The distinctive effect of providing syllables in letter fluency testing: Literate vs. illiterate elderly persons

Available online at www.sciencedirect.com

ScienceDirect Speech Communication 70 (2015) 42–48 www.elsevier.com/locate/specom

The distinctive effect of providing syllables in letter fluency testing: Literate vs. illiterate elderly persons Jung Wan Kim a, Si Kyung Lee a, Ji Hye Yoon b,⇑ b

a Department of Speech Pathology, College of Rehabilitation Sciences, Daegu University, Gyeongsan, Republic of Korea Department of Speech Pathology and Audiology, Audiology and Speech Pathology Research Institute, College of Natural Sciences, Hallym University, Chuncheon, Republic of Korea

Received 15 September 2014; received in revised form 17 March 2015; accepted 26 March 2015 Available online 2 April 2015

Abstract The aim of the study reported in this paper was to investigate the effects of providing syllables in a letter fluency test. We compared the performance of 81 healthy elderly participants in a phonemic fluency test to their performance in a syllabic fluency test. These tests formed subtests within a letter fluency test in which participants were given either an initial phoneme or an initial syllable as a cue for word retrieval. Based on their scores on the literacy level test, we divided the participants into four groups, namely pure-illiterate, semi-illiterate, literate, and high-level literate. The pure-illiterate and semi-illiterate participants produced more correct words in the syllabic fluency test than in the phonemic fluency test. In contrast, literate and high-level literate participants produced more correct words in the phonemic fluency test than in the syllabic fluency test. An important implication of these findings is that the effect of providing phonemes/syllables to assist word retrieval might differ among people with varying literacy levels. Providing an initial phoneme to literate people leads to activation of many words beginning with the phoneme. Literates might use a strategy for retrieval of the appropriate words from the plentiful list of candidate words. However, providing an initial syllable to literates might lead to the activation of fewer words, as the pool of candidate words becomes smaller, constraining the word-generating capability. On the other hand, providing the direct and concrete information of a syllable might be more helpful to those with low literacy skills who lack access to such a strategy. Ó 2015 Elsevier B.V. All rights reserved.

Keywords: Phoneme; Syllable; Word retrieval; Literacy level

1. Introduction It is reported by UNESCO (2009) that 20% of the world’s population is illiterate, a percentage which may be regarded as relatively high. The range of percentage of literacy varies widely from one country to another, with the lowest percentage (20%) occurring in Mali, and the highest (99%) in Japan and Finland (UNESCO, 2009). In ⇑ Corresponding author at: Department of Speech Pathology and Audiology, Hallym University, 1 Hallymdaehak-gil, Okcheon-dong, Chuncheon, Kangwon 200-702, Republic of Korea. Tel.: +82 33 248 2224; fax: +82 33 256 3420. E-mail address: [email protected] (J.H. Yoon).

http://dx.doi.org/10.1016/j.specom.2015.03.003 0167-6393/Ó 2015 Elsevier B.V. All rights reserved.

Korea, the percentage of literacy in the total population is 99.8%, which implies a low illiteracy percentage of 0.2% (Lee et al., 2002). However, as a result of factors like the Korean War and poverty, elderly people over 65 years of age are likely to make up the majority of illiterate people in Korea. Over the years, the relationship between illiteracy and cognition/language in this population has received significant attention. A number of previous studies have reported on the effect of literacy on cognitive processing (Ardila et al., 2010; Lezak et al., 2004). In particular, researchers have used brain imaging to compare the neural pathways involved in problem solving in literate and illiterate groups

