DRAGENwiki · Article

Natural Environment Teaching (NET)

Sources checked. Clinical review pending.

Natural environment teaching (NET) is teaching that happens during play and daily routines, at moments the learner's own interests create, usually as the partner to more structured table teaching.

NET in 60 seconds

  • It happens where life happens. NET is teaching spread through the day, at good moments inside ordinary activities. [1]
  • It grew out of play-based language teaching. Sundberg and Partington, who wrote NET into a language curriculum, said it is based mainly on the natural language paradigm of Koegel and colleagues. [1]
  • It cares about why the learner talks. NET tries to teach each kind of talk, like asking and naming, under the conditions it needs in real life. [1]
  • It is often mixed with table teaching. In a study of 142 toddlers, the groups that got NET, alone or with table teaching, gained more in adaptive skills than the group that got table teaching only. [4]
  • Teaching inside routines is an evidence-based practice. A 2020 national review lists naturalistic intervention, teaching embedded in everyday routines, as an evidence-based practice. [5]

Jump to: What it is · Where the name comes from · How NET differs from other naturalistic teaching · What a NET moment looks like · What the research shows · Mixing NET and table teaching · Choice, assent and the learner's lead · Who can do it · An example · Common myths · Words to know

What it is

NET is the name verbal behavior programs use for teaching in the flow of a real day: during play, snack time and other everyday moments. [1]

The idea is simple. When a learner wants something, notices something or is busy with something they enjoy, that moment is a chance to learn. The adult uses it to teach one small skill, and the natural result of that skill is the reward. [1][3]

  • At the table: the adult usually picks the item, gives an instruction and hands over a reward.
  • In NET: the learner's interest picks the moment, and the activity itself pays off.

That difference matters. In one study, teaching that followed the child's own wanting was compared with teaching where the instructor chose the item. The two ways led to different results for the same children. [3]

NET is still planned teaching. The adult knows the goal, how much help to give and how to write down what happened. In research, teaching inside play used set targets, session-by-session data and a clear rule for when a skill was learned, just like table teaching. [2]

Where the name comes from

  • Natural environment training. Sundberg and Partington (1998) used the term natural environment training (NET) for instruction "throughout the day at opportune moments in naturally occurring contexts." [1]
  • Its roots. They said NET is based mainly on the natural language paradigm (NLP), a play-based way to teach language that Koegel, O'Dell and Koegel introduced in 1987. [1]
  • A family tree. NLP later became the first step of a larger package called pivotal response training (PRT). [1]
  • Teaching or training. Some newer research says natural environment teaching. A 2024 study of toddlers uses that name. [4] The letters are the same, and so is the idea.
  • An older phrase. The words natural environment teaching are older still. They appear in the title of a 1982 article by Halle, listed in source 1's references. [1]

NET comes from programs built on Skinner's analysis of language. Professionals who follow that analysis call their programs applied verbal behavior (AVB). Sundberg and Partington (1998) recommended combining naturalistic teaching with table teaching in a verbal behavior curriculum. [1]

How NET differs from other naturalistic teaching

NET belongs to a family of naturalistic teaching approaches. Other members include incidental teaching, the natural language paradigm, PRT and the mand-model procedure. [1] They share a goal: skills that show up in real life, not only in the teaching room. [1]

What sets NET apart is its focus on the kinds of verbal behavior Skinner described.

  • Each kind of talk gets its own teaching. NET makes a real effort to teach each kind of verbal behavior under the conditions that belong to it. [1]
  • Asking follows wanting. A request (a mand) is taught at moments when the learner wants something.
  • Naming follows what is there. A label (a tact) is taught so that the thing itself is what prompts the word. Early teaching may use extra cues alongside the object. Then the extra cues fade until the object alone is enough. [1]
  • A close cousin. Source 1 says the mand-model procedure may share more features with NET than the other approaches it reviewed. [1]

Other naturalistic approaches often aim at more than language. Many of them also target social play, such as starting an interaction or sharing attention. Several are built to make teaching less unpleasant, with fun ways that help children join in willingly. [1] The authors noted that social behavior and lower problem behavior are not necessarily explicit parts of NET. [1]

They also named a gap. PRT had written manuals, including ones friendly to parents. In 2006, the authors called for more published step-by-step guidelines for NET, which could help quality and help parents and siblings take part. [1]

What a NET moment looks like

Every learner is different, and a BCBA tailors the plan. Most NET moments share this shape.

  1. Notice what the learner wants right now. Watch where their eyes, hands and feet go. A quick preference check helps you find out. In one study, teaching items came from each child's top choices. [3]
  2. Set up a reason to talk. Keep a favorite toy in sight but just out of reach. Hand over a puzzle with one piece missing. Pause a song right before the best part. Each one gives the learner a reason to communicate.
  3. Wait a moment. Give the learner a real chance to start. Many learners need a few seconds.
  4. Help just enough. If they need help, give it, then fade it over time so the real moment takes over. [1] See prompting and fading.
  5. Let the activity pay off. The reward is what they asked for, or the fun that keeps going. In the requesting study, each correct response earned a specific, matching reward. [3]
  6. Write it down. A quick tally after the moment is enough. A prompt level data sheet works well.

