There is an increasing recognition within the aquaculture industry that understanding the behaviours of farmed animals can improve feeding. In the production of Pacific white shrimp (Litopenaeus vannamei), inefficiencies in feeding activity often result in high feeding costs due both to slow feeding habits and high costs of fish meal as the main protein source in commercial feeds. Particular substances in feed stimulate different phases of feeding, meaning commercial diets must include these chemical cues that are recognised by the shrimp. A wide range of feed attractants have been developed for shrimp, but little is understood about their effects on shrimp behaviour. More knowledge of shrimp behaviours in response to different attractants will enable effective measures for assessing feed attraction and subsequent feed uptake and has been the focus of my Ph.D. funded by Skretting.
Studies of shrimp feeding behaviour have generally used direct observations or video recording followed by observation, with the latter minimizing disturbances to shrimp. However, as manual scoring of behaviour can be time-consuming, automated observations (such as video tracking) may be a quicker and more feasible way to analyse behaviour. For that reason, the use of Ethovision XT to measure L. vannamei feeding behaviour was investigated in response to different commercial feeds. Automated observations made by this tracking software were compared to manual scoring and a high agreement was found between the two methods. Thus, Ethovision XT offers a more suitable approach than manual observations for measuring L. vannamei feeding behaviour. There were several advantages provided by the software, such as an increase in the speed of analysis, removal of observer subjectivity and the possibility of analysing additional behavioural information (for example velocity, distance travelled).
Ethovision XT was able to distinguish between behavioural responses of shrimp to different commercial feeds and allowed us to associate the occurrence of certain behaviours with the level of diet attraction. As a result, a protocol has been developed to quantify and rank the attractiveness of feed to shrimp using Ethovision XT. Further research will be carried out to implement this automated protocol for testing shrimp feed attractants under commercial conditions.