Repeatability and reliability of exploratory behavior in proactive and reactive zebrafish, Danio rerio

Abstract

Behavioral responses to novel situations often vary and can belong to a suite of correlated behaviors. Characteristic behaviors of different personality types (e.g. stress coping styles) are generally consistent across contexts and time. Here, we compare the repeatability and reliability of exploratory behaviors between zebrafish strains selectively bred to display contrasting behavioral responses to stressors that represent the proactive-reactive axis. Specifically, we measure exploratory behavior of individual fish in an open field test over five weeks. We quantified the stationary time, average swimming speed and time spent by a fish in the center area. We found a number of strain differences for each behavioral measure. Stationary time was the most repeatable and reliable measure for assessing proactive-reactive behavioral differences. Reactive zebrafish generally showed the highest reliability and repeatability of exploratory behavior compared to proactive zebrafish and a separate wild caught strain. Given the increased interest in the evolutionary consequences and proximate mechanisms of consistent individual differences, it will be important to continue to investigate how different selective pressures may influence expression of stress coping styles and their effects on the consistency of an animal’s behavior.

Introduction

Animals are frequently faced with a variety of stressors to their survival and reproductive efforts and typically employ behavioral and physiological responses to overcome them. While an individual’s behavioral response has been thought to be labile in variable environments, studies show that it can be consistent. Consistency of behavioral and physiological responses have both a between and within-individual component where the response in one context is often predictive of the animal’s response in a disparate context and across time1,2,3,4,5. In response to stressors, many studies document consistent differences between individuals across contexts in behaviors like aggression, boldness and sociability but the temporal component within the same context has not been thoroughly examined4,5,6,7,8.

Across many taxa there are two alternative correlated suites of behavioral and physiological responses to stressors known as proactive and reactive stress coping styles9,10. Proactive individuals are characterized by actively engaging stressors, having a feed-forward memory process, low glucocorticoid stress response, and low behavioral flexibility. In contrast, reactive individuals are more sensitive to environmental cues with reduced exploration of novel environments, a higher glucocorticoid stress response, and higher behavioral flexibility10. Both coping styles represent adaptive responses to environmental challenges and are maintained within a population due to fitness trade-offs in a variable environment1,5,11,12,13. Selective pressures such as predation and immune challenge can constrain behavioral variation within a population and strengthen correlations between traits, thereby facilitating correlated behaviors and physiology14,15,16.

Selectively bred strains of proactive and reactive behavioral phenotypes are often used to investigate the proximate mechanisms of stress coping styles and estimate heritability17. Few studies have investigated repeatability (between-individual consistency) and reliability (within-individual consistency) of behaviors in stress coping styles across multiple time points. Studies looking at aggressive behavior of selectively bred mice (Mus musculus) suggest that proactive individuals with low behavioral flexibility show greater reliability across trials than reactive individuals18,19. Similarly, proactive great tits (Parus major) establish more rigid routines during an open field test (OFT) and novel object test while reactive birds showed a higher degree of behavioral plasticity and lower reliability20. While some studies report conflicting results in teleosts21,22, selectively bred reactive trout (Oncorhynchus mykiss) display higher behavioral flexibility when a novel object was introduced during a feeding task23. Despite these observations, the repeatability and reliability of stress coping styles across time has not been well-established. Further, it is also not well understood how artificial selection may affect repeatability and reliability of behavior.

Exploratory behavior in a novel environment can be used to assess the magnitude of a stress response and categorize an individual into a stress coping style. Variation in exploratory behavior is widely studied and often utilizes the well-established OFT24,25,26. Within-individual variation of exploratory and other behaviors over time are influenced by factors such as prior experience, age, and motivation7,8. To investigate within-individual consistency across time, studies have used the reliability estimate, which measures the predictability of an animal’s performance on a measured variable over time relative to others within a population (0 = no predictability, 1 = perfect predictability). In human and animal personality studies, reliability estimates tend to range between 0.7–0.8527. A frequently used estimate to measure consistent differences between individuals is repeatability. The repeatability of a behavior is defined as the intraclass correlation coefficient and is calculated as the ratio of between-individual variance and the sum of between- and within-individual variance. Overall, exploratory behavior in a novel environment is repeatable in many avian, rodent, and teleost species with repeatability values ranging from 0.2–0.528,29,30. In addition to exploratory behavior, other behaviors such as thigmotaxis and movement speed have been commonly used to assess the magnitude of behavioral stress response and could also be indicative of an individual’s stress coping style31,32. Thigmotaxis and movement speed are generally repeatable like exploratory behavior33. It is unknown which behavior (exploratory behavior, thigmotaxis, or movement speed) is most suitable for use as an indicator of an individual’s stress coping style.

Zebrafish (Danio rerio) are a promising teleost system to understand the causes and consequences of correlated behavioral variation6,34,35. Both wild and laboratory strains of zebrafish display the proactive and reactive stress coping styles, and have distinct and heritable genetic architectures11,16,36,37,38,39,40,41,42. Proactive zebrafish are typically dominant and have higher reproductive success43,44. We have previously shown that selectively bred proactive and reactive zebrafish strains show consistent behavioral differences across a variety of contexts between the strains41. Additionally, artificial selection of exploratory behavior will constrain morphological evolution and glucocorticoid responses31,45. While the consistency of exploratory behavior has not been extensively studied in individual strains of proactive and reactive zebrafish, several studies have suggested that other boldness, aggression, and locomotor behaviors are generally consistent across contexts and time, and are influenced by selective pressures33,46,47,48. Thus, zebrafish can provide unique insights into underlying mechanisms of behavioral variation in coping with stress and subsequently how variation can be constrained by selective forces acting on populations.