J.W. Kim et al. / Speech Communication 70 (2015) 42–48

(Castro-Caldas et al., 1998; Deloche et al., 1999). In addition, research has suggested that neural activation related to linguistic function might be affected by literacy (Castro-Caldas et al., 1998; Deloche et al., 1999). Reading and writing require the analysis of words into syllables and syllables into phonemes, as well as the synthesis of phonemes into syllables and words. Other phonological awareness skills involved in reading and writing include grapheme-to-phoneme conversion processes, and both verbal and visual memory. Thus, it seems clear that literacy may impact all spheres of cognitive-linguistic functioning. In clinical settings, word fluency tests requiring verbal output are commonly used to evaluate the linguistic and cognitive abilities of those who are literate and illiterate. In a semantic fluency test, the clinician provides a particular semantic category and the client is required to respond by producing a word in that category. In a letter fluency test, the clinician provides the name of the initial letter (i.e., [ei] for the letter A) and the client is required to respond by producing a word that begins with that letter. Responding appropriately to such word fluency tests requires integrated processing at semantic, lexical, and phonological levels. In particular, a letter fluency test requires phonological awareness in terms of both phonemes and letters. Phonological awareness includes the ability to recognize that words are comprised of a number of sound units, as well as an understanding of the connection between sounds and printed letters. Due to a lack of knowledge of the phoneme–grapheme relationship in the composition of a word (Morais et al., 1979; Reis and Castro-Caldas, 1997), illiterate and semi-illiterate people have been found to perform poorly in letter fluency tests (Kosmidis et al., 2004a; Montiel and Matute, 2006). According to Montiel and Matute (2006) and Morais et al. (1987), illiterate participants in their research performed better at the syllabic level (the syllable being a larger linguistic unit than the phoneme). These researchers, however, did not systematically investigate the difference between the phonemic and syllabic conditions in their word fluency testing. Moreover, the findings reflect the performance of participants using an alphabetic language system. Similar evaluation of performance among populations with non-alphabetic systems ought to be based on languagespecific characteristics. In the Korean case, it has been suggested that the syllable is the basic storage unit for words in the lexicon (Kwon et al., 2005; Nam et al., 1997) because the Korean syllable is easily recognizable by its separation from other sets of characters. From this perspective, providing a syllable in a Korean word fluency test may embody a languagespecific manner of cueing for Korean people and might be more effective in aiding word retrieval than would a letter or a phoneme. With this in mind, we hypothesized that providing a syllable to Korean individuals might maximize their word production during letter fluency testing, regardless of literacy level. According to literacy level, we compared their

43

performance in a phonemic fluency test with that in a syllabic fluency test, both of which fall within a letter fluency test. Based on our findings, however, it appeared that providing one syllable was only effective for the illiterate participants. Performance on phoneme and syllable fluency tests differed from one literacy level group to another. 2. Methods 2.1. Participants We initially recruited 90 healthy elderly (HE) people living in Korea between June 2013 and January 2014. In much of the gerontological literature, people older than 60 years of age constitute the “elderly” segment of the population (Meisheri, 1992; Prakash, 1999). According to Korea’s Welfare of the Aged Act (Ministry Health and Welfare, 2013), however, “elderly” refers to people aged 65 years and older, taking retirement age and activity level into account. Accordingly, we included people aged 65 years and older in our HE group. The Korean Mini-Mental State Examination (KMMSE, Kang, 2006) was administered to all participants. The mean score on the K-MMSE was 22.08 (±4.30). Nine adults with K-MMSE scores less than 1 standard deviation (SD) from the norm was excluded due to a probable influence of decreased cognitive function on their language test scores. Thus, 81 HE people (6 males and 75 females) ranging in age from 65 to 89 years participated in this study. The mean age of participants was 77.03 (5.56) years. Their mean number of years of education was 3.26 (±4.29). All participants were right-handed and spoke standard Korean. In order to measure literacy levels, a literacy test designed by Kim et al. (2014), consisting of six items, was conducted (see Appendix A). This test includes two items from the reading subtest of the Screening Test for Aphasia and Neurologic-Communication Disorders (STAND, Kim et al., 2009), two items requiring answers after reading medication instructions (testing paragraph level comprehension), one item involving writing one’s name (testing word level production), and one item involving writing at least three features of one’s appearance (testing sentence level production). Based on their scores on the literacy test, we divided the participants into four groups: (1) pure illiterate: cannot read and write at all, can occasionally write their own names (scoring 0–1 out of 6); (2) semi-illiterate: can read to a certain extent, can write their names, but cannot write beyond a short sentence level (scoring 2–3 out of 6); (3) literate: can read and write at sentence level (scoring 4–5 out of 6); and (4) high-level literate: can read and write perfectly at paragraph level (scoring a perfect 6). Demographic information for the 81 participants is shown in Table 1. Statistical examination of the homogeneity in terms of age, years of education, and K-MMSE score among the four literacy groups revealed statistically significant differences for all variables.

