Evidence of dual route model in adult reading

Introduction

Dual route model is the study that defines the changes in the mode of interpreting printed words to sound. Therefore dual route model is associated with the study of reading and spelling and the various assumptions that arise out of the investigation. These assumptions highlight the fact that the processing of the written language is always accomplished by two separate methods called the lexical and non-lexical routes. These two methods are interactive in nature. Reading and spelling in the lexical method always depends on the creation of word specific orthographic and phonological memory representation.

All the current models of word recognition are based on either the dual route models or connectionist theories. The way a word recognition model accounts for different types of reading behaviours, is the word of success of every model. The dual model theory has withstood most of the tests associated with reading behaviours. However some psychologists do not agree with this theory, instead they suggest a combination of dual as well as other models for deriving positive results in the field of reading behaviours. A number of psychologists have researched in this field and published their results.

Prominent among them is the study reports of Coltheart and Bates, who have demonstrated that an equation derived from dual-route theory accurately predicts reading performance in young normal readers with reading impairment due to developmental dyslexia or stroke. Studies conducted by John Marshal also have much significance. Marshal concentrated on the fact that the method of skilled reading can be used, particularly the dual route model, for interpreting acquired dyslexia as well as developmental dyslexia.

The basic concept of the dual model explains that non words and irregular words are two independent routes that lead from the written word to the noise. There is also the direct access route which is also known as the lexical route. This route deals with the pronunciation of irregular words. In this method reading is always continued through the lexicon for the purpose of pronouncing words correctly. Then there is the Grapheme phoneme conversion route which is also called the sub lexical route. In this route reading is continued without accessing the lexicon.

All regular words and irregular words should be read from the lexicon, this is the suggestion given by the dual route model. A strong judgement in this regard is that the steadiness of the word always effects the periods it takes to name it. Moreover there is also a relation with its rate of recurrence. Low recurring words are more affected than high recurring words.

Some words having the same frequency are not made similar, particularly inconsistent words. Moreover regular words are sometimes pronounced irregularly. According to experts a simple dual route model cannot explain all these outcomes. A dual route model can be used for the prediction of real words though the lexican route. However we can see that steadiness, rate of recurrence and uniformity effects reaction times.

The suggestion that non words are read by the GPC route is not correct. Strong findings suggest that non words processing s are affected by the lexicon. Previous normal reading process of a person is affected by the impact of acquired dyslexia the affected person will suffer some kind of a shock in the left side of his brain. In this case the dual route model predicts a double disassociation.

With respect to surface dyslexia, a person makes over regularisation mistakes for irregular words. The person can read words and non words if they are regular. The person will not show any case of image ability. This is due to the fact that the route remains undamaged.

Another type of dyslexia which is commonly found is the Phonological dyslexia.

In this situation, patients are poor to read non words. Moreover real words are easily read. Experts suggest that, here the GPC route may be damaged and the lexical route is undamaged.

It is often argued that the Phonological dyslexia and the surface dyslexia are separate. They do not associate themselves. Therefore the dual route model is supported. But in other types of dyslexia the situation can be more complicated.

Another type of dyslexia is the deep dyslexia. This type is characterised by poor reading of non words, semantic reading mistakes for real words, and problems associated with the GPC route. Moreover the patients under this category have an injury to a system which usually allows them to read through a semantic system.

This type of dyslexia cannot be regarded as a syndrome. But in some circumstances it has to be regarded because evidences support it. Another point of argument about this type of dyslexia is that it may be a result of reading using the right hemisphere of the brain. There are assumptions that this type of dyslexia is a syndrome which is caused by a wound in the left hemisphere. If that is correct then the conclusion should be that reading is possible through semantic system in such a way that the dual route model cannot be taken into account.

According to psychologist Knight R.A from his journal article titled, 'Language, history and use', there are alternatives to the dual route model also. They are the three routes model, Analogy theories and the Connectionist models. In the three route model, the lexical route is split into GPC route, the route through lexicon and the route direct from the lexicon. Analogy theories are single route theories. a major drawback of this theory is that it is sometimes unpredictable regarding the pronunciation of non words. The connectionist models are single route theories where there is no storage for individual words. The drawback of this model is that it also fails to predict the pronunciation of non words (Knight, Language History and Use).

The dual route model has become a standard model for both normal as well as abnormal reading of adults. There is very strong evidence that there is twofold disassociation of phonological and surface dyslexia in the dual route model. The dual route model had assumptions that the conversion of grapheme-phoneme depends on a theoretical set of rules which is similar to the plans in a logical system. Other models of normal and abnormal reading development were developed on the basis of the dual route model of mature reading. The dual route theory is a very comprehensive theory. It explains both normal and abnormal adult reading and as well as normal and abnormal reading development (Pennington, 2002, p.292).

In the semantic system, spoken as well as written words always turn on mechanically the matching theoretical representations. All familiar words used whether they are customary or non customary can be processed through the lexical route. This is possible in terms of their letter indicated by sound relationships. However the non lexical route makes use of the sub word level process which is based on the sound- spelling association rules success of non lexical route always depends on non words and regular words that is always associated with the English phoneme grapheme conversion laws but when there is irregular words which does not obey these set of rules, it fails to produce a correct response.

All the efforts undertaken to read or write through non lexical method can result in regularization mistakes. Studies by psychologists Rapcsak, Henry and others revealed that dual route model contains purposeful mechanisms that are exclusive to both the lexical and the non lexical routes but no one can say that the two procedures are totally independent. Dual route theory always maintains that lexical route will help in delivering a true response for irregular words. The non lexical route is essential for the reading and spelling of non words correctly.

For the interpretation of the written language of persons having acquired alexia / agraphia, dual route models have provided a powerful hypothetical structure. It is possible to identify the damaged cognitive module of neurological patients by specifying the practical architecture of the written language processing system. For example, the destruction that is caused to the lexical route will give rise to surface dyslexia distinguished by disproportionate complexity in reading and spelling irregular words. However damage to the non lexical route will result in phonological dyslexia, distinguished by very poor reading and spelling of non words. Apart from all these applications, dual route theory can also be used for producing quantitative predictions related to reading and spelling.

It is possible to predict the accuracy of a persons and the performance on irregular words as well as non words. The basic concept of dual route models is that using lexical method irregular words can only be read correctly whereas by using non lexical strategy non words can be read correctly. The capability of lexical and non lexical routes are proved when a person will be able read accurately proportions of irregular words. According to dual route theory, both the lexical and non lexical route can process regular words and therefore the accuracy of reading can be predicted (Rapcsak, Henry, Teague, Carnahan & Beeson. 2007. p.2519-2524)

A study conducted by John marshal, another renowned psychologist in the year 1984 clearly emphasised that the fact that the method of skilled reading can be used, particularly the dual route model, for interpreting acquired dyslexia as well as developmental dyslexia. This claim became controversial and many other researchers argued that the dual model is quite in appropriate in the way to understand development in reading and developmental disorders in reading. However further research in this area by psychologists Castles, Bates & Coltheart revealed that John Marshals claim was true (Castles, Bates & Coltheart. 2006. pp. 871-892 (22)).

Many researches have been undertaken for checking the effectiveness of dual route model. The dual route cognitive models of reading assume that the lexical and sub lexical routes are functionally different processes. Researchers Joubert, Beauregard and others have tested this hypothesis by using three experimental conditions. One of them was lexical and the other two were sub lexical. Frequent words represented the lexical condition whereas in the first sub lexical condition was represented by non words made of low frequency sub lexical units. These words did not resemble real words. Rare and regular words consist of the second condition. Comparisons were made between these reading tasks with a baseline condition which consisted of consonant string viewing. Moreover silent pronunciation of one or two of these consonants was also done along with this process.

By analysing the results of this experiment it was found that the lexical condition which consisted of regular words generates many different peaks of activations in the border area of the left super marginal region. Whereas the other two sub lexical tasks were engaged always in the area of the left prefrontal inferior cortex (Joubert, Beauregard, Walter, Bourgouin, Beaudoin, Leroux, et al. 2004. p. 9-20).

Renowned psychologists Hanley and Gard advocated that developmental surface dyslexia appears in adults. They took reading and spelling tests on two adult students and the results varied. Both these adults were affected significantly by spelling regularity. The psychologists arrived at the conclusion that the irregular words are harder to study that difficult ones, therefore a person is forced on to depend on lexical strategy during the course of reading (Funnel. 2000. P.138).

The dual route models have always come under attack in recent years the theories framed by dual route experts are not always effective in adults. Most of the criticisms in the dual route model have come in the field of adult readings. As a result of criticisms levelled against dual route models, connectionist's models were developed (Treiman. 1993. p.34)

Studies made on languages which were opaque suggested that the lexical and non lexical processing consists of independent routes. Where as in the case of transparent languages, there is no strong evidence for independent routes. The dual route in transparent languages like in opaque languages has created an arena for discussion. By studying the neurologically damaged patients who are unable to read unknown words or non words, the existence of these routes can be seen clearly (Plasencia, Dorado & Serrano. 2008. p. 48 -54).

According to a study conducted by two well known psychologists, Castles and Coltheart, process deficit models of disability became popular due to the developmental problems associated with reading disability associated with brain damages. The reasons for the popularity of the process models is because of the simplicity of its framework moreover the approach of a process deficit model gives a basis for a logic of instructions stuck by the strength and weakness of the readers. But unfortunately many researches have failed to provide evidences that support process models. The fact stated in the process model that disabled readers can be categorized into distinct types and the diagnosing task of distinguishing disabled readers from the poor readers has no supportive evidence (New Directions in Research, 2006).

In another experiment by Bertelson, two hypotheses were tested with relation to the development of access towards lexical meaning. They were initial indirect access and initial direct access. Conclusive proof relating to the testing pointed out that these hypotheses were not strongly supported. And it also recommended that the direct and indirect access could be used for early reading.

The involvement of these procedures always changed during its course of development. The previous assumption about standard dual model was that both accesses assumed the basic theoretical tests, went wrong. Based on evidence derived from these tests, the standard dual route model was always criticised. Instead of the standard dual model, a single model was found suitable (Bertelson, 1987. p.113).

The dual route cascaded model is a computational model related to visual word recognition and loud reading. It is the only computational model that is used to perform the two tasks i.e., lexical reading and aloud reading. The DRC model has its own limitations. A common claim attributed to this model is that it can explain everything. But this claim becomes false as far as the case of reading aloud and visual lexical decision is concerned. With various paradigms that are masked, facts suggest that phonological properties of very temporarily presented stimuli can influence performance in reading aloud and lexical decision tasks. The dual route cascading model or the DRC model is used for pronunciation computing from print through lexical and non lexical procedures. Non words cannot be read correctly under the lexical procedure because they are not present in the lexicons. (World Recognition Process in Reading. 2005).

Many studies have been undertaken to prove the effectiveness of the dual models. In one study done by psychologists Basso, Burgio and Prandoni, of the neurological clinic, Milan University, a group consisting of 21 people were selected having specific disorders and treated with a dual route model. Another group consisting of 23 persons were treated with methods using single route models. By comparing the treated patients from both the groups, it was found that the first group treated with the dual method showed significant improvement than the second group in many areas. But the benefits received from reading of isolated words, homophones etc did not improve the reading of passages. This suggested that treatments using dual models should include exercises containing passages or sentences (Tressoldi. 2000).

In another test, persons who did not have previous knowledge about French were asked to learn 12 French words. After getting their representations, they were given pictures and were asked to write the baseline. This experiment was repeated for three times. After 10 minutes these pictures were presented again and were asked to write words. After a week, these participants were again requested to do the same exercise. The results showed significant difference. The out come was that when irregular orthographic representations are acquired, it is supported by knowledge of orthographic representations (BASSO, BURGIO, & PRANDONI. 1999.p.405-412).

Arguments against dual route model theories.

Connectionist modelling techniques have spawned the debate between single and dual route accounts of the cognitive process. The necessity of dual route processing accounts in the psychology of the language has been questioned by connectionist's models. Some psychologists have also suggested that a single route model is enough for the learning of past tense formations and idiosyncratic formations which are rule based. It is also sufficient to learn rule based method of pronunciations and idiosyncratic pronunciations.

Basically there are two types of criticisms that are levelled against the dual route models. The first one is that morphology is the rule which directs association of the bilingual's lexical representations. It is argued that morphology can alone produce clear effects of independence in a sole undifferentiated structure. The second criticisms said about the dual model are that the word recognition in one language may be assisted if it is similar to a word in another language. Connectionist modelling techniques have spawned the debate between single and dual route accounts of the cognitive process.

The necessity of dual route processing accounts in the psychology of the language has been questioned by connectionist's models. Some psychologists have also suggested that a single route model is enough for the learning of past tense formations and idiosyncratic formations which are rule based is also sufficient to learn rule based method of pronunciations and idiosyncratic pronunciations.

There are also other types of criticisms that are levelled against the dual route models. The first one is that morphology is the rule which directs association of the bilingual's lexical representations. It is argued that morphology can alone produce clear effects of independence in a sole undifferentiated structure. The second criticism said about the dual model is that when motivations show some kind of resemblance between two languages, the necessity of word recognition arises. Moreover cross language meddling is found even though there is clear proof for independence of lexical presentations (Thomas, Bilingualism and the Single route).

Conclusions

Connectionists theories have originated as a result of the debate between single and dual route models in the area of language processing. The debate between single route and dual route has often arisen when the necessity of postulation of a rule based processing mechanism that would motivate aspects of cognition. The fundamental problem here is that all rules have exceptions and all processors based on rules should have an exception mechanism to accompany it. The other side of this debate between dual and single model is that whether there is necessity to postulate one or two realistic processing mechanisms to motivate a given cognitive ability.

