Development of lexical tone awareness in Chinese children with and without dyslexia

Contemporary Educational Psychology 49 (2017) 203–214

Contents lists available at ScienceDirect

Contemporary Educational Psychology journal homepage: www.elsevier.com/locate/cedpsych

Empirical study

Development of lexical tone awareness in Chinese children with and without dyslexia Li-Chih Wang a,⇑, Duo Liu a, Kevin Kien-Hoa Chung b,a, Hsien-Ming Yang c a

Department of Special Education and Counselling at The Education University of Hong Kong, Hong Kong Department of Early Childhood Education at The Education University of Hong Kong, Hong Kong c Department of Special Education at National University of Tainan, Taiwan b

a r t i c l e

i n f o

Article history: Available online 10 February 2017 Keywords: Lexical tone awareness Chinese dyslexia Primary school ages

a b s t r a c t This paper describes two studies that examined the lexical tone awareness of Chinese children both with and without dyslexia at different primary school ages. Study 1 examined the contributions of lexical tone awareness to distinguish children with and without dyslexia with respect to their Chinese character reading skills. Two hundred and seventy Chinese children participated in Study 1. Ninety of these were children with dyslexia (equally recruited from second, fourth, and sixth grades). Moreover, ninety children functioned as a chronological-age control group, and an additional ninety children functioned as a reading-level control group. The participants were tested for nonverbal intelligence, Chinese character reading, and cognitive-linguistic skills and lexical tone awareness. Our results revealed a later developmental ceiling in Chinese children with dyslexia than in those without dyslexia. Furthermore, children’s lexical tone awareness could serve to distinguish children with dyslexia from typically developing children in all primary school years. Study 2 compared the lexical tone awareness and Chinese character reading skills of Chinese children with dyslexia both before and after introducing the Perceptual Training Method. The participants in this study consisted of all the participants with dyslexia from Study 1, and the measurements were the Chinese character reading test and the lexical tone awareness task from Study 1. Our results revealed that only second-grade children with dyslexia gained substantially from the training on both lexical tone awareness and character naming, whereas those in the fourth grade obtained a significant improvement only on lexical tone awareness. Ó 2017 Elsevier Inc. All rights reserved.

1. Introduction Developmental dyslexia is characterized by difficulties in word reading and spelling despite normal intelligence and adequate formal education (Catts & Kamhi, 2005; Chung, Ho, Chan, Tsang, & Lee, 2010). Much research on dyslexia demonstrated that it involves multiple cognitive-linguistic deficits; the majority of these studies focused on phonological processing, including phonological awareness, rapid automatized naming (RAN), and verbal short-term memory (STM) in most languages tested thus far (e.g., Ellis, 2014; Melby-Lervåg, Lyster, & Hulme, 2012; Wagner & Torgesen, 1987). However, in Chinese, apart from phonological awareness, RAN, and verbal STM, deficits in the awareness of another component in phonological processing,

⇑ Corresponding author. E-mail addresses: [email protected] (L.-C. Wang), [email protected] (D. Liu), [email protected] (K.K.-H. Chung), [email protected] (H.-M. Yang). http://dx.doi.org/10.1016/j.cedpsych.2017.02.002 0361-476X/Ó 2017 Elsevier Inc. All rights reserved.

namely, lexical tones, which are pitch patterns that signify differences in the meanings of monosyllabic words with identical phonetic segments in Chinese (Tong, Tong, & McBride-Chang, 2015), have also been linked to problems in Chinese character reading (e.g., Cheung et al., 2009; Li & Ho, 2011; Zhang et al., 2012). Given that Chinese is a tonal language, lexical tone plays a major role in conveying differences in the meanings of Mandarin Chinese words. In Chinese, lexical tones are pitch patterns that influence different levels of language processing and are particularly related to linguistic domain (Gandour, 1978)—ranging from segmental analysis of words to suprasegmental processing of sentence structure. In natural speech, lexical tones assist listeners in distinguishing the meanings of words with identical phonetic segments and constrain syntactic interpretation. However, to date, few attempts have been made to explore the relationships between lexical tone awareness and the awareness of other components in phonological processing (i.e., phonological awareness, RAN, and verbal STM) in explaining individual differences in reading—despite the fact that reading

204

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

has been often viewed as an execution and integration of multiple cognitive-linguistic components (Kendeou & Trevors, 2012; van den Broek & Espin, 2012; Ziegler & Goswami, 2005). To bridge this gap, a focus on lexical tone awareness would be of substantial benefit in the study of Chinese character reading among Mandarinspeaking children in Taiwan. Thus, the present paper aimed to examine the development of lexical tone awareness of Chinese children with and without dyslexia at different primary school ages. 1.1. Lexical tone awareness and its relationship with reading in Chinese Lexical tone is a suprasegmental feature of the Chinese language linked to a syllable. Compared to phonemic or segmental levels of speech, including isolated vowels and consonants, features at the suprasegmental level cannot be easily analyzed as distinct segments but rather are viewed as belonging to a syllable or word, for example, prosody in English and lexical tone in Chinese (Silverman et al., 1992; Zhang & McBride-Chang, 2010). In alphabetic languages such as English, the awareness of stress in words across stress-focused phrases is a type of suprasegmental level feature, as is lexical tone in tonal languages such as Chinese (Chung, McBride-Chang, Cheung, & Wong, 2013). There are four lexical tones in Mandarin Chinese, namely, Tone 1: high level, Tone 2: high rising, Tone 3: falling-rising, and Tone 4: high falling. A syllable with different tones can represent different meanings. For example, the syllable /xi/ can represent four different meanings with four different tones: /xi1/ 西 (Western), /xi2/ 習 (study), /xi3/ 喜 (like), and /xi4/ 細 (thin). Although there are no symbols that indicate tone in a character, an awareness of tone is important to distinguish the meaning of a syllable to identify a correct Chinese character in its contextual use, and it helps readers to distinguish homophones (e.g., both 喜 [like] and 洗 [wash] are pronounced as /xi3/) (Tong et al., 2011) and find the relationship between the pronunciation of the whole character and the phonetic radicals that convey different tones (e.g. the phonetic radical of 嬉 /xi3/ is 喜 /xi3/) (Chen, Shu, Wu, & Anderson, 2003). Considering the functions of lexical tone reviewed above, being highly familiar with lexical tone information may enhance readers’ accuracy in reading Chinese characters and learning new Chinese vocabulary (e.g., Tong et al., 2015). This view is supported by a few studies that reported the unique contribution of lexical tone awareness to reading performance with statistically controlled reading-related abilities such as phonological awareness and RAN (e.g., McBride-Chang, Tong, et al., 2008; Shu, Peng, & McBrideChang, 2008; Tong et al., 2015). The relationships between lexical tone awareness and reading were shown to be even stronger than the association between phonemic awareness and reading in some studies (e.g., McBride-Chang, Lam, et al., 2008). Given the importance of lexical tone awareness to Chinese reading, those faced with difficulties in reading might also perform worse on lexical tone awareness. Thus far, lexical tone awareness has been found to be important in distinguishing Chinese children with and without dyslexia in a limited number of studies. For example, Cheung et al. (2009) found that children with dyslexia with a mean age of ten years performed significantly worse than their chronological-age (CA) control group, although not worse than a reading-level (RL) control group, on a lexical tone awareness task in which a speech continuum consisting of 5 tokens of the Cantonese syllable /xi/ realized at different frequency ranges was constructed. Similar to Cheung et al. (2009), Wang, Huss, Hämäläinen, and Goswami (2012) found that Chinese children with dyslexia with an average age of 9.9 years showed significantly

