Artificial visual obstruction, antipredator vigilance, and predator detection in the thirteen-lined ground squirrel (Spermophilus tridecemlineatus)

Antipredator vigilance behaviour is affected by a number of variables, but studies reporting experimental manipulation of visual obstructions are generally lacking in the literature. We examined the individual antipredator behaviour of free-ranging thirteen-lined ground squirrels in 7 city parks in Lincoln, Nebraska, USA as they fed on peanut butter within small plexiglass boxes. Boxes differed in the position and amount of visual occlusion for a squirrel feeding in it: opaque, clear, opaque-sides, and opaque-top (also opaque on the end). We experimentally controlled environmental and contextual variables that have been identified as affecting vigilance behaviour. We filmed 38 ground squirrels as they fed in the boxes, and the results indicate that ground squirrels treat lateral occlusion as functionally identical to complete occlusion. Additionally, there were no significant differences in the squirrels' behaviours in the clear versus the opaque-top box. In 35 approach tests, a person walked toward individual ground squirrels as they foraged in one of the 4 box types. The distance at which the squirrels reacted did not differ among the 4 box types. The results of these two experiments suggest that thirteen-lined ground squirrels compensate for visual occlusion, primarily through increased withdrawals coupled with low-level vigilance postures and that the position of visual occlusion can be of greater importance than the absolute amount of occlusion.