The Survey of Favorite Interests and Activities: Assessing and understanding restricted interests in children with autism spectrum disorder

Abstract

Restricted interests are an established diagnostic symptom of autism spectrum disorder. While there is considerable evidence that these interests have maladaptive consequences, they also provide a range of benefits. This article introduces a new instrument, the Survey of Favorite Interests and Activities, and uses it to examine the nature of restricted interests in autism spectrum disorder. Respondents report substantial benefits of restricted interests as well as areas of difficulty. The Survey of Favorite Interests and Activities assesses Social Flexibility, Perseveration, Respondent Discomfort, Adaptive Coping, and Atypicality. All scales have Cronbach’s α > 0.70. Age and socioeconomic status have little effect on Survey of Favorite Interests and Activities scales; nor does gender with the exception of interest Atypicality. The expected pattern of correlations with existing scales was found. Research and clinical implications are discussed.

Keywords

Asperger’s autism spectrum disorder circumscribed interests restricted and repetitive behavior restricted interests

Autism spectrum disorder (ASD) is a diverse neurodevelopmental condition characterized by two broad categories of symptoms: impaired social communication and restricted, repetitive patterns of behavior, interests, or activities (American Psychiatric Association, 2013). Affecting approximately 1 in 65 American children, ASD is associated with intellectual disability in about one-third of cases (Christensen and al., 2016). However, there is a subset of individuals with ASD who have intact language and cognitive functioning; these people may have been diagnosed with Asperger’s disorder under the previous diagnostic system (American Psychiatric Association, 2000). There is by no means a clear divide between “high-functioning” and “low-functioning” ASD, but there are meaningful average differences in symptom manifestation when comparing different IQ and language groups, such as the classic distinction between so-called “lower order” repetitive behaviors, such as lining up objects, and “higher order” repetitive behaviors, such as engaging in a circumscribed interest (Szatmari et al., 2006; Turner, 1999). Many of the latter behaviors presuppose age-appropriate or advanced cognitive abilities, such as reading and fluent speech, which are less prevalent among people with ASD who have comorbid intellectual disability. As such, it is meaningful to talk about restricted interests in high-functioning autism spectrum disorder (HFASD), with the understanding that this is a flexible category (Volker, 2012).

Restricted interests as an autism symptom are defined as “abnormal in intensity or focus” (APA, 2013). Factor analyses of autism symptoms typically group restricted interests with the other Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5) Category B diagnostic criteria, including other repetitive sensorimotor behaviors and an insistence on sameness (Norris et al., 2012; Shuster et al., 2014). Further analysis has found that that restricted interests are most closely associated with other “higher order” symptoms, such as perseveration, a resistance to change, repetitive speech on a single topic, and reluctance to try other activities (South et al., 2005; Szatmari et al., 2006). Abnormal focus on a restricted interest may interfere with effective social interaction and reduce the child’s willingness to attend to developmentally necessary activities (Anthony et al., 2013; Boyd et al., 2007). Restricted interests may also contribute to a general peer perception that the child with HFASD is “odd,” further decreasing opportunities for socialization. Some research suggests that restricted interest behavior is associated with increased anxiety and maladaptive coping (Spiker et al., 2012), though the direction of causality is unclear. Perseverative, and even aggressive, behaviors may occur if restricted interests are frustrated or interrupted. A small minority of individuals may even break the law in pursuit of their interests (Haskins and Silva, 2006). In rare cases, the restrictive interest itself may be intrinsically maladaptive, such as an interest in an illegal or harmful behavior. To address all of these challenges, intervention programs have targeted restricted interests for management or reduction (Boyd et al., 2012; Lopata et al., 2010; Sofronoff et al., 2004).

However, many adults with HFASD have reported a strong attachment to their restricted interests and have expressed anger or frustration at the thought of therapeutic techniques to diminish them (Bagatell, 2010). Many see a distinction between services meant to help them achieve goals they personally value, and services whose primary goal is to “normalize” them (Parsloe, 2015). They would prefer that their odd—but not harmful—behaviors or interests be accommodated rather than therapeutically altered (Hurlbutt & Chalmers, 2002; Parsloe, 2015). As such, people with HFASD point to the established benefits of restricted interests. They report that when a partner who shares their interest can be located, this is a very effective way to promote social interaction (Muller et al., 2008). Others note that restricted interests have practical uses, including development of an employable skill (Grandin and Duffy, 2004). A qualitative study of six HFASD adults and their families examined the impact of restricted interests on their lives (Mercier et al., 2000). Participants and their family members reported that the interests could become “addictive,” take up too much time or money, or become socially problematic. However, respondents identified several positive elements of the restricted interests, including promoting happiness, providing a constructive hobby, acting as a source of achievement or identity, and developing knowledge that could be applied in the social or occupational realms. Further research found that children with HFASD report using their interests as a source of positive affect and self-confidence (Winter-Messiers, 2007). Indeed, increased intensity of restricted interests has been found to be negatively correlated with depression in children with ASD (Stratis and Lecavalier, 2013). Others have found that adults with HFASD use their interests to cope with anxiety (Trembath et al., 2012). Nonsocial interest-related stimuli have been found to be significantly more motivating to youth with ASD than to typically developing children (Watson et al., 2015). A recent review of the literature found that use of restricted interests to enhance classroom instruction resulted in increased academic task performance and improved social engagement (Gunn and Delafield-Butt, 2016). This was true for both antecedent-based approaches, in which the restricted interest was built into the task or request, and consequence-based approaches, in which access to the preferred interest was contingent on task completion. The authors found potential downsides to interest-based approaches as well, specifically instances of weak generalization and parental discomfort.

Overall, restricted interests can neither be described as exclusively beneficial nor exclusively detrimental. Clinicians need to assess the consequences associated with restricted interests before determining whether and how to intervene. If an interest has meaningful positive and negative aspects, for example, an appropriate treatment plan will acknowledge both. It may be possible to reduce the negative effects without diminishing the interest itself, thereby preserving the positive elements. In some cases, negative outcomes, such as sibling embarrassment, occur outside of the child. By clearly identifying this as the area of concern, the practitioner could then decide whether it is most appropriate to intervene directly with the patient (by reducing or hiding interest behavior) or at the level of the family system (by targeting embarrassed sibling). Unfortunately, current assessment of restricted interests denies researchers and clinicians the opportunity to capture the range of associated consequences. Some assessment tools, such as the Yale Survey of Special Interests, allow the user to gather information on the topic or focus of the child’s interests, but provide only brief evaluation of the interests’ effects (Klin et al., 2007). An evaluation of 24 existing measures of restricted and repetitive behaviors examined their utility as clinical endpoints (Scahill et al., 2015). Of the five most strongly recommended measures, two are primarily for people with intellectual disabilities and one was originally designed for obsessive–compulsive disorder; as such, their coverage of complex restricted interest behavior is minimal. In fact, all five strongly recommended measures have limited coverage of restricted interest behavior and treat it as primarily pathological. For example, the Repetitive Behavior Scale–Revised (RBS-R) asks caregivers to rate whether a behavior “does not occur” is “mild,” “moderate,” or “severe”—there is no option for the caregiver to indicate that the behavior occurs and is harmless or beneficial. Nor is there specificity as to the types of positive or negative outcomes associated with restricted interests. The goal of this project was to develop a framework for understanding the functional impact of restricted interests and an assessment tool that can be used to apply that framework in a clinical or research context.

