Individual Rank Order Research Articles

Individual timing consistency across long-distance songbird migrations.

Migration timing in long-distance migratory birds plays an essential role in individual survival and fitness and is thought to be driven by circannual routines cued by photoperiod with some plasticity to environmental conditions. We examined the individual order of migration timing in purple martins (Progne subis), a neotropical migratory songbird that travels between breeding sites throughout eastern North America and nonbreeding sites in Brazil. Migration timing data were collected for 295 different individual purple martins over 9 years using light-level geolocators deployed at breeding sites across the range. We used linear mixed-effect models to examine the influence of the rank order of individual departure dates in one season on the rank order of four subsequent migration events while controlling for the effects of breeding latitude, sex, and age. Overall, we found evidence for consistent individual timing that can extend across 8 months and 12,000-24,000 km of migration. Individual rank order of migration timing in purple martins was generally conserved across migrations with consistent timings between fall departure dates from, and spring arrival dates to the breeding site the following year (0.28 ± 0.03, 95% CI 0.22-0.34), as well as at a finer scale across fall migration (0.33 ± 0.05, 95% CI 0.23-0.43), over the stationary nonbreeding period (0.39 ± 0.04, 95% CI 0.31-0.47), and across spring migration (0.03 ± 0.001, 95% CI 0.028-0.032). These results demonstrate that purple martins exhibit consistency in individual migration timing throughout the annual cycle that is likely driven by inherent individual circannual schedules. We additionally found that migration distance played a significant role, as the consistency of individual timing lessened over longer distances. Understanding how individual birds time migrations and if individuals are consistent between events can provide insight into how birds respond to shifts in their environment with climate change.

Jekyll and Hyde: Day–night personality differences in the water scorpionNepa hoffmanni

AbstractDifferences between daytime and night-time behaviors are common in insects due to potential trade-offs between predator avoidance and foraging. While diel effects on average behavior are well documented, it remains unclear whether individuals differ in how they adjust their behavior between daytime and night-time. In this study, we focused on two light dependent behaviors, activity and tonic immobility, in the water scorpion (Nepa hoffmanni). Each individual was assayed for each behavior six times during the day and six times at night. Then, we explored diel differences in (1) average behavior, (2) among-individual variation in behavior, (3) individual rank order of behavioral expression, and (4) the sign and magnitude of behavioral syndromes (among-individual correlations between behaviors). At night, N. hoffmanni was more active and exhibited shorter periods of tonic immobility. Among-individual variation in activity was greater at night than during the day, but among-individual variation in tonic immobility did not differ between day and night. Individuals also differed in their diel plasticity in activity but not in tonic immobility. Thus, both (1) day–night differences in among-individual variation and (2) individual variation in diel plasticity contributed to the light-dependent sign of the behavioral syndromes. Our findings highlight the need for individual-level studies to fully elucidate diel influences on animal behavior. We suggest that day–night differences in behavioral syndromes as well as in selection might be key to behavioral evolution, especially in animals active during both day and night.

Why Does Construct Validity Matter in Measuring Implementation Fidelity? A Methodological Case Study

In reading intervention research, implementation fidelity is assumed to be positively related to student outcomes, but the methods used to measure fidelity are often treated as an afterthought. Fidelity has been conceptualized and measured in many different ways, suggesting a lack of construct validity. One aspect of construct validity is the fidelity index of a measure. This methodological case study examined how different decisions in fidelity indices influence relative rank ordering of individuals on the construct of interest and influence our perception of the relation between the construct and intervention outcomes. Data for this study came from a large state-funded project to implement multi-tiered systems of support for early reading instruction. Analyses were conducted to determine whether the different fidelity indices are stable in relative rank ordering participants and if fidelity indices of dosage and adherence data influence researcher decisions on model building within a multilevel modeling framework. Results indicated that the fidelity indices resulted in different relations to outcomes with the most commonly used fidelity indices for both dosage and adherence being the worst performing. The choice of index to use should receive considerable thought during the design phase of an intervention study.

Dealing with Bad Risk in Cost-Effectiveness Analysis: The Cost-Effectiveness Risk-Aversion Curve.

