). Language and academic abilities in children with selective mutism

1. Infant and Child Development Inf. Child. Dev. 18: 271–290 (2009) Published online 18 May 2009 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/icd.624 Language and Academic Abilities in Children with Selective Mutism Matilda E. Nowakowskia, Charles C. Cunninghamb, Angela E. McHolmb, Mary Ann Evansc, Shannon Edisonc, Jeff St. Pierred, Michael H. Boyleb and Louis A. Schmidta,Ã a Department of Psychology, Neuroscience and Behaviour, McMaster University, Hamilton, Ont., Canada b Department of Psychiatry and Behavioural Neurosciences, McMaster University, Hamilton, Ont., Canada c Department of Psychology, University of Guelph, Guelph, Ont., Canada d Child and Parent Resource Institute (CPRI), London, Ont., Canada

3. We examined receptive language and academic abilities in

4. children with selective mutism (SM; n 5 30; M age 5 8.8 years),

6. anxiety disorders (n 5 46; M age 5 9.3 years), and community

8. controls (n 5 27; M age 5 7.8 years). Receptive language and

10. academic abilities were assessed using standardized tests

12. completed in the laboratory. We found a signiﬁcant group by

14. sex interaction for receptive vocabulary scores such that within

15. females, the SM and mixed anxiety groups had signiﬁcantly

17. lower receptive vocabulary scores than community controls. We

19. also found that children with SM and children with anxiety

21. disorders had signiﬁcantly lower mathematics scores than

23. community controls. Despite these differences in mathematics

25. and receptive vocabulary performance, children with SM and

27. children with anxiety disorders still performed at age-level

28. norms, while more children in the community control group

30. performed above age-level norms. Findings suggest that despite

32. their speaking inhibition in the school setting, children with SM

34. are still able to attain the receptive vocabulary and academic

36. abilities that are expected at their age levels. Copyright r 2009

37. John Wiley & Sons, Ltd. Selective mutism (SM) is a disorder in which children fail to speak in certain situations, usually outside the home, despite speaking normally in other situations, usually the home (APA, 1994). Although SM is most commonly diagnosed upon school entry when the demands for children to speak outside of the home increase, SM usually ﬁrst appears in the preschool years (Cunningham, *Correspondence to: Louis A. Schmidt, Department of Psychology, Neuroscience and Behaviour, McMaster University, Hamilton, Ont., Canada L8S 4K1. E-mail: schmidtl @mcmaster.ca Copyright r 2009 John Wiley & Sons, Ltd.

38. 272 M.E. Nowakowski et al. McHolm, Boyle, & Patel, 2004; Steinhausen & Juzi, 1996). The prevalence of SM is estimated at between 0.7% and 2% of children (Bergman, Piacentini, & McCracken, 2002; Elizur & Perednik, 2003; Kopp & Gillberg, 1997; Kumpulainen, Rasanen, Rasska, & Somppi, 1998), with a higher prevalence for immigrant children (Elizur & Perednik, 2003). SM is usually more common in girls than in boys with a ratio ranging between 1.2:1 and 2.6:1 (Black & Uhde, 1995; Dummit et al., 1997; Kristensen, 2000; Steinhausen & Juzi, 1996) and often comorbid with other disorders, including anxiety disorders (Bergman et al., 2002; Cunningham et al., 2004; Cunningham, McHolm, & Boyle, 2006; Dummit et al., 1997; Manassis et al., 2007; Steinhausen & Juzi, 1996; Vecchio & Kearney, 2005; Yeganeh, Beidel, & Turner, 2006; Yeganeh, Beidel, Turner, Pina, & Silverman, 2003), communication disorders (Dummit et al., 1997), developmental delay (Elizur & Perednik, 2003; Steinhausen & Juzi, 1996), and oppositional defiant disorder (Steinhausen & Juzi, 1996; Yeganeh et al., 2006). However, the aetiology of SM is not well understood due to the heterogeneous presentation of the disorder, and the fact that it appears to be influenced by many factors including, genetics, temperament, psychological well-being, attainment of developmental milestones and social factors (Cohan, Price, & Stein, 2006). There is much debate about the conceptualization and classification of SM, which is currently found under the ‘Other Disorders of Childhood’ category in the Diagnostic and Statistical Manual 4th Edition (APA, 1994). Some researchers have argued that SM should be viewed as a distinct anxiety disorder (Anstendig, 1999; Sharp, Sherman, & Gross, 2007; Vecchio & Kearney, 2005) or an extreme variant of social phobia (Yeganeh et al., 2006). This view is supported by similarities in the definition of SM and social phobia, both characterized by an intense fear of social situations in which embarrassment may occur as well as a lack of inhibition and anxiety in the home setting (APA, 1994). As well, a large number of studies have found that SM co-occurs at high rates with anxiety disorders, especially social phobia. For instance, both Vecchio and Kearney (2005) and Dummit et al., (1997) using sample sizes of 15 and 50 respectively, reported that 100% of the selectively mute children in their studies met the diagnostic criteria for social phobia and around 50% of the selectively mute children met the criteria for a second anxiety disorder. Another study by Black and Uhde (1995) found that 97% of the children who were diagnosed with SM met the diagnostic criteria for social phobia. Longitudinal studies have also found that individuals with a childhood history of SM experienced shyness and symptoms of social anxiety as adults and adolescents despite being ‘cured’ from their SM (Joseph, 1999). As well, family studies have found that parents of children with SM have higher rates of anxiety disorders, especially social phobia (Black & Uhde, 1995), compared with parents of normally developing children (Kumpulainen, 2002). Numerous researchers (Black & Uhde, 1995; Dummit et al., 1997) have highlighted the similarities between children with SM and children classified as behaviourally inhibited. According to Kagan, Reznick, & Snidman (1987), a failure or reluctance to speak is one of the defining characteristics of behavioural inhibition. Studies have found that parents of children with SM retrospectively report that their children were always shy from their first years of life (Garcia, Freeman, Francis, Miller, & Leonard, 2004). Steinhausen and Juzi (1996) reported that 85% of the children with SM in their sample exhibited high levels of shyness. Given that there is evidence that children with behavioural inhibition are at an increased risk for the development of anxiety disorders (Oberklaid, Sanson, Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

