Database Reference
In-Depth Information
The components of a model are expected to interact with each other. Such inter-
actions engender feedback processes.
Feedback
describes the process wherein one
component of the model initiates changes in other components, and those modi-
fications lead to further changes in the component that set the process in motion.
For example, everything else equal, an increase in the size of a population leads
to increases in the number of births, which in turn leads to an increase in the size
of the population. In positive feedback, the original modification leads to changes
that reinforce the component that started the process and typically lead the system
away from its initial state. The resulting dynamics are often referred to as “explosive
dynamics”—a phrase frequently used in modified forms, such as when we refer to
a “population explosion.”
Feedback is said to be negative when the modification in a component leads other
components to respond by counteracting that change. For example, an increase in
a medication dosage may help fight a disease initially, such that lower doses of
medication are required later. Negative feedback is often the engine that drives a
system toward a steady state. The word
negative
does not imply a value judgment—
it merely indicates that feedback tends to negate initial changes.
People from different disciplines perceive the role and strength of feedback
processes differently. Economists, for example, are typically preoccupied with mar-
ket forces that lead to equilibrium in the system. Therefore, the models are dom-
inated by negative feedback mechanisms, such as price increases in response to
increased demand. The work of ecologists and biologists, in contrast, is frequently
concerned with positive feedback, such as that leading to insect outbreaks or the
dominance of hereditary traits in a population.
Most systems contain both positive and negative feedback; these processes are
different and vary in strength. For example, as more people are born in a rural area,
the population may grow faster (positive feedback). However, as the limits of avail-
able arable land are reached by agriculture, the birth rate slows, at first perhaps for
psychological reasons but eventually for reasons of starvation (negative feedback).
Nonlinear relationships complicate the study of feedback processes. An example
of such a nonlinear relationship would occur when a control variable does not in-
crease in direct proportion to another variable but changes in a nonlinear way. Non-
linear feedback processes can cause systems to exhibit complex—even chaotic—
behavior.
A variety of feedback processes engender complex system behavior, and some of
these will be covered later in this topic. For now, we develop a simple model, which
illustrates the concepts of state variables, flows, and feedback processes. Discussion
will then return to some “principles of modeling” that will help you to develop the
model building process in a set of steps.
Besides feedback, two other real-world properties make purely mental models
truly impractical. They are delays and randomness. The response to an action is
often delayed in time with the delayed effect arriving sometimes at the most in-
opportune time or too late for real effect. The time between the recognition of the
onset of a serious contagious disease and the implementation of a vaccine program
to stave off its worst effects can be great—so great that the negative effects of the
