home index by author index by subject

Talk; reading

Saturday stream 5 Session 09.00 - 11.10 Length 25 minutes

Phonological perseverations in nonword reading in severe developmental dyslexia

Bo Jorgensen

University of Copenhagen. boj@greve.dk

Abstract

Brain-imaging studies have shown that developmental dyslexics have lowered activation in left temporo-parietal areas during both reading and phonological processing. Damage in temporo-parietal areas has earlier been shown to cause recurrent perseverations. Recently, a connection between cognitive deficits and perseverations have been found; perseverations tend to stay within cognitive areas with deficits. The present study aimed to see if there is a higher occurrence of recurrent phonological perseverations in developmental dyslexia. Fifteen students in ninth grade referred to special dyslexia instruction and fifteen controls in ninth grade at a community school were tested with a 34 item nonword reading test. The numbers of correctly read words and of perseverative errors were counted. There was no overlap between the groups in perseverations (range in the dyslexia group was 3-17 and in controls 0-1), and the difference was clearly significant (p < 0.01, Mann-Whitney U-test). The study indicates that the often-mentioned nonword reading deficit in developmental dyslexia is composed of more than core phonological deficits, and that the reading deficit in developmental dyslexia can be further boosted by errors stemming from recurrent phonological perseverations.

Brain-imaging studies have shown that developmental dyslexics have lowered activation in left temporo-parietal areas during both reading (Rumsey et al 1997) and phonological processing (Shaywitz et al 1998, Horwitz et al 1998). Areas in left temporo-parietal cortex have earlier been found to be involved in phonological processing (Binder et al 1996), which has been considered the core deficit in dyslexia (Bradley & Bryant 1983). Recently, Rumsey and colleagues (Rumsey et al 1999) showed that dyslexia severity could be predicted from blood flow in left angular gyrus during phonological processing, indicating that dyslexia is the result of a dysfunction in angular gyrus. The obvious problem in this line of thought is that dyslexics, who have phonological problems, tend to use different strategies than phonological in both reading and phonological tasks, and therefore could be expected to exhibit lower blood flow in brain areas involved in phonological processing, even if there was no neurological dysfunction present. The only secure conclusion in this line of research is therefore that good reading relies on good phonological processing.

Damage in temporo-parietal areas has earlier been shown to cause recurrent perseverations (Albert & Sandson 1986), characterised by reoccurrences of words, syllables or phonemes in otherwise rather fluently speech (Sandson & Albert 1984). Recently, a connection between cognitive deficits and perseverations have been found; perseverations tend to stay within cognitive areas with deficits (Cohen & Dehaene 1998). Cohen & Dehaene (1998) theorised that recurrent perseverations can be the results of problems at the input level. If a new stimuli is not strongly enough perceived, it will not override older stimuli, causing the person to persevere at the output level. Therefore, if a person has phonological processing deficits caused by dysfunction in the areas of the brain processing incoming phonologically coded information, we could expect that person to persevere at the phonological output level.

The present study aimed to see if there is a higher occurrence of recurrent phonological perseverations in developmental dyslexia during processing of novel phonological information, thereby validating the connection between dyslexia and left temporo-parietal dysfunction.

Method

Fifteen students in ninth grade referred to special dyslexia instruction and fifteen controls in ninth grade at a community school were tested individually with a 34-item nonword reading test. All were of normal to superior intelligence tested with Wisc (dyslexics) and Raven's Progressive Matrices (controls). The reading deficit of the dyslexics was according to the school severe, a reading age delay ranging from 5 to 8 years.

The numbers of correctly read words and of perseverative errors were counted. The testing took place at the schools during lesson time.

The test was developed for the occasion, and consisted if 34 items ranging from two to seven letters. All were readable in Danish.

Results

There was a significant difference between the groups in number of correctly read items (Mann-Whitney U-test p < 0.01), as expected.

Nonword reading, dyslexics Nonword reading, controls

There was no overlap between the groups in perseverations (range in the dyslexia group was 3-17 and in controls 0-1), and the difference was clearly significant (p < 0.01, Mann-Whitney U-test).

Perseverations in nonword reading, controls and dyslexics Range, perseverations, dyslexics

All perseverations were of recurrent type, phonemes and syllables from earlier items popping up in later items.

Example:

The correlation between perseverations and correctly read words was -0.499 (p = 0.58).

Discussion

The difference between the groups in number of correctly read items was expected, as Rack et al (1992) have shown that nonword reading is severely impaired in dyslexia. The difference between the groups in number of perseverations was clear. This indicates that the often-mentioned nonword reading deficit in developmental dyslexia is composed of more than core phonological deficits. Even though a perseveration automatically led to a wrongly read item, the correlation stayed moderate, explaining only approximately 25 % of the variance in nonword reading performance. This indicates, that the main problem in nonword reading in dyslexia is the phonological processing deficit, but that the reading deficit in developmental dyslexia can be further boosted by errors stemming from recurrent phonological perseverations. The study also supports the claim that dyslexia is the result of a dysfunction in left temporo-parietal areas; recurrent perseverations of phonemes and syllables have been found in patients with damage in left temporo-parietal cortex and not in patients with damage elsewhere (Sandson & Albert 1986, Cohen & Dehaene 1998).

This study is part of a larger project comprising the master thesis of the author at University of Copenhagen.