Replication Data for: The Associative System of Early-Learned Hebrew Verbs and Body Parts: A Comparative Study with American English

Paper abstract: This paper compares the associative system of early-learned verbs and body parts in Hebrew with previously published data on American English (Maouene, Josita, Shohei Hidaka & Linda B. Smith. 2008. Body parts and early-learned verbs. Cognitive Science 32(7). 1200–1216). Following the methodology of the former study, 51 Hebrew-speaking college students gave the first body part that came to mind for each of 103 early-learned Hebrew verbs, 81 of which were translational equivalents. Rate of convergence and divergence and underlying patterns were used to make inferences about the constraints at work. Overall convergence (92.3% of the Hebrew data and 93.7% of the English data) reveal similar entropy levels, comparable semantic field shapes of verbs organized by body parts and similar general cluster patterns of verbs by body parts. Most divergence lies in the infrequent responses (offered fewer than 1% of the time) which arise around body parts that are internal, very detailed, very general categorically, used in figurative language, uniquely provided and tend to be subject to cultural taboos. This is a new contribution, as previous work has not quantified the relative proportion of convergent to divergent associations. We discuss how these findings support neural and developmental continuity and stability in the verbal system with respect to the categorization of verbs by body parts cross-culturally.
Dataset description: The dataset consists of two matrices: the first matrix is composed of 101 early-learned American English verbs by 61 body parts as provided by 50 American English speakers. The second matrix displays 103 early-learned Hebrew verbs by 73 body parts provided by 51 Hebrew speakers. The English data collections was done at Indiana University Bloomington, IN, in 2004. A partial analysis of the data set was published between 2006 and 2008. The data collection in Hebrew was done at Ono Academic College (in Hebrew: הקריה האקדמית אונו) located in Kiryat Ono, Tel Aviv region, Israel, in 2012. This is the first publication for this data set.

本文对比了希伯来语中早期习得动词与身体部位的联想系统，并与此前已发表的美式英语相关研究数据（Maouene, Josita, Shohei Hidaka & Linda B. Smith. 2008. Body parts and early-learned verbs. Cognitive Science 32(7): 1200–1216）进行对照。本研究沿用前述研究的实验范式，招募51名希伯来语母语大学生，让其针对103个希伯来语早期习得动词，说出脑海中最先浮现的身体部位，其中81个动词为两类语言间的翻译对等词。本研究通过聚合率、分化率及潜在关联模式，对影响联想系统的约束机制进行推论。整体聚合度方面，希伯来语数据集的聚合占比为92.3%，美式英语数据集为93.7%，二者熵水平相近，按身体部位归类的动词语义场结构相似，且按身体部位划分的动词总体聚类模式一致。多数分化现象出现在低频反应（占比不足1%）中，这类反应所关联的身体部位多为内部器官、细节过细或范畴过于宽泛的部位，或用于比喻性表达、具有独特性且易涉及文化禁忌的部位。本研究首次量化了聚合联想与分化联想的相对占比，填补了此前相关研究的空白。我们还探讨了上述发现如何支持跨文化背景下，按身体部位归类动词的言语系统在神经与发展层面的连续性与稳定性。

本数据集包含两个矩阵：第一个矩阵由50名美式英语使用者提供的61个身体部位，与101个早期习得美式英语动词对应组成；第二个矩阵由51名希伯来语使用者提供的73个身体部位，与103个早期习得希伯来语动词对应组成。美式英语数据采集工作于2004年在美国印第安纳州印第安纳大学伯明顿分校完成，该数据集的部分分析结果曾于2006至2008年间发表。希伯来语数据采集工作于2012年在以色列特拉维夫地区基里亚特奥诺的奥诺学术学院（希伯来语：הקריה האקדמית אונו）完成，本数据集为首次正式发表。