Can control be a function of behavior?

Discussions regarding control as a function of behavior tend to be a reoccurring topic on behavior analytic forums.  These discussions typically include a practitioner’s description of a scenario consisting of a client’s engagement in problem behavior following requests to complete tasks that have been identified as high preference activities.  Due to an absence of problem behavior during the client’s independent engagement with the same high preference activities and the presence of problem behavior when these activities are initiated by others, practitioners suggest the client’s problem behavior is maintained by the client’s wish to control his/her environment.

Sad little boyLet’s tackle this problem by first addressing what we mean by ‘control’.  A quick Google search for its definition resulted in the following, “the power to influence or direct people’s behavior or the course of events.”  That definition sounds reasonable.  However, we are behavior analysts and our explanations for behavioral phenomena should be based on the science of behavior.  So what does our science say about control?  The discussion of control by Cooper, Heron, and Heward (2007) directly references the functional relation.  A functional relation exists when manipulations to one event (the independent variable) can reliably produce changes to another event (the dependent variable).

Let’s apply this definition to the above scenario by asking – What exactly is our client doing when she is attempting to ‘control’ her environment?  In the presence of a specific event (i.e., the practitioner’s request), the client engages in problem behavior (that has produced certain results in the past).  By assessing the situation in this manner, attempting to identify the actual function of the client’s behavior becomes a more apparent solution.

This is a good time to briefly discuss the relevant functions of behavior.  Generally speaking, the purpose of operant behavior is to either make something appear (access via positive reinforcement) or to make something go away (escape via negative reinforcement).  These processes can occur under socially mediated contexts (such as the one described in the scenario above) or automatic ones (which includes sensory stimuli).  This means the client is engaging in problem behavior to either escape the practitioner’s request (regardless if the requested activity had been identified in the past as high preferred) or to access another activity.

The next question to address is – Why is the client engaging in problem behavior in the presence of high preference tasks?  The short answer is

the task is not a high preferred task at that moment.

The presence of various motivating operations (MO) may have reduced the reinforcing efficacy of the specific task.  Here are a few suggestions for why a once high preference task may no longer be preferred when paired with a practitioner’s request.

  • Are you interrupting the client’s engagement in another activity when delivering tasks demands? If so, task interruption might be the problem. The occurrence of a simultaneous, ongoing activity may have reduced the client’s preference for the practitioner requested activity.
  • The client’s preference for specific activities may be sensitive to the time of day. An activity that is high preferred in the morning might not be preferred in the afternoon or evening.  Time of day is the relevant MO in this situation.
  • The client might have a history of others presenting her with aversive tasks. Her history may have generalized to all requests from others.  In this situation, the practitioner’s request is an MO reducing the client’s preference for the activity.

We have discussed why control cannot be the function of behavior, as the term refers to a relationship between environmental events.  We determined the function of problem behavior in the above scenario must either be in the form of escape from an aversive event or access to an appetitive event.  Finally, we discussed the role of MOs in the possible reduction of client tasks preferences.  In a future post, we will discuss some strategies to address the problems listed above.

Do you have questions or comments about this topic?  Share them with us and stay tuned for more from BAM Network!


Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis. (2nd ed.) Upper Saddle River, NJ: Pearson Education Inc.

Leave a Reply