Methods

This study was approved by the Rochester Institute of Technology Human Subjects Research Office.

Participants

In all, 170 parents of children with HFASD were recruited to participate in this study. All parent raters were required to be literate in English, have a child between the ages of 6 and 17 years with HFASD, and to have not participated in the study previously (even if they have more than one child with HFASD). Participants were recruited with the assistance of the Interactive Autism Network (IAN) Research Database at the Kennedy Krieger Institute, Baltimore, Maryland by messaging the subset of their database whose children were the appropriate age and who had previously reported phrase speech and no intellectual disability—a total of 2615 families, 352 of whom elected to click through to the study (13.5%). Of those 352, 90 elected not to initiate the study, 97 were screened out or discontinued, and 165 completed. Additional participants were recruited via advertisements distributed online and through community agencies. Community-recruited parents were compensated for their time with a USD$20 gift voucher; at the recommendation of IAN, parents recruited through this resource were compensated with a USD$10 gift voucher and a summary table of their child’s scores on the established instruments used in the study. Parents were able to complete surveys at their own pace, taking breaks as needed. However, there was no provision for parents to return to the survey after exiting the browser window. To be eligible for participation, the child must have received a diagnosis of ASD from a psychologist or a physician. The categorization of some children with ASD as “high-functioning” is complicated and can vary considerably across studies. Nonetheless, the nature of this study required the child to have the capacity for complex interest-related behaviors rarely seen in individuals with limited speech or very low IQ. As this study did not assess children directly and relied on parent report, a simple two-part test was used. A child was considered “high-functioning” if he or she was presently able to use fluent speech and exhibited at least one of the following characteristics: (1) had an individually administered IQ test score ⩾85, (2) was at or near grade level in reading, or (3) was at or near grade level in math. Children with ASD who carried concurrent diagnoses of intellectual disability, psychotic disorder, bipolar disorder, or neurological disability were excluded.

Child demographics are listed in Table 1. As expected, the ASD group was majority male. Respondents included 161 mothers (94.7%), 8 fathers (4.7%), and 1 grandmother (0.6%). Socioeconomic status (SES) was assessed by years of parent education, keyed such that a high-school diploma or General Educational Development (GED) was considered 12 years. The ASD sample participants’ specific diagnoses included 48 (28.2%) autistic disorder, 60 (35.3%) Asperger’s disorder, 34 (20.0%) pervasive developmental disorder not otherwise specified (PDD-NOS), and 44 (25.9%) ASD. (Note that many participants had more than one ASD diagnosis.) In all, 94.1% had received special education services and 56.6% had one or more comorbid diagnoses, the most common being attention-deficit disorder (ADD)/attention-deficit hyperactivity disorder (ADHD; 34.1%) and anxious/obsessive–compulsive disorders (22.4%).

Table 1.

Participant demographic characteristics.

	n = 170
Age (years)	11.81 (2.78)
Parent education	15.58 (1.93)
Gender	121 males
Gender	49 females
Race/ethnicity	146 White, non-Hispanic
	1 African American
	12 Hispanic
	2 Asian American
	9 multiracial

Materials

Survey of Favorite Interests and Activities

A preliminary set of 68 items, reflecting a wide range of adaptive and maladaptive outcomes potentially associated with restricted interests, was initially generated and administered to all participants. Potential items were generated based on a review of the literature regarding the impacts of restricted interests as well as informal data, such as statements made by autistic adults regarding their interests, and qualitative parent reports. The goal was to create a broad list which would encompass as many types of effects as possible, even those which affect relatively few children. Parents were first prompted to think about their child’s favorite interests, with the guideline provided that “interests” could be a topic, specific media, hobby, toy, or activity. Several examples were provided to ensure that all respondents understood the breadth of potential interests. Parents were prompted to think of up to three of their child’s strongest interests. This was done to promote consideration of the child’s range of interest-related behavior, as well as to prevent respondents from responding based on a low-ranking or marginal interest. All Survey of Favorite Interests and Activities (SOFIA) items use a 5-point Likert-type scale ranging from 1 (Never; not at all true) to 5 (Very frequently; totally true) (Appendix 1). To assess whether the item content was sufficiently comprehensive, after completing the survey, parents were asked to generate any other positive or negative effects of their child’s interests. These comments were compared to item content by both the author and an independent rater to check for novel themes.

Established scales

Several additional scales were administered concurrently. Two common measures of restricted and repetitive behaviors in ASDs, the Repetitive Behavior Questionnaire (RBQ; Honey et al., 2012) and the Repetitive Behavior Scale–Revised (RBS-R; Lam and Aman, 2007), were used as standard assessments of the frequency and severity of restricted/repetitive behaviors. The RBS-R offers both a total score and six subscale scores, one of which (Restricted Behaviors) is most directly relevant to the construct of restricted interests. All RBQ and RBS-R scores are keyed such that higher scores indicate more severe symptoms. The Social Responsiveness Scale (2nd ed.; SRS-2; Constantino and Gruber, 2012) was used as a general measure of autism symptomatology. SRS-2 scores are given in T-score units with higher scores indicating greater autistic psychopathology. The Behavior Assessment System for Children (2nd ed.; BASC-2; Reynolds and Kamphaus, 2004) Leadership subscale was used as a measure of the child’s ability to pursue meaningful and fulfilling goals, both alone and in groups. The BASC-2 Child form was used for participants aged 6–11 years, while the parallel Adolescent form was used for participants aged 12–17 years. Both forms provide age-adjusted T-scores in which higher scores indicate more adaptive behavior.

