Communicative competence and metalinguistic ability: performance by children and adults with autism spectrum disorder.

Plant et al. (2007) gave the Test of Language Competence-Expanded to people with autism. They wanted to see if old DSM-IV labels like Asperger or high-functioning autism predicted scores.

The group included both kids and adults. Everyone completed the same TLC-E tasks that check things like idioms, jokes, and grammar rules.

Across the board, autistic participants scored lower than typical peers. The test showed wide scatter inside the autism group, but the old AS vs. HFA labels did not line up with the scatter.

In plain words, the test revealed real language differences, yet the subgroup names told us nothing useful.

Gotham et al. (2014) extends this picture. They tested people who had lost the autism diagnosis and found near-typical language, but those people still leaned on verbal memory to keep up. This tells us TLC-E style tasks can flag subtle, lasting quirks even after outward "recovery."

Cardillo et al. (2021) digs into why the gaps exist. They ran similar pragmatic tasks and showed that theory-of-mind skill, not executive function, is the main engine behind the score drop. The 2007 paper saw the drop; the 2021 paper explains it.

Kaland et al. (2008) and Verté et al. (2006) echo the null on subtyping. Both tried to split AS, HFA, and PDD-NOS on theory-of-mind and executive-function tests and also found no meaningful label-based gaps, backing the 2007 claim that DSM-IV subtypes are weak guides.

Stop using Asperger vs. high-functioning labels to guess language skill. Instead, run a quick TLC-E or similar task to see each client's real profile. If scores are low, add theory-of-mind teaching to your plan, not just rote language drills. The test gives you clear, child-specific targets, and the label on the chart does not.