2. Literature Review
2.6 Summary
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
42
that this difference may result in different lexicon network, which may make priming overextension have different impact on children. After all, English words mostly carry phonological information while Chinese words are often more than that. Thus, the category hint in Chinese should be more than English, which may cause children less vulnerable to making errors.
2.6 Summary
In the literature review section, we reviewed three types of errors. The first one is category errors. Category errors happen when children still do not find identical features as adults to define a concept. They just use one or two features so that they will overextend several things at the same time. The second one is pragmatic errors.
Pragmatic errors mean that children use some known words to replace some unknown words. This happens because children still do not have sufficient vocabulary in their very young age. Therefore, they use some other related words which are similar with target words for children to make a substitution. Besides, we also introduce another distinct but related phenomenon: metaphors. Metaphors happen when children want to express the similarity between two objects. Since the features they base on to make metaphors are often perceptual similarity, it is hard to distinguish metaphors from overextensions. The criterion to distinguish them is that children will abandon the overextending usage when they have acquired the correct word while the usage of
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
43
metaphors will be retained. Finally, the third one is retrieval errors. Retrieval errors occur when children have acquired a certain word but they use another word to name that object due to some retrieval problems. The reasons often come from the effect of previously retrieved words or children‟s insufficient practice for that word. Besides, the issue whether children‟s overextensions come from their comprehension or production problems is also reviewed. Generally, researchers all think that children‟s performance errors do not necessarily reflect that they also have errors in their comprehension. In fact, children often have better comprehension than their production. And then we reviewed the shape bias. Shape is very crucial in children‟s lexical development. It is more effective than objects‟ function or concepts for children‟s category learning. It also has relation with the quantities of nouns which children have acquired. Besides, English syntactic cues seem to direct children‟s attention to objects‟ shape. And Gershkoff-Stowe et al.‟s (2006) study is reviewed.
Gershkoff-Stowe et al. (2006) thought that the underlying mechanism of these three types of errors may be the same. They may be all related to the spreading activation mechanism and can be explained unitarily. And they conducted two experiments in their study. They found that the underlying mechanisms of three error types should be the same. Besides, recent activation and perceptual similarity would affect children‟s naming. Older children were less susceptible to previously retrieved words than
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
44
younger children. Finally, the characteristics of Chinese are introduced. The characters in Chinese often represent not only phonological units but also semantic units. The meaning of a compound often can be understood through its components.
Under the study, we will base on Gershkoff-Stowe et al. (2006) work, follow the similar experimental design and procedure, and adopt the similar coding system to test whether Taiwan Mandarin children will yield the similar results or not.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
45
Chapter 3 Methodology 3.1 Experiment 1
The experiment design generally followed the work in Gershkoff-Stowe et al.
(2006).
In experiment 1, we first show three pictures to children, including one target priming picture, and two fillers. And they will be shown a 3-dimensional test object, and asked “What is this?” as shown in figure 3.
Picture naming
target prime filler filler Object naming
lollipop
Figure 3. The Demonstration of Object Naming Task
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
46
In figure 3, we can clearly see the stimuli that children will see during one of the trial in the experiment. In the picture naming phase, yan3jing4 „glasses‟ and wa4zi5 „socks‟
are two fillers that are not related to the test object. And the target priming picture is qiu2pai1 „racket‟, which is similar with the test object bang4bang4tang2 „lollipop‟ in
the way that they both have one big circle and one stick under that circle.
Target priming pictures are perceptually similar to the test objects in order to examine whether this picture will affect children‟s word selection. Thus, experiment 1 is designed to have two sets: set A and set B. The fillers in both sets were the same, but the target priming pictures were different. Both of the target priming pictures in set A and B are perceptually similar to test object. Experiment is designed this way for confirming that priming will affect children‟s word selection. If this is true, the error naming in the first and second set will differ. For example, in the figure 3, the target priming picture is qiu2pai1 „racket‟. And we will have another target priming picture in the other set jing4zi5 „mirror‟ which also shared the criterial perceptual features with the test object. In this way, if the answers children offer in the two sets differ, namely, children give qiu2pai1 „racket‟ in one set, and jing4zi5 „mirror‟ in the other, we can corroborate that priming task indeed will affect children‟s word selection.
