In these lecture notes, I explain how we can think of motivation in terms of a relatively simple mechanism called a control system, which is at the heart of basic physiological motivations such as those involved in eating, drinking, and seeking warmth or coolness. To make this mechanism easier to understand, I begin by describing a relatively simple man-made mechanism: a home heating and air-conditioning system. When we're done, you should understand how such mechanisms contribute to the direction and intensity of behavior. But first, a word of caution about the danger of circularity in motivational explanations of behavior.
As B. F. Skinner noted some years ago, it is all too easy to fall into the trap of thinking that you have explained a behavior simply because you have invented a possible motive -- "He's eating because he's hungry." The trap is that, having found what seems like a satisfactory answer to your question, you do not look any further. However, you have not really explained anything, because the only evidence you have that your explanation is correct is the very behavior that your explanation was invented to explain! In fact, it very well could be that he is eating for some other reason -- perhaps he is attempting to win a prize for consuming the greatest number of pizzas in an hour, for example.
To break the circle, you need evidence of hunger that is independent of the fact that you wish to explain by reference to hunger. For example, you may know that the person has not had an opportunity to eat for the past 24 hours. How do you know that he is hungry? "Because he has not eaten for 24 hours!" Note that the observation that the person has not eaten for 24 hours is independent of the observation that he is now eating. Of course, you still have work to do to establish that this explanation is the correct one, but at least you are not engaging is circular reasoning.
As it happens, heat tends to leak down a temperature gradient, from regions of higher temperature to regions of lower temperature, at a rate that depends on (a) the quality of the insulation, and (b) the difference in temperature. Heat begins to leak through the cabin walls and windows to the outside. However, the stove is still producing heat faster than it can leak out, so the room temperature continues to rise. Eventually, however, when the room temperature is, say, 70 degrees, there is so large a difference between inside and outside temperature (40 degrees) that heat is now leaking out as fast as the stove can produce it. The system comes to an equilibrium and the temperature levels off at 70 degrees. Satisfied with the temperature, you bed down for the night.
Around midnight, while you are fast asleep, an arctic cold-front blasts through and the outside temperature plummets from 30 degrees to minus 20. What happens to the temperature inside the cabin? Because there is now a much greater difference between inside and outside temperature, heat begins to leak out of the cabin at a greatly accelerated rate. The inside temperature falls rapidly, eventually leveling off at a new equilibrium temperature that re-establishes the same 40 degree difference between inside and outside temperature as before. The new inside temperature is 20 degrees, and you wake up freezing.
The Franklin stove-system doesn't really do what you want it to do -- keep the cabin at a comfortable temperature. In fact, any disturbance to the outside temperature (any change up or down) will eventually produce an equally large disturbance to the inside temperature. For fix this problem, you will need to replace the stove with a more complex system.
We'll rip out the Franklin stove and replace it with a good, high-capacity furnace. We'll also add a new device: a thermostat. The thermostat measures the room temperature and compares this to the set point for room temperature, which in this case we will assume to be 70 degrees. From these two inputs the thermostat produces a single output, called the error signal. In this example the error signal is used to drive the furnace. When the error signal is zero (no difference between set point and room temperature) the furnace is off. When the error signal is positive (room temperature below set point), the furnace is switched on.
The outside temperature is 30 degrees F and you switch on the furnace. Heat blasts into the cabin and the room temperature rises rapidly. If the furnace were left on, the temperature in the cabin might reach over 95 degrees, but after a short while, the room temperature reaches the 70 degree set point, the error signal becomes zero, and the furnace switches off. Because the room temperature is now higher than the outside temperature, heat is leaking out of the cabin and in a short while the inside temperature falls to around 68 degrees, enough to make the error signal become sufficiently positive to switch on the furnace again. The furnace cycles on and off like this, keeping the temperature between 68 and 70 degrees. You bed down for the night.
Again that arctic cold front comes through and the outside temperature falls to minus 20. What happens inside? There is a much greater rate of heat loss so when the furnace comes on it struggles to raise the temperature back to 70 degrees. Eventually the temperature reaches set point and the furnace switches off. With the rapid heat loss, it isn't long before the room temperature is 68 degrees and the furnace switches on again. We observe that the furnace is now spending a much higher percentage of the time on and generating heat. The extra heat this produces is enough to exactly balance off the higher rate of heat loss following the passage of the front.
Notice that the system has reacted to the disturbance -- the increased rate of heat loss -- by increasing the rate of heat production. The response of the system to the disturbance has an effect on room temperature that is opposite to that produced by the disturbance. When a system reacts to a disturbance in a way that tends to oppose the effect of the disturbance, we say that the system has negative feedback. Negative feedback tends to stabilize a system against the effects of disturbances.
Although our furnace system will oppose disturbances that tend to bring the room temperature below the set point, it will do nothing about disturbances that tend to bring the room temperature above the set point. For protection against the latter, we need to add a second control system, with its own thermostat and its own output mechanism -- an air conditioner. In this system, when the room temperature goes above set point, an error signal is generated that turns on the air conditioner, which then acts to pump the excess heat out of the room. If the room temperature is cooler than the outside temperature, heat will tend to leak in through the walls and windows, and the air conditioner will then pump any excess heat back out. Again we have negative feedback, with the disturbance tending to raise the room temperature above set point and the system acting to oppose this disturbance by pumping the heat out, thus lending to lower the room temperature.
At the same time that these purely physiological mechanisms are automatically brought into play, something else occurs. You become uncomfortable, and begin to experience a desire to do something about it. Moreover, the discomfort is specific: you know that your are cold. So, you become motivated to act, and to act in a specific way. You will now tend to do those things that, in the past, you have found help you to become comfortable again. You run in place, put on some extra clothing, grab a cup of hot chocolate, curl up by a hot fire. A specific error signal has become positive, and this activates specific behaviors that tend to oppose the effect of the disturbance to your body temperature. This accounts for the direction of your behavior -- why you are jumping into a hot bath rather than raiding the refrigerator. Moreover, the larger the error (the colder you feel), the more strongly you will act to get your body temperature back up. This accounts for the intensity of your behavior.
Notice that the behavioral actions have the same negative feedback relationship to the effects of disturbance as the automatic, physiological actions do -- behavior becomes part of the negative feedback look that acts to counter the effects of the disturbance.
When your body temperature rises above the set-point, a different set of automatic, physiological actions come into play. You begin to sweat, an excellent mechanism for reducing body temperature because the evaporation of water from your skin surface carries off large amounts of body heat. The blood vessels near the surface of your skin dilate, allowing the warm blood to circulate better near the skin surface, where its heat can be removed more quickly. Your metabolic rate is dialed down, making you feel lethargic, unwilling to exercise; this reduces heat production. At the same time that these purely physiological changes are taking place, you begin to feel a specific discomfort associated with being "too hot," and, now motivated to do something about it, you seek ways to cool off, most of which you have previously learned by experience. You strip off excess clothing, get out of the sun, down a cold drink, jump into the pool, turn on the car air-conditioner. Again your actions become part of a negative feedback loop, helping to counteract the disturbance to your body temperature. And like the home heating and air-conditioning system, two different systems respond to changes in body temperature, each with its own set of compensatory actions, one system to handle deviations below set point, the other to handle deviations above set point. This is a general feature of biological systems, which usually must employ separate systems to handle deviations from set point in opposite directions.