Task interspersal for individuals with autism and other neurodevelopmental disorders

Rapp and colleagues scanned every task-interspersal paper they could find for kids with autism or other delays.

They did not run new kids; they simply grouped old studies and graded the proof.

The goal: see if mixing easy and hard trials really helps learning, and where the science gaps are.

Only a handful of small studies exist, and most lack strong designs.

No clear yes-or-no verdict could be drawn; the authors call for component analyses and long-term data.

Vladescu et al. (2021) show smaller sets (3-6 items) beat big sets (12) in tact training—evidence that trial structure matters, yet Rapp’s pool rarely tested set size.

Perez et al. (2020) toilet-trained the kids with a simple 30-min package, proving everyday skills can move without fancy tactics; Rapp notes similar low-intensity TI mixes are under-studied.

Kodak et al. (2022) found a 10-minute scanning assessment predicts who will fail receptive-ID teaching—exactly the kind of component analysis Rapp says is missing from TI work.

Together these papers do not clash; they simply fill different puzzle pieces Rapp’s map left blank.

You now know the TI shelf is half-empty. Before you weave easy trials into hard ones, baseline the learner’s scanning (Kodak), keep exemplar sets small (Vladescu), and track maintenance. Treat TI as an open question, document every tweak, and you’ll help build the evidence Rapp wants.