poorer performance on awareness of lexical tone than did control groups with similar reading levels (RL control group) and similar chronological ages (CA control group). However, the group differences found by Cheung et al. (2009) and Wang et al. (2012) were not consistent with Li and Ho (2011). Although Li and Ho (2011) also found that Chinese children with dyslexia performed significantly worse than a CA control group in lexical tone awareness, no such difference was found between the group with dyslexia and an RL control group. Therefore, even though poor lexical tone awareness of Chinese children with dyslexia has been found consistently, there is still some controversy over whether a significant difference exists between lexical tone awareness of children with dyslexia and their RL group. One of the most likely reasons for the inconsistency described above is the varying ages of the participants in the different studies. The participants with dyslexia in the studies of Cheung et al. (2009) and Wang et al. (2012) were approximately ten years old (i.e., Mean age = 10;5 and 10;0), while the RL groups in these two studies were approximately eight years old (i.e., Mean age = 8;11 and 8;6). However, in the study by Li and Ho (2011), the children with dyslexia were only approximately eight years old (i.e., Mean age = 8;11), and they used an RL group with a mean age of approximately seven (i.e., Mean age = 7;4). Considering the rapid development of brain executive functions before the end of primary school (Anderson, 2002), a difference of two years at primary school ages may cause a large developmental gap in many aspects, which might be the reason for the inconsistent findings above. To date, only a few studies have demonstrated the developmental trajectory of lexical tone awareness in typically developing children. For instance, Ciocca and Lui (2003) compared children who were four, six, and ten years of age and adults on lexical tone awareness. They found that the development of lexical tone awareness did not extend beyond approximately ten years of age (i.e., the lexical tone awareness performance of ten-year-old children was not significantly different from that of adults), and their results have also been supported by other studies (e.g., Sze, 2004). The findings of developments in lexical tone awareness described above are in line with the development of auditory temporal processing, which is considered to be an underlying mechanism of lexical tone awareness (Zhang & McBride-Chang, 2010). For auditory temporal processing, the previous evidence in alphabetic languages showed that typically developing children tend to reach the developmental ceiling at approximately nine years old, while children with dyslexia tend to reach levels similar to those of typically developing children by ten to eleven years of age (e.g., Hautus, Setchell, Waldie, & Kirk, 2003). This view is supported by recent studies conducted in Chinese (e.g., Wang & Yang, 2016). Almost all studies on the role of lexical tone processing in Chinese reading have been conducted in Cantonese, which contains six contrastive lexical tones (Matthews & Yip, 1994). Although both Mandarin and Cantonese are tonal languages and are dialects of Chinese languages (Peng, Zhang, Zheng, Minett, & Wang, 2012), the lexical tone systems in Mandarin and Cantonese are diverse, which may obstruct the direct application of the results concerning the development of lexical tone awareness to the Mandarin population. Therefore, further investigation on the increasing trend in the development of lexical tone awareness as age increases is necessary for Mandarin. In addition, the strong associations between phonological processing and Mandarin/Cantonese lexical tone awareness at different ages has been revealed by previous studies (e.g., Chen, 2001; McBride-Chang, Tong, et al., 2008; Shu et al., 2008; Tong et al., 2015; Xu, 1991), and it has also been found that the development of lexical tone awareness may be a result of the growth in metalin-

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

guistic abilities, including phonological processing abilities (e.g., Lin, Wang, & Shu, 2013). Therefore, it is reasonable to take phonological processing abilities into consideration when examinations of lexical tone awareness are undertaken. However, because few, if any, studies have demonstrated the relationships of Mandarin lexical tone awareness and all other phonological processing abilities (phonological awareness, verbal STM, and RAN) of children with and without dyslexia across primary school ages, the examination of these associations was a prerequisite for the further comparisons in this study. Furthermore, comparisons of lexical tone awareness in children with and without dyslexia at primary school ages after controlling for phonological processing skills could enhance our understanding of whether children with dyslexia also have developmental delays in lexical tone awareness—similar to their delays in other phonological processing skills—and, if so, whether they could reach similar levels of lexical tone awareness as those without dyslexia by their late primary school years. Taken as a whole, the literature reviewed here reveals some significant gaps that call for additional exploration. It is worthwhile to enhance our knowledge of the function of lexical tone awareness to distinguish children with dyslexia from younger, typically developing children with similar reading levels despite the fact that the results of most studies have shown that the lexical tone awareness of children with dyslexia is significantly worse than their peers with similar chronological ages. 1.2. Intervention on lexical tone awareness In addition, to further examine the causal role of lexical tone awareness to reading at different primary school ages, it is necessary to introduce an intervention that can effectively improve lexical tone awareness. Moreover, a significant predictor may not be an effective factor during intervention. For example, children with dyslexia performed significantly worse on both phonological awareness and RAN than those without dyslexia. However, an intervention based on phonological awareness improved reading, whereas training in RAN did not, even though RAN is a predictor of reading (Hulme & Snowling, 2013). Therefore, the present study also aimed to test whether the impaired lexical tone awareness of children with dyslexia at different primary school ages could be improved using an experimental study design. To date, a few effective teaching methods have been proposed and demonstrated that focus on improving students’ lexical tone awareness. The Perceptual Training Method is one of these effective methods. The positive effects of the Perceptual Training Method on lexical tone awareness have been demonstrated by several studies (e.g., Cooper & Wang, 2012, 2013; Francis, Ciocca, Ma, & Fenn, 2008; Wang, Spence, Jongman, & Sereno, 1999); the targets included both native younger children and second language learners. The Perceptual Training Method can be considered as two distinct teaching procedures, namely, identification and same/different discrimination training (Wayland & Li, 2008). During identification training, participants hear a single stimulus in each trial and are asked to identify the stimulus in terms of two categories. In contrast, during same/different discrimination training, participants hear two stimuli in each trial and are asked to determine whether those two stimuli are instances of the same category or examples of two different categories (Wayland & Li, 2008). However, because the same/different discrimination training procedure was considered more attractive and effective (Morosan & Jamieson, 1989), the present study included only the same/different discrimination training procedure from the Perceptual Training Method to examine the role of lexical tone awareness in children with dyslexia at different primary school ages.

205

1.3. Research goals Thus, the goal of the present study was to examine the lexical tone 1 awareness deficits of Chinese children with dyslexia in primary schools. In total, three specific goals were addressed. The first research goal was to examine whether there were significant relationships between lexical tone awareness and other phonological processing abilities (phonological awareness, verbal STM, and RAN) in Chinese children with and without dyslexia at three primary school ages (Study 1). The second research goal was to examine whether dyslexic children have lexical tone awareness deficits compared to agematched or reading-matched groups, and, if so, whether the deficit pattern is affected by age (Study 1). The third research goal was to examine whether lexical tone intervention could improve lexical tone awareness and character reading among dyslexic children and whether the intervention effects are moderated by grade (Study 2). A comparison before and after introducing the same/different discrimination training procedure of the Perceptual Training Method on lexical tone awareness and character reading was conducted.

2. Study 1 2.1. Methods 2.1.1. Participants Two hundred and seventy Mandarin-speaking, Mandarinreading children participated in this study. Ninety of these (30 participants with an average age of 8.11 years; 30 participants with an average age of 9.97 years; and 30 participants with an average age of 11.86 years) were children with dyslexia. These children were referred by resource classrooms from primary schools in Taiwan. In the present study, all eligible children with dyslexia had a standard score with a standard deviation below 1.5 on the Chinese Character Recognition Test (Huang, 2001) and scored in the last 25th percentile of achievement in the subject of Chinese but had normal intelligence (an IQ score greater than 85). The criteria for dyslexia in the present study also involved normal or correctedto-normal vision and hearing and the absence of attention deficit disorder with hyperactivity. Furthermore, considering the possible influence of hearing problems to the outcome of this study, the medical histories of the hearing conditions of the children with dyslexia were carefully taken into consideration by questioning their parents and teachers. In contrast, ninety typically developing children (30 participants with an average age of 7.87 years; 30 participants with a mean age of 10.13 years; and 30 participants with an average age of 12.02 years) were sampled from primary schools and matched to the children with dyslexia on IQ, gender, and age. These children achieved a standard score with a standard deviation above 0.8 on the Chinese Character Recognition Test (Huang, 2001) (i.e., this was the CA control group). Another ninety children (30 participants with an average age of 6.07 years; 30 participants with a mean age of 7.91 years; and 30 participants with an average age of 10.22 years) also consisted typically developing children. The youngest group (average age of 6.07) was selected from kindergartens. Their teachers and parents reported that their performances were appropriate to their chronological ages. Because there is no suitable standardized measurement of character reading for this group, teachers’ and parents’ observations (via interviewed both of them) were used to confirm that the children showed no evidence of delay. These ninety children were matched with the children with dyslexia on IQ, gender, and character reading performance (i.e., this was the RL control

206

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

group). Table 1 shows the descriptive data for all participants in Study 1. 2.2. Materials The participants were administered both standardized tests and researcher-created tasks. Four standardized tests were used, including one nonverbal intelligence measurement, one character-reading measurement, and two cognitive-linguistic skills measurements. Participants were also evaluated on two researcher-created tasks that included a lexical tone awareness task and a verbal STM task. Participants were tested individually over a few sessions. Within each session, tasks were counterbalanced across participants. The details of these measures are as follows: 2.2.1. Nonverbal IQ test Raven’s Progressive Matrices–Parallel These standardized tests of nonverbal IQ were used to represent the participants’ nonverbal IQ in this study. For the participants aged under eight years old, Raven’s Colored Progressive Matrices–Parallel Form, which contains 36 items, was used, while Raven’s Standard (and Colored) Progressive Matrices–Parallel Form, which contains 60 items, was used to test the participants aged eight years or above. The items in both tests are arranged in increasing difficulty. Each item consists of a visual target matrix with one missing part. The children are asked to select from six to eight alternatives to fill a missing patch in a visual matrix. Nonverbal IQ was estimated based on the Taiwanese norm established by the Chinese Behavioral Science Corporation in 2006 (Chen & Chen, 2006). Based on our sample, Cronbach’s a coefficient was 0.78. 2.2.2. Character reading measure The Chinese Character Recognition Scale, developed by Huang (2001), is a standardized test in which participants read aloud or write the pronunciations of visually presented Chinese characters in the list. During the test, the participants were asked to utter or write the sounds as quickly as possible; the teachers do not provide cues or hints to the participants. The test arrangement includes 20 lines; each line includes 10 Chinese characters, totaling 200 Chinese characters. The applicable targets of this test are first to ninth graders. Considering the applicable for huge sample size in this study, the participants were given a seven-minute time limitation. Based on our sample and procedures, Cronbach’s a coefficient was 0.75.

