The effects of behavioral history on response acquisition with immediate and delayed reinforcement.

Snycerski et al. (2004) worked with rats learning a new lever press.

Half the animals got water right away. The other half waited 15 seconds.

Before testing, some rats sat through five sessions where water arrived no matter what they did.

The 15-second wait slowed learning.

Surprise twist: free water sessions later helped the delayed group catch up.

Yet letting rats merely see the lever before training hurt them—but only when reinforcement was delayed.

Taras et al. (1993) explain why tokens or praise must quickly signal the next reinforcer. Their delay-reduction theory backs the 15-s penalty Susan found.

Rosenthal et al. (1980) also used lever pressing, but they shortened electric shocks instead of giving water. Both studies show timing, not size, drives learning.

Schmidt et al. (1969) showed that once shock trains responding, the response can linger without payoff. Susan’s work adds that early free rewards can either fix or foul later delayed payoff, depending on what the animal learned about the lever.

For BCBAs, the message is simple: deliver the chip, praise, or break now, not later. If you must use a token or point system, chain it so the wait never feels like 15 s. And watch out for accidental "free reinforcement" history—too much non-contingent attention before teaching a skill can poison delayed consequences later.