A foraging trial was conducted 1 h after the shoaling trial ended
following the method described by Nassef et al. (2010) with slight
modification. To measure the feeding rate, fifteen fish from each
treatment and control group were individually transferred to an
aquarium (30 cm high 30 cm wide 50 cm long) containing
the exposure concentration of sertraline for treated fish, or no sertraline
for control fish. An incomplete vertical glass partition was
placed in the test aquarium, leaving a narrow space for introducing
the fish (Fig. S2). All the sides of the tanks, except the observational
one, were covered with black curtains in order to avoid disturbance
to the fish. Prior to placing the fish in the test box, 10 similarly
sized (10–15 mm) midge larvae were placed in the bottom on
the left side. The recording process commenced when the fish
passed the vertical partition and moved toward the midge larvae.
Each trial was digitally recorded using a video camera for 10 min
in each trial. If the fish did not move toward the midge larvae
within 5 min of starting the recording, the glass partition was
removed. The latencies for each fish to capture the first, the 5th,
and the last (10th) midge larvae were observed. After the initial
observation period of 10 min, an additional 10 midge larvae were
offered and the number of midge larvae remaining after 20 min
was recorded. Feeding behavior was thus quantified both in terms
of feeding rate and food consumption