發(fā)布時間：2017-12-28 19:28

  本文關(guān)鍵詞：聲旁家族對漢字詞匯識別的影響　出處：《西南大學(xué)》2016年碩士論文　論文類型：學(xué)位論文

  更多相關(guān)文章： 聲旁家族 語音總體激活 交互激活模型

【摘要】：漢字是一種復(fù)雜而又精密的文字,其中形聲字的比例占到半數(shù)以上,在漢字中是非常常見的。形聲字中的聲旁能夠表達漢字的語音信息,對漢字的語音總體激活有著重要的影響,在漢字詞匯識別中發(fā)揮著重要的作用。研究者們針對漢字的詞匯識別,提出了三種不同的理論模型,分別是:直通模型、語音中介模型和雙通路模型。直通模型支持直接由字形信息激活語義信息從而到達語義通達,不需要經(jīng)過語音激活的過程;語音中介模型的觀點是字形信息最先激活語音信息,再通過這一語音中介傳播興奮到語義表征,到達語義通達;而雙通路模型則認為這兩個通路可以同時存在且并行加工,每個通道都有可能激活語義,主要取決于字詞頻率、任務(wù)范式和被試水平等因素。這三種不同理論模型引起了巨大的討論,研究者們做了大量實驗去支持自己的理論假設(shè)。其中,語音在詞匯識別過程中的作用問題成為了討論的焦點之一。形聲字中的聲旁能夠提供語音信息線索,在識別漢字的過程中,這些語音信息能夠影響漢字的語音總體激活水平。研究發(fā)現(xiàn),聲旁家族對詞匯識別有可能表現(xiàn)為抑制效應(yīng),也有可能表現(xiàn)為促進效應(yīng)。交互激活模型的側(cè)抑制機制嘗試對抑制效應(yīng)做出解釋:在詞匯識別過程中,視覺刺激信息需要經(jīng)過特征水平、字母水平和詞語水平的表征,且在相鄰水平之間的傳遞是雙向的。當(dāng)靶字激活了與其自身相應(yīng)的單元即靶字單元,這些靶字單元也會激活與其具有同一聲旁的家族成員字單元,這些被激活的靶字單元和成員字單元相互競爭,相互抑制,直到最終獲得最高激活水平的靶字單元被挑選出來,獲得正確識別和通達。這意味著,當(dāng)靶字涉及的聲旁家族成員字越多,其被激活的分心單元即成員字單元也越多,受到的側(cè)抑制水平也就越高,導(dǎo)致靶字單元越難從分心單元中識別,最終反應(yīng)越慢,出現(xiàn)家族抑制效應(yīng)。Andrews的自上而下反饋理論嘗試對促進效應(yīng)做出解釋:在詞匯識別過程中,靶字單元被激活的同時,也會激活其聲旁家族成員字單元,然后這二者就會激活其亞詞匯層級的字母單元及字母串單元,這些被激活的字母及字母串單元又可以返回加強靶字單元的激活。當(dāng)靶字的聲旁家族成員越多,這些相同的字母及字母串單元的激活被不斷增強,那么其相似語音激活程度也越高,這些被激活的亞詞匯單元對靶字單元的反饋和再激活就越強,產(chǎn)生促進效應(yīng)。雖然對這兩種不同的模型存在了一定的爭論,但是這兩者之間并不是相互排斥的,都屬于識別的內(nèi)部過程,起促進作用還是抑制作用可能主要取決于這兩個加工誰占優(yōu)勢地位。本研究提出假設(shè):聲旁家族對詞匯識別的作用與某些特殊的實驗任務(wù)有關(guān)。在詞匯判斷任務(wù)和命名任務(wù)這兩種難度較小,可以依賴于相似語音的總體激活水平完成實驗的任務(wù),表現(xiàn)為促進作用,而在語義分類任務(wù)這種難度較大,無法單純依靠相似語音的總體詞匯激活水平而需要精確的單個靶字單元語義信息激活完成實驗的任務(wù),表現(xiàn)為抑制效應(yīng)。本研究采用聲旁家族作為實驗材料,選用語義分類任務(wù),詞匯判斷任務(wù)和命名任務(wù)這三個不同的實驗范式,考察聲旁在漢字詞匯識別中的作用,以探討在漢字詞匯識別中,語音總體激活在其中的作用。實驗一中,通過操控聲旁家族大小和規(guī)則性兩種條件,考查在語義分類任務(wù)中聲旁家族和漢字規(guī)則性對詞匯識別的影響。實驗采用2(聲旁家族大小:大,小)X2(規(guī)則性:規(guī)則、不規(guī)則)的兩因素被試內(nèi)設(shè)計。實驗要求被試判斷所呈現(xiàn)的刺激是否是動物字,因變量為正確反應(yīng)時和正確率。實驗數(shù)據(jù)指出:在正確率上,聲旁家族大小和規(guī)則性的交互作用顯著,在聲旁大家族條件下,與不規(guī)則字相比,規(guī)則字的正確率明顯更低;在規(guī)則字條件下,與聲旁小家族相比,聲旁大家族的正確率明顯更低。實驗結(jié)果支持了交互激活模型。實驗二是在實驗一程序的基礎(chǔ)上,刪除了序列中的一個空屏,以消除這個空屏可能對被試的反應(yīng)時間的影響而進行的實驗,其余的實驗材料和實驗任務(wù)都不變。實驗二結(jié)果與實驗一結(jié)果趨勢一致,同樣支持了交互激活模型。實驗三中,操控不含有聲旁的筆畫家族字大小,在排除語音總體激活的條件下進行詞匯判斷任務(wù),考察在詞匯判斷任務(wù)中,沒有聲旁語音線索提示的漢字加工機制,與后面的實驗進行比對。實驗為單因素(筆畫家族大小:大,小)被試內(nèi)設(shè)計,要求被試快速準(zhǔn)確地判斷所呈現(xiàn)的靶字是不是真字,記錄反應(yīng)時和正確率。實驗結(jié)果顯示,在排除語音總體激活的條件下,家族效應(yīng)表現(xiàn)為抑制作用。實驗四中,繼續(xù)采用實驗一中的聲旁家族材料,考察在命名任務(wù)中聲旁家族大小和漢字規(guī)則性對詞匯識別的作用。實驗任務(wù)要求被試快速而準(zhǔn)確地念出所呈現(xiàn)靶字,記錄反應(yīng)時和正確率。實驗數(shù)據(jù)指出:在反應(yīng)時上,聲旁家族大小的主效應(yīng)顯著,聲旁大家族的正確反應(yīng)時顯著大于聲旁小家族;規(guī)則性的主效應(yīng)顯著,規(guī)則字的正確反應(yīng)時比不規(guī)則字明顯更快。在正確率上,聲旁家族大小和規(guī)則性的交互作用顯著,具體是:在聲旁小家族條件下,與不規(guī)則字相比,規(guī)則字的正確率明顯更高,在規(guī)則字條件下,與聲旁小家族相比,聲旁大家族的正確率明顯更小。基于以上四個實驗數(shù)據(jù)結(jié)果,可以得到如下結(jié)論:聲旁家族效應(yīng)在漢字詞匯識別中表現(xiàn)為抑制效應(yīng),這種效應(yīng)在規(guī)則字條件下更明顯。聲旁家族的抑制效應(yīng)與實驗任務(wù)無關(guān)。
