1 Department of Psychology, Guangzhou University of Chinese Medicine, Guangzhou 510006, China
2 School of Education, Guangzhou University, Guangzhou 510006, China
3 International School, Guangdong Food and Drug Vocational College, Guangzhou 510520, China
4 School of Interpreting & Translation Studies, Guangdong University of Foreign Studies, Guangzhou 510420, China

摘要：

谐音型歇后语是汉语特殊的语汇表达形式, 对其加工常常需要通达后一语节的同音线索来完成语音、语义隐喻映射。本研究通过2个实验探讨相对熟悉度和同音线索类型对谐音型歇后语加工的作用。实验1结果表明, 在高熟悉度条件下, 被试对语汇的加工策略取决于歇后语的同音线索类型。当后一语节为同音字时, 加工较为快速; 当后一语节为谐音字时, 通达歇后语隐喻意义的路径受阻。对熟悉度较高的谐音型歇后语进行加工, 语义通达表现出预存性; 在缺乏同音线索的条件下, 语料加工时间较长, 反应准确率较低, 支持概念隐喻模型和概念合成假说。而对熟悉度较低的歇后语进行加工, 被试更倾向于采用即时(on-line)策略, 支持概念合成理论和基于心理模型的语用推理假说。实验2结果表明, 呈现“错误同音”线索对歇后语的加工形成干扰, 不支持概念合成理论和基于心理模型的语用推理理论。总的结果表明, 相对熟悉度和同音线索类型影响谐音型歇后语的通达, 谐音歇后语加工需要同时激活语音和语义两条通道。

语用推理理论

Abstract:

Two-part allegorical sayings are a typical language form in Chinese. Understanding two-part allegorical saying involves the ability to understand figurative meanings. Chinese two-part allegorical sayings convey figurative meanings by activating either homophonic or conceptual associations. Homophonic associations are realized based on a conceptual connection between the two homophonic expressions: the second part of the sayings and the expression of the idiomatic meaning. Within the example of Lao tai tai shang ji wo (老太太上鸡窝)—ben dan (笨蛋), a situation is described as an old lady (lao tai tai or老太太) is about to walk towards a henhouse (shang ji wo or上鸡窝), which is reflected in the second part that the purpose of doing this is “heading for eggs” (ben dan or奔蛋). The intended interpretation of the saying “an idiot” (ben dan or笨蛋) could not be worked out without the help of a very crucial apparatus—sound association; that is, “heading for eggs” is pronounced the same with “an idiot” in Chinese with respect to the same segmental combinations and tone patterns. Within the paradigm of sound association, the meaning identified in the source domain (the first part; in our example, the old lady’s behavior) is also observed in the target domain (the second part; in our example, the figurative meaning of the old lady’s behavior) in a metaphoric way through mapping between the two domains, resulting in a shifting from a concrete concept to an abstract one. Mapping, which was described by Lakoff and Johnson in their Conceptual Metaphor Theory, has been considered a powerful theory in interpreting metaphors. Fauconnier proposed Conceptual Blending Theory, emphasizing that mapping happens across spaces via connecting counterparts in the input mental spaces. In our example, it connects one mental space contained the image of an old lady walking towards a henhouse and another mental space describing the purpose of carrying out this behavior. Then the mapping happens when the mental apparatus identifies the sound similarity and generates the intended meaning. Meanwhile, the knowledge of recognizing implicature ( Xu, 2005 ) in pragmatic inference also plays a crucial role in processing two-part allegorical sayings. From this perspective, Chinese two-part allegorical sayings are one of the ideal languages. The successful understanding of them couldn’t be accomplished without considering how people interpret in their real usage. There are three theories relevant to interpreting of Chines two-part allegorical sayings, but what we wonder is which theory is more powerful in explaining the processing of homophonic two-part allegorical sayings in terms of various degrees? Does sound association play a crucial role in the processing? In order to answer these questions, two experiments were designed by using eye-movement instrument: experiment 1 investigated the effect of various degrees of familiarity on the processing of two different types of back parts (homophonic association/ phonography), for example, 老太太上鸡窝—奔蛋 is phonography because there is no metaphoric inference between front and back parts, but 老太太上鸡窝—笨蛋 is with homophonic association because the implied meaning (笨蛋) is inferred from the words (奔蛋) through sound similarity. We asked the participants to judge the semantic relatedness between front and back parts and we found that the judgment was determined by the type of back parts, that is, the homophone facilitated the participants’ judgment because of the sound association; while phonography forced participants to infer the implied meaning of the sayings. Meanwhile, participants took longer time to process the sayings with high familiarity and made more errors in the judgment task, the reason of which might be caused by the negative effect of long-term memory. The result supported the Conceptual Metaphor theory and Conceptual Blending theory. However, participants adopted a quite different processing strategy called the on-line processing strategy when the sayings were with low familiarity. The result supported the Pragmatic Inference theory. Experiment 2 investigated how various intonations affected the judgment of semantic relatedness between front and back parts. The results showed that the characters with the same sound pattern but not with the same intonation (e.g. 笨、本、奔) exerted different influences on the judgment. Specifically, the character “本”, which does not fit into the meaning of any of the two parts, did not play a role in the processing. The result does not support the Conceptual Blending theory.

