Characterising the Long-Term Language Impairments of Children Following Cerebellar Tumour Surgery by Extracting Psycholinguistic Properties from Spontaneous Language.

Following cerebellar tumour surgery, children may suffer impairments of spontaneous language. Yet, the language processing deficits underlying these impairments are poorly understood. This study is the first to try to identify these deficits for four levels of language processing in cerebellar tumour survivors. The spontaneous language of twelve patients who underwent cerebellar tumour surgery (age range 3-24 years) was compared against his or her controls using individual case statistics. A distinction was made between patients who experienced postoperative cerebellar mutism syndrome (pCMS) and those who did not. Time since surgery ranged between 11 months and 12;3 years. In order to identify the impaired language processing levels at each processing level (i.e., lexical, semantic, phonological and/or morphosyntactic) nouns and verbs produced in the spontaneous language samples were rated for psycholinguistic variables (e.g., concreteness). Standard spontaneous language measures (e.g., type-token ratio) were calculated as well. First, inter-individual heterogeneity was observed in the spontaneous language outcomes in both groups. Nine out of twelve patients showed language processing deficits three of whom were diagnosed with pCMS. Results implied impairments across all levels of language processing. In the pCMS-group, the impairments observed were predominantly morphosyntactic and semantic, but the variability in nature of the spontaneous language impairments was larger in the non-pCMS-group. Patients treated with cerebellar tumour surgery may show long-term spontaneous language impairments irrespective of a previous pCMS diagnosis. Individualised and comprehensive postoperative language assessments seem necessary, given the inter-individual heterogeneity in the language outcomes.

The development of number line estimation in children at risk of mathematics learning difficulties: A longitudinal study.

Children with mathematics learning difficulties (MLD) show poorer performance on the number line task, but how performance on this task relates to other mathematical skills is unclear. This study examined the association between performance on the number line task and mathematical skills during the first 2 years of school for children at risk of MLD. Children (N = 100; Mage = 83.63 months) were assessed on four occasions on the number line task and other mathematical skills (math fluency, numerical operations, and mathematical reasoning). Estimation patterns were analyzed based on the representational shift and proportional judgment accounts separately. More consistent longitudinal trends and stronger evidence for differences in mathematical skills based on estimation patterns were found within the representational shift account. Latent growth curve models showed accuracy on the number line task as a predictor of growth in some mathematical skills assessed. We discuss impacts of methodological limitations on the study of estimation patterns.

Spoken verb learning in children with language disorder.

The current study examined spoken verb learning in elementary school children with language disorder (LD). We aimed to replicate verb learning deficits reported in younger children with LD and to examine whether verb instrumentality, a semantic factor reflecting whether an action requires an instrument (e.g., "to chop" is an instrumental verb), influenced verb learning. The possible facilitating effect of orthographic cues presented during training was also evaluated. In an exploratory analysis, we investigated whether language and reading skills mediated verb learning performance. General language skills and verb learning were assessed in Dutch children with LD and age-matched typically developing controls (n = 25 per group) aged 8 to 12 years (M = 9;9 [years;months], SD = 1;3). Using video animations, children learned 20 nonwords depicting actions comprising 10 instrumental and 10 noninstrumental verbs. Half of the items were trained with orthographic information present. Verb learning was assessed using an animation-word matching and animation naming task. Linear mixed-effects models showed a main effect of group for all verb learning measures, demonstrating that children with LD learned fewer words and at a slower rate than the control group. No effect of verb instrumentality, presence of orthographic information, or the included mediators was found. Our results emphasize the importance of continued vocabulary instruction in elementary school to strengthen verb encoding. Given that our findings are inconsistent with the overall literature showing an orthographic facilitation effect, future studies should investigate whether participants pay attention to the written word form in learning contexts with moving stimuli.

Serial position effects in graphemic buffer impairment: An insight into components of orthographic working memory.

This study investigated the nature of graphemic buffer functioning and impairment, through analysis of the spelling impairment shown by GEC, a man with acquired dysgraphia and clear characteristics of graphemic buffer impairment. We discuss GEC's error patterns in relation to different processes of orthographic working memory. This is the first study to show the contribution of these processes in one individual through performance on different spelling tasks. GEC's spelling errors in writing to dictation showed a linear serial position effect, including deletions of final letters. These "fragment errors" can be explained as the result of information rapidly decaying from the buffer (reduced temporal stability). However, in tasks that reduced working memory demands, GEC showed a different error distribution that may indicate impairment to a different buffer process (reduced representational distinctiveness). We argue that different error patterns can be a reflection of subcomponents of orthographic working memory that can be impaired separately.