Test objects in this experiment were composed of two categories. One was familiar items and the other was unfamiliar ones. The purpose of experiment 1 was to
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
47
examine if priming has the same effect for children on those familiar items with known names and unfamiliar items with unknown names. If this is true, this could imply that retrieval errors, which involve known lexicons, have the same underlying mechanism with pragmatic and category errors which involve unknown lexicons.
3.1.1 Procedure
Every child would undergo six trials in the experiment 1. Each trial was composed of two phases. The first phase is picture naming as priming, during which children would see one target priming picture, and then two fillers one by one as mentioned above. And the experimenter would ask children zhe4shi4she2me5 „what is this‟ after showing each picture to them. If children offer an answer, the next pictures would show to them until three pictures in a trial have shown up. If not, the experimenter would tell them the name of the picture and ask children to repeat it.
The second phase is object naming, during which the experimenter would present children with a test object and ask them to label it. If children do not reply, then the experimenter would ask the question again and wait for the answer for around ten seconds. If children still keep silent, we would proceed to the next trial. After children completed all the six trials, they were rewarded by some candies for their cooperation in the experiment. All the processes in the experiment were recorded and we also wrote down the responses which children gave in order to prepare for the subsequent
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
48
analysis.
About the order of three priming pictures, their showing sequences were not fixed, and were arranged to appear equally in the first, second, and third position.
3.1.2 Subjects
Twenty two children participated in this experiment. Their age ranged from 2;0 to 2;11 (mean=2;5). They were all normal children with no development delay or brain damages and their first languages were all Mandarin Chinese. They were recruited from six kindergartens, one church and several personal contacts in Taipei.
3.1.3 Materials
The stimuli in the experiment 1 included 12 test objects, and 24 priming pictures.
Test objects were all 3-dimensional and realistic items or toys in the true world.
Twenty four priming pictures consisted of 12 fillers and 12 target priming pictures.
Every picture was realistic again and colorful, and presented on card to every children.
The contents of 12 fillers were as follows: spoon, broom, fork, bottle, banana, truck, dog, plane, faucet, remote control, glasses and socks. They were all perceptually different from the test objects so that they served as fillers rather than target primes.
The pictures of twelve fillers shown to children were shown in the appendix.
The target priming pictures were designed to have two sets: set A and set B.
Pictures in both sets had similar shape with the test objects according to adults‟
‧
judgments which would be shown in the next section. Experiment had two conditions.
Half of the children would participate in Familiar Test Object Condition, and half would be in Unfamiliar Test Object Condition. Thus, of the 12 test objects, 6 were familiar things to children, and the other 6 were unfamiliar to them. In each trial, three priming pictures, including two fillers and one target prime, were paired with one test object, either familiar or unfamiliar depending on which conditions that children were assigned to. The test objects and target primes used in the experiment one are shown in the following table.
Table 3. The Stimuli in the Experiment 1
Test objects Prime
Children would be shown two fillers, and one target prime either from set A or set B
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
50
in one trial. For example, in the Familiar Test Object Condition, a child would be primed either by he2zi5 „box‟ or gu3 „drum‟, and two fillers, and finally shown a test object dan4kao1 „cake‟. On the other hand, if children participate in the Unfamiliar Test Object Condition, they would be primed by he2zi5 „box‟ or gu3 „drum‟, and two fillers, and finally shown a test object xiao1qian1bi3ji1 „pencil sharpener‟. Besides, in both conditions, children were further divided into two parts. One was assigned to Set A, and the other was to set B to examine whether priming task would indeed affect children‟s word selections. Likewise, for the unfamiliar condition, the situation was the same. Namely, children assigned to this condition half was shown a target prime from set A, and half was from set B. And all of them were shown two fillers, and finally saw an unfamiliar test object. To sum up, in this experiment, each child was presented with 12 fillers, 6 target primes either from set A or set B, and 6 test objects (familiar or unfamiliar). The pictures were grouped into six triads, and each included two fillers and one target prime.
