语音象征的产生机制：基于敏感期的先天后天作用模型

Bouba-Kiki效应(简称BK效应)指语音和形状特征之间的映射关系。针对BK效应的产生机制,先天论和后天论之间争论激烈。先天论的观点认为人们对语音象征的敏感性是出生时便存在的一种语言机制。而后天论的观点则强调语音象征是语言经验的产物。上述理论均获得大量研究证据的支持,且均无法完全否定对方。这表明,上述理论可能均未完整揭示语音象征的产生机制。鉴于此,针对BK效应的产生机制,梳理先天论与后天论的支持证据,并率先提出语言相关的BK效应敏感期假设。同时,梳理了支持BK效应敏感期的初步研究证据及可能的影响因素。进而,以语言相关的BK效应敏感期假设为基础,提出语音象征产生的先天后天相互作用模型,以整合以往研究中的矛盾。最后,展望了语音象征的未来研究进展和方向。

The mechanism of sound symbolism: Innate and acquired interaction model based on the sensitive period

The Bouba-kiki effect (i.e., BK effect) indicates the mapping between phonemes and shape features. The innate theory and acquired theory have heated debates on the mechanism of the BK effect. The innate theory holds that sensitivity to sound symbolism is inborn. On the contrary, the acquired theory holds that sound symbolism is the product of linguistic experience. The above-mentioned theories are both supported by many studies and neither of them can completely refute the other. This suggests that none of these theories fully reveal the mechanisms of the emergence of sound symbolism. Therefore, the present review systematically reviews previous findings that support the innate and acquired theories on the mechanism of the BK effect, respectively. Then, the present review first proposes the hypothesis of the sensitive period of the language-related BK effect and reviews recent studies that have provided preliminary evidence for the sensitive period of the BK effect. In addition, the present review also reviews potential factors that may affect the sensitive period of the BK effect. Specifically, the present study holds that the BK effect has a sensitive period that is similar to the critical period of language. The BK effect may be influenced by linguistic experience and the distribution of arbitrariness in language. On one hand, the early language learning experience may enhance the sensitivity of infants toward the BK effect. However, with the increase of age and language experiences, the sensitivity of the sound symbolism for adults gradually decreases. On the other hand, children mainly learn symbolic vocabulary in the early stage of language learning and children’s sound symbolism sensitivity gradually increases in this stage. When the proportion of arbitrary vocabulary gradually increases, children’s sensitivity to the sound symbolism gradually decreases. Furthermore, to integrate the above debates, the present review proposes the innate and acquired interaction model for sound symbolism based on the sensitive period of the language-related BK effect. This model holds that sound symbolism is supported by innate mechanisms and acquired experiences. In other words, sound symbolism is simultaneously influenced by the plasticity of the brain and the accumulated language experiences. In this way, children may have an innate “sound symbolism cognitive structure,” which enables them to map and integrate multi-modal inputs. Therefore, children are sensitive to all potential sound symbolism. In the early stage of language learning, the iconicity in language is prominent. Meanwhile, the sound symbolism cognitive structure can be triggered by children’s language expression and the sound symbolism schema is built through assimilation during this period. This process increases the brain’s sensitivity to sound symbolism. However, with the increase of vocabularies, the arbitrariness of the form-meaning mappings also increases. When the sound symbolism cognitive structure fails to assimilate the new vocabulary, the arbitrary schema is built through accommodation. In the accommodation process, the brain’s sensitivity to sound symbolism gradually decreases. As mentioned above, assimilation and accommodation simultaneously participate in language learning. The difference between assimilation and accommodation is that accommodation is gradually enhanced during the whole language learning period, while assimilation is only gradually enhanced during the sensitive period of sound symbolism and gradually decreased after this period. Eventually, with the increase of age and language experiences, assimilation and accommodation reach a dynamic balance. In particular, the proportion of the language’s arbitrariness and iconicity gradually tends to be stable and the sound symbolism cognitive structure is adapted to the external information. As a result, assimilation and accommodation maintain a relatively balanced state. Finally, the present study proposes future research directions for the sensitive period of sound symbolism.