Key words: homophonic two-part allegoric sayings, eye-movement, concept metaphor theory, conceptual blending theory, pragmatic inference theory

语义性质	低熟悉度	高熟悉度	t
熟悉度	1.89 (0.29)	3.02 (0.59)	-8.29
可表象度	4.34 (0.56)	4.64 (0.61)	-1.68
语义一致度	4.40 (0.64)	4.68 (0.67)	-1.45
预测度	0.04 (0.07)	0.07 (0.10)	-1.23
前一语节笔画数	43.65 (12.34)	44.26 (9.65)	-0.19
后一语节(同音)笔画数	21.39 (7.39)	23.35 (7.35)	-0.90
后一语节(谐音)笔画数	21.09 (7.24)	22.48 (8.44)	-0.60
同音字笔画数	9.65 (3.26)	8.87 (2.14)	0.96
谐音字笔画数	9.35 (3.73)	7.96 (2.88)	1.42
前一语节长度	5.70 (1.11)	5.43 (0.99)	0.84
后一语节长度	2.65 (0.98)	3.17 (1.07)	-1.72
同音字书写概率	0.77 (0.24)	0.83 (0.20)	-0.88

表1 两类材料语义性质评定各项结果的平均值和标准差

语义性质	低熟悉度	高熟悉度	t
熟悉度	1.89 (0.29)	3.02 (0.59)	-8.29
可表象度	4.34 (0.56)	4.64 (0.61)	-1.68
语义一致度	4.40 (0.64)	4.68 (0.67)	-1.45
预测度	0.04 (0.07)	0.07 (0.10)	-1.23
前一语节笔画数	43.65 (12.34)	44.26 (9.65)	-0.19
后一语节(同音)笔画数	21.39 (7.39)	23.35 (7.35)	-0.90
后一语节(谐音)笔画数	21.09 (7.24)	22.48 (8.44)	-0.60
同音字笔画数	9.65 (3.26)	8.87 (2.14)	0.96
谐音字笔画数	9.35 (3.73)	7.96 (2.88)	1.42
前一语节长度	5.70 (1.11)	5.43 (0.99)	0.84
后一语节长度	2.65 (0.98)	3.17 (1.07)	-1.72
同音字书写概率	0.77 (0.24)	0.83 (0.20)	-0.88

