Defining Behaviors for Intervention

Why Define Behavior

A critical piece to every behavior plan or behavior intervention is the “target behavior”. In the majority of cases, this is the behavior of concern that the intervention is meant to decrease and it will often be paired with a functionally equivalent replacement behavior. For example, a target behavior labeled “Hits” may be defined as “forcefully swings an open or closed fist in the direction of her head or chin, swings objects in the direction of her head/chin, or swings her head in the direction of a wall or desk with or without making contact” (pulled from an actual behavior support plan referenced in the 2013 article by Smith, Lambert, & Moore). In contrast to this, a replacement behavior labeled “Places” may be defined as “gently places open palms on head or chin, places open palms on objects, or gently lowers and places head against object”.

In both examples, the target behavior and the replacement behavior are described such that a member of the team of interveners (staff, family, etc.) would be likely to identify the behavior as it is occurring such that it can be counted for data collection and responded to with a consequence procedure. With this in mind, we can generally state that we define behaviors so that interveners can count them and use them as a cue for when to implement a behavior change procedure.

The Trouble with Defining Behavior

Now that we’ve agreed on a general “why”, let’s revisit the target behavior labeled “Hits” used above. It reads, “forcefully swings an open or closed fist in the direction of her head or chin, swings objects in the direction of her head/chin, or swings her head in the direction of a wall or desk with or without making contact”. Take a moment and envision this behavior. Maybe even forcefully swing your arm a time or two being careful not to strike a person or object in your vicinity. Now take another moment and imagine someone is watching you as you swing your arm. Does that person also believe the speed of the arm was “forceful”? Does the arc of the arm fall within the definition of a “swing” or is it more of a “whirl”? Was the swing in the “direction” of the head or more towards the shoulder?

Here we see how a behavioral definition, which was pretty clear just a moment ago, can be rife with uncertainty. This is troublesome even when there is just one intervener who is unsure if a behavior “counts”. However, we often instill teams of interveners to measure behavior and intervene. If each person conceptualizes the target behavior differently, then how can we be sure it is counted reliably and accurately? Are there strategies to increase the likelihood that each subjective experience of the target behavior aligns with the actual event? If so, how can these increase agreement, decrease confusion, and ultimately improve treatment fidelity?

Considerations for Defining Behavior

While no source has established the “best” way to define target behaviors for intervention, there are several factors deemed valuable to the process. Several of these are reviewed below.

1. Present Tense, Active Voice – Use the present tense and an active voice form of a verb in your behavioral definition. Verbs that demonstrate action on an object ending in “s” usually meet this criteria. “Touches head”, “Throws ball”, “Sits bottom”, “Raises hand”, “Drops body”, etc. For those with experience with Celeration Charting and Precision Teaching, these may seem familiar.
2. Remove or Limit Adverbs – Adverbs can be great qualifiers in some behavior plans, especially ones that are specifically intervening on the intensity of a behavior. However, if intensity is not a qualifier, adverbs can create confusion. For example, “Quickly raises hand” introduces confusion to the observer. They may ask, “Did that count as quick enough?” when collecting data. Remove the confusion by removing the adverb. “Raises hand” is a fine target behavior without extra qualifiers.
3. Develop Exemplars – Within your team, demonstrate exactly what counts and does not count as an instance of the behavior and review these with all interveners. This is especially useful if the behavior may look different across various contexts or when adverbs cannot be removed from the definition. It is also especially useful when the behavior may not match the words. Even the phrase “raises hand” can lead to some confusion. Was the hand “high” enough? Was the bend to the arm within an “acceptable range”? Does it count if “both” hands are raised at the same time? Having staff review and observe various forms of the behavior can significantly improve detection and, therefore, increase both the quality of the data and correct implementation of the behavior change intervention.
4. Update Definitions regularly – It should come as no surprise that behaviors change over time. In most cases, the behavior is gradually shaped such that the most effective version of the response is selected more and more frequently. Because of this, we should expect changes in the behavior over time and update our definitions accordingly.
5. Build Directly into Interventions – Incorporating these pieces from the outset of designing an intervention can be quite helpful when training an intervener. While this can take many forms, consider the ABC Chart format below:

Notes:

– FR3 refers to a Fixed-Ratio schedule with delivery on the third response.

– FR1 refers to a Continuous schedule with delivery on every response.

Wrap-up

Again, it should be repeated that no “best” way to define a behavior has been established. In fact, if such a best way existed, the field as a whole would be in a very different place. The suggestions and discussion above offer one starting point to a larger discussion on defining behaviors. Hopefully, they are also useful in some way to you, the reader, regarding the creation of behavioral definitions, treatment planning, and intervention.

Until we meet again, happy definition writing!

Disclaimer:

Please note that the strategies and items discussed here do not constitute clinical advice and should not be used in place of treatment interventions developed by clinical professionals and multidisciplinary teams in your place of work. While some of the suggestions here may be incorporated into a well-designed treatment package, every treatment should be individualized to address the needs of the individual served.

References

Smith, G. D., Lambert, J. V., Moore, Z. (2013). Behavior description affects accuracy and reliability. The Journal of General Psychology, 140(4), 269-281.