Improving the Reading Quality of Dyslexics

The Neurocognitive Laboratory, headed by Assoc. Prof. Zvia Breznitz, has been attacking the question of Why Johnny—or in this case, Yonatan--Can’t Read from, in a sense, a physical rather than a mental direction. The process of normal and defective reading involves, it has been found, the element of speed in information processing. The work in this laboratory complements the work being done in the reading lab (see Focus, Spring 97).

The question for investigation was this: Does the processing efficiency of the various internal processors in the process of identifying words also stem from the rate of processing, and thereby affect the reading pace of individuals and reading quality in general?

The three processors involved in processing information for reading are (1) the visual-orthographic, for processing the visual dimension of the written word; (2) the auditory-phonological, for the sound dimension of the word; and (3) the semantic, for the sense dimension of the word. The Neurocognitive Laboratory has been conducting systematic research on the pace and quality of processing within and between these processors. Examined were groups of 10-12 year olds and 22-26-year-old students, both dyslexic and normal readers.

The first series of tests involved 30 regular male readers and 30 dyslexic males, all university students. Does the does the documented slower reading of dyslexics emanate from a basically slower pace of processing by one or more of the processors? Electrophysiological and behavioral measures were employed to find out.

The former involved Event-Related Potentials (ERP) derived from an EEG with 24 sampling channels that was administered when the subjects undertook various assignments. The ERPs are crests in the recorded waves resulting from external stimuli; the time of the appearance of the wave (latencies) and the wave height (amplitudes) were measured. Behavioral measures were preciseness and reaction time for the different experimental tasks.

The ERP measurements enabled the receipt of direct information on brain activity in real time while the students were performing the assignments and tracking the various stages of brain activity as information was being processed: the conceptual stage, that of short-term memory updating, and that of semantic processing. The behavioral measures gave information on the quality of performance accuracy and the reaction times of the examinees in relation to the task at hand.

The stimuli were of various processing levels and were presented in the visual and auditory channels both separately and integrated. For an examination of the rate and quality of processing at low levels of processing, the test stimuli consisted of non-linguistic information (for example, barking and horns) and linguistic information, such as phonemes (the sound of letters) and graphemes (letters) of a simple processing level. To examine the pace and quality of processing information at higher levels of orthographic and phonological processing, the stimuli were linguistic-semantic at the level of word and sentence.

At the word level, experimental tasks included the ability to distinguish between rhyming pairs of words, between homonyms, between homomorphisms, between actual and made-up words, and so forth. The respondents were asked to answer such questions as the following: “Before you is a pair of words. Press the right button if they rhyme or press the left button if they don’t.”

At the sentence level, sentences consisted of short narratives with either a proper or an improper ending. Respondents had to read the sentences both at their own pace and at an accelerated rate.

Each student tested was first examined individually in the laboratory for two hours. After being attached to an EEG machine, he or she was request to perform the research tasks as the EEG waves were recorded.

Analysis of the ERPs recorded on the EEG was carried out by means of special programs developed for this purpose. These test results showed that the rate of processing in the auditory-phonological channel among both populations was slower than in the visual-orthographic channel. The gaps between these two information channels were larger among the dyslexic group of readers, whose rate of processing in the auditory-phonological channel was much slower than that of the control group.

The differentiation in processing rates focused mainly on the stimulus conception stage. The rise in the brain-wave associated with conceptual processing that was recorded came later in the auditory channel than in the visual and later in the phonological than in the orthographic channel. These differences were greater among the dyslexic readers.

The gap in conceptual processing between the two groups grew larger as the research task demanded simultaneous inter-sensory processing. In assignments in which semantic processing is also required, the recorded wave identified with semantic processing also appears later among the dyslexic readers.

It should be mentioned that the behavioral accuracy in performing visual-orthographic processing tasks was almost complete among both populations. The accuracy in auditory-phonological assignments was almost full among regular readers, but not so among the dyslexic group. The behavioral reaction times, too, for auditory-phonological tasks were longer among dyslexic readers. The research date point to clear correlations among this latter group between the latency of conceptual processing components and behavioral response times of simple linguistic auditory assignments and phonological assignments.

The phonological-processing failures among dyslexics possibly stem, among others, from the slow basic rate of processing in this channel, a rate that harms the visual-othographic channel. This phenomenon is found to be valid for both populations—the dyslexic and the regular readers. Apparently it does not interfere with the reading quality of the latter. Thus it is possible that the slowness of the auditory-phonological channel alone does not constitute a factor in reading difficulty; rather, the size of the gap in the processing rate of these two channels may possibly constitute the factor adversely affecting reading. This gap is large within the dyslexic population.

The visual-orthographic information and the auditory-phonological information must be carried out in full synchronism. Since the visual (holistic) information is faster in essence, it must wait for the slower auditory (serial) information. This action necessitates synchronizing the processing rates in these channels. In light of the time limitations that the information-processing system imposes on information processing in the reading act, a large enough gap in the processing rate between these two channels can disrupt the synchronism and interfere with the synchronizing of information. The auditory information arrives too late at its meeting place with the visual information, and no desirable paradigm is created of reading representations, the pace of reading is slowed down, and the quality of reading is harmed.

On the basis of this research, it seems clear that the pace of the normal self-processing of information among dyslexics is, on the one hand, a factor that interferes with the quality and efficiency of the act of reading. On the other hand, other laboratory studies have found that speeding up the processing rate makes the quality of dyslexics’ reading in general, and that of persons with reading difficulties in particular, more efficient. Thus, one of the practical conclusions of these studies is that it is possible to train and to accelerate the information-processing rate of both dyslexics and of those with reading difficulties, thereby improving reading quality.

With the aim of examining the hypothesis of the information processing rate as a basic factor for understanding the developmental phenomenon of dyslexia, a replication of the study described is now being carried in the Children’s Reading Laboratory at the University. Children with developmedyslexia and two control groups comparable in chronological and reading age are being studied.





Back to Table of Contents