Videoconferencing Interventions for Children with Reading and Spelling Difficulties: A Pilot Study.

Background: Many children in rural and remote areas do not have access to professionals providing literacy interventions. However, delivery of services through videoconferencing would increase access and choice for end users. Introduction: This pilot study investigated the efficacy of videoconferencing literacy interventions. As videoconferencing platforms become easier and cheaper to use, this form of telehealth delivery is increasing in popularity. However, there is currently no strong evidence base to support this practice. Materials and Methods: We studied 18 children, aged 7-12 years, with poor reading and/or spelling, and whose literacy interventions were videoconferenced into their homes and/or schools. Children were tested on three reading measures: (1) reading words, (2) reading nonwords, and (3) letter-sound knowledge, twice before their interventions commenced and once after their intervention concluded. Results: Children's raw and standardized scores on 2 of 3 outcome measures increased significantly more during the intervention than in the no-intervention period before their training commenced. Discussion and Conclusions: This study demonstrates that videoconferencing is a promising delivery mode for literacy interventions, and the results justify running a larger, randomized controlled trial.

Too harts, won sole: Using dysgraphia treatment to address homophone representation.

Previous spoken homophone treatment in aphasia found generalization to untreated homophones and interpreted this as evidence for shared phonological word form representations. Previous written treatment of non-homophones has attributed generalization to orthographic neighbours of treated items to feedback from graphemes to similarly spelled orthographic word forms. This feedback mechanism offers an alternative explanation for generalization found in treatment of spoken homophones. The aim of this study was to investigate the mechanism underpinning generalization (if any) from treatment of written homophones. To investigate this question a participant with acquired dysgraphia and impaired access to orthographic output representations undertook written spelling treatment. Generalization to untreated items with varying degrees of orthographic overlap was investigated. Three experimental sets included homographs (e.g., bank-bank), heterographs (e.g., sail-sale), and direct orthographic neighbours (e.g., bath-path). Treatment improved written picture naming of treated items. Generalization was limited to direct neighbours. Further investigation of generalization found that items with a greater number of close neighbours in the treated set showed greater generalization. This suggests that feedback from graphemes to orthographic word forms is the driving force of generalization. The lack of homograph generalization suggests homographs do not share a representation in the orthographic lexicon.

Phonics training for English-speaking poor readers.

BACKGROUND: The reading skills of 16% of children fall below the mean range for their age, and 5% of children have significant and severe reading problems. Phonics training is one of the most common reading treatments used with poor readers, particularly children. OBJECTIVES: To measure the effect of phonics training and explore the impact of various factors, such as training duration and training group size, that might moderate the effect of phonics training on literacy-related skills in English-speaking poor readers. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, 12 other databases, and three trials registers up to May 2018. We also searched reference lists of included studies and contacted experts in the field to identify additional studies. SELECTION CRITERIA: We included studies that used randomisation, quasi-randomisation, or minimisation to allocate participants to a phonics intervention group (phonics training only or phonics training plus one other literacy-related skill) or a control group (no training or non-literacy training). Participants were English-speaking poor readers with word reading one standard deviation below the appropriate level for their age (children, adolescents, and adults) or one grade or year below the appropriate level (children only), for no known reason. Participants had no known comorbid developmental disorder, or physical, neurological, or emotional problem. DATA COLLECTION AND ANALYSIS: We used standard methodological procedures expected by Cochrane. MAIN RESULTS: We included 14 studies with 923 participants in this review. Studies took place in Australia, Canada, the UK, and the USA. Six of the 14 included studies were funded by government agencies and one was funded by a university grant. The rest were funded by charitable foundations or trusts. Each study compared phonics training alone, or in conjunction with one other reading-related skill, to either no training (i.e. treatment as usual) or alterative training (e.g. maths). Participants were English-speaking children or adolescents, of low and middle socioeconomic status, whose reading was one year, one grade, or one standard deviation below the level expected for their age or grade for no known reason. Phonics training varied between studies in intensity (up to four hours per week), duration (up to seven months), training group size (individual and small groups), and delivery (human and computer). We measured the effect of phonics training on seven primary outcomes (mixed/regular word reading accuracy, non-word reading accuracy, irregular word reading accuracy, mixed/regular word reading fluency, non-word reading fluency, reading comprehension, and spelling). We judged all studies to be at low risk of bias for most risk criteria, and used the GRADE approach to assess the quality of the evidence.There was low-quality evidence that phonics training may have improved poor readers' accuracy for reading real and novel words that follow the letter-sound rules (standardised mean difference (SMD) 0.51, 95% confidence interval (CI) 0.13 to 0.90; 11 studies, 701 participants), and their accuracy for reading words that did not follow these rules (SMD 0.67, 95% CI 0.26 to 1.07; 10 studies, 682 participants). There was moderate-quality evidence that phonics training probably improved English-speaking poor readers' fluency for reading words that followed the letter-sounds rules (SMD 0.45, 95% CI 0.19 to 0.72; 4 studies, 224 participants), and non-word reading fluency (SMD 0.39, 95% CI 0.10 to 0.68; 3 studies, 188 participants), as well as their accuracy for reading words that did not follow these rules (SMD 0.84, 95% CI 0.30 to 1.39; 4 studies, 294 participants). In addition, there was low-quality evidence that phonics training may have improved poor readers' spelling (SMD 0.47, 95% CI -0.07 to 1.01; 3 studies, 158 participants), but only slightly improve their reading comprehension (SMD 0.28, 95% CI -0.07 to 0.62; 5 studies, 343 participants). AUTHORS' CONCLUSIONS: Phonics training appears to be effective for improving literacy-related skills, particularly reading fluency of words and non-words, and accuracy of reading irregular words. More studies are needed to improve the precision of outcomes, including word and non-word reading accuracy, reading comprehension, spelling, letter-sound knowledge, and phonological output. More data are also needed to determine if phonics training in English-speaking poor readers is moderated by factors such as training type, intensity, duration, group size, or administrator.