39. Language and Academic Abilities in Children 273 Pedlow, & Prior, 1993), many researchers have suggested that behavioural inhibition may be an early predictor of SM. Despite the similarities in aetiology and presentation between SM and anxiety disorders (especially social phobia), SM differs from anxiety disorders in that it is associated with higher rates of expressive language difficulties and developmental delays. Steinhausen and Juzi (1996) found that 38% of the children with SM in their sample exhibited problems with expressive language, articulation, and stuttering. Similarly, McIness, Fung, Manassis, Fiksenbaum, & Tannockand (2004) reported that compared with children with social anxiety, the speech of children with SM was linguistically simpler and shorter. In terms of general developmental delays, Kristensen (2000) found that 68% of their sample of children with SM met the criteria for a developmental delay. Thus, difficulties in expressive language abilities as well as general developmental delays appear to be distinguishing factors between SM and social phobia. Further research comparing children with SM to those with anxiety disorders is necessary to better understand the similarities and differences in how the disorders present themselves and impact children’s functioning. One area of importance in terms of its impact on children’s functioning is whether SM affects children’s performance in school. Verbal participation at school is considered an important part of the learning process (Daly & Korinek, 1980). Given that SM is characterized by a lack of speech, it is reasonable to suspect that children with SM may score lower on tests of academic abilities due to their inability to ask questions when they require clarification. This inability may in turn result in a lack of homework completion and an accumulation of deficits due to not receiving the clarification necessary early on and falling be- hind as the material becomes more difficult. As well, much incidental learning occurs in the school setting through social and collaborative conversations with peers. Owing to their lack of speech, children with SM may miss out on such conversations. Past studies investigating academic abilities in children with SM have yielded varying results. For example, some researchers have reported that children with SM perform significantly lower than community controls on academic measures (Bergman et al., 2002; Kristensen & Oerbeck, 2006; Schwartz, Freedy, & Sheridan, 2006), especially language abilities (Manassis et al., 2007; McInnes et al., 2004). In comparison, other studies have found no differences between children with SM and community controls (Cunningham et al., 2004). The discrepancies between studies may be explained by three factors: (1) the use of different standardized measures, (2) the utilization of different testing methods such that some studies relied on methods that required verbal responses while others used methods that allowed for non-verbal responses, and (3) different foci of interest in terms of academic skill sets. In a sample of 52 children with SM and 52 community controls, Cunningham et al., (2004) reported no differences between groups on standardized tests of math and reading or on teacher reports of children’s math, reading, and overall academic abilities. On the other hand, Manassis et al., (2007) recently found that children with SM (n 5 44) performed significantly less well than children with anxiety disorders (n 5 28) and community controls (n 5 19) on a standardized test measuring receptive vocabulary abilities. However, a closer look at the group means for the receptive vocabulary standardized scores in this study showed that children with SM tended towards average performance for their age levels, whereas children with anxiety disorders and community controls tended to perform at a higher level. Further, McInnes et al., (2004) investigated a small Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

40. 274 M.E. Nowakowski et al. sample of seven children with SM and seven children with social phobia. The children’s expressive language abilities were tested by having children retell stories to their parents that they heard in the laboratory and in the home setting and analyzing the content of their narratives. The authors found that, compared with the social phobia group, the children with SM had narratives that were significantly shorter, less detailed, and linguistically simpler (McInnes et al., 2004). Based on the current literature, it appears that children with SM have lower receptive vocabulary and expressive language abilities (although not in the clinical range) but normal math and reading abilities. However, to date, most research (with the exception of Cunningham et al., 2004) has focused on expressive and receptive language abilities. The current study attempted to expand on the past research in three ways. First, we investigated a wide range of academic areas (receptive language, mathematics, reading, and spelling) in one sample. Comparing across previous studies of different academic areas to obtain an understanding of the overall academic performance of children with SM is difficult given that different researchers may use different criteria to identify their samples. Consequently, a study investigating all academic areas in one sample provides stronger insight into the academic performance in children with SM. Such insight is important to better formulate treatment methods for these children. Second, in addition to a control group, we also included a group of children with anxiety disorders as a comparison. With the exception of the study by Manassis et al., (2007), no other study has looked across all three groups (SM, anxiety, and control) to determine differences and similarities in academic abilities. Third, academic areas other than language have been largely neglected in the literature on children with SM (notwithstanding Cunningham et al., 2004). Thus, we attempted to look across all academic areas to get a fuller picture of how SM affects academic performance. We compared children with SM to children with anxiety disorders, and community controls on spelling, math, reading, and receptive vocabulary abilities using standardized methods that did not require the children to provide the experimenter with verbal responses. We addressed the following two questions: (1) Do children with SM differ from children with anxiety disorders, and community controls on standardized measures of reading, math, spelling, and receptive vocabulary abilities? and (2) Given that SM is slightly more common in females than males, are there sex differences within and between groups on the standardized measures of academic abilities? Past studies have shown that children with SM perform lower on receptive vocabulary tests compared with children with anxiety disorders and community controls but that these differences are not in the clinical range. Thus, we further hypothesized that children with SM would perform lower than children with anxiety disorders and community controls on our measure of receptive vocabulary ability but that these differences would not be in the clinical range. Given the scarcity of research on the performance of children with SM in areas other than language, we did not make any specific predictions as to how the three groups would compare in their performance on standardized measures of mathematics, spelling, and reading. However, we were interested in investigating whether the pattern of lower scores on language tests in children with SM would extend to other academic areas. Similarly, given that no past studies have reported group by sex interactions in the determination of group differences for academic performance, we made no specific predictions about possible sex by group interactions and rather used the present study as an opportunity to explore possible interactions. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

41. Language and Academic Abilities in Children 275 METHOD Participants Participants were 103 children and their parents recruited from local children’s mental health agencies. This included one agency that had a regional SM programme as well as a Child Database contained in the Department of Psychology, Neuroscience and Behaviour at McMaster University that comprises the names of healthy community children recruited at birth from McMaster University Medical Centre and St. Joseph’s Healthcare, Hamilton, Ontario. Participant demographics are presented in Table 1. All primary caregivers completed a package of diagnostic questionnaires (described below) that was used to determine group membership for the current study. We focused on children between the ages of 6 and 10 years old because this is the age range at which children with SM are usually ﬁrst diagnosed (Kumpulainen et al., 1998; Standart & LeCouteur, 2003) and many of the primary caregivers who sought support from the local mental health agency specializing in SM had children in this age range. Given that we did not want signiﬁcant age differences between groups, we focused our recruitment from the Child Database of normally developing children on the same age range. SM group The inclusion criteria for the SM group were as follows: (1) the primary caregiver indicated that the child failed to speak (i.e. never talks) in two or more situations on the Speech Situations Questionnaire—Parent Version (SSQ-Parent; Cunningham et al., 2006; Cunningham et al., 2004) or the teacher indicated that the child failed to speak (i.e. never talks) in two or more situations on the Speech Situations Questionnaire—Teacher version (SSQ-Teacher); (2) the lack of speaking was not due to a communication disorder; and (3) the lack of speaking persisted for a minimum of 1 month. Thus, the SSQ questionnaires completed by the primary caregiver and teacher were focused on identifying cases of stable, contextual lack of speech in a variety of situations in the general community (parent questionnaire) and in the school setting (teacher questionnaire). In total, 30 children (14 males, 16 females), all from the mental health agencies, met these criteria. The mean scores for children in the SM group on the SSQ-Parent and SSQ-Teacher were 17.44 (range 5 9–30) and 4.92 (range 5 0–14), respectively. As indicated by the ranges, the maximum possible values for the parent and teacher SSQ were reached. This was due to two participants whose primary caregivers indicated that their children spoke normally in all situations but whose teachers indicated that they never spoke in two or more situations in the school setting. Similarly, there were seven children whose parents indicated that they failed to speak in two or more situations, but whose teachers indicated that they spoke normally in the school setting. All of the children in the SM group received at least one rating from either their primary caregiver or their teacher indicating that they never spoke in some aspect of the school setting (i.e. in the classroom to their friends, in the classroom to the teacher, in front of the class, etc.). The primary caregiver also completed the internalizing section of the Com- puterized Diagnostic Individual Schedule for Children (C-DISC IV; Shaffer, Fisher, Lucas, Dulcan, & Schwab-Stone, 2000) over the phone to assess for comorbid internalizing disorders. The C-DISC was not available for 2 (6%) of the children in the SM group because a time could not be scheduled with the primary caregiver to complete the C-DISC over the phone. Of the 28 children with SM for Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