3.1.4 Coding
All the answers which children gave were classified for the subsequent analysis.
And the classification followed Gershkoff-Stowe et al‟s (2006) study. The answer types were categorized as follows:
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
51
(1) Correct answers:
Correct answers include all the specifically correct and some unspecific but still correct answers. Specifically correct answers were those responses corresponding to most adults‟ expectations. And unspecific but still correct answers refer to some inexact naming. Sometimes children used a general name to include the object they saw. For example, when children see tian2tian2quan1 „donut‟, both tian2tian2quan1 „donut‟ and mian4bao1
„bread‟ should be counted as the correct answer type although the former one
are more specifically correct. Other correct responses included some naming which was not a direct name for that object. Instead, children just used some words to modify the objects they saw. For example, a child named Ling2dang1 „bell‟ as ding1ding1dang1 „jingles‟. This is not an overextension
error because children just used other conceptually related ways to label the objects so that this would be counted as „correct‟ type as well.
(2) Omission:
“Omission” means that children did not give any answer. They just omitted it.
(3) Erroneous answers:
A. Extraneous:
“Extraneous” means the response which children gave was not related to the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
52
pictures in the priming task in a trial. Children‟s labeling was not based on previously primed items. Rather, they may be based on other cues, such as function, shape, color or so on.
B. Control:
“Control” means the response that children gave did not show up in the
priming task but show up in the other set of priming task. For example, if one child in prime set A named gua4shi4 „hanging ornament‟ as qiu2 „ball‟, this should be a “control” error because he/she did not be primed by qiu2 „ball‟.
Qiu2 „ball‟ actually should be the target prime in set B.
C. Target:
“Target” means that the erroneous naming came from the target prime. For
example, ling2dang1 „bell‟ was named as qiu2 „ball‟ since the target prime in the priming task was qiu2 „ball‟.
D. Filler:
“Filler” refers to the response which was affected by the fillers rather than
target prime.
3.1.5 Adults’ similarity judgments
Before the object naming task experiment was conducted to children, 10 adults had participated for similarity judgments among 12 test objects and 24 prime pictures
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
53
which were composed of 12 fillers and 12 target primes. Similarity scales ranged from 1 to 10. Scale 1 means that two things were low similarity while 10 means high similarity. Adults were asked to grade the similarity using the 10 points scale. Five of ten were assigned to familiar condition judgments and the other five were assigned to unfamiliar condition judgments. The result showed that in the familiar condition, the mean of adults‟ rating for the similarity of target primes and objects was 8.08, and the similarity of fillers and objects was 1.31. According to ANOVA, for the shape similarity to the objects, target primes and fillers were significantly different, F(1,118)=1129.01, p<0.001. On the other hand, in the unfamiliar condition, the
similarity rating of target primes and objects was 8.38, and the similarity of fillers and objects was 1.33. The similarity of target primes to the objects and fillers to the objects in unfamiliar condition also reached significance, F(1,118)=909.20, p<0.001.
And for the comparing of two conditions (familiar vs. unfamiliar), it did not differ significantly, F(1,238)=0.11, p=0.74>0.05.
3.1.6 Word comprehension
After children have finished the naming tasks in the experiment 1, they would continue the word comprehension task. In this task, they had to index which object the experimenter was asking. For example, the experimenter would ask children na3yi2ge5shi4dan4gao1 „Which one is the cake?‟, and children had to point out
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
54
which one is the cake. For the comprehension task, children had to choose one answer from three objects, in which one was the target answer, and the other two were the distracters. And all the target objects in this task were the test objects which were presented to children in the previous experiment. All of the objects shown to children in word comprehension task are presented in the appendix.