44

J.W. Kim et al. / Speech Communication 70 (2015) 42–48

Table 1 Demographic features of pure-illiterate, semi-illiterate, literate, and high-level literate individuals.

Age Years of education K-MMSE

Pure-illiterate (n = 23)

Semi-illiterate (n = 23)

Literate (n = 19)

High-level literate (n = 16)

F

79.04 (4.42) .17 (.58) 18.04 (2.88)

76.44 (5.48) 1.70 (2.34) 21.30 (3.25)

77.42 (5.54) 4.47 (3.50) 24.89 (2.62)

73.67 (6.53) 10.25 (3.79) 26.83 (2.29)

2.768* 42.685*** 33.121***

Mean (SD), K-MMSE: Korean Mini-Mental State Examination. * p < .05. ** p < .01. *** p < .001.

2.2. Test materials Word fluency tests require a person to produce as many words as possible, meeting a set of criteria in a set amount of time (Gladsjo et al., 1999). Such tests are generally composed of two types of subtests, namely semantic fluency and letter fluency tests, and are known to be differentially sensitive to different disease states because they require different cognitive processes (Feyereisen et al., 1991). Letter fluency tests are particularly sensitive to frontal lobe dysfunction (Miceli et al., 1981; Ramier and He´caen, 1970). By contrast, semantic fluency tests are affected by deterioration in the structure of semantic knowledge, which is said to characterize Alzheimer’s disease (Butters et al, 1987; Coslett et al., 1991). 2.2.1. Semantic fluency test In this study, participants were asked to respond with as many animal names as possible during a period of 1 min, and then with as many shop items as possible. Admissible words in the animal and shop categories were counted as correct responses. 2.2.2. Letter fluency test The Korean phonemes , , and , and the Korean syllables , , and were presented for phonemic and syllabic fluency testing respectively, with participants having 1 min to provide as many words as possible beginning with each phoneme or syllable. These phonemes and syllables were selected from word lists commonly used in Korean studies of this nature (Kang et al., 2003), and according to an investigation of the frequency of occurrence of Korean phonemes and consonants (Shin, 2008). Responses were recorded verbatim with the conventional restriction that proper nouns, numbers, and multiple forms of the same root word were not permitted. 2.3. Statistical analysis Effects and interaction of demographic variables were identified using 3-way ANOVA. Then, level of education, age, and gender were controlled through ANCOVA, and differences in performance in each test across literacy level groups were compared and analyzed. The score for each

test was converted to an age- and education-adjusted score for post hoc analysis. In addition, a t-test was performed for comparison of the differences across literacy levels in performance on the phonemic and syllabic fluency tests, using the adjusted scores. 3. Results Results for the semantic and letter fluency (phonemic and syllabic) tests are shown in Table 2. Statistically significant differences in performance across literacy groups were found for all tests except the animal task of the semantic fluency test. Bonferroni-adjusted post hoc comparisons revealed that the high-level literate group scored best on the shop task of the semantic fluency test, better than the pure-illiterate (p < .001) and semi-illiterate group (p < .01). Furthermore, statistically significant differences in performance across the four groups were found for all three phonemes and three syllables in the letter fluency test. Across all tests, the high-level literate group achieved the highest scores. A t-test was performed in order to ascertain whether there was a difference in performance on the phonemic and syllabic fluency tests across literacy level groups. The results showed statistically significant differences across literacy groups when comparing their performance across the three phonemes and the three syllables, except in the case of the semi-illiterate group with (/shee-ot/) vs. (/sa/), and the case of the literate group with (/ee-ng/) vs. (/ah/) (Fig. 1). Note, however, that the picture looks different if we consider the total number of words produced for each task type (phoneme vs. syllable) by participants in each of the four groups. The total number of words produced in the syllabic fluency tests was greater than the number produced in the phonemic fluency test for the pure-illiterate and semi-illiterate groups, whereas the total number of words produced in the phonemic fluency tests was greater than those produced in the syllabic fluency tests for the literate and high-level literate groups. 4. Discussion The results of the current study reveal the effect of the literacy level of Korean healthy elderly people on their

J.W. Kim et al. / Speech Communication 70 (2015) 42–48

45

Table 2 Semantic, letter fluency test scores of pure-illiterate, semi-illiterate, literate, and high-level literate individuals.