42. 276 Table 1. Demographic characteristics for the full sample (N 5 108) Selective mutism Mixed anxiety Community control Statistic p-Value (n 5 30) (n 5 46) (n 5 27) Percent female 53 48 56 w2(2) 5 0.47 p 5 0.79 Mean Age of Child (years) 8.8 (3.5) 9.3 (2.8) 7.8 years (2.4) F(2, 100 ) 5 2.25 p 5 0.11 Percent in regular classroom 97% 100% 100% w2(2) 5 2.39 p 5 0.30 Copyright r 2009 John Wiley & Sons, Ltd. Age range of primary caregiver (years)a 19–39 (n 5 12), 19–39 (n 5 18), 19–39 (n 5 12), w2(4) 5 2.74 p 5 0.60 40–64 (n 5 18) 40–64 (n 5 28) 40–64 (n 5 15) Percent English spoken at home 90 100 100 w2(2) 5 7.78 p 5 0.02 Percent of participants who fall in each w2(2) 5 28.98 p 5 0.004 of the following total income levels Less than $15 000 0 4 0 $15 000–$30 000 7 15 0 $30 000–$45 000 13 9 0 $45 000–$60 000 0 15 0 $60 000–$75 000 37 9 28 $75 000–$90 000 23 20 40 Greater than $100 000 20 28 32 a Differences between sample sizes for groups and the number of responses are due to missing data. M.E. Nowakowski et al. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

43. Language and Academic Abilities in Children 277 which the C-DISC was available, 15 (54%) had one or more comorbid internalizing disorders. The C-DISC was used to determine the diagnoses of anxiety disorders because it has been shown to have good reliability in assessing anxiety disorders (Shaffer et al., 2000). However, the C-DISC has not been used in the past for diagnosing SM, and its reliability for this diagnosis has not been established. Accordingly, we used the parent and teacher-report questionnaires to identify children with SM and the C-DISC to identify comorbid anxiety disorders. The mean age of the children in the SM group was 8.8 years (S.D. 5 3.46 years). With the exception of one child, all of the children with SM were in a regular classroom and followed the regular academic programme. In all but three of the homes, English was the language most commonly spoken (see Table 1). Mixed anxiety group The inclusion criteria for the mixed anxiety group were that the children had to have one or more anxiety disorder diagnoses on the C-DISC IV (social phobia, separation anxiety, speciﬁc phobia, panic disorder, post-traumatic stress disorder, obsessive compulsive disorder, or agoraphobia) and no diagnosis of SM, as determined by teacher and parent ratings on the SSQs. Forty-six children (24 males, 22 females), 11 (24%) of whom were recruited from the child database, met these criteria. The mean scores for the parent and teacher SSQ questionnaire for the mixed anxiety group were 26.82 (range 5 8–30) and 12.89 (range 5 5–14), respectively. Of the 46 children in the mixed anxiety group, 25 (54%) had one anxiety disorder while 21 (46%) had two or more anxiety disorder diagnoses according to the C-DISC. It should be noted that although we had originally intended to have a pure social phobia group as our anxiety comparison group, we found that the high comorbidity of anxiety disorders made the attainment of such a group very difﬁcult. Given the wide range of anxiety disorders that we found in our sample, we decided to create a mixed anxiety group based on the criteria outlined above. The mean age of the children was 9.28 years (S.D. 5 2.78 years). All the children were in a regular classroom and followed the regular classroom programme. In all the families, English was spoken most commonly in the home (see Table 1). Community control group The selection criteria for the community control group were that they were: (1) healthy children from the Child Database; (2) had no anxiety disorders as determined by the C-DISC IV; and (3), had no diagnosis of SM, as determined by teacher and parent ratings on the SSQ. The mean scores for the parent and teacher SSQ questionnaires were 29.39 (range 5 27–30) and 13.21 (range 5 10–14), respectively. Twenty-seven children (12 males, 15 females) met these criteria. The mean age of the children was 7.8 years (S.D. 5 2.4 years). All the children were in a regular classroom and followed the regular academic programme. In all of the homes, English was the most common language spoken (see Table 1). Procedures This study was part of a larger study investigating the aetiology, familial background, academic performance, and behavioural and psychophysiological characteristics of children with SM. The present paper focused only on the academic and receptive vocabulary measures that were collected. All procedures Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

