Establishing sound blending in moderately mentally retarded children: implications of verbal instruction and pictorial prompting.

Sanders et al. (1989) worked with four children who had moderate intellectual disability. The kids were six to eight years old and could not yet blend consonant-vowel-consonant sounds like 'cat'.

Each child got two teaching styles in turn. One style gave spoken hints only. The other style added picture cards on top of the hints. The team tracked how fast each child learned new blends and could use them in untaught words.

Verbal-only teaching won. Children reached mastery in fewer trials and made fewer errors when pictures were left out.

When pictures were added, learning slowed for both trained blends and brand-new words. The blocking effect showed up fast and stayed.

Logan et al. (2000) saw the same picture-block in sight-word reading. Five of six students learned words faster when flashcards showed text alone. Together, the two studies say 'strip the visuals' for both phonics and whole-word tasks in kids with moderate ID.

Singh et al. (1990) dug into why pictures hurt. Their stimulus-control work proved that prior pairing of a picture with a word creates conditioned inhibition. That mechanism explains the slower blending scores R et al. saw a year earlier.

Schneider et al. (1967) and Fantino (1968) set the stage by showing that smooth, errorless prompting beats trial-and-error for learners with ID. Verbal-only trials in the target study follow the same low-error logic and get the same clean gains.

If you teach reading to students with moderate intellectual disability, drop the picture prompts. Run brief verbal trials, check blending after every few attempts, and move to new sounds as soon as accuracy hits 90 percent across two sets. You will save session time and build stronger generalization without extra materials.