2.2.3. Lexical tone awareness task The lexical tone awareness task was designed using the oddity paradigm (Deacon & Kirby, 2004). This paradigm has frequently been used in designing lexical tone tasks in previous studies (e.g., Li & Ho, 2011; Lin et al., 2013). In each trial, participants are asked to choose which of three orally presented Chinese characters differ from the others in terms of their lexical tones; there are no visually presented materials in this task. To consider the difficulties of different items in each trial, the character frequencies were balanced based on reports from Ministry of Education in Taiwan. There are 60 total trials in this task, and the syllables of the three items in each trial, including onset and rhyme, are entirely distinct from one another. In this study, the answers to all trials were equally divided into four lexical tones and presented in a mixed fashion to avoid boring the participants. Later, the answers were extracted by the authors from recordings made during the trials. This test is an individual test, and all stimuli are real words that exist in Zhuyin Fuhao in Taiwan. Three approaches were used to ensure task reliability and validity: Cronbach’s a coefficient, re-reliability, and criterion validity. Cronbach’s a coefficient was 0.94, which was tested using thirty third graders. Re-reliability was 0.90, tested twice using thirty third graders over a four-week period. The task criterion used the periodic test scores for Chinese language in the school, and the criterion validity was 0.68. 2.2.4. Phonological awareness test Tzeng, Chen, and Hsieh (2006) developed the Phonological Awareness Test Battery (screening) to test the phonological awareness of students in primary and secondary school. Considering that another independent and more comprehensive task was designed to test the participants’ lexical tone awareness as discussed above, only four tasks were used in this study: phoneme isolation (including both vowel and consonant), phoneme deletion, and phoneme blending, with ten items respectively in each task. In both the vowel and consonant phoneme isolation tasks, the participants were asked to identify the vowel/consonant in a compound sound. In the phoneme deletion task, the participants were asked to delete the on-set sound from a compound sound and then pronounce the rest of the sound. Finally, in the phoneme blending task, the participants were asked to combine two or three sounds together that were provided orally by the examiner. The composite score was the combination of the standardized scores for all four tasks used in this study. Based on the original data of this test, the Cronbach’s a coefficients were between 0.60 and 0.73 (p < 0.01) for all the subtests,

Table 1 Descriptive statistic of all participants in Study 1. Dyslexia

CA

RL

F(2, 87)/t

Post-hoc (Bonferroni)

0.71 0.67 0.80

203.55** 243.42** 198.39**

CA = dyslexia > RL CA = dyslexia > RL CA = dyslexia > RL

101.80 101.55 100.75

10.21 9.70 11.09

1.29 1.50 0.21

CA = RL = dyslexia CA = RL = dyslexia CA = RL = dyslexia

– 21.14 29.19

– 6.81 7.14

8.79** 11.45** 174.20**

CA > dyslexia CA > RL = dyslexia CA > RL = dyslexia

M

SD

M

SD

M

SD

Age Second gradersa Fourth gradersa Sixth gradersa

8.11 9.97 11.86

0.76 0.52 0.94

7.87 10.13 12.02

0.72 0.81 0.89

6.07 7.91 10.22

IQ Second gradersa Fourth gradersa Sixth gradersa

99.32 99.55 100.31

16.52 14.29 12.90

101.59 100.84 102.53

9.71 11.07 8.85

Character reading Second gradersa Fourth gradersa Sixth gradersa

13.50 19.86 28.43

5.18 6.36 8.44

21.55 29.19 35.12

6.81 7.09 7.02

Note. CA = chronological age control; RL = reading level control; scores of character reading were raw data. a The grades here indicated dyslexic and CA children’s grades. ** p < 0.01.

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

and the relationships between this task and other reading comprehension measurements and academic achievements in Mandarin were expected to be significant. Based on our sample, Cronbach’s a coefficient of the composite score was 0.72. Moreover, this test is a standardized test intended for participants of primary school ages. Therefore, because the RL control group for the second graders were kindergarten children, they did not receive this phonological awareness test in the present study. 2.2.5. Rapid automatized naming test The Rapid Automatized Naming Test was developed by Tzeng, Chang, Chien, and Ling (2011). It consists of four types of stimuli: object, digit, number, and Zhuyin Fuhao. Participants are asked to name the stimuli in left-to-right order as quickly and accurately as possible, and then, the mean total naming time is calculated. Based on the original data of this test, the retest reliabilities were 0.77–0.91 (p < 0.01) for all subtests after a two-week interval. Based on our sample, Cronbach’s a coefficient was 0.86. 2.2.6. Verbal short-term memory task This task was adapted from the digit span portion of the Wechsler Intelligence Scale for Children—IV (WISC–IV). Digits were assigned randomly to lists with two constraints: no digit was repeated in a list and the lists contained no successive numbers (e.g., 2–3). The lists varied in length from 4 to 7 numbers. Two lists of each length were presented in total. After practice with a list containing three numbers, testing continued until the child made mistakes on both lists of a given length. The score was the length of the longest list participants could recall correctly. Examiners read the lists at a rate of one word per second, and children were told to listen carefully because they would be asked to repeat each list. Based on the original data of this test, the retest reliabilities were 0.87 (p < 0.01) after a four-week interval. Based on our sample, Cronbach’s a coefficient was 0.79. 2.3. Procedure Before formal testing began, three steps were completed. First, all researcher-created tasks (i.e., the lexical tone awareness task and the verbal STM task) were designed, and the preliminary examinations were completed. Also, all examiners (who were all either part- or full-time special educators in the participants’ schools) were trained for at least five sessions to learning the procedures for administering all tests used in the present study. Finally, informed consent was obtained from the parents of each child participant. 3. Results To answer the first question, separated correlational analyses of lexical tone awareness and other phonological processing abilities of children with and without dyslexia at different primary school ages were employed. The results are integrated into Table 2. Our results indicated the same pattern of the relationships of lexical tone awareness and other phonological processing abilities of children with and without dyslexia in second grade and fourth grade. That is, lexical tone awareness significantly correlated to all phonological processing abilities (phonological awareness, verbal STM, and RAN). In contrast, the patterns were slightly different between children with and without dyslexia in sixth grade. Specifically, the association between lexical tone awareness and verbal STM was significant for children with dyslexia whereas it was not significant for those without dyslexia.

207

To more precisely compare whether the correlations of lexical tone awareness and phonological processing, we followed the method used by Silver, Hittner, and May (2004), which is a modification of Dunn and Clark’s z using a backtransformed average Fisher’s Z procedure with a one-tailed correlation. For the second graders, our results revealed no significant difference in the association of lexical tone awareness with phonological awareness (z = 0.40, p = 0.35 > 0.05) or with RAN (z = 1.12, p = 0.13 > 0.05) between children with and without dyslexia; however, the association of lexical tone awareness with verbal STM between children with and without dyslexia showed a significant difference (z = 1.83, p = 0.03 < 0.05) although both were significant correlations. For the fourth graders, no significant difference in the relationships of lexical tone awareness and phonological awareness (z = 0.51, p = 0.31 > 0.05), verbal STM (z = 0.00, p = 0.50 > 0.05), or RAN (z = 1.71, p = 0.24 > 0.05) between children with and without dyslexia. Finally, for the sixth graders, there was no significant difference in the relationships of lexical tone awareness and phonological awareness (z = 1.24, p = 0.11 > 0.05) and RAN (z = 0.85, p = 0.20 > 0.05) between children with and without dyslexia. It is worth noting that the difference in the relationships of lexical tone awareness and verbal STM between children with and without dyslexia was also not significant (z = 1.13, p = 0.11 > 0.05). However, the patterns between the two groups were different, which is significant for children with dyslexia but not for children without dyslexia. To examine the second research goal, a 3 (Grade: three grades)  2 (Category: with and without dyslexia) two-way ANCOVA was employed to compare lexical tone awareness by controlling for the IQ and phonological processing abilities of children with and without dyslexia in the three primary school grades. The results are shown in Table 3. A significant interaction between Grade and Category on lexical tone awareness was found. In other words, the simple main effects of either Grade or Category on lexical tone awareness were necessary. Regarding the Grade, the results revealed that CA group in second grade performed significantly worse than the CA group in fourth grade (p < 0.001) and in sixth grade (p < 0.001), but there was no significant different between the CA group in fourth grade and the CA group in sixth grade (p = 0.189 > 0.05). In addition, the children in the RL group in second grade performed significantly worse than those in the fourth-grade group (p < 0.001) and in the sixth-grade group (p < 0.001) on lexical tone awareness. Moreover, there was also a significant difference between the RL of the fourth-grade group and the sixth-grade group (p < 0.001). Finally, for the children with dyslexia, fourth graders outperformed second graders on lexical tone awareness (p < 0.001) while sixth graders with dyslexia performed better than fourth graders with dyslexia on lexical tone awareness (p = 0.015 < 0.05). Regarding the Category, which involved comparisons of the lexical tone awareness of the groups with dyslexia and the two control groups, the results revealed that the performance of second graders with dyslexia was significantly worse than both the CA control group (p < 0.001) and the RL control group p < 0.001), and this significantly worse performance on lexical tone awareness was also found in fourth grade and sixth grade between children with dyslexia and CA control groups (4th: p < 0.001; 6th: p < 0.001) and RL control groups (4th: p = 0.003 < 0.01; 6th: p < 0.001). In sum, our results revealed that lexical tone awareness associated significantly with phonological processing—especially phonological awareness and RAN, and the patterns of children with and without dyslexia are similar. Also, lexical tone awareness of

