Language and learning in selective mutism: An exploratory study

Introduction

Selective mutism(SM) is considered a rare condition. It is estimated to be less than 1% of psychiatric clinic populations. Children with SM will speak in some situations (eg at home) but will remain mute in others (eg in school). There are essentially three ways of understanding SM. In the past, SM was considered a result of family psychopathology as well as intrapsychic unresolved conflicts and was treated with psychodynamic and family therapies. Recent interest has been focused on understanding SM as a childhood equivalent of social phobia along the lines of an anxiety disorder. Some authors have also focused on the presence of developmental disorders in children with SM. Many recent studies have concentrated on examining the anxiety component of the etiology of SM. There have been no published reports of systematic review of learning and language abilities in children with SM. Part of this is due to the difficulty and challenges in testing children with SM. We have developed a clinical protocol for assessing the cognitive, academic and language abilities using a combination of non-verbal techniques in mute situations and narrative techniques in non-mute situations.

Objectives

· To become familiar with assessment procedures that can be used in SM,

· To understand the hypothesized relationships between SM and anxiety disorders,

· To become familiar with important research questions in the area of SM, and

· To use a clinical assessment protocol to compare the clinical profile, language and learning abilities of children with SM and a similar group of children with social phobia (SP) but no SM.

Method

Twenty six children who were seen at the Hospital for Sick Children were recruited in our study. 17 had a diagnosis of SM while 9 had a diagnosis of SP. There were altogether 10 boys and 16 girls. The children’s ages ranged from 6-15. The children must be able to understand and speak English. Children from immigrant families with English as a second language were included if they have been shown to use English in the non-mute situations. We have chosen to include children who use English as a second language (ESL) as a significant number of children with SM in Canada come from an immigrant background. All the children had a normal audiometric screen. The assessment procedure consists of a full day assessment at the hospital. Prior to the appointment, the child is provided with a wordless picture book of a frog story and a narrative tape with instructions to listen to the tape while reading the book and then retelling it to the parent. This was followed up with 10 comprehension questions. The retelling was tape recorded and brought along for analysis. A qualified psychiatrist interviewed every child along with at least 1 parent. Diagnosis was made using a semi structured interview using DSM IV criteria. This is supported using a structured clinical diagnostic interview based on DSM IV criteria administered through a computerised program by either parent. (Diagnostic Interview for children and adolescents-revised parent version)(DICA-R-P). Both the child and the parent were asked to complete several rating scales following the interview. The child then completed a cognitive, academic and language assessment by a researcher administered non-verbally. For expressive language testing, the child is asked to read a 2^nd and 3^rd frog story in increasingly fearful situations (with parent and without parent). Finally the child was given a fourth story in which they were expected to write down the retelling and answer the comprehension questions.

Standardised rating scales used include the Conner’s rating scale-revised (CRS-R) to be filled by parents and teacher, the Multidimensional Anxiety Scale for Children (MASC), the Social Anxiety Scale for Children (SASC), Children’s Global Assessment Scale (CGAS). Cognitive assessment used the Weschler Intelligence Scale for Children-III, performance subtests Academic assessment was based on the Wide Range Achievement Test-3 (WRAT-3), the spelling subtest of the Peabody Individual Achievement Test and the operations subtest of the Key Maths. An evaluation of speech and language involve obtaining information from parents, standardised testing and an audiotape of the child’s narrative. Standardised testing included a test of phonemic awareness (The Lindamood Auditory Conceptualisation Test, LACT) and receptive language tests (Peabody Picture Vocabulary Test, PPVT, Clinical Evaluation of Language Fundamentals CELF-3 and, The Test for Auditory Comprehension of Language-Third Edition TACL-3). We used the Strong Narrative Assessment Procedure (SNAP), a standardised narrative elicitation task based on easy to use tape-recorded stimulus stories that accompany wordless picture books. The retelling and responses are transcribed and analysed for fluency, length, syntax, cohesion, story grammar and comprehension.

Results

Data was analysed using the Statistical Package for Social Sciences (SPSS). There was a greater preponderance of girls than boys in the SM group compared to the SP group. The age range of the 2 groups were similar but the SM children were slightly younger (Table 1). Only 1 child had a step parent. In terms of the anxiety measures (Table 2), SM children and their families tend to rate anxiety lower compared to the SP group. Functional ratings were surprisingly similar in the 2 groups although SM children were generally functioning worse in terms of speech in all situations although the differences were not significant. Cognitive and academic measures did not reveal significant differences between SM and SP groups (Table 3). However, there were differences between the SM and SP groups particularly in the language measures compared to the cognitive and academic measures. A small number (3) of SM children had a language disorder. The narrative assessment showed that deficiencies in expressive language were identified despite the child being described as speaking normally.

Table 1: Demographic characteristics

	SM (N=17)	SP (N=9)
Ethnicity Caucasian Asian Mixed Hispanic	11 3 1 1	7 1 1 0
Gender Girls Boys	11 6	5 4
Age Range Mean SD	6-15 9.41 2.60	7-15 11.33 2.50

Table 2: Anxiety and Functional Measures

Instrument	SM		SP
	Mean	SD	Mean	SD
Anxiety ratings
MASC (T score)	50.44	10.49	53.56	12.26
SASC-R (Child)	46.47	19.95	47.22	15.79
SASC-R (Parent)	48.82	14.30	57.33	13.80
Connors (Anxiety) Parent	62.00	11.29	67.11	15.81
Connors (Anxiety) Teacher	57.59	17.16	69.56	15.22
Functional ratings
CGAS (0-100)	64.35	10.48	69.67	7.38
SMQ (Family)	2.10	0.44	2.58	0.28
SMQ (School)	1.25	0.89	1.52	0.49
SMQ (Other )	1.31	0.70	1.35	0.60

Table 3: Cognitive, Academic and Language Measures

Instrument	SM		SP
	Mean	SD	Mean	SD
Cognitive tests
WISC III	105.65	17.33	105.11	17.95
Academic tests
WRAT Spelling	101.94	19.56	112	12.74
WRAT Maths	96.71	20.27	101.33	7.69
PIAT Spelling	105.82	13.86	103.56	10.36
Key Maths	100.55	20.53	95.22	9.71
Language tests
PPVT	88.76	23.12	109.44	18.35
CELF (Concepts and directions subtest)	10.35	2.87	12.56	2.74
TACL Quotient (N=11)	77.55	15.27	87.44	12.29
LACT	105.65	17.33	105.11	17.95

Conclusions

We have demonstrated that selectively mute children can be assessed using standardized tests. We intend to continue this work in characterizing these children. Based on the initial findings, there appears to be no significant differences in anxiety ratings between SM and SP children suggesting that these 2 conditions may share similar origins. SM has been described as a more severe variant of SP but this does not seem to be true as SM children rated lower (but not significantly) in anxiety scales. Instead they may represent alternative coping mechanisms for the same distressing affect, one group (SP) avoids social situations entirely; while the other group(SM) enters them, but silently to reduce the risk of humiliation. Some differences do emerge, especially in the language indices. This may either reflect the lack of practice in language or is an inherent difficulty found in SM children. Our findings need further investigation in a larger sample.