[Abstract]:Chinese characters is a kind of complex and precise words, the phonogram accounted for more than half, is very common in Chinese characters. The phonetic characters can express phonetic speech information Chinese characters, the Chinese characters voice overall activation plays an important role, plays an important role in the Chinese characters in Chinese word recognition. The researchers have proposed three different theoretical models for Chinese character recognition, namely, direct through model, voice medium model and double channel model. Through the model is directly supported by font information semantic information to activate semantic access, do not need to go through the process of phonological activation; phonological mediation models is the first font information activate voice information through the voice communication intermediary excitement to semantic representation, semantic access to and; dual route model is that these two pathways can be simultaneously the existence and parallel processing, each channel may activate semantics, mainly depends on word frequency, and test the factors equal task paradigm. The three different theoretical models have caused great discussion, and the researchers have done a lot of experiments to support their own theoretical assumptions. Among them, the role of speech in the process of vocabulary recognition has become one of the focus of discussion. The characters in the phonetic speech information can provide clues, in the process of identifying Chinese characters in the speech information can affect the overall level of Chinese characters voice activation. The study found that the phonetic family has showed an inhibitory effect on word recognition, are also likely to show effect. The lateral inhibition mechanism of interaction activation model attempts to explain the inhibitory effect: in the process of word recognition, visual stimulus information needs to be characterized by feature level, letter level and word level, and the transmission between adjacent levels is bidirectional. When the target word activates its corresponding unit that is the target word units, members of the family unit unit will also activate these target words and have the same sound, activation of these units and members of the target word word units compete with each other, mutual inhibition, until finally obtain the highest level of activation of the target word unit is selected to obtain the correct recognition and understanding. This means that when more members of the family of the target word phonetic word relates to the distraction activated member unit unit more, also the higher the level of the lateral inhibition, lead to more difficult to identify the target word unit distracted from the unit, the final reaction is slower, the inhibitory effect of family. Andrews top-down feedback theory attempts to explain the facilitation effect in word recognition process, the target word unit is activated at the same time, will also activate the phonetic word family unit, then the two will activate the sublexical level letter letter string unit and unit activation activation of these letters and letter string unit it can return to the target word units to strengthen the. The more radical family members when the target word, activate the same letters and letter string unit is increasing, so the similar voice activation degree is higher, activation of these sub lexical unit feedback on the target word unit and then activate the stronger, promoting effect. Although there are some disputes about the two different models, they are not mutually exclusive. They