The conditioning of textual responses using "extrinsic" reinforcers.

Researchers wanted to know if preschoolers learn to read faster when they get tokens plus candy instead of just smiles and praise. They worked with typical four-year-olds in a lab. The kids sat at a table and looked at cards with simple words like 'cat.'

When the child said the word correctly, the teacher either gave a token and a small edible or just said 'good job.' The team flipped back and forth between the two conditions four times to be sure any change was caused by the reinforcers.

Kids reached the reading goal quicker on days they earned tokens and candy. When only social praise was used, learning slowed. Each time the extrinsic rewards came back, reading scores jumped again.

The pattern held across every reversal, showing the token-plus-edible package, not chance, drove the gain.

Weinsztok et al. (2023) recently pooled dozens of DTT studies and found the same bottom line: tangible or edible reinforcers usually beat praise for speed of skill acquisition. The 1962 lab result is one of the earliest single cases in that trend line.

Baer et al. (1984) later asked, 'Does the reinforcer still work if it comes after a delay?' Their preschoolers kept matching words and actions even when the candy arrived late, showing the token-edible effect survives real-world timing hiccups.

Hart et al. (1980) moved the idea into daycare free-play. They used incidental teaching instead of tokens, yet still saw language bloom. Together the studies trace a path: start with extrinsic rewards for fast acquisition, then fade to natural social or activity rewards once the skill takes off.

If a client stalls during early reading or echoic drills, swap in tokens backed by a bite-size reinforcer right away. Run a quick reversal probe—praise only for two sessions—to verify the extrinsic package is really the active ingredient. Once responding is steady, thin the edible ratio and shift to social or activity rewards, knowing the skill should maintain.