PERIPHERAL ASPECTS OF WRITING CAN BE DIFFERENTIALLY AFFECTED BY SENSORIAL AND ATTENTIONAL DEFECT: EVIDENCE FROM A PATIENT WITH AFFERENT DYSGRAPHIA AND CASE DISSOCIATION Maria Caterina Silveti (Neuropsychology Service, Institute of Neurology, Catholic University, Rome, Italy) ABSTRACT This study reports a patient with right posterior cerebral atrophy who was affected by afferent dysgraphia, characterised by case dissociation: cursive better than upper-case print. The patient also had severe visuocostructional deficit and simultanagnosia. The hypothesis advanced to explain this dissociation is that cases are not equally dependent on visual and kinaesthetic control. Thus, if visual and kinaesthetic feedback (or the attention to them) are impaired by cerebral lesions, upper case can be expected to be more impaired than cursive. INTRODUCTION All writing disorders that originate from deficits of the graphemic level and downstream are included in the so-called peripheral dysgraphia (Ellis, 1988; Margolin and Goodman-Schulman, 1992), as opposed to central dysgraphias due to lesions that affect lexical and sublexical processes up to the production of an abstract graphemic representation of the word (or non-word). According to current cognitive models on writing processing, each component letter of a word stored at the graphemic level requires the activation of the appropriate allographic unit within an allographic level of representation. In fact, graphemes can assume different physical forms for writing in upper or lower case, in script or cursive. The graphic motor pattern (Margolin, 1984; Ellis, 1988; Margolin and Goodman-Schulman, 1992) will specify "size, position and order of strokes" (Ellis, 1988) requested to realise the allograph. The general validity of this model, which was articulated to account for data on normal letter level slips of the pen (Ellis, 1988), receives support from pathology. Patients have been described with selective deficit at the grapheme level (Miceli, Silveri and Caramazza, 1985, 1987; Caramazza, Miceli, Villa et aI., 1987; Hillis and Caramazza, 1989), at the allographic level (Rosati and De Bastiani, 1979; Rothi and Heilman, 1981; Kapur and Lawton, 1983; Goodman and Caramazza, 1986; De- Bastiani and Barry, 1989; Patterson and Wing, 1989) and at the graphic motor pattern level, both in activating the adequate motor patterns (Hatfield and Patterson, 1983; Baxter and Warrington, 1986; Papagno, 1992) or in executing them (Hecaen, 1967; Hecaen and Marcie, 1974; Lebrun, 1976; Margolin and Binder, 1984; Lebrun, 1985; Ellis, Young and Plude, 1987). Cortex, (1996) 32, 155-172 156 Maria Caterina Silveri This study describes a patient, RI, with difficulties in executing graphic motor patterns. This deficit, originally called spatial (Hecaen, Penfield, Bertrand et aI., 1956) or afferent (Lebrun, 1976) dysgraphia, has been repeatedly described in the past, in association with right hemisphere lesion (Hecaen, 1967; AssaI and Zander, 1969; Hecaen and Marcie, 1974). At variance with dysgraphias associated with lesions in the left hemisphere, afferent dysgraphia has been interpreted as a spatial disorder that affects the "act of writing" (Lebrun, 1976) rather than as a linguistic deficit, since the right hemisphere subserves the processing of visuospatial information and proprioceptive information involved by the spatial components of writing. The main characteristics of afferent dysgraphia consist (Ellis et aI., 1987) in the tendency to omit and reduplicate letters and strokes, in the difficulty to write horizontal lines and in the tendency to write on the right hand side of the page. The patient hereby described, who suffered from a focal progressive degenerative pathology involving the posterior areas of the right hemisphere (De Renzi, 1986), presented a typical afferent dysgraphia, and, in addition, a dissociation between preserved ability to write in cursive case and difficulty to write in upper-case print. Upper case and lower case can be differentially affected by brain lesions whit a predominant impairment of the former (Patterson and Wing, 1989) or of the latter (DeBastiani and Barry, 1986). Patterson and Wing (1989) hypothesised that the principal candidate locus for this dissociation was the allographic level (or physical letter code) and excluded downstream levels such as the graphic motor pattern. In cognitive models of writing, afferent dysgraphia has been interpreted (Margolin, 1984; Ellis et aI., 1987) as due to the inability to correctly execute graphic motor patterns, a performance which requires visual and kinaesthetic feedback in order to monitor handwriting movements. If RI's deficit is at the level of graphic motor patterns, this would demonstrate