Development of observational learning during school formation in jack mackerel Trachurus japonicus juveniles

Interspecific differences in developmental mode determine early cognitive abilities in teleost fish

Most studies on developmental variation in cognition have suggested that individuals are born with reduced or absent cognitive abilities, and thereafter, cognitive performance increases with age during early development. However, these studies have been mainly performed in altricial species, such as humans, in which offspring are extremely immature at birth. In this work, we tested the hypothesis that species with other developmental modes might show different patterns of cognitive development. To this end, we analysed inhibitory control performance in two teleost species with different developmental modes, the zebrafish Danio rerio and the guppy Poecilia reticulata, exploiting a simple paradigm based on spontaneous behaviour and therefore applicable to subjects of different ages. Zebrafish hatch as larvae 3 days after fertilisation, and have an immature nervous system, a situation that mirrors extreme altriciality. We found that at the early stages of development, zebrafish displayed no evidence of inhibitory control, which only begun to emerge after one month of life. Conversely, guppies, which are born after approximately one month of gestation as fully developed and independent individuals, solved the inhibitory control task since their first days of life, although performance increased with sexual maturation. Our study suggests that the typical progression described during early ontogeny in humans and other species might not be the only developmental trend for animals' cognition and that a species' developmental mode might determine variation in cognition across subjects of different age.

A mixed-species shoal including fewer heterospecifics is preferred by jack mackerel Trachurus japonicus juveniles

Mixed-species shoals of fishes were reported from various environments such as coral reefs and temperate lakes. The driving mechanism of forming mixed-species shoal can be either haphazard or a preferential choice by an individual. Here, we tested if jack mackerel Trachurus japonicus, a temperate marine pelagic fish, could choose a shoal composition using vision based on a choice experiment. Shoal preference of jack mackerel juveniles was evaluated by presenting shoals comprising one to four heterospecifics (half-lined cardinal Ostorhinchus semilineatus) out of four shoal members to the focal fish. When both conspecific and heterospecific shoals were presented to a single jack mackerel, the focal fish showed a preference for the conspecific rather than heterospecific shoal, indicating that they can choose a conspecific shoal based on the visual information. However, when a focal fish encountered both conspecific and mixed-species shoals including fewer heterospecifics (one out of four individuals), the latter was preferred. The result implies that jack mackerel can judge a preferable composition of shoal using visual information. The inclusion of fewer heterospecifics may be exploiting oddity effects, whereby minority species are exposed to relatively high vulnerability to predators, as well as resulting in reduced competition for food while maintaining predator vigilance in a shoal.

Observational learning and irreversible starvation in first-feeding zebrafish larvae: is it okay to copy from your friends?

Starvation is one cause of high mortality during the early life stages of many fish species. If larvae do not learn to feed, or if no food is available during early stages, irreversible starvation occurs and larvae reach the Point of No Return (PNR), the developmental period/age when they will not feed even if food is available. Fish larvae may learn to how to feed by observing conspecifics or through personal/individual experience with prey items that are encountered. We examined food acquisition in first-feeding zebrafish larvae to determine the impact of delayed feeding and identify the time of irreversible starvation and the PNR. Next, we examined how feeding ability, and the PNR, is altered by either observational learning or previous experience, to determine which paradigm facilitates successful feeding.Our data indicate that zebrafish larvae learn to feed, with the PNR at 7-8 days postfertilization (dpf). Exposure to prey items immediately after hatching (3-5 dpf) results in the highest survival rates. Zebrafish larvae learning to feed by observing conspecifics also had high survival, though the PNR was not changed. In contrast, previous experience with prey items caused an earlier PNR and reduced survival. Overall, these results that indicate feeding is a learned behavior in zebrafish larvae and interacting with/observing conspecifics during the early larval period is a better predictor of feeding ability than previous experience with food.