Extrapolation accuracy underestimates rule learning: Evidence from the function-learning paradigm

Understanding the development of non-linear processes such as economic or population growth is an important prerequisite for informed decisions in those areas. In function-learning paradigm, people's understanding function rule that underlies to-be predicted process typically measured by means extrapolation accuracy. Here we argue, however, even though accurate necessitates rule-learning, reverse does not necessarily hold: Inaccurate exclude rule-learning. Experiment 1 shows more than one third participants who would be classified “exemplar-based learners” based on their accuracy were able to identify correct shape and slope a rule-selection demonstrating rule. 2 higher proportions learning ruleapplication paradigm due (i) priori probabilities guess paradigm; nor it (ii) lack simultaneous access all values paradigm. We conclude application tantamount assessing xlearning via underestimates proportion learners experiments.