Several pelagic fish species are known to associate with floating objects. However, quantitative information on the main factors that drive this associative behaviour is still lacking. Small pelagic fish offer a particularly interesting case study for this phenomenon, since the small spatial scales involved in their association with floating objects allow experimental data to be collected at a fine scale. Here, we monitored twelve acoustically-tagged bigeye scads, Selar crumenophthalmus (Carangidae), (mean fork length 16.4 cm, SD 2.1 cm) around a floating object moored in shallow water (15 m depth) playing the role of a coastal fish aggregating device (FAD). To quantify the role played by variations in current and daylight, we calculated the speed distribution, pair correlation function and group polarization for the tagged fish hourly, from midday to dusk (13:00–18:00), for varying current strengths and daylight conditions. We found that the current induced a shift in the position of the aggregation upstream of the FAD, at distances that increased with the current strength. We found evidence of an expansion and a higher coordination in the aggregation at dusk, with increasing speed, distance among conspecifics and alignment. We discuss possible scenarios in which group polarization increases at dusk and suggest complementary measurements for future experiments that could confirm our findings.

We monitored temporal changes in the magnitude of altered host behaviour in minnows (Pimephales promelas) experimentally infected with metacercariae of a brain-encysting trematode (Ornithodiplostomum ptychocheilus). This parasite develops and then encysts in a region of the brain that mediates the optomotor response (OMR), an innate behaviour that links visual stimuli with motor performance. The OMR of infected and uninfected minnows was evaluated between 0 and 10 weeks post-infection (p.i.), an interval spanning the development period of metacercariae to infectivity in birds. Trials involved monitoring the time an individual minnow spent following a spinning drum that had been painted with alternating black and white stripes. At 2 and 4 weeks p.i., infected minnows followed the drum 40% less often than controls. Differences between controls and infected fish declined thereafter, and were undetectable by 10 weeks p.i. Both control and infected fish habituated equally rapidly to the spinning drum. However, the difference in performance between controls and infected fish was 29% for experienced fish and 48% for fish that had never experienced the drum. Because maximum parasite-induced reduction in OMR coincided with the period of maximum parasite development, the behavioural effects are most likely due to unavoidable pathology in the brain associated with developing larvae.