Procedures

All participants gave informed consent and established eligibility before completing the survey set listed above. A preliminary factor analysis was completed to establish the five interpretable scales. As the goal was to identify common variance associated with underlying constructs, exploratory factor analysis (EFA) was selected as the most appropriate procedure (Norris and Lecavalier, 2010). Estimates of the number of participants required to obtain a stable solution vary depending on the number of items, number of factors, and the strength of factor loadings. While 200 participants is a commonly cited minimum, considerably fewer cases can be used when factor loadings are strong, as was found in this study (Guadagnoli and Velicer, 1988). The Likert-type scale used by the SOFIA is essentially ordinal in nature, so a polychoric correlation matrix was used (Norris and Lecavalier, 2010). A review of the data found significant skew in individual item response distributions, requiring the use of principle axis analysis for factor extraction (Fabrigar et al., 1999; Norris and Lecavalier, 2010). (Alternate techniques (i.e. maximum likelihood estimation) provide the statistical means of determining the optimum solution in the form of goodness-of-fit indices, but require a level of normality not supported by the data.) Promax rotation was used to maximize distinctions among the factors. Determination of the number of factors was made on the basis of the eigenvalue >1 criterion, visual analysis of the scree plot, and a comparison between obtained eigenvalues and the 95th percentile of a random eigenvalue distribution (Glorfeld, 1995). Once the number of factors was identified, factor loading matrices were generated for the identified number of factors ±2, after which all matrices were examined for interpretability.

Once interpretable scales were established, further statistical analysis examined the intercorrelations among the scales, their reliability, and their relationship with external variables, including demographics. Since this was an exploratory analysis, the decision was made to interpret each statistical test individually, at an alpha level of 0.05. However, to guard against Type I errors, effect sizes (r or partial η²) and exact p values are provided. A total of 55 tests of significance were performed in this article. If the Bonferroni correction were applied, only p values less than approximately 0.001 would be significant.

Results

SOFIA scale structure

An EFA of the SOFIA was run to establish a provisional scale structure for further investigation. Initial analysis using the maximum correlation method found that prior communalities for all 68 original items range from 0.36 to 0.84 (M = 0.60). As noted above, a polychoric item correlation matrix was used via principle axis factoring for initial factor extraction. The scree plot was suggestive of 3–6 factors, while the eigenvalue >1 criterion was met by 12 factors. Parallel analysis identified six factors whose eigenvalues exceeded the 95th percentile of a random eigenvalue distribution; a six-factor solution was also identified by Velicer’s minimum average partial test (O’Connor, 2000). As such, factor loading models including 6 ± 2 factors were created; consideration was given to a three-factor model as well, based on the scree plot. In each model, items which did not load on any factor (all loadings ⩽ 0.30) were identified as were items which loaded onto multiple factors (2 or more loadings >0.30). Models included 1–4 non-loading items, and 12–18 multiple-loading items. Similar factors recurred across solutions. The five-factor model was judged most interpretable, such that each scale assessed a single, meaningful construct and contained at least three items. The five-factor solution accounted for 75.02% of the common variance. These five factors comprised 53 items which made up the final form. Three items were removed for failing to load on any factor and 12 items were removed for demonstrating loadings of a magnitude of 0.30 or greater on multiple scales. Factor loadings from the retained items are listed in Table 2. Factors were named by selecting a title which best represented the included items, putting more weight on items that loaded more strongly. The five factors include Social Flexibility (12 items; capacity for engagement in others’ interests), Perseveration (17 items; rigid and excessive interest behavior), Respondent Discomfort (11 items; parental disapproval of interests or interest-related behaviors), Adaptive Coping (10 items; use of interests to enhance affect or develop skills), and Atypicality (3 items; degree to which the interests are uncommon). The six-factor model was similar to the five-factor model, but attempted to divide the Adaptive Coping Scale into two subelements; this resulted in a sixth subscale that contained only two unique items. Internal consistency of the scales was established via Cronbach’s α coefficients, as listed in Table 3. Means and standard deviations for the HFASD sample are provided as well. All scales are normally distributed, with the exception of the Atypicality scale, which showed strong positive skew.

Table 2.

Item summaries and EFA factor loadings.

	Social flexibility	Perseveration	Respondent discomfort	Adaptive coping	Atypicality
35. Asks about peer activities	89	9	5	–13	–10
20. Asks about peer media	84	–3	9	–6	–7
28. Asks peers questions	82	9	–6	–11	–8
32. Learns about others’ interests	81	–4	–4	–13	10
8. Asks about peer’s interests	75	–8	10	10	0
43. Asks about family interests	75	–14	7	0	9
53. Attentive to others’ interests	73	–8	–5	–4	6
12. Engages in others’ interests	67	–15	8	–2	–2
17. Stops own activities to join others	61	–14	0	2	12
34. Does what peers choose	60	–4	–1	–12	–30
3. Aware of peer interests	54	–12	13	3	–17
52. Peers want to join child’s interest	49	11	–9	24	–19
5. Irrelevant interest speech	–7	67	–14	–11	0
25. Repetitive interest speech	–5	67	–1	–11	5
45. Demanding	0	65	8	13	8
48. Defiant or immature	–11	61	12	–2	–8
39. Assumes others share interest	15	60	0	1	4
26. Interest increases persistence	–4	59	–3	23	–23
37. Ignores safety rules	6	59	20	–14	8
22. Frustration	4	58	–8	–11	21
51. Driven instead of playful	–1	58	8	11	16
19. Interest speech is relevant	11	–57	7	16	12
33. Interest increases frustration tolerance	6	54	–7	28	–10
47. Upset if prevented from interest	–22	52	11	14	–14
24. Inattentive to other tasks	–12	49	18	–6	–5
6. Stopping causes distress	–18	47	20	1	–5
18. Treats opinions as facts	–13	38	21	27	14
27. Switches from interest activity	27	–36	–17	23	9
2. Resists other activities	–17	34	–8	–8	2
41. Respondent embarrassed by interests	–5	1	77	3	13
50. Resp. discourages interests	5	7	76	–18	–1
4. Resp. discourages public displays of interests	16	–4	72	12	8
23. Resp. discourages public speech	1	10	71	7	9
36. Interests cause resp. to worry	–8	11	71	0	6
7. Child must leave interests behind	13	–3	68	–16	9
46. Interest are unhealthy	–4	13	66	–19	–4
44. Cannot be successful with interests	3	2	64	–29	11
30. Interests not shared with parent	–1	–2	60	–1	–3
15. Interests lead to negative people or situations	13	26	47	–1	11
11. Resp. encourages alternate interests	0	–6	43	–10	–7
10. Genuinely happy	–15	–11	–14	72	–12
21. Major source of happiness	–27	–4	–9	66	–16
16. Child shows best qualities	2	–7	–28	65	11
1. Child works hard	–5	–2	5	62	3
49. Interests reduce stress	–22	11	0	60	–1
9. Creative in interests	8	–25	0	52	14
14. Possible career	11	–5	–17	52	10
29. Interest talk is clear	25	–14	13	45	12
42. Imagination	13	26	–29	41	6
40. Computer skill	11	27	4	39	–5
13. Unusual for sex	–6	6	18	7	66
38. Unusual for age	–15	14	21	15	60
31. More common for opposite sex	8	–6	14	7	57

Decimal points are omitted. All cells with values of magnitude ⩾0.30 are shaded.