How DRAGEN does step 2. In DRAGEN's practice standard, a set-up never holds back food, comfort, rest, the bathroom or anything else a person needs. A toy on a shelf is fine. A BCBA reviews every set-up.

What the research shows

Teaching inside play can work as well as table teaching. Geiger and colleagues (2012) taught two 4-year-old boys to touch the right picture when it was named. They compared traditional table teaching with the same trials woven into a game or toy activity each boy liked. For both boys, the two ways were equally effective and efficient. [2]

Learners may have a preference, and you can ask. During part of that study, each boy got to choose which way he was taught. One boy liked both ways the same. The other showed more positive affect with play-based teaching and chose it more often. [2]

Asking for things. Jennett, Harris and Delmolino (2008) compared table teaching with mand training for six children aged 3 to 6. In mand training, the child's wanting picked the moment. [3]

  • Five of the six children made more independent requests and learned to request faster with mand training. [3]
  • The same five had slightly better eye contact during table teaching, and fewer challenging behaviors during mand training. [3]
  • One child did not learn to request in either condition. [3]
  • Four of the six parents said their children were asking for more things at home, which had not been seen before training. [3]
  • Parents rated both methods well and slightly favored mand training on everything except eye contact. [3]

A larger study of toddlers. Yanchik, Vietze and Lax (2024) studied 142 toddlers aged 16 to 35 months who received DTT, NET or both. [4] Children who got NET, or NET plus DTT, showed significant improvements in adaptive skills and fewer barrier behaviors compared with the DTT-only group. [4] The authors add that DTT may play a needed foundation role for children with more significant delays. [4]

Skills that travel. Naturalistic approaches were built to help skills show up in new places, following Stokes and Baer (1977). Many experimental studies of naturalistic teaching measure this directly. Source 1 reports that they show better effects than more structured table teaching, citing a 1987 study. [1]

What the national review says. The 2020 review by the National Clearinghouse on Autism Evidence and Practice lists naturalistic intervention as an evidence-based practice. It defines it as techniques embedded in the learner's typical activities and routines, and it places PRT inside that group. It does not use the name NET. [5]

Mixing NET and table teaching

You do not have to choose one or the other.

  • The authors behind NET recommended a mix. Sundberg and Partington suggested combining structured trials with naturalistic teaching. [1]
  • The toddler study points the same way. NET plus DTT was one of the two groups that improved more than DTT alone. [4]
  • Table teaching can still be a foundation. The same authors suggest DTT may be needed for some children with more significant delays. [4]
  • Each format had strengths. In the requesting study, table teaching gave slightly better eye contact, while mand training gave faster requesting and fewer challenging behaviors. [3]

A helpful way to think about the mix: what does this learner need to learn, and where will they need to use it? A skill that has to work at the park should be practiced at the park, or somewhere that looks a lot like it. [1] Your BCBA can plan the mix with you, and you can ask why a skill is taught at the table or in play.

Choice, assent and the learner's lead

This section is DRAGEN's practice standard, built on DRAGEN's four pillars: autonomy, beneficence, nonmaleficence and justice.

  • Follow, do not force. NET works because the learner wants something. If the moment passes, let it go and wait for the next one.
  • "No" is information. A learner who walks away, says "no" or picks something else is telling you something. That is not "noncompliance," and it is never a treatment target.
  • Keep safety separate from dissent. If someone could get hurt, keep everyone safe first. A learner's "no" is not a safety problem.
  • Watch for frustration. In the requesting study, a few children cried or showed frustration when teaching used items they did not want at that moment, or when a strongly wanted item was taken away. [3] Treat signs like these as a reason to change the plan.
  • Let the learner choose how they learn. One study let each child choose the style of teaching during part of the study. [2] Choice can be part of the plan.

Assent never replaces the consent a parent or guardian gives. See assent and assent withdrawal.

Who can do it

  • Parents, teachers and siblings. Naturalistic approaches lean on the people already in the learner's life, called natural change agents. [1] You can learn NET.
  • RBTs. In many programs, a registered behavior technician runs NET moments during sessions, following the written plan.
  • BCBAs. A BCBA picks the targets, the kind of help and how to measure, then reviews the data and changes the plan when it is not working.

A BCBA reviews every clinical decision. Coaching for parents can use behavioral skills training: instructions, a model, practice and feedback.

An example

This is a made-up example. It is not a real child.

Learner A is four and loves toy cars and a ramp. The BCBA's plan has two NET goals: ask for "car," and name colors when looking at the cars.

  • The set-up. A parent puts the ramp on the floor and keeps the cars in a clear box nearby. Learner A can see them.
  • The moment. Learner A reaches for the box. The parent waits about three seconds. Learner A says "ca." That sound is the step the plan accepts for now, so the parent says "Car," and hands one over right away. Learner A sends it down the ramp.
  • A label. As the red car zooms, the parent asks, "What color?" Learner A says "red." Over the next weeks, the plan fades the question, so seeing the red car is enough.
  • A "no." After ten minutes, Learner A pushes the box away and walks to the couch. The parent stops. The walk-away is honored.
  • The data. The parent jots a tally: five asks, two after a model and three alone; three colors named after the question. The BCBA looks at the tallies each week and adjusts.

