Changes in animal movement (frequency or speed of locomotion) following exposure to a toxicant are frequently considered a biomarker of contaminant exposure and are some of the most widely reported behavioral results in toxicological literature. However, the ecological consequences of such behavioral changes, such as effects on toxicant transfer in foodwebs, are far less well understood, complicated in part by the short-term nature of laboratory experiments and the lack of complementary field studies where the nature of toxicant exposure is more complex. Here we examine whether naturally exposed individuals of the round goby, a benthic, site-loyal fish, move in a manner similar to conspecifics from less contaminated habitats. In the laboratory, round goby from a relatively cleaner site showed greater activity and exploration than goby from two highly contaminated sites. Male fish were more active than females but the site effects were similar in both sexes. In contrast to laboratory findings, a field mark-recapture study of 881 round goby showed that fish from the cleaner site did not move greater distances or exhibit shorter residence times within the site than round goby from highly contaminated sites. Our results indicate that while behavioral changes in the laboratory may be one of several useful diagnostics of toxicant exposure of wild-exposed animals, they do not necessarily translate readily into measurable differences in a natural context. Thus, the potential fitness consequences of toxicant exposure based on behavioral changes need to be assessed carefully.