Understanding metaphor: A relational frame perspective.

The team worked with three autistic children. They used a three-step lesson: first the child names a picture, then names its feeling category, then matches feeling pictures while the teacher adds short verbal hints. No one practiced metaphors directly.

The goal was to see if this quick chain could spark brand-new metaphor talk like 'my heart is a balloon' without further teaching.

All three kids began using fresh metaphorical emotion words after the lesson. The new talk stayed in place later with no extra training.

The study shows a brief equivalence-based sequence can give untrained metaphor language to young autistic learners.

Dixon et al. (2017) got the same result with rival-model videos instead of match-to-sample. Both papers show autistic children can gain metaphor talk; one uses equivalence, the other uses screen models.

Richman et al. (2001) seems to disagree. That same-year study found 'failure on metaphorical inferences' in high-functioning older kids. The gap is about task demand: the target taught basic metaphor links, while M et al. tested tricky pragmatic inferences in fluent speakers.

Fabbretti et al. (1997) set the groundwork. Their equivalence-equivalence work with typical adults and kids gave the theory that Magiati et al. (2001) later applied to autism.

You can add a five-minute tact→category tact→match-to-sample routine to your table time. It may unlock metaphor talk without long drills. Pair it with later video models for variety, and save harder inferential tasks for when the learner has stronger context skills.