In this study, we examined repeatability and reliability of three estimates of exploratory behavior in individual zebrafish during weekly OFT behavioral assays over five weeks. We used wild-derived strains selectively bred to display proactive and reactive stress coping styles to determine if (i) exploratory behavior is stable across time, (ii) repeatability and reliability measures differ between the stress coping style strains, and (iii) selectively bred strains are more repeatable or reliable than a separate wild caught population. By measuring variation of behavior within and between individuals, we can gain insight into factors contributing to the emergence and maintenance of stress coping styles in different populations11,49. As correlated traits constrain behavioral plasticity, they also have large implications regarding underlying genetics and heritability, which can lead to shifts in evolutionary trajectories50,51,52.

Methods

Subjects

We used three different zebrafish (Danio rerio) strains: wild caught (WC), high stationary behavior (HSB), and low stationary behavior (LSB). Wild caught fish were imported from North Bengal, India through a commercial supplier (Nebraska Aquatic Supply, Omaha, Nebraska, USA) and housed in the laboratory for 21 months before testing. The HSB and LSB strains were selected for stationary behavior (i.e. exploratory behavior) in an open field test and were 10 generations removed from a wild caught population from Gaighata in West Bengal, India41. The HSB and LSB strains display behaviors across multiple different behavioral assays, glucocorticoid responses, and morphology consistent with the reactive and proactive stress coping styles, respectively31,41,45,53,54. Additionally, HSB and LSB strains differ in neurotranscriptome profiles54,55. Females of both strains exhibit higher stationary time than males in an open field test41,56. We tested 28 individuals from LSB (N = 12 males, 16 females) and 27 each from HSB (N = 13 males, 14 females) and WC strains (N = 13 males, 14 females). LSB and HSB individuals were 13 months post-fertilization when testing began. Fish were individually housed in 3-liter tanks throughout the period of experiments on a recirculating water system (Pentair Aquatic Eco-Systems) using UV and solid filtration. Water temperature was set at 27 °C. Fish were kept on a 14:10 L/D cycle and fed twice a day with Tetramin Tropical Flakes (Tetra, USA). Morning feedings were prior to experiments on testing days.

Experiments

To test the repeatability and reliability of exploratory behavior, we used the open field test following established procedures26,41,54. Briefly, a plexiglass testing arena (30 × 30 × 10 cm) was filled with 4 L of system water. Animals were individually placed in the arena and video-recorded for 5 min. Each fish was tested once a week for five consecutive weeks between 8 and 10 h in the morning. The video recordings were analyzed with Noldus EthoVision XT (Noldus XT, Wageningen, Netherlands). For each fish, we quantified three estimates of exploratory behavior: stationary time, average swimming speed and time spent in the center. The subject was considered stationary if it was moving less than 0.5 cm/s and the center zone was defined as the 15 × 15 cm zone in the center of the chamber. We digitally measured standard length of each fish at end of the five weeks. There was a significant effect of strain on standard length (F2,79 = 35.84 p < 0.01). The WC strain (3.27 ± 0.05 cm) was significantly larger than the HSB (2.84 ± 0.04 cm; p < 0.01) and LSB lines (2.78 ± 0.05 cm; p < 0.01). HSB and LSB lines did not significantly differ in standard length (p = 0.42). Females (3.03 ± 0.04 cm) were significantly larger than males (2.90 ± 0.04 cm; F1,79 = 6.79 p = 0.01). All testing experiments were approved by the Institutional Animal Care and Use Committee of University of Nebraska at Omaha/University of Nebraska Medical Center (17-070-00-FC, 17-064-08-FC) and were performed in accordance with the relevant guidelines and regulations.

Statistics

Reliability of exploratory behavior across time and between-strain differences were tested using a repeated measures general linear model (GLM) in SPSS (Version 24). Sex and strain were included as between-subjects variables and standard length was controlled for by including it as a covariate. Since the assumption of sphericity was violated for each of the three exploratory behaviors we applied the Greenhouse-geisser correction. This did not change any statistical conclusions, therefore we only reported the uncorrected model. For the post-hoc comparisons of the estimated marginal means of the three estimates of exploratory behavior, we applied a Benjamini-Hochberg correction to reduce the likelihood of type I errors57.

To assess behavioral variation among strains, we used multilevel structural equation modeling in M-plus statistical analysis software58. This allowed us to control for shared associations between the behaviors in a single model, and more importantly test for differences in the variability within- and between-individuals across strains. First, we began with an unconditional model to assess the variability at each level (e.g. within- and between-individual variability). Next, the covariances between all the behaviors were included at each level. Then, we split the models by strain (LSB, HSB and WC) and constrained every part of the model in a step-wise fashion to ascertain which sources of variability were significantly different across the strains (first at the within-individual level and then the between-individual level). A constraint was considered to have worsened the model based on a significant chi-square test (p < 0.05). Any significantly worsening constraint reflects a difference in the estimates between strains.