SF-animal SF-shop LF – (/gi-awk/) LF – (/shee-ot/) LF – (/ee-ng/) LF – (/ka/) LF – (/sa/) LF – (/ah/)

Pure-illiterate M (SD)

Semi-illiterate M (SD)

Literate M (SD)

High-level literate M (SD)

F

8.83 (2.50) 9.91 (3.91) 1.30 (1.55) .96 (1.55) .57 (1.20) 1.96 (1.80) 2.09(2.02) 2.26(1.74)

10.30 (3.02) 10.57 (3.45) 2.22 (2.59) 2.39 (2.78) 2.09 (2.29) 2.96 (1.87) 3.26(1.91) 3.61(1.95)

11.68 (3.80) 12.16 (3.53) 4.95 (2.39) 6.32 (4.27) 4.84 (3.53) 4.26 (2.45) 5.32(1.11) 4.79(1.93)

15.17 (5.39) 16.08 (4.85) 9.08 (4.56) 11.17 (5.17) 9.83 (3.35) 7.25 (4.14) 8.25(4.05) 7.25(2.22)

1.255 3.448* 4.830** 6.833*** 8.237*** 2.693* 6.407** 6.045**

SF: semantic fluency, LF: letter fluency. * p < .05. ** p < .01. *** p < .001.

performance in word fluency tests.1 It is hardly surprising that the high-level literate group showed the best performance. Word fluency tests require participants to produce words cued by particular letters or specific semantic categories, whereas word retrieval in general is driven by the specific concept to be expressed. Word fluency tests are essentially meta-linguistic in nature as they require participants to perform a controlled search of the lexicon. Metalinguistic knowledge is likely to be influenced by education and learning. Therefore, it might be suggested that a higher level of education leads to better performance on such tests. In terms of error patterns, illiterate people showed non-words or real words starting with other (non-target) phonemes or syllables, while literate people produced proper nouns starting with target phonemes or syllables. Among word fluency tests, letter fluency tests are regarded as more difficult for illiterate people (Ardila et al., 2010). In general, Hangul learning begins at the preschool age (approximately 4–7 years). At this time, children do not learn consonants and vowels separately, but learn to perceive syllable types in a form in which vowels are combined with consonants. Later, as they pass through reading and writing lessons in elementary school, they acquire the phonological awareness needed for the analysis of the qualities of phonemes contained in each syllable. Much research has been conducted on the correlation between reading development and phonological awareness. It has been suggested that children develop a variety of other skills beyond reading and writing through formal schooling (Brama˜o et al., 2007; Petersson et al., 2001), such as vocabulary, executive functioning, working memory, associative learning, and visuospatial and visuomotor skills. Such learning may be related to improved

1

One thing that we need to mention here is that, unlike the consonant “ ” and “ ” , the consonant “ ” does not have sounding information in Korean phonetic system. Thus, the words which should be retrieved from the “ ” are actually started from following vowels not from the consonant “ ” in pronunciation (sounding) sense. In terms of the difference on sounding information among these, that might affect to the results of any group.

performance in many neuropsychological tests (Kosmidis et al., 2004b; Vlahou and Kosmidis, 2002). Phonological processing is an area of cognitive functioning that has drawn considerable attention from researchers in the field of illiteracy (Kosmidis et al., 2006). In previous studies, one of the difficulties experienced by illiterate people involves the explicit processing of information on the basis of phonological characteristics (Kosmidis et al., 2004a). Thus, it would appear that the pure-illiterate and semi-illiterate healthy elderly people in the current study, who had fewer than 2 years of schooling, had not acquired the necessary knowledge of grapheme–phoneme correspondence. An interesting result in the current study was that performance on phoneme and syllable fluency tests differed from one literacy level group to another. Hypothetically, providing one syllable might assist direct access to the lexicon for word retrieval, possibly more so than providing one phoneme. Therefore, we expected that providing one syllable would lead to higher word production in all participants, regardless of literacy level. However, it appeared that providing one syllable was only effective for the illiterate participants. We suggest a two-fold explanation for the contrasting response patterns across literacy levels. First, it may be explained from the perspective of a word retrieval strategy. There are several psycholinguistic word retrieval models, which present slightly different points of view in terms of activation flow. However, the concept of components related to word retrieval processing is similar across models (Ellis and Young, 1998; Levelt et al., 1991). Within these models, the stages of the letter fluency process are as follows: (1) recognition of the instruction (to produce words beginning with a particular letter), (2) access to the phonological–lexical–semantic representation, (3) lexical entries are retrieved based on orthographic representations, and (4) the entries are converted to a motor articulatory program for verbal output. Through this process, both appropriate candidate words (i.e., words beginning with particular letters) and less appropriate words (i.e., words beginning with other letters) in the phonological output lexicon are activated. In order to retrieve appropriate

