The SAGE Handbook of Dyslexia

Gavin Reid, Angela Fawcett, Frank Manis, Linda Siegel, Gavin Reid, Angela Fawcett, Frank Manis, Linda Siegel

• Publisher: SAGE Publications Ltd
• Year: 2008
• ISBN:
  9781473971769
• Edition: 1
• Topic: Bildung
• Subtopic: Lernschwierigkeiten

The SAGE Handbook of Dyslexia is a comprehensive overview of a complex field. It is a rich, critical assessment of past and present theory and current research, which also looks to the future. The editors have brought together key figures from the international academic world - both researchers and practitioners - to examine the relationships between theoretical paradigms, research and practice, and to map new areas of research.

The book has 5 main sections:

-neurological/genetic perspectives

-cognitive and learning perspectives

-educational influences

-beyond school

-international perspectives.

PART I

Neurological/Genetic Perspectives

1

The Origin of Dyslexia: The Asynchrony Phenomenon

Zvia Breznitz

This chapter will

i)	discuss synchronization in speed and content as a factor contributing to fluency in the regular and impaired reading process; and
ii)	present an innovative concept for our understanding of dyslexia

Reading is a highly composite cognitive task that involves word decoding and comprehending printed or written materials. Effective word decoding is a prerequisite for comprehension. In most cases, the reading process is successful. However, in approximately 10–15 per cent of the population it is not. Reading failures are commonly termed developmental dyslexia. Different theories have been proposed in an attempt to understand the dyslexia phenomenon. Some theories are descriptive and focus on the inaccuracy and slowness/dysfluency of word reading among dyslexics (see British Psychological Society, 1999: 18 for their definition of dyslexia), while others aim to explain dyslexia (Lyon et al., 2003). An abundance of data has pointed to impairments in phonological processing as the cause of poor word reading (e.g., Share, 1995). The lack of orthographic, semantic, syntactic, and morphological skills has also been considered (e.g., see Adams, 1990 for a review). Other evidence has indicated that dyslexia has a neurological basis (Lyon et al., 2003).

This chapter will present an innovative concept for our understanding of dyslexia, which refers to its foundations. Breznitz’s recent asynchrony theory proposes that dyslexia is an outcome of the failure to synchronize the various brain entities activated during the reading process (see Breznitz, 2002; Breznitz & Misra, 2003; Breznitz, 2006). The asynchrony theory is based on the idea that the word reading process relies on various sources of information with regard to printed materials. This information arrives from different brain entities. These entities have different biological structures. They are activated in separate areas of the brain, and they process information in a different manner and at different speeds. Furthermore, the reading activity requires the flow of relevant information from one brain area (the posterior lobes) to another (the frontal lobe). Whereas the posterior lobes are responsible for perception and physical processing of a stimulus, the frontal lobe provides meaning and motoric pronunciations to the stimulus (Sousa, 2000). In addition, word decoding relies on the transfer of information between brain hemispheres (Sousa, 2000). The left hemisphere processes information in a sequential manner and specializes in linguistic processing. This hemisphere contains Wernicke’s area, where the mental lexicon is stored, and Broca’s area, which is responsible for language pronunciation. The right hemisphere for most right handed individuals processes information in a holistic way and specializes in the identification of visual patterns (Sousa, 2000). Reading as a linguistic activity requires work in both hemispheres (Marsolek, Kosslyn, & Squire, 1992) and diverse areas of the brain need to communicate in a timely fashion.

Moreover, reading is a cognitive process activated through the various stages of the information processing system, including perception, processing, and output (Atkinson & Shiffrin, 1971). The act of reading must be sufficiently fast to operate within the constraints of limited capacity and rapid decay of the information processing entities (Perfetti, 1985). Above all, word decoding during reading is an inflexible action. In most alphabetic languages each grapheme matches only one phoneme. This grapheme-phoneme correspondence necessitates precision in content and time. This complexity constitutes a major challenge for the human brain as each brain entity activated in this process operates on a different time scale (Breznitz, 2002; Breznitz & Misra, 2003; Breznitz, 2006 for review). As a result, the integration and synchronization in time of the information arriving from the various brain entities, at all levels and stages of activation, is essential for successful word reading to occur.

