In France, many pressures threaten diadromous species and lead to their decline, some being even in danger of extinction. Dampening pressures on these migratory species is thus of great concern for their conservation. To restore the aquatic continuity near dams, fishways and fish ladders are built to allow fish passage. However, these sites constitute bottlenecks frequented by lots of catfish Silurus glanis. Catfish is an alien species that, due to its very big size, is the only one able to predate these adult migratory species in rivers. On several rivers, catfish has been shown to predate up to 80% of migrators when they were going up the river to spawn, which considerably limits the efficiency of fishways and fish ladders. Among fishes, catfish hears excellently; typically, recreational fishermen hit the water surface with a specific wooden tool (« clonk ») to attract catfish. However, in intensive fishing areas, it has been reported that, on the contrary, the clonk triggered an escape response, highlighting the great learning abilities of this species. Our work aims to test the possibility to control catfish behaviour by phonotaxis and to apply it to its local management. At first, different kinds of sounds (eg congener vocalizations, clonk) will be collected in an experimental natural lake and in other ecosystems thanks to citizen science. Once it will have been checked that they specifically target catfish, the most attractive of them will be used to control catfish behaviour. This consists in attracting catfish with these stimuli, and then generate an unpleasant experience (a light electric shock for example), so that catfish learns to associate the attractive acoustic stimuli to a noxious experience; catfish should then escape just by hearing the initially attractive stimuli. Operant conditioning and its efficiency will be assessed in an experimental natural lake where catfish individuals are continuously tracked. Once the method will be well tuned, it will be tested in real conditions near a fishway in presence of diadromous species. The method will be refined to become a management tool, targeting specifically catfish and limiting its predation pressure on endangered migratory species at sites with high stakes.