Modeling means you use the AAC system while you speak. You are not quizzing. You are showing how a word is found and what it can do in a real sentence.

Pick one routine you already do every day. Breakfast, getting dressed, or the drive to school is enough. During that routine, say a short line and tap the same words: I want milk, you go, shoes on. Repeat the same handful of sentences for a week.

Talk first, then tap. If you stare at the screen and go quiet, the child learns that the device pauses the conversation. Keep your voice moving and point to two or three words that carry the meaning.

Wait after you model. Count slowly to ten before you add another prompt. Many children need that pause to look, think, and try. Filling the silence with more questions turns the moment into a test.

Accept any response. A look, a reach, a sound, or one word on the board all count. If you only celebrate a full sentence, you teach them that partial tries are wrong. Expand what they offered: they tap go, you say and tap you go car.

Do not ask them to say it again for someone who just walked in. Performance makes the device feel like homework. Let the new person see you model, and leave the child free to join or not.

Keep a tiny note of the words you actually modeled, not the words you meant to add someday. At the end of the week, the useful next step is more practice with those words, in the same places, until they are easy.