This dataset was collected using seabed stereo-baited remote underwater video systems (Stereo-BRUVS). Seabed stereo-BRUVS consist of two GoPro cameras mounted 80 cm apart on a horizontal base bar, each converging at an angle of four degrees to a common focal point. A galvanised steel mesh bait cage containing ~1 kg of crushed pilchards Sardinops sagax is attached to the end of a 1.5-m-long bait arm. Seabed stereo-BRUVS are deployed at least 200 m apart for a minimum of 60 minutes. 

All videos were analysed using Eventmeasure software. Video processing commenced once seabed stereo-BRUVS had settled on the seabed, for a period of 60 minutes. All animals entering the field of view were identified to the lowest possible taxonomic level, and abundance was estimated using the conservative abundance metric MaxN, which is the maximum number of individuals of a given taxon in a single frame (Cappo et al. 2006). Fork length was measured in stereo with individuals measured where they were well positioned relative to the camera and not occluded by other individuals. An Excel database was then created containing the identifications and measurements (where possible) of all animals. 

All observations of individuals from the family Lethrinidae were extracted from this database. These observations were reanalysed using EventMeasure to determine frequency of different colour phases (number of individuals currently in each colour phase), physiological colour change (individuals actively displaying physiological colour change while in frame) and the association of physiological colour change with feeding behaviour (individuals displaying colour change while feeding or approaching to feed). Pictures taken from the videos were subsequently processed using Adobe Photoshop 22.2 adjusting brightness, saturation and contrast. Distinction was made between two different colour phases, light (silver/pale) and any form of dark colour (partly or fully dark). No quantitative differentiation of patterns among dark colour phases was made, however, different patterns were qualitatively described. Individuals displaying any form of dark colour phase were counted at the moment of MaxN (maximum number of individuals observed in the frame at one time) or the highest cumulative MaxN moment in which physiological colour change occurred or dark colour phases were present to avoid multiple counts of the same individuals. Individuals that showed active physiological colour change were also marked during dark phases.