In 1997, Wolfram Schultz published a discovery about dopaminergic neurons in primates that changed the understanding of motivation: these neurons do not fire when a reward is received — they fire when a reward is predicted. And they fire more when the reward is unexpected. When the reward becomes completely predictable, the firing disappears.
This dopaminergic prediction-error mechanism has direct implications for game design and learning motivation: guaranteed rewards quickly lose their motivational power. Uncertain rewards — dependent on variable performance — keep the dopaminergic system engaged much longer.
The overjustification effect: when rewards destroy intrinsic motivation
Lepper, Greene and Nisbett documented in 1973 a paradoxical phenomenon: children who received external rewards for activities they already did for pleasure engaged with those activities less after the rewards were removed. The external reward had transformed an intrinsically motivated activity into an instrumentally motivated one. Deci, Koestner and Ryan's 1999 meta-analysis confirmed this across 128 studies: tangible, expected rewards for interesting activities robustly reduce intrinsic motivation.
Self-Determination Theory: what truly motivates
Deci and Ryan identified three basic psychological needs whose satisfaction sustains intrinsic motivation: autonomy (feeling that choices are one's own), competence (feeling that challenges are surmountable with real effort) and relatedness. Well-designed educational games satisfy all three: the player chooses difficulty (autonomy), challenges increase as competence grows (competence) and progress is visible (sense of trajectory).
Progressive difficulty and the optimal challenge interval
Csikszentmihalyi described "flow" as the point where challenge is slightly above current ability — enough to require real effort, not enough to produce anxiety. In this state, focus is intense and the sense of progress is continuous. What makes flow dopaminergically relevant: at the optimal challenge interval, outcomes are uncertain. The dopaminergic system remains active because the result is not yet predictable.
Educational Quiz with four difficulty levels, Math Adventure with progressively increasing speed requirements, and Jigsaw Puzzle with four levels (24 to 96 pieces) all implement this principle structurally.
Conclusion
Schultz's dopamine research, Lepper's overjustification effect and Deci and Ryan's SDT converge on the same conclusion: lasting motivation to learn depends on genuine uncertainty about outcomes, real autonomy over the process and progressively challenged competence. Guaranteed, easy and frequent rewards do the opposite.