ICBTT2002The 1st International Conference on Business & Technology Transfer Top page in Japanese
Top Rationale Key words Topics Day's program Place Committee Sponsor & supports Program
Lecture Registration form Call for papers E-mail photos 1.Banquet 2.Technical Tour
  ICBTT2004 Technology & Society Division, JSME

Diffusion of Innovations and Their Forming Process within a Social System


This study aims to investigate the diffusion process of innovations into a social system. Innovations change their aspects while they diffuse into it. Various social groups interact with each other to promote or prevent the diffusion of innovations, using various communications channels. In addition, due to the rapid progress of communications technologies on the whole, communications channels themselves are said to become innumerable, which affect the diffusion process of innovations that should be formed while they diffuse into the social system. Firstly, this paper develops the diffusion process of an innovation that is formed while it diffuses within a social system, mentioning the interaction between various social groups. Secondly the author reconstructs the innovation diffusion process, and finally discusses the relationship between the critical mass and dominant design.

Key Words:Innovation Diffusion Process, Social System, Critical Mass, Dominant Design

1. Introduction
How does a technological innovation diffuse into a social system? In many cases even now a linear process model are applied in order to explain it; that is, some sort of an institute invents an innovation, which is handed to the manufacturing section, where an artifact or so-called a product is designed and produced. After that the artifact diffuses into the social system, with the rate of adoption going over the critical mass, sometimes through the assistance of change agents or marketing people. However, due to the rapid increase and the diversity of innovations, the remarkable progress of communications technology and various other reasons, the linear process model does not necessarily match with the reality. Among others with shedding light on the innovation process more fully, the linear process model seems to be looked at critically. Especially in this model, the fact that artifacts, which embody an innovation, change their aspects while it diffuses into the social system is not expressed explicitly, but it is treated as a postulate and both the innovation and artifacts does not change so much as affecting the innovation diffusion process.
This study investigates the diffusion process of innovations into a social system, where innovations change their aspects while they diffuse into it. Various social groups interact with each other to promote or prevent the diffusion of innovations, using various communications channels. Firstly, this paper proposes the diffusion process of a technological innovation that is formed while it diffuses within a social system, taking into consideration the interactions between various social groups. Secondly the author reconstructs the innovation diffusion process, and finally discusses the relationship between the critical mass and dominant design.

2. Developing the innovation diffusion model
The technological innovation is in many cases embodied by artifacts, which are either products in terms of manufacturing or commodities in the market place, usually not altering their forms easily. Rogers(1) describes the diffusion of an innovation as the "process by which an innovation is communicated through certain channels over time among the members of a social system." It is composed of four main elements; an innovation, communications channels, a social system and time. An innovation is an idea perceived as new by the members of the social system, which is defined as "a set of interrelated units that are engaged in joint problem-solving to accomplish a common goal."(2) Communications channels are usually divided into two categories; mass communications channel and interpersonal communications channel or more generically an external influence factor and an internal influence factor.
According to the linear process model at least explicitly, an innovation, a social system and communications channels are all given elements. That is, these elements do not change while innovation diffusion process occurs. In actuality, they are a function of time, especially in the age of the rapid increase and the diversity of innovations as well as the wide spread of various communications channels. Therefore, this paper assumes that the diffusion process of a technological innovation is formed while it diffuses within a social system, taking into consideration the interactions between various social groups.
Based on the Bass(3) model, this paper assumes that,

  1. Potential innovation's adopters decide to adopt or reject it all the time, being affected by the internal and external influence factors.
  2. Innovation's providers try not only to make potential adopters adopt it for the increase of the diffusion rate, but also to enhance its efficiency and/or effect.
Then, an innovation diffusion model that would consider the affect of a social system could be as follows,


x: number of innovation adopters by the member of a social system,
n: (discrete) elapsed time, and,
K: population of a social system.

The left side of the equation represents the number of innovation adopters at time n+1. The first member of the right side of the equation is the number of innovation adopters at time n. The second one is new innovation adopters caused by internal influence factors a. The third one is new innovation adopters caused by external influence factors b. And the fourth one is the discontinuance number of the innovation between time n and n+1.
Coefficients a, b, and c are parameters, fluctuating with the passing of time based on the change in the innovation itself, population size, number of adopters and so on. Not only do they fluctuate, but also the innovation providers can operate these parameters, with enhancing the efficiency and effect of the innovation, lowering the price and so on.

