BUS603: Managing People

The Basic Operant Model

According to operant conditioning theory, we learn to behave in a particular fashion because of consequences that resulted from our past behaviors. The learning process involves three distinct steps (see Table 7.2). The first step involves a stimulus (S). The stimulus is any situation or event we perceive that we then respond to. A homework assignment is a stimulus. The second step involves a response (R), that is, any behavior or action we take in reaction to the stimulus. Staying up late to get your homework assignment in on time is a response. (We use the words response and behavior interchangeably here.) Finally, a consequence (C) is any event that follows our response and that makes the response more or less likely to occur in the future. If Colleen Sullivan receives praise from her superior for working hard, and if getting that praise is a pleasurable event, then it is likely that Colleen will work hard again in the future. If, on the other hand, the superior ignores or criticizes Colleen's response (working hard), this consequence is likely to make Colleen avoid working hard in the future. It is the experienced consequence (positive or negative) that influences whether a response will be repeated the next time the stimulus is presented.

Reinforcement occurs when a consequence makes it more likely the response/behavior will be repeated in the future. In the previous example, praise from Colleen's superior is a reinforcer. Extinction occurs when a consequence makes it less likely the response/behavior will be repeated in the future. Criticism from Colleen's supervisor could cause her to stop working hard on any assignment.

There are three ways to make a response more likely to recur: positive reinforcement, negative reinforcement, and avoidance learning. In addition, there are two ways to make the response less likely to recur: nonreinforcement and punishment.

Making a Response More Likely

According to reinforcement theorists, managers can encourage employees to repeat a behavior if they provide a desirable consequence, or reward, after the behavior is performed. A positive reinforcement is a desirable consequence that satisfies an active need or that removes a barrier to need satisfaction. It can be as simple as a kind word or as major as a promotion. Companies that provide "dinners for two" as awards to those employees who go the extra mile are utilizing positive reinforcement. It is important to note that there are wide variations in what people consider to be a positive reinforcer. Praise from a supervisor may be a powerful reinforcer for some workers (like high-nAch individuals) but not others.

Another technique for making a desired response more likely to be repeated is known as negative reinforcement. When a behavior causes something undesirable to be taken away, the behavior is more likely to be repeated in the future. Managers use negative reinforcement when they remove something unpleasant from an employee's work environment in the hope that this will encourage the desired behavior. Ted doesn't like being continually reminded by Philip to work faster (Ted thinks Philip is nagging him), so he works faster at stocking shelves to avoid being criticized. Philip's reminders are a negative reinforcement for Ted.

Approach using negative reinforcement with extreme caution. Negative reinforcement is often confused with punishment. Punishment, unlike reinforcement (negative or positive), is intended to make a particular behavior go away (not be repeated). Negative reinforcement, like positive reinforcement, is intended to make a behavior more likely to be repeated in the future. In the previous example, Philip's reminders simultaneously punished one behavior (slow stocking) and reinforced another (faster stocking). The difference is often a fine one, but it becomes clearer when we identify the behaviors we are trying to encourage (reinforcement) or discourage (punishment).

A third method of making a response more likely to occur involves a process known as avoidance learning. Avoidance learning occurs when we learn to behave in a certain way to avoid encountering an undesired or unpleasant consequence. We may learn to wake up a minute or so before our alarm clock rings so we can turn it off and not hear the irritating buzzer. Some workers learn to get to work on time to avoid the harsh words or punitive actions of their supervisors. Many organizational discipline systems rely heavily on avoidance learning by using the threat of negative consequences to encourage desired behavior. When managers warn an employee not to be late again, when they threaten to fire a careless worker, or when they transfer someone to an undesirable position, they are relying on the power of avoidance learning.

Making a Response Less Likely

At times it is necessary to discourage a worker from repeating an undesirable behavior. The techniques managers use to make a behavior less likely to occur involve doing something that frustrates the individual's need satisfaction or that removes a currently satisfying circumstance. Punishment is an aversive consequence that follows a behavior and makes it less likely to reoccur.

