By Fred Nickols

From time to time I fancy myself a toolmaker to knowledge workers. By “tool” I mean any device that assists the worker in doing the work. One of the more common classes of tools used by many knowledge workers are the models, diagrams, and schematics they use to represent the phenomena with which they grapple. One of my favorite tools in this sense of that word is the GAP-ACT model of behavior and human performance shown in Figure 1. This column focuses on that model.

Performer and Situation PX

Figure 1.The GAP-ACT Model of Human Behavior and Performance

People, to use William T. Powers’ (1989) term, are “living control systems,” which is to say they seek to control–or at least influence–various aspects of their environment. In other words, they target certain variables in their environment. In Figure 1, these targeted variables are represented by the letter T. For a salesperson, the variable in question might be the number of sales or number of new accounts or perhaps the profitability of a given sale. For a manager overseeing a processing operation, the targeted variable might be the error rate or quality of output or perhaps the volume of output. For you, the reader, when driving your car down the highway, one such variable might be the position of your car in a particular lane.

In their efforts to control or influence T, people set goals that define the desired state of T. Goals are represented in Figure 1 by G. The salesperson might have a goal of five new accounts per month or a dollar sales volume of $50,000 per month; the operations manager might set a goal of an error rate of .05% or less; and you probably opt to have your car in the center of the lane (or somewhere else depending on what is going on around you).

All that anyone knows of the actual state of any variable he or she has targeted, he or she knows by way of perceptions. This is true whether the person is directly observing T or reading a report on the condition of T. Perceptions are represented in Figure 1 by P.

As part of their effort to influence or control T, people compare their perceptions of T with their goals for T. Any difference or discrepancy between the perceived state of T and the goal state for T is represented in Figure 1 by the lowercase d on the line leading from the juncture of P and G. An unacceptable discrepancy results in action. Action is represented by A in Figure 1.

These first three elements, goals (G), actions (A), and perceptions (P) make up the GAP portion of the GAP-ACT model.

Actions are taken so as to bring the perceived state of T into alignment with the goal state for T. These actions are purposeful, and purposeful action has a name: intervention. In Figure 1, purposeful action or intervention is represented by the lowercase i on the line leading from A to the juncture of C and T.

Were there no other factors affecting T, our actions would have the desired effect; namely, closing any gap between our goal or desired value of T and our perception of its actual value. But there are other actors and factors at work. The salesperson’s goal for T might be complicated by the customer’s perception of the quality or the prices of the products or services being sold. It might also be confounded by the actions of competitors. The manager’s efforts to realize an error rate of .05% or less might be thwarted by the skill levels of the people carrying out the process in question or perhaps the quality of input materials to the process. And your efforts to keep your car in the middle of your lane might be interfered with by other motorists, road hazards, or even a blowout. Our efforts to control or influence some targeted variable are almost always susceptible to disturbances by complicating and confounding conditions. These complicating conditions are identified in Figure 1 by the letter C.

We act, then, so as to influence T. But T is also influenced by C and, thus, any changes in T represent the net effect of our actions and these other influences. Ordinarily, we are able to adjust and adapt our actions so as to counter, offset, or negate these other factors. On occasion, we are not. The disturbances to T posed by C overwhelm our actions and T does not align with our goal state. The salesperson loses too many sales to a competitor who has a better quality and lower-priced product. The manager finds that employee turnover results in a lack of skilled and knowledgeable staff, making process performance suffer. And you get run out of your lane by a reckless motorist.

Fortunately, for the most part, we are able to adjust to any disturbances posed by C. We manage to keep T aligned with G as revealed to us via P. Through it all, we vary our actions to as to hold results constant. The actions, conditions, and target elements constitute the ACT portion of the GAP-ACT model.

In summary, if we see a GAP, we ACT. An easy way to remember the elements in this model is the name itself: GAP-ACT. A gap is a difference between a goal state (G) and the perceived state (P) of some variable we wish to influence or control. Actions (A) are intended to affect the targeted variable (T). Other confounding and complicating conditions (C) also affect T. This requires us to vary our actions so as to keep our perceptions of T aligned with our goals for T.

So What?

“So what?” you might well ask at this point. Well, answering that question is the task I will tackle in my next column. In it, I will focus on how to put this model to good use.

Powers, W. T. (1989). Living control systems. Gravel Switch, KY: The Control Systems Group, Inc.

FredNickolsAbout the Author
Fred Nickols, CPT, is a performance improvement professional and the managing partner of Distance Consulting LLC. A former Navy technician, instructor, and programmed instruction writer turned consultant and executive, he is a longtime member of ISPI and a frequent contributor to its various publications. He can be contacted via email at, or you may visit his website at