208

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

Table 2 Correlations of lexical tone awareness, phonological awareness, verbal short-term memory, and rapid automatized naming in children with and without dyslexia at different primary school ages after controlling for IQ. 1

2

3

Second graders 1. Lexical tone awareness 2. Phonological awareness 3. Verbal short-term memory 4. Rapid automatized naming

– 0.56** 0.37* 0.54**

0.63** – 0.45* 0.45*

0.71** 0.58** – 0.52**

Fourth graders 1. Lexical tone awareness 2. Phonological awareness 3. Verbal short-term memory 4. Rapid automatized naming

– 0.57** 0.55** 0.55**

0.47** – 0.58** 0.52**

0.55** 0.40* – 0.53**

Sixth graders 1. Lexical tone awareness 2. Phonological awareness 3. Verbal short-term memory 4. Rapid automatized naming

– 0.67** 0.09 0.37*

0.44* – 0.33 0.49**

0.40* 0.18 – 0.04

4 0.72** 0.54** 0.37* – 0.67** 0.35+ 0.65** – 0.55** 0.53** 0.19 –

Note. The numbers highlighted in italics are for the CA group; the numbers that are not highlighted are for the dyslexic group. + p < 0.06. * p < 0.05. ** p < 0.01.

Table 3 Comparison of different children’s lexical tone awareness with controlling IQ and phonological processing (with main effects and interactions from the 2  3 ANCOVA). Measurement

Dyslexia

CA

(N = 30) M a

Second graders Fourth gradersa Sixth gradersa a ** ***

22.82 29.13 33.25

RL

(N = 30) SD 4.46 6.23 4.54

M 33.72 38.68 38.83

Main effect

(N = 30) SD 3.18 4.14 3.37

M 27.89 32.97 38.80

Interaction

Grade SD 3.61 3.31 4.52

21.65

***

Grade  Category

Category

g2p

F

0.16

g2p

F 18.50

***

0.14

g2p

F 6.55

**

0.08

The grades here indicated dyslexic and CA children’s grades. p < 0.01. p < 0.001.

both children with dyslexia and RL group increase by ages, while the growth of lexical tone awareness of CA group stopped growing after fourth grade. Finally, children with dyslexia underperform both CA and RL groups in all primary school ages. 4. Discussion The goals of the present study were to examine the relationships of lexical tone awareness and phonological processing and, then, to compare the development of lexical tone awareness in children with and without dyslexia at different primary school ages. The first goal was to examine the relationships of lexical tone awareness and phonological processing, including phonological awareness, verbal STM, and RAN. Our results were in line with previous studies, which reported strong associations between lexical tone awareness and phonological processing—especially phonological awareness (McBride-Chang, Tong, et al., 2008; Shu et al., 2008; Tong et al., 2015) and RAN (e.g., Shu et al., 2008)—and our findings were consistent between children with and without dyslexia. Our results here matched the idea proposed by some researchers (e.g., Li, Anderson, Nagy, & Houcan, 2002; Tzeng et al., 2006)— that the strong associations of lexical tone awareness with other phonological processing skills may indicate that lexical tone awareness should be understood as a concept under the umbrella of phonological processing in Chinese. On the other hand, the role of verbal STM on lexical tone awareness was proposed and supported by previous studies (Xu, 1991). In general, it was thought that lexical tone information was

encoded in short-term memory (Lee, Vakoch, & Wurm, 1996). Our results partially supported this point of view because of the significant relationships of lexical tone awareness and verbal STM of children both with and without dyslexia at all three primary school ages. In our results, however, this association seems to be slightly inconsistent between children with and without dyslexia when they reach the late primary school ages (i.e., sixth grade) although both were significant. This group difference might be explained as simply that the underlying mechanisms of lexical tone awareness of Chinese children with and without dyslexia are diverse. However, this discrepancy remains unclear and needs further examination in the future. Our second research goal was to compare the development of lexical tone awareness in children with and without dyslexia at different primary school ages. First, our results showed that children with dyslexia performed significantly worse than those without dyslexia (both CA group and RL group) at three primary school ages. This result was slightly inconsistent with the findings from Li and Ho (2011), in which the significantly worse performance of lexical tone awareness of children with dyslexia was found only when comparing them to the CA group—but not to the RL group. A possible reason for this inconsistency is the different ways the children with dyslexia learn Chinese in the two studies. Children from Hong Kong and Macau, where Cantonese is the native language, learn to read Chinese characters using the ‘‘look-and-say” method, without a system of phonetic symbols to label the pronunciations. This approach emphasizes rapid retrieval of the names of Chinese characters and may strengthen the importance of rapid naming to

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

Hong Kong children’s reading skills (Ho, 2010). In contrast to the approach in Hong Kong and Macau, people on mainland China (Pin Yin) and in Taiwan (Zhuyin Fuhao), where Mandarin is the native language, also read Chinese characters, but learn a system of phonetic symbols that not only provide visual cues about the sounds but also function as a type of phonological awareness training (Lin et al., 2010). The different methods for learning Chinese character were also found to influence the cognitive portfolio of those with dyslexia. More specifically, the major deficit of children with dyslexia who learn Chinese through the phonetic system (e.g., Mainland China and Taiwan) was mainly found in relation to their phonological awareness, while children with dyslexia who learn Chinese without learning the phonetic system (e.g., Hong Kong and Macau) were found to have a major deficit in their rapid naming and orthographic knowledge but not in their phonological awareness (Luan & Ho, adapted from Ho, 2010). Ho (2010) speculated that the high percentage of children with rapid naming deficits found in Hong Kong might be due to the look-and-say teaching method used in Hong Kong, which requires readers to rapidly retrieve information from their lexicon. In other words, the children with dyslexia in Li and Ho’s (2011) study who were recruited from Hong Kong might not show such severe deficits on lexical tone awareness as those with dyslexia in our study (who were recruited from Taiwan) because of the differences in the ways they learn Chinese. Furthermore, our results showed that the development of the typically developing children’s lexical tone awareness in Mandarin also stopped at between eight to ten years old, which coincides with the findings of previous studies focused on developments of auditory temporal processing (e.g., Hautus et al., 2003; Wang & Yang, 2016) and Cantonese lexical tone awareness (e.g., Ciocca & Lui, 2003; Sze, 2004). On the other hand, the results of the present study indicate that the development of lexical tone awareness in children with dyslexia is slightly different from that of typically developing children. The children with dyslexia in the sixth grade performed better than those in the second and fourth grades on lexical tone awareness. Similarly, the children with dyslexia in the fourth-grade performed significantly better than those in the second grade. Because there was no indication that the children with dyslexia had stopped developing lexical tone awareness in the sixth grade, the implication is that these children probably do not reach their ceiling in the development of tone awareness until at least the end of primary school. However, this determination requires follow-up evidence from the students in secondary school ages. Considering the results above occurred after controlling for relevant phonological processing abilities, one of the possible reasons for the later development of lexical tone awareness in children with dyslexia is that it simply reflects their developmental lag in auditory temporal processing (Wang & Yang, 2016). However, it is worth noting that the development ceiling for auditory temporal processing in Chinese children with dyslexia found by Wang and Yang (2016) occurs at late primary school ages (between ten to twelve years old), an age obviously earlier than our findings here. Furthermore, the significantly worse performance of the dyslexic group than that of either the CA group or the RL group represented a severe problem in lexical tone awareness at all primary school ages, rather than a purely developmental lag involving auditory temporal processing. Therefore, the deficits of lexical tone awareness of Chinese children with dyslexia might be affected by other possible influential factors such as verbal working memory (e.g., Wong & Perrachione, 2007) or acoustical similarity (e.g., Tsao, 2008). A more definitive answer must wait for further examinations. In sum, our findings above imply that lexical tone awareness is a major problem for Chinese children with dyslexia and that this