Developmental graphemic buffer dysgraphia in English: A single case study.

A single case study is reported of a 10-year-old, English-speaking boy, L.S., who presented with spelling errors similar to those described in acquired graphemic buffer dysgraphia (GBD). We used this case to evaluate the appropriateness of applying adult cognitive models to the investigation of developmental cognitive disorders. The dual-route model of spelling guided this investigation. L.S. primarily made "letter errors" (deletions, additions, substitutions, transpositions, or a combination of these errors) on words and nonwords and in all input (aural and visual) and output modalities (writing, typing, oral spelling); there was also some evidence of a length effect and U-shaped serial position curve. An effect of lexical variables on spelling performance was also found. We conclude that the most parsimonious account is an impairment at the level of the graphemic buffer and without systematic cognitive neuropsychological investigation, the nature of L.S.'s spelling difficulty would likely have been missed.

Development of children's identity and position processing for letter, digit, and symbol strings: A cross-sectional study of the primary school years.

Letter recognition and digit recognition are critical skills for literate adults, yet few studies have considered the development of these skills in children. We conducted a nine-alternative forced-choice (9AFC) partial report task with strings of letters and digits, with typographical symbols (e.g., $, @) as a control, to investigate the development of identity and position processing in children. This task allows for the delineation of identity processing (as overall accuracy) and position coding (as the proportion of position errors). Our participants were students in Grade 1 to Grade 6, allowing us to track the development of these abilities across the primary school years. Our data suggest that although digit processing and letter processing end up with many similarities in adult readers, the developmental trajectories for identity and position processing for the two character types differ. Symbol processing showed little developmental change in terms of identity or position accuracy. We discuss the implications of our results for theories of identity and position coding: modified receptive field, multiple-route model, and lexical tuning. Despite moderate success for some theories, considerable theoretical work is required to explain the developmental trajectories of letter processing and digit processing, which might not be as closely tied in child readers as they are in adult readers.

Challenges in the use of treatment to investigate cognition.

The use of data from people with cognitive impairments to inform theories of cognition is an established methodology, particularly in the field of cognitive neuropsychology. However, it is less well known that studies that aim to improve cognitive functioning using treatment can also inform our understanding of cognition. This paper discusses a range of challenges that researchers face when testing theories of cognition and particularly when using treatment as a tool for doing so. It highlights the strengths of treatment methodology for testing causal relations and additionally discusses how generalization of treatment effects can shed light on the nature of cognitive representations and processes. These points are illustrated using examples from the Special Issue of Cognitive Neuropsychology entitled Treatment as a tool for investigating cognition.