3.Interpreting the innovation diffusion model
An innovation is formed while it diffuses into the social system. The author proposes the innovation diffusion process as follows from the point of the self-organizing view, based on the case study of Japanese word processors in Japan around 1980(4). That is,

  1. An innovation appears that overthrows the existing technological paradigm.
  2. After the appearance of an innovation, engineers or relevant professionals witnessing the innovation conduct various kinds of trial and error to compete with each other.
  3. Under passing through the turmoil period, the social system in which an innovation has been implemented becomes changing its structure to self-organize.
  4. Once an innovation has surpassed the irreversible phase, obtaining the dominant design, it begins diffusing into the social system and is finally embedded in it.

Self-organizing systems are, as Krugman(5) put it, these "systems that, even when they start from an almost homogeneous or almost random state, spontaneously form large-scale patterns."
In accordance with the concept shown by Rogers that the innovation diffusion process is composed of four elements (an innovation, a social system, communications channels, and time), the author interprets the innovation diffusion process in terms of the self-organizing system. Firstly, members of the social system is divided into both actual and potential innovation adopters, innovation providers, policy makers in central and local government, and other relevant social groups. The population sizes as well as members of the social system undergo a change, as the innovation diffuses into the social system. An innovation and artifacts that embody it alter their aspects in the process of the innovation diffusion; especially, artifacts sometimes undergo the remarkable transfiguration. The usage of communications channels at one time differs from an innovation diffusion phase at another. Especially in case of the diffusion process of communications technology, communications channels themselves come to alter their features.
Figure 1 shows the innovation diffusion process, taking into considerations such relationships. When an innovation appears that overthrows the existing technological paradigm and members of the social system pay attention to the artifact that embodies the innovation, they interpret its meaning in various ways. Even the innovation providers have a scarce idea whether the potential adopters would accept it or not. At this time such professionals, firms or other organizations as have a potential to become innovation providers may examine if they should develop and produce the similar product (=artifact) against the first innovation provider. Policy makers would try to make the innovation diffusion foster if they would infer its spread into the social system is worthwhile, whilst if not they might try to suppress its diffusion. Relevant social groups on which the innovation seems to have a significant influence also interpret it in various ways and may try to urge its development toward the direction favorable to them.

Fig. 1 A chart of the innovation-diffusion process.

If the innovation diffusion process entered in the irreversible one after passing through the turmoil period, it would become diffusing into the social system successfully. In order to be said to diffuse successfully, it is commonly necessary that not only the dominant technology appears corresponding with the innovation, but also the adoption rate of the innovation exceeds the critical mass where the sustainable diffusion process starts among the members of the social system. The social system has spontaneously formed large-scale patterns in accordance with the innovation from the viewpoint of the self-organizing system. Next, the author would examine the relationship between the dominant design and critical mass, considering the formation process of large-scale patterns.

