Elsevier

Research in Developmental Disabilities

Volume 32, Issue 6, November–December 2011, Pages 2748-2757
Research in Developmental Disabilities

Speech perception in noise deficits in Japanese children with reading difficulties: Effects of presentation rate

https://doi.org/10.1016/j.ridd.2011.05.035Get rights and content

Abstract

We examined the effects of presentation rate on speech perception in noise and its relation to reading in 117 typically developing (TD) children and 10 children with reading difficulties (RD) in Japan. Responses in a speech perception task were measured for speed, accuracy, and stability in two conditions that varied stimulus presentation rate: high rate and same rate conditions. TD children exhibited significantly more stable responses in the high rate condition than they did in the same rate condition. Multiple regression analyses indicated that response stability in the high rate condition accounted for a unique amount of variance in reading and mora deletion. As a group, children with RD performed less accurately than did TD children in the high rate condition, but not in the same rate condition. Findings suggest that the dependence of speech perception on stimulus context relates to reading proficiency or difficulty in Japanese children. The influences of phonology and orthography of language on the relationships between speech perception and reading are discussed.

Highlights

► We examined the relation between reading and speech perception in Japanese children. ► We focused on the effects of presentation rate on speech perception in noise. ► Dependence of speech perception on stimulus context relates to reading proficiency. ► Perceptual deficits in children with reading difficulties depend on stimulus context.

Introduction

Developmental reading difficulty (RD), or dyslexia, is a failure to acquire age-appropriate reading skills, despite educational opportunity and at least statistically average intelligence (Snowling, 2000). Although the causal factors of these difficulties are still intensely debated, the most widely held theory currently implicates phonological deficits (Vellutino, Fletcher, Snowling, & Scanlon, 2004). That is, these individuals face difficulties in developing conscious awareness of segmental speech units, or manipulating them, so that they fail to map the orthographic code onto phonological code.

While phonological deficit theory is well established, others have argued that auditory processing deficits precede, and are causal of, phonological deficits. For example, it has been found that individuals with RD can exhibit problems in a variety of psychoacoustic tasks, including frequency discrimination (Banai & Ahissar, 2004), fast temporal processing (Tallal, 1980), or amplitude modulation detection (Goswami et al., 2002). In recent years, there has been additional focus on deficits in speech perception apart from general auditory processing (for a review see Rosen, 2003). Several studies have found deficits in individuals with RD in multiple aspects of speech perception. For instance, Adlard and Hazan (1998) found that a sub group of children with RD showed poor performance in a speech discrimination test. Similarly, McArthur, Ellis, Atkinson, and Coltheart (2008) reported that some individuals within a sample of children with RD showed abnormal thresholds in another speech discrimination task.

In addition to these findings, a number of studies found clear speech perception deficits in background noise. In the initial study of this line, Brady, Shankweiler, and Mann (1983) showed that nine-year-old poor readers performed less accurately than age-matched good readers on a repetition task requiring identification of words presented in noise. Similarly, Chermak, Vonhof, and Bendel (1989) found that adults with a learning disability consistently showed speech perception deficits in various types of noise conditions. Indeed, the majority of studies on the subject has consistently shown speech perception deficits in noise among individuals with RD in alphabetic languages as well as in nonalphabetic languages such as Japanese (Boets et al., 2007, Ziegler et al., 2009, Sakaue, 1999; but see also Robertson, Joanisse, Desroches, & Ng, 2009).

Although the hypothesized nature of these deficits differ in detail, most researchers assumed that deficits in auditory processing or speech perception result in (or result from) impoverished phonological representations, which provoke difficulty in learning to read (Boada & Pennington, 2006). Recently, an alternative viewpoint was suggested that speech perception deficits surface only as a function of stimulus context (e.g., Ahissar et al., 2006, Chandrasekaran et al., 2009). Ahissar et al. (2006) observed in their study that individuals with dyslexia did not benefit from repetitive presentation of the same auditory stimulus, and proposed the “perceptual anchor” theory. Similarly, Chandrasekaran et al. (2009) found that dyslexic children exhibited impairments in their ability to modify representation in repetitive stimulus contexts in a speech perception task using auditory brainstem responses. They suggested this impairment may contribute to a deficit in noise-exclusion (Sperling, Lu, Manis, & Seidenberg, 2005), a hallmark symptom of developmental dyslexia. These researchers encouraged a deeper examination of the dynamic processes in speech perception. In the present study, therefore, we investigated the presence of speech perception in noise deficits that has been repeatedly reported in alphabetic languages and in Japanese children with RD. Specifically, we focused on the context-dependent processes in speech perception in noise, through the effects of stimulus presentation rate.

The acquisition of reading is thought to be a culture-specific skill. Indeed, it is assumed that phonological properties and orthographic consistency strongly influence the acquisition of reading (Ziegler & Goswami, 2006). In the Japanese language, the segmental unit of speech sounds is based on a subsyllabic unit, the mora (Otake, Hatano, Cutler, & Mehler, 1993). Moreover, due to the constraint that a mora usually consists of a consonant and a vowel (Kubozono, 1999), the amount of required phonological analysis of speech sounds for Japanese listeners is thought to be relatively low. On the other hand, in the Japanese writing system, phonologic and logographic scripts, “Kana” and “Kanji,” respectively, are used concurrently. Therefore, in contrast to an alphabetic language such as English, Kana has a highly consistent orthography as each character usually represents one mora. Given that the importance of auditory processing in reading may vary according to orthographic consistency (Georgiou, Protopapas, Papadopoulos, Skaloumbakas, & Parrila, 2010), these characteristics of Japanese language may influence the relationship between reading and speech perception in Japanese children. Indeed, in an earlier epidemiological study, it was reported that the prevalence of reading problems in Japanese children was very low (.98%) compared to those reported in some alphabetic languages (Makita, 1968; see Uno, Wydell, Haruhara, Kaneko, & Shinya, 2009 for more details). However, compared to alphabetic languages, few studies exist on nonalphabetic languages of the relationship between reading and speech perception.

