bs_bs_banner
Geriatr Gerontol Int 2014; 14: 911–917
ORIGINAL ARTICLE: EPIDEMIOLOGY, CLINICAL PRACTICE AND HEALTH
Effect of literacy level on cognitive and language tests in Korean illiterate older adults JungWan Kim,1 Ji Hye Yoon,2 Soo Ryon Kim3 and HyangHee Kim3,4 1
Department of Speech and Language Pathology, College of Rehabilitation Sciences, Daegu University, Gyeongsan, 2Department of Speech Pathology and Audiology, Audiology and Speech Pathology Research Institute, College of Natural Sciences, Hallym University, Chuncheon, 3 Graduate Program in Speech and Language Pathology, Yonsei University, and 4Department and Research Institute of Rehabilitation Medicine, Yonsei University College of Medicine, Seoul, Korea
Aim: Illiterate individuals represent a significant proportion of the world’s population. Acquisition of reading and writing skills influences the functional status of the brain, and consequently alters the performance on cognitive and language tests. Thus, it is important to identify the degree of the impact of levels of both illiteracy and education as potential confounders on test performance in people with neurological communication disorders. Methods: A total of 203 community-dwelling older adults aged 65 years and older were recruited for the present study. Participants were classified into four groups based on the literacy level; pure illiterate (n = 29), semi-illiterate (n = 67), literate (n = 75) and high-level literate (n = 32). The participants completed the Mini-Mental State Examination, Boston Naming Test, Controlled Oral Word Association Test (animal), verb naming, and sentence comprehension tests. Results: The pure illiterate group showed the lowest performance on all five tests. Regression analysis showed that literacy level was the variable that best predicted the performance on cognitive and language tests. Conclusions: These findings suggest that literacy in performance on cognitive and language tests is an important factor in neuropsychological evaluations for older adults. Geriatr Gerontol Int 2014; 14: 911–917. Keywords: cognition, elderly, language, literacy.
Introduction Illiterate individuals represent a significant proportion of the world’s population.1 In 2011, the illiteracy rate of adults aged 15 years and older was 16% of the world’s population.2 In South Korea, as a result of social (i.e. lack of opportunity to attend school, poverty, etc.) and personal (i.e. learning difficulties) reasons, the illiteracy rate is approximately 20% in members of the population over the age of 70 years. In particular, the illiteracy rate is higher in agricultural areas, and 26.4% of older adults aged 65 years and older are illiterate.3
Accepted for publication 15 October 2013. Correspondence: Professor HyangHee Kim PhD, Graduate Program in Speech and Language Pathology, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Korea. Email: [email protected]
© 2013 Japan Geriatrics Society
Although the acquisition of oral language is a natural phenomenon during language development, literacy skills are not acquired naturally. Thus, great effort is required in order to obtain and maintain literacy competencies. The ability to read and write, which is achieved in childhood for most people, could change the morphological and functional status of the human brain,4,5 influencing several linguistic and non-linguistic functions.6–8 Functional imaging studies have reported that literacy and level of education might have an influence on pathways used in the brain for problem solving.9 This implies that neural networks of cognition in illiterate individuals are different from those of literate individuals.10 In addition, illiteracy and lower levels of education constitute risk factors for dementia.11 Higher risk for development of Alzheimer’s disease has been shows in illiterate and less educated individuals, perhaps as a result of a vulnerability to cognitive deterioration.9,12 Illiteracy and lack of formal education attainment have dramatic effects on neuropsychological performance;13–15 therefore, illiteracy represents an important variable in research into dementia.16,17 doi: 10.1111/ggi.12195
| 911
JW Kim et al.