3.2 Experiment 2
In the experiment 2, the effects of previously retrieved words and perceptual similarity on children‟s naming were compared between 2- and 4-year-old children.
The design of experiment 2 followed again the study in Gershkoff-Stowe et al. (2006), but we modified the priming pictures. There were three groups in this experiment.
One was pictures with high similarity to test objects, another was pictures with low similarity, and the other one was unrelated pictures to the test objects. The third group acted as a control group to compare and contrast with the first and second groups to see the effects on children‟s naming when there is no perceptual similarity between test objects and priming. We expect that children who participated in this condition will produce more extraneous answers since they are not primed by related concepts.
And high and low similarity priming were used to examine how the priming and similarity affect children‟s naming behaviors.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
55
In this experiment, novel objects were used. The novel objects which we used in this experiment were from the study of Gershkoff-Stowe et al. (2006). And for the target names in high-similarity and low-similarity conditions were decided from adults‟ imagination and judgments, which would be illustrated in the section 3.2.3.
The purpose of using novel objects in this experiment was to test the differences of word choosing between 2- and 4-year-old children. We expected 4-year-old children would be less susceptible to previously retrieved words since it‟s reasonable that 4-year-old children have more mature vocabulary and concept base than younger children, they may have better performances toward naming.
3.2.1 Procedure
Children at each age level were evenly distributed to three sets. All the procedures in this experiment were the same as the experiment 1. The only difference between these two experiments was that the experiment 2 had eight trials rather than six. And after the end of the experiment, there was no comprehension test for objects since there is no real name for the novel objects in this experiment.
3.2.2 Subjects
Twenty two 2-year-old (mean=2;5) and 25 4-year-old ranged from 3;11 to 4;11 (mean=4;4) children participated in this experiment. They were all normal Chinese-speaking children and had no brain damage before. Two-year-old children
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
56
were the same as the experiment 1. And 4-year-children were recruited from six kindergartens in Taipei.
3.2.3 Materials
The novel test objects followed the study of Gershkoff-Stowe et al. (2006). And all of the priming pictures were modified. They were decided from adults‟
imagination and similarity judgments.
Ten adults participated in the imagination task. Eight novel objects were shown to them. And they were asked to write down what they thought of based on shape similarity when they saw these novel items. Each adult was asked to write down two to three items they could think of and these items must have some perceptual similarity with those novel objects presented to them. After 10 adults had completed, we collected all the answers they had provided. After that, the corresponding pictures of those objects were found. By comparing with the target object, we thought of some unrelated items by our intuition for ready to be 16 fillers and unrelated targets in the experiment 2. And all the unrelated and related items were judged by other 10 adults for their similarity to target objects. These 10 adults did not participate in the imagination task before. In the adult judgment task, adults were shown a sheet full with many pictures. They were asked to grade the similarity between the target objects and the pictures we had chosen. The scale was ranged from 1 to 10. Scale 1
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
57
represented the lowest similarity and 10 represented the highest. These 10 adults were asked to mark a point which would show the similarity between the target object and the judged item. After grading, the total points of each item were counted. The one which receives most points was the item used in the high similarity condition, the one receiving lower points was the item used in the low similarity condition, and the ones receiving least points were the items used in the unrelated control and fillers. The mean rating of high similarity items was 6.73 (S.D=2.64). The mean rating of low similarity items was 4.35 (S.D=2.28), and the mean of unrelated items was 1.09
represented the lowest similarity and 10 represented the highest. These 10 adults were asked to mark a point which would show the similarity between the target object and the judged item. After grading, the total points of each item were counted. The one which receives most points was the item used in the high similarity condition, the one receiving lower points was the item used in the low similarity condition, and the ones receiving least points were the items used in the unrelated control and fillers. The mean rating of high similarity items was 6.73 (S.D=2.64). The mean rating of low similarity items was 4.35 (S.D=2.28), and the mean of unrelated items was 1.09