We examined the potential for a trait-mediated indirect interaction (TMII), whereby a behavioural response to a predator (crab Cancer borealis) alters the per capita effect of a reacting species (sea urchin Strongylocentrotus droebachiensis) on its algal food (kelp Saccharina latissima), in field and laboratory experiments. We used timelapse video to monitor sea urchin movement in response to either an attractant food cue (kelp) or a repellant predation-risk cue (crabs) placed in small cages at the centre of 7m2 circular plots in a wave-exposed sea urchin barrens off Nova Scotia. We found that the presence of crabs in the cage did not reduce sea urchin grazing rates on kelp compared to a control without crabs. We observed similar results using a similar experimental design in large tanks (900 L) under controlled flow conditions in the laboratory. We conclude that C. borealis does not significantly inhibit the foraging activity of S. droebachiensis under turbulent flow conditions in nature, and that a TMII is unlikely to contribute to a trophic cascade in the shallow rocky subtidal zone of the Northwest Atlantic.