44. 278 M.E. Nowakowski et al. were approved by the McMaster University Health Sciences Research Ethics Board. All children and primary caregivers were tested at the Child Emotion Laboratory at McMaster University. Upon arrival at the laboratory, the primary caregiver and child were briefed about the procedures and consent forms were signed. The primary caregiver also provided the name of the child’s current school and grade teacher. All of the measures in the laboratory were completed in a quiet room with a video camera that was visible to both the child and primary caregiver/experimenter. With the exception of the Reading Recognition subtest, all the measures were completed with the experimenter while the primary caregiver was in the other room. The laboratory visit took approximately 2 h. In order to decrease the length of the laboratory visit, the parents were contacted after the visit to complete the C-DISC over the phone. The children received a toy at the end of the study as a token of our appreciation for their participation. Following the completion of the laboratory visit, a package of questionnaires was sent to the child’s teacher asking him/her to complete the enclosed questionnaires within a week of receiving them and to send them back to the laboratory in the included self-addressed envelope. The teachers were not compensated for their participation. Parent and Teacher Measures Diagnostic questionnaires completed by primary caregiver The primary caregiver completed the SSQ-Parent (Cunningham et al., 2004; Cunningham et al., 2006) as a measure of the child’s speaking patterns. The SSQ-Parent is a 15-item questionnaire in which parents rate their children’s speaking in a variety of situations, including the home, school, and community, and to a range of different people, including parents, friends, teachers, and strangers on a 3-point scale (0 5 never talks, 1 5 whispers, 2 5 talks in a normal voice). For the purpose of diagnosing children with SM, the primary caregiver had to select never talks (i.e. 0) for a minimum of two situations. The internal consistency of the SSQ—Parent Version in a past study was 0.82 (Cunningham et al., 2006) and in the present study was 0.92. The primary caregiver also completed the internalizing disorders portion of the C-DISC IV (Shaffer et al., 2000). The C-DISC IV is a structured diagnostic interview based on the DSM-IV that assesses children for 34 psychiatric disorders. It has been shown to have good reliability (Shaffer et al., 2000). The C- DISC IV was administered to the primary caregiver over the phone by a trained research assistant after the laboratory visit. To decrease the amount of time the assessment took and due to the fact that the present study was focused on internalizing disorders, only the internalizing disorders section of the C-DISC IV was administered. The children were assessed for the following internalizing disorders: panic disorder, generalized anxiety disorder, social phobia, speciﬁc phobia, separation anxiety, obsessive compulsive disorder, post-traumatic stress disorder, agoraphobia, and major depression. Diagnostic questionnaires completed by teachers The teachers completed the SSQ-Teacher to evaluate student’s speech patterns in the school setting. The SSQ-Teacher is a 7-item questionnaire in which teachers assess children’s speech in a variety of school settings, such as the playground, the hallway and the classroom, and to a variety of individuals, including friends Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

45. Language and Academic Abilities in Children 279 and teachers, on a 3-point scale (0 5 never talks, 1 5 whispers, 2 5 talks in normal voice). For the purpose of diagnosing children with SM, the teacher had to select never talks (i.e. 0) for a minimum of two situations. The internal consistency of the SSQ-Teacher in the present study was 0.95. Child Academic Skill and Receptive Vocabulary Measures PPVT-III The receptive vocabulary of the child was measured through the Peabody Picture Vocabulary Test—3rd Edition (PPVT-III; Dunn & Dunn, 1997), which comprises 204 words sequenced to be increasingly challenging. The experimenter says a target word and the child is presented with four line drawings and has to indicate which one of the line drawings corresponds to the word. The PPVT-III was introduced to the children as a picture game. The PPVT-III raw scores were converted to standard age-normed scores. The internal consistency of the PPVT- III has been estimated at 0.90 (Dunn & Dunn, 1997). PIAT-R The Reading Recognition, Mathematics, and Spelling subtests of the Peabody In- dividual Achievement Test—Revised (PIAT-R; Dunn & Markwardt, 1998) were conducted as standardized measures of children’s oral reading, mathematics, and spelling abilities. These subtests were completed only by children who were 5 years old or older, given that the youngest age for which the assessment is standardized is 5 years. The internal consistency of the subtests of the PIAT-R ranges from the low to mid 0.90 s. All the assessments from thePIAT-R were videotaped. The Reading Recognition subtest comprises 100 words that are arranged in order of difficulty. The child is presented with 12 words on a page and is asked to read them out loud to the tester. Unlike the other subtests, given that this task requires verbal responses the children completed it with their primary caregiver while the experimenter was out of sight. The primary caregiver was instructed by the experimenter while the child was in an adjoining room to read the provided instructions to the child asking the child to read the presented words from left to right. The primary caregiver was told to indicate on the provided form whether the child read the word correctly but to not let the child know his/her accuracy. Testing was complete when the child made five mistakes on seven consecutive words. The experimenter entered the room with the primary caregiver to set up the materials for the reading recognition subtest and told the child that he/she would be doing a word game with his/her parent and then left the room. The Mathematics subtest comprises 100 math questions that are arranged in order of difficulty. The experimenter reads a math question to the child, and the child is presented with four numbered options, one of which is the correct answer and three that are distracters. The PIAT-R-M was introduced to the children as a number game. The children indicated their answers by either saying the number that corresponded to the correct answer or pointing to the correct answer. The Spelling subtest comprises 100 words that are arranged in order of difficulty. The experimenter reads a target word, reads a sentence that contains the target word, and then reads the target word again. The child is presented with four numbered options, one of which is the target word spelled correctly and three of which are distracters. The subtest was introduced to the children as a word game. The children indicated the correct answer either by saying the number that corresponded to the correct answer or pointing to the correct answer. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

46. 280 M.E. Nowakowski et al. Data Loss There was some variability in the number of participants who completed the standardized tests due perhaps to fatigue or anxiety in the children, or a lack of availability of the testing materials. Of the 26 children with SM who were 5 years old or older and were eligible to complete all subtests of the PIAT-R, 7 (27%) failed to complete the Reading Recognition subtest, 1 (4%) failed to complete the Mathematics subtest, and 1 (4%) failed to complete the Spelling subtest. Of the 30 children with SM, 5 (17%) failed to complete the PPVT-III (4 of these were due to an initial lack of availability of testing materials). Of the 44 mixed anxiety children who were 5 years old or older and eligible to complete all subtests of the PIAT-R, 3 (7%) failed to complete the Reading Recognition subtest. Of the full 46 children in the mixed anxiety group, 3 (6%) failed to complete PPVT-III (2 of these were due to a lack of available testing materials). Of the 23 children in the community control group who were 5 years old or older and eligible to complete all subtests of the PIAT-R, 1 (4%) failed to complete the Reading Recognition subtest. All the children in the community control group completed the PPVT-III. RESULTS Preliminary Analyses There was a statistically significant difference on language spoken at home (w2(2) 5 7.78, p 5 0.02; see Table 1). All the children in the community control and mixed anxiety groups spoke English most frequently at home while 3 of the 30 children with SM spoke a different language at home. However, when we eliminated the three children with SM who spoke a different language at home, the differences between the groups for the standardized tests of academic performance were still statistically significant. Therefore, we included the three children with SM who spoke a different language at home in the analyses presented here. There were no significant correlations between the standardized scores for the academic measures (PPVT-III, reading recognition, mathematics, and spelling subtests of the PIAT-R) and income levels. Therefore, we did not consider the differences in income levels as confounding factors in our analyses. Academic Standardized Measures Means and standard deviations for all the dependent variables are shown in Table 2. Table 2. Means and standard deviations for the PPVT-III standardized scores among males and females in the three groups PPVT-III Male Female Selective Mixed Control Selective Mixed Control mutism anxiety (n 5 12) mutism anxiety (n 5 14) (n 5 12) (n 5 24) M (S.D.) (n 5 13) (n 5 19) M (S.D.) M (S.D.) M (S.D.) M (S.D.) M (S.D.) 102.67 108.83 108.58 105.92 102.53 116.07 (12.87) (11.27) (14.00) (9.27) (13.76) (9.02) Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

