Cognate translation priming with Chinese–Japanese bilinguals: No effect of interlingual phonological similarity.

Abstract

Previous masked translation priming studies, especially those with different-script bilinguals, have shown that cognates provide more priming than noncognates, a difference attributed to cognates’ phonological similarity. In our experiments employing a word naming task, we examined this issue for Chinese–Japanese bilinguals in a slightly different way, using same-script cognates as primes and targets. In Experiment 1, significant cognate priming effects were observed. The sizes of the priming effects were, however, statistically not different for phonologically similar (e.g., 信赖/xin4lai4/-信頼/shiNrai/) and dissimilar cognate pairs (e.g., 保证/bao3zheng4/-保証 /hoshoR/), suggesting no impact of phonological similarity. In Experiment 2, using exclusively Chinese stimuli, we demonstrated a significant homophone priming effect using two-character logographic primes and targets, indicating that phonological priming is possible for two-character Chinese targets. However, priming only emerged for pairs that had the same tone pattern (e.g., 守卫/shou3wei4/-首位/shou3wei4/), suggesting that a match in lexical tone is crucial for observing phonologically based priming in that situation. Therefore, Experiment 3 involved phonologically similar Chinese–Japanese cognate pairs in which the similarity of their suprasegmental phonological features (i.e., lexical tone and pitch-accent information) was varied. Priming effects were statistically not different for tone/accent similar pairs (e.g., 关心/guan1xin1/-関心/kaNsiN/) and dissimilar pairs (e.g., 满足/man3zu2/-満足/maNzoku/). Our results indicate that phonological facilitation is not involved in producing cognate priming effects for Chinese–Japanese bilinguals. Possible explanations, based on underlying representations of logographic cognates, are discussed.

Visual word recognition of Chinese–Japanese bilinguals: limited role of phonology

“…Regarding ERP results, a significant difference was found between cognate priming and unrelated priming within the 90–170 ms and 170–270 ms window, whereas the difference between phonologically similar cognates and phonologically dissimilar cognates was not significant in the 170–270 ms window, suggesting a limited role of phonology in logographic scripts processing. These findings stand in contrast to studies on alphabetic scripts and syllabary scripts (e.g., Grainger et al, 2006 ; Nakayama et al, 2012 ; Okano et al, 2013 ; Jouravlev et al, 2014 ; Ando et al, 2015 ) but are in line with a behavioral study where no phonological priming effect was found in Chinese-Japanese bilinguals ( Liu et al, 2023 ).…”

“…They tested the reaction time, and behavioral data showed no significantly varying extents of priming effects across phonologically analogous (e.g., 信赖/xin4lai4/−信頼/shinrai/) and disparate cognate pairs (e.g., 保证/bao3zheng4/−保証/houshou/), indicating the absence of phonological similarity effect. From the studies of Zhang et al (2019) and Liu et al (2023) , it appears that phonological similarity has little impact on the processing of logographic scripts (e.g., Chinese), contrasting with findings from alphabetic script research.…”

“…This component links to orthographic and phonological processing in the mental lexicon and has been supported by additional studies (e.g., Holcomb and Grainger, 2007 ; Okano et al, 2013 ). Given that prior research ( Liu et al, 2023 ) suggests a weaker phonological activation for logographic scripts, Chinese-Japanese bilinguals may not exhibit cross-linguistic phonological facilitation. Whether such an effect can be observed will depend on the experimental outcomes.…”

“…To date, only one relevant study has been identified. Liu et al (2023) (in their Experiment 1) examined Chinese–Japanese bilinguals using same-script cognates (Chinese and Japanese kanji) as primes and targets with a masked word naming task. They tested the reaction time, and behavioral data showed no significantly varying extents of priming effects across phonologically analogous (e.g., 信赖/xin4lai4/−信頼/shinrai/) and disparate cognate pairs (e.g., 保证/bao3zheng4/−保証/houshou/), indicating the absence of phonological similarity effect.…”

“…As Japanese kanji characters were borrowed from Chinese, the readings of many kanji characters reflect the readings of the Chinese characters when they were integrated into Japanese. Meanwhile, the readings of some Chinese characters have since changed in China, whereas the original readings have typically been retained in Japan ( Liu et al, 2023 ). Therefore, in Chinese-Japanese cognates, some words are phonologically very similar (e.g., 安心/an1xin1/−安心/ansin/), while others are phonologically quite dissimilar (e.g., 条件/tiao2jian4/−条件/jyouken/), even though they share the same orthography and semantics.…”