Elsevier

Cognition

Volume 108, Issue 1, July 2008, Pages 243-262
Cognition

Brief article
Morpheme-based reading aloud: Evidence from dyslexic and skilled Italian readers

https://doi.org/10.1016/j.cognition.2007.12.010Get rights and content

Abstract

The role of morphology in reading aloud was examined measuring naming latencies to pseudowords and words composed of morphemes (roots and derivational suffixes) and corresponding simple pseudowords and words. Three groups of Italian children of different ages and reading abilities, including dyslexic children, as well as one group of adult readers participated in the study. All four groups read faster and more accurately pseudowords composed of root and suffix than simple pseudowords (Experiment 1). Unlike skilled young and adult readers, both dyslexics and younger children benefited from morphological structure also in reading aloud words (Experiment 2). It is proposed that the morpheme is a unit of intermediate grain size that proves useful in processing all linguistic stimuli, including words, in individuals with limited reading ability (dyslexics and younger readers) who did not fully develop mastering of whole-word processing. For skilled readers, morphemic parsing is useful for reading those stimuli (i.e., pseudowords made up of morphemes), for which a whole-word lexical unit does not exist; where such whole-word lexical units do exist, skilled readers do not need to rely on morphological parsing because they can rely on a lexical (whole-word) reading unit that is larger than the morpheme.

Introduction

In languages as different as English, Finnish and Italian, elementary school children use the morphemic constituents of words in performing tasks on written stimuli such as fragment completion (Feldman, Rueckl, DiLiberto, Pastizzo, & Vellutino, 2002), word definition (Bertram et al., 2000, Burani et al., 2006) and lexical decision (Burani, Marcolini, & Stella, 2002). Sensitivity to word morphology develops early in childhood (Carlisle and Nomabody, 1993, Casalis and Louis-Alexandre, 2000) and is present to some extent also in impaired readers (Casalis et al., 2004, Elbrö and Arnbak, 1996, Leong and Parkinson, 1995).

Morphological knowledge in word comprehension and production tasks does not necessarily imply that morphemes also play a role as processing units in a print-to-sound decoding task such as reading aloud. Little information on this topic is available on children. Studies in opaque orthographies, such as English, Danish and French, followed the assumption that in such orthographies (in which word spelling is to some degree morphologically governed) knowledge of morphemes may help the child in assigning the correct word pronunciation (Seymour, 1997, Verhoeven and Perfetti, 2003). With untimed stimulus presentation, the presence of known morphemes in a word, such as stems and affixes, may affect young readers’ accuracy in reading aloud, mainly when morphologically complex words are phonologically and semantically transparent with respect to the base word (Carlisle and Stone, 2003, Elbrö and Arnbak, 1996, Laxon et al., 1992), or when suffixes are frequent and productive (Mann & Singson, 2003).

Up-to-date, only one study has assessed the role of morphology in children’s reading aloud a transparent language. Burani et al. (2002) showed that young Italian readers in third to fifth grades could benefit from the presence of morphemes similarly to adult readers (see review in Burani & Laudanna, 2003). In a naming task, pseudowords made up of a root and a derivational suffix in a combination not existing in Italian (e.g., DONNISTA, ‘womanist’) were read faster and more accurately than simple pseudowords matched for orthographic familiarity (e.g., DENNOSTO).

To our knowledge, no study has investigated whether in transparent orthographies morphemes are effective units also in reading aloud words. For an experienced reader, parsing a word into morphemic sub-parts may be an efficient strategy when it is not familiar. For low-frequency words, the recourse to higher-frequency constituents (morphemes) may facilitate processing (Burani & Thornton, 2003).

Apart from frequency, the adoption of morphemes as reading units may also be constrained by the joint effect of word length and reader’s ability. For both adult and young skilled readers, 7–10 letter words can be processed in one shot (Hutzler and Wimmer, 2004, Rayner et al., 1976). However, developmental dyslexics experience difficulties in processing such stimuli as whole units. Italian dyslexics’ eye fixations reveal fractionated text scanning, with a prevalence of small amplitude saccades (De Luca, Di Pace, Judica, Spinelli, & Zoccolotti, 1999). These eye-fixation patterns result in an extremely slow and analytical reading strategy and in marked stimulus length effects (Spinelli et al., 2005, Zoccolotti et al., 1999, Zoccolotti et al., 2005a) affecting similarly words and non-words (De Luca, Borrelli, Judica, Spinelli, & Zoccolotti, 2002; see also Hutzler & Wimmer, 2004). This pattern of reading performance resembles that of children at an early stage of learning to read (Zoccolotti et al., 2005a).

The present study assessed reading aloud of morphologically complex words and pseudowords in Italian children of different reading abilities, including developmental dyslexics. Our hypothesis was that morphemic constituents (roots and suffixes) could help dyslexics to read aloud both pseudowords and words. Morphemes may be efficient reading units for dyslexics because they have an intermediate size between graphemes, which lead to extremely slow and analytical processing, and words, which for dyslexics are too large units to be processed as a whole. Similar benefits from morpheme-based reading aloud of familiar words may be present in younger readers, but may not occur for skilled readers due to their ability to process larger reading units as a whole. Word-based reading avoids parsing and assembling costs connected to morpheme-based reading (for costs of morphemic parsing in lexical decision, see Laine et al., 1999, Traficante and Burani, 2003).