47. Language and Academic Abilities in Children 281 Figure 1. Standardized scores for the Peabody Picture Vocabulary Test—3rd Edition for the selective mutism (n 5 30), mixed anxiety (n 5 46), and community control (n 5 27) groups among females and males. The error bars are 95% confidence intervals. PPVT-III To examine receptive vocabulary differences among the three groups and the sexes, a 3 Â 2 analyses of variance (ANOVA) was conducted with Group (SM, Mixed Anxiety, and Community Control) and sex (male and female) as the independent variables and PPVT-III standard score as the dependent measure. There was a significant main effect for group (F(2, 88) 5 3.51, p 5 0.03, Z2 ¼ 0:074) p as well as a significant group by sex interaction on the standard score of the PPVT-III (F(2, 88) 5 3.02, p 5 0.05, Z2 ¼ 0:064; see Figure 1). Given the significant p interaction, we conducted simple main effects to decompose the interaction (using LSD tests to compare relevant means). Results indicated that for girls, both the SM (n 5 13; M 5 105.92, S.D. 5 9.27) and the mixed anxiety (n 5 19; M 5 102.53, S.D. 5 13.76) groups attained significantly lower standardized scores on the PPVT-III compared with the control group (n 5 14; M 5 116.07, S.D. 5 9.02, p 5 0.03 and 0.002, respectively). In contrast, for boys there were no significant differences between the SM (n 5 12; M 5 102.67, S.D. 5 12.87) and mixed anxiety (n 5 24; M 5 108.83, S.D. 5 11.27; p 5 0.15) groups, the SM and control groups (n 5 24; M 5 108.58, S.D. 5 14.00; p 5 0.23), or the mixed anxiety and control groups (p 5 0.95). We also conducted w2 tests to look at differences between groups within each sex in the frequency of children in each group who scored at or below the 25th percentile or at or above the 75th percentile on the PPVT-III. Among girls, there was a significant difference between groups with only 5 out of 13 (i.e. 38%) girls from the SM group and 5 out of 19 (i.e. 26%) of girls from the mixed anxiety group scoring at or above the 75th percentile, whereas 12 out of 14 (i.e. 86%) girls Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

48. 282 M.E. Nowakowski et al. from the community control group scored at or above the 75th percentile (w2(2) 5 12.03, p 5 0.002). We found no significant differences between groups in the number of girls who scored at or below the 25th percentile on the PPVT-III (w2(2) 5 2.55, p 5 0.28). For boys, we found no significant differences between groups in the number of children who scored at or below the 25th percentile (w2(2) 5 0.67, p 5 0.72) or at or above the 75th percentile (w2(2) 5 0.17, p 5 0.92) on the PPVT-III. PIAT-R To examine differences between groups and sexes on the standardized scores for the PIAT-R subtests, a two-way multivariate analysis of variance (MANOVA) was conducted with Group (SM, Mixed Anxiety, Community Control) and Sex (male and female) as the independent variables and the three PIAT-R subtests (reading recognition, math, and spelling) as the dependent variables. The MANOVA failed to yield a significant group by sex interaction, but did yield a statistically significant main effect of group, which was followed up with AN- OVA on the dependent variables. All significant ANOVAs were further followed up using post hoc Tukey HSD tests given that the equal variances assumption was met. There was a significant main effect for group on the PIAT-R (F(6, 152) 5 3.58, p 5 0.002, Z2 ¼ 0:13). The follow-up one-way ANOVAs on the dependent mea- p sures showed that the groups differed significantly only on the math subtest of the PIAT-R (F(2, 77) 5 6.39, p 5 0.003, Z2 ¼ 0:14; see Figure 2). Both the SM p (M 5 96.00, S.D. 5 14.08) and mixed anxiety groups (M 5 100.45, S.D. 5 12.45) had significantly lower mathematics scores than the community control group Figure 2. Standardized scores for the Mathematics Assessment of the Peabody Individual Assessment Test—Revised for the selective mutism (n 5 30), mixed anxiety (n 5 46), and community control (n 5 27) groups. The error bars are 95% confidence intervals. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

49. Language and Academic Abilities in Children 283 Table 3. Means and standard deviations for all three groups for the reading, math, and spelling subtests of the PIAT-R PIAT-R Selective mutism Mixed anxiety Community control (n 5 30) (n 5 46) (n 5 27) M S.D. M S.D. M S.D. Reading Recognitiona 112.26 20.13 106.93 12.17 109.64 13.08 Matha 96 14.08 100.45 12.45 110.86 14.83 Spellinga 103 14.35 104.48 13.89 106.27 12.49 a Analyses completed on the following sample sizes due to missing data: Selective mutism (n 5 19), Mixed Anxiety (n 5 42), Control (n 5 22). (M 5 110.86, S.D. 5 14.83) (Table 3). There were no significant differences between the SM and the mixed anxiety groups (p 5 0.47). There were no significant differences between the groups on the spelling (SM and mixed anxiety, p 5 0.92; SM and control, p 5 0.73; mixed anxiety and control, p 5 0.88) and reading recognition subtests of the PIAT-R (SM and mixed anxiety, p 5 0.40; SM and control, p 5 0.84; mixed anxiety and control, p 5 0.77). Using w2 tests, we also found a significant difference between groups in the number of participants who scored at or above the 75th percentile on the mathematics subtest of the PIAT-R (w2(2) 5 12.71, p 5 0.002). Only 1 child (i.e. 4%) from the SM group and 10 (i.e. 23%) children from the mixed anxiety group scored at or above the 75th percentile on the math subtest of the PIAT-R, while 11 children (i.e. 48%) from the community control group scored at or above the 75th percentile on the math subtest of the PIAT-R. There were no significant differences in the number of participants who scored at or below the 25th percentile on the math subtest of the PIAT-R (w2(2) 5 2.08, p 5 0.35). There were no other differences between groups in the distribution of scores for the subtests of the PIAT-R. DISCUSSION The goal of this study was to explore the language abilities of children with SM. We found that girls with SM and mixed anxiety performed significantly lower on tests of receptive vocabulary skills as compared with community controls, but there were no significant differences between groups for boys. We also found that children with SM and mixed anxiety scored significantly lower than community controls on tests of mathematics skills (regardless of sex). There were no significant differences between children with SM and children with mixed anxiety on any of the other standardized measures. However, despite the statistically significant differences, the average receptive vocabulary standardized scores for girls with SM and mixed anxiety were age-appropriate compared with normative data (i.e. 106 and 102). Similarly, the standardized scores for the mathematics for children with SM and children with mixed anxiety were average or age-appropriate compared with normative data (i.e. 96 and 100, respectively). Similar results were found when looking at the percentage of children scoring at or below the 25th percentile and at or above the 75th percentile for receptive vocabulary and mathematics performance. That is, fewer individuals with SM and mixed scores Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

