فراموشی رشد: یک الگوی جدیدی از تفکیک با حافظه اپیزودیک دست نخورده
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|33601||2004||18 صفحه PDF||سفارش دهید||12481 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Neuropsychologia, Volume 42, Issue 6, 2004, Pages 764–781
A case of developmental amnesia is reported for a child, CL, of normal intelligence, who has intact episodic memory but impaired semantic memory for both semantic knowledge of facts and semantic knowledge of words, including general world knowledge, knowledge of word meanings and superordinate knowledge of words. In contrast to the deficits in semantic memory, there are no impairments in episodic memory for verbal or visual material, assessed by recall or recognition. Lexical decision was also intact, indicating impairment in semantic knowledge of vocabulary rather than absence of lexical representations. The case forms a double dissociation to the cases of Vargha-Khadem et al. [Science 277 (1997) 376; Episodic memory: new directions in research (2002) 153]; Gadian et al. [Brain 123 (2000) 499] for whom semantic memory was intact but episodic memory was impaired. This double dissociation suggests that semantic memory and episodic memory have the capacity to develop separately and supports models of modularity within memory development and a functional architecture for the developmental disorders within which there is residual normality rather than pervasive abnormality. Knowledge of arithmetical facts is also spared for CL, consistent with adult studies arguing for numeracy knowledge distinct from other semantics. Reading was characterised by difficulty with irregular words and homophones but intact reading of nonwords. CL has surface dyslexia with poor lexico-semantic reading skills but good phonological reading skills. The case was identified following screening from a population of normal schoolchildren suggesting that developmental amnesias may be more pervasive than has been recognised previously.
1.1. Residual normality There is an on-going theoretical debate concerning the brain organisation of children with developmental disorders. Some argue that the organisation of cognitive systems in such cases may be fundamentally different from that of the normal child such that there is pervasive change to the system without residual normality (Thomas & Karmiloff-Smith, 2002). However, this view runs counter to many neuropsychological studies in which focal impairments have been described in children who appear to have intact residual skills (e.g. Anderson, Northam, Hendy, & Wrennall, 2001; Temple, 1997a and Temple, 1997b). Moreover, the patterns of deficit and skill in such neuropsychological studies appear to integrate with models of normal development, suggesting impairments of components of a modular system (Clahsen & Temple, 2003; Temple & Clahsen, 2002). Using such neuropsychological approaches, children who have specific reading difficulties despite normal intelligence are well documented and the form and character of these impairments has been systematically outlined in relation to the normal reading system (e.g. Castles & Coltheart, 1996 and Jackson & Coltheart, 2001). Similarly, children who have developmental dysgraphias and developmental dyscalculias have been documented in detail and interpreted in relation to focal deficits with modular spelling and arithmetical systems (Rovet et al., 1994 and Sokol et al., 1994). Specific impairments in domains outside literacy have also been discussed. Developmental prosopagnosias in those of otherwise normal intelligence have been outlined (Ariel & Sadah, 1996, Campbell, 1992, Elgar & Campbell, 2001 and Temple, 1992) as well as selective impairments in the perception of movement (Ahmed & Dutton, 1996) and selective impairments in the perception of location (McCloskey, Rapp, Yantis, & Rubin, 1995). In each of these cases, the disorders may take several different forms and have been explained in relation to selective impairment in the development of components of modular systems underlying face processing and visual perception (Temple, 1997a and Temple, 1997b). If modular organisation of the sort argued for in these studies is a fundamental underlying principle in brain development, then one would expect to see focal developmental disorders in all cognitive domains. The current paper concerns a case study of such a disorder of memory. 