Among-individual behavioural variation in the ornamental red cherry shrimp, Neocaridina heteropoda

Abstract

Personality variation, defined as among-individual differences in behaviour that are repeatable across time and context, is widely reported across animal taxa. From an evolutionary perspective, characterising the amount and structure of this variation is useful since differences among individuals are the raw material for adaptive behavioural evolution. However, behavioural variation among individuals also has implications for more applied areas of evolution and ecology-from invasion biology to ecotoxicology and selective breeding in captive systems. Here, we investigate the structure of personality variation in the red cherry shrimp, Neocaridina heteropoda, a popular ornamental species that is readily kept and bred under laboratory conditions and is emerging as a decapod crustacean model across these fields, but for which basic biological, ecological and behavioural data are limited. Using two assays and a repeated measures approach, we quantify behaviours putatively indicative of shy-bold variation and test for sexual dimorphism and/or size-dependent behaviours (as predicted by some state-dependent models of personality). We find moderate-to-high behavioural repeatabilities in most traits. Although strong individual-level correlations across behaviours are consistent with a major personality axis underlying these observed traits, the multivariate structure of personality variation does not fully match a priori expectations of a shy-bold axis. This may reflect our ecological naivety with respect to what really constitutes bolder, more risk-prone, behaviour in this species. We find no evidence for sexual dimorphism and only weak support for size-dependent behaviour. Our study contributes to the growing literature describing behavioural variation in aquatic invertebrates. Furthermore, it lays a foundation for further studies harnessing the potential of this emerging model system. In particular, this existing behavioural variation could be functionally linked to life-history traits and invasive success and serve as a target of artificial selection or bioassays. It thus holds significant promise in applied research across ecotoxicology, aquaculture and invasion biology.

Keywords: animal personality; behavioural syndrome; behavioural variation; boldness; cherry shrimp; decapod.

FIGURE 1

The tank setups used for (a) Open field trials (OFT) and (b) Food and shelter trials (FST) showing a side view on the left and an overhead view on the right. The starting set up for both assays has the shrimp to be tested placed inside the black acclimation tube.

FIGURE 2

Repeatabilities for OFT and FST traits estimated from initial univariate models (blue points) and refitted models that condition on additional fixed effects of size and sex (red points). Error bars depict estimates ±1SE.

FIGURE 3

Among‐individual correlation structure between traits as determined from multivariate model estimates of (a) ID the variance–covariance matrix and (b) ID _sex:mass , the corresponding matrix conditional on sex and size. Ellipse shape and colour denote the strength and sign of each correlation. Light grey background shading indicates sets of correlations among traits measured within each assay type (FST, OFT), while across‐assay correlations have white backgrounds.

FIGURE 4

Trait loadings on the leading vectors of ID (blue points) and ID _sex:mass (red points). Error bars denote 95% confidence intervals obtained using a parametric bootstrap (n = 1000) and loadings can be considered nominally significant if CI do not cross zero (solid vertical line).

FIGURE 5

A graphical illustration of the relationship between multivariate personality, mass, and sex. Each point represents an individual's predicted behavioural score id _max plotted against mean weight, with colours denoting assigned sex (female, male, unknown). Behavioural scores are calculated for each individual as id _max .i ^T where i is a column vector containing the best linear unbiased predictions (BLUPs) of individual deviations from each trait mean. Also shown for illustrative purpose are overall (grey dashed) and sex‐specific regressions of behavioural score on mean weight. Behavioural scores are (uncertain) model predictions and no statistical inference is intended.