50. 284 M.E. Nowakowski et al. at or above the 75th percentile on the standardized tests of mathematics and receptive vocabulary (girls only) compared with the community controls. Language Skills In terms of receptive language, it may appear that girls with SM and mixed anxiety are neither significantly impaired nor significantly advanced in their receptive vocabulary abilities, whereas comparison girls were above average in these abilities. However, research has also shown that receptive vocabulary abilities as measured by the PPVT-III tend to be inflated, resulting in children with receptive language difficulties often not being identified as clinically impaired (Ukrainetz & Duncan, 2000). For example, William (1998) reported that children obtained standardized scores that were about 10 points higher on the PPVT-III than on the PPVT-R (Dunn & Dunn, 1981). Moreover, this score inflation was especially evident in children between the ages of 4 and 10 (the age of focus for the present study). Therefore, it may be the case that the children with SM are in fact performing below their age-appropriate norms but this is not being revealed through the standardized scores on the PPVT-III. Notwithstanding the group by sex interaction, there was a significant main effect for group on receptive vocabulary ability with children with SM scoring significantly lower (but within the age-appropriate range) than community controls. This is consistent with previous studies that have found lower scores on the PPVT-III in children with SM (Kristensen & Oerbeck, 2006; Manassis et al., 2007; McInnes et al., 2004). Unlike these previous studies, however, we did not find any significant differences between the SM and the mixed anxiety groups in receptive vocabulary scores. One possible explanation for this discrepancy in results is that in the study conducted by McInnes et al., (2004), the SM group had children who had only a diagnosis of SM and the social phobia group comprised children who had only a diagnosis of social phobia. In contrast, over half of the children in our SM group had one or more comorbid anxiety disorders and our mixed anxiety group consisted of a number of different anxiety disorders. In addition, even though the study conducted by Manassis et al., (2007) included children with SM who had a comorbid anxiety disorder, the majority of these children had comorbid social phobia. In contrast, in our study there was a wide variety of comorbid disorders for children with SM, with the most common being specific phobia. As well, differences in age between the SM group and the anxiety group in previous studies (e.g. Manassis et al., 2007) may have also influenced results. 1 The main effects we noted for group on receptive vocabulary are also consistent with previous work investigating the expressive and receptive language abilities of temperamentally shy children (Rubin, 1982; Spere, Schmidt, Theall- Honey, & Martin-Chang, 2004). Moreover, these findings also provide further evidence for the similarities that have been highlighted by a number of researchers (Black & Uhde, 1995; Dummit et al., 1997) between children with SM and Kagan’s (1987) conceptualization of the behaviourally inhibited child. Like previous studies, we also found that children with SM performed significantly lower than controls on receptive vocabulary tests. Based on standardized scores, these differences did not reflect serious receptive vocabulary deficits (e.g. Spere et al., 2004), although, as discussed above, this may be a reflection of the score inflation that is seen in the PPVT-III. Our study, however, suggests that the picture is more complex as we also found a significant group by sex interaction, whereby only girls with SM and females with mixed anxiety Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

51. Language and Academic Abilities in Children 285 performed significantly lower than community control on receptive vocabulary. Again, however, this significant difference did not place the females with SM or mixed anxiety as having a deficit in receptive vocabulary abilities based on standardized scores but rather reflected the fact that the community control girls were high average on receptive vocabulary performance. This finding is consistent with research that has shown that girls tend to be more advanced than same-aged boys in their language development, especially with regards to vocabulary (Bornstein, Hahn, & Haynes, 2004). One mechanism that has been suggested for this advanced language ability in females is that the activities that are stereotyped as being female and that young girls are likely to engage in promote collaboration and proximity to others (Caldera, Huston, & O’Brien, 1989). Consequently, girls who do not engage in such activities are unlikely to attain the advanced language abilities, helping to explain why only the girls in the SM and mixed anxiety groups scored significantly lower on receptive vocabulary compared with females in the community control group. Mathematic Skills With regards to mathematics, the only other study that we are aware of that has specifically investigated mathematics skills in children with SM was conducted by Cunningham et al., (2004), who found no significant differences in mathematics abilities between children with SM and community controls. There are two possible explanations for the discrepancies in findings on mathematics abilities between the present study and the study conducted by Cunningham et al., (2004). First, the children with SM in the study conducted by Cunningham et al., (2004) had a mean age of 7.1 years, and thus were younger than the children with SM in our study, who had a mean age of 8.8 years. It is possible that the differences that are observed between children with SM and community controls on mathematics standardized scores are not seen early in the child’s education but manifest themselves as the children get older and academic demands in mathematics increase. 2 Second, in the study conducted by Cunningham et al., (2004), the standardized math scores for both groups of children were actually lower than the scores in the present study. The difference was that the standardized math scores for the community controls in the Cunningham et al., (2004) study were substantially lower than those of the community control group in the current study. A possible explanation for the differences in these standardized scores is that Cunningham et al., (2004) used a more sophisticated sampling strategy to compose the community control group and had a higher participation rate. The tendency for fewer children with SM to score above their age levels for mathematics may be a reflection of their lack of speaking in school situations. Research has shown that children who explain and discuss new concepts with their classmates, enhance their learning (Nattive, 1994). Thus, is likely that children elaborate and solidify their learning of mathematical concepts by discussing the concepts with their classmates. However, children with SM typically will not engage in such discussions. Consequently, these children may lack the solidifi- cation of concepts that occurs through discussion. Furthermore, children with SM may be less likely to ask teachers questions when they do not understand something. Given that mathematics concepts tend to build upon each other, children with SM may experience more difficulties with mathematics due to their lack of ability to ask for help when needed. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

52. 286 M.E. Nowakowski et al. Spelling and Reading Skills There were no significant differences between children with SM and community controls on spelling and reading abilities. This result suggests that children with SM have specific academic areas where they may score lower than community controls (i.e. math and receptive vocabulary). It is possible that parental involvement is reflected in the results. Parents view the school as primarily responsible for the teaching of mathematics but often see the home as playing a key role in the development of literacy skills (Evans, Fox, Creamos, & McKinnon, 2004). Parents of children with SM may take a particularly active approach in developing their children’s reading abilities given that their children fail to participate verbally in the school setting, and work with their children at home to teach them to read. These parental interventions in the home setting where children with SM are most comfortable may help to ensure that they learn to read at the same level as their peers. Given that little past research has looked at spelling and reading abilities in children with SM, these results need further replication to fully understand their meaning and implications. Strengths and Limitations The current study had a number of strengths. First, it appears to be the first study to examine all aspects of academic performance in one sample of children with SM. Second, we used methods to evaluate children’s academic abilities that did not rely on verbal responses. With the exception of the reading assessment, which was done with the primary caregiver, the nature of all the assessment was such that they could be completed by the child either verbally or non-verbally. Thus, the lack of speaking found in children with SM was not a confounding factor in evaluating their academic abilities. Third, given the rarity of SM, obtaining sample sizes large enough to confidently assess differences between groups is a challenge. The current study had a relatively large group of children with SM, thus enabling the statistical power required for valid comparisons between groups. Despite the aforementioned strengths, the current study also has a number of limitations. Although, as mentioned above, our sample was quite large con- sidering the rarity of the disorder, our investigation of group by sex interactions for academic performance resulted in comparisons of fairly small sample sizes. Future studies with larger samples are necessary to replicate our results. Further, given that SM is highly comorbid with other anxiety disorders, we were not able to have a pure SM group. Therefore, it is possible that we did not observe differences between the SM and mixed anxiety groups because over half of the children in the SM group also had an anxiety disorder. As well, due to the high rate of comorbidity, we cannot determine whether the difference between the SM group and the community controls is due to the speaking inhibition of the children with SM or due to their anxiety levels. 3 The results need to be replicated with a larger and more pure sample to fully delineate the effect of SM as opposed to anxiety per se on receptive vocabulary ability and academic performance in reading, mathematics, and spelling. Second, although the children with SM were not required to provide verbal responses for the academic tasks, the presence of the experimenter may have resulted in high levels of anxiety and inhibition in the children. Thus, the group differences found for mathematics and receptive vocabulary abilities could possibly be a reflection of general inhibition rather than a lack of ability or Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

