Acta Neurol Scand 2014: 130: 312–318 DOI: 10.1111/ane.12228
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd ACTA NEUROLOGICA SCANDINAVICA
Facial memory deficits in myotonic dystrophy type 1 Kleberg JL, Lindberg C, Winblad S. Facial memory deficits in Myotonic dystrophy type 1. Acta Neurol Scand: 2014: 130: 312–318. © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd. Objectives – To evaluate facial memory ability (FMA) in patients with myotonic dystrophy type 1 (DM1). We also explored the relationship between FMA and neuropsychological data, disease-related factors, and CTG repeat expansion size. Materials and methods – Patients with DM1 (n = 33) and healthy subjects (n = 30) were tested with the faces task of the Rivermead Behavioural Memory Test – Extended version (RBMT-E) and an additional set of neuropsychological tests. Clinical data were collected, and CTG repeat size was quantified in blood lymphocytes. Results – Low results on the faces task were more common in patients with DM1 compared with healthy subjects (P < 0.05), with 36% of the patients showing a poor/impaired performance. DM1 patients with deficits in FMA performed significantly worse on tests measuring visual-construction ability and memory. Furthermore, these patients more often falsely recognised unknown faces as known. Deficits in FMA were not associated with any disease-related factor, including CTG repeat expansion size. Conclusions – These findings revealed deficits in FMA in the DM1 group, which was associated with reduced construction- and visual memory ability.
Introduction
Myotonic dystrophy type 1 (DM1) is a slowly progressive neuromuscular disorder considered to be the most prevalent muscular dystrophy in adults (1, 2). The disease is multisystemic, frequently involving muscular, ocular, cardiac, gonadal, endocrine and central nervous (CNS) abnormalities (1, 3). DM1 is transmitted in an autosomal dominant manner and is caused by an expanded and unstable trinucleotide CTG repeat localised to the 3′untranslated region of the dystrophia myotonica protein kinase (DMPK) gene on chromosome 19 (4). Unaffected individuals have ≤49 CTG repeats, whereas expansions associated with DM1 ranges from 50 to >2500 repeats. Repeat length correlates with disease severity and inversely with age of onset. Consequently, longer expansions are associated with more severe symptoms and earlier disease onset (1). In some cases, the first signs of the disease 312
J. L. Kleberg1,2, C. Lindberg3, S. Winblad1,3 1 Department of Psychology, University of Gothenburg, Gothenburg, Sweden; 2Department of Psychology, Uppsala University, Uppsala, Sweden; 3Neuromuscular Centre, Sahlgrenska University Hospital, Gothenburg, Sweden
Key words: cognition; CTG repeats; facial memory; myotonic dystrophy type 1 S. Winblad, Department of Psychology, University of Gothenburg, Box 500, 405 30 Gothenburg, Sweden Tel.: + 46 31 786 16 92 Fax: + 46 31 786 46 28 e-mail: [email protected] Accepted for publication January 14, 2014
are detected during early age (congenital and juvenile variants) and in others during adolescence, adulthood or old age (classical and mild variants). A number of studies have demonstrated brain anomalies as well as cognitive abnormalities associated with DM1 (for reviews, see 5, 6). Extensive and widespread white-matter pathology have been reported (7) with insular, frontal and temporal areas most prominently affected (8, 9), while grey matter decrease has been observed in cerebral lobes, thalamus and putamen (8). Cognitive deficits include mental retardation in congenital and juvenile DM1 (10) and reduced visual-constructive ability and dysexecutive behaviour in childhood and classical variants (11–13). Furthermore, a number of studies have demonstrated impaired social cognition and low sociability (14– 16). Disrupted social participation has been observed, including a limited social network and a diminution of social support associated with
Myotonic dystrophy and facial memory increased CTG repeats (17, 18). This disruption has been explained by extremity weakness and fatigue (17). It is known that difficulties in processing, and especially the ability to remember faces, are associated with social anxiety and withdrawal from social interaction (19, 20). Facial memory is defined as the ability to recognise previously seen faces, and severe inability is termed prosopagnosia or face blindness (21). This condition may be the consequence of brain damage localised to the temporal and occipital lobes, bilaterally or in the right hemisphere (22). A highly heritable condition called developmental prosopagnosia is associated with similar behavioural features (21). When healthy individuals perceive and memorise faces, holistic strategies are typically used and facial recognition accuracy is strongly influenced by holistic processing (23). This means that healthy individuals perceive the face as a complex configuration or a ‘Gestalt’. By contrast, less effective analytical (detail-based) strategies are employed by subjects with acquired prosopagnosia and by subjects with developmental brain abnormalities, including autism (24, 25). Experimental manipulations that disrupt holistic processing are detrimental for subsequent face memory in healthy subjects, but not in patients with prosopagnosia. This suggests that deficient holistic processing is a cause of FMA impairments and prosopagnosia (24). Furthermore, the tendency to falsely recognise unknown people as known in social settings is associated with both heritable and acquired face memory impairments (21, 26). Different brain areas are associated with different aspects of the processing of faces (27). Accordingly, the ability to recognise facial emotional expressions and to remember faces is generally subserved by distinct cognitive systems (28). In a previous study (14), we demonstrated a facial emotion recognition deficit in patients with DM1, which correlated with CTG-repeat expansions size, specific neuropsychological functions and personality dimensions associated with low sociability. So as to explore additional factors possibly affecting sociability, we measured DM1 patient’s ability to remember faces, using a standardised test. Our main hypothesis was that subjects with DM1 would show impairments in FMA. We further hypothesised FMA to be negatively related to CTG repeat length of the DMPK gene and disease duration. To further explore the relationship between FMA impairments and other cognitive functions, we included a neuropsychological test battery.