Common myths

Myth: NET is just play. NET has goals and data. In research, teaching inside play used set targets, session-by-session data and a clear rule for when a skill was learned. [2] NET also aims at specific kinds of talk under specific conditions. [1]

Myth: NET and table teaching are rivals. The authors behind NET recommended mixing it with structured trials. [1] In the toddler study, NET plus DTT was one of the two groups that did better. [4]

Myth: NET always works better. In the play study, both formats were equally effective and efficient. [2] In the requesting study, table teaching gave slightly better eye contact, and one child learned with neither method. [3]

Myth: In NET, the adult just waits for things to happen. The adult plans which skills to teach and under which conditions, and then fades extra help on purpose. [1]

Words to know

  • NET (natural environment teaching, or training): teaching during play and daily routines, at moments the learner's interest creates. [1][4]
  • Naturalistic teaching approaches: a family of methods that teach in everyday settings, such as incidental teaching, NLP and PRT. [1]
  • Applied verbal behavior (AVB): programs that organize language teaching around Skinner's kinds of verbal behavior. [1]
  • Verbal operant: one kind of verbal behavior in Skinner's analysis, such as a request or a label. [1]
  • Mand: a request, set off by what the person wants and kept going by getting that specific thing. [1]
  • Tact: a label or comment, set off by something the person sees, hears or feels. [1]
  • Embedded teaching: teaching trials woven into an activity, such as a game. [2]
  • Natural change agents: people in the learner's everyday life, such as parents, teachers and siblings, who do the teaching. [1]
  • Generalization: using a skill in new places, with new people or new materials. [1]

Need a PDF?

Create it here, for free: the DRAGEN app. Use it at your next session.

Or print one we already made, also free:

More from DRAGEN

Related DRAGENwiki pages: naturalistic teaching · incidental teaching · discrete trial teaching · pivotal response treatment · mand · tact · motivating operations · establishing operation · stimulus fading · generalization · preference assessment · prompting and fading · parent training · assent

What remains unknown

  • The two studies that compared formats head to head were small: two children [2] and six children [3]. How well their results hold for many learners is not known from these studies.
  • All six children in the requesting study had earlier experience with table teaching, and with the sitting and attending skills it usually teaches. [3] Results may differ for learners without that history.
  • The toddler study's abstract does not say how children were placed into groups, so the groups may have differed at the start. [4]
  • In 2006, published step-by-step guidelines for NET were limited, and the authors asked for more. [1] Whether newer manuals close that gap is not covered by the sources on this page.
  • Which learners do best with which mix of NET and table teaching is not settled. [4]
  • This page is general information, not medical advice. It was last checked on November 12, 2026.

Sources

  1. LeBlanc, L. A., Esch, J., Sidener, T. M., & Firth, A. M. (2006). Behavioral language interventions for children with autism: Comparing applied verbal behavior and naturalistic teaching approaches. The Analysis of Verbal Behavior, 22(1), 49–60. https://doi.org/10.1007/BF03393026 · https://pmc.ncbi.nlm.nih.gov/articles/PMC2774588/
  2. Geiger, K. B., Carr, J. E., LeBlanc, L. A., Hanney, N. M., Polick, A. S., & Heinicke, M. R. (2012). Teaching receptive discriminations to children with autism: A comparison of traditional and embedded discrete trial teaching. Behavior Analysis in Practice, 5(2), 49–59. https://doi.org/10.1007/BF03391823 · https://pmc.ncbi.nlm.nih.gov/articles/PMC3592489/
  3. Jennett, H. K., Harris, S. L., & Delmolino, L. (2008). Discrete trial instruction vs. mand training for teaching children with autism to make requests. The Analysis of Verbal Behavior, 24(1), 69–85. https://doi.org/10.1007/BF03393058 · https://pmc.ncbi.nlm.nih.gov/articles/PMC2779923/
  4. Yanchik, A. A., Vietze, P., & Lax, L. E. (2024). The effects of discrete trial and natural environment teaching on adaptive behavior in toddlers with autism spectrum disorder. American Journal on Intellectual and Developmental Disabilities, 129(4), 263–278. https://doi.org/10.1352/1944-7558-129.4.263 · https://pubmed.ncbi.nlm.nih.gov/38917993/
  5. Steinbrenner, J. R., Hume, K., Odom, S. L., Morin, K. L., Nowell, S. W., Tomaszewski, B., Szendrey, S., McIntyre, N. S., Yücesoy-Özkan, S., & Savage, M. N. (2020). Evidence-based practices for children, youth, and young adults with autism. The University of North Carolina at Chapel Hill, Frank Porter Graham Child Development Institute, National Clearinghouse on Autism Evidence and Practice Review Team. https://ncaep.fpg.unc.edu/wp-content/uploads/EBP-Report-2020.pdf

Pages that link here