Among other strategies to improve fish welfare in rearing environment, domestication and/or selective breeding was proposed to minimize fish responsiveness to husbandry practices. To verify this hypothesis on a recently domesticated specie, the sea bass, Dicentrarchus labrax, L., an experiment was realized, using four populations differing according to their level of domestication or selection: one population produced from wild parents (Wild), one population produced from parents domesticated for one generation (Domesticated) and two produced from parents selected for growth for one generation (Selected A and Selected B). The experiment was carried out over 91 days with 600 fish (50 fish per tank, 150 fish per population). After a control period, the fish were submitted from day 35 and during 56 days to a stress treatment including frequent and random application of 4 acute stressors (pursuing fish with a net during 1 min, switching off the light for 2 s during the day or, conversely, switching on the light for 2 s during the night, and overflying a bird predator silhouette above the tank during 30 s). The two variables that were measured, i.e.: fish self-feeding behavior and growth performance [at days (D) 14, 35, 63, and 91] were both altered, albeit differentially according to populations, by the stress treatment. During the first stress period (from D35 to D63), all groups modified their feeding rhythm and highly increased their feed intake while their growth rate decreased (Domesticated and both Selected fish groups) or remained stable (Wild). During the second stress period (from D64 to D91) fish continued to modify their feeding rhythm (being more and more diurnal) and increased again their feed intake; conversely to what happened during the first stress period, here, these modifications were associated with an improvement of the growth rate of all populations. During the whole experiment, both Selected groups and Domesticated fish were always characterized by a higher body mass, specific growth rate and body condition factor than Wild fish. In conclusion, and according to the results of this study, a first generation of domestication or selection improved fish growth performance but, at this early stage do not modify behavioral responses to repeated acute stress exposure. (C) 2010 Elsevier B.V. All rights reserved.