Studies have been carried out for determination of differences in performance on cognitive–language tests of the elderly depending on schooling and literate ability. Previous studies have reported that the level of education is significant in prediction of Mini-Mental State Examination (MMSE)18 scores; and, in particular, the cut-off point of illiterate subjects is close to 13 points, which is lower than that of educated subjects.19 In a study comparing language performance of illiterate adults, adults who had 1–2 years of elementary education15 and adults who had 3–4 years of elementary education, no difference in performance on naming, repetition and semantic verbal fluency tests was observed among groups. However, the illiterate group showed significantly lower performance on language comprehension (i.e. token test) and phonemic verbal fluency tests. When research findings on illiterate adults who showed low performance on naming, verbal fluency,20,21 memory test7,22 and visual construction tasks23,24 are considered together, differences in performance on language tests between literate and illiterate individuals are not limited only to tasks involving use of phonemic skills.13 Illiterate individuals might be at a disadvantage in testing situations, both because the testing environment is unfamiliar25,26 and because they lack appropriate testtaking strategies.27 As the MMSE was developed for literate older adults, illiterate older adults are often diagnosed as having dementia.28 According to one report, 79% of uneducated Korean adults are illiterate.3 This could be reinterpreted as 21% of uneducated adults have a certain degree of literate ability. In addition, because people in Korea tend to feel ashamed of having no official education or little official education, level of education is not proportional to literacy level, and it is more likely for the uneducated or those with fewer years of education. Diverse neuropsychological tests used for diagnosis of dementia, including the MMSE and the Boston Naming Test, do not present any norm with regard to literacy level. Thus, determination of the degree of impact of levels of both illiteracy and education as potential confounders in test performance in those with neurological communication disorders is important. The purpose of the present study was to compare the performance of four groups of older adults with different literacy levels on cognitive–language tests.
Methods
a cognitive–language test over a long period of time, random sampling was impossible. Thus, tests were carried out only for those among the elderly who wished to be evaluated. They had no history of neurological or psychiatric illness. The Korean Mini-Mental State Examination (K-MMSE)29 was administered to all participants. The mean score on the K-MMSE was 24.31 (± 4.34). Nine adults with K-MMSE score less than −1 standard deviation (SD) of the norm were excluded because of the probable influence of decreased cognitive function on the language test score. Thus, 203 HE (64 males, 139 females) ranging in age from 60 to 91 years participated as subjects in the study. The mean age was 74 ± 6.9 years. The mean number of years of education was 6.92 ± 4.35. All participants were right-handed, speaking standard Korean. This study was approved by the institutional review board of Yonsei University College of Medicine (No. 1–2011-0061). For measurement of the literacy level, a literacy test consisting of six items was carried out. The test includes two items from the reading subtest of the Screening Test for Aphasia and Neurologic-communication Disorders (STAND),30 two items requiring selection of answers after reading medicine instructions (phrase level), one item involving writing one’s name (word level) and one item involving writing at least three features of one’s appearance (sentence level). Based on the performance score on the literacy test, we divided the participants into four groups: (i) pure illiterate: cannot read and write at all, can occasionally write their own names (0∼1 out of a perfect score of 6); (ii) semi-illiterate: can read to a certain extent, can write their names, but cannot write beyond a short sentence level (2∼3 out of 6); (iii) literate: can read and write at sentence level (4∼5 out of 6); and (iv) high-level literate: can read and write perfectly at phrase level (a perfect score of 6). Demographic information for the 203 participants is shown in Table 1.
Procedure and scoring Mini-Mental State Examination MMSE is a popular measure for screening for mental impairment, particularly in the elderly.18 The test items were designed for assessment of orientation, attention and calculation, memory, and language ability. The score is the total number of correct answers, and the maximum score is 30. It has high test–retest reliability, and is very sensitive to the presence of dementia.
Participants We initially recruited 212 healthy elderly (HE) persons living in the community within Korea between May 2011 and December 2012. Because of the nature of the present study, which carried out not a simple survey but 912 |
Controlled Oral Word Association Test The Controlled Oral Word Association Test (COWAT) evaluates the spontaneous production of words under specific conditions (i.e. phonemic, semantic) in a given © 2013 Japan Geriatrics Society
Literacy on cognitive-language tests
Table 1 Demographic features of illiterate, semi-illiterate, literate, and high-level literate individuals
Age (years) Years of education
Pure-illiterate (n = 29)
Semi-illiterate (n = 67)
Literate (n = 75)
High-level literate (n = 32)
F
P
79.48 (6.68) 1.86 (3.08)
76.67 (5.91) 5.26 (3.60)
71.47 (5.59) 8.65 (3.30)
69.09 (6.06) 10.72 (3.04)
25.248 47.482
0.000*** 0.000***
*P < 0.05, **P < 0.01, ***P < 0.001. Mean (Standard Deviation),
period of time.31 In the present study, the participants were asked to respond with as many animal names as possible in 30 s.32 Admissible words in the animal category were counted as correct responses. The test is known to be useful for identification of impairment of executive function of the frontal lobe.33
Short-form of the Boston Naming Test The purpose of the Short-form of the Boston Naming Test (SF-BNT) is to assess visual naming ability.34 The short version includes 15 items of common objects of the BNT. In order to extract 15 items, we applied Item Response Theory (IRT), ran the WINSTEPS 3.70 program,35 and retrieved information on item goodness-of-fit and item level of difficulty. Items with 1.2 of infit/outfit and sex differences were excluded. Three points are given for each response, with a total score of 45 points.