Cost-effectiveness analysis has been advocated and is widely used to inform policy and decision makers in setting priorities for resource allocation. Since the costs and effects of health care interventions are uncertain, much research interest has focused on handling uncertainty in cost-effectiveness analysis. The most widely used method to summarize uncertainty in cost-effectiveness analysis is the cost-effectiveness acceptability curve, which estimates the probability that an intervention is cost effective for a wide range of threshold ratios. However, by estimating the uncertainty associated with incremental costs and effects, information about the uncertainty associated with the costs and effects of the individual programs is lost, which may be important to inform risk-averse decision makers. In the present paper, we suggest to penalize the expected net monetary benefit (NMB) of a program for its downside risk (i.e. bad risk), which preserves the uncertainty of the individual programs and rank orders programs according to their risk-adjusted NMB. The cost-effectiveness risk-aversion curve (CERAC) is introduced, which estimates the net benefit-to-risk ratio for a wide range of threshold rations. The CERAC is a helpful additional tool to inform decision and policy makers who are risk averse, and can easily be constructed using the results of a cost-effectiveness analysis.

Lifetime low behavioural plasticity of personality traits in the common vole (Microtus arvalis) under laboratory conditions

AbstractIndividual differences in behaviour, referred to as animal personality, are consistent across time and contexts. Nevertheless, personality traits show behavioural plasticity, much like many other phenotypic traits. In the present study, we examined the relationship between personality traits and behavioural plasticity in the common vole (Microtus arvalis) under stable, long‐lasting laboratory conditions. A total of 94 voles were tested in the classic open field test, designed to measure seven behavioural parameters (distance moved, grooming, immobility, rearing, running, scanning and walking duration) during a three‐minute test. A total of 60 voles formed the experimental group and were tested at four different time points over their lifetime (1st, 3rd, 5th and 7th month); 34 voles formed the control group and were tested only once at the 7th month. All voles were of the same age. Based on principal component analysis (PCA), two ordination axes were determined: “exploration” and “activity.” For further analyses, “distance moved” and “scanning duration” were selected from the first axis and “walking duration” from the second. Using linear mixed‐effect models (LMMs), we found highly significant random intercepts (i.e. personality traits) in all three behavioural parameters. However, evidence for behavioural plasticity was only found in the distance moved parameter, as determined from the random slope, and correlations between personality trait (intercept) and plasticity (slope) were not significant for any trait. During the experiment, variances of random effects were high and remained essentially the same, whilst the rank order of many individuals changed. Based on fixed effect slopes and a comparison with the control group, habituation was only significant for “walking duration.” The observed low behavioural plasticity could mirror stable (laboratory) conditions that result in the manifestation of original trait settings (genetic, early postnatal) or their gradual overcoming. These findings provide a starting point for further tests on free‐living voles.

Increasing reward prospect motivates switching to the more difficult task.

Computerized working-memory training (WM), despite typically yielding large practice effects in the training task, transfers reliably almost only to similar tasks and barely transfers to Fluid Intelligence (Gf). We hypothesized that WM training tasks gradually become less related to Gf due to the development of task-specific skills that reduce reliance on WM. As a result, what is being trained in the advanced stages of training is weakly related to WM and Gf. This hypothesis leads to predicting that with training progression, there would be a gradual change in the rank-ordering of individuals (quasi-simplex) in the per-session scores of the training task coupled with a trend in reduction in Gf loadings of these scores. We reanalyzed individual differences in persession scores in the training task from two moderately large-scale published studies. Results show that, as predicted, the correlations between per-session scores decreased with increasing temporal distance between sessions, suggesting a quasi-simplex pattern indicative of a gradual change in the rank-ordering of individuals. However, contrary to the prediction, the training tasks maintained or even tended to increase their Gf loading with training progression. We provide post-hoc accounts for these results, some which challenge prevalent assumptions beyond the attempt to improve Gf through improving WM.

