ABA Fundamentals

A comparison of fixed and repetitive models to teach object imitation to children with autism

Halbur et al. (2023) · Journal of Applied Behavior Analysis 2023
★ The Verdict

Use the same fixed model across trials when teaching object imitation to preschoolers with autism — it’s usually faster.

✓ Read this if BCBAs running discrete-trial or table-time sessions for young kids with autism.
✗ Skip if Clinicians working on vocal or gross-motor imitation only.

01Research in Context

01

What this study did

Halbur et al. (2023) asked a simple question: when you model how to copy an action with a toy, should you do it the same way every time or switch it up?

They worked with preschoolers with autism. Each child got both kinds of modeling in an alternating-treatments design.

Fixed modeling meant the adult always stacked the blocks the exact same way. Repetitive modeling changed the block pattern each trial.

02

What they found

In 10 out of 16 head-to-head comparisons, kids learned faster when the model never changed. Two comparisons favored the changing model, and four showed no difference.

The takeaway: sticking with one clear model is usually the quicker path to mastery.

03

How this fits with other research

Deshais et al. (2020) ran almost the same study and got the opposite winner — repetitive models came out on top. The two papers form a direct replication with flipped results.

The clash looks real, but the kids and methods differed slightly. Halbur used an alternating-treatments setup; Deshais used a simpler single-case design. Small details can swing the winner.

Bloh et al. (2025) also used an alternating-treatments design, but compared human versus animated video models. They found individual preference matters — a reminder that the "best" model can vary by learner.

04

Why it matters

When you teach object imitation, start with a fixed model. It’s easy to do and Halbur’s data show it wins more often than not. If the child stalls, try switching to a varied model — Deshais proved that can work too. Track the data and let the learner’s progress pick the final strategy.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Pick one clear toy action and model it identically for five straight trials — then graph the child’s correct responses.

02At a glance

Intervention
discrete trial training
Design
alternating treatments
Sample size
6
Population
autism spectrum disorder
Finding
mixed

03Original abstract

A well-established imitative repertoire can facilitate the acquisition of functional communication, social behaviors, and observational learning. Although early intensive behavioral intervention programs for young children with autism incorporate imitation training, learners with autism may exhibit difficulties in acquiring an imitative repertoire. Few studies have evaluated the types of models responsible for acquisition when teaching imitation to children with autism. A preliminary evaluation with fixed and repetitive model targets suggested that children with autism may acquire imitation more rapidly when taught with repetitive models (Deshais & Vollmer, 2020). The purpose of the current study was to compare the rates of acquisition when teaching with repetitive and fixed models for six children with autism. The findings suggested that (a) fixed models resulted in the most efficient acquisition for 10 of 16 comparisons, (b) fixed and repetitive models had similar efficacy for four comparisons, (c) and the repetitive condition was most efficient for two comparisons.

Journal of Applied Behavior Analysis, 2023 · doi:10.1002/jaba.993