Repeatability was defined as the intraclass correlation coefficient (R), which was calculated as the ratio of between-individual variance and the sum of between- and within-individual variance59. We calculated R based on variance components estimated from the multilevel structural equation model. Based on several literature meta-analyses60,61 we describe repeatability values as follows: low repeatability R ≤ 0.2; moderate repeatability 0.2 < R < 0.4; and high repeatability R ≥ 0.4. Reliability measures were estimated as the inter-trial reliability measure in SPSS (Version 24). Values > 0.8 were considered highly reliable and indicate that individuals maintained rank order across the five weeks of testing. Of note, repeatability and reliability values are deemed significantly different from a comparison value if they do not fall within that comparison value’s 95% confidence interval. All statistical tests were two-tailed, and were conducted with an alpha level of 0.05.

Results

Repeatability and reliability of exploratory behavior across time

There was a significant between-subjects effect of strain for stationary time (F2,79 = 15.75 p < 0.01, Table 1). The WC strain spent significantly less time stationary than the LSB (p < 0.01) and HSB strains (p < 0.01). The LSB strain also spent significantly less time stationary than the HSB strain (p = 0.03), as expected. Further, there was high repeatability for stationary behavior in the HSB (R = 0.71) and LSB (R = 0.56) strains, while the WC strain was moderately repeatable (R = 0.28). All three repeatability values were significantly different from each other for stationary time (Table 2). There was a significant effect of strain on swimming speed (F2,79 = 3.37 p = 0.04, Table 1). The WC strain swam significantly faster than the HSB (p = 0.02), but not the LSB strain (p = 0.26). HSB and LSB lines did not significantly differ in mean swimming speed (p = 0.13). Further, all strains showed high repeatability for mean swimming speed (WC: R = 0.40; HSB: R = 0.59; LSB: R = 0.55). The HSB strain had significantly higher repeatability than the WC, but not the LSB strain for mean swimming speed (Table 2). For time spent in the center there was an effect of strain (F2,79 = 40.73 p < 0.01, Table 1).The WC strain spent significantly less time in the center than the LSB (p < 0.01) and HSB strains (p < 0.01). The LSB strain spent significantly more time in the center zone than the HSB strain (p = 0.01). The HSB (R = 0.46) strain was significantly more repeatable than the WC (R = 0.21) and LSB (R = 0.10) strains. The LSB strain’s repeatability was not significantly different from zero (Table 2). All other between-individual factors were not significant.

Table 1 Results of repeated measures GLM for behavioral estimates across time and post hoc tests. For post-hoc results, estimated marginal mean (EMM) values with different superscript letters indicate significant differences.
Table 2 Repeatability values (intraclass correlation (95% confidence intervals)) of each behavior by strain (Abbreviations: HSB, high stationary behavior; LSB, low stationary behavior; WC, wild caught).

There were significant interaction effects of week*strain (F4, 300 = 2.97 p < 0.01) and week*standard length (F4,300 = 2.95 p = 0.02) for time spent in the center (Table 1, Fig. 1). Further, the HSB strain was significantly more reliable (Reliability = 0.85) than the LSB (Reliability = 0.58) and WC (Reliability = 0.57) strains for time spent in the center (Table 3). There were no significant effects of week, nor any significant interaction effects for stationary time (all p > 0.05; Fig. 1). The HSB strain was significantly more reliable (Reliability = 0.94) compared to WC (Reliability = 0.70), but not LSB (Reliability = 0.86) fish for stationary time (Table 3). There were no significant effects of week, nor any significant interaction effects for mean swimming speed (all p > 0.05; Fig. 1). The HSB strain (Reliability = 0.90) was significantly more reliable than the WC (Reliability = 0.80), but not the LSB fish (Reliability = 0.85) for mean swimming speed (Table 3).

Figure 1
figure1

Stationary time (a), mean swimming speed (b), and time in center (c) performance across five weeks of testing for each zebrafish strain. Points represent weekly means (±1 SE). The blue circle, red triangle, and green square points represent the low stationary behavior (LSB), high stationary behavior (HSB), and wild-caught (WC) strains, respectively. There was a significant strain*time effect for time spent in the center.

Table 3 Reliability values (95% confidence intervals) of each behavior by strain (Abbreviations: HSB, high stationary behavior; LSB, low stationary behavior; WC, wild caught).

Behavioral variation across strains

The unconditional model revealed that between 50.03–60.78% of the variability in mean swimming speed, stationary time, and time spent in the center was at the between-individual level. Not surprisingly, this represented a significant proportion of within- and between-individual variability for all three estimates of exploratory behavior (p < 0.05). The resulting model provided adequate fit (χ2(3) = 1.15, p > 0.05, CFI = 1.00, RMSEA < 0.01, SRMR(within) < 0.01, SRMR(between) = 0.10). After constraining the variability at the within-individual level, a number of differences emerged (Table 4). Specifically, the HSB strain had significantly less variability in both mean swimming speed and stationary time at the within-individual level. Meanwhile, the WC strain had significantly less variability in time spent in the center. At the between-individual level, the LSB strain had significantly higher variability in mean swimming speed. Time spent in the center was significantly more variable in the HSB strain. Lastly, the WC strain had significantly lower variability in stationary behavior.

Table 4 Multi-level modeling differences in the sources of variability as a function of strain.

Discussion

An animal’s tendency to explore during unpredictable or risky situations is indicative of its stress coping style. It is unclear whether behavioral traits of a stress coping style are repeatable and reliable. Here, we found effects of strain, but not sex or standard length, on within- and between-individual variation for three estimates of exploratory behavior. All three behavioral estimates were generally repeatable and reliable. Overall, the selectively bred strains of zebrafish showed higher repeatability and reliability values compared to the WC fish. Further, the HSB strain showed remarkably high repeatability and reliability for all three behaviors, and had significantly less within-individual variability compared to LSB and WC strains for both swimming speed and stationary behavior. Stationary time was the most repeatable behavioral measure and was consistent across time. Time spent in the center zone showed the greatest variability across weeks compared to stationary time and swimming speed. While several studies have found sex-specific behavioral variation of mating and aggressive behaviors46,47,60, we did not observe any sex differences in exploratory behavior.