53. Language and Academic Abilities in Children 287 knowledge (although it is not clear why this would affect just mathematics and receptive vocabulary). Third, again due to the high comorbidity of different anxiety disorders, our mixed anxiety group was a combination of different types of anxiety disorders. There was also differing levels of impairment in the mixed anxiety group (e.g. single vs multiple anxiety disorders). Fourth, the inclusion criteria for our community control group were very strict, given that the children not only had to have no diagnoses on the C-DISC, but they also had to be recruited from the child database. This may have limited the generalizability of our results in two ways: (a) it is possible that the differences that we saw between the children with SM and the community controls were not due to SM but rather due to a self-selection bias whereby children from mental health agencies may have lower academic scores, in addition to other difficulties, compared with community controls; (b) it is possible that the strict inclusion criteria for the community controls resulted in the community controls being selected from a non-representative population that had high average academic scores, given that our community controls had substantially higher standardized mathematics scores than the community controls in the study conducted by Cunningham et al., (2004). Fifth, the groups differed significantly in their income levels with the community control group having significantly higher incomes than the SM and mixed anxiety groups. Even though we did not find any significant correlations between income level and academic performance in our data, it is still possible that income level may have influenced the results given that families who have higher income levels may have more resources to put towards the educational development of their children. Sixth, we used only one measure of receptive language ability: the PPVT-III. This test has been shown to produce inflated standardized scores, resulting in children with vocabulary difficulties being missed or misidentified as being low average but within age-appropriate norms. Therefore, our lack of clinical significance between the girls with SM and community controls has two possible interpretations: (1) it may be a true lack of clinical significance; and (2) it may be a reflection of score inflation. Future studies should look at more than one standardized receptive language measure to delineate the clinically significant differences between the groups on receptive vocabulary performance. Finally, given that both the parent and teacher versions of the SSQ have only been used in two other studies to date, there is no test–retest reliability data on the questionnaire, as they have not been utilized in any longitudinal studies. This is a limitation of the study as we relied on the questionnaire for our group classification. Our results suggest that, despite their lack of speaking in the school setting, children with SM are still able to acquire the necessary academic knowledge to achieve average performance for their ages, with regards to reading, mathematics, and spelling. Given that we found no significant differences between children with SM and children with mixed anxiety, and that our results for receptive vocabulary abilities resemble those of children with temperamental shyness, our study provides further support for the idea that SM can be con- ceptualized as sharing commonalities with behaviourally inhibited children. Future studies should take a longitudinal approach to investigate how these children’s academic abilities develop over time and whether they continue to attain the academic levels expected for their age as they get older. Further, language assessment using more than one standardized test would be helpful in further delineating the meaning of the lack of receptive language deficits as measured through the PPVT-III. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

54. 288 M.E. Nowakowski et al. Notes 1. We did conduct an exploratory analysis comparing children with SM only and children with SM who also had comorbid anxiety disorders on the academic measures. The only significant difference for the math subtest of the PIAT-R (t(21) 5 À2.21, p 5 0.04), where children with SM only (M 5 93.33, S.D. 5 8.57) performed significantly lower than children with SM and comorbid anxiety (M 5 110.55, S.D. 5 9.08). This finding was somewhat puzzling as one would expect that comorbidity would lead to greater functional impairment. However, the analyses were performed on a very small sample (n 5 9 for SM only, n 5 14 for SM and anxiety). Thus, a future study with a larger sample size is necessary to further analyze and interpret the differences between children with SM only and children with selective mutsim and comorbid anxiety. 2. To further evaluate this hypothesis, we correlated math scores with children’s age for the SM group and found a significant negative correlation between age and standardized score on the math subtest of the PIAT-R (r(43) 5 À0.44, po0.001). There were no significant correlations between age and math standardized score for the mixed anxiety or community control groups. This provides at least some preliminary evidence that the mathematics performance of children with SM may decrease as children get older. A longitudinal study looking at the mathematics performance of children with SM would be helpful in further understanding this relation. 3. Although we did attempt to parse the children into those who presented only with SM and those who presented with SM and comorbid anxiety, the sample sizes for our comparisons were small, thus not allowing us to make any strong conclusions about the implications of comorbid anxiety disorders in the academic performance of children with SM. ACKNOWLEDGEMENTS This research was supported by a grant from the Ontario Mental Health Foundation (OMHF). The authors would like to thank Lindsay Bennett, Sue McKee, Renee Nossal, and Jamie Sawyer for their assistance with data collection and coding. We would also like to thank many children their parents and teachers for their participation in the study. REFERENCES American Psychiatric Association. (1994). Diagnsotic and statistical manual of mental disorders (4th ed.). Washington, DC: American Psychiatric Association. Anstendig, K. D. (1999). Is selective mutism and anxiety disorder? Rethinking its DSM-IV classification. Journal of Anxiety Disorder, 13, 417–434. Bergman, L. R., Piacentini, J., & McCracken, J. T. (2002). Prevalence and description of selective mutism in a school-based sample. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 938–946. Black, B., & Uhde, T. W. (1995). Psychiatric characteristics of children with selective mutism: A pilot study. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 847–856. Bornstein, M. H., Hahn, C., & Haynes, M. O. (2004). Specific and general language performance across early childhood: Stability and gender considerations. First Language, 24, 267–304. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

