Positive Behavior Support (PBIS)
and behavior intervention

The core ideas behind behavior intervention in special education — the function of behavior, positive behavior support, functional behavior assessment and ABC data, replacement behavior, prompting, and independent correct responding — explained in plain terms, with a note on where Avatar Coach fits and where it deliberately stops.

Behavior has a reason — where intervention starts

Behavior intervention is not about punishing a challenging behavior until it stops. It is an educational approach: observe why the behavior happens, change the environment accordingly, and teach a new behavior that works better.

The reason behind a behavior is called its function. Five functions cover most of what teachers see, put in everyday words:

• Attention — "notice me"
• Access to items or activities — "I want that"
• Sensory — "that feels good"
• Escape from a task — "I don't want to do this"
• Escape from discomfort — "this hurts, or it's too loud, too bright"

The same hitting means opposite things depending on function, and calls for opposite responses. If it is attention-maintained, you reduce the reaction to the behavior and teach a better way to get attention. If it is escape-maintained, you look first at task difficulty and how the task is presented. That is why good intervention always begins with observation and data.

What PBIS is — prevention and teaching instead of punishment

Positive Behavioral Interventions and Supports (PBIS, also called PBS) replaces after-the-fact punishment with prevention and instruction. It is not a slogan about praising more; it is a structured process grounded in assessment of function.

1. Identify — define the behavior concretely: what it looks like, when it happens, how often.
2. Assess function — form a hypothesis about why, from observation data.
3. Plan — combine prevention (environment), instruction (replacement behavior), and reinforcement.
4. Implement, review, revise — run the plan, look at the data, adjust.

At school level this is organized in tiers: universal support for all students (Tier 1), small-group support (Tier 2), and individualized support (Tier 3). In the Republic of Korea, provincial offices of education publish official implementation materials, and this is standard practice in the field.

Functional behavior assessment and ABC data — the most labor-intensive step

The process of identifying function is called functional behavior assessment (FBA). Its most basic tool is ABC data collection: what happened immediately before the behavior (A, antecedent), the behavior itself (B), and what followed (C, consequence).

Enough of these entries support a hypothesis: "during [clean-up time], in order to [escape the task], the child [throws blocks]." ABC data does not prove cause — it generates hypotheses. The more entries, and the more accurate they are, the sturdier the hypothesis.

The problem is time. This is the single most labor-intensive step in behavior intervention. A teacher typically carries a paper form for one to two weeks. Recording while simultaneously teaching is not realistic, and notes written from memory after school are less accurate.

One distinction worth keeping: FBA is not the same as experimental functional analysis (FA). The two terms are often used interchangeably, and they should not be. Avatar Coach supports observational data collection, not experimental analysis.

Replacement behavior — teach, don't just suppress

The center of PBIS is not stopping a behavior but teaching a better way to achieve the same purpose. A child who hits to get attention learns to call the teacher. A child who screams to get an object learns to point, sign, or hand over a picture card.

What matters is that a replacement behavior is not simply "any nice behavior." It must be functionally equivalent — it has to achieve the same purpose — or the child has no reason to switch. Compared to the challenging behavior, the replacement must be:

• easier to do,
• followed by a better outcome,
• followed by that outcome immediately, and
• reinforced consistently.

Goals are written the same way — as something the child does, not something they stop doing.

Success rate is not enough — prompting, errorless learning, independent responding

"90% success" tells you very little on its own. All of that 90% could be successes where an adult was guiding the child's hand.

Prompting — help comes in degrees

A prompt is help given when a child cannot yet do something alone: a gesture, a picture, a spoken cue, hand-over-hand guidance. Help has intensity. You start with enough support to guarantee success, then reduce it one step at a time — prompt fading. A prompt is not a correction after an error; it is support given before one.

Errorless learning — don't let failure accumulate

Errorless learning minimizes the experience of getting it wrong: give enough support up front to succeed, then withdraw it gradually. It is not "let mistakes slide" — it is a design that prevents the error with a prompt before it happens. This matters especially for children who readily begin avoiding a task once failure repeats.

Independent correct response rate — what the child did alone

This is why behavior intervention tracks independent correct response rate — successes achieved without a prompt — separately from overall success rate. If overall success holds steady while independent responding climbs, the child is stepping down the ladder of support. Without recording how much help was given, that growth is invisible.

How Avatar Coach implements this

Each activity in Avatar Coach is designed as a container for practising a replacement behavior — motor imitation, recognizing and expressing emotion, following instructions, waiting for a turn, asking for help, calming with deep breathing. Each activity states which function and which replacement behavior it trains, so a teacher can match an activity to a child's target behavior.

And while the child plays, the observation record a teacher would otherwise write by hand accumulates on its own.

• Target behavior registration — the teacher tags behavior type and function per child.
• Trial-level automatic records — for every trial: success or not, the type and number of prompts, response time, and the replacement behavior practised.
• Automatic checklist — records are compiled into a replacement-behavior checklist with four states: able / able with help / difficult / no opportunity. The table a teacher used to fill in by hand gets filled in by play data.
• Success and independence kept apart — overall success rate and independent correct response rate are shown separately, so a prompted success is never presented as something the child did alone.
• Report drafts — the accumulated record becomes a behavior observation report and a draft IEP plan for IEP present levels.

The activities follow the same principle. An incorrect answer does not produce a penalty or eliminate the child; it routes through guidance to a successful path, after which support is withdrawn — errorless learning. Reinforcement is points and badges rather than elaborate effects: low sensory load is the default, not an option.

Frequently asked questions

Further reading

The concepts in this guide are aligned with publicly available implementation materials. Terminology follows international usage (PBIS, FBA, ABC data, replacement behavior, prompt fading) so that it maps onto both Korean provincial guidelines and practice elsewhere.

• Center on PBIS (pbis.org) — tiered framework, implementation practices
• Korean provincial offices of education — behavior intervention implementation materials (behavior type and function classification)
• Korea National Institute for Special Education — behavior support materials