Materials and methods Subjects
Thirty-three patients (17 male, 16 female) with DM1 who were consecutively attending the Neuromuscular Centre at the Sahlgrenska University Hospital agreed to participate. Inclusion criteria were a molecular confirmation of the disease and an age between 15 and 65 years. In all cases, symptoms had first appeared after 10 years of age. All participants had normal or corrected-tonormal vision. Patients with a history of major psychiatric or somatic illness, acquired brain injury or alcohol misuse were excluded. The control group consisted of 30 individuals (14 male, 16 female) recruited from three schools and workplaces in Gothenburg. None of the participants in the control group had any disease or were on any medication or drug that could be suspected of affecting the test results. All participants gave informed consent and the local ethical committee approved of the study. There were no significant differences in age between DM1 patients and health controls (mean = 39.5 vs. 37.1 years) or gender proportion. However, the control group had longer education as measured in years (mean = 13.7) than the patient group (mean = 11.1) (P < 0.01). The mean CTG repeat expansion size was 595 (min = 100, max = 2000). Mean time elapsed as symptoms first appeared according to patients self-report was 16.2 years (SD = 11). Facial recognition task
The faces subtest from the Extended Rivermead Behavioural Memory Test (RBMT-E) (29) was used to measure FMA. The RBMT-E has good psychometric properties (29, 30). In the faces subtest, the participant is asked to focus on 15 greyscale photographs depicting different human faces for 3 s per face. The participant is asked to estimate whether the persons on the pictures are older or younger than 40 years of age. After a ten-min delay, the same pictures are presented along with 15 distracters (other faces), and the participant is asked to tell whether the picture was part of the original set or not. The faces subtest gives a correct recognition score (0–15), a false recognition score (0–15) and a total score (the number of falsely recognised faces subtracted from the number of correctly recognised faces, 0–15). Raw scores can be converted to profile scores based on the results of 188 control subject’s performance (29). A raw score ≤11 indicates 313
Lundin Kleberg et al. poor/impaired facial memory ability. RBMT-E was initially developed for detecting everyday memory problems rather than with the aim to supply cognitively ‘pure’ measures. As a result, the pictures used in the faces subtest include features such as hair, beard and glasses. Neuropsychological tests
The neuropsychological examination included four tests with good psychometric properties (30). Tests were chosen to assess broad aspects of neuropsychological functions within a limited time frame. The Block design test (31) is as a test of visuospatial organisation ability. The subject′s task is to use blocks to construct replicas of a model design presented by the examiner. Each block has two white and two red sides, and two half-red half-white sides with the colours divided along the diagonal. Nine items is presented in order of increasing difficulty with seven trials giving time bonus points (total score = 51). The Vocabulary test (31) requires the subject to define 35 words listed in order of difficulty. Every right answer gives two scores per word (total score = 70). The test has been recognised as an excellent measure of general mental ability including a strong association with verbal ability (30). The Rey Auditory Verbal Learning Test (RAVLT) (32) is a test of verbal learning and recall and consists of a list of 15 concrete nouns. The words are read aloud by the examiner, and the patient is asked to recall as many of them as possible. RAVLT includes five trials and words that have been recalled on previous trials are to be repeated. The variable used in the analysis is the sum of recalled words on the five learning trials (total score = 75). The Rey Complex Figure Test (RCFT) (33) examines perceptual organisation and visual memory. The test includes a trial requiring the subject to copy a complex figure on a sheet of paper, and a free recall trial performed 3 min after completion. Each picture is analysed using a scoring system (total score on each trial = 36). Procedure
Patients were examined by an experienced neuropsychologist (SW) at the Sahlgrenska university hospital. Healthy subjects were examined by the first author (under the supervision of SW) at schools and workplaces. The test procedure was performed in the same order in both groups. In all cases, participants completed testing in a quiet environment with adequate lightning. RBMT-E 314
Faces stimuli were presented on a stationary computer to the patients and on a lap top computer to the control group. The computer screens were of comparable size and resolution. The assessment took approximately 1.5 h to complete with a short break to avoid the effects of fatigue. During the examination, information on onset and duration of the disease were collected from the patient. This information was later compared with available medical records. Genetic analysis
