In the cuing effect, peripheral cuing facilitates recognition of a target stimulus presented at the cued position. It is often assumed that allocation of spatial attention on the cued position by a peripheral precue enhances a quality of representation at the cued position. However, peripheral precue inevitably interacts with a target visually, such as a contrast effect between the precue and the target. Since visual performance is modulated by the flanking stimulus positioned near the target, this kind of interaction could explain the cuing effect without assuming allocation of attention. The purpose of this study is to clarify whether the cuing effect is induced by an attentional process or by a visual interaction. We examined the cuing effect by using different types of peripheral cue in the target discrimination task. In our task, observers reported the orientation of the target Gabor patch presented at peripheral retina. In the peripheral condition, one peripheral precue was presented beside the position of the target. In the peripheral-central condition, two precues were presented sequentially. The first precue was presented beside the position of the target at periphery, then the second precue was presented at the central retina. Since the duration and the position of the first cue was the same between two cuing conditions, similar effect of visual interaction is expected between conditions. If the cuing effect is due to visual interaction, no difference of performance between two conditions is predicted. If the cuing effect is due to an attentional process, the performance of the peripheral-central condition may be less accurate, as a result of being pulled attention away from the target position by the onset of the second precue. We found that discrimination performance was deteriorated in the peripheral-central condition than in the peripheral condition. We conclude that presenting the peripheral cue allocates covert attention in an automatic way.