46

J.W. Kim et al. / Speech Communication 70 (2015) 42–48

Fig. 1. Differences in performance with phoneme/syllable initiation on letter fluency test scores of pure-illiterate, semi-illiterate, literate, and highlevel literate individuals. * p < .05, ** p < .01, *** p < .001.

words, activation at the phonological–lexical representation level inhibits the activation of words that do not share initial phonemes and aids the selection of the most appropriate word among candidate words that do share the initial phoneme. The cognitive processes involved, such as activation, inhibition, and word selection, are known to be influenced by literacy level (Ratcliff et al., 1998). The Korean dictionary (Lee, 2014) contains approximately 20,000 words starting with the phoneme . When the Korean consonant is combined with the vowel , as in the syllable , the number of words with this initial syllable decreases substantially, from approximately 20,000 to 1600 words. The larger the number of appropriate candidate words, the higher the possibility of word

production by people in word fluency tests (Yoon et al., 2010). Thus, there might have been a smaller number of candidate words in Korean syllabic conditions than in Korean phonemic conditions. Interpreting the results within the framework of word retrieval models, providing an initial phoneme to literate people leads to activation of many words beginning with the phoneme. Then, literates might use a strategy for retrieval of the appropriate words from the plentiful list of candidate words. However, providing an initial syllable to literates might lead to the activation of fewer words, as the pool of candidate words becomes smaller, constraining the word-generating capability. On the other hand, providing a syllable may be more effective for illiterate people. Because illiterate people might not use the literate selection strategy effectively (Kim et al., 2007), it is more useful to have a smaller pool of words activated than a plentiful list of candidate words. The second part of our explanation for illiterate participants showing better performance with syllables than with phonemes comes from the perspective of language-specific characteristics. English letters that comprise words are horizontally arranged and syllables forming words are not visually segmented. However, in written Korean, basic consonants and vowels are combined to form more complex characters, sometimes referred to as block characters, which contain exactly one syllable. A syllable is easily recognizable by its separation from other sets of characters. For example, in the case of (/san/), (which is equivalent to the English letter ‘/s/’) is combined with (which is equivalent to the English letter ‘/a/’) and (which is equivalent to the English letter ‘/n/’) to form a word that means “mountain”. Due to this visuospatial aspect, it has been suggested that visuospatial representations of block syllabic forms as a “picture-like image” may be rigidly stored in the lexicons of Korean people (Kim et al., 2007; Yoon et al., 2012), and indeed that the syllable is the minimum storage unit in the Korean lexicon (Nam et al., 1997; Kwon et al., 2005). Consequently, Korean syllable might play a special role in invoking lexical entities for word retrieval in illiterate people who lack the phonemic knowledge. The significance of this study is two-folds. First, this is the first study in which the performance of people of varying literacy levels is systematically compared on syllabic and phonemic fluency tests. The central implication of the results of this study is that language-specific characteristics of Korean syllables may play a role in word retrieval over and above the traditional word retrieval strategies suggested in the literature, and that literacy level may determine the contribution of each. Second, the aging of society has been increasingly accelerated with the passage of time. This study suggests that cognitive training as part of clinical applications for the elderly may be assisted by the presentation of syllabic or phonemic cues hierarchically or differentially according to literacy level. In terms of the language-universal application, further research is needed