1.2. Acquired versus developmental amnesias Disorders of memory in children were first described in detail over a decade ago, but it is only recently that there has been increased focus on them (Temple, 2002). The studies reported to date differentiate between acquired amnesias in children who have sustained injury or disease and developmental amnesias in which the impairment arises in the absence of an acquired lesion and any deterioration in skills. In practice, there is an issue about where, in this bipartite division, cases should be classified who have sustained injury or disease in utero or at birth. Such cases are acquired in the sense that an explicit aetiology is known but are developmental in the sense that the abnormality is present from birth and its impact emerges as the child acquires skills without any subsequent loss of these skills, however, the literature has not addressed this dichotomy and traditionally classifies the disorders one way or the other. The case of amnesia in a child following a prenatal stroke has been labelled developmental (Maurer, 1992 and Temple, 2002), as have cases following birth anoxia (Vargha-Khadem, Gadian, & Connelly, 1997; Vargha-Khadem, Gadian, & Mishkin, 2001; Gadian et al., 2000 and Baddeley et al., 2001). 1.3. Impairments of both semantic and episodic memory Of those disorders classified as acquired amnesias of childhood, the first detailed case was that of Ostergaard (1987) who described a 10-year-old child, CC, who had sustained an anoxic episode with consequent damage to the left hippocampus. Semantic memory for facts was impaired, as was episodic memory, but procedural memory was intact. Acquired amnesia with impairment of both semantic and episodic memory was also described in a 9-year-old child, TC, following acute encephalopathy (Wood, Brown, & Felton, 1989). A further case with impairments of both episodic and semantic memory is reported by Broman et al. (1997). MS had severe asthma from an early age and suffered a respiratory arrest when eight years old. An MRI scan indicated loss of volume bilaterally in the medial temporal area, with some evidence of a shortened hippocampus. The authors suggested his profile closely resembled that of the adult case, HM (Corkin, 1984 and Scoville & Milner, 1957). In cases of developmental amnesia, in which there has been no known injury post-natally or at birth, impairments of both semantic memory and episodic memory have also been described. MS, a 22-year-old man reported by De Renzi and Lucchelli (1990), was born two months early with a birth weight of 250 g. EEG indicated bilateral bursts of theta and delta waves with a left frontotemporal propensity but there was otherwise no know neurological abnormality. MS complained of memory problems since childhood with difficulty in remembering the names and faces of familiar people, places, foreign language words, song lyrics, poems, mathematical formula and tables. Intelligence was normal, with a Verbal IQ of 110 and a Performance IQ of 111. Semantic memory for past events and famous names was impaired as was recognition memory for famous faces. Episodic memory was also impaired with difficulty in learning new verbal and nonverbal material, including story recall. Developmental amnesia with impairments of both episodic and semantic memory was also reported by Maurer (1992). NS had a CT scan when she was 9 years old that revealed low density areas in both temporal fossa indicating absence of the left temporal lobe and the pole and medial parts of the right temporal lobe. The aetiology, as mentioned above, was thought to be a prenatal stroke. Although both semantic memory and episodic memory were impaired, procedural memory was intact. Using simple stimulus-response training, NS was repeatedly trained to learn the names of people she met everyday. Results were effective as she was able to learn some names (Maurer, 1992). Temple, 1997a and Temple, 1997b reported a further case of developmental amnesia with both semantic memory impairment and episodic memory impairment. Julia was a 12-year-old girl for whom gestation appeared normal, and milestones were reached without significance. Memory difficulties were evident during pre-school years. At the age of six, Julia was diagnosed with temporal lobe epilepsy. There was impairment of semantic memory for factual knowledge and difficulties in episodic memory and the acquisition of new verbal and non-verbal material. Procedural knowledge, as assessed by recall of automated sequences like the alphabet and days of the week was normal. 