High exploratory behavior in a novel environment is characteristic of the proactive stress coping style9,10. Previous studies have demonstrated that selection on exploratory behavior can strengthen correlations between other stress coping behaviors in other contexts, glucocorticoid levels, and morphology31,41,45. Here, we show that artificial selection can also constrain behavioral variation in populations across time. HSB fish showed significantly higher stationary behavior than LSB animals, which is consistent with previous studies41,54. While the WC strain was moderately repeatable (R = 0.29), both of the LSB (R = 0.56) and HSB (R = 0.71) strains showed high repeatability values for stationary time (Table 2). There was no effect of time, and selectively bred fish maintained rank order across the five time points (Reliability > 0.8, Table 3). Further, the HSB line showed significantly less within-individual variability compared to LSB and WC strains (Table 4). Emergence of consistent individual differences in the presence of selective pressures are also documented in other species14,15. Field crickets (Gryllus integer) exposed to a common bacterial pathogen showed increased repeatability in their tendency to explore a novel environment15. Similarly, boldness and aggression behaviors were correlated in wild sticklebacks (Gasterosteus aculeatus) only after exposing the population to predation14. In great tits, correlation between exploratory behavior and stress physiology emerged through selectively bred proactive and reactive birds but not wild individuals62. These studies suggest that selection can influence expression of repeatability and reliability by potentially placing survival costs on individuals. Predation has frequently been identified as one of the strongest ecological pressures that can influence the repeatability and reliability of animal behavior11,16,47. We speculate that by selecting for an ecologically relevant behavioral response to a stressor in the HSB and LSB lines within the lab, it may have been simulating selection of behavioral responses to predation in the wild. Low exploratory behavior may be adaptive and directly selected for in environments with high predation. This could explain why the HSB fish showed the highest repeatability and reliability values for a majority of the behavioral estimates. Altogether, this indicates that artificial selection may act similarly to natural selection and increase the repeatability and reliability of behavior.

Increased thigmotaxis in many species is indicative of higher stress levels and an aversion to being exposed in the center of a novel environment63,64. Despite being a measure of stress and type of exploratory behavior, thigmotaxis has not been commonly used to predict an animal’s stress coping style. We observed that the repeatability of time in the center in the HSB, WC, and LSB strains was high, low, and not repeatable, respectively (Table 2). Similarly, HSB fish maintained rank order (Reliability = 0.85), whereas LSB and WC fish did not (Reliability < 0.6; Table 3). The WC strain had the highest amount of thigmotaxis (indicating high stress levels) but spent the least amount of time stationary (suggestive of proactive stress coping style). The combination of low repeatability and reliability for all but the HSB strain and conflicting behavioral interpretations in the WC strain suggest that thigmotaxis may not be an accurate proxy measure for proactive-reactive tendencies.

It should be noted that the WC individuals were imported from North Bengal, India, which is a different location than the founding animals used to generate the HSB and LSB strains (West Bengal, India). The local ecological factors that might have contributed to shaping the WC animals’ behavior are not known. It is possible that some population differences or behavioral correlations may only emerge under certain local environmental conditions16,46,65. Several studies that investigated wild zebrafish populations found that predation and water flow can explain population differences in behavioral correlations between boldness, aggression and activity16,46. While these studies examined between-population differences, similar patterns have been observed within populations across time in collared flycatcher birds (Ficedula albicollis). Behavioral correlations appeared and disappeared across years that coincided with changes in the density and age composition of the bird population, which could reflect changes in resource availability65. Our results suggest that the ecological pressures acting on the WC population shaped different behavioral patterns compared to those resulting from artificial selection.

Studies show that a faster swimming speed and larger body size are suggestive of a proactive coping style31,66. There were no significant differences between the HSB and LSB strains in standard length or mean swimming speed. Although a previous study showed that the LSB strain swims faster and has a larger caudal region compared to the HSB strain, this was examined using morphometrics and within a startle-response paradigm31. Measuring standard length does not allow for inferring size of specific body areas (e.g. caudal region). In our study we also measured average swim speed over a five minute period within the open field test and did note evoke a startle-response. All three strains were highly repeatable for average swimming speed (Table 2), similar to wild trout (Salmo trutta)33. There was no effect of time across weeks (Table 1), and all strains maintained rank order (Table 3) for mean swimming speed. Further, HSB fish showed higher repeatability and reliability values than the WC strain, as well as significantly lower within-individual variability compared to LSB and WC strains (Table 4). Similar to stationary time, the selection force shaping the behavior may have a greater effect on the HSB than the LSB strain. It is possibly related to environmental contexts where high or low exploratory behavior would be adaptive and selected for. However, with no significant difference in swimming speed between the HSB and LSB strains, other measures such as maximum velocity or acceleration should be considered in future studies for assessing stress coping style differences.

