There exist many ways to reduce the occurrence of unwanted behavior besides punishment and extinction. Perhaps the best initial approach to decrease unwanted behavior is to reinforce some competing alternative behavior differentially while simultaneously simply ignoring the unwanted one (Skinner, 1953; Kazdin, 1989). There are three basic schedules of differential reinforcement: (1) differential reinforcement of other behavior (DRO), (2) differential reinforcement of incompatible behavior (DRI), and (3) differential reinforcement of low rate (DRL).
The schedule for differential reinforcement of other behavior (DRO) provides reinforcement for any behavior provided that the unwanted one does not occur during a fixed period of time. DRO scheduling is especially suited to nuisance behaviors occurring at a high frequency. For example, puppies exhibiting excessive mouthing tendencies might accept petting for 5 seconds or so before engaging in the undesirable habit. It is important to determine this baseline interval accurately before beginning the training process. The first step is to reward the puppy after 5 seconds regardless of what he is doing as long as he is not mouthing and has not mouthed for at least 5 seconds. Once the 5-second requirement is mastered, the DRO schedule can be lengthened through gradual increments of duration, until he accepts longer periods of attention without mouthing. The DRO schedule directly impacts on the mouthing behavior by reinforcing other behavior occurring in its absence.
An important drawback of the DRO schedule is that an equally unwanted behavior might be inadvertently reinforced (Foxx, 1982a). The puppy in the foregoing example is reinforced at the end of a fixed period, provided that he does not mouth regardless of other behavior that might be going on. At the moment of reinforcement, the puppy may not be mouthing but he might be barking or jumping up, undesirable behaviors that could be easily strengthened as a result of reinforcement. Another drawback of the DRO schedule is that it does not require that the puppy learn anything new to replace mouthing—it only requires that the puppy not mouth or bite.
These problems can be mitigated by introducing a schedule for differential reinforcement of incompatible behavior (DRI) after the DRO schedule has reduced the frequency of the unwanted behavior into workable dimensions. Under the DRI schedule, the puppy is rewarded only if it performs an incompatible target behavior that is both moti-vationally and physically opposed to the unwanted behavior. In the aforementioned case of excessive mouthing, the target behavior might be licking. DRO and DRI schedules can be implemented together. For instance, the puppy can be reinforced after a predetermined interval of time, provided that no mouthing has occurred and that it licks at the end of the period.
The selection of reinforcement can help make this strategy even more effective. In many cases, the most desirable reinforcer is an opportunity to perform the unwanted behavior in a more acceptable form. This could be arranged by substituting an alternate object and activity in place of biting on hands. In the case of excessive mouthing, providing the puppy with a tennis ball, together with gentle tug and fetch games, is a quite satisfying outlet and alternative to mouthing on one's hands. This is a very constructive alternative, since ball play serves an important role in the puppy's future training. Excessive or persistent mouthing is often associated with dominance testing and may require additional training efforts to fully resolve.
The schedule for differential reinforcement of low rate (DRL) is similar to the DRO schedule in that a certain interval of time must pass between opportunities for reinforcement. In the case of DRL, the dog must emit a predetermined number of targeted responses over a fixed interval period or the entire interval is reset, thus further delaying re-
inforcement. The DRL schedule is also similar to a fixed interval schedule, except that any responses exceeding the required contingency reset the interval. DRL schedules are sometimes used in controlling excessive social behaviors that need to be reduced in frequency but not eliminated altogether. Though of technical interest in the laboratory, the DRL schedule is rarely employed in the management of dog behavior.
Was this article helpful?