According to Breznitz (2006), a gap in speed of processing (SOP) between the different brain entities activated in the word decoding process may prevent the precise synchronization of information necessary for an accurate process. This idea lies at the heart of the asynchrony theory, which suggests that the wider the SOP gap between the different brain entities; the more severe the word decoding failure will tend to be (Breznitz, 2002; 2006). There are several preconditions for the asynchrony phenomenon to occur:

1. more than one system, area of brain activation and/or stage of cognitive operation (referred to as entities) are involved in the processing task;
2. there are differences in the speeds at which each entity processes information;
3. the SOP of the various entities is not sufficiently coordinated to allow effective integration.

ASYNCHRONY THEORY OF DYSLEXIA

More than one entity is involved in the word decoding process

Word decoding relies on different brain entities that can be distinguished on the basis of three levels of involvement in the process:

1. the biological brain systems;
2. the cognitive processes;
3. the alphabetic units of various sizes obtained from printed materials.

The three levels of activation

The biological level
The biological level refers to a domain-general notion of processing, of which reading is considered a part. It consists of several entities activated during reading from the entryway to the human brain along the different stages of activation within and between the various neural pathways of the brain. At the end of this process, it is necessary for the brain to provide a solution, which appears in the form of output accompanied, in many cases, by motoric features. The entities at this level are the visual and auditory-acoustic modalities. Each entity is represented by specific brain locations and has a different length and structure to its neural pathway, with the visual pathway being longer than the auditory. The visual system processes information in a holistic manner, while processing in the auditory system is sequential. Each contributes to the reading process at a different point in time and at a different speed. The visual entity is the first to start decoding. It is triggered by printed materials and then works in parallel with the auditory entity. The performance of each entity is subject to its intrinsic capabilities, to the successes and failures of other components, and to the manner in which the entities are synchronized in terms of their speed (rapidity) and content (accuracy).

The cognitive processes
The cognitive processes combine two levels of operation, the domain-general and domain-specific processes. The domain-general processes assembled at the biological level are responsible for distinguishing, selecting, perceiving, categorizing, storing, and retrieving information. The domain-specific processes assembled in the biological modalities activate the orthographic, phonological-recoding, and semantic processes, which are specific to the word decoding process. At the initial stage of activation, attention is first allocated to the printed materials prior to the operation of the biological systems (at the entryway), and then different levels of cognitive operation are assembled along with the activation of the biological processes.

The domain-general processes operate sequentially, while the domain-specific processes function in an interactive manner until word meaning is obtained (Seidenberg and McClelland’s PDP model, 1989; Harm & Seidenberg, 2004). There has been considerable debate concerning the mechanisms involved in mapping word meaning from print. Harm & Seidenberg (2004) attempted to resolve this longstanding debate by constructing and testing a computational model based on connectionist principles. According to this model, the meaning of a word is a ‘pattern of activation over a set of semantic units that develops over time based on continuous input from both orthography>semantics and orthography>phonology> semantics components’ (p. 663). Unlike previous accounts, in which the two pathways operate independently, in this model, both pathways determine meaning simultaneously, and have are mutually dependant on each other. The performance of each component is subject to both its intrinsic capabilities and the successes and failures of other components. An important assumption incorporated in the model is the notion that the reader’s task is to compute meanings both accurately and quickly. Although the orthography>phonology>semantics pathway has a clear speed disadvantage, as it involves an extra ‘step’ compared to the direct pathway (orthography>semantics), the direct pathway actually takes longer to learn.

The asynchrony theory suggests that in an attempt to process information adequately, the system’s speed attribute becomes a crucial factor in the further development of activation in the pathways. Moreover, more than one system is activated in each pathway and as such, synchronization is required and can only be achieved if the SOP gap between systems is minimal.

