One of the reasons that mental models are so complex is simply that the real-world systems that humans are trying to understand are highly complex. Over the years, system dynamicists have identified characteristics that seem to appear again and again in real-world systems -- particularly social systems. They have found that: (1) symptoms of a problem are often separated from the actual problem by time and space; (2) complex systems often behave counter to human intuition; (3) policy intervention in complex systems can frequently yield short-term successes but long-term failure, or the reverse; (4) internal system feedback often counters external policy intervention;(5) it is better to structure a system to withstand uncertain external shocks than to try to predict those external shocks; (6) real-world complex systems are not in equilibrium and are continually changing. These characteristics arise due to the nonlinear stock, flow, feedback structures of social systems.

A System Problem and Its Symptoms are Separated By Time and Space

One of the reasons that problems are often hard to solve in social systems is that they are frequently separated from their causes by both time and space. This is a result of both system delays (due to stocks) and system interconnectedness (due to feedback loops). Consider once again, Figure 19, shown below. Investment performance in an insurance company (shown in the top right of the figure) is shown to be directly influenced by portfolio mismatch, but only after a significant delay, and indirectly influenced by many factors that are far removed (spatially) from the firm's financial performance (e.g., sales person skills). Managers in actual insurance companies may or may not be aware of these delays and connections, but their existence ensures that, if a problem arises with an insurance firm's investment performance, deciding what to do to permanently correct it will be difficult indeed. Mentally keeping track of the various interactions and delays, over time, is next to impossible in such a system.

Figure 19: Causal Loop Diagram of a Model Examining the Growth or Decline of a Life Insurance Company. The fact that system problems and system symptoms are separated by time and space implies the need for a holistic or systems approach to problem solving. Although this is perhaps obvious to someone reading this book, it is the exact opposite of the traditional approach to problem solving practiced in most of science and in most organizations. The traditional approach to problem solving is the reductionist approach, which involves breaking down a system experiencing a problem into "manageable" pieces and then analyzing each piece in isolation. The reductionist approach ignores the connections between a system's pieces, because the behavior of an entire system is thought to be merely the sum of the behaviors of its parts. This view is implicitly a "linear" view of the world as the behavior of a linear system is, indeed, merely the sum of the behavior of its parts. The behavior of a nonlinear system, however, is more than just the sum of its parts. A nonlinear system can only be analyzed in its entirety, with the connections between its parts being as important as the parts themselves. An old Sufi allegory known as the "Blind Ones and the Matter of the Elephant" illustrates the folly of the reductionist approach to problem solving, and the usefulness of a "nonlinear" approach to problem solving, quite nicely.

Beyond Ghor was a city. All its inhabitants were blind. A king with his entourage arrived nearby; he brought his army and camped in the desert. He had a mighty elephant, which he used in attack and to increase the people's awe. The populace became anxious to learn about the elephant, and some sightless from among this community ran like fools to find it. Since they did not know even the form or shape of the elephant, they groped sightlessly, gathering information by touching some part of it. Each thought that he knew something because he could feel a part. When they returned to their fellow-citizens, eager groups clustered around them, anxious, misguidedly, to learn the truth from those who were themselves astray. They were asked about the form, the shape, of the elephant, and they listened to all they were told. The man whose hand had reached the ear said, "It is a large, rough thing, wide and broad, like a rug." One who had felt the trunk said, "I have the real facts about it. It is like a straight and hollow pipe, awful and destructive."One who had felt its feet and legs said, "It is mighty and firm, like a pillar." Each had felt one part out of many. Each had perceived it wrongly. Idries Shah. 1969. "The Blind Ones and the Matter of the Elephant," p. 25. In: Tales of the Dervishes. New York: E. P. Dutton.

Counterintuitive Behavior

Nonlinear stock-flow-feedback systems frequently behave in ways that are counterintuitive or different from that which a decision maker's unaided mental model would suggest -- particularly in response to the dynamics of a policy change. Forrester notes, for example, that both the places where "high leverage points" (i.e., places at which a policy change can permanently alter a system's behavior) are located, and the direction in which a decision maker must "push" a system to change it's behavior, are counterintuitive. Frequently, the counterintuitive behavior of social systems is due to the existence of negative feedback loops. As an example, consider a small sailboat with a rudder in the stern. If the sailor operating the boat wishes to turn the bow (front) to the starboard (right), he or she must turn the rudder, located in the stern, to port (left). In other words, he or she must intervene with a policy change (turn the rudder) in a place (the stern) different from where the change will manifest itself (the bow) and in a direction (port) opposite from the one desired for the entire system (starboard). A sailboat is a negative feedback system in the sense that there is a desired direction the sailor wishes to point the bow of the boat and an actual direction the bow of the boat is pointing. If a discrepancy develops between the desired direction and the actual direction, corrective action (moving the rudder) is taken. A second example of a simple negative feedback loop system that can exhibit counterintuitive behavior is a thermostat. Consider the following thought experiment. If a person is given a cigarette lighter (i.e., a small heat source) and an ice cube (i.e., a small source of cold) and asked to apply one to the sensor on the thermostat to make a room warmer, which one should he or she choose? The answer, of course, is the ice cube because applying a cold source to the thermostat's sensor would cause the furnace to turn on and heat the room.

Better Before Worse or Worse Before Bettery

Another characteristic of complex feedback systems is that policy changes can frequently make them better before making them worse, or worse before making them better. Again, this is due to the long run effects of feedback. Ignoring a system's long run feedback effects can lead to policies that yield unintended consequences (see Figure 20 below).