Table 3.

Scale interpretations, internal consistency coefficients, and descriptive statistics.

Scale—items	High scores indicate	Cronbach’s α	Total mean (SD)	Mean item score	Range
Social Flexibility—12	Child is able and willing to engage in others’ interests	0.913	27.36 (8.95)	2.28	12–57
Perseveration—17	Child has rigid and excessive interest–related behavior	0.806	59.14 (12.09)	3.48	17–84
Respondent Discomfort—11	Parent disapproves of interests	0.865	23.14 (7.52)	2.10	12–44
Adaptive Coping—10	Interests produce positive affect or useful skills	0.737	39.44 (5.57)	3.94	22–50
Atypicality—3	Interests are unusual	0.736	5.95 (3.03)	1.98	3–15

SD: standard deviation.

Items are rated on a scale from 1 (never; not at all true) to 5 (very frequently; totally true).

Intercorrelations among the five scales range from negligible to moderate. Unsurprisingly, they do not form a single overall composite, as there is no reason to believe that the diverse impacts of restricted interests reflect a unitary underlying construct. The strongest relationship is between Social Flexibility and Perseveration, indicating that these are related, though distinct, constructs. Respondent Discomfort is more likely if the child is high on Perseveration and/or Atypicality. Of note, Adaptive Coping is relatively unrelated to all other scales with the exception of Respondent Discomfort, indicating that the beneficial elements of restricted interests seem to occur largely independently of negative aspects.

Relationships between SOFIA scores and autism severity

The Social Flexibility, Perseveration, and Atypicality scales appear directly related to established features of restricted interests in HFASDs, and the data bear this out. Correlations between these scales and the SRS-2 fall in the moderate range and in the expected direction (r_SF×SRS-2 (170) = –0.551, p < 0.001; r_PERS×SRS-2 (170) = 0.652, p < 0.001; r_ATYP×SRS-2 (170) = 0.436, p < 0.001). Respondent Discomfort with child interests occurs in many situations that are unrelated to ASD, such as child interest in media the parent finds inappropriate, but appears to be mildly related to HFASD severity (r_RD×SRS-2 (170) = 0.220, p = 0.004). Adaptive Coping is unrelated to degree of ASD symptoms (r_AC×SRS-2 (170) = –0.011, p = 0.891). In other words, meaningful benefits associated with restricted interests are as likely to occur in youth with severe HFASD as in those with milder conditions.

Demographic effects

In this sample, no significant correlation was found between child age and severity of autism as measured by the SRS-2 (r(170) = –0.119, p = 0.123), nor between SES and SRS-2 (r(170) = –0.066, p = 0.394). As such, zero-order correlations were run between age/SES and the five SOFIA scales. All findings were nonsignificant with the exception of a weak, negative correlation between age and Perseveration (r(170) = –0.171, p = 0.026). Girls in this sample scored somewhat higher than boys on the SRS-2 (t(169) = 4.191, p < 0.001, d = 0.72), so autism severity was retained as a covariate when assessing gender effects on SOFIA scales. Gender exerted a nonsignificant effect on four out of the five SOFIA scales, but girls scored significantly higher than boys on the Atypicality scale (F(1,167) = 26.917, p < 0.001, partial η² = 0.139), after controlling for autism severity. Finally, the impact of race was assessed. Given the sample size, race was dichotomized to White, non-Hispanic (n = 146) and non-White (n = 24). The race variable was not significantly related to autism severity and so SRS-2 was not retained as a covariate. There were no significant racial differences on any of the five SOFIA scales.

Validity

After completing the SOFIA, parents were to write down any additional positive or negative impacts of their child’s restricted interests. The vast majority of comments addressed outcomes already measured by the SOFIA or that were off-topic. When asked about positive impacts, 17 parents gave no response or denied any positive effects, 143 gave specific examples of constructs addressed by the SOFIA, and 10 gave novel responses. When asked for further negative impacts, 29 gave no response or denied negative effects, 134 gave specific examples of constructs addressed by the SOFIA, and 7 gave novel responses. Broadly, this suggests that the SOFIA addresses the great majority of important ways in which restricted interests affects functioning. Looking at the novel responses, 14 (9 positive and 5 negative) referred to motor skills, physical fitness, or time spent sedentary. This suggests that this construct is an important outcome of restricted interests absent from the SOFIA. The remaining novel comments were eclectic: sleep disruption (negative), learning morality (positive), and lying about age (negative).

Existing scales to assess restricted interests focus on quantifying their frequency and overall maladaptive nature, with little specification of the types of problems incurred nor any assessment of their beneficial effects. As such, they are expected to correlate moderately with the Social Flexibility, Perseveration, and Atypicality subscales, as these reflect the features traditionally associated with restricted interests in HFASDs. No correlation is expected with Adaptive Coping, because positive elements are not addressed in existing scales. Respondent Discomfort is a more complicated scale to interpret. Intercorrelations, as noted in Table 3, show that parents feel more negatively about atypical interests, those associated with perseverative behavior, and those that generate few positive effects. However, these correlations are small and account for relatively little of the variance. As such, parents’ disapproval is likely to be no more than weakly related to standard restricted interest scales.

All five SOFIA scales were examined for correlation with the RBQ and RBS-R. The strongest correlations were found with the Perseveration scale (r_PERS×RBQ(170) = 0.563, p < 0.001; r_PERS×RBS-R(170) = 0.590, p < 0.001), which directly assesses repetitive and inflexible interest behaviors. Weak-to-moderate negative correlations were found with Social Flexibility (r_SF×RBQ(170) = –0.372, p < 0.001; r_SF×RBS-R(170) = –0.392, p < 0.001) and positive correlations with Atypicality (r_ATYP×RBQ(170) = 0.384, p < 0.001; r_ATYP×RBS-R(170) = 0.356, p < 0.001), showing that these elements of restricted interests are only somewhat accounted for by existing measures. Respondent Discomfort was unrelated to the RBQ (r_RD×RBQ(170) = 0.128, p = 0.128) and very weakly related to the RBS-R (r_RD×RBS-R(170) = 0.182, p = 0.018). The existing scales showed no relationship with Adaptive Coping (r_AC×RBQ(170) = 0.042, p = 0.585; r_AC×RBS-R(170) = –0.095, p = 0.219). The RBS-R includes a Restricted subscale which focuses primarily on restricted interests. The correlations were repeated using only the RBS-R Restricted subscale, and the same pattern was found (Table 4).

