The Reading Program uses theoretical models of reading as the basis for investigating how children learn to read, why some children have so much difficulty (developmental dyslexia), and how such difficulties may best be treated. These models are applied to the understanding of, and can inform development of treatment options for, different kinds of reading disorders that are seen in formerly skilled readers after brain damage (acquired dyslexia). The neural basis of both normal and impaired reading can be investigated using brain imaging techniques such as event-related potentials.
- The relationship between poor reading, poor self-concept, and poor emotional health
- Computational modelling of reading
- Learning to read in different languages
- Attentional deficits in dyslexia
- Crowding effects in young readers and subtypes of developmental dyslexia
- Using brain stimulation to enhance spelling training in adults
- Using cognitive theories to inform our understanding of left ventral occipitotemporal dysfunction in dyslexia
- Letter processing in typical reading and dyslexia
Difficulty learning to read can result in negative consequences for children that go well beyond their ability to read and write. This difficulty can impact the childs self-concept and emotional health. This program comprises three studies that aim to better understand the relationships between different types of reading and spelling, and different types of self-concept and emotional health in poor readers. In our first study, we discovered thatpoor phonological decoding and poor sight word reading were specifically associated with poor academic self-concept. We also found that a large minority, around one-third, of poor readers had poor academic self-esteem. In our second study, a systematic review revealed that poor reading was more reliably associated with anxiety than depression. In our third study, we found that a number of types of reading were associated with various types of reading self-concept, but that reading comprehension alone appeared to be related to anxiety and depression. Considered together, these findings suggest that resilience training might help protect some poor readers from self-concept or emotional health problems and boost theefficacy of some types of reading treatment for some poor readers.
This ongoing project continues to focus on the second version of the Dual RouteCascaded (DRC) computational model of reading: DRC 2.0. The model is being tested using data from published experimental studies of reading aloud and visual word recognition. The project also aims to extend the success of DRC in modelling the Roman alphabet to the development of DRC-style computational models of reading aloud and visual word recognition in languages that do not use the Roman alphabet: Japanese, Russian and Greek. Japanese represents a unique challenge because it uses three different writing systems. Russian, written in the Cyrillic alphabet, is also challenging because it differs greatly from English with respect to the type of alphabetic writing it employs. DRC-style computational models of reading aloud and visual word recognition have been constructed for Japanese, Greek and Russian. Despite its success in offering an account of skilled reading, DRC has not explicitly modelled the process of acquiring reading skill. We are developing a new DRC model, entitled learning-DRC, or L-DRC, which models reading acquisition.
Research has repeatedly shown that it is harder to learn to read in English than in any other alphabetic language. One of the most important reasons is the fact that the mappings between print and speech are much less reliable in English than in other languages. To better understand the impact of this unreliability (or orthographic depth) on reading acquisition, we have further specified the definition of orthographic depth. In addition, we have demonstrated the importance of another cross-linguistic factor, namely differences in the degree of similarity between words. The results of our studies have implications for reading instruction in different alphabetic languages. We are currently starting a new project that compares reading acquisition in English and Mandarin - a non-alphabeticlanguage.
We continueto examine multiple aspects of attention in dyslexia, including temporal attention, crowding, and attentional dyslexia. Our research findings suggest that the preparation of attention may be slower in some individuals with dyslexia. This is consistent with a meta-analysis of the attentional blink research focused on dyslexia and follow-up experiments manipulating preparation time in the attentional blink in typical readers. We have begun exploring whether video game experience might alter speed of preparation and, in turn, may supplement existing reading interventions. Our research has also examined the effect of visual attention span and text spacing (crowding) on reading accuracy and fluency. Finally, we have studied the phenomenon of attentional dyslexia.People with attentional dyslexia swap letters between words, for example dark part may be read as park dart or dart park.
