Refinement and objectivity in magnetoreception research (#209)
It has been argued that animal behaviourists should not shy away from the animal experimentation debate, and actively seek to adopt the 3R’s principles of animal research. Also, the adoption of research methods which eliminate confirmation bias, such as by designing experiments where observers are blind to the treatments or taking objective measures of behaviour, has been proposed as important for the understanding of animal behaviour. Thirty-two day-old domestic chicks were fed for 7 days from a dish placed above either a copper wire coil that produced an experimental magnetic field (magnetic coil), or a dish placed above a coil for which the wires were run in both directions, thereby producing no magnetic field (Control coil). From day 8 to 12, hungry chicks were placed in a square arena without food once a day, with a magnetic coil in one corner and a control coil in the opposite corner. A camera positioned above the centre of the arena recorded the chicks’ positions. The location of the centroid of each chick was automatically recorded using a statistical threshold technique to separate the light coloured chicks from the dark background. Every 5th frame (i.e. every 0.2 seconds) was processed to yield 1500 centroid locations for each chick for each day. The centroid locations were transformed to actual distances from the centre of the magnetic coil, following correction for the geometric distortion introduced by the lens. Chicks previously fed above a magnetic coil spent significantly more time within 12.5cm of the magnetic coil’s centre than chicks previously fed above the control coil (P=0.016). This study presents an automated protocol for recording spatial distribution of a bird in response to a magnetic stimulus using a domesticated species that is readily available and easy to handle and house.