J.W. Kim et al. / Speech Communication 70 (2015) 42–48

to clarify the effect of providing syllables for word retrieval in other languages, such as English or Italian. In addition, with respect to neuro-anatomical correlates, many reports have demonstrated activation of different brain regions across tasks in illiterate and literate people. The present results suggest that it may be helpful to identify the brain regions activated by the provision of syllabic and phonemic cues. Acknowledgement This work was supported by a 2012 Academic Research Grant funded by Daegu University (Grant number: 20120243). Appendix A. Literacy test 1. Read the following phrase. (1 point) A mouse has fallen into the water 2. Read the following sentence and choose the correct answer. (1 point) Where did it fall into? h cave h alcohol h water h fire [Paragraph about method of taking medicine.] 3. When should adults take this medicine and how much should they take? (1 point) 4. Who can take this medicine? (1 point) 5. Write your name. (1 point) 6. Write down at least three features of your appearance (in a complete sentence). (1 point) References Ardila, A., Bertolucci, P.H., Braga, L.W., Castro-Caldas, A., Judd, T., Kosmidis, M.H., Matute, E., Nitrini, R., Ostrosky-Solis, F., Rosselli, M., 2010. Illiteracy: the neuropsychology of cognition without reading. Arch. Clin. Neuropsychol. 25, 689–712. Brama˜o, I., Mendonc¸a, A., Faı´sca, L., Ingvar, M., Petersson, K.M., Reis, A., 2007. The impact of reading and writing on visuo-motor integration task: a comparison between illiterate and literate subjects. J. Int. Neuropsychol. Soc. 13, 359–364. Butters, N., Granholm, E., Salmon, D.P., Grant, I., Wolfe, J., 1987. Episodic and semantic memory: a comparison of amnesic and demented patients. J. Clin. Exp. Neuropsychol. 9, 479–497. Castro-Caldas, A., Petersson, K.M., Reis, A., Stone-Elander, S., Ingvar, M., 1998. The illiterate brain. Learning to read and write during childhood influences the functional organization of the adult brain. Brain 121, 1053–1063. Coslett, H.B., Bowers, D., Verfaellie, M., Heilman, K.H., 1991. Frontal verbal amnesia: phonological amnesia. Arch. Neurol. 48, 949–955. Deloche, G., Souza, L., Braga, L.W., Dellatolas, G., 1999. A calculation and number processing battery for clinical application in illiterates and semiliterates. Cortex 35, 503–521.