Note that managers have another alternative, known as nonreinforcement, in which they provide no consequence at all following a worker's response. Nonreinforcement eventually reduces the likelihood of that response reoccurring, which means that managers who fail to reinforce a worker's desirable behavior are also likely to see that desirable behavior less often. If Philip never rewards Ted when he finishes stocking on time, for instance, Ted will probably stop trying to beat the clock. Nonreinforcement can also reduce the likelihood that employees will repeat undesirable behaviors, although it doesn't produce results as quickly as punishment does. Furthermore, if other reinforcing consequences are present, nonreinforcement is unlikely to be effective.

While punishment clearly works more quickly than does nonreinforcement, it has some potentially undesirable side effects. Although punishment effectively tells a person what not to do and stops the undesired behavior, it does not tell them what they should do. In addition, even when punishment works as intended, the worker being punished often develops negative feelings toward the person who does the punishing. Although sometimes it is very difficult for managers to avoid using punishment, it works best when reinforcement is also used. An experiment conducted by two researchers at the University of Kansas found that using nonmonetary reinforcement in addition to punitive disciplinary measures was an effective way to decrease absenteeism in an industrial setting.

Schedules of Reinforcement

When a person is learning a new behavior, like how to perform a new job, it is desirable to reinforce effective behaviors every time they are demonstrated (this is called shaping). But in organizations it is not usually possible to reinforce desired behaviors every time they are performed, for obvious reasons. Moreover, research indicates that constantly reinforcing desired behaviors, termed continuous reinforcement, can be detrimental in the long run. Behaviors that are learned under continuous reinforcement are quickly extinguished (cease to be demonstrated). This is because people will expect a reward (the reinforcement) every time they display the behavior. When they don't receive it after just a few times, they quickly presume that the behavior will no longer be rewarded, and they quit doing it. Any employer can change employees' behavior by simply not paying them!

If behaviors cannot (and should not) be reinforced every time they are exhibited, how often should they be reinforced? This is a question about schedules of reinforcement, or the frequency at which effective employee behaviors should be reinforced. Much of the early research on operant conditioning focused on the best way to maintain the performance of desired behaviors. That is, it attempted to determine how frequently behaviors need to be rewarded so that they are not extinguished. Research zeroed in on four types of reinforcement schedules:

Fixed Ratio. With this schedule, a fixed number of responses (let's say five) must be exhibited before any of the responses are reinforced. If the desired response is coming to work on time, then giving employees a $25 bonus for being punctual every day from Monday through Friday would be a fixed ratio of reinforcement.

Variable Ratio. A variable-ratio schedule reinforces behaviors, on average, a fixed number of times (again let's say five). Sometimes the tenth behavior is reinforced, other times the first, but on average every fifth response is reinforced. People who perform under such variable-ratio schedules like this don't know when they will be rewarded, but they do know that they will be rewarded.

Fixed Interval. In a fixed-interval schedule, a certain amount of time must pass before a behavior is reinforced. With a one-hour fixed-interval schedule, for example, a supervisor visits an employee's workstation and reinforces the first desired behavior she sees. She returns one hour later and reinforces the next desirable behavior. This schedule doesn't imply that reinforcement will be received automatically after the passage of the time period. The time must pass and an appropriate response must be made.

Variable Interval. The variable interval differs from fixed-interval schedules in that the specified time interval passes on average before another appropriate response is reinforced. Sometimes the time period is shorter than the average; sometimes it is longer.

Which type of reinforcement schedule is best? In general, continuous reinforcement is best while employees are learning their jobs or new duties. After that, variable-ratio reinforcement schedules are superior. In most situations the fixed-interval schedule produces the least effective results, with fixed ratio and variable interval falling in between the two extremes. But remember that effective behaviors must be reinforced with some type of schedule, or they may become extinguished.