In this study on interstimulus contingencies the probability of presenting a target stimulus at a certain location was varied in dependency of the occurrence of an accessory stimulus at the same or at the opposite location. A focussed attention paradigm was used: Subjects were asked to make an eye movement as quickly and accurately as possible to a visual target and to use an accessory auditory stimulus only as a possible cue for the location of the visual target. The auditory stimulus was presented in the same or in the opposite hemisphere at the same horizontal eccentricity (15 deg) as the visual target via a virtual auditory display. The probability that the auditory stimulus occurred in the same hemisphere as the visual target was 20, 50 or 80%, resp., in different blocks of trials. In addition, the visual and the auditory stimulus were presented with varying SOA's. Trials with visual targets only were interleaved in the bimodal trials, whereas trials with auditory targets only were measured in separate blocks in the first and last session of the experiment. Saccadic reaction times in the bimodal trials showed a facilitative effect in the 80% condition in comparison to the 20 or 50% condition. This effect was especially apparent when the auditory stimulus was presented before the visual target. The results will be discussed in the context of a probability distribution inequality (race inequality) by Miller (1982) which builds a boundary condition between race models and models of coactivation.