Can verbal working memory training improve reading?

The aim of the current study was to determine whether poor verbal working memory is associated with poor word reading accuracy because the former causes the latter, or the latter causes the former. To this end, we tested whether (a) verbal working memory training improves poor verbal working memory or poor word reading accuracy, and whether (b) reading training improves poor reading accuracy or verbal working memory in a case series of four children with poor word reading accuracy and verbal working memory. Each child completed 8 weeks of verbal working memory training and 8 weeks of reading training. Verbal working memory training improved verbal working memory in two of the four children, but did not improve their reading accuracy. Similarly, reading training improved word reading accuracy in all children, but did not improve their verbal working memory. These results suggest that the causal links between verbal working memory and reading accuracy may not be as direct as has been assumed.

Generalisation after treatment of acquired spelling impairments: A review.

This paper provides a comprehensive review of treatment studies of acquired dysgraphia and the occurrence of generalisation after this treatment. The aim is to examine what determines the occurrence of generalisation by investigating the link between the level of impairment, the method of treatment, and the outcome of therapy. We present the outcomes of treatment with regard to generalisation in 40 treatment studies. We derive general principles of generalisation which provide us with a better understanding of the mechanism of generalisation: (1) Direct treatment effects on representations or processes; (2) interactive processing and summation of activation; and (3) strategies and compensatory skills. We discuss the implications of these findings for our understanding of the cognitive processes used for spelling. Finally, we provide suggestions for the direction of further research into this important area, as a better understanding of the mechanism of generalisation could maximise treatment effects for an individual with acquired dysgraphia.

Are comorbidities of poor reading related to elevated anxiety in children?

Studies of the association between dyslexia and mental health have typically tried to minimise the influence of dyslexia comorbidities on the outcomes. However, in the "real world", many children with dyslexia have these comorbidities. In this study, we tested (1) if children with dyslexia with three common comorbidities - inattention, hyperactivity, language difficulties - experience more anxiety than children with dyslexia without these comorbidities; and (2) if any type of comorbidity is related to a certain type of anxiety (reading, social, generalised, or separation). The data of 82 children with dyslexia (mean age = 9 years and 4 months; 25 girls) were analysed using Fisher exact tests, which revealed that those with inattention (40.54%) or hyperactivity (42.30%) were statistically significantly more likely to experience elevated anxiety than children with dyslexia without these comorbidities (8.11 and 14.28%, respectively). This was not the case for language difficulties (24.5% versus 30%). Spearman ρ correlations (α = .05) indicated significant moderate relationships between inattention and reading anxiety (.27), social anxiety (.37), and generalised anxiety (.24); and between hyperactivity and social anxiety (.24) and generalised anxiety (.28). There were no significant correlations between language and anxiety. Examination of highly inter-correlated variables suggested a specific relationship between one type of comorbidity (inattention) and one type of anxiety (reading anxiety).

Language outcomes in children who underwent surgery for the removal of a posterior fossa tumor: A systematic review.

BACKGROUND: Children who underwent posterior fossa tumor removal may have spoken or written language impairments. The present systematic review synthesized the literature regarding the language outcomes in this population. Benefits of this work were the identification of shortcomings in the literature and a starting point toward formulating guidelines for postoperative language assessment. METHODS: A systematic literature search was conducted, identifying studies with patients who had posterior fossa surgery before 18 years of age. Included studies were narratively synthesized to understand language outcomes by language function (e.g., phonology, morphosyntax) at a group and individual level. Furthermore, the influence of several mediators (e.g., postoperative cerebellar mutism syndrome (pCMS), tumor type) was investigated. A critical evaluation of the language assessment tools was conducted. RESULTS: The narrative synthesis of 66 studies showed that a broad spectrum of language impairments has been described, characterized by a large interindividual heterogeneity. Patients younger at diagnosis, receiving treatment for a high-grade tumor and/or radiotherapy and diagnosed with pCMS seemed more prone to impairment. Several gaps in language assessment remain, such as a baseline preoperative assessment and the assessment of pragmatics and morphosyntax. Further, there were important methodological differences in existing studies which complicated our ability to accurately guide clinical practice. CONCLUSION: Children who had posterior fossa surgery seem to be at risk for postoperative language impairment. These results stress the need for language follow-up in posterior fossa tumor survivors.

Getting to grips with the heterogeneity of developmental dyslexia.

