Effects of Competing Messages: A Laboratory Simulation

This paper presents a series of experiments in which subjects were exposed to series of competing messages and were asked to make choices based on the messages they had received. Choice-making proved to be extremely senstive to certain properties of the pre-choice message streams. When some messages were delivered more often than others, choices were almost exactly proportional to each message's share of the message stream, up to a point of diminishing returns. In addition, messages proved most effective when delivered near the choice-point and when they were delivered in "flights" or "bursts."