4. Forming the dominant technology and attaining to the critical mass
Before talking about the dominant design and critical mass, let the author confirm their definition. On the one hand, the critical mass "occurs at the point at which enough individuals have adopted an innovation so that the innovation's further rate of adoption becomes self-sustaining.(6)" On the other hand, as Utterback(7) put it, a dominant design "in a product class, the one that competitors and innovators must adhere to if they hope to command significant market following." In addition, the dominant design is not necessarily the optimum one, but is some sort of a satisfied solution and results in a specific artifact, emerging from the technological possibility and the selection process by a market.
Thereupon, the critical mass in the innovation diffusion process is the irreversible point at which it is hardly practicable to return to the previous situation as well as large-scale patterns are formed in terms of the social system. What the diffusion process passes through the irreversible point is such conditions that the social system is structurized around the innovation, namely, the innovation is embedded in the social system and recognized widely as indispensable. In the end, members of the social system cannot live their active lives effectively and efficiently without it. Whereas, the appearance of the dominant design indicates that various interpretations on the innovation converge at some specific design in terms of technological possibilities and the members' propensity. Ultimately, hardly other design than the dominant one can survive any more.
In this sense, the critical mass is of the social system's perspective, while the dominant technology is of the technological perspective. If so, how does the social system achieve at the critical mass point? How does the dominant design appear in the social system? Firstly, as shown in Fig. 1, various members of a social system communicate with each other and are exposed to the influence of mass communication, interpreting an artifact that embodies the innovation. There are roughly two models. The one is the contagion model, in which an innovation spreads from mouth to mouth, as influenza is infectious. Many scholars analyze the innovation diffusion process based on this model. The other is structural equivalence model, in which whether structurally equivalent members in the social system introduce the innovation or not plays an significant role when a member adopt it. (8)
As is well known, communications technology such as Internet has made a rapid progress, so that tremendously various communications channels exist now compared with decades ago. Therefore, these two propagation processes take place simultaneously even though it depends on the type of an innovation.
Other potential innovation providers also interpret the artifact that embodies an innovation, when it first comes onto the market. Observing the adopters' response, and investigating the new product, they may put a new product on the market, and enter rivalry with the first innovation provider. When many members of the social system take much notice of the innovation and some of them begin adopting it, diverse varieties of the innovation get to appear. As time goes by, the more an artifact that gains an advantage over others obtains much potential adopters, the more its design must be refined. Partly because the bandwagon effect sometimes arises, the design of the innovation converges specifically, that becomes the dominant design. For example, Arthur(9) shows that even when there are two technologies and it is not known ex ante which of them are superior, the one technology becomes predominant while the other one ceases to exist, based on the stochastic theory. Moreover, he certifies that when one technology accidentally gets to be superior to the other, the additional investment makes it more advantageous so that the lock-in mechanism to one technology finally takes place. Meanwhile, Bak(10) shows that when the situation has changed to one technology's advantage, the innovation get to converge promptly on it, citing the self-organized criticality theory.
Finally, what is the relationship between the critical mass and the dominant design? In the diffusion theory based on the linear process model, after the appearance and settlement of an innovation and an artifact that embodies it, the innovation does not change through the diffusion process. In other words, the dominant design exists when the diffusion process starts. For example, when an early innovation diffusion research was conducted around 1950, concerned with hybrid corn among farmers in the US(11), it was assumed that the innovation and its artifact, hybrid corn, does not change through the diffusion process. The same thing can be said of the famous diffusion study on the medicine among medical specialists. (12)
Namely, these studies could be conducted under the "ideal" condition. In actuality, the innovation and its artifacts change on occasion, especially when information and communications technology has made a rapid progress and diffused widely. In other words, the innovation itself is formed while itdiffuses into the social system.


  1. Rogers, Everett M., Diffusion of Innovations (Fourth edition), New York: The Free Press (1995).
  2. Rogers: op. cit.
  3. Bass, F., A New Product Growth for Model Consumer Durables, Management Science, 15(5), (1969), p215-227.
  4. Mitsufuji, T., How an Innovation Is Formed: A Case Study of Japanese Word Processors, Technological Forecasting and Social Change (2002) (in Press).
  5. Krugman, Paul, Self-Organizing Economy, Blackwell Publishing (1996).
  6. Rogers: op. cit.
  7. Utterback, J., Mastering the Dynamics of Innovation, Boston: HBS Books (1994).
  8. Burt, Ronald S., Social Contagion and Innovation: Cohesion Versus Structural Equivalence, American Journal of Sociology, 92, (1987) p1287-1335.
  9. Arthur, B., Self-Reinforcing Mechanisms in Economics (in Anderson, Arrow and Pines (Eds.), The Economy As an Evolving Complex System, Perseus Books (1988).)
  10. Bak, Per, How Nature Works: The Science of Self-Organized Criticality, Springer-Verlag (1996).
  11. Griliches, Z., Hybrid Corn: An Exploration in the Economics of Technological Change, Econometrica, 25, (1957) p501-523.
  12. Coleman, J. S., H. Menzel, and E. Katz, The Diffusion of an Innovation among Physicians, Sociometry, 20, (1957) p253-270.