Colored apparel and its potential influence on heterosexual attraction

AbstractThe objective of this study is to determine if men would follow the “red effect” when choosing colors for women to wear on a date, and also to determine if the colors that men would wear when going on a date would be the same as the colors that females (their date) would wish them to wear. A set of psychophysical data was generated from this experiment, where participants were asked to rank a set of 10 colored samples based on preference for each question asked. There were three different sets of colored samples. The set of colored samples given to the participant depended on the question. A total of five questions were asked. Scaling analysis was done on the data to organize a set of items according to preferences providing values, an interval scale (Z values), that correspond to the relative perceptual differences among the stimuli. The Z values were graphed to show the general preference of colors for women to wear, and the preference of colors for men to wear. A Spearman's rank‐order correlation coefficient (SRCC) was calculated comparing each individual's rank order with the mean rank order for that specific question. An average Spearman's rank order was calculated for each question and each gender in order to determine the variability in answers. Scaling results indicate that men follow the “red effect,” but women preferred to wear other colors such as turquoise, blue, or yellow depending on the outfit. Males and females agreed that no matter the colored bottoms (denim or black), blue was the preferred color top for men to wear. SRCC results showed a lot of variability between individual answers and the mean answer indicating that participants' rankings did not necessarily agree with general color preferences presented in the scaling analysis. While scaling analysis might suggest certain color preferences such as men following the “red effect” and women preferring to wear blue, the poor correlation found using SRCC between the individual answers and the mean rank orders suggests that color preferences for each individual are inherently unique.

Intergenerational continuity and stability in early family formation.

This study examines intergenerational continuity (mean level similarity) and stability (maintenance of rank ordering of individuals) in age and marital status at the time of becoming a young parent using prospective data from 3 generations of 585 families. G2 participants were recruited at the age of 5 years and followed until the age of 28, by which time 227 had become parents themselves. The findings suggest that despite dramatic intergenerational discontinuities with young adults, on average, now being more likely to be unmarried and older at the time of becoming parents than in previous generations, intergenerational stability in age and marital status at the time of becoming a young parent is still substantial. This intergenerational stability was, for the most part, not moderated by demographic, familial, or behavioral factors, suggesting that a developmental, multigenerational perspective is necessary to understand what has previously been considered a largely demographic issue. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Mechanisms of working memory training: Insights from individual differences

Computerized working-memory training (WM), despite typically yielding large practice effects in the training task, transfers reliably almost only to similar tasks and barely transfers to Fluid Intelligence (Gf). We hypothesized that WM training tasks gradually become less related to Gf due to the development of task-specific skills that reduce reliance on WM. As a result, what is being trained in the advanced stages of training is weakly related to WM and Gf. This hypothesis leads to predicting that with training progression, there would be a gradual change in the rank-ordering of individuals (quasi-simplex) in the per-session scores of the training task coupled with a trend in reduction in Gf loadings of these scores. We reanalyzed individual differences in per-session scores in the training task from two moderately large-scale published studies. Results show that, as predicted, the correlations between per-session scores decreased with increasing temporal distance between sessions, suggesting a quasi-simplex pattern indicative of a gradual change in the rank-ordering of individuals. However, contrary to the prediction, the training tasks maintained or even tended to increase their Gf loading with training progression. We provide post-hoc accounts for these results, some which challenge prevalent assumptions beyond the attempt to improve Gf through improving WM.

Development of math anxiety and its longitudinal relationships with arithmetic achievement among primary school children

The aim of this study is to examine the development of two separable aspects of math anxiety, anxiety about math-related situations and anxiety about failure in math, and their cross-lagged relationship with arithmetic achievement. The mean level of anxiety about math-related situations decreased among second, third, and fourth graders, and the level of anxiety about failure in math declined among third, fourth, and fifth graders. The rank-order of individuals was more stable in arithmetic achievement than in either aspect of math anxiety. Arithmetic achievement predicted later anxiety about failure in math, but neither aspect of math anxiety predicted later achievement. The results underline the importance of paying attention to math anxiety because anxiety about math-related situations seems to be as stable in primary school as it is in secondary school students. It is important to provide sufficient educational support and take into account affective factors related to learning from the beginning of schooling.

A14654 Prevalence and relative factors of metabolic syndrome in occupational population of heavy industry in northern China