DNA was extracted from peripheral blood lymphocytes and analysed for the expansion of the CTG repeat in the DMPK gene. The analyses were performed with PCR and southern blot using the probe PM10M6 (4). The size of the CTG expansions was assessed visually from exposed x-ray films, as described earlier (10). Statistical analysis
Data were analysed using PASW base 18 (SPSS Inc, Chicago, IL, USA). In cases where an impression of significant deviations from normal distribution was supported by a Shapiro–Wilk test (P < 0.05), nonparametric Mann–Whitney U-tests was used for group comparisons. The chisquare test was used to compare gender distribution. To analyse the possible contribution of other neuropsychological functions on facial memory ability, we used the Kruskal–Wallis test in between group comparisons. Mann–Whitney tests with a Bonferroni correction were used for post hoc comparisons. Results
Low total scores on RBMT-E Faces were more prevalent in the patient group, and the difference in distribution of test scores was significant (U = 347, P < 0.05). The clinical significance of this finding can be estimated by converting RBMT-E Faces scores to profile scores (29). As shown in Fig. 1, 12 patients (36%) achieved results indicating poor (10, 11) or impaired (≤9) face memory as compared to four subjects (13%) in the control group. The results were negatively skewed and deviated significantly from normal distribution. Distribution above the median appears similar, reflecting a ceiling effect in both groups. As shown in Fig. 2, patients with impaired or poor results on the RBMT-E Faces (DM1-1) (n = 12) had lower correct recognition scores
Myotonic dystrophy and facial memory Table 1 Neuropsychological test results for the patients and control groups DM1-1 (n = 12) Block design Vocabulary RAVLT RCFT copy RCFT memory
Figure 1. Distribution of RBMT-E Faces profile scores between patients with DM1 (n = 33) and the healthy control group (n = 30).
16 14
TRUE FALSE
Median
12 10 8 6 4 2 0
DM1-1
DM1-2
Controls
Groups
Figure 2. True and false recognition scores on the RBMT-E Faces test. Median scores are shown separately for the patient groups with poor/impaired results (DM1-1), average/ good/exceptionally good (DM1-2) and the healthy control group.
(median = 11.5) as compared to both DM1 patients with average/good/exceptionally good results (DM1-2) (n = 21, median = 15) and healthy controls (n = 30, median = 14). The differences were significant (U = 22, P < 0.001 and U = 31, P < 0.001, respectively). Furthermore, false recognition scores were higher in the DM1-1 group (median = 3) as compared to the DM1-2 group (median = 1) and the healthy control group (median = 1). The differences were statistically significant (U = 41, P = 0.001 and, U = 76, P = 0.003, respectively). No statistical differences were observed between the DM1-2 patients and the healthy control persons on these measures. To examine the role of examined neuropsychological functions for face memory, test performance of DM1-1 was compared with DM1-2 and healthy controls. As shown in Table 1, patients with facial memory deficits (DM1-1) scored significantly lower (adjusted Mann–Whitney test, P < 0.016) than healthy controls on tests
12.7 39.2 47.9 21.9 10.5
(10.2) (10.3) (11.6) (11.4) (10.6)
[11]** [40] [48.5] [23.5]** [7]*,**
DM1-2 (n = 21) 16.9 (9.4) [19]† 42 (12.9) [42] 52.1 (8.5) [53]† 31.1 (5.9) [32] 18.7 (8.5) [15]
Controls (n = 30) 32 41.8 44.8 32.5 20.7
(8,6); [31.5] (10.8) [42] (8,2) [44.5] (3) [33.5] (6.1) [20.8]
P
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd ACTA NEUROLOGICA SCANDINAVICA
Facial memory deficits in myotonic dystrophy type 1 Kleberg JL, Lindberg C, Winblad S. Facial memory deficits in Myotonic dystrophy type 1. Acta Neurol Scand: 2014: 130: 312–318. © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd. Objectives – To evaluate facial memory ability (FMA) in patients with myotonic dystrophy type 1 (DM1). We also explored the relationship between FMA and neuropsychological data, disease-related factors, and CTG repeat expansion size. Materials and methods – Patients with DM1 (n = 33) and healthy subjects (n = 30) were tested with the faces task of the Rivermead Behavioural Memory Test – Extended version (RBMT-E) and an additional set of neuropsychological tests. Clinical data were collected, and CTG repeat size was quantified in blood lymphocytes. Results – Low results on the faces task were more common in patients with DM1 compared with healthy subjects (P < 0.05), with 36% of the patients showing a poor/impaired performance. DM1 patients with deficits in FMA performed significantly worse on tests measuring visual-construction ability and memory. Furthermore, these patients more often falsely recognised unknown faces as known. Deficits in FMA were not associated with any disease-related factor, including CTG repeat expansion size. Conclusions – These findings revealed deficits in FMA in the DM1 group, which was associated with reduced construction- and visual memory ability.