The aim of this study was to better understand the heterogeneity of developmental dyslexia by identifying the most common deficits in the reading systems of children with dyslexia with different poor word reading profiles. We classified the word reading profiles of 138 children with developmental dyslexia using nonword and irregular-word reading tests and then used independent experimental tests to explore the cognitive deficits within their word reading systems. The most common deficit associated with primary sublexical impairment (i.e., poor nonword reading) was poor grapheme-phoneme conversion (GPC) knowledge. The most common deficits associated with primary lexical impairment (i.e., poor irregular-word reading) were an impaired orthographic lexicon plus impaired links between this lexicon and the phonological lexicon and semantic knowledge. Finally, the most common deficits associated with mixed reading impairment (i.e., poor nonword reading and poor irregular-word reading) were poor GPC knowledge, an impaired orthographic lexicon, poor links between this lexicon and the phonological lexicon and semantic knowledge, and poor phonological output. We discuss the implications of these findings for theories of reading and for the diagnosis and treatment of dyslexia.

Pirates at parties: letter position processing in developing readers.

There has been much recent interest in letter position coding in adults, but little is known about the development of this process in children learning to read. Here, the letter position coding abilities of 127 children in Grades 2, 3, and 4 (aged 7-10 years) were examined by comparing their performance in reading aloud "migratable" words (e.g., bread, three, diary) with that in reading aloud words without potential for migration (e.g., heard, boot, bitter). Across all three grade levels, children made many more errors on migratable words than on non-migratable words, and the proportion of migration errors did not decrease with increasing grade level. Within each grade, a tendency to make a high proportion of migration errors was not associated with deficits in other reading subprocesses or with general lexical guessing; indeed, it was associated with strong lexical reading skills. We conclude that letter position coding is generally fragile in developing readers and that this interacts with lexical knowledge to produce migration errors in reading.

Phonics training for English-speaking poor readers.

BACKGROUND: Around 5% of English speakers have a significant problem with learning to read words. Poor word readers are often trained to use letter-sound rules to improve their reading skills. This training is commonly called phonics. Well over 100 studies have administered some form of phonics training to poor word readers. However, there are surprisingly few systematic reviews or meta-analyses of these studies. The most well-known review was done by the National Reading Panel (Ehri 2001) 12 years ago and needs updating. The most recent review (Suggate 2010) focused solely on children and did not include unpublished studies. OBJECTIVES: The primary aim of this review was to measure the effect that phonics training has on the literacy skills of English-speaking children, adolescents, and adults whose reading was at least one standard deviation (SD), one year, or one grade below the expected level, despite no reported problems that could explain their impaired ability to learn to read. A secondary objective was to explore the impact of various factors, such as length of training or training group size, that might moderate the effect of phonics training on poor word reading skills. SEARCH METHODS: We searched the following databases in July 2012: CENTRAL 2012 (Issue 6), MEDLINE 1948 to June week 3 2012, EMBASE 1980 to 2012 week 26, DARE 2013 (Issue 6), ERIC (1966 to current), PsycINFO (1806 to current), CINAHL (1938 to current), Science Citation Index (1970 to 29 June 2012), Social Science Citation Index (1970 to 29 June 2012), Conference Proceedings Citation Index - Science (1990 to 29 June 2012), Conference Proceedings Citation Index - Social Science & Humanities (1990 to 29 June 2012), ZETOC, Index to Theses-UK and Ireland, ClinicalTrials.gov, ICTRP, the metaRegister of Controlled Trials, ProQuest Dissertations and Theses, DART Europe E-theses Portal, Australasian Digital Theses Program, Education Research Theses, Electronic Theses Online System, Networked Digital Library of Theses and Dissertations. Theses Canada portal, www.dissertation.com, and www.thesisabstracts.com. We also contacted experts and examined the reference lists of published studies. SELECTION CRITERIA: We included studies that use randomisation, quasi-randomisation, or minimisation to allocate participants to either a phonics intervention group (phonics alone, phonics and phoneme awareness training, or phonics and irregular word reading training) or a control group (no training or alternative training, such as maths). Participants were English-speaking children, adolescents, or adults whose word reading was below the level expected for their age for no known reason (that is, they had adequate attention and no known physical, neurological, or psychological problems). DATA COLLECTION AND ANALYSIS: Two review authors independently selected studies, assessed risk of bias, and extracted data. MAIN RESULTS: We found 11 studies that met the criteria for this review. They involved 736 participants. We measured the effect of phonics training on eight outcomes. The amount of evidence for each outcome varied considerably, ranging from 10 studies for word reading accuracy to one study for nonword reading fluency. The effect sizes for the outcomes were: word reading accuracy standardised mean difference (SMD) 0.47 (95% confidence interval (CI) 0.06 to 0.88; 10 studies), nonword reading accuracy SMD 0.76 (95% CI 0.25 to 1.27; eight studies), word reading fluency SMD -0.51 (95% CI -1.14 to 0.13; two studies), reading comprehension SMD 0.14 (95% CI -0.46 to 0.74; three studies), spelling SMD 0.36 (95% CI -0.27 to 1.00; two studies), letter-sound knowledge SMD 0.35 (95% CI 0.04 to 0.65; three studies), and phonological output SMD 0.38 (95% -0.04 to 0.80; four studies). There was one result in a negative direction for nonword reading fluency SMD 0.38 (95% CI -0.55 to 1.32; one study), though this was not statistically significant.We did five subgroup analyses on two outcomes that had sufficient data (word reading accuracy and nonword reading accuracy). The efficacy of phonics training was not moderated significantly by training type (phonics alone versus phonics and phoneme awareness versus phonics and irregular word training), training intensity (less than two hours per week versus at least two hours per week), training duration (less than three months versus at least three months), training group size (one-on-one versus small group training), or training administrator (human administration versus computer administration). AUTHORS' CONCLUSIONS: Phonics training appears to be effective for improving some reading skills. Specifically, statistically significant effects were found for nonword reading accuracy (large effect), word reading accuracy (moderate effect), and letter-sound knowledge (small-to-moderate effect). For several other outcomes, there were small or moderate effect sizes that did not reach statistical significance but may be meaningful: word reading fluency, spelling, phonological output, and reading comprehension. The effect for nonword reading fluency, which was measured in only one study, was in a negative direction, but this was not statistically significant.Future studies of phonics training need to improve the reporting of procedures used for random sequence generation, allocation concealment, and blinding of participants, personnel, and outcome assessment.