Objectives: To investigate the prevalence and relative factors of metabolic syndrome (MS) for the heavy industry workers in northern coastal district of China. Methods: We performed an observational study on the heavy industry workers in Leting County, Hebei Province of China. The standard epidemiologic method for cardiovascular diseases was used. Items of survey included questionnaire, physical examination, fasting blood sample collection. Metabolic syndrome was defined in accordance with the International Diabetes Federation (IDF) criteria in 2004. Results: Totally 2045 participants (men 1651, women 394) with aged 37.5 ± 8.2 yrs were included in the data analysis. The mean age was 37.3 ± 8.3 yrs and 38.1 ± 8.0 years (P > 0.05) for men and women respectively. The total prevalence of MS was 19.1%, for men and women were 20.4% and 13.7%, respectively (P = 0.002). When compared the differences of prevalence of MS between age groups, there were no significant differences in men, but not in women. In the subjects with MS, a higher prevalence of alcohol drinking, snoring, higher salt taste, lower dairy intake and lower tea drinking were observed. The individual components rank order of MS in men was raised triglycerides, raised blood pressure, raised fasting glucose and reduced HDL cholesterol, and the order in women was raised triglycerides, reduced HDL cholesterol, raised blood pressure, and raised fasting glucose. Conclusion: The prevalence of MS in occupational population of heavy industry was higher than normal population, especially in young men below 30 years old. Unhealthy lifestyle such as alcohol drinking, higher salt taste, less drinking tea and so on should be the risks of higher prevalence of MS. Intervention to the risk factors of MS would be important in the prevention and control of cardiovascular diseases.

A Closer Look: Examining Distance within Status Hierarchies

Status hierarchies are a central part of group life. Scholars have devoted considerable attention to understanding the effects of status hierarchies. However, there is unresolved debate about the promises and perils of status hierarchies. In this symposium, we revisit a core assumption in how status hierarchies are typically conceptualized in the literature. Rather than viewing status solely in relative terms, as a rank-ordering of individuals on the basis of prestige, respect, and esteem, we introduce new theory and research that invokes the concept of status distance --- the absolute magnitude of status differences between individuals in a group. Across 5 methodologically diverse papers, we examine the individual, dyadic, and group-level effects of status distance. By exploring status distance from multiple angles, our hope is to open up new theoretical terrain and stimulate novel research questions. The contingent effects of scapegoating in work teams: The moderating role of hierarchy stability Presenter: Stuart Bunderson; Washington U. Presenter: Steven Gray; The U. of Texas at Austin Presenter: Gerben S. Van Der Vegt; U. of Groningen The interactive effect of status distance and newcomer status on willingness to help teammates Presenter: Sarah Doyle; U. of Arizona Presenter: Beth Polin; Eastern Kentucky U. Presenter: Robert B. Lount; Ohio State U. A qualitative investigation of the relationships between leaders and their second-in-commands Presenter: Benjamin Dow; The U. of Texas at Austin Perspective taking and employee deviance: The role of leaders dominance and prestige Presenter: Hemant Kakkar; London Business School Presenter: Niro Sivanathan; London Business School Do we agree on whos in charge here? Status dissensus antecedents and impact on team performance Presenter: Catarina Fernandes; Harvard Business School Presenter: Sujin Jang; INSEAD

Plasticity in activity and latency to explore differs between juvenile Atlantic cod Gadus morhua across a temperature gradient.

In the current study activity and latency to explore, as well as the correlation of these traits, were examined in individually marked juvenile Gadus morhua at 7, 10 and 13° C. It was concluded that individual rank order of both traits was maintained across temperature but that the level of change differed between individuals.

The trajectory of the blood DNA methylome ageing rate is largely set before adulthood: evidence from two longitudinal studies.

The epigenetic clock, defined as the DNA methylome age (DNAmAge), is a candidate biomarker of ageing. In this study, we aimed to characterize the behaviour of this marker during the human lifespan in more detail using two follow-up cohorts (the Young Finns study, calendar age i.e. cAge range at baseline 15-24years, 25-year-follow-up, N=183; The Vitality 90+ study, cAge range at baseline 19-90years, 4-year-follow-up, N=48). We also aimed to assess the relationship between DNAmAge estimate and the blood cell distributions, as both of these measures are known to change as a function of age. The subjects' DNAmAges were determined using Horvath's calculator of epigenetic cAge. The estimate of the DNA methylome age acceleration (Δ-cAge-DNAmAge) demonstrated remarkable stability in both cohorts: the individual rank orders of the DNAmAges remained largely unchanged during the follow-ups. The blood cell distributions also demonstrated significant intra-individual correlation between the baseline and follow-up time points. Interestingly, the immunosenescence-associated features (CD8+CD28- and CD4+CD28- cell proportions and the CD4/CD8 cell ratio) were tightly associated with the estimate of the DNA methylome age. In summary, our data demonstrate that the general level of Δ-cAge-DNAmAge is fixed before adulthood and appears to be quite stationary thereafter, even in the oldest-old ages. Moreover, the blood DNAmAge estimate seems to be tightly associated with ageing-associated shifts in blood cell composition, especially with those that are the hallmarks of immunosenescence. Overall, these observations contribute to the understanding of the longitudinal aspects of the DNAmAge estimate.