Introduction
Myotonic dystrophy type 1 (DM1) is a slowly progressive neuromuscular disorder considered to be the most prevalent muscular dystrophy in adults (1, 2). The disease is multisystemic, frequently involving muscular, ocular, cardiac, gonadal, endocrine and central nervous (CNS) abnormalities (1, 3). DM1 is transmitted in an autosomal dominant manner and is caused by an expanded and unstable trinucleotide CTG repeat localised to the 3′untranslated region of the dystrophia myotonica protein kinase (DMPK) gene on chromosome 19 (4). Unaffected individuals have ≤49 CTG repeats, whereas expansions associated with DM1 ranges from 50 to >2500 repeats. Repeat length correlates with disease severity and inversely with age of onset. Consequently, longer expansions are associated with more severe symptoms and earlier disease onset (1). In some cases, the first signs of the disease 312
J. L. Kleberg1,2, C. Lindberg3, S. Winblad1,3 1 Department of Psychology, University of Gothenburg, Gothenburg, Sweden; 2Department of Psychology, Uppsala University, Uppsala, Sweden; 3Neuromuscular Centre, Sahlgrenska University Hospital, Gothenburg, Sweden
Key words: cognition; CTG repeats; facial memory; myotonic dystrophy type 1 S. Winblad, Department of Psychology, University of Gothenburg, Box 500, 405 30 Gothenburg, Sweden Tel.: + 46 31 786 16 92 Fax: + 46 31 786 46 28 e-mail: [email protected] Accepted for publication January 14, 2014
are detected during early age (congenital and juvenile variants) and in others during adolescence, adulthood or old age (classical and mild variants). A number of studies have demonstrated brain anomalies as well as cognitive abnormalities associated with DM1 (for reviews, see 5, 6). Extensive and widespread white-matter pathology have been reported (7) with insular, frontal and temporal areas most prominently affected (8, 9), while grey matter decrease has been observed in cerebral lobes, thalamus and putamen (8). Cognitive deficits include mental retardation in congenital and juvenile DM1 (10) and reduced visual-constructive ability and dysexecutive behaviour in childhood and classical variants (11–13). Furthermore, a number of studies have demonstrated impaired social cognition and low sociability (14– 16). Disrupted social participation has been observed, including a limited social network and a diminution of social support associated with
Myotonic dystrophy and facial memory increased CTG repeats (17, 18). This disruption has been explained by extremity weakness and fatigue (17). It is known that difficulties in processing, and especially the ability to remember faces, are associated with social anxiety and withdrawal from social interaction (19, 20). Facial memory is defined as the ability to recognise previously seen faces, and severe inability is termed prosopagnosia or face blindness (21). This condition may be the consequence of brain damage localised to the temporal and occipital lobes, bilaterally or in the right hemisphere (22). A highly heritable condition called developmental prosopagnosia is associated with similar behavioural features (21). When healthy individuals perceive and memorise faces, holistic strategies are typically used and facial recognition accuracy is strongly influenced by holistic processing (23). This means that healthy individuals perceive the face as a complex configuration or a ‘Gestalt’. By contrast, less effective analytical (detail-based) strategies are employed by subjects with acquired prosopagnosia and by subjects with developmental brain abnormalities, including autism (24, 25). Experimental manipulations that disrupt holistic processing are detrimental for subsequent face memory in healthy subjects, but not in patients with prosopagnosia. This suggests that deficient holistic processing is a cause of FMA impairments and prosopagnosia (24). Furthermore, the tendency to falsely recognise unknown people as known in social settings is associated with both heritable and acquired face memory impairments (21, 26). Different brain areas are associated with different aspects of the processing of faces (27). Accordingly, the ability to recognise facial emotional expressions and to remember faces is generally subserved by distinct cognitive systems (28). In a previous study (14), we demonstrated a facial emotion recognition deficit in patients with DM1, which correlated with CTG-repeat expansions size, specific neuropsychological functions and personality dimensions associated with low sociability. So as to explore additional factors possibly affecting sociability, we measured DM1 patient’s ability to remember faces, using a standardised test. Our main hypothesis was that subjects with DM1 would show impairments in FMA. We further hypothesised FMA to be negatively related to CTG repeat length of the DMPK gene and disease duration. To further explore the relationship between FMA impairments and other cognitive functions, we included a neuropsychological test battery.