眼动指标	高熟悉-同音字		高熟悉-谐音字		低熟悉-同音字		低熟悉-谐音字
	前语节	后语节	前语节	后语节	前语节	后语节	前语节	后语节
总注视时间(ms)	1336 (212)	738 (182)	1443 (264)	816 (208)	1324 (274)	789 (180)	1372 (232)	803 (202)
总注视次数	6.80 (1.19)	3.23 (0.82)	7.55 (1.33)	3.34 (0.78)	7.20 (1.65)	3.40 (0.82)	7.02 (1.13)	3.28 (0.81)
首次注视时间(ms)	232 (49)	258 (61)	223 (51)	258 (63)	214 (45)	252 (51)	229 (44)	262 (68)
首次凝视时间(ms)	720 (223)	515 (137)	715 (219)	497 (179)	689 (211)	521 (161)	668 (183)	521 (155)
首次凝视次数	3.82 (1.05)	2.17 (0.52)	3.83 (0.88)	2.00 (0.46)	3.78 (1.11)	2.21 (0.55)	3.56 (0.88)	2.10 (0.48)
第二次注视时间(ms)	169 (38)	227 (52)	165 (31)	240 (77)	177 (36)	235 (53)	173 (45)	239 (57)
第二次凝视时间(ms)	539 (169)	396 (162)	608 (159)	389 (130)	524 (127)	429 (226)	573 (154)	400 (124)
第二次凝视次数	2.62 (0.77)	1.78 (0.48)	2.98 (0.70)	1.73 (0.49)	2.68 (0.70)	1.92 (0.78)	2.79 (0.69)	1.73 (0.43)

表3 不同条件下目标刺激的各项眼动指标数据的平均值和标准差

眼动指标	高熟悉-同音字		高熟悉-谐音字		低熟悉-同音字		低熟悉-谐音字
	前语节	后语节	前语节	后语节	前语节	后语节	前语节	后语节
总注视时间(ms)	1336 (212)	738 (182)	1443 (264)	816 (208)	1324 (274)	789 (180)	1372 (232)	803 (202)
总注视次数	6.80 (1.19)	3.23 (0.82)	7.55 (1.33)	3.34 (0.78)	7.20 (1.65)	3.40 (0.82)	7.02 (1.13)	3.28 (0.81)
首次注视时间(ms)	232 (49)	258 (61)	223 (51)	258 (63)	214 (45)	252 (51)	229 (44)	262 (68)
首次凝视时间(ms)	720 (223)	515 (137)	715 (219)	497 (179)	689 (211)	521 (161)	668 (183)	521 (155)
首次凝视次数	3.82 (1.05)	2.17 (0.52)	3.83 (0.88)	2.00 (0.46)	3.78 (1.11)	2.21 (0.55)	3.56 (0.88)	2.10 (0.48)
第二次注视时间(ms)	169 (38)	227 (52)	165 (31)	240 (77)	177 (36)	235 (53)	173 (45)	239 (57)
第二次凝视时间(ms)	539 (169)	396 (162)	608 (159)	389 (130)	524 (127)	429 (226)	573 (154)	400 (124)
第二次凝视次数	2.62 (0.77)	1.78 (0.48)	2.98 (0.70)	1.73 (0.49)	2.68 (0.70)	1.92 (0.78)	2.79 (0.69)	1.73 (0.43)

眼动指标	不呈现同音字		呈现同音字		呈现错误同音字
	前语节	后语节	前语节	后语节	前语节	后语节
总注视时间(ms)	812 (119)	748 (153)	771 (110)	769 (149)	791 (116)	820 (110)
总注视次数	4.32 (0.73)	3.42 (0.60)	4.15 (0.61)	3.58 (0.64)	4.21 (0.65)	3.87 (0.63)
首次注视时间(ms)	186 (37)	253 (53)	186 (39)	241 (38)	185 (39)	235 (36)
首次进入时间(ms)	262 (88)	749 (173)	288 (100)	723 (157)	266 (104)	755 (181)
首次凝视时间(ms)	448 (76)	512 (154)	428 (75)	550 (142)	434 (82)	586 (124)
首次凝视次数	2.53 (0.45)	2.26 (0.44)	2.47 (0.40)	2.50 (0.52)	2.51 (0.46)	2.71 (0.50)