209

problem probably cannot be resolved by natural development across the primary school ages. 5. Study 2 5.1. Introduction Study 1 disclosed the relationships among lexical tone awareness, phonological processing, and character reading in different grades. Furthermore, a significant difference between Chinese children with and without dyslexia on lexical tone awareness was found after controlling for phonological processing at different primary school ages. This finding provided some preliminary evidence on the deficits of lexical tone awareness of Chinese children with dyslexia at different primary school ages. Nevertheless, the correlational nature of the data in Study 1 preliminarily indicated the predictive ability of a lexical tone awareness deficit to reading problems but was not strong enough to disclose a relationship between the deficit in lexical tone awareness and reading problems. Therefore, a comparison before and after introducing the same/different discrimination training procedure of the Perceptual Training Method on lexical tone awareness and character reading to examine the role of lexical tone awareness of Chinese children with dyslexia in character reading abilities during primary school was conducted in Study 2. 5.2. Methods 5.2.1. Participants Considering that a significant improvement in the lexical tone awareness of typically developing children was found at approximately ten years of age in previous studies (e.g., Ciocca & Lui, 2003; Sze, 2004), those children without serious reading problems might not need additional help during primary school. Therefore, the training in Study 2 was implemented only for those children with dyslexia. All the children with dyslexia from the different grades in Study 1 were randomly assigned to experimental and control groups by grade. There were six groups in total: (1) a second-grade experimental group, (2) a second-grade control group, (3) a fourthgrade experimental group, (4) a fourth-grade control group, (5) a sixth-grade experimental group, and (6) a sixth-grade control group. The children in the experimental group and groups for each grade showed no significant differences in age, IQ, or character reading ability. Table 4 shows the descriptive data for all participants in Study 2. 5.3. Materials The participants in this study were evaluated on the character reading test and lexical tone awareness task in Study 1. 5.4. Procedure After the initial period of data collection, to thoroughly examine the effect of this training, the Perceptual Training Method was introduced into the groups to train all the children with dyslexia in the experimental groups. All participants in the experimental group received a total of 16 training sessions; each session lasted approximately 15 min. Considering children’s rapidly developing cognitive abilities—especially attentional control—from early primary school ages to middle primary school ages (Anderson, 2002), the training providers were asked to provide a three-minute short break when they found that the participants were unable to maintain sufficient

210

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

Table 4 Descriptive statistic of all participants in Study 2. Experimental group

Control group

T(14)

Comparisons

M

SD

M

SD

Age Second graders Fourth graders Sixth graders

8.02 10.17 11.74

1.62 2.04 1.39

7.94 9.82 11.97

1.88 1.43 1.10

0.40 1.05 0.86

Experimental = Control Experimental = Control Experimental = Control

IQ Second graders Fourth graders Sixth graders

88.42 90.38 92.77

11.90 14.16 15.94

91.34 89.72 92.08

13.55 13.69 14.79

1.95 1.08 0.61

Experimental = Control Experimental = Control Experimental = Control

attention. Apart from the flexibility of implement in each session, the training sessions were consistently conducted in the morning before the first class or in the afternoon after lunchtime. Also, all the training sessions were lasting three and four weeks. Each training session was divided into first and second phases; the durations of these phases were flexibly adjusted depending on the participants’ familiarity with the first phase training. The details of the first and second phases in the training program were introduced below. The training providers were full- or part-time qualified special educators in Taiwan; most of them served as the special education teachers in the schools attended by the participants. Furthermore, to ensure the training quality, the learning objectives and recommended activities were provided and discussed by the authors with all training providers in advance of the sessions. In addition, the authors arranged one trial lesson for each training provider to ensure that the provided training would reach the expected fidelity level. Finally, a checklist containing each step of the training session was distributed to the training providers to help ensure that their training included identical crucial elements both for the purposes of this study and to enhance the fidelity of the intervention. To improve the participants’ lexical tone awareness, a Perceptual Training Method was designed by adapting Wang et al. (1999) and Wang, Jongman, and Sereno (2003) for repeatedly exposing the participants to different—but limited—numbers of lexical tones. By following this repeated exposure, the participants were expected to further develop more automatic and unconscious reactions to the presented stimuli (Shiffrin & Schneider, 1977). Furthermore, as discussed above, the task that involved discrimination of same/different tones was shown to be both more attractive and more effective (Morosan & Jamieson, 1989). In this type of Perceptual Training Method, the participants were asked to determine different stimuli either simultaneously or in succession are the same or different. The Perceptual Training Method has two phases in each period. In the first phase, the task was a two-alternative forced-choice discrimination. There were ten possible pairs of stimuli (including Tone 1-Tone 1, Tone 1-Tone 2, Tone 1-Tone 3, Tone 1-Tone 4, Tone 2-Tone 2, Tone 2-Tone 3, Tone 2-Tone 4, Tone 3-Tone 3, Tone 3Tone 4, and Tone 4-Tone 4). The participants were asked to indicate (1) whether the two stimuli were the same or different, and if different, then (2) they were asked to indicate what the two stimuli were. During the process, the participants were given corresponding feedback when they provided correct and incorrect answers. In the second phase of the training program, the task was modified to a three-alternative forced-choice discrimination with 19 possible pairs of stimuli (including Tone 1-Tone 1-Tone 1, Tone 1-Tone 1-Tone 2, Tone 1-Tone 1-Tone 3, Tone 1-Tone 1-Tone 4, Tone 1-Tone 2-Tone 2, Tone 1-Tone 2-Tone 3, Tone 1-Tone 2Tone 4, Tone 1-Tone 3-Tone 3, Tone 1-Tone 3-Tone 4, Tone 2-

Tone 2-Tone 2, Tone 2-Tone 2-Tone 3, Tone 2-Tone 2-Tone 4, Tone 2-Tone 3-Tone 3, Tone 2-Tone 3-Tone 4, Tone 2-Tone 4-Tone 4, Tone 3-Tone 3-Tone 3, Tone 3-Tone 3-Tone 4, Tone 3-Tone 4Tone 4, and Tone 4-Tone 4-Tone 4). Compared to the twoalternative forced-choice items administered in the first phase, the three-alternative forced-choice items are much more difficult to discriminate. Similar to the first phase, participants were asked to indicate (1) whether the three given stimuli were identical and, if not, then (2) to indicate which one was different. During the process, the participants were given corresponding feedback to both correct and incorrect answers. In contrast to the experimental groups, the children with dyslexia in control groups participated in the regular activities in their classroom during the intervention phases for the experimental groups. The reason children in the control groups received no additional intervention in this study is that they were matched to control for natural maturation factors and test-retest effect rather than simply comparing the effectiveness of different interventions. After the training period, another data collection period was introduced. All participants’ performances on lexical tone awareness and character reading were measured again to evaluate whether they had improved in comparison to the pretest.

6. Results To examine the third research goal of this study, the intergroup comparisons and intragroup comparisons of lexical tone awareness and character reading of the participants in the experimental and control groups at different primary school ages (see Table 5) were addressed separately. Two 3 (Grade: three grades)  2 (Group: experimental and control) two-way analyses of covariance (ANCOVAs) were first conducted for the lexical tone awareness and character reading posttests. The IQ and pretest scores on the two tasks served as covariates for intergroup comparisons for each ANCOVA as suggested by Dimitrov and Rumrill (2003) to reduce the error variance and eliminate systematic bias. Furthermore, to test the assumption of homogeneity of regression coefficients, a design with the factor  covariate interaction was introduced (García-Berthou, 2001). The results indicated that the interaction between both factors (independent variables [Grade and Group]) and both covariates (IQ and pretest scores) were not significant on either lexical tone awareness (Grade ⁄ IQ: F[2, 80] = 0.20, p = 0.82 > 0.05; Grade ⁄ Pertest score: F[2, 80] = 0.85, p = 0.43 > 0.05; Group ⁄ IQ: F [1, 80] = 0.24, p = 0.88 > 0.05; Group ⁄ Pertest score: F[1, 80] = 1.33, p = 0.25 > 0.05) or character reading (Grade ⁄ IQ: F[2, 80] = 0.20, p = 0.82 > 0.05; Grade ⁄ Pertest score: F[2, 80] = 1.55, p = 0.22 > 0.05; Group ⁄ IQ: F[1, 80] = 0.11, p = 0.75 > 0.05; Group ⁄ Pertest score: F[1, 80] = 0.37, p = 0.54 > 0.05). In other words, our data did not violate homogeneity of regression assumption and the ANCOVA was appropriate for further proceeding.