Materials and methods Subjects
Thirty-three patients (17 male, 16 female) with DM1 who were consecutively attending the Neuromuscular Centre at the Sahlgrenska University Hospital agreed to participate. Inclusion criteria were a molecular confirmation of the disease and an age between 15 and 65 years. In all cases, symptoms had first appeared after 10 years of age. All participants had normal or corrected-tonormal vision. Patients with a history of major psychiatric or somatic illness, acquired brain injury or alcohol misuse were excluded. The control group consisted of 30 individuals (14 male, 16 female) recruited from three schools and workplaces in Gothenburg. None of the participants in the control group had any disease or were on any medication or drug that could be suspected of affecting the test results. All participants gave informed consent and the local ethical committee approved of the study. There were no significant differences in age between DM1 patients and health controls (mean = 39.5 vs. 37.1 years) or gender proportion. However, the control group had longer education as measured in years (mean = 13.7) than the patient group (mean = 11.1) (P < 0.01). The mean CTG repeat expansion size was 595 (min = 100, max = 2000). Mean time elapsed as symptoms first appeared according to patients self-report was 16.2 years (SD = 11). Facial recognition task
The faces subtest from the Extended Rivermead Behavioural Memory Test (RBMT-E) (29) was used to measure FMA. The RBMT-E has good psychometric properties (29, 30). In the faces subtest, the participant is asked to focus on 15 greyscale photographs depicting different human faces for 3 s per face. The participant is asked to estimate whether the persons on the pictures are older or younger than 40 years of age. After a ten-min delay, the same pictures are presented along with 15 distracters (other faces), and the participant is asked to tell whether the picture was part of the original set or not. The faces subtest gives a correct recognition score (0–15), a false recognition score (0–15) and a total score (the number of falsely recognised faces subtracted from the number of correctly recognised faces, 0–15). Raw scores can be converted to profile scores based on the results of 188 control subject’s performance (29). A raw score ≤11 indicates 313
Lundin Kleberg et al. poor/impaired facial memory ability. RBMT-E was initially developed for detecting everyday memory problems rather than with the aim to supply cognitively ‘pure’ measures. As a result, the pictures used in the faces subtest include features such as hair, beard and glasses. Neuropsychological tests
The neuropsychological examination included four tests with good psychometric properties (30). Tests were chosen to assess broad aspects of neuropsychological functions within a limited time frame. The Block design test (31) is as a test of visuospatial organisation ability. The subject′s task is to use blocks to construct replicas of a model design presented by the examiner. Each block has two white and two red sides, and two half-red half-white sides with the colours divided along the diagonal. Nine items is presented in order of increasing difficulty with seven trials giving time bonus points (total score = 51). The Vocabulary test (31) requires the subject to define 35 words listed in order of difficulty. Every right answer gives two scores per word (total score = 70). The test has been recognised as an excellent measure of general mental ability including a strong association with verbal ability (30). The Rey Auditory Verbal Learning Test (RAVLT) (32) is a test of verbal learning and recall and consists of a list of 15 concrete nouns. The words are read aloud by the examiner, and the patient is asked to recall as many of them as possible. RAVLT includes five trials and words that have been recalled on previous trials are to be repeated. The variable used in the analysis is the sum of recalled words on the five learning trials (total score = 75). The Rey Complex Figure Test (RCFT) (33) examines perceptual organisation and visual memory. The test includes a trial requiring the subject to copy a complex figure on a sheet of paper, and a free recall trial performed 3 min after completion. Each picture is analysed using a scoring system (total score on each trial = 36). Procedure