all belong to the internal process of recognition. The promotion or inhibition is likely to be decided by those two processes. This study put forward the hypothesis: the effect of phonetic family on word recognition related to some special tasks. The judgment task and the naming task two less difficulty in vocabulary, can depend on the overall similarity of voice activation level to complete the experiment task, performance for the role, and the larger the difficulty in semantic classification task, can not simply rely on similar overall vocabulary speech activation level and single target word semantic information unit accurate activation completed the experimental task, showed an inhibitory effect. This study used the phonetic family as experimental material, using a semantic categorization task, lexical decision task and naming the three different experimental paradigms, the effects of sound on the Chinese characters in Chinese word recognition, in order to explore Chinese characters in Chinese word recognition, voice activation plays a role in general. In the first experiment, through two kinds of conditions and rules of phonetic manipulation of family size, the effect of sound on the semantic classification task in family and Chinese characters rules on word recognition. The experiment adopted 2 (phonetic family size: large, small (X2) Rules: regular and irregular) two factors within subjects design. The experiment asked the subjects to judge whether the stimuli presented were animal words, and the variables were correct and correct. Experimental data show that: in the correct rate, the interaction of family size and the phonetic rules significantly, around family conditions, compared with the irregular word, the correct rate was significantly lower in the rules of words; word rule conditions, compared with the sound small family, large family of phonetic correctness rate is significantly lower. The experimental results support the interactive activation model. Experiment two, based on the first experiment, deleted an empty screen in the sequence to eliminate the influence of the empty screen on the response time of the subjects. The rest of the experimental materials and tasks were unchanged. The result of experiment two is consistent with the result of experiment one, and it also supports the interactive activation model. Experiment 3, control does not contain the phonetic word for stroke family size, lexical decision task in the exclusion of speech overall activation conditions, effects in lexical decision task, no phonetic cues of speech Chinese characters processing mechanism, compared with the following experiment. The single factor (family size: large, small strokes) within subjects design, subjects were required to rapidly and accurately determine the target word is not true, recording the reaction time and correct rate. The experimental results show that the conditions for eliminating the overall activation of the speech
【學(xué)位授予單位】：西南大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：B842.2

【相似文獻】

相關(guān)期刊論文 前10條

1 方愛鄉(xiāng);中日漢字詞匯趣談之三[J];日語知識;2000年07期

2 張侃,吳美榮;非漢字詞的使用與規(guī)范研究[J];江西社會科學(xué);2004年07期

3 曹s，

本文編號：1347155