Extrapolation accuracy underestimates rule learning

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

A priori and a posteriori · Artificial neural network · Decision rule · Epistemology · Extrapolation · Function (biology) · Learning rule · Machine learning · Rule-based system · Slide rule · Statistics · Artificial Intelligence · Bayesian Modeling and Causal Inference · Cognitive Science and Mapping · Computer Science · Decision-Making and Behavioral Economics · Mathematics

• BayestestR

  Open Access•Dominique Makowski, Mattan Ben-Shachar et al.•Journal of Open Source Software•2019

  Add to list