1.4. Dissociations in episodic and semantic memory impairment The possibility of a dissociation between impairments of semantic and episodic memory in children emerged in the study of a 15-year-old girl, VT, who had amnesia following a mild head injury (Maravita, Spadoni & Parma, 1995). Although initially VT had impaired episodic and semantic memory, when reassessed 23 months later, episodic memory remained impaired but semantic memory had improved substantially. It appeared that she had partially recovered previously learnt material, as her learning of semantic knowledge was easier for material that she had been studying prior to her injury. Even so, her performance on semantic memory tasks never reached the previous level of proficiency and there remained a degree of impairment in both episodic and semantic domains (Maravita et al., 1995). A more marked dissociation between episodic and semantic memory impairment was described by Vargha-Khadem et al., 1997 and Vargha-Khadem et al., 2002. They reported cases of children who suffered anoxic episodes at birth and had hippocampal damage on MRI. The authors argued that these were the first reported cases of developmental amnesia in children whose semantic skills were within normal limits, with some intact declarative abilities despite episodic memory impairment. Gadian et al. (2000) assessed a further three children, in addition to two from the Vargha-Khadem et al. (1997) study. All exhibited the profile of impaired episodic memory but relatively preserved semantic memory. MRI revealed bilateral hippocampus atrophy and reduced grey matter in the putamen, with abnormality in the thalamus and midbrain. The authors proposed that the perirhinal, entorhinal and parahippocampal cortices are able to support context-free semantic memories but not context-rich episodic memories, for which the hippocampus is required (Vargha-Khadem et al., 1997 and Vargha-Khadem et al., 2002). For both semantic and episodic memory to be severely affected the hippocampus and adjoining cortices would both be impaired. In the cases described above, there are examples of impairment of both episodic and semantic memory and cases of intact semantic memory and impaired episodic memory but there are no descriptions of intact episodic memory despite impaired semantic memory. Such a childhood case, regardless of acquired or developmental aetiology, would form a double dissociation to the cases of Vargha-Khadem et al., 1997 and Vargha-Khadem et al., 2002 and Gadian et al. (2000) and thereby strengthen arguments for modularity between the episodic and semantic memory systems and their relative independence within development. It is such a case of impaired semantic memory but intact episodic memory, in a child, which is the focus of the current paper. A further dissociation within episodic memory itself has been described by Casalini, Brizzolara, Cavallaro, and Cipriani (1999). ON was a 9-year-old girl with developmental amnesia who had no neurological symptoms, but her father had similar difficulties. ON had poor semantic memory for facts. Her verbal episodic memory was also poor with difficulty in learning new verbal material but nonverbal memory and learning was relatively good indicating a modality specific episodic impairment. In a further case study, described in a separate paper, a case study of a child similar to ON is discussed (Temple & Richardson, forthcoming). 1.5. Impact of amnesia upon language in development A further issue which arises in development is the impact of any amnesic disorder upon language development. Amnesia in adults usually spares language skills, however, the ramifications of acquired or developmental amnesia in children may be qualitatively different. In adults, language skills have been acquired prior to the injury or disease which induces amnesia and these established skills usually remain intact. In children, language skills are still maturing. If the child is acquiring language with an impaired memory substrate, a semantic memory impairment could impinge on the development of the semantic representations that underpin language, resulting in impaired acquisition of vocabulary. Thus, in the child there are issues about the independence and interdependence of semantic systems which can be studied in a way that is not possible in adulthood. Impairments in semantic knowledge of vocabulary in children have been described in cases where other semantic memory skills are also impaired. Thus, for example, in addition to impaired semantic memory for facts, CC had impaired semantic memory for vocabulary (Ostergaard, 1987). MS who had impaired semantic knowledge of facts also had impaired semantic knowledge of words (Broman et al., 1997). Similarly, in addition to impaired semantic knowledge of facts, Julia had impaired semantic knowledge of words (Temple, 1997a and Temple, 1997b). She had difficulties with both word recognition and word finding from the outset and prior to the onset of her epilepsy, Julia’s class teacher commented that she “sometimes has difficult remembering words or what a word actually means”. Language was characterized by an anomia such that at the age of 12;8 her naming age was 5;3. ON, described by Casalini et al. (1999) also had impaired memory for words and had difficulties in naming which affected her spontaneous speech. Whilst these may be coincidental co-existing