Joint control and the generalization of selection-based verbal behavior.

Lowenkron (1991) wrote a theory paper. He asked: why can a child hear a new word once and later point to the right picture?

He said the answer is not simple if-then rules. Instead, two verbal acts happen at the same time. The child silently says the word and also hears herself. When both match the picture, she points. He called this moment joint control.

He argued that you need spoken words first. Without them, the child cannot talk to herself inside her head.

The paper does not give data. It gives a new story: selection-based tasks only work when the learner already has spoken words.

Joint control is the bridge that lets old spoken skills guide new pointing or matching.

Hozella et al. (2022) later tested the idea with five students with autism. They taught the kids to whisper the steps of a job. All students learned to pick the right items and finish the task. This real-life test shows the 1991 idea can be used in classrooms.

Vosters et al. (2020) asked: is the silent talk really needed? They blocked speech by asking people to hum. Visual tasks fell apart, but tactile tasks did not. This means joint control matters more for pictures than for touch. The lab result sharpens the 1991 claim.

Petursdottir et al. (2023) update the old picture. They say today we should worry less about speak-vs-point and more about icon size, array clutter, and how much a symbol looks like its referent. Their paper keeps the 1991 engine but swaps the road map.

If a learner cannot match new pictures to words, first check spoken skills. Teach the child to echo, name, and self-cue. Then use that inner voice to drive selection tasks. This order can save weeks of trial-and-error drills.