Behavioral trials were conducted in the same order and not randomized. First, we examined behavior upon introduction to a new environment, as a proxy for stress (Chin et al., 2018 (link)). Second, we assayed aggression in response to a mirror. These first two trials were conducted on the same day in 19 L tanks separated by opaque dividers to prevent fish from seeing conspecifics during experiments. All trials were recorded continuously using one Wyze Cam v2 for each tank.
Next, the fish were acclimated to circular arenas for 70 h to test for wall-following behavior over a 1 h window. The fish were then returned to their 19 L experimental tanks, starved for three days, and tested for food consumption. After all behavioral measurements, fish were weighed, and photographs were taken to measure neuromast number, melanophore number, standard length, head depth, and eye diameter. All trials were conducted with tanks at room temperature (20–22 °C).
All video recordings of the trials were analyzed using EthoVision XT 15 (Noldus) behavioral tracking software to track subjects and calculate variables of interest. To ensure accurate tracking, each recording was reviewed manually, with erroneous data removed and subsequently interpolated using EthoVision XT 15 (Noldus). Fish videos with greater than 20% missing data were removed from their respective datasets prior to statistical analysis.
Next, the fish were acclimated to circular arenas for 70 h to test for wall-following behavior over a 1 h window. The fish were then returned to their 19 L experimental tanks, starved for three days, and tested for food consumption. After all behavioral measurements, fish were weighed, and photographs were taken to measure neuromast number, melanophore number, standard length, head depth, and eye diameter. All trials were conducted with tanks at room temperature (20–22 °C).
All video recordings of the trials were analyzed using EthoVision XT 15 (Noldus) behavioral tracking software to track subjects and calculate variables of interest. To ensure accurate tracking, each recording was reviewed manually, with erroneous data removed and subsequently interpolated using EthoVision XT 15 (Noldus). Fish videos with greater than 20% missing data were removed from their respective datasets prior to statistical analysis.