Previous studies suggest that proactive individuals with low behavioral flexibility and rigid behavior patterns are more consistent than reactive individuals18,19,20. Our results show the opposite pattern. HSB individuals had significantly less within-individual variability in stationary time and mean swimming speed across five weeks (Table 4). Further, the HSB strain had the highest repeatability and reliability values for each of the behavioral measures (Tables 2 and 3). It is possible that the conflicting observations with prior studies can be attributed to different focal behavior measured or the strength of selection on the behavior. For example, in two mice lines bidirectionally selected for divergent aggressive behaviors towards conspecifics, proactive individuals showed higher consistency of aggressive behaviors compared to reactive individuals18,19. In trout bidirectionally selected for divergent cortisol reactivity to confinement stress, reactive trout displayed higher behavioral flexibility across trials when a novel object was introduced to a feeding task23. The current study used zebrafish strains bidirectionally selected for opposing exploratory behavior in response to a novelty stressor. Thus, the behavioral flexibility trait of stress coping styles appear to vary across stressor types (e.g. conspecific, novel object, novel environment) and selectively bred traits (e.g. behavioral and physiological).

It is also important to be able to separate out changes in consistency across time from those that could be due to changes in contexts between assay time points7. In studies examining repeatability and reliability within a social context such as aggressive behaviors18,19 and predator inspection67,68, it is difficult to ensure consistency of behaviors and motivation of live stimulus animal across trials. Thus, the context the focal animal experiences may subtly vary across testing periods and make it difficult to understand if results are due to stimulus animal behavioral state or within-individual variation over time. In using the OFT we removed potential confounds of varying contexts over time and therefore are confident we measured within-individual variation. Even when these changing contexts are taken into account, other internal developmental factors can also influence the repeatability and reliability of behavior. Despite using a similar methodology as the current study, selectively bred proactive great tit birds were more consistent over time in exploratory behavior relative to the reactive birds20. It is noteworthy that the two assay time points were across developmentally distinct periods (once in juveniles and once in adults), which has been shown to influence repeatability60. In the current study, all zebrafish were sexually mature adults at the time of testing and were assayed over five weeks, which could explain the inconsistency of our findings.

There are many key considerations when estimating repeatability and reliability of animal behavior. Often studies have estimated repeatability by using two measurements for each individual, which can overlook any behavioral changes that may occur over longer periods of time or multiple observations69. This is especially important given that a central assumption of stress coping styles is that a behavioral phenotype is maintained over time, despite the observation that animal behavior can be very labile over small durations61. Here, we found that stationary time was the most repeatable and reliable estimate of exploratory behavior over five repeated observations. Additionally, the artificially selected proactive and reactive strains showed higher repeatability and reliability values compared to the wild caught population. This suggests that in populations under high levels of selection, a single to a few measurements for the examined behaviors can be a sufficient representation of that individual’s behavior. While increasing the number of repeated trials allow for more robust repeatability estimates, short inter-trial intervals can lead to habituation or other forms of associative learning. These types of learning can have confounding effects on estimates of repeatability and reliability7,70. If animals habituate during repeated measurements, there is weaker construct validity of assessing consistency of stress coping behaviors. We found no evidence of habituation in our study with a one week intertrial interval (Fig. 1). Overall, behavioral repeatability and reliability metrics are important for insight into selective pressures that may support development of stress coping styles. Future studies should test the fitness consequences associated with high or low behavioral consistency. Additionally, while stress coping styles have been well documented to have a genetic basis, the neural and molecular mechanisms are only beginning to be explored. Artificial bidirectional selection can serve as a complementary approach to understanding proximate mechanisms underlying consistent individual differences in stress coping behaviors.

Data availability

All data generated or analyzed during this study are included in this published article and its Supplementary Information files.

References

  1. 1.

    Dingemanse, N. J., Both, C., Drent, P. J. & Tinbergen, J. M. Fitness consequences of avian personalities in a fluctuating environment. Proc. R. Soc. B Biol. Sci. 271, 847–852 (2004).

    Article  Google Scholar 

  2. 2.

    Sih, A. & Bell, A. M. Chapter 5 Insights for Behavioral Ecology from Behavioral Syndromes. Advances in the Study of Behavior 38, 227–281 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

  3. 3.

    Sih, A., Bell, A. M., Johnson, J. C. & Ziemba, R. E. Behavioral Syndromes: An Integrative Overview. Q. Rev. Biol. 79, 241–277 (2004).

    Article  PubMed  Google Scholar 

  4. 4.

    Sih, A. et al. Animal personality and state-behaviour feedbacks: A review and guide for empiricists. Trends in Ecology and Evolution 30, 50–60 (2015).

    Article  PubMed  Google Scholar 

  5. 5.

    Dall, S. R. X. & Griffith, S. C. An empiricist guide to animal personality variation in ecology and evolution. Front. Ecol. Evol. 2 (2014).

  6. 6.

    Baker, M. R., Hofmann, H. A. & Wong, R. Y. Neurogenomics of Behavioural Plasticity in Socioecological Contexts, https://doi.org/10.1002/9780470015902.a0026839.

  7. 7.

    Stamps, J. A. Individual differences in behavioural plasticities. Biol. Rev. 91, 534–567 (2016).

    Article  PubMed  Google Scholar 

  8. 8.

    Stamps, J. A., Briffa, M. & Biro, P. A. Unpredictable animals: Individual differences in intraindividual variability (IIV). Anim. Behav. 83, 1325–1334 (2012).

    Article  Google Scholar 

  9. 9.

    Øverli, Ø. et al. Evolutionary background for stress-coping styles: Relationships between physiological, behavioral, and cognitive traits in non-mammalian vertebrates. Neuroscience and Biobehavioral Reviews 31, 396–412 (2007).