Table 4.

Scale intercorrelations.

Scale		Perseveration	Respondent Discomfort	Adaptive Coping	Atypicality
Social Flexibility	r	−0.429***	−0.129	0.097	−0.264***
Social Flexibility	p	<0.001	0.094	0.210	<0.001
Perseveration	r		0.348***	0.066	0.279***
Perseveration	p		<0.001	0.393	<0.001
Respondent Discomfort	r			−0.316***	0.229**
Respondent Discomfort	p			<0.001	0.003
Adaptive Coping	r				0.060
Adaptive Coping	p				0.435

p < 0.01, ***p < 0.001.

The Adaptive Coping subscale has been largely unrelated to the covariates discussed thus far, as it does not assess the problematic or symptomatic elements of restricted interests. The BASC-2 Leadership subscale was felt to assess a meaningfully similar, though not identical, construct. As seen in Table 5, the Leadership scale correlated moderately with both Social Flexibility (r_SF×LEAD(170) = 0.509, p < 0.001) and Adaptive Coping (r_AC×LEAD(170) = 0.400, p < 0.001).

Table 5.

Relationship between SOFIA scores, autism severity, demographics, and existing scales.

		Social Flexibility	Perseveration	Respondent Discomfort	Adaptive Coping	Atypicality
SRS-2	r	−0.551***	0.652***	0.220**	−0.011	0.436***
SRS-2	p value	<0.001	<0.001	0.004	0.891	<0.001
Age	r	0.070	−0.171	0.146	0.130	−0.057
Age	p value	0.914	0.026*	0.057	0.091	0.458
SES	r	0.008	0.001	−0.115	0.072	0.077
SES	p value	0.914	0.993	0.136	0.350	0.321
Gender, controlling for SRS-2	F	0.310	0.598	0.209	1.533	26.917
	p value	0.579	0.440	0.648	0.217	<0.001***
	Partial η²	0.002	0.004	0.001	0.009	0.139
Race	F	0.001	1.556	0.079	0.728	0.400
	p value	0.976	0.214	0.779	0.395	0.528
	Partial η²	0.000	0.009	0.000	0.004	0.002
RBQ	r	−0.372***	0.563***	0.128	0.042	0.384***
RBQ	p value	<0.001	<0.001	0.095	0.585	<0.001
RBS-R Total	r	−0.392***	0.590***	0.182*	0.095	0.356***
RBS-R Total	p value	<0.001	<0.001	0.018	0.219	<0.001
RBS-R Restricted	r	−0.348***	0.519***	0.253**	−0.043	0.296***
RBS-R Restricted	p value	<0.001	<0.001	0.001	0.578	<0.001
BASC-2 Leadership	r	0.509***	−0.278***	−0.184*	0.400***	−0.076
BASC-2 Leadership	p value	<0.001	<0.001	0.016	<0.001	0.327

SOFIA: Survey of Favorite Interests and Activities; SRS-2: Social Responsiveness Scales, 2nd edition; SES: socioeconomic status; RBQ: Repetitive Behavior Questionnaire; RBS-R: Repetitive Behavior Scales–Revised; BASC-2: Behavior Assessment System for Children.

p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

This study examined the features of restricted interests in a sample of 170 school-age children with HFASD, using the novel instrument SOFIA. The five-factor, analytically derived scales yielded intercorrelations ranging from −0.429 to 0.348, with several nonsignificant values. As such, the SOFIA does not provide a meaningful overall composite score. Instead, its five scales function relatively independently. The factor analysis presented was intended to provide a framework for examining the patterns and relationships in the data, but is not the only meaningful way to categorize the items. To aid in examining alternate approaches, means and standard deviations are provided in Table 6 for all items, including those that were removed from this article’s analysis. It is notable that 6 of the 7 highest rated items—those with a mean between 4.00 and 5.00—are about positive impacts of restricted interests: happiness, emotional coping, and skill development. This emphasizes the adaptive role that restricted interests play in the lives of youth with ASD, even if they may be simultaneously diagnostically significant or associated with negative impacts.

Table 6.

Item means and standard deviations.