Verb naming There are three categories of verbs: dynamic verbs, motion verbs and change-of-state verbs. Verbs were chosen based on the following criteria: (i) those in the Korean basic vocabulary dictionary; (ii) those without synonyms; and (iii) those can depict the given action in video. We presented a 5-s video clip on a laptop and gave the following instruction: “Watch this video clip and tell me what happens.” If there was no response, the clip was replayed. We used a multidimensional scoring system, from 0 to 3 points for each item.
of education, age and sex were controlled through and comparison and analysis of differences in performance on cognitive and language tests depending on the level of literacy was carried out. The score on the cognitive and language tests was converted to an age-, education- and sex-adjusted score for post-hoc analysis. Linear regression analyses were also carried out in order to examine the effects of literacy level, years of education, age, sex, and depression status on cognitive and language performance. ANCOVA,
Results The results for performance on cognitive and language tests are shown in Table 2. The pure illiterate group showed significantly lower scores than the other three groups on the MMSE, COWAT, SF-BNT, verb naming and sentence comprehension. In Bonferroni-adjusted post-hoc comparisons, significant differences in each task were observed among the four groups (Fig. 1). Among all tasks, the high-level literate group showed the highest performance. Effects of literacy level, years of education, age, sex, and depression status on cognitive and language performance were examined through linear regression analyses (Table 3). As a result, each regression equation included five variables; however, the variable that actually had an influence on all five tests was literacy level. Age was valid in determining performance on SF-BNT, verb naming and sentence comprehension tasks. GDS and years of education had an influence only on the MMSE score.
Sentence comprehension A total of 12 multiple choice sentence questions were given to each participant for selection of the appropriate drawing. The questions were varied with different kinds of grammatical/morphological complexity, such as passive and causative verbs, double negatives, tenses, and the like. A total score of 24 points was given (2 for immediate response; 1 for delayed response; 1 for incorrect response).
Statistical analyses Using three-way ANOVA, effects and interaction of demographic variables were identified. Then, level © 2013 Japan Geriatrics Society
Discussion Associations between literacy and cognitive ability have been well documented.36 During the aging process, cognitive functions (i.e. memory, execution and language of the elderly) decline. However, the decline is considered more rapid among older adults who are at a low reading level.37 In a clinical setting, most neuropsychological tests used for dementia patients are influenced by years of education. Because many test items are developed with the assumption of reading and writing capability of examinees, literacy levels have a significant influence as well. Thus, the accuracy of discriminative diagnosis not | 913
JW Kim et al.
Table 2 Cognitive and language test scores of illiterate, semi-illiterate, literate, and high-level literate individuals
MMSE COWAT SF-BNT Verb naming Sentence comprehension
Pure-illiterate
Semi-illiterate
Literate
F
P
M(± SD)
High-level literate M(± SD)
M(± SD)
M(± SD)
17.36 (3.68) 6.61 (2.58) 15.46 (8.04) 33.07 (5.91) 12.21 (4.31)
23.65 (3.67) 9.06 (2.58) 25.12 (9.67) 35.46 (6.13) 15.09 (4.35)
26.38 (2.31) 9.99 (2.45) 34.58 (8.12) 39.23 (3.83) 19.69 (3.16)
27.31 (1.96) 10.59 (3.26) 36.25 (5.81) 40.38 (3.40) 21.00 (2.38)
26.075 3.922 14.188 3.127 13.516
0.000*** 0.010*** 0.000*** 0.027* 0.000***
*P < 0.05, **P < 0.01, ***P < 0.001. COWAT, Controlled Oral Word Association Test (animal naming, 30 s administration time); MMSE, Mini-Mental Status Examination; SF-BNT, short-form of the Boston Naming Test.