An untapped resource: treatment as a tool for revealing the nature of cognitive processes.

This paper focuses on the role of treatment in cognitive neuropsychological research, arguing that treatment for cognitive impairments should be viewed as a powerful methodology for developing, evaluating, and extending cognitive theories. We suggest that the key aim of cognitive neuropsychology should be characterized as the use of data from the investigation and treatment of individuals with cognitive disorders to develop, evaluate, and extend theories of normal cognition. To support this assertion, this paper discusses examples of how treatment studies have informed theory. The major methodological tool is generalization logic, both generalization across items and generalization across tasks. However, an alternative is to use case series methodology to test predicted correlations between particular cognitive skills and response to treatment. These methods enable explicit testing of a theory or discrimination between theories, focusing on the nature of cognitive representations, the architecture of the cognitive system, and the acquisition of cognitive skills.

The effect of an integrated reading and anxiety intervention for poor readers with anxiety.

A recent systematic review has reported that poor reading is reliably associated with anxiety. However, we currently lack evidence-based intervention for children who have both poor reading and anxiety (PRAX). In this study, we tested a new PRAX intervention in 8- to 12-year-old children using a double-baseline intervention case series design. Analyses of both group and individual data revealed that 12 weeks of PRAX intervention significantly improved children's reading and spelling accuracy, and significantly reduced both anxiety disorders and symptoms. These results support PRAX intervention as a treatment for comorbid reading and anxiety problems in children and pave the way to a randomised controlled trial.

Predicting generalization in the training of irregular-word spelling: treating lexical spelling deficits in a child.

This paper presents a single case study investigating the mechanisms underlying generalization of treatment benefits to untrained words in spelling. Brunsdon, Coltheart, and Nickels (2005) observed that untreated words that improved tended to be those whose errors were closest to being correct prior to treatment. These words also tended to be high in written frequency. The present study employed the same treatment techniques as those used by Brunsdon et al. with K.M., a developmental surface dysgraphic. During a first treatment the characteristics of words whose spelling improved without specific training were identified. These characteristics were then used in a second treatment to test whether it was possible to predict generalization. The results showed that treatment generalization to untreated irregular words was best predicted by neighbourhood size and frequency. We suggest that the processes underlying treatment generalization are based on the interaction between the orthographic lexicon and the graphemic buffer. Clinical implications are discussed.