    Article  PubMed  CAS  Google Scholar 

  10. 10.

    Koolhaas, J. M., de Boer, S. F., Coppens, C. M. & Buwalda, B. Neuroendocrinology of coping styles: Towards understanding the biology of individual variation. Frontiers in Neuroendocrinology 31, 307–321 (2010).

    Article  PubMed  CAS  Google Scholar 

  11. 11.

    Moretz, J. A., Martins, E. P. & Robison, B. D. Behavioral syndromes and the evolution of correlated behavior in zebrafish. Behav. Ecol. 18, 556–562 (2007).

    Article  Google Scholar 

  12. 12.

    Sih, A., Bell, A. & Johnson, J. C. Behavioral syndromes: An ecological and evolutionary overview. Trends in Ecology and Evolution 19, 372–378 (2004).

    Article  PubMed  Google Scholar 

  13. 13.

    Smith, B. R. & Blumstein, D. T. Fitness consequences of personality: A meta-analysis. Behavioral Ecology 19, 448–455 (2008).

    Article  Google Scholar 

  14. 14.

    Bell, A. M. & Sih, A. Exposure to predation generates personality in threespined sticklebacks (Gasterosteus aculeatus). Ecol. Lett. 10, 828–834 (2007).

    Article  PubMed  Google Scholar 

  15. 15.

    DiRienzo, N., Niemelä, P. T., Hedrick, A. V. & Kortet, R. Adult bacterial exposure increases behavioral variation and drives higher repeatability in field crickets. Behav. Ecol. Sociobiol. 70, 1941–1947 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  16. 16.

    Martins, E. P. & Bhat, A. Population-level personalities in zebrafish: Aggression-boldness across but not within populations. Behav. Ecol. 25, 368–373 (2014).

    Article  Google Scholar 

  17. 17.

    Van Oers, K., De Jong, G., Van Noordwijk, A. J., Kempenaers, B. & Drent, P. J. Contribution of genetics to the study of animal personalities: A review of case studies. Behaviour 142, 1185–1206 (2005).

    Article  Google Scholar 

  18. 18.

    Benus, R. F., Daas, S., den, Koolhaas, J. M. & van Oortmerssen, G. A. Routine Formation and Flexibility in Social and Non-Social Behaviour of Aggressive and Non-Aggressive Male Mice. Behaviour 112, 176–193 (1990).

    Article  Google Scholar 

  19. 19.

    Benus, R. F., Bohus, B., Koolhaas, J. M. & van Oortmerssen, G. A. Heritable variation for aggression as a reflection of individual coping strategies. Experientia 47, 1008–1019 (1991).

    Article  PubMed  CAS  Google Scholar 

  20. 20.

    Carere, C., Drent, P. J., Privitera, L., Koolhaas, J. M. & Groothuis, T. G. G. Personalities in great tits, Parus major: Stability and consistency. Anim. Behav. 70, 795–805 (2005).

    Article  Google Scholar 

  21. 21.

    Castanheira, M. F., Herrera, M., Costas, B., Conceição, L. E. C. & Martins, C. I. M. Can We Predict Personality in Fish? Searching for Consistency over Time and across Contexts. PLoS One 8 (2013).

  22. 22.

    Laursen, D. C. et al. Behavioural responses to hypoxia provide a non-invasive method for distinguishing between stress coping styles in fish. Appl. Anim. Behav. Sci. 132, 211–216 (2011).

    Article  Google Scholar 

  23. 23.

    Ruiz-Gomez, M. et al. Response to environmental change in rainbow trout selected for divergent stress coping styles. Physiol. Behav. 102, 317–322 (2011).

    Article  CAS  Google Scholar 

  24. 24.

    Carter, A. J., Feeney, W. E., Marshall, H. H., Cowlishaw, G. & Heinsohn, R. Animal personality: What are behavioural ecologists measuring? Biol. Rev. 88, 465–475 (2013).

    Article  PubMed  Google Scholar 

  25. 25.

    Godwin, J., Sawyer, S., Perrin, F., Oxendine, S. & D. Kezios, Z. Adapting the Open Field Test to Assess Anxiety-Related Behavior in Zebrafish. Zebrafish Protocols for Neurobehavioral Research, https://doi.org/10.1007/978-1-61779-597-8_13 (2012).

  26. 26.

    Perals, D., Griffin, A. S., Bartomeus, I. & Sol, D. Revisiting the open-field test: what does it really tell us about animal personality? Anim. Behav. 123, 69–79 (2017).

    Article  Google Scholar 

  27. 27.

    Gosling, S. D. From mice to men: What can we learn about personality from animal research? Psychol. Bull. 127, 45–86 (2001).

    Article  PubMed  CAS  Google Scholar 

  28. 28.

    Boulton, K. et al. How integrated are behavioral and endocrine stress response traits? A repeated measures approach to testing the stress-coping style model. Ecol. Evol. 5, 618–633 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  29. 29.

    Dingemanse, N. J. et al. Variation in personality and behavioural plasticity across four populations of the great tit Parus major. J. Anim. Ecol. 81, 116–126 (2012).

    Article  PubMed  Google Scholar 

  30. 30.

    Schuster, A. C., Carl, T. & Foerster, K. Repeatability and consistency of individual behaviour in juvenile and adult eurasian harvest mice. Sci. Nat. 104 (2017).

  31. 31.

