Using optical illusions in the shoulder of a cycle path to affect lateral position

•Effects of different virtual objects in the shoulder on bicyclists’ lateral position were evaluated.•The illusions were not effective in changing cycling behaviour.•If objects were noticed, cyclists rode closer to them. An important factor in single-sided accidents of older cyclists is that they ride off the cycle path onto the verge. Two experiments were performed to assess the feasibility of using virtual 3D objects in the verge to affect the lateral position of bicyclists. In the first experiment, different virtual objects were placed in the shoulder and 1150 passing bicyclists were observed using a fixed camera. The (standard deviation of the) lateral position and speed in four conditions with virtual objects differing in colour, structure, or 3D effect were compared with a control condition in which no virtual objects were applied. In a second experiment, the behaviour of 32 bicyclists aged 50years or older was measured by mounting two digital action cameras with GPS on the participants’ bicycles. The participants cycled a route of approximately 12km in which several locations were passed, one of these contained 15 virtual objects similar to the ones used in the first experiment placed in the shoulder of the cycle path. Cyclist behaviour was compared with behaviour at a control location consisting of a solitary two-way cycle path with a grass shoulder. Results indicate that the virtual objects in the tested format had little overall effect on cyclists’ behaviour. However, bicyclists were positioned closer to the virtual objects and the shoulder when they looked at the objects or when they reported that they saw them while cycling. This suggests that the overall visibility of the object design may have been too conservative.