	Mean^a	SD
1. When it comes to her or his interests, my child is more willing to work hard and stick with it than usual.	4.43	0.83
2. My child resists engaging in activities outside of her or his interests.	3.76	1.06
3. My child is aware of peers’ interests.	2.95	1.05
4. I discourage my child from displaying his or her interests in public (e.g. by talking about the interest at school, by wearing clothes that advertise the interest, etc.).	1.68	0.91
5. My child brings up her or his interests when they are not related to the topic others are discussing.	3.91	1.18
6. When asked to stop or change activities related to these interests, my child becomes very upset, anxious, or stressed.	3.59	1.21
7. I know that to be happy, my child will have to leave these interests behind.	1.69	0.93
8. My child asks other kids questions about their interests.	2.28	1.04
9. My child engages in her or his interests in creative ways, rather than copying exactly from an instruction manual, book, TV show, or movie.	3.48	1.19
10. These interests seem to make my child genuinely happy.	4.72	0.54
11. I try to encourage my child to develop different interests.	3.83	1.00
12. My child offers to engage in someone else’s interest (that my child doesn’t share).	1.99	0.98
13. People think my child’s interests are unusual for her or his sex.	1.84	1.27
14. I can imagine these interests leading to a career for my child someday.	3.29	1.36
15. My child’s interests lead her or him to negative people or situations.	1.78	0.97
16. My child shows some of her or his best qualities when pursuing these interests.	3.86	0.98
17. My child voluntarily stops engaging in her or his interests to spend time with other people.	2.04	0.96
18. My child rigidly defends opinions about these interests as if her or his opinions were facts.	3.64	1.31
19. My child shares information about her or his interests only when it is relevant.	2.26	1.03
20. My child asks other kids what kinds of books, movies, TV shows, etc. they like.	2.30	1.17
21. On the whole, my child’s interests are a major source of happiness in her or his life.	4.61	0.55
22. My child’s interests make her or him frustrated (for example, when pieces are missing, when other people don’t understand, etc.).	3.09	1.36
23. I discourage my child from talking about his or her interests in public.	1.65	0.84
24. My child is so focused on these interests that he or she doesn’t pay attention to school or other important tasks.	2.99	1.30
25. My child shares the same information about her or his interests over and over.	3.54	1.34
26. My child is more willing to attempt an unpleasant task if it is tied to her or his interests.	3.84	1.15
27. Without direction from an adult, my child decides to switch from her or his interests to another activity.	2.65	1.14
28. My child asks follow-up questions when peers talk about their interests.	2.25	1.11
29. When my child talks about these interests, he or she gives enough details or background information so the listener understands.	3.06	1.24
30. I wish my child and I had more of the same interests in common.	2.94	1.29
31. My child’s interests are far more common among people of the opposite sex (i.e. my son’s interests are more common among girls or my daughter’s interests are more common among boys).	1.78	1.17
32. My child seeks information about someone else’s interest that my child doesn’t share.	1.76	0.91
33. My child is more willing to face a difficult or frustrating task if it is tied to her or his interests.	3.85	1.16
34. My child plays with agemates, doing activities the other children choose.	2.54	1.09
35. My child asks peers what kinds of toys, games, or activities they like.	2.36	1.09
36. Sometimes, my child’s interests worry me.	2.46	1.32
37. My child’s focus on these interests is so extreme, it causes her or him to fail to follow safety rules (e.g. does not react to a fire alarm when playing chess, walks into street without looking when playing a handheld video game).	2.21	1.28
38. People think my child’s interests are unusual for her or his age.	2.32	1.31
39. My child assumes that other people share her or his interests (e.g. assumes that others will like the same movies, gifts).	3.84	1.13
40. These interests have led my child to develop skills with computers or electronics.	4.09	1.02
41. I’m embarrassed by my child’s interests.	1.44	0.79
42. These interests have led my child to spend more time being imaginative or creative than he or she would have otherwise.	3.61	1.28
43. My child asks about the interests and activities of family members.	2.22	1.09
44. I know that to be successful, my child will have to leave these interests behind.	1.92	1.17
45. When it comes to her or his interests, my child makes many demands.	2.81	1.38
46. I think my child’s interests are unhealthy.	1.84	1.05
47. Preventing my child from engaging in her or his interests causes her or him to be very upset.	4.05	1.09
48. My child’s behavior becomes unusually defiant or immature when denied the opportunity to engage in her or his interests (i.e. more than is typical for my child).	3.55	1.35
49. My child can engage in her or his interests to reduce or manage stress.	4.28	0.84
50. I try to discourage some of my child’s current interests.	1.90	1.10
51. When my child gets engaged in her or his interests, he or she seems driven instead of playful, like there is a goal that must be met.	3.41	1.38
52. My child can describe or engage in her or his interests in a way that makes other children want to join in.	2.65	1.06
53. My child appears attentive when peers are talking about their interests (that my child does not share).	2.02	0.93
Items removed during factor analysis
54. My child correctly identifies a gift that a peer or family member would like.	2.98	1.16
55. My child blends her or his interests with those of other children.	2.82	1.16
56. Other children tease or exclude my child because of her or his interests.	2.25	1.15
57. These interests have led my child to develop a practical skill, like cooking or auto repair.	2.91	1.36
58. Other adults have criticized my child’s interests.	2.17	1.23
59. These interests have caused my child to practice math or science skills much more than he or she would have otherwise.	2.56	1.41
60. My child interacts with other people on the internet to engage in his or her interests	2.58	1.55
61. My child is aware of who does and does not share her or his interests.	3.04	1.21
62. My child has developed friendships because of her or his interests.	3.06	1.31
63. Engaging in these interests helps my child relax when he or she is worried or upset.	4.53	0.71
64. My child engages in these interests alone, without friends or family.	3.94	1.02
65. When my child engages in these interests with other people, he or she enjoys it more than engaging in the interest alone.	3.26	1.21
66. My child engages in these interests by interacting with same-age peers.	2.86	1.23
67. These interests have caused my child to practice reading or writing much more than he or she would have otherwise.	3.21	1.45
68. My child becomes angry, defensive, or offended when another person does not share or dislikes these interests.	2.28	1.28

Raw (unreversed) means are given for all items. Items were rated on a 5-point Likert scale from 1 (never, not at all true) to 5 (very frequently, totally true).

As expected, a cluster of items—the Perseveration scale—emerged, reflecting the maladaptive features that have been previously discussed in the literature (e.g. excessive time spent on the interest and repetitive interest behaviors). The mean item score of 3.48 is toward the “agree” end of the scale, indicating that parents rated their children as having maladaptive perseverative behaviors relating to their restricted interests. The Social Flexibility scale comprised a related, but distinct cluster of items which focused on the child’s ability and willingness to engage in someone else’s interest. These items combine a capacity to shift focus away from one’s own interests with a readiness for some social engagement. The mean Social Flexibility scale score is 2.28, toward the “disagree” end of the scale, indicating that children with ASD tended to lack in Social Flexibility. Perseveration and Social Flexibility collectively assess restricted interest-related impairment, although they are keyed in opposite directions. It is notable that these two factors are separate and only moderately correlated. A child may struggle with one, but not the other. Similarly, an intervention targeted at one cannot be assumed to improve the other.

Positive elements of restricted interests formed the Adaptive Coping factor. Many parents reported that their children derived positive affect, were spontaneous and creative, and developed useful skills while engaging in their restricted interests. This scale was significantly correlated with the BASC-2 Leadership subscale, which assesses “skills associated with accomplishing academic, social, or community goals, including the ability to work with others” (Reynolds and Kamphaus, 2004: 60). This factor is orthogonal to the Perseveration and Social Flexibility scales, that is, maladaptive interest behaviors are neither directly nor inversely predictive of beneficial elements. Clearly, if an interest has few negative aspects (high Perseveration, low SF) and many positive ones (high AC), there is no need for intervention. If the inverse is true, intervention is clearly merited. However, when an interest has both negative and positive aspects, clinicians are well served by examining the full breadth of intervention options and ensuring that the positive aspects are retained or at least replaced. For example, perseverative speech on a restricted interest topic can be reduced without requiring the child to relinquish the interest as a whole (Fisher et al., 2013).