that the processes implicated in upper case and in lower case writing are separated at a lower level than the allographic level. Recently, Kartsounis (1992) reported a patient with better preservation of upper case in respect to lower case writing and attributed the dissociation to defective access to graphic motor patterns for lower case, hence suggesting the independence of the motor programs for executing upper and lower cases. In the present patient the opposite dissociation is viewed as dependent on the greater role that the impairment of visual and/or kinaesthetic functions plays on upper case print in comparison to cursive. Thus RI's deficit would not refer to a deficit within the writing process but to the effect on writing of a more general visuospatial disorder. CASE REPORT Clinical History RI was a 57-year old woman, teacher in the primary school, admitted at the Department of Neurology on December 1989. At the end of 1988 she begun to complain of a progressive Peripheral dysgraphia with case dissociation 157 Fig. I - MRI peljiJrllled Oil December lYYJ. difficulty in object identification that obliged her to retire a few months later. She consulted an eye specialist that excluded peripheral damage and suspected a central neurological deficit. At admission, RI was well oriented in time and in place. The neurological examination was normal, except for a visual agnosic disorder, and occasional left extinction to tactile and visual double stimulation. CT scan showed a mild atrophy in the right parietal and occipital areas. The patient was discharged with the diagnosis of focal degenerative brain pathology. She was controlled as out-patient on May 1990 and September 1990. On December 1990, she was re-admitted to the hospital and submitted to a single photon computed tomography with 99Tc-HMPAO, which showed decreased perfusion in the posterior areas of the right hemisphere. She was further examined as out-patient on October 1991 and February 1992. Visual spatial ability progressively got worse, while other cognitive functions remained generally good. Her relatives confirmed that, with the exception due to the visual problems, RI successfully attended everyday activities. In the last control, RI seemed to be a little slower, and to have some difficulties in moving in space. However, she was still oriented in time and in place and had preserved linguistic and memory functions. The patient was examined again on December 1993. She was markedly impaired in all visuoconstructional and visuospatial tasks. Writing and reading were virtually impossible. Severe limb apraxia, but not oral apraxia, was detected. Visual fields were normal at clinical examination, but the patient systematically extinguished the left stimulus in tactile and visual double stimulation and clearly neglected the left half of space when she performed visual tasks. Visual analysis of pictures was typical of severe simultanagnosia. Verbal memory was moderately impaired. Naming on oral description as well as spontaneous speech and episodic memory remained good. MRI confirmed the right posterior atrophy (Figure 1). Neuropsychological Examination The more extensive evaluation was done on December 1990, when the patient was admitted to the hospital for the second time. 158 Maria Caterina Silveri Language No sign of aphasia was present. Spontaneous speech was flawless and adequate to the patient's level of education. She was extremely accurate in describing her deficit, showing to be aware of her impairment. Confrontation naming was poor because of visual errors. She was able to recognise only 28 out of 70 pictures of objects and animals, and 10 out of 20 visually presented objects, whereas she immediately recognised all of them when they were presented in the tactile modality. The fact that responses were referred to part of the object suggested simultanagnosia. She produced a normal number of items in a word fluency task. Language comprehension could not be formally evaluated with visual stimuli, but her ability to comprehend and to execute complex verbal orders was normal. She was always very accurate when asked to describe the structural and functional features of verbally given stimuli. She could write correctly though slowly in cursive, only demonstrating some difficulties in maintaining horizontal lines and the regularity of the margins. Upper case writing was poor. No tendency to neglect the left-hand side of the page was observed. Reading was dominated by visual errors. Only occasional neglect errors were produced. There was no part of speech effect but only a slight effect of frequency and length. No tendency to letter-by-Ietter reading was detected. Reading sentences was virtually impossible. Memory