211

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214 Table 5 Descriptive data of lexical tone awareness and character reading of experimental and control groups before and after perceptual training method. Grades

Experimental group Pretest

Control group Posttest

Pretest

Posttest

M

SD

M

SD

M

SD

M

SD

Lexical tone awareness Second graders Fourth graders Sixth graders

23.51 28.79 32.97

3.43 5.43 4.39

29.43 32.36 33.31

7.08 4.28 5.57

23.17 30.23 33.84

3.07 7.38 6.82

22.78 30.12 30.74

2.55 6.93 8.09

Character reading Second graders Fourth graders Sixth graders

13.26 21.19 26.73

4.88 5.31 7.63

20.36 19.63 25.86

5.64 4.79 8.48

15.42 21.09 31.26

5.61 4.81 6.93

15.73 21.86 28.89

6.12 5.08 4.01

The results of the two-way ANCOVAs revealed that there was no significant interaction between Grade and Group (F = 1.22; p = 0.30 > 0.05) on posttest lexical tone awareness. Therefore, the main effects of these factors were calculated separately because the differences of posttest lexical tone awareness from Grade (F = 8.83; p < 0.01; g2p = 0.10) and Group were both significant (F = 5.39; p < 0.01; g2p = 0.12). The main effects of Grade on posttest lexical tone awareness revealed that second graders performed significantly worse than both fourth graders (p < 0.01) and sixth graders (p < 0.01) but that the fourth graders’ performance was not significantly different from that of the sixth graders’ (p = 0.63 > 0.05). The main effects of Group on posttest lexical tone awareness revealed that the performance of the experimental group was significantly better than that of the control group (p < 0.01). Furthermore, the results of another two-way ANCOVA indicated that there was a significant interaction between Grade and Group (F = 3.82; p = 0.03 < 0.05) on posttest character reading. Therefore, the simple main effect of Grade and Group on posttest character reading was subsequently examined via ANCOVAs (after controlling for IQ and pretest scores) and independent t tests. On one hand, for the simple main effect Grade on posttest character reading, the performance of the experimental groups in three grades were not significantly different (F = 2.35; p = 0.11 > 0.05; g2p = 0.10) whereas there were significant differences among the three grades in the control group performances (F = 14.44; p < 0.001; g2p = 0.41). The detailed pattern of the grade difference in the character reading ability of the control group was that second graders performed significantly worse than both fourth graders (p < 0.01) and sixth graders (p < 0.001) and fourth graders performed significantly worse than sixth graders (p < 0.01). On the other hand, for the simple main effect of Group on posttest character reading, the results showed that the performance of the experimental group was significantly better than that of the control group only for those in second grade (t = 2.15; p = 0.04 < 0.05; Cohen’s d = 0.79; Hedges’ g = 0.77) but for those not in fourth grade (t = 1.24; p = 0.23 > 0.05; Cohen’s d = 0.45; Hedges’ g = 0.44) or in sixth grade (t = 1.25; p = 0.22 > 0.05; Cohen’s d = 0.46; Hedges’ g = 0.44). Furthermore, the intragroup improvement under different conditions was examined via repeated measured ANCOVA to compare the mean differences between pretest and posttest performances at different primary school ages. Our results indicated that there was a significant interaction between the pretest-posttest difference and Grades for lexical tone awareness of the experimental group (F = 4.10; p = 0.049 < 0.05; g2p = 0.09); therefore, the paired sample t tests were used to test the simple main effect. Significant improvements on lexical tone awareness were found for those in second grade (t = 2.52; p = 0.03 < 0.05; Hedges’ g = 1.04) and in fourth grade (t = 2.26; p = 0.04 < 0.05; Hedges’ g = 0.71) but not for those in sixth grade (t = 0.18; p = 0.86 > 0.05; Hedges’ g = 0.07).

In contrast, there was no significant interaction between the pretest-posttest difference and Grade for the lexical tone awareness of the control group (F = 0.78; p = 0.38 > 0.05; g2p = 0.18), and the pretest-posttest difference was also not significant (F = 0.21; p = 0.65 > 0.05; g2p = 0.01). Furthermore, similar examinations were also conducted to test the intragroup improvements in character reading under different conditions. The results reported a significant interaction between the pretest-posttest difference and Grade for the character reading of the experimental group (F = 5.48; p = 0.02 < 0.05; g2p = 0.11); therefore, the paired sample t tests were used to test the simple main effect. A significant improvement on lexical tone awareness was found for those in second grade (t = 4.02; p < 0.01; Hedges’ g = 1.32) but not for those in fourth grade (t = 0.84; p = 0.42 > 0.05; Hedges’ g = 0.30) or in sixth grade (t = 0.27; p = 0.79 > 0.05; Hedges’ g = 0.08). In contrast to the experimental group, the interaction between the pretest-posttest difference and Grade for character reading of the control group was not significant (F = 0.78; p = 0.38 > 0.05; g2p = 0.18), and the pretest-posttest difference was also not significant (F = 0.47; p = 0.50 > 0.05; g2p = 0.11). The Study 2 results revealed that some of the children with dyslexia benefitted from the Perceptual Training Method. Specifically, those in the second grade who received training performed significantly better on both lexical tone awareness and character reading ability, whereas the fourth graders who received training benefitted significantly only on lexical tone awareness. In contrast, no significant improvement was observed on either lexical tone awareness or character reading ability for those in all grades who did not receive training. It is noteworthy that significant differences could be found only on the lexical tone awareness and character reading ability between the control and experimental groups in second grade, but not in fourth grade.

7. Discussion The goal of Study 2 was to examine the response to an effective intervention, namely, the Perceptual Training Method, on the lexical tone awareness and character reading ability of Chinese children with dyslexia at different primary school ages. The results of Study 2 revealed that children with dyslexia in the second grade benefitted significantly on both lexical tone awareness and character reading ability from the Perceptual Training Method, whereas those in the fourth grade obtained a significant improvement only on lexical tone awareness. In contrast, the children with dyslexia in the sixth grade received no benefit from the Perceptual Training Method. It is also noteworthy that there was a significant improvement on lexical tone awareness for the fourth-grade children with dys-

212

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

lexia in the experimental group after receiving the intervention, but that posttest comparison of the control group and experimental group on lexical tone awareness was not significant. The slightly inconsistent results of the two examinations indicate that the benefits accrued by fourth graders with dyslexia from the Perceptual Training Method might not be as strong as those accrued by second graders. The findings above, which showed that the second graders earned significant benefits on both lexical tone awareness and Chinese character reading, that the fourth graders mainly gained significant benefits on lexical tone awareness, and that the sixth graders gained no benefit from the training, revealed an apparent developmental trend that might be explained by considering the development of reading. The ‘‘Read-to-learn stage” starts around fourth grade as proposed by Chall (1996) in her Stages of Reading Development, in which primary school-aged children face increasingly difficult and complex reading materials after fourth grade and, consequently, a fourth-grade slump could occur. Therefore, reading development might have a different influence on the effects of the Perceptual Training Method on lexical tone awareness and character reading for different ages, especially on fourth and sixth grade students in primary school. Because improvement of both lexical tone awareness and character reading could be found mainly in the second graders, the lexical tone awareness deficit of second graders with dyslexia was partially confirmed. In other words, a strong relationship between the improvements of lexical tone awareness and reading was found only in second grade; this pattern could not be found at the fourth and sixth grade levels. Therefore, lexical tone awareness deficit training might be useful only during early primary school for children with dyslexia. The results emphasize the importance of lexical tone awareness to Chinese character reading at ages similar to those found in Study 1. The result that only early primary school children with dyslexia can benefit from the Perceptual Training Method reflects the suggestions of Shaywitz et al. (1999)—that intervention with younger dyslexic students should have remediation as the goal, whereas management for older children with dyslexia is most often based on accommodation. However, unlike the stable and consensual hypotheses of dyslexia in alphabetic languages, where deficits in phonological awareness and rapid naming can be found across all ages (e.g., Hatcher & Hulme, 1999; Manis, Seidenberg, & Doi, 1999; Norton & Wolf, 2012; Shaywitz et al., 1999), the stability and validity of the stronger relationship between lexical tone awareness deficit and character reading for Chinese dyslexics might be questioned due to the obvious age differences we found in Study 2. Therefore, further investigations and experiments on this topic are necessary for future studies.

