Determining the cognitive biases behind a viral linguistic universal: the order of multiple adjectives

Abstract When people are asked to create a phrase with the elements {blue, earrings, beautiful}, they produce ‘beautiful blue earrings’. Several theories have been proposed about origins of this universal tendency order multiple adjectives in specific way: an innate hierarchy designated positions for each category adjectives, sensitivity definiteness adjectival denotation, availability and psychological closeness adjective attributes speaker, encoding subjective vs. objective properties, adjective’s phonological weight. Although these strong descriptive power, often focus on what happens at phenotypic level without explaining cognitive needs trigger behavior. Through timed task that measures acceptability ‘Adjective-Adjective-Noun’ sequences either comply or violate it, we adduce evidence high violating orders, whose processing did not take longer than compliant as should happened if former were non-canonical. The results suggest ordering preferences exist but invariable, one would expect linguistic was involved. We track origin synergistic interplay three biases: Zipf’s Law, Intolerance Ambiguity, Novel Information Bias. Last, show manifestation is sensitive statistical distribution input data, resulting variation even among speakers same language.