An experiment was conducted to investigate whether polyphagous predators dispersing from a within field overwintering refuge (‘beetle bank’) could reduce cereal aphid numbers in an adjacent crop of winter wheat. To assess the impact of predators emigrating from the beetle bank on populations of Sitobion avenae in the crop, polyphagous predator densities were manipulated using exclusion barriers placed at set distances away from the beetle bank. Results from the pitfall trap data indicated that the exclusion barriers were effective in significantly reducing the number of ‘boundary’ Carabidae, ‘open-field’ Carabidae, Staphylinidae, Linyphiidae and Lycosidae in the enclosed areas compared to the control areas by 56, 31, 46, 24 and 57%, respectively. Aphid numbers were significantly greater in the enclosures, where they were 34% higher compared to the control areas during the aphid peak. The impact of polyphagous predation decreased with increasing distance away from the beetle bank and the distribution of polyphagous predators in relation to the beetle bank varied over time. The results are discussed in terms of the peak activity periods of the different predatory groups and the consequences for integrated crop management (ICM).