Intergroup comparison ***P
Geriatr Gerontol Int 2014; 14: 911–917
ORIGINAL ARTICLE: EPIDEMIOLOGY, CLINICAL PRACTICE AND HEALTH
Effect of literacy level on cognitive and language tests in Korean illiterate older adults JungWan Kim,1 Ji Hye Yoon,2 Soo Ryon Kim3 and HyangHee Kim3,4 1
Department of Speech and Language Pathology, College of Rehabilitation Sciences, Daegu University, Gyeongsan, 2Department of Speech Pathology and Audiology, Audiology and Speech Pathology Research Institute, College of Natural Sciences, Hallym University, Chuncheon, 3 Graduate Program in Speech and Language Pathology, Yonsei University, and 4Department and Research Institute of Rehabilitation Medicine, Yonsei University College of Medicine, Seoul, Korea
Aim: Illiterate individuals represent a significant proportion of the world’s population. Acquisition of reading and writing skills influences the functional status of the brain, and consequently alters the performance on cognitive and language tests. Thus, it is important to identify the degree of the impact of levels of both illiteracy and education as potential confounders on test performance in people with neurological communication disorders. Methods: A total of 203 community-dwelling older adults aged 65 years and older were recruited for the present study. Participants were classified into four groups based on the literacy level; pure illiterate (n = 29), semi-illiterate (n = 67), literate (n = 75) and high-level literate (n = 32). The participants completed the Mini-Mental State Examination, Boston Naming Test, Controlled Oral Word Association Test (animal), verb naming, and sentence comprehension tests. Results: The pure illiterate group showed the lowest performance on all five tests. Regression analysis showed that literacy level was the variable that best predicted the performance on cognitive and language tests. Conclusions: These findings suggest that literacy in performance on cognitive and language tests is an important factor in neuropsychological evaluations for older adults. Geriatr Gerontol Int 2014; 14: 911–917. Keywords: cognition, elderly, language, literacy.
Introduction Illiterate individuals represent a significant proportion of the world’s population.1 In 2011, the illiteracy rate of adults aged 15 years and older was 16% of the world’s population.2 In South Korea, as a result of social (i.e. lack of opportunity to attend school, poverty, etc.) and personal (i.e. learning difficulties) reasons, the illiteracy rate is approximately 20% in members of the population over the age of 70 years. In particular, the illiteracy rate is higher in agricultural areas, and 26.4% of older adults aged 65 years and older are illiterate.3
Accepted for publication 15 October 2013. Correspondence: Professor HyangHee Kim PhD, Graduate Program in Speech and Language Pathology, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Korea. Email: [email protected]
© 2013 Japan Geriatrics Society
Although the acquisition of oral language is a natural phenomenon during language development, literacy skills are not acquired naturally. Thus, great effort is required in order to obtain and maintain literacy competencies. The ability to read and write, which is achieved in childhood for most people, could change the morphological and functional status of the human brain,4,5 influencing several linguistic and non-linguistic functions.6–8 Functional imaging studies have reported that literacy and level of education might have an influence on pathways used in the brain for problem solving.9 This implies that neural networks of cognition in illiterate individuals are different from those of literate individuals.10 In addition, illiteracy and lower levels of education constitute risk factors for dementia.11 Higher risk for development of Alzheimer’s disease has been shows in illiterate and less educated individuals, perhaps as a result of a vulnerability to cognitive deterioration.9,12 Illiteracy and lack of formal education attainment have dramatic effects on neuropsychological performance;13–15 therefore, illiteracy represents an important variable in research into dementia.16,17 doi: 10.1111/ggi.12195
| 911
JW Kim et al.
