Three groups of 18 children were selected for this study, one group with autism spectrum disorders (ASD), one group with developmental delays in which ASD was ruled out (DD), and one group with typical development (TD), from a pool of 3026 children who were screened with the Communication and Symbolic Behavior Scales Developmental Profile (CSBS DP, Wetherby & Prizant, 2002) Infant-Toddler Checklist under 24 months of age. The CSBS DP Behavior Sample was videotaped on selected children as a second-level evaluation during the second year of life. The Infant-Toddler Checklist had a sensitivity and specificity of 88.9% for this sample of children. Significant group differences were found on the Infant-Toddler Checklist and the Behavior Sample, however, these differences did not distinguish children with ASD and DD with high accuracy. The videotapes of the Behavior Sample were reanalyzed to identify red flags of ASD. Nine red flags differentiated children in the ASD group from both the DD and TD groups and four red flags differentiated children in the ASD Group from the TD group but not the DD group. These 13 red flags were found to discriminate the three groups with a correct classification rate of 94.4%.

Chromosomal copy number variants (CNV) are the most common genetic lesion found in autism. Many autism-associated CNVs are duplications of chromosome 15q. Although most cases of interstitial (int) dup(15) that present clinically are de novo and maternally derived or inherited, both pathogenic and unaffected paternal duplications of 15q have been identified. We performed a phenotype/genotype analysis of individuals with interstitial 15q duplications to broaden our understanding of the 15q syndrome and investigate the contribution of 15q duplication to increased autism risk. All subjects were recruited solely on the basis of interstitial duplication 15q11.2-q13 status. Comparative array genome hybridization was used to determine the duplication size and boundaries while the methylation status of the maternally methylated small nuclear ribonucleoprotein polypeptide N gene was used to determine the parent of origin of the duplication. We determined the duplication size and parental origin for 14 int dup(15) subjects: 10 maternal and 4 paternal cases. The majority of int dup(15) cases recruited were maternal in origin, most likely due to our finding that maternal duplication was coincident with autism spectrum disorder. The size of the duplication did not correlate with the severity of the phenotype as established by Autism Diagnostic Observation Scale calibrated severity score. We identified phenotypes not comprehensively described before in this cohort including mild facial dysmorphism, sleep problems and an unusual electroencephalogram variant. Our results are consistent with the hypothesis that the maternally expressed ubiquitin protein ligase E3A gene is primarily responsible for the autism phenotype in int dup(15) since all maternal cases tested presented on the autism spectrum.

Three studies were conducted to evaluate the validity and reliability of the three measures of the Communication and Symbolic Behavior Scales Developmental Profile (CSBS DP): (1) a one-page parent-report checklist; (2) a four-page follow-up caregiver questionnaire (CQ); and (3) a behavior sample (BS), which is a face-to-face evaluation of the child. Participants for these studies were drawn from a pool of 603 children for the checklist and CQ (ages 6-24 months) and 364 children for the BS (ages 12-24 months). Study 1 examined the concurrent relationship of standard scores for the checklist, CQ, and BS. Large correlations were found between the checklist and CQ and moderate to large correlations were found between each of the parent report tools and the BS. Study 2 examined test-retest stability by comparing the raw and standard scores over a 4-month retest interval. The results indicated significantly greater retest raw scores but no significant differences between standard scores from test to retest for the checklist, CQ and BS, providing evidence that the three measures detect growth over short periods but produce relatively stable rankings of children. Study 3 examined the concurrent and predictive relationship of the three CSBS DP measures and children's outcomes on standardized tests of receptive and expressive language at 2 years of age. Moderate to large correlations were found between all of the CSBS DP measures and language outcomes at 2 years of age. Multiple regression analyses indicated that the three composites were a significant predictor of receptive and expressive language outcomes. The findings from these three studies support the use of the CSBS DP as a screening and evaluation tool for identifying children with developmental delays at 12 to 24 months of age.

This descriptive study evaluates the speech disfluencies of 8 verbal children between 3 and 5 years of age with autism spectrum disorders (ASD). Speech samples were collected for each child during standardized interactions. Percentage and types of disfluencies observed during speech samples are discussed. Although they did not have a clinical diagnosis of stuttering, all of the young children with ASD in this study produced disfluencies. In addition to stuttering-like disfluencies and other typical disfluencies, the children with ASD also produced atypical disfluencies, which usually are not observed in children with typically developing speech or developmental stuttering. (Yairi & Ambrose, 2005).

The purpose of this study was to examine the expressive prosody of young verbal children with autism spectrum disorder (ASD). Method: Vocalizations were examined in 14 children between 24 and 68 months of age: 7 diagnosed with ASD and 7 with typical development, matched for expressive language abilities. To obtain precise measures of prosody, spontaneous speech samples were obtained and then coded using The Prosody Voice-Screening Profile (PVSP; Shriberg, Kwiatkowski, & Rasmussen, 1990). Results: The 2 groups did not differ in their production of rate, loudness, or pitch. The group with ASD did, however, produce atypical stress patterns significantly more often, such as misplacement of stress in multisyllabic words or reduced stress. Conclusion: Results of the current investigation highlight the importance of considering prosody, specifically misplaced stress, when assessing and treating young children with ASD.