8. General discussion Considering the results of Study 1 and 2 together, the problems of lexical tone awareness of children with dyslexia appeared across primary school ages, but only those at early primary school ages could benefit from effective remediation on lexical tone awareness. In theory, the developmental pattern of lexical tone awareness found here is consistent with the development of cognitivelinguistic skills described in the previous studies (e.g., Chen & Tzeng, 1999, 2005; Huang & Chan, 1997; Tzeng, Chien, Chang, Chou, & Lien, 2005). According to Anderson (2002), due to the developmental patterns of primary school aged children’s executive functions—especially information processing, cognitive flexibility, and attentional control—children’s performances may change rapidly during the early primary school years. This finding makes it possible to reasonably infer that the large age differences

in lexical tone awareness may be partially attributable to the development of these cognitive abilities. Also, the importance of lexical tone awareness could be taken into practical consideration in schools. In Taiwan, children learn Zhuyin Fuhao from the beginning of primary school, and the Chinese characters in their textbooks have Zhuyin Fuhao marks attached until the children reach the fourth grade. Therefore, early primary school aged children are assumed to be very familiar with Zhuyin Fuhao, of which lexical tone is one part; however, previous studies have shown associations between deficiencies in Zhuyin Fuhao and Chinese children with dyslexia (Hu & Catts, 1998), and the findings of the present study further demonstrated deficits in lexical tone awareness in Chinese children with dyslexia. Furthermore, considering the strong relationship found between lexical tone awareness and Chinese character reading skills in early primary school ages in this paper, teachers should be encouraged to provide explicit instruction on lexical tone—instruction which is currently usually not the central component of Zhuyin Fuhao— as early as possible, especially for those children who face severe reading problems. The present study had at least four limitations. First, for the variable aspects of lexical tone, awareness is just one aspect of this comprehensive ability; production is the other aspect (e.g., Wang et al., 2003). The mental processing of these two aspects of lexical tone could be very different. However, during a child’s development, awareness of lexical tone is more fundamental and occurs earlier than its production (Wong, Schwartz, & Jenkins, 2005). In other words, awareness is more essential and has a higher priority when conducting lexical tone awareness capabilities examinations. However, even though lexical tone awareness occurs earlier, the unique importance of lexical tone production has also been demonstrated for children’s character reading abilities (e.g., Wang et al., 2003; Wong et al., 2005). Therefore, lexical tone production should be taken into consideration in future studies. Second, all items in the phonological awareness test used in this study apply mainly to the phoneme-level and onset-rhyme level of phonological awareness. Although this test is a standardized test that is widely used in Taiwan, it still lacks syllable-level phonological awareness, which has been shown to contribute significantly to children’s (especially kindergartners’) ability to read Chinese characters and words and English words (Chow, McBride-Chang, & Burgess, 2005). Therefore, this lack may lead to some limitations of the explanations of phonological awareness in this study. Therefore, we suggest that future studies should consider including syllable-level phonological awareness. Third, even though a cross-sectional design can compare the performance of participants in different grades, a thorough demonstration of the developmental trend of lexical tone production is limited by this design, especially for clearly identifying causal inference from different factors (Bland, 2000). In other words, even though IQ, age, and reading levels of the participants in the present study were controlled, the development pattern of the participants’ lexical tone awareness could still possibly have been affected by individual participants in different grades. To examine the development of lexical tone awareness and gain more detailed information (e.g., intercept and slope) a longitudinal study focused on this issue should be conducted in the future. Finally, previous studies have reported associations between lexical tone awareness and segmental phonological awareness (e.g., Zhang & McBride-Chang, 2014), and such associations were also found in this study. Given the well-established role of segmental phonological awareness in Chinese word reading, it may be that the improvement of character reading was not due to the improvement of lexical tone awareness per se, but instead to the improvement of segmental phonological awareness. The same principle applies to cognitive constructs. It seems impossible to exclude

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

these possibilities without comparing segmental phonological awareness and cognitive measures before and after training. Furthermore, also a potential limitation about the intervention fidelity in this study is that the items in checklist we issued for ensuring the intervention fidelity focus on only the teaching steps of the intervention for the experimental groups but other important components, such as the documents of teaching for control group (O’Donnell, 2008) and the participants’ responsiveness (e.g., Hasson, 2010). 9. Conclusions Despite the limitations listed above, the results of the present study are useful because they show the later developmental ceiling of Chinese children with dyslexia than those without dyslexia. In addition, the performance of lexical tone awareness could also be useful for distinguishing dyslexic from typically developing children, especially during the early primary school years. Finally, this finding was expanded in Study 2 to show positive responses to effective intervention on lexical tone awareness and character reading ability in younger children with dyslexia (second graders), but not for older children. This paper’s findings enhance our understanding of Chinese children with dyslexia, especially during the early primary school years and can be extended to help plan remediation or even diagnosis. However, for a better understanding of the importance and development of lexical tone in dyslexia, future research with a longitudinal design should be conducted to examine the awareness and production of lexical tone in typically developing and dyslexic Chinese children. Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work described in this article was partially supported by the Research Support Scheme (2016–2017) of the Department of Special Education and Counselling at the Education University of Hong Kong. Declaration of interest The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. References Anderson, P. (2002). Assessment and development of executive function (EF) during childhood. Child Neuropsychology, 8, 71–82. Bland, J. M. (2000). An introduction to medical statistics (3rd ed.). Oxford, UK: Oxford University Press. Catts, H. W., & Kamhi, A. G. (2005). Language and reading disabilities (2nd ed.). Needham Heights, MA: Pearson/Allyn and Bacon. Chall, J. S. (1996). Stages of reading development (2nd ed.). Orlando, FL: Harcourt Brace College Publishers. Chen, C. S. (2001). Comparison of cognitive components of word recognition of the second graders with and without word recognition difficulties. Bulletin of Special Education, 21, 215–237. Chen, J. H., & Chen, H. Y. (2006). Raven’s progressive matrices – Parallel manual. Taipei, Taiwan: Chinese Behavioral Science Corporation. Chen, X., Shu, H., Wu, N., & Anderson, R. C. (2003). Stages in learning to pronounce Chinese characters. Psychology in the Schools, 40, 115–124. Chen, S., & Tzeng, S. (1999). Phonological abilities in reading disabled elementary graders. Bulletin of Special Education, 17, 205–224. Chen, S., & Tzeng, S. (2005). The role of rapid automatized naming and phonological awareness in classifying Chinese children with dyslexia – A case study of remedial teaching. In L.-Y. Hung, C.-C. Wang, & C.-Y. Chen (Eds.), Overcoming learning difficulties-studies of assessments and implications (pp. 179–214). Taipei, Taiwan: Psychological Publishing.

213

Cheung, H., Chung, K. K. H., Wong, S. W. L., McBride-Chang, C., Penney, T. B., & Ho, C. S. H. (2009). Perception of tone and aspiration contrasts in Chinese children with dyslexia. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 50, 726–733. Chow, B. W. Y., McBride-Chang, C., & Burgess, S. (2005). Phonological processing skills and early reading abilities in Hong Kong Chinese kindergarteners learning to read English as a second language. Journal of Educational Psychology, 97, 81–87. Chung, K. K. H., Ho, C. S.-H., Chan, D. W., Tsang, S. M., & Lee, S. H. (2010). Cognitive profiles of Chinese adolescents with dyslexia. Dyslexia, 16, 2–23. Chung, K. K. H., McBride-Chang, C., Cheung, H., & Wong, S. W. L. (2013). General auditory processing, speech perception and phonological awareness skills in Chinese-English biliteracy. Journal of Research in Reading, 36, 202–222. Ciocca, V., & Lui, J. (2003). The development of the perception of Cantonese lexical tones. Journal of Multilingual Communication Disorders, 1, 141–147. Cooper, A., & Wang, Y. (2012). The influence of linguistic and musical experience on Cantonese word learning. Journal of the Acoustical Society of America, 131, 4756–4769. Cooper, A., & Wang, Y. (2013). Effects of tone training on Cantonese tone-word learning. Journal of the Acoustical Society of America, 134, EL133–EL139. Deacon, S. H., & Kirby, J. R. (2004). Morphological awareness: Just ‘‘more phonological”? The roles of morphological and phonological awareness in reading development. Applied Psycholinguistics, 25, 223–238. Dimitrov, D. M., & Rumrill, P. D. Jr., (2003). Pretest-posttest designs and measurement of change. Work, 20, 159–165. Ellis, A. W. (2014). Reading, writing and dyslexia: A cognitive analysis. Hove, UK: Psychology Press. Francis, A. L., Ciocca, V., Ma, L., & Fenn, K. (2008). Perceptual learning of Cantonese lexical tones by tone and non-tone language speakers. Journal of Phonetics, 36, 268–294. Gandour, J. T. (1978). The perception of tone. In V. Fromkin (Ed.), Tone: A linguistic survey (pp. 41–76). New York: Academic Press. García-Berthou, E. (2001). On the misuse of residuals in ecology: Testing regression residuals vs. the analysis of covariance. Journal of Animal Ecology, 70(4), 708–711. Hasson, H. (2010). Systematic evaluation of implementation fidelity of complex interventions in health and social care. Implementation Science, 5(1), 67. Hatcher, P. J., & Hulme, C. (1999). Phonemes, rhymes, and intelligence as predictors of children’s responsiveness to remedial reading instruction: Evidence from a longitudinal intervention study. Journal of Experimental Child Psychology, 72, 130–153. Hautus, M. J., Setchell, G. J., Waldie, K. E., & Kirk, I. J. (2003). Age-related improvements in auditory temporal resolution in reading-impaired children. Dyslexia, 9, 37–45. Ho, C. S.-H. (2010). Understanding reading disability in Chinese: From basic research to intervention. In M. H. Bond (Ed.), Handbook of Chinese psychology (2nd ed., pp. 109–121). New York: Oxford Press. Hu, C.-F., & Catts, H. W. (1998). The role of phonological processing in early reading ability: What we can learn from Chinese. Scientific Studies of Reading, 2, 55–79. Huang, H. S. (2001). Chinese character recognition test. Taipei, Taiwan: Psychological Publishing. Huang, H., & Chan, H. (1997). An analysis of phonemic awareness, word awareness and tone awareness among dyslexia children. Bulletin of Special Education and Rehabilitation, 5, 125–138. Hulme, C., & Snowling, M. J. (2013). Learning to read: What we know and what we need to understand better. Child Development Perspectives, 7, 1–5. Kendeou, P., & Trevors, G. (2012). Learning from texts we read: What does it take? In M. J. Lawson & J. R. Kirby (Eds.), The quality of learning (pp. 251–275). Cambridge, UK: Cambridge University Press. Lee, Y. S., Vakoch, D. A., & Wurm, L. H. (1996). Tone perception in Cantonese and mandarin: A cross-linguistic comparison. Journal of Psycholinguistic Research, 25, 527–542. Li, W., Anderson, R. C., Nagy, W., & Houcan, Z. (2002). Facets of metalinguistic awareness that contribute to Chinese literacy. In W. Li, J. S. Gaffney, & J. L. Packard (Eds.), Chinese children’s reading acquisition (pp. 87–106). New York: Springer. Li, W. S., & Ho, C. S.-H. (2011). Lexical tone awareness among Chinese children with developmental dyslexia. Journal of Child Language, 38, 793–808. Lin, D., McBride-Chang, C., Shu, H., Zhang, Y., Li, H., Zhang, J., ... Levin, I. (2010). Small wins big: Analytic pinyin skills promote Chinese word reading. Psychological Science, 21(8), 1117–1122. Lin, C. Y., Wang, M. I. N., & Shu, H. U. A. (2013). The processing of lexical tones by young Chinese children. Journal of Child Language, 40, 885–899. Manis, F. R., Seidenberg, M. S., & Doi, L. M. (1999). See Dick RAN: Rapid naming and the longitudinal prediction of reading subskills in first and second graders. Scientific Studies of Reading, 3, 129–157. Matthews, S., & Yip, V. (1994). Cantonese: A comprehensive grammar. London: Routledge. McBride-Chang, C., Lam, F., Lam, C., Doo, S., Wong, S. W. L., & Chow, Y. Y. Y. (2008). Word recognition and cognitive profiles of Chinese pre-school children at risk for dyslexia through language delay or familial history of dyslexia. Journal of Child Psychology and Psychiatry, 49, 211–218. McBride-Chang, C., Tong, X., Shu, H., Wong, A. M. Y., Leung, K.-W., & Tardif, T. (2008). Syllable, phoneme, and tone: Psycholinguistic units in early Chinese and English word recognition. Scientific Studies of Reading, 12, 171–194.