Studies have been carried out for determination of differences in performance on cognitive–language tests of the elderly depending on schooling and literate ability. Previous studies have reported that the level of education is significant in prediction of Mini-Mental State Examination (MMSE)18 scores; and, in particular, the cut-off point of illiterate subjects is close to 13 points, which is lower than that of educated subjects.19 In a study comparing language performance of illiterate adults, adults who had 1–2 years of elementary education15 and adults who had 3–4 years of elementary education, no difference in performance on naming, repetition and semantic verbal fluency tests was observed among groups. However, the illiterate group showed significantly lower performance on language comprehension (i.e. token test) and phonemic verbal fluency tests. When research findings on illiterate adults who showed low performance on naming, verbal fluency,20,21 memory test7,22 and visual construction tasks23,24 are considered together, differences in performance on language tests between literate and illiterate individuals are not limited only to tasks involving use of phonemic skills.13 Illiterate individuals might be at a disadvantage in testing situations, both because the testing environment is unfamiliar25,26 and because they lack appropriate testtaking strategies.27 As the MMSE was developed for literate older adults, illiterate older adults are often diagnosed as having dementia.28 According to one report, 79% of uneducated Korean adults are illiterate.3 This could be reinterpreted as 21% of uneducated adults have a certain degree of literate ability. In addition, because people in Korea tend to feel ashamed of having no official education or little official education, level of education is not proportional to literacy level, and it is more likely for the uneducated or those with fewer years of education. Diverse neuropsychological tests used for diagnosis of dementia, including the MMSE and the Boston Naming Test, do not present any norm with regard to literacy level. Thus, determination of the degree of impact of levels of both illiteracy and education as potential confounders in test performance in those with neurological communication disorders is important. The purpose of the present study was to compare the performance of four groups of older adults with different literacy levels on cognitive–language tests.
Methods
a cognitive–language test over a long period of time, random sampling was impossible. Thus, tests were carried out only for those among the elderly who wished to be evaluated. They had no history of neurological or psychiatric illness. The Korean Mini-Mental State Examination (K-MMSE)29 was administered to all participants. The mean score on the K-MMSE was 24.31 (± 4.34). Nine adults with K-MMSE score less than −1 standard deviation (SD) of the norm were excluded because of the probable influence of decreased cognitive function on the language test score. Thus, 203 HE (64 males, 139 females) ranging in age from 60 to 91 years participated as subjects in the study. The mean age was 74 ± 6.9 years. The mean number of years of education was 6.92 ± 4.35. All participants were right-handed, speaking standard Korean. This study was approved by the institutional review board of Yonsei University College of Medicine (No. 1–2011-0061). For measurement of the literacy level, a literacy test consisting of six items was carried out. The test includes two items from the reading subtest of the Screening Test for Aphasia and Neurologic-communication Disorders (STAND),30 two items requiring selection of answers after reading medicine instructions (phrase level), one item involving writing one’s name (word level) and one item involving writing at least three features of one’s appearance (sentence level). Based on the performance score on the literacy test, we divided the participants into four groups: (i) pure illiterate: cannot read and write at all, can occasionally write their own names (0∼1 out of a perfect score of 6); (ii) semi-illiterate: can read to a certain extent, can write their names, but cannot write beyond a short sentence level (2∼3 out of 6); (iii) literate: can read and write at sentence level (4∼5 out of 6); and (iv) high-level literate: can read and write perfectly at phrase level (a perfect score of 6). Demographic information for the 203 participants is shown in Table 1.
Procedure and scoring Mini-Mental State Examination MMSE is a popular measure for screening for mental impairment, particularly in the elderly.18 The test items were designed for assessment of orientation, attention and calculation, memory, and language ability. The score is the total number of correct answers, and the maximum score is 30. It has high test–retest reliability, and is very sensitive to the presence of dementia.
Participants We initially recruited 212 healthy elderly (HE) persons living in the community within Korea between May 2011 and December 2012. Because of the nature of the present study, which carried out not a simple survey but 912 |
Controlled Oral Word Association Test The Controlled Oral Word Association Test (COWAT) evaluates the spontaneous production of words under specific conditions (i.e. phonemic, semantic) in a given © 2013 Japan Geriatrics Society
Literacy on cognitive-language tests
Table 1 Demographic features of illiterate, semi-illiterate, literate, and high-level literate individuals
Age (years) Years of education
Pure-illiterate (n = 29)
Semi-illiterate (n = 67)
Literate (n = 75)
High-level literate (n = 32)
F
P
79.48 (6.68) 1.86 (3.08)
76.67 (5.91) 5.26 (3.60)
71.47 (5.59) 8.65 (3.30)
69.09 (6.06) 10.72 (3.04)
25.248 47.482
0.000*** 0.000***
*P < 0.05, **P < 0.01, ***P < 0.001. Mean (Standard Deviation),
period of time.31 In the present study, the participants were asked to respond with as many animal names as possible in 30 s.32 Admissible words in the animal category were counted as correct responses. The test is known to be useful for identification of impairment of executive function of the frontal lobe.33
Short-form of the Boston Naming Test The purpose of the Short-form of the Boston Naming Test (SF-BNT) is to assess visual naming ability.34 The short version includes 15 items of common objects of the BNT. In order to extract 15 items, we applied Item Response Theory (IRT), ran the WINSTEPS 3.70 program,35 and retrieved information on item goodness-of-fit and item level of difficulty. Items with 1.2 of infit/outfit and sex differences were excluded. Three points are given for each response, with a total score of 45 points.
