The effect of stopping before turning on the direct observational measure of whole body turning bias.