetics present a more optimistic view than is warranted by a close examination of their references. Also, because of the absence of adequate control experiments, we question whether successful selective breeding for learning ability in the blow fly has yet been demonstrated. Much of our criticism of the two reviews arises from their failure to distinguish between validly measuring learned behavior in individual flies and the inability to infer from group performance that a given individual has learned. We discuss an approach that can be used to measure learning in individual flies as well as in subsequent selective breeding for learning ability.

Abstract: Environmental factors more common in twins than singletons appear to be disadvantageous with respect to intelligence. If these factors contribute to within-pair differences in IQ, the effect should be found more in the cotwin lower in IQ (the low twin) than in the cotwin higher in IQ (the high twin). In this case, the variance of the low twins should exceed that of the high twins. This hypothesis was tested for five samples of monozygotic (MZ) twins and three samples of dizygotic (DZ) twins but was confirmed only for the MZ samples. These results suggest that there are disadvantageous environmental factors unique to MZ twins which contribute to within-pair variability. Such factors will cause heritability coefficients obtained using MZ twins to be underestimated.

Abstract: Microhabitat preferences ofDrosophila pseudoobscura strains were examined in a Waddington maze, with an emphasis on learning how early environment affected adult habitat choice. The genotypes were roughly those expected in a natural population; the environmental variables included light, temperature, and food. It was found that (1) the different genotypes chose habitats differently; (2) early experience affected subsequent habitat choice; and (3) the effect of early experience was complex, as preference for one niche dimension (temperature) was reinforced by experience with the generally preferred value, preference for another niche dimension (light) was weakened by experience with the generally preferred value, and preference for other niche dimensions (food) was generally unaffected by experience. In this study the contribution to the total chi square was about equal from genotype and from environment. The significance of these findings for studies of dispersal and population structure of natural populations is discussed.

Abstract: We examined the interaction of the albino locus with the maternal environment on the behavioral development of two coisogenic strains of mice. Subjects of the pigmented C57BL/6 strain (=B6+/+) and of the albino C57BL/6c2J strain (=B6c/c) were either fostered by a mother of their own strain or cross-fostered at birth to an F1 hybrid dam. They were compared for the amount and daily distribution of activity displayed during 48 h in a seminatural device at weaning and when 75 days old. Food hoarding in the nest and food consumption at the food-search place were also recorded in adult subjects. When animals were fostered by a mother of their own strain, albino mice were more active and less nocturnal than pigmented mice at both ages. They hoarded less food in the nest and ate more at the food-search place. Most of these differences disappeared when both strains were fostered by an F1 dam. The amount of activity displayed during 48 h increased between 21 and 75 days of age. This increase was affected by cross-fostering to an F1 dam in B6c/c mice only. The developmental pattern of daily distribution of activity was changed by F1 dams in B6+/+ mice only. Whereas these influences of F1 dams produced subjects resembling the mother's phenotypic score, maternal effects on hoarding behavior in B6c/c mice produced subjects which did not resemble their foster mother. The results are discussed in terms of different possible ways of hereditary transmission of behavior and some methodological consequences are emphasized.

Abstract: The present data illustrate an important characteristic of the mate recognition system of individuals ofDrosophila melanogaster. This system showed a high degree of stability, and the stability occurred even though the populations under investigation were divergent in other characteristics. We review the results of other studies of this type, discuss our results in terms of the general characteristics of specific-mate recognition systems, and account for its stability in terms of the structure of the communication system.

Abstract: Cardiovascular responses (changes in heart rate, systolic and diastolic blood pressure) of 103 normal young adult males (46 European, 30 Japanese, 27 Chinese) to a test drink of alcohol were analyzed. The two Oriental groups did not differ in their mean responses (measured as postdrink value minus baseline value). When these two groups were pooled as “Orientals,” they differed very significantly from Europeans in their responses. Each of the three groups showed marked between-individual variability in alcohol response for each cardiovascular parameter, in the absence of obvious environmental differences. Repeated-measures ANOVA on these and other data, plus a direct genetic study in mice of the heritability of alcohol-induced change in heart rate, indicates that the broad-sense heritability of such changes in humans is in the region 0.3 to 0.5.

Abstract: Holliday & Hirsch (this issue) now agree that “Quinnet al. (1974) have demonstrated learning [inDrosophila] with group data, and their inability to identify individual differences (IDs) in performance does not invalidate their conclusion that some individuals in the population must have learned.” However, they consider it important, if not necessary, to show that anindividual fly has learned. In response to Holliday and Hirsch, this paper discusses why it is not necessary to measure learning in individual fruit flies before searching for underlying biochemical mechanisms.

Abstract: Adult male mice from the substrains DBA/2JNmg and DBA/2J were individually placed into a novel environment and compared for nine behavioral components. The substrains differed in three vertically oriented exploratory acts, the latter showing the highest rates. On the basis of the breeding history of strain DBA/2, it is concluded that over the past years one or more mutations with an effect on behavior have occurred in at least these sublines of an old and widely used inbred mouse strain.

Abstract: The hypothesis that negative assortative mating occurs as a mechanism limiting inbreeding between genetically related individuals ofDrosophila melanogaster was tested. In order to avoid bias linked to using inbred lines, experiments made use of the F1 hybrid progeny between lines rendered homozygous on chromosomes 1, 2, and 3. No negative assortative mating was found, but significant additive variation was observed between lines for orientation, vibration, copulation latencies, and copulation duration. There was no consistency of results, either among parameters or between sexes from the same line. It is therefore unlikely that the variations observed are due merely to quantitative differences in “vigor”. Since all lines originated from the same wild population, these differences are a possible estimate of natural variation in sexual behavior.