Tank size alters mean behaviours and individual rank orders in personality traits of fish depending on their life stage

Animals adjust their behaviour over time and contexts to cope with ecological and environmental variation. However, the presence of consistent between-individual differences in behaviour (i.e. personality) suggests that individual behaviour may be less flexible than previously thought. Here, we tested whether the size of the experimental tank and the ontogenetic stage of individuals affect estimates of average and consistent individual variation in activity and risk-taking behaviours in the eastern mosquitofish, Gambusia holbrooki, a fish model widely utilized for behavioural research. We measured risk taking and activity in juvenile, subadult and adult mosquitofish in a standard open-field test, in which the size of test tank varied linearly. We found strong evidence that spatial constraints alter mosquitofish behaviour. In particular, we observed that average activity increased with tank size, while the willingness of fish to take risks was independent of tank size. Moreover, juvenile fish exhibited, on average, lower risk-taking behaviours than older individuals. We highlight that the use of differently sized tanks may result in unequal variation in the average behaviour between juvenile and older fish, with escaping abilities of juvenile fish being underestimated in small environments. Most interestingly, we observed that variation in tank size triggered changes in the individual rank order for both risk taking and activity in juvenile fish, thus altering their personality estimates. In contrast, adult fish maintained consistent individual rank orders across all tank sizes. This study supports the hypothesis that behavioural repeatability increases with age, suggesting that personality estimates on adult animals may be less vulnerable to variation in laboratory contexts than those on juvenile ones.

Variation in behavioural plasticity regulates consistent individual differences in Enallagma damselfly larvae

Plastic behavioural responses by individuals to different conditions and consistent individual differences in mean behaviour across situations both contribute to variation in a population. The relationship between behavioural plasticity and consistent individual differences is not clearly understood but may help predict personality variation in animals. High variation in mean behaviour and low variation in individual plastic responses will tend to maintain the rank order of individuals across situations and so permit consistent individual differences. Conversely, low variation in mean behaviour and high variation in plastic responses, by changing the rank orders of individuals, will erode consistent individual differences. Thus, selection that reduces variation in individual plastic responses should increase the opportunity for consistent individual differences in a population. We tested for relationships between heterogeneous predation regimes, the mean and variance of behavioural plasticity and consistent individual differences among three species groups of larval damselflies. Larvae of Enallagma signatum probably face consistent predation from fish over successive generations, whereas Enallagma ebrium/hageni and Enallagma annexum/boreale face a changing predation regime over generations either from fish or larval dragonflies. The behavioural reaction norms of larvae in repeated exposure trials to cues from a predatory fish, dragonfly larvae or no predator differed between species groups. Enallagma ebrium/hageni expressed the most consistent plastic response to predator cues, less variability in plasticity and greater consistent individual differences across cues compared to more variable plastic responses and low consistent individual differences in E. signatum. Selection on behavioural plasticity may enhance consistent individual differences in E. ebrium/hageni whereas relaxing selection on plasticity may reduce consistent individual differences in E. signatum. More generally, selection on plastic behaviour may enhance behavioural types while selection on mean behaviour may reduce behavioural types in animal populations.

Hypoxia tolerance variance between swimming and resting striped bass Morone saxatilis.

Individual striped bass Morone saxatilis were each exposed in random order to aquatic hypoxia (10% air saturation) either while swimming at 50% of their estimated critical swimming speed (Ucrit ) or while at rest until they lost equilibrium. Individuals were always less tolerant of hypoxia when swimming (P < 0.01); the average fish was over five times more tolerant to the same hypoxia exposure when not swimming. There was no relationship between an individual's rank order of hypoxia tolerance (HT) under the two flow regimes, suggesting that different factors determine an individual's HT when at rest than when swimming.

The Short Questionnaire to Assess Health-Enhancing (SQUASH) Physical Activity in Adolescents: A Validation Using Doubly Labeled Water.