impairments, it is also possible that there is a common underlying semantic impairment in each area of disorder. The semantic impairment which is the basis for impaired knowledge of facts may also underlie impaired knowledge of vocabulary. The only reported dissociation in development between semantic memory impairment for facts and semantic memory impairment for words is the case MS (De Renzi & Lucchelli, 1990). As discussed above, MS had impaired semantic memory for facts but semantic knowledge of vocabulary, for this 22-year-old man, was described as normal. There is comment in the study that there were difficulties with the names of places, people and certain foreign language words which means that the issue of a limitation in vocabulary knowledge cannot be dismissed completely but the case raises the possibility that semantic impairments for facts and vocabulary may be dissociable even in development. However, in the case reported here the semantic memory impairment for facts co-exists with impairment in semantic knowledge of vocabulary. 1.6. Impact of amnesia upon literacy in development Contemporary models of reading identify a lexico-semantic reading route which utilises a semantic system of representations of words in order to subsequently activate their pronunciations (e.g. Coltheart, Rastle, Perry, Langdon, & Ziegler, 2001). Semantic impairment in reading would affect the development of this lexico-semantic reading route and would lead to the prediction of a particular pattern of developmental reading difficulty, namely surface dyslexia (Coltheart et al., 1983 and Hanley et al., 1992; Temple, 1984, Temple, 1985, Temple, 1997a and Temple, 1997b; Samuelsson, 2000). In developmental surface dyslexia, the impaired development of lexico-semantic reading systems leads to over reliance on phonologically (“sounding out”)-based reading systems. This leads to increased difficult with irregular and exception words (e.g. pint, yacht,) which violate rules of orthographic to phonological translation. There may also be an increased error rate with regular words because of the ambiguities of some of these rules in English but the difficulty with irregular words is the most marked. A further characteristic of reading in surface dyslexia is homophone confusion, where the meanings assigned to homophones (e.g. sale–sail) are confused. This is attributed to the over-reliance on the sound pattern of the words rather than their semantic representations, with meaning being derived after the phonological representation for the word. Connectionist models also predict surface dyslexia in association with semantic impairment ( Plaut, McClelland, Seidenberg, & Patterson, 1996; Patterson et al., 1996). Surface dyslexia has been described in studies of adults with semantic dementia who have impairments of semantic memory ( Patterson & Hodges, 1992 and Patterson et al., 1996). There has been no systematic published analysis of the pattern of reading skills in cases of semantic memory impairment in children, however, there are a number of reports of reading difficulty. CC had impaired semantic memory for facts and words but was also reported as having reading and spelling difficulties (Ostergaard, 1987). Similarly, MS had impaired semantic memory for facts and words and also had poor reading and spelling (Broman et al., 1997). MS had impaired semantic memory for facts and was also described as having reading and writing skills which were poorly developed for his age and were slow and laborious (De Renzi & Lucchelli, 1990). Our own case Julia (Temple, 1997a and Temple, 1997b) in an unpublished investigation was found to be surface dyslexic. An early case of surface dyslexia was also reported in a child, NG, who had temporal lobe epilepsy and semantic memory impairment (Temple, 1984). These cases raise the issue of whether such an interrelationship might appear in other cases of semantic memory impairment in children. In the case reported here, impaired semantic memory extends to impairment in the lexico-semantic reading route and a pattern of surface dyslexia is demonstrated in reading. 