The proposed definition of communication systems as complementary informative processes enables one to accurately model and understand a range of communication phenomena. With this model, we can describe and predict the characteristics of previously unstudied systems. While the linear hierarchy presented above captures simple communication environments, the usefulness of the model may be improved significantly if we allow a single process to have multiple processes beneath it (which we refer to as multiple feet) and multiple processes above a single process (which we refer to as multiple heads). Doing this allows claims to be made about multiple heads and feet that more closely reflect the complexity of real world settings. In addition, graph theoretic communication models may be applied to complex structures, providing a wider range of both quantitative and qualitative models of communication.
Consider that the author might be able to communicate with you through this publication, or through email, hand-written correspondence, or by telephone. Each different medium can be represented by a different foot. The author's language component can be understood as having below it several feet, each consisting of one or more communication processes connecting the author and the reader.
Modeling social communication systems may be facilitated by using multiple ``feet" in the model. A group of people having a particular relationship is referred to as a social network, which may be studied qualitatively [Cha92] or quantitatively [WF94]. The study of social networks can be viewed as the study of which processing feet exist for each individual and what other feet in what other hierarchies individuals use for communication. For example, assume that each individual has a hierarchy and that there are three levels on each hierarchy: high (H), medium (M), and low(L):
The diffusion of information or of an innovation is the transmission of information to some or all of the members of a domain of individuals, with the communication channels going from one individual to one or more others, with the recipient in turn possibly transmitting to one or more others, with time delays often occurring between transmissions [MP85,Rog83]. The diffusing information may spread from one person to another through any of a number of communication media and at varying rates [Cha86,Jen82,LB75,MP85,Rog83].
Person to person diffusion is modeled by each person having a foot connected to several other people's feet, each connection representing a possible point of contact between this individual and another. Diffusion through an individual thus depends on the individual receiving the information and the decision being made to pass along the information.
Multiple heads may occur within the processing of information by an individual human. Information that I receive may be processed in different ways and viewed differently by the author, who is a male, a father, and a teacher, all of which may affect different ways the information is processed given the different roles. These different ways of knowing allow one to communicate with someone else with the same way of knowing, while communication with someone else will result in inverse-function noise affecting what is received.
Consider a situation where we have two overlapping groups of language speakers, with one group speaking French and one group speaking Spanish. There is also a group of people using email, and a group not using email. We may model this situation as in Figure 2 which shows multi-headed and multi-footed processes.
There are consequences to modeling a communication system as a multi-headed multi-footed monster. Such a model claims that hierarchies of processes ``meet" and that there is a common meeting point that is perhaps critical to the effective operation of the communication system. In addition, meeting points that bracket a set of processes above and below suggests that each of these processes may be treated as a single type of functional unit. The existence of these units has a natural support and the study of these units may be more profitable than the study of units with lesser degrees of support. Each unit may evolve, increasing the rate of survival and reproduction of those creatures with this function.