Figure 20: Ignoring the Long Run Effects of Feedback Can Lead to Unintended Consequences Consider the example of congestion in a city's expressway system. If the policy response is to significantly expand the capacity of the system, the short run result is a lessening of congestion. But, as information about the improved, even pleasant-to-use, expressway system begins to affect longer run decisions such as where people choose to live (e.g., people move to the suburbs and commute to work via the improved expressway system), the congestion can return and even be worse than before. Of course, the effect can occur in the opposite direction as, closing down some of a city's expressway to create congestion can, in the long term, cause people to resettle near the city and/or to build and use public transportation.

Policy Resistance

Frequently, a nonlinear feedback system will respond to a policy change in the desired manner for a short period of time, but then return to its pre-policy-change state. This occurs when the system's feedback structure works to defeat the policy change designed to improve it. Policy resistance is caused by a system's negative feedback processes. Consider the example of the percentage of white students attending Boston schools with nonwhite students, before and after mandatory busing was instituted as a policy change. The data for the years 1968 - 1992 is shown in Figure 21. Inspection of the figure reveals that the percentage of white students attending Boston schools with nonwhite students shot up immediately after the busing policy was instituted in 1974, but then gradually declined so that by 1982 it had return to its pre-busing level.

Figure 21: Percentage of White Students Attending School with NonWhite Students in Boston Policy resistance occurs when a policy is applied to a system dominated by negative feedback processes and the policy change does not alter the desired states of the negative loops. In the case of the percentage of white students attending Boston schools with nonwhite students, the busing policy did not change the desire of white parents to have their children attend school primarily with other white students. In other words, it did not change the parents' desired percentage of white students in the schools to which they sent their children. Thus, after busing was instituted, many white families gradually moved to the suburbs and enrolled their children in primarily white schools.

Unpredictability

Many decision makers spend enormous amounts of time and money trying to develop models to precisely predict or forecast the future state of a system. From a system dynamics point of view, however, this is a poor use of a decision maker's resources. There are two reasons for this. The first is that it is impossible, in principle, to precisely predict the future state of a nonlinear feedback system, except in the very short term. The second is that, even if it were possible to predict the future state of a nonlinear feedback system, a decision maker's resources are better spent trying to predict the behavior mode of a system in response to a proposed policy change, and in trying to redesign the stock-flow-feedback structure of a system so that it behaves well, regardless of what happens in the future.

Figure 22: A System Continuously Shocked by External Forces Any real system, whether it is a physical system, biological system, or social system, is continuously shocked or buffeted by external (exogenous) forces. An aircraft, for example is shocked by bursts of wind. The human body is shocked by changes in temperature (e.g., stepping into a cold shower). A firm is shocked by sudden changes in the demand for its product due to, say, changes in customer tastes and preferences. Figure 22 depicts a system that is being continuously shocked by external forces. When a decision maker attempts to forecast the future state of a system, he or she is essentially trying to forecast the timing, magnitude, and direction of the incoming shocks that are perturbing the system, so that he or she can respond to them. The implied mind-set of the decision maker is that the shocks, rather than the stock-flow-feedback structure of the system, are responsible for the behavior of the system. The decision maker's search for the causes of the system's problems is outward or towards the shocks, rather than inward or towards the stock-flow-feedback structure of the system. From a system dynamics point of view, the decision maker's resources would be better spent trying to redesign the stock-flow-feedback structure of the system so that it responds well to shocks, regardless of when they arrive, how large their magnitude, or in what direction they push the system. The system dynamics perspective is an inward or endogenous point of view. In order to illustrate how it is impossible, in principle, to forecast the future state of a nonlinear feedback system, consider the following experiment. Figure 23, below, is a simple system dynamics model depicting the interaction between elephants and hunters. In the experiment, this model is defined to be the "real world system." Next, an exact copy of the "real world system" is made. The "model" is perfect in the sense that its nonlinear stock-flow-feedback structure, its parameters, its distribution of random variates, and its initial values, are identical to those of the "real world system." The "model" is thus more perfectly specified than any actual social system model could ever be in the true real world. (click on figure to run simulation)

Figure 23: Elephant-Hunter Model The experimental simulation of the "model" and "real world system" is set up so that, in period twenty, the "model" begins utilizing a random number stream that has been initiated by a seed value different from the one that initiated the random number stream that is being used by the "real world system". In other words, before period twenty, the "model" and "real world system" are completely identical. After period twenty, the "model" and "real world system" are identical in every way except for the seed values that initiate the streams of random numbers exciting their behaviors.

Figure 24: Simulated Time Series Plot of Hunters from the "Model" and "Real World System" Figure 24 is a time series plot of the simulated stock of hunters from both the "model" and "real world system." Before period twenty, the two curves are clearly identical and overlay perfectly. After period twenty, however, they begin to diverge significantly. Indeed, from about period thirty forward, the perfectly specified "model" of the "real world system" predicts the correct number of hunters in the system, only by chance. The conclusion is thus that, even a perfectly specified model cannot predict the future state of a nonlinear feedback system, except in the very short term. Point prediction is thus impossible in principle.

Disequilibrium

As was pointed out earlier, all actual social systems exist in a state of disequilibrium. This implies that actual social systems possess on-going pressures for change, and stocks that decouple flows and allow inflows to differ from outflows. It also implies that using equilibrium-based modeling techniques, such as many of those used in economics and management, will frequently not yield insights that directly relate to the dynamic behavior of actual social systems.