1.7. Impact of amnesia upon arithmetical skills in development Studies of memory impairments in adults suggest that general semantic knowledge about the world may dissociate from semantic knowledge about arithmetic. Cappelletti, Butterworth, and Kopelman (2001) and Butterworth, Capalletti, and Kopelman (2002) have described a case of semantic dementia in a patient IH, a former banker, who had severely impaired semantic knowledge including the meanings of words and yet retention of calculation abilities, dependent upon knowledge of numerical facts. There was sparing of the ability to process the single semantic category of numbers. Knowledge of word meaning was severely impaired with inability to generate any items in category fluency tasks, inability to provide any accurate definitions of words, and severe anomia in spontaneous speech. In contrast, single and multi-digit addition and subtraction were excellent. Knowledge of multiplication tables though impaired in comparison to controls was nevertheless 73% accurate, demonstrating a markedly higher level of ability than seen in vocabulary tasks. In literacy, there was severe impairment in reading and writing non number words but retained ability to write number words (Cappalletti, Kopelman, & Butterworth, 2002). IH could write three but not tree, and seventeen but not event. These literacy skills form a double dissociation with the developmental case, Paul ( Temple, 1989), who could read and write non-number words well but had digit dyslexia. Paul could read orchestra, colonel and physics but made errors reading three, nine and five. The study of IH ( Butterworth et al., 2002; Cappelletti et al., 2001, 2002) raises the further issue of whether semantic memory impairment in children would always extend to encompass arithmetical semantic knowledge or whether general semantic knowledge about the world and arithmetical semantic knowledge may dissociate in development. Temple (1994) described twins with significantly impaired semantic knowledge of numerical facts who had excellent general semantic knowledge about the world and good vocabulary knowledge. The reverse dissociation has not previously been described but is demonstrated in the case reported here. 1.8. Intact intelligence The cases of acquired amnesia in children indicate limitations in the impact of the memory impairment upon other domains. Broman et al. (1997) argued that despite his amnesia, MS had syntactic and logical reasoning skills which had matured to adult proficiency. TC (Wood et al., 1989 and Ostergaard & Squire, 1990) also made scholastic progress, albeit limited. Some of the current studies suggest that acquired amnesia may be compatible with normal intelligence (Vargha-Khadem et al., 2002). The developmental case MS (De Renzi & Lucchelli, 1990) also had good normal intelligence on formal psychometric assessment. If developmental amnesia can co-exist with normal intelligence then there may be developmental amnesics within the normal school population. It is perhaps surprising that a larger number of cases of developmental amnesia have not been reported. One possibility is that the developmental amnesias are rare and that the reported cases reflect the incidence. Another possibility is that there are children with developmental amnesia whose cases have not been identified. In pursuit of this issue, the current study involved screening normal children attending normal schools to determine whether within this population, children of normal intelligence with specific impairments of memory could be detected. The screening did indeed detect such cases and the study reports in detail one of the cases. The new case of developmental amnesia reported here presents a child who has intact episodic memory in the face of impaired semantic memory. It forms a double dissociation to the cases of Vargha-Khadem et al., 1997, Vargha-Khadem et al., 2002 and Gadian et al., 2000). Although this contrast is of cases where there is explicit pathology and the case here which is clearly developmental, the Gadian et al. (2000) study concerned impairment acquired at birth. Thus, memory in these cases had not developed normally to a point after which it became impaired. There were acquired lesions but there had been no period of normal cognitive development. Both the Gadian et al. (2000) study and the case reported here describe children for whom there is no evidence of a period of normal memory development. In both, the Gadian et al. (2000) study and the study here memory development is impaired but the impairment is selective in relation to the components of memory that are affected. As discussed above, semantic memory impairment in development may affect the acquisition of vocabulary and also literacy. The case of developmental amnesia reported here therefore explores not only different elements of memory but also knowledge of words and vocabulary and the nature of literacy development. It demonstrates impaired semantic knowledge of vocabulary and an impaired lexico-semantic reading system with surface dyslexia. The domain of numerical semantic knowledge appears preserved. The paper presents the results of a case study of CL, a child who was screened from a population of normal schoolchildren as having weak memory skills. Performance is compared to normal controls from the same sample of school children matched to CL for IQ, but with normal semantic and episodic memories on the screening measures employed.