Furthermore, it is not clear whether there are two pathways that are activated during normal reading development. As during most activities, the brain searches for the most economical way to process information. It is conceivable that there is only one pathway used to process information during reading, the indirect pathway (orthography>phonology>semantic) which leads to an effective process. Over the years, the speed at which information is transferred from the orthographic to the semantic system via the phonological system is sped up. In other words, the recoding of printed materials in the phonological system continues to exist throughout, even during advanced word decoding in an inhibitory way, although it becomes very fast, barely manifested and hardly measurable with currently available research measures. Furthermore, the act of recoding the linguistic unit in the phonological system has an advantage, it has been trained over the years by two actions: spoken language and reading. It is possible that the dual training actions of this system assists speed of information processing to the extent that activation of the phonological step becomes barely noticeable during the normal course of word decoding among skilled readers the dual training actions of this system assists speed of information processing to the extent that activation of the phonological step becomes hardly noticeable during the normal course of word decoding among skilled readers.

The alphabetic level
The alphabetic level refers to domain-specific processing, which is an objective process, external to the reader and depends on the level of printed materials. The printed material units provide content to the biological and cognitive entities. This level includes all of the reading-related subtasks triggered by, and derived from, the printed materials followed by activation in the mental lexicon. It includes all levels of the alphabetic code (letters, sounds, and syllables), the various levels of linguistic units (words and connected text) and their phonological, orthographic, semantic, syntactic, and morphological representations and processes. Activation of the biological systems is triggered by these printed linguistic units, and elevates the various forms of processing to the different cognitive levels.

Successful reading requires a form of ‘dialogue’ between the different brain entities involved in this process. During the normal course of word decoding, the dialogue takes place at different stages of activation along the neural pathways. This places an additional workload on the decoding activity as it requires different levels of synchronization of the brain entities at different points in time. When the decoding skills are not automatic, are less stable, or have yet to progress through the developmental stages, and when the orthographic representations of words in the mental lexicon have yet to stabilize, the various entities are required to cooperate continuously in order to send whatever information is available (even if inaccurate) regarding the linguistic unit being processed. In an impaired or undeveloped word decoding process, the information in one or more entities may not be completely accurate or available. This may constitute an additional obstacle for precise synchronization between entities. Moreover, partial or incomplete input from one entity might diminish the incoming information arriving in the other entities.

Word decoding complexity is increased further by the fact that from an evolutionary perspective, the human brain has existed for around 60,000 years, and the alphabetic code for only 5000 years. Consequently, the ability to read is not part of our evolutionary heritage. No biological brain system has been developed specifically for the reading process, so the activation of reading must rely on systems that were developed for different tasks. This complexity poses a major challenge for the human brain, and proves too much for some readers.

Speed of processing differences between different brain entities

What are the speeds of processing of the brain entities involved in word decoding? Research indicates that non-linguistic auditory information arrives in the auditory cortex after 30 ms (Heil et al., 1999), whereas visual information arrives in the visual cortex after 70 ms (Schmolesky et al., 1998). In other words, based on the natural operation of these biological entities, auditory stimuli leave the ‘entryway’ and arrive in the brain faster than visual stimuli. This may be due to the different lengths of the neural pathways, brain sites, and structures of each modality. However, at the linguistic level there is evidence that at least from the word level onwards, linguistic information is processed in the auditory channel in a temporal-serial manner. In other words, the information concerning the phonemes that make up a word arrives sequentially (Rosenzweig and Bennett, 1996). In contrast, word processing in the visual channel is a holistic and simultaneous process (Willows et al., 1993), suggesting that visual processing at this level might be faster than auditory processing.

It is important to note that most existing studies have used behavioural measures of reaction time. This means that the information concerning this entire sequence of cognitive activity has only been provided at the conclusion of processing; in the reader’s output. This stage only occurs after the completion of sensory, cognitive and motor processes (Bentin, 1989; Brand...