The Extended Alone Condition in Functional Analysis

The extended alone condition is a test used in a functional analysis. In it, the person is left alone with no attention or demands. The clinician watches to see if a behavior keeps going anyway. If it does, the behavior may be its own reward.

This condition sounds simple, but it carries real weight. BCBAs and RBTs use it to understand why a behavior happens. Yet some experts now question its use with dangerous behavior. This page explains the purpose and the growing debate.

What the condition is for#

A functional analysis tests why a behavior happens. It sets up different situations and watches the behavior in each. The alone condition removes people, tasks, and toys. Nothing in the room can reinforce the behavior from the outside.

The goal is to spot automatic reinforcement. That means the behavior feels good on its own, with no help from others. Matt Harrington frames the classic clinical question this way.

Should we use an extended alone or ignore condition to screen out for automatic reinforcement? From the talk. Matt Harrington

If the behavior drops when alone, other people likely drive it. If it stays high when alone, the behavior may be self-reinforcing. That single test can shape the whole treatment plan. Knowing the function points you toward the right treatment. It also steers you away from plans that would not work.

How it works as a screen#

Clinicians often add the alone condition as a screening step. It helps rule automatic reinforcement in or out early. Harrington describes it as a standard add-on.

Other things are like adding an extended alone to screen out for automatic reinforcement. From the talk. Matt Harrington

The word "extended" points to the length. The person may be alone for a long stretch. The idea is to give the behavior time to show up without any social payoff. A short look might miss a behavior that comes and goes. Some behaviors only appear after a while alone. The longer window is meant to catch those cases.

This is why the method can feel harsh. To learn if a behavior is self-reinforcing, you wait it out. You let the behavior happen without stepping in. For many behaviors that is fine. For dangerous behavior it raises hard questions.

The ethical concern#

Some behaviors can hurt the person doing them. Head-hitting is one clear example. Leaving a child alone with that behavior means letting harm continue. Dr. Rajaraman names this problem plainly.

Our extended alone conditions and functional analysis is something I've been thinking about recently, where we put a child in a room who engages in dangerous behavior with the idea of being, let's see what would happen if we let them hit their head. From the talk. Dr. D2 Rajaraman

Put that way, the discomfort is obvious. The test asks you to observe harm to gather data. The data may be useful, but the cost can be steep. Rajaraman lands on a strong stance.

I'm actually feeling pretty comfortable saying that that is trauma-inducing, not trauma-informed. From the talk. Dr. D2 Rajaraman

This is a real shift in thinking. Trauma-informed care asks us to reduce harm during assessment. An extended alone condition with dangerous behavior may do the opposite. That tension is why the field is rethinking the method.

Weighing risk against information#

The condition can give clear answers, which is its appeal. Knowing a behavior is automatic changes the whole plan. You would not chase attention or escape if neither drives it. That knowledge can prevent months of wrong treatment.

But information is not the only value at stake. Safety and dignity matter just as much. A test that harms the person to learn something needs a hard second look. Many clinicians now ask if a safer method can answer the same question.

There are gentler options for assessment. Some use a brief version of the condition with tight safety limits. Some skip it and use other data to judge function. The right choice depends on the behavior and the risk. You can hear more assessment thinking in Confessions of a New Behavior Analyst in Functional Analysis.

A method in transition#

The extended alone condition still has a place in some work. For low-risk behaviors it can be safe and useful. The concern grows only when the behavior is dangerous. Then the balance tips toward caution.

The bigger lesson is about how the field grows. Old methods get reviewed as our values sharpen. Trauma-informed care pushes us to ask new questions. A test that once felt normal can look different today. Good clinicians stay open to that kind of change.

Safer ways to answer the same question#

The goal of the condition is real information. You want to know if a behavior is self-reinforcing. There is often more than one way to learn that. A clinician can choose a safer path to the same answer.

One option is a very brief version of the test. You limit the time and set firm safety rules. You stop the moment risk climbs too high. This can still hint at whether the behavior is automatic.

Another option leans on other data. Caregiver interviews can describe when the behavior happens. Watching the behavior in daily life adds more clues. Together these can support a good hypothesis without an extended alone test.

The key is to match the method to the risk. A low-risk behavior may fit a standard condition. A dangerous behavior calls for extra care. The safest method that answers your question is usually the right one.

FAQ#

What does the extended alone condition test for?

It tests whether a behavior is kept going by automatic reinforcement. The person is left alone with no attention, tasks, or toys. If the behavior continues, it may be its own reward. If it stops, other people likely drive it.

Why is the extended alone condition controversial?

Because it can mean watching a person harm themselves. For dangerous behavior like head-hitting, the clinician withholds a response. Some experts now call this trauma-inducing rather than trauma-informed. They push for safer ways to learn the same information.

Is there a safer alternative to the alone condition?

Often, yes. Clinicians can use brief versions with strict safety limits. They can also rely on interviews and other data to judge function. The best choice depends on the behavior and how risky it is. A low-risk behavior may fit a standard condition just fine. A dangerous behavior calls for the safest method that still answers your question.