The reading aloud of both pseudowords and words composed of morphemes (roots and derivational suffixes) was compared to that of simple pseudowords and words with no root + suffix structure. Three groups of Italian children of different ages, with and without reading difficulties, were tested along with a group of adult readers. The aim was to show that, in a transparent orthography, readers of different skills may take advantage of reading units (morphemes) of larger than the single grapheme grain size. All groups of readers, irrespective of reading skill, were expected to take advantage of morphemic units in reading aloud pseudowords: the presence of morphemes in a pseudoword would result in shorter reading latencies and higher reading accuracy than grapheme-based reading. Only less skilled readers, i.e., younger readers and dyslexics, who have not developed efficient whole-word reading ability, were expected to rely on morphemic constituents when reading aloud words. In contrast, skilled readers, both children and adults, should read as fast and accurately both morphologically complex and simple words, because of their capacity to process both types of words as whole units.

Section snippets

Experiment 1. Reading aloud pseudowords

In the first experiment, the reading aloud of pseudowords made up of morphemes was contrasted with the reading of simple pseudowords.

Experiment 2. Reading aloud words

In Experiment 1, pseudowords were considered, contrasting pseudowords made up of morphemes and control pseudowords. In Experiment 2, words were considered, contrasting derived words and simple words; the aim was to answer whether morphemes are effective processing units (and equally effective reading units for readers of different reading skills) also when a larger reading unit (i.e., the word) is present. Morphemic constituents (roots and suffixes) in a word may affect positively the reading

General discussion

Confirming previous evidence (Burani et al., 2006, Burani and Laudanna, 2003, Burani et al., 2002), both adults and skilled children read faster and more accurately pseudowords composed of a root and a suffix. Less skilled (dyslexic and younger) readers also engaged in morphemic processing to supplement grapheme-phoneme decoding. These results add up to recent data showing that lexical units are available to Italian developmental dyslexics (Barca, Burani, Di Filippo, & Zoccolotti, 2006) as well

Acknowledgments

This study was supported by a grant from the International Dyslexia Association. Cristina Burani and Pierluigi Zoccolotti are members of the Marie Curie RTN “Language and Brain” (European Commission, MRTN-CT-2004-512141). The authors thank Despina Paizi for useful comments.

References (48)

  • K. Rayner et al.

    What guides a reader’s eye movements?

    Vision Research

    (1976)
  • P. Zoccolotti et al.

    Word length effect in early reading and in developmental dyslexia

    Brain and Language

    (2005)
  • P. Angelelli et al.

    Characteristics of writing disorders in Italian dyslexic children

    Cognitive and Behavioral Neurology

    (2004)
  • L. Barca et al.

    Context-sensitive rules and word naming in Italian children

    Reading and Writing: An Interdisciplinary Journal

    (2007)
  • R. Bertram et al.

    The role of derivational morphology in vocabulary acquisition: Get by with a little help from my morpheme friends

    Scandinavian Journal of Psychology

    (2000)
  • D. Brizzolara et al.

    Do phonological and rapid automatized naming deficits differentially affect dyslexic children with and without a history of language delay? A study on Italian dyslexic children

    Cognitive and Behavioural Neurology

    (2006)
  • C. Burani et al.

    Frequency, not age of acquisition affects Italian word naming

    European Journal of Cognitive Psychology

    (2007)
  • C. Burani et al.

    Naming morphologically complex pseudowords: A headstart for the root?

    The Mental Lexicon: A Journal of Cross-Disciplinary Research

    (2006)
  • C. Burani et al.

    Orthographic complexity and word naming in Italian: Some words are more transparent than others

    Psychonomic Bulletin and Review

    (2006)
  • C. Burani et al.

    Morpheme-based lexical reading: Evidence from pseudo-word naming

  • C. Burani et al.

    The interplay of root, suffix and whole-word frequency in processing derived words

  • J.F. Carlisle et al.

    Phonological and morphological awareness in first grades

    Applied Psycholinguistics

    (1993)
  • J.F. Carlisle et al.

    The effects of morphological structure on children’s reading of derived words in English

  • J.F. Carlisle et al.

    Exploring the role of morphemes in word reading

    Reading Research Quarterly

    (2005)
  • Cited by (118)

    • Morphological segmentation of nonwords in individuals with acquired dyslexia

      2022, Journal of Memory and Language
      Citation Excerpt :

      Cognitive neuropsychological investigations of the role of morphology in reading has largely focused on reading morphologically complex familiar words. Comparatively little work has focussed on the role of morphology in nonword reading (for evidence from children with dyslexia, see Burani, Marcolini, De Luca, & Zoccolotti, 2008). The examination of morphologically complex nonwords has several advantages.

    • Morphological decomposition compensates for imperfections in phonological decoding. Neural evidence from typical and dyslexic readers of an opaque orthography

      2020, Cortex
      Citation Excerpt :

      However, despite these impairments it was shown that dyslexic readers can benefit from morphological interventions to facilitate reading (Bowers et al., 2010; Goodwin & Ahn, 2013). Our results are consistent with behavioral findings from Hebrew and other languages showing that when the morphological structure is salient (i.e., has high orthographic or semantic transparency or high frequency) impaired readers rely on morphological information in reading more than unimpaired readers (Burani, 2010; Burani et al., 2008; Cavalli et al., 2017; Elbro & Arnbak, 1996; Leikin & Even Zur, 2006; Vellutino et al., 2004). Interestingly, mixed results were found in a recent study (Kimel & Ahissar, 2019) that examined dyslexic Hebrew readers learning novel verbs with or without a familiar morphological structure.

    View all citing articles on Scopus
    View full text