The remaining two factors assess the context of the interest, rather than the child’s behavior itself. The Respondent Discomfort scale assesses the degree to which the parent reports negative attitudes about the child’s interest. This does not, in and of itself, represent a symptom of restricted interests or of ASD in general. Instead, this factor is worth assessing because parental disapproval of children’s interests may be the primary motivating factor for clinical intervention. Only 12.2% of cases responded with a mean item score of 3.0 (neutral) or higher on this subscale, indicating that few parents disapprove of their child’s interests. This scale touches on the concern of some adults with ASD that restricted interests may be arbitrarily “normalized” based on adult preference rather than child well-being (Parsloe, 2015). In this vein, some cases of elevated Respondent Discomfort may be due to parents’ inflexibility or restrictive expectations. However, Respondent Discomfort is positively related to Perseveration and negatively related to Adaptive Coping. Thus, in many cases, high Respondent Discomfort may reflect the parent’s realistic view that the interest is unhelpful in the child’s day-to-day life. This scale is also correlated with Atypicality, suggesting that parents may disapprove of interests because they are uncommon for the child’s age or sex. This may be because the parent is personally uncomfortable with the child’s unusual interests or because the parent fears that others will react negatively. Again, clinicians may help parents determine whether their fears are realistic and whether the costs of intervening outweigh the benefits in a child’s particular case. These issues are further complicated as Respondent Discomfort may elevate in response to an interaction between HFASD symptoms and typical childhood challenges. For example, parents often restrict children’s access to media with offensive language; parents may be even more uncomfortable with such media if the child with HFASD exhibits frequent delayed echolalia. Furthermore, to understand this scale, it is important to recognize that parental disapproval of restricted interests need not be tied to ASD. Common reasons for parents to have a negative attitude toward any child’s interests, unrelated to ASD, might include conflict with the parent’s values, perceived inappropriate content for the child’s age, lack of overlap with parent interests, or exposing the child to a negative peer group. Thus, clinicians should not assume that parental concerns about the interests of a child with ASD are inherently ASD-related. Just as when working with parents of typically developing children, clinicians should help parents of children with ASD determine whether their objections are reasonable and developmentally appropriate.

The final scale—Atypicality—contained only three items and reflects the degree to which the interests themselves are unusual. This scale showed strong positive skew, with a modal score of 3, the minimum possible score, occurring in 26.5% of cases. Only 13.5% of cases received scores greater than 9, meaning the per-item average was at least somewhat toward “agree” on a 5-point scale. Ergo, most children’s interests were not reported by their parents to be abnormal in focus. This is consistent with prior research showing that although bizarre interests are particularly memorable, they do not comprise the majority of restricted interests in HFASD (Winter-Messiers, 2007). When interests are rated as abnormal in focus, they are mildly predictive of negative interest effects (lower Social Flexibility and higher Perseveration) as well as elevations in Respondent Discomfort and overall autism symptomatology. Girls received higher average scores than boys on this scale, even when controlling for overall autism severity. Prior findings on this topic are mixed. School-age girls with ASD have been found less likely than their male peers to meet the restricted interest diagnostic criterion, but more likely to have restricted interests categorized as “random” (Hiller et al., 2014). Other studies have found no significant gender difference in restricted interests across this age group (Joseph et al., 2013; Rivet and Matson, 2011). A likely reason for these different findings is the variation in exactly what is being measured and how. When measured by the SOFIA, the actual atypicality of girls’ interests is not assessed, but rather the parent’s sense of how unusual the interest is. Thus, the difference may lie not in the girls’ actual interests but in the parents’ expectations for girls’ typicality. In addition, it should be noted that 2 of the 3 items on this scale refer to the degree to which interests are gender-atypical. Accordingly, this finding may indicate that girls with ASD have interests that are less traditionally feminine, but not necessarily odd in a general sense.

An examination of the SOFIA’s content validity asked parents to provide any consequences of restricted interests not already described by the SOFIA. Out of 340 potential responses, only 17 reflected novel elements, indicating that the SOFIA has addressed the major outcomes associated with restricted interests. However, 14 of the 17 novel comments fell into a gross motor theme (developing coordination, overly sedentary, etc.), indicating that future revisions of the scale should add one or more items reflecting this concept.

A major strength of this study is the focus on the effects of restricted interests, rather than simply their presence or absence, investigated across a moderately large (n = 170) sample. However, there are limitations that must be acknowledged and which will hopefully guide future research. As noted above, a total of 55 tests were performed, raising concerns about the Type I error rate. However, most findings which were significant at the 0.05 level were also significant at the Bonferroni-corrected 0.001 level. Those that were not may be considered provisional and are especially important targets for future replication. Additionally, the sample was primarily Caucasian (86%). Further research should incorporate more diverse participants to determine whether these findings replicate across racial and ethnic groups.

Individuals with HFASDs commonly have restricted interests, as noted in the official diagnostic criteria (APA, 2013), but unlike many symptoms, restricted interests are not inherently maladaptive. They can certainly be associated with a broad range of restricted and repetitive behaviors which are well-assessed by existing instruments (Scahill et al., 2015); these instruments play an important role in diagnosing ASD and in quantifying the overall burden associated with repetitive behaviors. However, they do little to specify exactly why a restricted interest appears to the rater to be problematic. For example, this article shows that Respondent Discomfort is only mildly correlated with traditional autistic perseverative difficulties; thus, an adult may report a restricted interest to be a problem for reasons that are largely unrelated to child’s autism. This is not to say that such a complaint would necessarily be invalid; instead, that researchers and clinicians will have a more accurate view of the nature and scope of the child’s issues if they identify the specific of concerns associated with restricted interests, rather than a general sense of frequency or severity. Restricted interests are also associated with adaptive benefits. At the time of this writing, there is no other quantitative instrument to assess these. As such, there is no way for researchers to assess whether programs targeted to ameliorate the negative effects of restricted interests have iatrogenically reduced the positive effects as well. The SOFIA Adaptive Coping scale by no means encompasses all of the possible benefits associated with restricted interests, but by capturing some, it provides a way to begin quantifying, examining, and discussing the complex range of outcomes associated with restricted interests.

Supplemental Material

AUT742140_Lay_Abstract – Supplemental material for The Survey of Favorite Interests and Activities: Assessing and understanding restricted interests in children with autism spectrum disorder

Supplemental material, AUT742140_Lay_Abstract for The Survey of Favorite Interests and Activities: Assessing and understanding restricted interests in children with autism spectrum disorder by A Smerbeck in Autism

Footnotes

Appendix 1 Funding

The author(s) received an internal grant from the Rochester Institute of Technology to support the research presented in this article.

References

American Psychiatric Association. (2000) Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

American Psychiatric Association. (2013) Diagnostic and statistical manual of mental disorders (5th ed.). Washington, DC: Author.

Anthony

Kenworthy

Yerys

et al . (2013) Interests in high-functioning autism are more intense, interfering, and idiosyncratic than those in neurotypical development. Development and Psychopathology 25: 643–652.

Bagatell

(2010) From cure to community: transforming notions of autism. Ethos 38: 33–55.