Verb naming There are three categories of verbs: dynamic verbs, motion verbs and change-of-state verbs. Verbs were chosen based on the following criteria: (i) those in the Korean basic vocabulary dictionary; (ii) those without synonyms; and (iii) those can depict the given action in video. We presented a 5-s video clip on a laptop and gave the following instruction: “Watch this video clip and tell me what happens.” If there was no response, the clip was replayed. We used a multidimensional scoring system, from 0 to 3 points for each item.
of education, age and sex were controlled through and comparison and analysis of differences in performance on cognitive and language tests depending on the level of literacy was carried out. The score on the cognitive and language tests was converted to an age-, education- and sex-adjusted score for post-hoc analysis. Linear regression analyses were also carried out in order to examine the effects of literacy level, years of education, age, sex, and depression status on cognitive and language performance. ANCOVA,
Results The results for performance on cognitive and language tests are shown in Table 2. The pure illiterate group showed significantly lower scores than the other three groups on the MMSE, COWAT, SF-BNT, verb naming and sentence comprehension. In Bonferroni-adjusted post-hoc comparisons, significant differences in each task were observed among the four groups (Fig. 1). Among all tasks, the high-level literate group showed the highest performance. Effects of literacy level, years of education, age, sex, and depression status on cognitive and language performance were examined through linear regression analyses (Table 3). As a result, each regression equation included five variables; however, the variable that actually had an influence on all five tests was literacy level. Age was valid in determining performance on SF-BNT, verb naming and sentence comprehension tasks. GDS and years of education had an influence only on the MMSE score.
Sentence comprehension A total of 12 multiple choice sentence questions were given to each participant for selection of the appropriate drawing. The questions were varied with different kinds of grammatical/morphological complexity, such as passive and causative verbs, double negatives, tenses, and the like. A total score of 24 points was given (2 for immediate response; 1 for delayed response; 1 for incorrect response).
Statistical analyses Using three-way ANOVA, effects and interaction of demographic variables were identified. Then, level © 2013 Japan Geriatrics Society
Discussion Associations between literacy and cognitive ability have been well documented.36 During the aging process, cognitive functions (i.e. memory, execution and language of the elderly) decline. However, the decline is considered more rapid among older adults who are at a low reading level.37 In a clinical setting, most neuropsychological tests used for dementia patients are influenced by years of education. Because many test items are developed with the assumption of reading and writing capability of examinees, literacy levels have a significant influence as well. Thus, the accuracy of discriminative diagnosis not | 913
JW Kim et al.
Table 2 Cognitive and language test scores of illiterate, semi-illiterate, literate, and high-level literate individuals
MMSE COWAT SF-BNT Verb naming Sentence comprehension
Pure-illiterate
Semi-illiterate
Literate
F
P
M(± SD)
High-level literate M(± SD)
M(± SD)
M(± SD)
17.36 (3.68) 6.61 (2.58) 15.46 (8.04) 33.07 (5.91) 12.21 (4.31)
23.65 (3.67) 9.06 (2.58) 25.12 (9.67) 35.46 (6.13) 15.09 (4.35)
26.38 (2.31) 9.99 (2.45) 34.58 (8.12) 39.23 (3.83) 19.69 (3.16)
27.31 (1.96) 10.59 (3.26) 36.25 (5.81) 40.38 (3.40) 21.00 (2.38)
26.075 3.922 14.188 3.127 13.516
0.000*** 0.010*** 0.000*** 0.027* 0.000***
*P < 0.05, **P < 0.01, ***P < 0.001. COWAT, Controlled Oral Word Association Test (animal naming, 30 s administration time); MMSE, Mini-Mental Status Examination; SF-BNT, short-form of the Boston Naming Test.
Intergroup comparison ***P
