Preference for and effects of variable-as opposed to fixed-reinforcer duration.

Researchers let pigeons choose between two keys. One key gave food for the same length of time every time. The other key gave food for different lengths of time.

The birds could switch keys whenever they wanted. The team counted pecks on each key to see which schedule the birds liked more.

The pigeons pecked more on the variable key. They stuck with it even when it gave less total food.

The birds picked the gamble even when it paid less. The surprise of not knowing the next prize kept them playing.

English et al. (1995) ran the same test with water instead of food. The pigeons still chose the variable schedule. The preference holds across different kinds of reinforcers.

Andrade et al. (2014) showed you can thin human rewards to every few days and still keep step counts high. That study used fixed schedules, yet the birds here loved the opposite. The difference is purpose: F et al. wanted to save tokens; M et al. wanted to see what animals pick when given a choice.

Lovaas et al. (1973) spread news clips and tokens across short sessions. They focused on learning, not preference. Together the papers show timing matters, but the question you ask—learning vs liking—shapes the schedule you pick.

Your client may choose an uncertain reward over a sure one even if it gives less. Build in small surprises—mystery prizes, random bonus tokens—to keep engagement high. Just watch the net payoff; variability helps motivation, but you still need enough reinforcement to maintain the skill.