Measuring and supporting language function for children with autism: evidence from a randomized control trial of a social-interaction-based therapy.

Researchers re-examined a 2015 randomized trial of MEHRIT, a play-based parent coaching program. They tracked the preschoolers with autism for six months.

Kids got either MEHRIT or standard community services. Sessions were videotaped at home. Coders counted real speech acts—requests, comments, answers—during free play with mom or dad.

Children in MEHRIT used longer sentences and more purposeful utterances than the control group. Gains showed up on mean length of utterance (MLUm) and number of speech acts, not on traditional vocabulary tests.

The extra progress was medium-sized and held at follow-up. Standard scores barely moved, but everyday talking did.

Aravamudhan et al. (2020) and Gwynette et al. (2020) also boosted expressive language in autism, but they used tabletop drills and instructive feedback. MEHRIT proves you can hit the same target while parents simply play on the carpet.

Tincani et al. (2020) warn that SGD studies over-focus on multiply-controlled mands. MEHRIT’s speech-act coding captures the tacts and intraverbals those studies miss, filling the gap the review flagged.

Together the papers show: measure what kids actually say, not just what tests ask.

Next time you run parent coaching, bring a stopwatch and code speech acts for five minutes. You’ll spot functional growth that standardized tests overlook, just like MEHRIT did. Share the clip with parents so they see the progress too.