While the primary language deficit in autism has been thought to be pragmatic, and in specific language impairment (SLI) structural, recent research suggests phenomenological and possibly genetic overlap between the two syndromes. To compare communicative competence in parents of children with autism, SLI, and down syndrome (DS), we used a modified pragmatic rating scale (PRS-M). Videotapes of conversational interviews with 47 autism, 47 SLI, and 21 DS parents were scored blind to group membership. Autism and SLI parents had significantly lower communication abilities than DS parents. Fifteen percent of the autism and SLI parents showed severe deficits. Our results suggest that impaired communication is part of the broader autism phenotype and a broader SLI phenotype, especially among male family members.

Reports of a relatively high prevalence of absolute pitch (AP) in autistic disorder suggest that AP is associated with some of the distinctive cognitive and social characteristics seen in autism spectrum disorders. Accordingly we examined cognition, personality, social behavior, and language in 13 musicians with strictly defined AP (APS) and 33 musician controls (MC) without AP using standardized interviews and tests previously applied to identify the broad autism phenotype seen in the relatives of autistic probands. These included the Pragmatic Rating Scale (PRS) (social aspects of language) the Personality Assessment Schedule (PAS) (rigidity, aloofness, anxiety/worry, hypersensitivity), and WAIS performance subtests (PIQ). On the basis of their behavior in the interviews, subjects were classified as socially eccentric, somewhat eccentric, or not eccentric. Forty-six percent of the APS, but only 15% of the MC, were classified as socially eccentric (p < .03). APS but not MC showed higher scores on block design than on the other PIQ tests (p < .06), a PIQ pattern seen in autism spectrum disorders. Although APS and MC did not differ significantly on other measures it is of note that APS mean scores on the PRS and PAS (5.69, 4.92) were almost twice as high as those for the MC (3.03, 2.45). Thus, musicians with AP show some of the personality, language, and cognitive features associated with autism. Piecemeal information processing, of which AP is an extreme and rare example, is characteristic of autism and may be associated as well with subclinical variants in language and behavior. We speculate that the gene or genes that underlie AP may be among the genes that contribute to autism.

Auditory problems are common in individuals with autism. Several previous studies have indicated that Auditory Integration Training (AIT), a sound-based intervention, may bring about significant improvement in autism. AIT entails listening to specially processed music for a total of 10 hours over a 10- to 20-day period. In the present study, 19 individuals with autism were assigned at random to either an experimental group (n = 9), who listened to AIT-processed music, or a placebo group (n = 10), who listened to the same but unprocessed music. All evaluators were blind to group assignment. Behavioral, electrophysiological (ERP), and audiometric measures were assessed prior to and following AIT. A significant decrease in Aberrant Behavior Checklist scores was observed in the experimental group at the 3-month follow-up assessment. Of the 19 participants, only three experimental group members and two placebo group members were able to perform the auditory P300 ERP task. All five participants showed abnormal P300 ERPs prior to the AIT listening sessions. Three months following AIT, all three treated participants showed a dramatic improvement in their auditory P300 ERP, whereas none of the participants in the placebo group showed change. The participants' poor language and attention skills precluded collecting sufficient data for formal statistical evaluation of the results from the battery of audiometric tests.