Functional analysis of precursors for serious problem behavior and related intervention.

Three adults with developmental disabilities showed early warning signs like whining or pacing before hitting or self-injury. The team first ran a functional analysis to see why the warning signs happened. Then they taught each adult to ask for what they wanted right when the warning sign appeared.

After learning to ask for items or breaks, all three adults used the new words instead of the warning signs. Serious problem behavior dropped to near zero for everyone. The simple request gave them the same payoff the warning sign used to give.

Foti et al. (2015) later used the same FA-then-FCT plan with nine children who have fragile X. They saw a meaningful improvement in severe behavior across home, school, and the store. The 2008 precursor idea still worked, just with more kids and more places.

Spackman et al. (2025) and Schieltz et al. (2022) moved the same steps online. Parents ran the FA and FCT at home while a coach watched on Zoom. Seventeen kids in Emily’s study and 199 in Schieltz’s still hit the meaningful improvement mark. The early-warning method travels well through a screen.

Corrigan et al. (1998) showed that FCT works best when each request has its own signal, like a green card for break and a red card for toy. Porter et al. (2008) used that trick to make the new request beat the warning sign every time.

You do not have to wait for the big blow-up. Watch for the small sign—tapping, humming, rocking—then teach a quick request. One session can replace a whole behavior chain before it starts. Use the same FA rules you already know, just shift the target to the earliest move you see.

A precursor behavior is an innocuous or low-severity behavior that reliably occurs just before a more severe problem behavior. It is the observable front edge of an escalation: the grumble before the scream, the fist clench before the hit, the pacing before the elopement, the increasingly loud vocalizations before self-injury.

Common examples reported across the literature include changes in vocal volume or tone, negative statements, motor agitation such as rocking or fidgeting, facial tension, dropping to the floor, and mild forms of the target response itself, like a light tap that precedes hard hitting. What counts as a precursor is individual, so it has to be identified for each learner, not assumed from a list.

Identification is empirical. In this study the researchers ran a descriptive assessment focused on transitional probabilities, asking how likely severe problem behavior was given that the candidate precursor had just occurred, and confirmed that the problem behavior reliably followed the precursor. That temporal reliability is what separates a true precursor from behavior that merely co-occurs with bad days.

The first reason is safety. If assessment and treatment can key off a harmless behavior, the learner, staff, and peers are never exposed to the dangerous one. This matters most in functional analysis, where a standard FA of severe aggression or self-injury deliberately evokes the behavior; an FA of the precursor can identify the same function while evoking only fist clenching or grumbling.

The second reason is the functional logic this study tested. If precursor and severe behavior are members of the same response class, meaning they are maintained by the same reinforcer, then intervening on the precursor is intervening on the class. The experimental analysis here supported exactly that: the precursor and the problem behavior served the same function for the participants.

The third reason is practical: precursors give staff a response window. A plan triggered by a precursor interrupts escalation at its cheapest point, before restraint, injury, or room clears are on the table.

The sequence in this study is a usable template. First, a descriptive assessment establishes that severe problem behavior reliably follows the candidate precursor. Second, a functional analysis is conducted with contingencies applied to the precursor, with test conditions such as attention, escape, and tangible arranged as usual, so the maintaining reinforcer can be identified without waiting for the severe behavior. Third, results are compared to confirm the precursor and problem behavior share a function.

Treatment then follows standard function-based logic. Here, participants were taught a functionally equivalent communication response, functional communication training, that produced the same reinforcer the precursor and problem behavior had been producing. Problem behavior decreased for all three individuals.

For practicing BCBAs the takeaway is workflow-level: when a referral involves dangerous topographies, ask early whether a reliable precursor exists. If it does, you may be able to run your FA, build your treatment, and monitor your outcomes largely on the safe member of the response class.