214

L.-C. Wang et al. / Contemporary Educational Psychology 49 (2017) 203–214

Melby-Lervåg, M., Lyster, S. A., & Hulme, C. (2012). Phonological skills and their role in learning to read: A meta-analytic review. Psychological Bulletin, 138, 322–352. Morosan, D. E., & Jamieson, D. G. (1989). Evaluation of a technique for training new speech contrasts. Journal of Speech Language and Hearing Research, 32, 501–511. Norton, E. S., & Wolf, M. (2012). Rapid automatized naming (RAN) and reading fluency: Implications for understanding and treatment of reading disabilities. Annual Review of Psychology, 63, 427–452. O’Donnell, C. L. (2008). Defining, conceptualizing, and measuring fidelity of implementation and its relationship to outcomes in K–12 curriculum intervention research. Review of Educational Research, 78(1), 33–84. Peng, G., Zhang, C., Zheng, H. Y., Minett, J. W., & Wang, W. S. Y. (2012). The effect of intertalker variations on acoustic-perceptual mapping in Cantonese and Mandarin tone systems. Journal of Speech, Language, and Hearing Research, 55, 579–595. Shaywitz, S. E., Fletcher, J. M., Holahan, J. M., Shneider, A. E., Marchione, K. E., Stuebing, K. K., ... Shaywitz, B. A. (1999). Persistence of dyslexia: The Connecticut longitudinal study at adolescence. Pediatrics, 104, 1351–1359. Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic human information processing: II. Perceptual learning, automatic attending and a general theory. Psychological Review, 84, 127–190. Shu, H., Peng, H., & McBride-Chang, C. (2008). Phonological awareness in young Chinese children. Developmental Science, 11, 171–181. Silver, N. C., Hittner, J. B., & May, K. (2004). Testing dependent correlations with nonoverlapping variables: A Monte Carlo simulation. The Journal of Experimental Education, 73(1), 53–69. Silverman, K., Beckman, M., Pitrelli, J., Ostendorf, M., Wightman, C., Price, P., . . . Hirschberg (1992). TOBI: A standard for labeling English prosody. Paper presented at the proceedings from ICSLP 1992: The second international conference on spoken language processing. Banff: Alberta, Canada. Sze, W. (2004). The development of tone perception in Cantonese-speaking children (Dissertation). Paper presented at the University of Hong Kong, Hong Kong. Tong, X., McBride-Chang, C., Wong, A. M. Y., Shu, H., Reitsma, P., & Rispens, J. (2011). Longitudinal predictors of very early Chinese literacy acquisition. Journal of Research in Reading, 34, 315–332. Tong, X., Tong, X., & McBride-Chang, C. (2015). Tune in to the tone: Lexical tone identification is associated with vocabulary and word recognition abilities in young Chinese children. Language and Speech, 58, 441–458. Tsao, F. M. (2008). The effect of acoustical similarity on lexical-tone perception of one-year-old mandarin-learning infants. Chinese Journal of Psychology, 50, 111–124. Tzeng, S., Chang, Y., Chien, S., & Ling, Y. (2011). Rapid automatized naming test. Taipei, Taiwan: Chinese Behavioural Science Corporation. Tzeng, S., Chen, S., & Hsieh, Y. (2006). Phonological awareness test battery (screen). Taipei, Taiwan: Ministry of Education.

Tzeng, S., Chien, S., Chang, Y., Chou, L., & Lien, Y. (2005). Predicting Chinese reading comprehension and word reading with early naming speed and phonological awareness measures: A four-year follow-up. In Bulletin of special education (Vol. 28, pp. 123–144). van den Broek, P., & Espin, C. A. (2012). Connecting cognitive theory and assessment: Measuring individual differences in reading comprehension. School Psychology Review, 41, 315–326. Wagner, R. K., & Torgesen, J. K. (1987). The nature of phonological processing and its causal role in the acquisition of reading skills. Psychological Bulletin, 101, 192–212. Wang, H. S., Huss, M., Hämäläinen, J. A., & Goswami, U. (2012). Basic auditory processing and developmental dyslexia in Chinese. Reading and Writing, 25, 509–536. Wang, Y., Jongman, A., & Sereno, J. A. (2003). Acoustic and perceptual evaluation of mandarin tone productions before and after perceptual training. Journal of the Acoustical Society of America, 113, 1033–1043. Wang, Y., Spence, M. M., Jongman, A., & Sereno, J. A. (1999). Training American listeners to perceive mandarin tones. Journal of the Acoustical Society of America, 106, 3649–3658. Wang, L. C., & Yang, H. M. (2016). Temporal processing development in Chinese primary school-aged children with dyslexia. Journal of Learning Disabilities. Wayland, R. P., & Li, B. (2008). Effects of two training procedures in cross-language perception of tones. Journal of Phonetics, 36, 250–267. Wong, P. C. M., & Perrachione, T. K. (2007). Learning pitch patterns in lexical identification by native English-speaking adults. Applied Psycholinguistics, 28, 565–585. Wong, P., Schwartz, R. G., & Jenkins, J. J. (2005). Perception and production of lexical tones by 3-year-old, mandarin-speaking children. Journal of Speech Language and Hearing Research, 48, 1065–1079. Xu, Y. (1991). Depth of phonological recoding in short-term memory. Memory & Cognition, 19, 263–273. Zhang, J., & McBride-Chang, C. (2010). Auditory sensitivity, speech perception, and reading development and impairment. Educational Psychology Review, 22, 323–338. Zhang, J., & McBride-Chang, C. (2014). Auditory sensitivity, speech perception, L1 Chinese, and L2 English reading abilities in Hong Kong Chinese children. Developmental Psychology, 50, 1001–1013. Zhang, Y., Zhang, L., Shu, H., Xi, J., Wu, H., Zhang, Y., & Li, P. (2012). Universality of categorical perception deficit in developmental dyslexia: An investigation of mandarin Chinese tones. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 53, 874–882. Ziegler, J. C., & Goswami, U. (2005). Reading acquisition, developmental dyslexia, and skilled reading across languages: A psycholinguistic grain size theory. Psychological Bulletin, 131, 3–29.