Some theorists posit that increased sensitivity to letter crowding may underlie reading deficits in developmental dyslexia. For example, in identifying the string of letters RKSFD while focusing on the central letter, crowding would affect the internal letters, particularly K and F, making these letters harder to identify. This theory of dyslexia predicts that other visual items with similar properties should also be affected, including digit strings such as 37526. Previously, with Associate Professor Sachiko Kinoshita we have examined the performance of university students with normal reading and those with developmental dyslexia on a task involving the identification of letters, symbols, and digits presented in strings like those above. It was found that participants with dyslexia were more prone to making errors in which they reported a letter at an incorrect position and these errors occurred particularly at the most crowded positions. We have now conducted the same task with young typically developing readers aged 7 to 12, as well as with children with letter position dyslexia. Letter position dyslexia is characterised by errors that change the positions of the internal letters, such as reading the word BEARD as BREAD. Preliminary results indicate that normal readers process letter strings and digit strings very similarly, while children with letter position dyslexia show differences from normal readers both in processing letters and digits. Ongoing work investigates the nature of these differences, the performance of children with phonological and surface dyslexia on this task, childrens performance on symbol strings, as well as any differences in performance with age.
We have extended our spelling training research to include adults without reading and spelling difficulties. In a series of studies, we have been training adults to spell difficult words that they are initially unable to spell. A novel aspect of this project is that we are investigating whether non-invasive brain stimulation can enhance the effectiveness of our spelling training program. That is, does the administration of brain stimulation during training increase the speed of learning orthographic knowledge and support its retention? We are using transcranial direct current stimulation (tDCS) to deliver a weak electrical current directly to the scalp to stimulate the left temporal cortex. This region is believed to be involved in various language processes, including orthographic processing. In addition to exploring the effects of tDCS on spelling performance, we are also interested in whether gains in spelling performance are reflected by changes in brain function. We are investigating this by measuring the lateralisation of language processes in the brain before and after the spelling training program, using functional transcranial Doppler ultrasound. This technique measures changes in blood flow in the brain. Two pilot studies found that tDCS resulted in small gains in spelling performance when administered during brief spelling tasks. In 2016, we will conduct our first full-scale training study. We hope to then extend this project to investigate whether tDCS can be used to enhance interventions for people with dyslexia and severe spelling problems.
Using cognitive theories to inform our understanding of left ventral occipitotemporal dysfunction in dyslexia
Individuals with dyslexia show reduced neural activation of the reading network. Whereas it is generally accepted that dysfunction of the dorsal system reflects phonological deficits, the dysfunction of the left ventral occipitotemporal (LvOT) system remains underspecified in relation to cognitive theories of dyslexia. Our project aims to map neural activation of this network onto specific cognitive processes, to determine whether reduced activation is a global marker of dyslexia, or evident only for certain stimulus types or tasks. To do so, we have tested adolescents with and without dyslexia on a range of cognitive and literacy measures and examined their brain activation during fMRI tasks that vary by stimulus type (written words/object pictures) and by task (overt naming/visual discrimination). These contrasts enable us to examine how atypical activation in dyslexia is modulated by these factors. The results will inform our understanding of how brain differences in LvOT relate to cognitive theories of dyslexia, with particular interest in whether the results support an orthographic or phonological interpretation of LvOT dysfunction. We are also examining connectivity between the dorsal and ventral components of the reading system, both during active tasks and at rest.
This project is focussed on early processing in letter and character strings. While skilled readers can effortlessly identify letters and order them, letter-position dyslexics find the task very difficult. For example, they might read wrap and pirates as warp and parties. Letter-by-letter reading is another form of dyslexia that is characterised by a strong sensitivity to letter-confusability, which results in very slow one letter at a time reading of words, something that skilled readers do not do. These findings show that letter identification and letter-position coding are not bullet-proof. In two studies we have found that people do not confuse letters for each other (A vs. H) as much as they confuse digits for letters (A vs. 4) but they confuse H for A if they only expect to see H. Further studies are probing how the visual similarity of transposed letters (LCOAL vs. LOACL) affects reading times in developing readers, skilled adult readers, and in letter-position dyslexics.
Upcoming CCD Seminars
- Wednesday 29th Mar,
Professor Ocke-Schwen Bohn,
"Second language speech learning: Do cross-language phonetic ..."
- Friday 31st Mar,
Professor Paula Fikkert,
"Umlaut in the history of West Germanic with particular focus on Dutch. ..."
- Wednesday 12th Apr,
Dr Danielle Colenbrander,
"Morphological instruction for children with reading and spelling ..."
- Wednesday 19th Apr,
"Beginner guide to Magnetoencephalography (MEG)"