Accurate assessment of physical activity energy expenditure (PAEE) among adolescents is important for surveillance, evaluating interventions, and understanding the relation between energy balance and normal physiological and behavioral growth and development. The purpose of this study was to examine the validity of the Short Questionnaire to Assess Health-Enhancing Physical Activity (SQUASH)13 for measuring PAEE among adolescents. The participants were seventeen adolescents (9 females; Mean age = 17.53; SD = 0.62). Energy expenditure was measured during a 9-day period with doubly labeled water (DLW). The SQUASH was self-administered on the morning of the 10th day and assessed commuting activities, leisure time activities, household activities, and activities at work and school over the previous 9 days. A Bland-Altman plot indicated that the SQUASH underestimated PAEE compared with DLW by a mean difference of 126 kcal·d-1 (95% limits of agreement: -1,207 to 1,459 kcal·d-1), representative of a 10% underestimation. The Spearman rank order correlation coefficient showed there was a significant association between the SQUASH and DLW (r = .50, P = .04), for estimating PAEE. When using a sample of highly active adolescents, the SQUASH is a valid self-report tool for measuring PAEE at the group and individual rank order level.

Types de dominance agressive dans une colonie constituée exclusivement de males chimpanzés : intensité, fréquence, durée et direction de l'agression

Living in groups entails several fitness benefits but also leads to conflicts of interest. The strategies that different species have evolved to manage these conflicts are diverse and in some species aggressive exchanges of various kinds are commonplace. In some species, aggressions are important contributing elements to the organization of their social systems, providing a social order that may serve to avoid future agonistic encounters. Male chimpanzees are known for being gregarious, for exchanging complex social interactions and for forming dominance relationships where aggressions of different intensities can be frequent. Aggressions are costly interactions as they are energetically demanding but also because they encompass social consequences and risks. In this study we aimed to investigate the relationship between aggressive dominance styles and the intensity, the frequency, the duration and the direction of the aggressive interactions in a group of ten young male chimpanzees (“Pan troglodytes verus”) housed at la Vallée des Singes in France. First, we investigated whether aggressive interactions were costly behaviors; for this, we classified aggressions into three intensity levels, low, middle and high; we expected to find that if they were energetically demanding, then a negative relation between intensity and duration of aggressive interactions should be obtained. Second, we investigated whether the rank order of individuals varied depending on whether we assessed occurrence of aggression or duration of aggressive episodes for each one of the three levels and for the pooled category. Next, we correlated the two rank orders thus obtained. Finally, we studied whether aggressive dominance rank orders were consistent across the various aggression-based measures. As expected, we found that low intensity aggressions lasted longer than did medium and high aggressive exchanges. With both measures, occurrence and duration, we found similar aggressive patterns for the three levels of aggressions and for the totality of aggressive behaviors. Finally, we found some minor but interesting differences in the aggression-based dominance rankings when we compared individual measures, dyadic measures, the h’ de Vries dominance index and the David’s Score method. Our results suggest that due to the consistency of several measures, aggressive interactions are valuable to study aggression dominance in this species and that these interactions are selectively used. We also believe that although probing into the aggressive dominance relationships of male chimpanzees can provide a more nuanced view of their social relationships, a fuller understanding of this species’ group structure and dominance hierarchies should also incorporate other agonistic behavioral interactions (e.g., submission) and affiliative measures. Partly funded by project grant SFRH/BD/69069/2010 to NAV and PSI2011-29016-C02-01 to FC.

The impact of male contraception on dominance hierarchy and herd association patterns of African elephants (Loxodonta africana) in a fenced game reserve

Overpopulation of African elephants (Loxodonta africana) in fenced reserves in South Africa is becoming increasingly problematic to wildlife managers. With growing opposition to culling and the high cost of translocation, alternative management strategies focusing on male elephants are being investigated. In this study, hormonal treatment via Gonadotropin Releasing Hormone (GnRH) suppression, and surgical treatment via vasectomy were trialled. Focusing on behavioural responses, we tested the male dominance hierarchy for transitivity, and examined the rank order of individuals in relation to age and contraceptive treatment received. Additionally, we studied association patterns between males within the male population and with the female herds.Findings suggest that the treatment of one individual with GnRH suppressant is affecting the rank order of the dominance hierarchy, though it is still transitive, yet fluid (Landau’s linearity index h=0.7), as expected in a normal elephant population. Between males, association patterns were found to be weak. However, some males had relatively strong associations with the female herds, with association indices between 0.25 and 0.41. This suggests that the reduction on births is resulting in the males spending atypically large amounts of time with the female herds. The future conservation implications of this population control mechanism are discussed.