    Kern, E. M. A., Robinson, D., Gass, E., Godwin, J. & Langerhans, R. B. Correlated evolution of personality, morphology and performance. Anim. Behav. 117, 79–86 (2016).

    Article  PubMed  PubMed Central  Google Scholar 

  32. 32.

    Leppänen, P. K., Ravaja, N. & Ewalds-Kvist, S. B. M. Twenty-three generations of mice bidirectionally selected for open-field thigmotaxis: Selection response and repeated exposure to the open field. Behav. Processes 72, 23–31 (2006).

    Article  PubMed  Google Scholar 

  33. 33.

    Adriaenssens, B. & Johnsson, J. I. Natural selection, plasticity and the emergence of a behavioural syndrome in the wild. Ecol. Lett. 16, 47–55 (2013).

    Article  PubMed  Google Scholar 

  34. 34.

    Conrad, J. L., Weinersmith, K. L., Brodin, T., Saltz, J. B. & Sih, A. Behavioural syndromes in fishes: A review with implications for ecology and fisheries management. Journal of Fish Biology 78, 395–435 (2011).

    Article  PubMed  CAS  Google Scholar 

  35. 35.

    Khan, K. M. & Echevarria, D. J. Feeling Fishy: Trait Differences inZebrafish (Danio Rerio) BT - Personality in Nonhuman Animals. In (eds Vonk, J., Weiss, A. & Kuczaj, S. A.)111–127, https://doi.org/10.1007/978-3-319-59300-5_6 (Springer International Publishing, 2017).

  36. 36.

    Ariyomo, T. O., Carter, M. & Watt, P. J. Heritability of boldness and aggressiveness in the zebrafish. Behav. Genet. 43, 161–167 (2013).

    Article  PubMed  Google Scholar 

  37. 37.

    Norton, W. H. J. et al. Modulation of Fgfr1a Signaling in Zebrafish Reveals a Genetic Basis for the Aggression-Boldness Syndrome. J. Neurosci. 31, 13796–13807 (2011).

    Article  PubMed  CAS  Google Scholar 

  38. 38.

    Oswald, M. E., Drew, R. E., Racine, M., Murdoch, G. K. & Robison, B. D. Is Behavioral Variation along the Bold-Shy Continuum Associated with Variation in the Stress Axis in Zebrafish? Physiol. Biochem. Zool. 85, 718–728 (2012).

    Article  PubMed  Google Scholar 

  39. 39.

    Oswald, M. E., Singer, M. & Robison, B. D. The Quantitative Genetic Architecture of the Bold-Shy Continuum in Zebrafish, Danio rerio. PLoS One 8 (2013).

  40. 40.

    Rey, S., Boltana, S., Vargas, R., Roher, N. & Mackenzie, S. Combining animal personalities with transcriptomics resolves individual variation within a wild-type zebrafish population and identifies underpinning molecular differences in brain function. Mol. Ecol. 22, 6100–6115 (2013).

    Article  PubMed  CAS  Google Scholar 

  41. 41.

    Wong, R. Y. et al. Comparing behavioral responses across multiple assays of stress and anxiety in zebrafish (Danio rerio). Behaviour 149, 1205–1240 (2012).

    Article  Google Scholar 

  42. 42.

    Zajitschek, S., Herbert-Read, J. E., Abbasi, N. M., Zajitschek, F. & Immler, S. Paternal personality and social status influence offspring activity in zebrafish. BMC Evol. Biol. 17 (2017).

  43. 43.

    Ariyomo, T. O. & Watt, P. J. The effect of variation in boldness and aggressiveness on the reproductive success of zebrafish. Anim. Behav. 83, 41–46 (2012).

    Article  Google Scholar 

  44. 44.

    Dahlbom, S. J., Lagman, D., Lundstedt-Enkel, K., Sundström, L. F. & Winberg, S. Boldness predicts social status in zebrafish (Danio rerio). PLoS One 6 (2011).

  45. 45.

    Russ, J. Differences in Behavioral and Physiological Responses to Stress in Zebrafish. (University of Nebraska at Omaha, 2018).

  46. 46.

    Roy, T. & Bhat, A. Population, sex and body size: determinants of behavioural variations and behavioural correlations among wild zebrafish Danio rerio. R. Soc. Open Sci. 5, 170978 (2018).

    Article  PubMed  PubMed Central  Google Scholar 

  47. 47.

    Roy, T., Shukla, R. & Bhat, A. Risk-Taking During Feeding: Between- and Within-Population Variation and Repeatability Across Contexts Among Wild Zebrafish. Zebrafish zeb. 2017, 1442, https://doi.org/10.1089/zeb.2017.1442 (2017).

    Article  Google Scholar 

  48. 48.

    Conradsen, C., Walker, J. A., Perna, C. & McGuigan, K. Repeatability of locomotor performance and morphology–locomotor performance relationships. J. Exp. Biol. 219, 2888–2897 (2016).

    Article  PubMed  Google Scholar 

  49. 49.

    Dingemanse, N. J. et al. Behavioural syndromes differ predictably between 12 populations of three-spined stickleback. J. Anim. Ecol. 76, 1128–1138 (2007).

    Article  PubMed  Google Scholar 

  50. 50.

    Bell, A. M. Future directions in behavioural syndromes research. Proc. R. Soc. B Biol. Sci. 274, 755–761 (2007).

    Article  Google Scholar 

  51. 51.

    Dochtermann, N. A. & Dingemanse, N. J. Behavioral syndromes as evolutionary constraints. Behav. Ecol. 24, 806–811 (2013).

