While a there is a literature addressing the issue of “Prompt Dependence, with respect to decreasing its occurrence, a fundamental problem exists regarding its definition.
Green .. identified it as ….
However, the term prompt dependence is fundamentally inconsistent with a behavior analytic perspective.
The term “prompt dependent” suggests that the individual is somehow requires prompts to be able to initiate or continue specific activities or steps of a task.
However this formulation has two fatal flaws, first it places the explanation for this “dependence” within the individual (they are dependent) rather than in the environment (consequences maintaining the behavior).
Second the behavior of concern is not correctly described as “dependence”, that is not a behavior, rather it is a characterization of how we “feel” about some other behavior. It is the actual behavior that the person engages in that we must turn our attention and our analysis. The actions (behaviors) we must concern ourselves with are the acts of stopping, or ceasing the actions that are desired. This should not be confused with the absence of behavior, as the topography of response can be specified and may take a number of forms that can be operationally defined.
It is axiomatic that the individual who is described as prompt dependent, actually has the desired response in their repertoire. We are concerned with the fact that they are not initiating or continuing to engage in a behavior that they have in their current repertoire. It may be more accurately described as stimulus control problem. Despite having the ability to perform this skill or behavior, the individual does not perform the behavior under the control of the conditions where we want the response to occur.
If we are to remain consistent with our science, this failure to demonstrate a skill or behavior that is in the current repertoire must be accounted for using basic behavioral principles. The correct unit of analysis in this context should be the stimulus – response- consequence or (SRC) paradigm. In the context of this paradigm, the first question to be asked is, does the individual understand that the targeted skill or behavior is to be performed in this context? Simply put does the presentation of the discriminative stimulus result in the production of the response? Clearly if we are describing the problem as prompt dependence, the answer is no. This means that there is not currently a reinforcement relationship between the occurrence of the response and the contingent delivery of reinforcement. It may also, suggest a reinforcement relationship between the undesired response (responses that are incompatible with the target response) and the consequence that occurs. The very fact that incorrect responses (responses incompatible with the target response) are occurring at a higher rate than target responses, suggests that a reinforcement relationship may exist.
The next unit of the paradigm to evaluate is the response. This analysis is relatively straight forward. What has to be determined is whether or not the individual can motorically perform the target response. Typically where prompt dependence is the concern, the response is in the repertoire and does not need to be established. If the response is not in the current repertoire, the solution is obvious, you must establish it directly.
Finally, the analysis must turn to the critical element in the S-R-C learning formulation, the consequence. We must next determine what consequence (response) is delivered contingent upon the occurrence of incorrect performance of the target behavior (not responding). The contingent delivery of prompting as a consequence for not responding must contrasted with the consequence provided for correct performance of the target response. When that data demonstrate higher levels of behavior for one consequence and low levels of the other response, the correct interpretation is that one consequence is a more effective reinforcer than the other.
When this lens is applied the correct analysis is to look at the consequences provided contingent upon stopping. For an individual described as prompt dependent, this consequence is typically prompting. The fact that this consequence reliably follows the occurrence of the behavior of stopping suggests that “prompting” may be the consequence maintaining “stopping”. The individual who is reliably producing prompting from multiple caregivers across a wide range of behavior may simply be maximizing access to those topographies of attention. This is at this point an hypothesis derived from direct observational data should then be subject to empirical validation. It is an empirical question not an opinion or perception question.
It is critical to keep in mind that when clinicians use the term “prompt dependent” they are describing their hypothesis, not an objective fact. If this hypothesis is mistaken for an objective fact, many problems with respect to treatment may occur. The first problem is that this hypothesis is not typically not formally evaluated to determine if the premise (that they are prompt dependent) is objectively true. In order to do this we must conduct and analysis of the skills that the individual has learned and demonstrates correctly and independently and contrast this with those where dependence is seen. Where evidence exists that the individual has learned some skills, behaviors, or routines and completes them without prompting, we must acknowledge that we cannot continue to view THEM as dependent in light of data demonstrate that they are not under some conditions and circumstances. In short, we must alter our hypothesis. Our description of the individual must become more nuanced. It is more accurate to describe of them as independent on some task and skills, but not on others and describe the conditions where these differences are observed. The willingness to look for dis-confirming data not just accept the current hypothesis uncritically is one of the most critical skills that behavior analysis brings to the treatment table. Where a person is labeled as dependent upon prompts despite the fact that objective evidence exists that this is not true in all conditions, suggests that dis-confirming data is being ignored. A truism of behavior analysis should be that inaccurate descriptions of behavior will invariably lead to ineffective treatment.
A second concern and equally important point is that behavioral interventions to decrease “dependence” do not logically flow from a belief that the problem lies within the learner. The science of behavior analysis postulates that the controlling variables of behavior are correctly found in the environment not within the person. The belief that the problem is within the learners is simply inconsistent with a behavioral analytic perspective. It is critically important that we remain faithful to our basic principles in the analysis of behavior.
When looking at the problem that is often described as prompt dependence we must begin with accurate descriptions of the actual behavior occurring, not subjective characterizations of those behaviors. This subjectivity can be seen when the descriptions of the individual’s behavior includes the use of descriptors such as; dependence or needing prompts. These terms do not describe behavior; they refer to our hypothesis of the “cause” of that behavior. This fundamental error in description leads treatment away from environmental variables (the environment) and towards inner causes. It may be more appropriate to describe these behaviors in terms of their consequences. There are a range of descriptions that may be more accurate to describe the phenomenon in terms of its effect on the environment. Some possible alternative descriptors might include; prompt producing behavior; prompt maintained behavior or prompt seeking behavior. Fundamentally, the “dependence” description/hypothesis is inconsistent with the science of behavior analysis and may decrease the likelihood of the implementation of function based treatment.