Equiprobability bias

Biological Parallel

Naive honeybees visiting flowers for the first time probe each species equally—yellow, purple, white—distributing visits uniformly until nectar data accumulates. This maximum entropy strategy (equal priors absent evidence) minimizes risk when pattern knowledge is zero. But humans overapply this to non-uniform realities: lottery balls (truly uniform) feel equivalent to childbirth sequences (falsely assuming boy-girl runs 'even out'), weather forecasts (30% rain ≠ 50-50), basketball shooting percentages (falsely expecting 60% shooter to alternate makes-and-misses evenly). The bias persists because ancestral environments offered sparse data—uniform sampling was the safest bet when you'd only see 10-20 instances before deciding.