Therefore, the present study aimed to address the following questions: (i) Is there evidence for context-dependent processes in speech perception in noise, and does it relate to reading in Japanese children?; (ii) Is there a speech perception in noise deficit in Japanese children with RD, and if so, is it related to context-dependent processes, rather than to the presence of external noise itself?

Section snippets

Participants

Participants in this study included 117 typically developing (TD) children and 10 children with RD (Table 1). All of them were Japanese natives and recruited from mainstream primary schools. For inclusion, all children with RD possessed difficulty in the acquisition of reading and writing, and had general intelligence levels within normal limits (>80) on a verbal or performance IQ estimate from the Japanese version of the Wechsler Intelligence Scale for Children, Third Edition (Japanese

Speech perception performance in TD children

Descriptive statistics for all variables in the TD group are shown in Table 2. Two-way analysis of variance (ANOVA) with grade and condition as independent variables revealed main effects for grade on RT, CV, and ER [RT: F(5, 111) = 10.41, p < .001; CV: F(5, 111) = 7.30, p < .001; ER: F(5, 111) = 6.40, p < .001]. All variables revealed developmental change. There was also a main effect for condition on CV [F(1, 111) = 4.92, p < .05]. Responses in the high rate condition were significantly more stable than

Discussion

This study aimed to examine the effect of stimulus context on speech perception and its relationship to reading in TD children and children with RD in Japan. Speech perception in noise tasks were administered within two conditions that varied stimulus presentation rate.

References (29)

  • K. Banai et al.

    Poor frequency discrimination probes dyslexics with particularly impaired working memory

    Audiology and Neuro-Otology

    (2004)
  • G.D. Chermak et al.

    Word identification performance in the presence of competing speech and noise in learning disable adults

    Ear and Hearing

    (1989)
  • U. Goswami et al.

    Amplitude envelope onsets and developmental dyslexia: A new hypothesis

    Proceedings of the National Academy of Science of the United States of America

    (2002)
  • Inoue, T., Higashibara, F., Okazaki, S., & Maekawa, H. Relationship between reading ability and phonological processing...
  • Cited by (15)

    • Evaluation of speech in noise abilities in school children

      2017, International Journal of Pediatric Otorhinolaryngology
      Citation Excerpt :

      Children are particularly vulnerable to the effects of noise in speech perception, yet little is known about the relationship mechanism between core processes and the perception of speech in noise [2]. Studies indicate that difficulties in speech perception can result in poor phonological representations, reading, memory and learning difficulties [3,4]. The relationship of speech perception deficit in noise with reading difficulties can be found even in children who use non-alphabetic language, such as Japanese children [4].

    • Perceptual context and individual differences in the language proficiency of preschool children

      2016, Journal of Experimental Child Psychology
      Citation Excerpt :

      Nevertheless, here context had a direct contribution to auditory and visual identification, memory for syllables, and rapid naming. Previous studies also showed direct contribution of perceptual context to syllable memory and identification (Inoue et al., 2011; Oganian & Ahissar, 2012) and to pseudo-word repetition and pseudo-word reading (Ahissar et al., 2006; Oganian & Ahissar, 2012) in older children and adults. The idea that sensitivity to perceptual context and language skills are intimately linked is based on the anchoring deficit hypothesis of dyslexia (Ahissar, 2007), and the current findings provide evidence to the claim that perceptual context and language are associated even prior to the onset of formal reading instruction.

    • Phonological memory and word learning deficits in children with specific language impairment: A role for perceptual context?

      2015, Research in Developmental Disabilities
      Citation Excerpt :

      Together these studies suggest that anchoring could play a role in perceptual as well as in language processes. Most relevant in the context of the present study, anchoring deficits in persons with dyslexia were observed across the range of tasks used to elicit anchoring – auditory discrimination, phonological memory and pseudoword reading tasks (Ahissar et al., 2006; Inoue, Higashibara, Okazaki, & Maekawa, 2011; Oganian & Ahissar, 2012). Here we aimed to determine whether anchoring is similarly impaired among children with SLI.

    • Anchoring in 4- to 6-year-old children relates to predictors of reading

      2012, Journal of Experimental Child Psychology
      Citation Excerpt :

      In contrast to the strong version of the anchoring deficit hypothesis, both typical and dyslexic readers exhibited an anchoring effect, leading the authors to rule out the potential contribution of anchoring to reading development (see also Ziegler, 2008). Nevertheless, a recent study in Japanese-speaking school-age children showed that syllable discrimination in an anchoring-affording condition, but not in a condition where anchoring was not possible, was related to reading and phonological processing (Inoue et al., 2011). If anchoring influences reading by facilitating or impeding the acquisition of the domain-specific rules on which skilled reading develops, the theory should be tested in prereading children.

    View all citing articles on Scopus
    View full text