    Article  Google Scholar 

  52. 52.

    van Oers, K. & Mueller, J. C. Evolutionary genomics of animal personality. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 365, 3991–4000 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  53. 53.

    Wong, R. Y., Oxendine, S. E. & Godwin, J. Behavioral and neurogenomic transcriptome changes in wild-derived zebrafish with fluoxetine treatment. BMC Genomics 14, 348 (2013).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  54. 54.

    Wong, R. Y., Lamm, M. S. & Godwin, J. Characterizing the neurotranscriptomic states in alternative stress coping styles. BMC Genomics 16, 425 (2015).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  55. 55.

    Wong, R. Y. & Godwin, J. Neurotranscriptome profiles of multiple zebrafish strains. Genomics Data 5, 206–209 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  56. 56.

    Wong, R. Y., McLeod, M. M. & Godwin, J. Limited sex-biased neural gene expression patterns across strains in Zebrafish (Danio rerio). BMC Genomics 15, 905 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  57. 57.

    Benjamini, Y., Drai, D., Elmer, G., Kafkafi, N. & Golani, I. Controlling the false discovery rate in behavior genetics research. Behav. Brain Res. 125, 279–284 (2001).

    Article  PubMed  CAS  Google Scholar 

  58. 58.

    Muthén, L. K. & Muthén, B. O. Mplus User’s Guide. J. Am. Geriatr. Soc. 2006, 676 (2007).

    Google Scholar 

  59. 59.

    Boake, C. R. B. Repeatability: Its role in evolutionary studies of mating behavior. Evol. Ecol. 3, 173–182 (1989).

    Article  Google Scholar 

  60. 60.

    Bell, A. M., Hankison, S. J. & Laskowski, K. L. The repeatability of behaviour: a meta-analysis. Animal Behaviour 77, 771–783 (2009).

    Article  PubMed  PubMed Central  Google Scholar 

  61. 61.

    Wolak, M. E., Fairbairn, D. J. & Paulsen, Y. R. Guidelines for estimating repeatability. Methods Ecol. Evol. 3, 129–137 (2012).

    Article  Google Scholar 

  62. 62.

    Baugh, A. T. et al. Corticosterone responses differ between lines of great tits (Parus major) selected for divergent personalities. Gen. Comp. Endocrinol. 175, 488–494 (2012).

    Article  PubMed  CAS  Google Scholar 

  63. 63.

    Lamprea, M. R., Cardenas, F. P., Setem, J. & Morato, S. Thigmotactic responses in an open-field. Brazilian J. Med. Biol. Res. 41, 135–140 (2008).

    Article  CAS  Google Scholar 

  64. 64.

    Stewart, A. M., Gaikwad, S., Kyzar, E. & Kalueff, A. V. Understanding spatio-temporal strategies of adult zebrafish exploration in the open field test. Brain Res. 1451, 44–52 (2012).

    Article  PubMed  CAS  Google Scholar 

  65. 65.

    Garamszegi, L. Z. et al. Among-year variation in the repeatability, within- and between-individual, and phenotypic correlations of behaviors in a natural population. Behav. Ecol. Sociobiol. 69, 2005–2017 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  66. 66.

    Brown, C., Jones, F. & Braithwaite, V. A. Correlation between boldness and body mass in natural populations of the poeciliid Brachyrhaphis episcopi. J. Fish Biol. 71, 1590–1601 (2007).

    Article  Google Scholar 

  67. 67.

    Mazué, G. P. F., Dechaume-Moncharmont, F. X. & Godin, J. G. J. Boldness-exploration behavioral syndrome: Interfamily variability and repeatability of personality traits in the young of the convict cichlid (Amatitlania siquia). Behav. Ecol. 26, 900–908 (2015).

    Article  Google Scholar 

  68. 68.

    Wilson, A. D. M. & Godin, J. G. J. Boldness and behavioral syndromes in the bluegill sunfish, Lepomis macrochirus. Behav. Ecol. 20, 231–237 (2009).

    Article  Google Scholar 

  69. 69.

    Biro, P. A. & Stamps, J. A. Using repeatability to study physiological and behavioural traits: ignore time-related change at your peril. Anim. Behav. 105, 223–230 (2015).

    Article  Google Scholar 

  70. 70.

    Biro, P. A. Do rapid assays predict repeatability in labile (behavioural) traits? Animal Behaviour 83, 1295–1300 (2012).

    Article  Google Scholar 

Download references

Acknowledgements

We are grateful to D. Revers and S. Roundtree for zebrafish husbandry. We thank J. Russ and J. Davila for helpful discussions. This study was supported by funds from the National Institutes of Health (R15MH113074), Nebraska EPSCoR First Award (OIA-1557417), Nebraska Research Initiative, and University of Nebraska Omaha start-up grants and Open Access Funds to RYW.

Author information

Affiliations

Authors

Contributions

R.Y.W. conceived the study. M.R.B. and A.C.G. conducted the behavioral testing and data collection. J.B.S. and M.R.B. performed the statistical analyses. All authors contributed to the writing of the manuscript.

Corresponding author

Correspondence to Ryan Y. Wong.

Ethics declarations

Competing Interests

The authors declare no competing interests.

Additional information

Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Baker, M.R., Goodman, A.C., Santo, J.B. et al. Repeatability and reliability of exploratory behavior in proactive and reactive zebrafish, Danio rerio. Sci Rep 8, 12114 (2018). https://doi.org/10.1038/s41598-018-30630-3

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.