Differential reinforcement

Differential reinforcement refers to (a) the procedure used in shaping new behaviors and (b) differential reinforcement procedures such as DRO or DRA used to reduce problem behavior. 

DRA is an umbrella term that refers to reinforcement of behaviors that are appropriate alternatives to a target behavior. DRI (Incompatible) is a subtype of DRA in that the alternative behavior is incompatible with (or cannot be done at the same time as) the problem behavior. More recently emphasis has been placed on selecting alternative behaviors that serve the same function as the problem behavior. Often these are communicative (i.e., DRC) behaviors (e.g., "May I have a cookie," as the alternative to stealing cookies) The term DRC (Communication) is not included in many textbooks. 

DRA is an umbrella term that refers to reinforcement of behaviors that are appropriate alternatives to a target behavior. DRI (Incompatible) is a subtype of DRA in that the alternative behavior is incompatible with (or cannot be done at the same time as) the problem behavior. More recently emphasis has been placed on selecting alternative behaviors that serve the same function as the problem behavior. Often these are communicative (i.e., DRC) behaviors (e.g., "May I have a cookie," as the alternative to stealing cookies) The term DRC (Communication) is not included in many textbooks. 

DRA/DRI is more effective than DRO. It has the advantage of building appropriate behavior and puts the focus of the behavior manager on appropriate behavior. DRO and DRD have a major disadvantages of focusing on the inappropriate behavior and providing attention for the inappropriate behavior.  

The "dead man's test" - If a dead man can do it, it is not a behavior. DRO and DRD base the reinforcement contingency on not emitting a behavior, which fails the dead man's test. This creates several disadvantages with DRO and DRD. 

1. No behavior is taught. 
2. It risks reinforcing unwanted behavior. 
3. It accentuates the negative behavior by monitoring it. This can reduce the behavior manager's attention on appropriate behavior and can provide attention to the inappropriate behavior. Adult attention is almost always a positive reinforcer of the behavior it is focused on, so DRO and DRD risk reinforcing the behavior that is being reduced with attention. 
4. DRO and DRD are less durable because the replacement behavior is not developed which will receive natural reinforcement. DRA/DRI increase a replacement behavior and DRL shapes the temporal relationship of a repeating behavior.

Drl is same space-responding DRL.

DRL (or spaced-responding DRL) can be used to reinforce responding at a low rate. When the goal is to eliminate a problem behavior, DRO or DRD is used and the criterion or time interval is often varied with time as the problem behavior decreases. DRD (DR of Diminishing rates) is a new and perhaps improved label for Full-Session DRL or Interval DRL, because both full-session DRL and interval DRL have a goal of diminishing or eliminating a behavior.

Full-Session DRL - (to reduce or eliminate problem behavior) reinforce at end of session if the target behavior occurred at or below the criterion 

Interval DRL - (to reduce or eliminate problem behavior) reinforce at the end of each interval if target behavior occurred at or below the criterion 

DRD - DR of Diminishing rates (to reduce or eliminate problem behavior) same as full-session DRL or interval DRL, without specifying single or multiple intervals

DRL (or Spaced-Responding DRL) - (to reduce responding to a low, stable rate) reinforce the first response after a predetermined minimum IRT

Differential reinforcement can use positive or negative reinforcement. DRA with negative reinforcement is sometimes referred to as differential negative reinforcement of alternative behavior (DRNA). Negative reinforcement can also be used for incompatible behavior (DRNI). DRNA/DRNI uses escape from a task demand for the reinforcer. 

DRO is relatively rapid at reducing problem behavior compared to extinction alone. It also does not require the development of a new behavior as in DRA or DRI.

The term "differential reinforcement" refers to responding that meets a particular criterion gets reinforced while responding that does not meet that criterion does not get reinforced. This procedure is used (a) in the shaping process to develop new behaviors and (b) in procedures such as DRO or DRA that are used to reduce problem behavior. Do not confuse "differential reinforcement" with specific types of differential reinforcement (e.g., DRO, DRL, DRI, DRA, DRH). 

DRD (Differential Reinforcement of Declining rates) is used by Mayer et al. (2012). Cooper et al. (2007) refer to this type of DR as full-session DRL and interval DRL. In this course, we will use the term DRD because it is a more accurate description. DRD has the purpose of reducing or eliminating a behavior. The term spaced-responding DRL (Cooper et al.), DRL (Mayer et al.) or DRL (low rate) refers to DR with the objective of producing a lower rate of responding, but not eliminating the behavior. This would be used on a behavior such as eating too rapidly. Obviously we do not want the behavior to be eliminated. In this course, we will use the term DRL for this type of DR. 

DRO typically has a less dense schedule of reinforcement compared to DRA/DRI and it is relatively simple to use with a minimum amount of training. 

DRI (Incompatible behavior) - reinforcement is contingent upon behavior that cannot be done at the same time as the problem behavior

DRL (Low rates) - reinforcement only after responding below a predetermined rate

DRO (Other [or Omission] of behavior) - reinforcement is provided contingent upon the non-occurrence of behavior

DRA (of Alternative behaviors) - reinforcement of behaviors that are appropriate alternatives to the target behavior.