Category Exemplar Production Norms for Hong Kong Cantonese: Instance Probabilities and Word Familiarity.

Bing Li,Qiduo Lin,Hoi Yan Mak,Chih-Mao Huang,Ovid J L Tzeng,Hsu-Wen Huang

doi:10.3389/fpsyg.2021.657706

Abstract

The lexical system of Hong Kong Cantonese has been heavily shaped by the local trilingual environment. The development of cultural- and language-specific norms for Hong Kong Cantonese is fundamental for understanding how the speaker population organize semantic memory, how they utilize their semantic resources, and what information processing strategies they use for the retrieval of semantic knowledge. This study presents a normative database of 72 lexical categories in Hong Kong Cantonese produced by native speakers in a category exemplar production task. Exemplars are enlisted under a category label, along with the instance probabilities and word familiarity scores. Possible English equivalents are given to the exemplars for the convenience of non-HKC speaker researchers. Statistics on categories were further extracted to capture the heterogeneity of the categories: the total number of valid exemplars, the number of exemplars covering 90% of the occurrence and the probabilities of the most frequent exemplars in each category. The database offers a direct lexical sketch of the vocabulary of modern Hong Kong Cantonese in a categorical structure. The category-exemplar lists and the comparative statistics together lay the foundations for further investigations on the Hong Kong Cantonese speaking population from multiple disciplines, such as the structure of semantic knowledge, the time-course of knowledge access, and the processing strategies of young adults. Results of this norm can be also used as a benchmark for other age groups. The database can serve as a crucial resource for establishing initial screening tests to assess the cognitive and psychological functioning of the Cantonese-speaking Hong Kong population in both educational and clinical settings. In sum, this normative study provides a fundamental resource for future studies on language processing mechanisms of Hong Kong Cantonese speaking population, as well as language studies and other cross-language/culture studies on Hong Kong Cantonese.

Highlights

Being categorical is a fundamental property of our knowledge of the world (Barsalou, 2003)
The frequency results from a category exemplar production task are reliable for indexing exemplar typicality
This paper presents a norming study of category instance production for 72 natural semantic categories in modern Hong Kong Cantonese, with instance probability and familiarity rating results

Summary

Introduction

Being categorical is a fundamental property of our knowledge of the world (Barsalou, 2003). There is a graded structure within categories (Lakoff, 1973; Rosch, 1975; Barsalou, 1985), which consists of a core that includes the most representative (i.e., high-typical) examples surrounded by the exemplars which are less representative (i.e., low-typical). Category members are not equal in terms of the “goodness” of their membership This non-equivalence of category members (Mervis and Rosch, 1981) is reflected in the probability that a member will be recalled in production tasks, or in the subjective rating of a proposed category member’s degree of typicality (Rosch, 1975). The exemplar production task conveniently provides both the exemplars and their typicality measurements at the same time

Methods

Results

Discussion

Conclusion