Assessing correspondence following acquisition of an exchange-based communication system.

Jeff et al. (2007) taught two children with developmental delay to hand over a picture card to ask for toys.

The team used a multiple-baseline design across kids. They first trained the card exchange, then checked if each child actually took the toy they had just asked for.

Both children learned to swap the picture for the item. One child always picked the same toy he had requested.

The other child often grabbed a different toy after handing over the card. He needed extra teaching to make his choice match his request.

Savaldi-Harussi et al. (2025) later got faster, bigger gains with toddlers using a smart-glove that lit up short video clips on the card. Their tool cut learning time and pushed success to 90-100%, showing the 2007 method can be upgraded.

Valentino et al. (2019) give a rule for when to start with picture exchange: if a child cannot copy two-syllable sounds, skip vocal mand training and go straight to cards. This extends Jeff et al. by showing who benefits most from the exchange system.

Drasgow et al. (2016) taught only spoken requests, while Jeff et al. used pictures. Both worked, but the picture route may be safer when vocal imitation is weak.

Always test correspondence right after you teach picture exchange. A child may hand you the cookie card yet reach for the bubbles. One quick check saves you from false positives and tells you when to add extra matching trials. Use Valentino’s brief vocal-imitation probe to decide if you should even start with pictures, and keep an eye on new tech like Gat’s smart-glove that could speed things up for your lowest-verbal learners.