Boyd

Conroy

Mancil

et al . (2007) Effects of circumscribed interests on the social behaviors of children with autism spectrum disorders. Journal of Autism and Developmental Disorders 37: 1550–1561.

Boyd

McDonough

Bodfish

(2012) Evidence-based behavioral interventions for repetitive behaviors in autism. Journal of Autism and Developmental Disorders 42: 1236–1248.

Christensen

Baio

Van Naarden Braun

et al . (2016) Prevalence and characteristics of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 sites, United States, 2012. MMWR. Surveillance Summaries 65: 1–23.

Constantino

Gruber

(2012) Social Responsiveness Scale, Second Edition (SRS-2). Torrance, CA: Western Psychological Services.

Fabrigar

Wegener

MacCallum

et al . (1999) Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods 4: 272–299.

10.

Fisher

Rodriguez

Owen

(2013) Functional assessment and treatment of perseverative speech about restricted topics in an adolescent with Asperger syndrome. Journal of Applied Behavior Analysis 46: 307–311.

11.

Glorfeld

(1995) An improvement on Horn’s parallel analysis methodology for selecting the correct number of factors to retain. Educational and Psychological Measurement 55: 377–393.

12.

Grandin

Duffy

(2004) Developing Talents: Careers for Individuals with Asperger Syndrome and High-Functioning Autism. Shawnee Mission, KS: Autism Asperger Publishing Company.

13.

Guadagnoli

Velicer

(1988) Relation of sample size to the stability of component patterns. Psychological Bulletin 103: 265–275.

14.

Gunn

KCM

Delafield-Butt

(2016) Teaching children with autism spectrum disorder with restricted interests: a review of evidence for best practice. Review of Educational Research 86: 408–430.

15.

Haskins

Silva

(2006) Asperger’s disorder and criminal behavior: forensic-psychiatric considerations. Journal of the American Academy of Psychiatry and the Law 34: 374–378.

16.

Hiller

Young

Weber

(2014) Sex differences in autism spectrum disorder based on DSM-5 criteria: evidence from clinician and teacher reporting. Journal of Abnormal Child Psychology 42: 1381–1393.

17.

Honey

McConachie

Turner

et al . (2012) Validation of the repetitive behaviour questionnaire for use with children with autism spectrum disorder. Research in Autism Spectrum Disorders 6: 355–364.

18.

Hurlbutt

Chalmers

(2002) Adults with autism speak out: perceptions of their life experiences. Focus on Autism and Other Developmental Disabilities 17: 103–111.

19.

Joseph

Thurm

Farmer

et al . (2013) Repetitive behavior and restricted interests in young children with autism: comparisons with controls and stability over 2 years. Autism Research 6: 584–595.

20.

Klin

Danovich

Merz

et al . (2007) Circumscribed interests in higher functioning individuals with autism spectrum disorders: an exploratory study. Research and Practice for Persons with Severe Disabilities 32: 89–100.

21.

Lam

KSL

Aman

(2007) The Repetitive Behavior Scale–Revised: independent validation in individuals with Autism spectrum disorders. Journal of Autism and Developmental Disorders 37: 855–866.

22.

Lopata

Thomeer

Volker

et al . (2010) RCT of a manualized social treatment for high-functioning autism spectrum disorders. Journal of Autism and Development Disorders 40: 1297–1310.

23.

Mercier

Mottron

Belleville

(2000) A psychosocial study on restricted interests in high-functioning persons with pervasive developmental disorders. Autism 4: 406–425.

24.

Muller

Schuler

Yates

(2008) Social challenges and supports from the perspective of individuals with Asperger syndrome and other autism spectrum disabilities. Autism 12: 173–190.

25.

Norris

Lecavalier

(2010) Evaluating the use of exploratory factor analysis in developmental disability psychological research. Journal of Autism and Development Disorders 40: 8–20.

26.

Norris

Lecavalier

Edwards

(2012) The structure of autism symptoms as measured by the autism diagnostic observation schedule. Journal of Autism and Development Disorders 42: 1075–1086.

27.

O’Connor

(2000) SPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test. Behavior Research Methods, Instruments, & Computers 32: 396–402.

28.

Parsloe

(2015) Discourses of disability, narratives of community: reclaiming an autistic identity online. Journal of Applied Communication Research 43: 336–356.

29.

Reynolds

Kamphaus

(2004) Behavioral Assessment System for Children. 2nd ed. Circle Pines, MN: American Guidance Service.

30.

Rivet

Matson

(2011) Gender differences in core symptomatology in Autism spectrum disorders across the lifespan. Journal of Developmental and Physical Disabilities 23: 399–420.

31.

Scahill

Aman

Lecavalier

et al . (2015) Measuring repetitive behaviors as a treatment endpoint in youth with autism spectrum disorder. Autism 19: 38–52.

32.

Shuster

Perry

Bebko

et al . (2014) Review of factor analytic studies examining symptoms of autism spectrum disorders. Journal of Autism and Developmental Disorders 44: 90–110.

33.

Sofronoff

Leslie

Brown

(2004) Parent management training and Asperger syndrome: a randomized controlled trial to evaluate a parent based intervention. Autism 8: 301–317.

34.

South

Ozonoff

McMahon

(2005) Repetitive behavior profiles in Asperger syndrome and high-functioning autism. Journal of Autism and Developmental Disorders 35: 145–158.

35.

Spiker

Lin

Van Dyke

et al . (2012) Restricted interests and anxiety in children with autism. Autism 16: 306–320.

36.

Stratis

Lecavalier

(2013) Restricted and repetitive behaviors and psychiatric symptoms in youth with autism spectrum disorders. Research in Autism Spectrum Disorders 7: 757–766.

37.

Szatmari

Georgiades

Bryson

et al . (2006) Investigating the structure of the restricted, repetitive behaviours and interests domain of autism. Journal of Child Psychology and Psychiatry 47: 582–590.

38.

Trembath

Germano

Johanson

et al . (2012) The experience of anxiety in young adults with autism spectrum disorders. Focus on Autism and Other Developmental Disabilities 27: 213–224.

39.

Turner

(1999) Annotation: repetitive behaviour in autism: a review of psychological research. Journal of Child Psychology and Psychiatry 40: 839–849.

40.

Volker

(2012) Introduction to the special issue: high-functioning autism spectrum disorders in the schools. Psychology in the Schools 49: 911–916.

41.

Watson

Miller

Hannah

et al . (2015) Increased reward value of non-social stimuli in children and adolescents with autism. Front Psychology 6: 1026.

42.

Winter-Messiers

(2007) From tarantulas to toilet brushes: understanding the special interests of youth with Asperger syndrome. Remedial and Special Education 28:140–152.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB