Semantic Impairment with and without Surface Dyslexia - Essay Example

Running Head: SEMANTIC IMPAIRMENT Semantic Impairment with and without Surface Dyslexia: Implications for Models of Reading Semantic dementia, in a nutshell, is characterized by the deterioration of one’s semantic memory – that which deals with meaning, understanding, and concept-based knowledge in general – while other cognitive functions remain intact (The Brain From Top to Bottom, accessed 12/01/09), and, while usually concurrent with, is distinctly different from similar disorders such as Alzheimer’s. It is produced by pathological substrates that mainly affect the frontal or temporal neocortex – a trait it shares with two other syndromes falling under frontotemporal dementia (FTD). It is associated with a number of conditions such as surface dyslexia, which will also be discussed in this paper. This paper’s objectives, therefore, are to discuss in detail the symptoms of both conditions, as well as the results of a case study conducted testing the correlation of the two. Also intended is an objective discussion of the case study’s strengths and weaknesses, especially where the results are concerned. Introduction The Association for Frontotemporal Dementias (accessed 12/01/09) further explains that those with semantic dementia retain the ability to speak fluently, but that they become more and more unable to express themselves intelligibly. Specifically, in the latter stages, they start to lose their understanding of even basic words – a serious impediment to communication, to be sure. In time, they start to experience behavioral, social and motor difficulties as well – all of which are common to FTDs in general. Dyslexia in general, adds Valerie Suydam (2008), is a language disorder in which one diagnosed has difficulty reading and/or writing properly, and is distinguishable from normal reading mistakes in that a dyslexic individual tends to commit the same mistake repeatedly. It has an equal chance of manifesting in both genders and of all intelligences. In particular, surface, or orthographic dyslexia, is a subtype in which words are pronounced the way they are spelled (such as “stra-yigt” for “straight”). This, according to Elise Caccappolo-van Viet et al (2004), is more prevalent in such languages as English and French due to the relationship of orthography with phonology – few words are spelled the same way they are pronounced. According to Anna Woolams et al (2007), the most glaring sign of surface dyslexia is the regularization error – words which are supposed to have exceptional spelling-to-sound correspondences are instead read according to how the word looks like it should be pronounced, rather than its actual pronunciation (“sew” being pronounced the same as “new”, for example). This is usually more common with low-frequency words, but can be observed even with high-frequency words once the condition has reached a certain point. It usually manifests in patients with static brain lesions, neurodegenerative diseases (such as semantic dementia, also discussed in this paper), and Alzheimer’s (it is important, however, to remember that surface dyslexia and Alzheimer’s disease are two distinct conditions). Whatever the case, though, sufferers of surface dyslexia also experience a significantly deteriorated semantic memory. Tracy O’Brien (2009) further explains how those suffering from surface dyslexia find it hard to store mental representations of words, especially phonetically irregular words (read: those spelled… unusually). The underlying problems usually relate to memory – specifically semantic memory – and coding skills. She also adds how sufferers of this condition are usually used to only one way to pronounce particular consonants – hence the example cited above. The Study Two cases were studied by Blazely et al (2005) in relation to these two disorders. The first, a 69-year old lady referred to in the article as EM, known for forgetting names and faces but not events such as birthdays and appointments. On a related note, she had little trouble memorizing routes, going about her business, or travelling long distances, after moving to a new home. Her family began to note her increasingly obsessive-compulsive behavior (having the exact same meals every breakfast, lunch and dinner, at the exact same time), as well as her over-idealization of the past. Further tests showed that while her day-to-day memory and spontaneous speech remained flawless (in the latter’s case, at least where articulation and grammar were concerned), her performance in the Boston Naming Test, wherein one was to name specific objects used in the test, was poor. Her verbal definitions of animals and object, as well as her performance in tasks of verbal generativity, were similarly less than satisfactory, yet nothing suggested she had visiospatial, constructional, or perceptual deficits. The second case, meanwhile, was PC, a 66-year old man with a history of agnosia for objects and faces, as well as more obsessive-compulsive behavior. For instance, in addition to the speech problem described above, his wife explained how he became increasingly rigid and inflexible, citing examples ranging from absolute refusal to change brands while doing groceries, to cleaning the pool up to 10 times a day. It was also discovered that he had surface dyslexia when he was asked to read aloud. Both subjects were given the same tests: written and spoken word-picture matching, word reading, and visual lexical decision, in which they were asked to choose a word’s correct spelling based on its pronunciation. All scores were then recorded and compared, with one fact glaringly obvious: though the scores of the two subjects varied, they shared a common point in that they scored highest on the high-frequency parts and lowest on the low-frequency parts. Two models were kept in mind while observing results: the DRC, or the Dual Route Cascaded model, and the one by Plaut, McLelland, Seidenberg, and Patterson, referred to as the PMSP model. The former was shown to include a direct lexical reading route from and between lexical orthography and phonology independent of word meanings, which is also supported by a number of other authors. Under such a model, degradation or even total loss of semantic memory could leave one’s ability to read irregular words intact. In contrast to this, the PMSP model proposes that when semantic memory goes, so does the ability to read irregular words. According to this model, semantically impaired patients also suffer from surface dyslexia and have impaired visual lexical decision – semantic impairment would, in effect, destroy the ability to recognize a string of letters as a real word. It proposes two mechanisms for reading aloud. The first, the phonological pathway, directly converts orthography to phonology, and is the pathway on which the reading of both words non-words depends. Meanwhile, the semantic pathway, as the name implies, entails reading via semantics, and is the pathway crucial to reading exception words aloud normally and properly. Based on the descriptions of both models, semantic dementia and surface dyslexia are described as having to occur together under PMSP, but are completely distinct from each other in the DRC model. However, this is still a possibility under the latter, as semantic damage coupled with an impaired pathway between the orthographical and phonological lexicons would result in both symptoms. Thus, the association of both these symptoms can in fact be applied to both models. More noteworthy is the dissociation between these symptoms – that is, wherein one symptom appears without the other. While this goes against the PMSP model, it fits just fine with the DRC model – according to it, damage to the semantic system only results in semantic impairment, without surface dyslexia. Blazely et al go on to pose a question first asked by Patterson and Hodges in 1992: whether successfully reading exception words aloud reflected the importance of word comprehension, or if the orthographic input was responsible. Data Analysis & Discussion In the case of the two subjects, both PC and EM were shown to have semantic dementia, judging from their scores in the test. However, it was a different story in the case of surface dyslexia. While PC had this, EM could read aloud just fine, and her visual lexical decision remained intact as well. This discrepancy could probably be explained in that semantic memory deteriorates gradually; therefore, its interaction with the lexical system would degrade at a similar rate, rather than in an instant. However, the tests showed that EM’s semantic dementia was even worse than PC’s, as shown by her lower scores. Therefore, her data remains inconsistent with the PMSP model. On the other hand, according to the DRC model, all the tests show is that the only impairment affecting EM is one affecting her semantic system. Under such a model, this would then impair name comprehension, both written and spoken picture-to-word matching, and picture naming; visual lexical decision and reading irregular words was spared. Thus, according to the same model, another impairment exists that afflicts PC with dyslexia. As for the results, since it was assumed that EM’s impaired performance was due to damage to her semantic system, it was expected that the items she succeeds in between tasks will have significant correlation. This was found to indeed have been the case upon further inspection; for all six pairs of tests, the successfully answered items between them had a positive and statistically significant relationship. This in turn proves that her impaired performance really was due to partial damage to her semantic system. The same was done in PC’s case, with addition of results from the visual lexical decision task. Five out of six phi coefficients analyzed in relation to EM’s data were found to have been significant, the last being only marginally so; it was thought that his case then mimicked EM’s in that his problems were caused by one particular problem in the semantic system. His performance in the tasks would seem to agree – his performances in visual lexical decision, as well as in NC, PN, and WPMS, were all found to have nonsignificant phi coefficients as expected. Likewise, his phi coefficient in WPMW was found to have been significant – also as expected. The relationships of accuracy between reading aloud irregular words was no longer tackled, mainly due to the lack of clarity with regard to the predictions to be made. That is, his misreading of words could either be due to failure at the orthographic lexicon, or at the semantic level along with the lexical nonsemantic route, or at the phonological lexicon. And since it was impossible to judge just how much his lexical nonsemantic route was impaired (if at all) as well as how able he was to access the phonological lexicon, it likewise became impossible to predict the correlations between reading aloud and the other tests. Blazely et al conclude their article saying that being able to read irregular words aloud despite semantic impairment suggests a lexical, rather than semantic route for reading. Specifically, if partial semantic information played a part in the subjects’ reading of irregular words, EM would not have done as well as she did, nor would PC have done as poorly. Limitations Despite the accuracy of their research, though, this student has noticed at least one, glaring issue worth thinking about. For a start, as would be expected, these tests were conducted in carefully controlled conditions, when at least one could be observed under normal circumstances (reading aloud, for one). Specifically, this student believes that that factor in particular may have affected the results in some way. For instance, maybe the knowledge of what the results were going to be used for affected the subjects’ performance, for better or worse. Regardless of what researchers may think, some people actually think that being asked to participate in research is a big deal – in fact, as shown by some experiments such as that of the Stanford Prison, these experiments may have a lasting impact on the participants. This may have happened in this case; who knows, maybe EM and PC’s performance may actually be different in their everyday routines. For instance, perhaps EM actually does struggle with reading aloud despite forcing herself not to. Such a limitation, however, could probably be extended to all experiments in general: that the most carefully controlled experiment still cannot substitute for results acquired in the subject’s natural environment. Conclusion Regardless of the above, however, this experiment is still a more or less accurate look at the relationship between semantic dementia and surface dyslexia. At least in the circumstances provided, one was shown not to be dyslexic despite suffering from semantic impairment. All in all, the study seems to have affirmed the DRC model. EM noticeably did not experience surface dyslexia, even if PC did – and his case, by the researchers’ own admission, was rather unusual. Indeed, it proves the idea that a lexical but not semantic route for reading aloud exists, while refuting the notion that semantic impairment comes with the destruction of orthographic and phonological representations. References Blazely, A.M. Ms, Coltheart, M. and Casey, B.J. (2005). Semantic impairment with and without surface dyslexia: Implications for models of reading, Cognitive Neuropsychology, 22(6), 695 -717. Caccappolo-van Viet, E., Miozzo, M., and Stern, Y. (2004). Phonological dyslexia without phonological impairment? Cognitive Neuropsychology, 21(8), 820-839. Memory and learning. (2009). The brain from top to bottom. Retrieved on December 2, 2009 from http://thebrain.mcgill.ca/flash/a/a_07/a_07_p/a_07_p_tra/a_07_p_tra.html#3 O’Brien, T. (2009). Types of dyslexia: three subtypes are orthographic, phonological, and mixed. Retrieved on December 2, 2009 from http://healthfieldmedicare.suite101.com/article.cfm/types_of_dyslexia Semantic dementia (accessed 12/02/09) Association for frontotemporal dementia. Retrieved on December 2, 2009 from http://www.ftd-picks.org/frontotemporal-dementias/disorders/semantic-dementia Suydam, V. (2008). Researching dyslexia: the brain and functional imaging technology. Retrieved on December 2, 2009 from http://disabilities.suite101.com/article.cfm/researching_dyslexia Woolams, A.M., Ralph, M.A.L., Plaut, D.C., and Patterson, K. (2007. SD-squared: on the association between semantic dementia and surface dyslexia’, Psychological Review, 114(2), 316-339. Read More