Verbal memory (immediate and delayed free recall of 15 words) was in the normal range: immediate recall: 37175; delayed recall: 9115 (cut-off: immediate recall: 35175; delayed recall: 7/15). Digit span was 6 forwards and 4 backwards. Verbal span (words) was between 4 and 5. Spatial span was 4 forwards. In spite of her difficulty in object recognition, she demonstrated a normal ability to perform a Visual Memory task in which 10 pictures of coloured objects and animals were first presented and then had to be recognised among 40 foils. Visuospatial and Praxic Tasks RI correctly performed, even though slowly, line and letter cancellation tasks without neglect for the left side. Left-right orientation was good. Copying threedimensional drawings was severely impaired. The performance was better with bidimensional drawings. Neither oral nor limb apraxia was detected. Over time, however, limb movements became progressively slower and clumsier. Visuoperceptual Tasks The patient was able to name 20 out of 20 colours, and to correctly point to 20 out of 20 colours in multiple choice tasks with three foils. She was able to identify geometrical figures (15/15) only when they were of small size (up to 2-3 cm). She could not read global letters composed of smaller letters, but she was able to read the small components letters. The patient never complained of difficulty in face recognition, nor in recognising relatives or friends or the Peripheral dysgraphia with case dissociation 159 doctors of the medical staff. She was able to identify 9 out of 10 famous faces and her score was borderline (39) at the Benton Facial Recognition Test. Writing and spelling On May 1990, at the first control after the discharge, she complained of writing disorders, mainly in upper-case print. She claimed that her pre-morbid writing in both upper and cursive case was perfect (she was a teacher!). She refused to write in lower script case, since she never used to. In the first session non-words were also given; writing difficulties were precisely of the same order than with words. In addition, she demonstrated the same proficiency in spelling aloud words and non-words. For these reasons, only words were given in the experimental testing. The same words were generally given for upper case, cursive writing and oral spelling. Almost all lists of words given were controlled for length, word frequency and grammatical class. Sometimes, additional words were given for cursive writing, in order to increase the number of errors available for the analysis. The ability to write single letters in the two cases was also explored. Writing with letter blocks was prevented by the patient's dyslexia and, mostly, by severe difficulty in visual exploration. Qualitative Aspects of Writing Samples of RI's writing in cursive and upper case collected over sessions are shown in Figure 2. In the first session (May 1990) (Figure 2A) cursive handwriting was good both for single letters and words, even if words tended to slop down on the right side of the line. No irregularity of the left-hand margin was observed. Letters had normal shape and size and their spatial relation within the word was respected. Writing speed was mildly reduced. When the patient was requested to write with eyes closed her performance deteriorate only marginally. Upper-case print was, on the contrary, laborious, and markedly slower than cursive. The component strokes of letters were properly selected but inserted in anomalous position. Spatial relations among the component letters of a word were not respected. The occasional errors produced in cursive handwriting were generally letter omissions, particularly, omissions of geminate letters. Errors produced in upper case included both stroke errors within letters and letter errors within words, particularly, omissions, mislocations and repetition of stroke and omissions and repetitions of letters. When the patient was requested to write in upper case with eyes closed, an increasing number of errors and a clear alteration of the spatial relationships among the strokes within a letter and of the component letters within a word was observed. On October 1990 (Figure 2B), the cursive did not show any significant worsening, while upper case writing was markedly deteriorated, with an increased number of errors. Although able to select the component strokes in proper sequence, RI found difficulty in identifying their correct direction and position. This was particularly evident with eyes closed; words were not legible for a person who did not follow the patient's hand while writing. On December 1990, Maria Caterina Silveri 160 DECEMBER 1990 (C) MAY 1990 (A) CF4 d; ~ "-V) quotidiano [DAILY] l1e~l~cLo ~'.Jt [MILLIONS] """'-"'-