Dual routes are not effective in all areas, particularly in the field of bilingual word recognition. To those areas where dual theories are insignificant, it is wise to work with connectionists models which are regarded as an exiting prospect. Single route models when compared with dual route models should overcome many hurdles to get accepted. However they can be used in situations where dual route models fail to produce results. The single route models always characterise stimulus response compatibility effects to the difficulty of transaction from stimulus to response whereas dual route models suggest that compatibility effects reveal in part an extra direct or routine creation route.

In consistency is one of the major drawbacks suffered by these models. Form the outcomes of studies undertaken; it is evident that the DRC model can stimulate the reading aloud task as well as the lexical decision task. The stimulation given by the DRC model to these tasks cannot be matched by any other models. Therefore the DRC model is regarded as the most successful computational model discovered so far in the field of reading behaviors.

Nowadays word recognition models are constructed on the basis of dual route as well as connectionist models. Dual route has its own advantages and limitations in adult reading. In cases where dual method is not effective, connectionists' models can be used. Or a combined format can be used. Combination of these two models will always give satisfactory results in word reading. (Shinoda. 2005).

References

BASSO, A. , BURGIO, F. & PRANDONI, P. 1999. Acquisition of output irregular orthographic

representations in normal adults: An experimental study. Retrieved from

http://journals.cambridge.org/action/displayAbstract;jsessionid=0B6AC8EC3378DCD2C42DFE0963CBF5B3.tomcat1?fromPage=online&aid=49571 on August 23, 2008.

Bertelson, P. 1987. The Onset of Literacy. Cambridge: MIT Press.

Castles, A.,Bates, T., & Coltheart, M. 2006. John Marshall and the developmental dyslexias.

Aphasiology, Volume 20, Numbers 9-11, -11 pp. 871-892(22). Retrieved from

http://www.ingentaconnect.com/content/psych/paph/2006/00000020/F0030009/art00004;jsessionid=kldvo0vfe03v.alice?format=print on August 22, 2008.

Funnel, E. 2000. Case Studies in the Neuropsychology of Reading. London: Psychology Press.

Joubert, S., Beauregard, M., Walter, N., Bourgouin, P., Beaudoin, G., Leroux, J.M. et al. 2004.

Brain and Language . Vol. 89 9-20. Retrieved from http://www.sciencedirect.com

Knight R.A. Language History and Use - Psycholinguistics. Retrieved from

http://www.rachaelanne.co.uk/teaching/psych/reading_03_HO.doc

on August 22, 2008.

New Directions in Research. A Transactional Perspective on Reading Difficulties and Response

to Intervention. 2006. Retrieved from

http://www.reading.org/Library/Retrieve.cfm?D=10.1598/RRQ.41.1.7&F=RRQ-41-1-McEneaney_2.html on August 23, 2008.

Pennington, B.F. 2002.The Development of Psychopathology. New York. Guliford Press.

Plasencia, P.M., Dorado, J.I. & Serrano, J.M. 2008. Evidence for Linguistic Deficit in Nonlexical

Processing in Reading. A Study of a Spanish-Speaking Patient. The Spanish Journal of

Psychology, Vol.11, No.1, 48 -54. Retrieved from

http://www.ucm.es/info/Psi/docs/journal/v11_n1_2008/art48.pdf on August 23, 2008.

Rapcsak S.Z, Henry M.L, Teague,S.L., Carnahan,S.D.& Beeson,P.M. June 18, 2007. Do Dual-

Route Models Accurately Predict Reading and Spelling Performance in Individuals with

Acquired Alexia and Agraphia? Neuropsychologia Vol. 45(11): 2519-2524. Retrieved from

http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1988783 on August 21, 2008.

Shinoda, K.2005. Speaker adaptation techniques for speech recognition using probabilistic

models. Retrieved from http://www3.interscience.wiley.com/journal/110546421/abstract

on August 23, 2008.

Thomas, M.S.C. Bilingualism and the Single route / Dual route debate. Retrieved from

http://www.psyc.bbk.ac.uk/people/academic/thomas_m/Thomas_Cogsci98.doc on August 22,

2008.

Treiman, R. 1993. Beginning to Spell. Washington, DC: Oxford University Press, US

Tressoldi, P.E. 2000. Treatment of Specific Developmental Reading Disorders, Derived from

Single- and Dual-Route Models. Retrieved from

http://ldx.sagepub.com/cgi/content/abstract/33/3/278 on August 23, 2008.

World Recognition Process in Reading. 2005. Retrieved from

http://www.blackwellpublishing.com/content/BPL_Images/Content_store/Sample_chapter/9781405114882/Snowling_sample_chapter1.pdf on August 23, 2008.