55. Language and Academic Abilities in Children 289 Caldera, Y. M., Huston, A. C., & O’Brien, M. (1989). Social interactions and play patterns of parents and toddlers with feminine, masculine, and neutral toys. Child Development, 60, 70–76. Cohan, S. L., Price, J. M., & Stein, M. B. (2006). Suffering in silence: Why a developmental psychopathology perspective on selective mutism is needed. Developmental and Behavioral Pediatrics, 27, 341–355. Cunningham, C. E., McHolm, A., Boyle, M. H., & Patel, S. (2004). Behavioral and emotional adjustment, family functioning, academic performance, and social relationships in children with selective mutism. Journal of Child Psychology and Psychiatry, 45, 1363–1372. Cunningham, C. E., McHolm, A. E., & Boyle, M. H. (2006). Social phobia, anxiety, oppositional behavior, social skills, and self-concept in children with speciﬁc selective mutism, generalized selective mutism, and community controls. European Journal of Child Adolescent Psychiatry, 15, 245–255. Daly, J. A., & Korinek, J. (1980). Interaction in the classroom: An overview. In D. Nimmo (Ed.), Communication yearbook (Vol. 4, pp. 515–532). New Brunswick: Transaction Books. Dummit, S. E., Klein, R. G., Tancer, N. K., Asche, B., Martin, J., & Fairbanks, J. A. (1997). Systematic assessment of 50 children with selective mutism. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 653–660. Dunn, L. M., & Dunn, L. M. (1981). Peabody Picture Vocabulary Test—Revised. Circle Pines, MN: American Guidance Service. Dunn, L. M., & Dunn, L. M. (1997). Peabody Picture Vocabulary Test—Third Edition. MN: American Guidance Service. Dunn, L. M., & Markwardt, F. C. (1998). Peabody individual achievement test—revised. USA: American Guidance Service. Elizur, Y, & Perednik, R. (2003). Prevalence and description of selective mutism in immigrant and native families: A controlled study. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 1451–1459. Evans, M. A., Fox, M., Creamos, L., & McKinnon, L. (2004). Beginning reading: The views of parents and teachers of young children. Journal of Educational Psychology, 96, 130–141. Garcia, A., Freeman, J., Francis, G., Miller, L. M., & Leonard, H. L. (2004). Selective mutism. In T. Ollendick (Ed.), Phobic and anxiety disorders in children and adolescents: A clinician’s guide to effective psychosocial and pharmacological interventions. London: Oxford University Press. Joseph, P. R. (1999). Selective mutism: The child who doesn’t speak at school. Pediatrics, 104, 308–310. Kagan, J., Reznick, J. S., & Snidman, N. (1987). The physiology and psychology of behavioral inhibition in children. Child Development, 58, 1459–1473. Kopp, S., & Gillberg, C. (1997). Selective mutism: A population based study: A research note. Journal of Child Psychology and Psychiatry, 38, 257–262. Kristensen, H. (2000). Selective mutism and comorbidity with developmental disorder/ delay, anxiety disorder, and elimination disorder. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 249–256. Kristensen, H., & Oerbeck, B. (2006). Is selective mutism associated with deﬁcits in memory span and visual memory? An exploratory case-control study. Depression and Anxiety, 23, 71–76. Kumpulainen, K. (2002). Phenomenology and treatment of selective mutism. CNS Drugs, 16, 175–180. Kumpulainen, K., Rasenen, E., Rasska, H., & Somppi, V. (1998). Selective mutism among second graders in elementary school. European Child and Adolescent Psychiatry, 11, 71–78. Manassis, K., Tannock, R., Garland, J. E., Minde, K., McInnes, A., & Clark, S. (2007). The sounds of silence: Language, cognition, and anxiety in selective mutism. Journal of the American Academy of Child and Adolescent Psychiatry, 46, 1187–1195. McInnes, A., Fung, D., Manassis, K., Fiksenbaum, L., & Tannock, R. (2004). Narrative skills in children with selective mutism: An exploratory study. American Journal of Speech- Language Pathology, 13, 304–315. Nattive, A. (1994). Helping behavior and math achievement gain of students using cooperative learning. Elementary School Journal, 94, 285–297. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

56. 290 M.E. Nowakowski et al. Oberklaid, F., Sanson, A., Pedlow, R., & Prior, M. (1993). Predicting preschool behavior problems from temperament and other variables in infancy. Pediatrics, 91, 113–120. Rubin, K. H. (1982). Social and social cognitive characteristics of young isolate, normal and sociable children. In K. H. Rubin & H. S. Ross (Eds.) Peer relationships and social skills in childhood (pp. 353–374). New York: Springer-Verlag. Schwartz, R. H., Freedy, A. S., & Sheridan, M. J. (2006). Selective mutism: Are primary care physicians missing the silence? Clinical Pediatrics, 45, 43–48. Shaffer, D., Fisher, P., Lucas, C. P., Dulcan, M. K., & Schwab-Stone, M. E. (2000). NIMH diagnostic interview schedule for children version IV (NIMH DISC-IV): Description, differences from previous versions, and reliability of some common diagnoses. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 28–38. Sharp, W. G., Sherman, C., & Gross, A. M. (2007). Selective mutism and anxiety: A review of the current conceptualization of the disorder. Journal of Anxiety Disorders, 21, 568–579. Spere, K. A., Schmidt, L. A., Theall-Honey, L. A., & Martin-Chang, S. (2004). Expressive and receptive language skills of temperamentally shy preschoolers. Infant and Child Development, 13, 123–133. Standart, S., & LeCouteur, A. (2003). The quiet child: A literature review of selective mutism. Child & Adolescent Mental Health, 8, 154–160. Steinhausen, H. C., & Juzi, C. (1996). Elective mutism: An analysis of 100 cases. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 606–614. Ukrainetz, T. A., & Duncan, D. S. (2000). From old to new: Examining score increases on the Peabody Picture Vocabulary Test—III. Language, Speech, and Hearing Services in Schools, 31, 336–339. Vecchio, J. L., & Kearney, C. A. (2005). Selective mutism in children: Comparison to youths with and without anxiety disorders. Journal of Psychopathology and Behavioral Assessment, 27, 31–37. William, K. T. (1998). Peabody Picture Vocabulary Test—III: What is new and different? Clinical Connection, 11, 6–8. Yeganeh, R., Beidel, D. C., & Turner, S. M. (2006). Selective mutism: More than social anxiety? Depression and anxiety, 23, 117–123. Yeganeh, R., Beidel, D. C., Turner, S. M., Pina, A. A., & Silverman, W. K. (2003). Clinical distinction between selective mutism and social phobia: An investigation of childhood psychopathology. Journal of the American Academy of Child and Adolescent Psychiatry, 42, 1069–1075. Copyright r 2009 John Wiley & Sons, Ltd. Inf. Child. Dev. 18: 271–290 (2009) DOI: 10.1002/icd

). Language and academic abilities in children with selective mutism

More Related Content

Similar to ). Language and academic abilities in children with selective mutism (20)

Recently uploaded (20)

). Language and academic abilities in children with selective mutism