Early behaviour can determine food intake and growth rate with important consequences for life history and survival in fishes. Egg size is known to affect growth rate of young Arctic charr but its influence on the development of behaviour is poorly documented. It is believed that egg size influence on growth and potentially on the behaviour of young fish decreases over time, minimized by the effects of social factors. Shortly after first feeding, we examined differences in mobility and foraging of Arctic charr in relation to egg size and social environment. The behaviour of juveniles from small and large eggs was compared five times over the course of development and in three different experimental settings: long‐term isolation (isolation before hatching), short‐term isolation vs. group rearing and mixed size group vs. homogeneous size groups. Egg size affected foraging behaviour and mobility of fish: fish coming from large eggs were more mobile and foraged more than fish coming from small eggs. Social environment affected foraging behaviour, mobility and space use: fish in a group were more mobile, foraged more and responded faster to food delivery than isolated fish. The interaction of egg size and social effects was seen primarily in foraging activities but did not affect mobility or space use. Large fish in groups foraged more than the three other groups: large fish in isolation, small fish in groups and small fish in isolation. Agonistic behaviour was rarely observed and there was no significant effect of group composition on agonistic behaviour. We discuss the importance of egg size and social effects at early stages of development with a focus on the evolutionary ecology of Arctic charr.

Most studies carried out with seabass under self-feeding conditions report an intriguing social structure that is built around the device and the food dispenser with three coexisting triggering categories: hightriggering (HT), low-triggering (LT) and zero-triggering (ZT) fish. However, neither sex nor feeding motivation or hierarchy can explain the establishment of this specialization. We characterised the personality of seabass with the commonly used restraint and open field tests and assessed the link between personality traits and individual triggering activity towards the self-feeder apparatus. We found no differences between triggering categories during the restraint test but high triggering fish were characterised as shyer than low-and zero-triggering fish during the open field test. Triggering activity was negatively correlated with exploratory capacities and boldness. This experiment provides for the first time evidence that high triggering status in seabass is correlated with personality traits, which could partly explain the social structure that builds around a self-feeder device. Highlights►European seabass personality (i.e. bold-shy and motivation to escape stressful situation) was characterized. ►Latency to emerge from a shelter and latency to escape during a restraint test were correlated. ►Placed under self-feeding, individual triggering activity level was higher in shy individuals.

To understand the genetic basis of coping style in European seabass, fish from a full factorial mating (10 females x 50 males) were reared in common garden and individually tagged. Individuals coping style was characterized through behavior tests at four different ages, categorizing fish into proactive or reactive: a hypoxia avoidance test (at 255 days post hatching, dph) and 3 risk-taking tests (at 276, 286 and 304 dph). We observed significant heritability of the coping style, higher for the average of risk-taking scores (h2 = 0.45 ± 0.14) than for the hypoxia avoidance test (h2 = 0.19 ± 0.10). The genetic correlations between the three risk-taking scores were very high (rA = 0.96–0.99) showing that although their repeatability was moderately high (rP = 0.64–0.72), successive risk-taking tests evaluated the same genetic variation. A mild genetic correlation between the results of the hypoxia avoidance test and the average of risk-taking scores (0.45 ± 0.27) suggested that hypoxia avoidance and risk-taking tests do not address exactly the same behavioral and physiological responses. Genetic correlations between weight and risk taking traits showed negative values whatever the test used in our population i.e. reactive individual weights were larger. The results of this quantitative genetic analysis suggest a potential for the development of selection programs based on coping styles that could increase seabass welfare without altering growth performances. Overall, it also contributes to a better understanding of the origin and the significance of individual behavioral differences.