47

Ellis, A.W., Young, A.W., 1998. Human cognitive neuropsychology. Erlbaum, Hillsdale, NJ. Feyereisen, P., Pillon, A., De Partez, M.-P., 1991. On the measures of fluency in the assessment of spontaneous speech production by aphasic subjects. Aphasiology 5, 1–21. Gladsjo, J.A., Schuman, C.C., Evans, J.D., Peavy, G.M., Miller, S.W., Heaton, R.K., 1999. Norms for letter and category fluency: demographic corrections for age, education, and ethnicity. Assessment 6, 147–178. Kang, Y., 2006. Normative data on the Korean Mini-Mental State Examination (K-MMSE) in Korean elderly. Kor. J. Clin. Psychol. 25, 1–12. Kang, Y., Chang, S.M., Na, D.L., 2003. Seoul Neuropsychological Screening Battery-II. Human Brain Research & Consulting, Seoul. Kim, H., Na, D.L., Park, E.S., 2007. Intransient vowel–consonant position in Korean dysgraphia: evidence of spatial–constructive representation. Behav. Neurol. 18, 91–97. Kim, H., Heo, J.H., Kim, D.Y., Kim, J., 2009. Screening Test for Aphasia & Neurologic-Communication Disorders. Hakjisa, Seoul. Kim, J.W., Yoon, J.H., Kim, S.R., Kim, H., 2014. Effect of literacy level on cognitive and language tests in Korean illiterate older adults. Geriat. Geront. Int. 14, 911–917. Kosmidis, M.H., Tsapkini, K., Folia, V., Vlahou, C.H., Kiosseoglou, G., 2004a. Semantic and phonological processing in illiteracy. J. Int. Neuropsychol. Soc. 10, 818–827. Kosmidis, M.H., Vlahou, C.H., Panagiotaki, P., Kiosseoglou, G., 2004b. The verbal fluency task in the Greek population: normative data, and clustering and switching strategies. J. Int. Neuropsychol. Soc. 10, 164–172. Kosmidis, M.H., Tsapkini, K., Folia, V., 2006. Lexical processing in illiteracy: effect of literacy or education? Cortex 42, 1021–1027. Kwon, Y., Kim, H., Sin, J., Nam, K., 2005. The representation type and unit in Korean spoken word recognition. Kor. Soc. Cognit. Sci., 141– 144 Lee, H., 2014. A Dictionary of the Korean Language. Minjungseorim, Korea. Lee, H.S., Lee, J.H., Ahn, D.H., Byeon, J.I., Park, S.O., Kwon, J.H., Lee, H.S., 2002. Korean Adult Illiteracy Survey. Korean Educational Development Institute, Seoul. Levelt, W.J.M., Schriefers, H., Vorberg, D., Meyer, A.S., Pechmann, T., Havinga, J., 1991. The time course of lexical access in speech production: a study of picture naming. Psychol. Rev. 98, 122–142. Lezak, M.D., Howieson, D.B., Loring, D.W., 2004. Neuropsychological Assessment. Oxford University Press, New York. Meisheri, Y.V., 1992. Geriatric services–need of the hour. J. Postgrad. Med. 30, 103–105. Miceli, G., Caltagione, C., Gainotti, G., Masullo, C., Silveri, M.C., 1981. Neuropsychological correlates of localized cerebral lesions in nonaphasic brain damaged patients. J. Clin. Neuropsychol. 3, 53–63. Ministry Health and Welfare, 2013. The Elderly Welfare Act . Montiel, T., Matute, E., 2006. La relacio´n entre alfabetizacio´n y la escolarizacio´n con el desempen”o en tareas verbales. In: Matute, E. (Ed.), Lectura y diversidad cultural. Universidad de Guadalajara, Guadalajara, Mexico. Morais, J., Cary, L., Alegria, J., Bertelson, P., 1979. Does awareness of speech as a sequence of phones arise spontaneously? Cognition 7, 323– 331. Morais, J., Bertelson, P., Cary, L., Alegria, J., 1987. Literacy training and speech segmentation. In: Bertelson, P. (Ed.), The Onset of Literacy. Cognitive Process in Reading Acquisition. MIT Press, Boston. Nam, K., Seo, K., Choi, K., Lee, K., Kim, T., Lee, M., 1997. The word length effect on Hangul word recognition. Kor. J. Exper. Cognit. Psychol. 9, 1–18. Petersson, K.M., Reis, A., Ingvar, M., 2001. Cognitive processing in literate and illiterate subjects: a review of some recent behavioral and functional neuroimaging data. Scand. J. Psychol. 42, 251–267. Prakash, I.J., 1999. Aging In India. World Health Organization, Geneva.

48

J.W. Kim et al. / Speech Communication 70 (2015) 42–48

Ramier, A., He´caen, H., 1970. Role respectif des atteintes frontales et de la late´ralisation le´sionnelle dans les de´ficits de la ‘‘fluence verbale’’. Revue Neurol. 123, 17–22. Ratcliff, G., Ganguli, M., Chandra, V., Belle, S., Seaberg, E., Pandav, R., 1998. Effects of literacy and education on measures of word fluency. Brain Lang. 61, 115–122. Reis, A., Castro-Caldas, A., 1997. Illiteracy: a cause for biased cognitive development. J. Int. Neuropsychol. Soc. 3, 444–450. Shin, J., 2008. Phoneme and syllable frequencies of Korean based on the analysis of spontaneous speech data. Kor. J. Commun. Disord. 13, 193–215. UNESCO. Literacy (retrieved 27.03.09).

Vlahou, C.H., Kosmidis, M.H., 2002. The Greek trail making test: preliminary normative data for clinical and research use. Psychol.: J. Hell. Psychol. Soc. 9, 336–352. Yoon, J.H., Na, D.L., Chin, J., Ahn, H., Lee, B.H., Suh, M.K., Kim, G.H., Kim, H., 2010. Is phonological–lexical representation preserved in moderate stage Alzheimer’s disease? Evidence from the efficacy of Korean syllabic cues. Alzheimer Dis. Assoc. Disord. 24, 334–338. Yoon, J.H., Kim, H., Seo, S.W., Chin, J., Kim, J.H., Lee, K.H., Kim, Y.W., Park, E.S., Suh, M.K., Na, D.L., 2012. Dysgraphia in Korean patients with Alzheimer’s disease as a manifestation of bilateral hemispheric dysfunction. J. Neurol. Sci. 320, 72–78.