Computer-based Tools for Rethinking Innovation
The epidemic model of the diffusion of innovations focuses our attention on the relationship between adoption decisions and social interactions, such as happens in word-of-mouth advertising. The epidemic models based on the SI disease model in Chapter 2 all assumed that each individual in some population is able to interact freely with any other individual. Real social interactions - whether formal or informal, held in the home, in the street or at work - are not like this, however. Some social interactions are more likely to occur than others. On the day one of us types this sentence in his office in Guildford, England, there is a good chance he will bump into his co-author later and discuss the work - our offices are situated in the same building. He actually shares an office with some other colleagues, but they are not co-authors, so do not share the same interest in discussions of this book. Hence two apparent reasons for social interactions are geographical co-location or physical proximity and a preference for discussing shared interests. Geography and interests are constraining our interaction possibilities - forcing the network of interaction relations to take on a particular structure. This chapter uses simulation models of diffusion through social interactions to show that the structure of these interactions - the social network architecture - matters a great deal to the outcomes of innovation diffusion.