Patients were examined by an experienced neuropsychologist (SW) at the Sahlgrenska university hospital. Healthy subjects were examined by the first author (under the supervision of SW) at schools and workplaces. The test procedure was performed in the same order in both groups. In all cases, participants completed testing in a quiet environment with adequate lightning. RBMT-E 314
Faces stimuli were presented on a stationary computer to the patients and on a lap top computer to the control group. The computer screens were of comparable size and resolution. The assessment took approximately 1.5 h to complete with a short break to avoid the effects of fatigue. During the examination, information on onset and duration of the disease were collected from the patient. This information was later compared with available medical records. Genetic analysis
DNA was extracted from peripheral blood lymphocytes and analysed for the expansion of the CTG repeat in the DMPK gene. The analyses were performed with PCR and southern blot using the probe PM10M6 (4). The size of the CTG expansions was assessed visually from exposed x-ray films, as described earlier (10). Statistical analysis
Data were analysed using PASW base 18 (SPSS Inc, Chicago, IL, USA). In cases where an impression of significant deviations from normal distribution was supported by a Shapiro–Wilk test (P < 0.05), nonparametric Mann–Whitney U-tests was used for group comparisons. The chisquare test was used to compare gender distribution. To analyse the possible contribution of other neuropsychological functions on facial memory ability, we used the Kruskal–Wallis test in between group comparisons. Mann–Whitney tests with a Bonferroni correction were used for post hoc comparisons. Results
Low total scores on RBMT-E Faces were more prevalent in the patient group, and the difference in distribution of test scores was significant (U = 347, P < 0.05). The clinical significance of this finding can be estimated by converting RBMT-E Faces scores to profile scores (29). As shown in Fig. 1, 12 patients (36%) achieved results indicating poor (10, 11) or impaired (≤9) face memory as compared to four subjects (13%) in the control group. The results were negatively skewed and deviated significantly from normal distribution. Distribution above the median appears similar, reflecting a ceiling effect in both groups. As shown in Fig. 2, patients with impaired or poor results on the RBMT-E Faces (DM1-1) (n = 12) had lower correct recognition scores
Myotonic dystrophy and facial memory Table 1 Neuropsychological test results for the patients and control groups DM1-1 (n = 12) Block design Vocabulary RAVLT RCFT copy RCFT memory
Figure 1. Distribution of RBMT-E Faces profile scores between patients with DM1 (n = 33) and the healthy control group (n = 30).
16 14
TRUE FALSE
Median
12 10 8 6 4 2 0
DM1-1
DM1-2
Controls
Groups
Figure 2. True and false recognition scores on the RBMT-E Faces test. Median scores are shown separately for the patient groups with poor/impaired results (DM1-1), average/ good/exceptionally good (DM1-2) and the healthy control group.
(median = 11.5) as compared to both DM1 patients with average/good/exceptionally good results (DM1-2) (n = 21, median = 15) and healthy controls (n = 30, median = 14). The differences were significant (U = 22, P < 0.001 and U = 31, P < 0.001, respectively). Furthermore, false recognition scores were higher in the DM1-1 group (median = 3) as compared to the DM1-2 group (median = 1) and the healthy control group (median = 1). The differences were statistically significant (U = 41, P = 0.001 and, U = 76, P = 0.003, respectively). No statistical differences were observed between the DM1-2 patients and the healthy control persons on these measures. To examine the role of examined neuropsychological functions for face memory, test performance of DM1-1 was compared with DM1-2 and healthy controls. As shown in Table 1, patients with facial memory deficits (DM1-1) scored significantly lower (adjusted Mann–Whitney test, P < 0.016) than healthy controls on tests
12.7 39.2 47.9 21.9 10.5
(10.2) (10.3) (11.6) (11.4) (10.6)
[11]** [40] [48.5] [23.5]** [7]*,**
DM1-2 (n = 21) 16.9 (9.4) [19]† 42 (12.9) [42] 52.1 (8.5) [53]† 31.1 (5.9) [32] 18.7 (8.5) [15]
Controls (n = 30) 32 41.8 44.8 32.5 20.7
(8,6); [31.5] (10.8) [42] (8,2) [44.5] (3) [33.5] (6.1) [20.8]
P