表6 在不同条件下目标刺激的各项眼动数据的平均值和标准差

眼动指标	不呈现同音字		呈现同音字		呈现错误同音字
	前语节	后语节	前语节	后语节	前语节	后语节
总注视时间(ms)	812 (119)	748 (153)	771 (110)	769 (149)	791 (116)	820 (110)
总注视次数	4.32 (0.73)	3.42 (0.60)	4.15 (0.61)	3.58 (0.64)	4.21 (0.65)	3.87 (0.63)
首次注视时间(ms)	186 (37)	253 (53)	186 (39)	241 (38)	185 (39)	235 (36)
首次进入时间(ms)	262 (88)	749 (173)	288 (100)	723 (157)	266 (104)	755 (181)
首次凝视时间(ms)	448 (76)	512 (154)	428 (75)	550 (142)	434 (82)	586 (124)
首次凝视次数	2.53 (0.45)	2.26 (0.44)	2.47 (0.40)	2.50 (0.52)	2.51 (0.46)	2.71 (0.50)

曹海波, 兰泽波, 高峰, 于海涛, 李鹏, 王敬欣. 词素位置概率在中文阅读中的作用：词汇判断和眼动研究 [J]. 心理学报, 2023, 55(2): 159-176. 卢张龙, 刘梦娜, 刘玉洁, 马盼盼, 张瑞平. 内隐序列学习表征机制探究：眼动证据 [J]. 心理学报, 2022, 54(7): 779-788. 鹿子佳, 符颖, 张慢慢, 臧传丽, 白学军. 中文词类信息在副中央凹中的加工 [J]. 心理学报, 2022, 54(5): 441-452. 刘洪志, 李兴珊, 李纾, 饶俪琳. 基于期望值最大化的理论何时失效：风险决策中为自己-为所有人决策差异的眼动研究 [J]. 心理学报, 2022, 54(12): 1517-1531. 唐晓雨, 崔鑫忠, 高敏, 袁梦莹. 视听时、空一致性对Pip-and-Pop效应的影响 [J]. 心理学报, 2022, 54(11): 1310-1324. 詹沛达. 引入眼动注视点的联合-交叉负载多模态认知诊断建模 [J]. 心理学报, 2022, 54(11): 1416-1423. 吴三美, 田良苏, 陈家侨, 陈广耀, 王敬欣. 中文阅读中无关言语效应的认知机制探究：眼动证据 [J]. 心理学报, 2021, 53(7): 729-745. 刘志方, 仝文, 张智君, 赵亚军. 语境预测性对阅读中字词加工过程的影响：眼动证据 [J]. 心理学报, 2020, 52(9): 1031-1047. 杨帆, 隋雪, 李雨桐. 中文阅读中长距离回视引导机制的眼动研究 [J]. 心理学报, 2020, 52(8): 921-932. 高晓雷, 李晓伟, 孙敏, 白学军, 高蕾. 藏语阅读中中央凹词频效应及对副中央凹预视效应的影响 [J]. 心理学报, 2020, 52(10): 1143-1155. 周蕾, 李爱梅, 张磊, 李纾, 梁竹苑. 风险决策和跨期决策的过程比较：以确定效应和即刻效应为例 [J]. 心理学报, 2019, 51(3): 337-352. 刘璐, 闫国利. 聋人阅读中的副中央凹视觉注意增强效应——来自消失文本的证据 [J]. 心理学报, 2018, 50(7): 715-726. 李宝珠, 魏少木. 广告诉求形式对产品反馈的影响作用： 基于眼动的证据 [J]. 心理学报, 2018, 50(1): 69-81. 刘志方, 张智君, 潘运, 仝文, 苏衡. 中文阅读中预视阶段和注视阶段内词汇视觉编码 的过程特点：来自消失文本的证据 [J]. 心理学报, 2017, 49(7): 853-865.