Curiosity is fundamental to young children’s learning, yet no brief, validated behavioral measure of epistemic curiosity is designed for field-based research with preschool-aged children. This article describes the development and validation of the Animal Game, a 10-trial behavioral task, that yields two scores: a Novelty Score, reflecting preference for new over familiar stimuli (interest-type curiosity), and a Novelty plus Knowledge-Seeking Score, additionally capturing the desire to learn more (deprivation-type curiosity). Data were collected between March 2023 and March 2024 from 594 socioeconomically diverse children in the United States (ages 3–6; M = 4.20 years; 52% female; parents 42% White, 28% Black, 21% Hispanic or Latino). Internal consistency was high (α = .85 and .83) and six-month test–retest reliability was low to moderate (r = .21–.23). Confirmatory factor analysis supported a two-factor structure, with scalar measurement invariance across child sex and parental education. Receptive vocabulary correlated more strongly with the Novelty plus Knowledge-Seeking Score (r = .181) than with the Novelty Score (r = .115), and the association was robust to a rich set of demographic and parental controls. The Animal Game can capture relevant dimensions of early childhood curiosity in field-based studies, pending replication in independent samples.