Effects of Serial Multiple Exemplar Training on Bidirectional Naming in Children with Autism

Salomonsen et al. (2024) tested a new way to teach bidirectional naming to preschoolers with autism. They used serial multiple exemplar training, or S-MET. One picture and its name are taught at a time in a strict order.

Four children aged three to five took part. The team tracked whether each child could later name the picture without being told and could pick the picture when they heard the name.

Three of the four children reached full bidirectional naming. The fourth child made clear gains but did not hit the mastery mark.

All skills were checked with new pictures to be sure the learning was real and not just memorizing one set.

Kim et al. (2023) got the same BiN goal with a different twist. They first gave lots of naming probes, then added mixed-operant instruction if probes alone failed. Both studies show MET works, but S-MET gives you a single clear package instead of a two-step plan.

Ptomey et al. (2021) used a computer to run multiple exemplar drills. Their Mandarin-speaking preschoolers also gained BiN. Salomonsen moves the work off the screen and into the natural play area, showing the method still works face-to-face.

Carnerero et al. (2014) reached BiN with only observational pairing: watch picture, hear name, done. That looks simpler, but it only worked after many trials. S-MET adds active naming practice and reaches mastery faster for most kids.

If you run early learner groups, S-MET gives you a tight script: teach one picture-name link at a time, test both directions, then move to the next. No extra software or probe weeks are needed. Try it next session with a small set of kitchen items or animals and track whether the child both names and selects each one without extra prompts.