A team recorded more than 200 children, ages 10 months to 4 years, for a period of six to eight (12 hr) days. Computer software filtered out crying, sneezes and coughs and focused on the sounds that resembled syllables. The group of children was comprised of: boys - 48 typically developing and 64 with autism, girls - 48 typical and 64 with autism. SES was taken into consideration.
The mean age of the studied groups; typically developing - 28.176 months, delayed - 28.963 months, autism - 35.928 months. Some typically developing children were as young as 10 months, and so were some of the delayed children. The autism group had some as young as 16 months, but none younger.
This research effort has claimed to have the ability to identify autism by analysis of speech patterns.
Onset of autism symptoms has some recently defined patterns. These identified patterns might indicate that a vocal analysis which intends to provide diagnostic indicators for early identification of autism would apply only to those in the autism spectrum who have experienced a pattern of plateau, or a pattern of no loss and no plateau; perhaps 56% of those within the autism spectrum.
Many parents find that their autism affected child at first developed speech normally and then experienced regression in speech ability (usually to non verbal) followed by onset of stereotypic behaviors. In cases of normal development being followed by such regression, early detection would be a moot point; the indicator for autism in those cases is the regression. This regression has been studied to be evidenced in 44 % of the autism spectrum.
• Regression (n=44%): A loss of previously acquired social, communication or cognitive skills prior to 36 months
• Plateau (n=17%): Display of only mild developmental delays until the child experiences a gradual to abrupt developmental halt that restricts further advancement of skills
• No Loss and No Plateau (n=39%): Display of early warning signs of autism spectrum disorders without loss or plateau
(From Kennedy Krieger Study on Autism Onset Patterns)
According to Automated vocal analysis of naturalistic recordings from children with autism, language delay, and typical development, "A typically developing child’s age can be predicted at high levels (accounting for more than two thirds of variance) based on this ﬁrst attempt at totally automated acoustic modeling with all-day recordings. The automated procedure has proven sufficiently transparent to offer suggestions on how the developmental age prediction worked."
Vocal analysis which results in autism identification is a little ways off, but worthy of further pursuit. The goal seems to be that of providing a means for earlier autism detection. Some propose that earlier detection will result in better lifelong outcome, since interventions can start sooner.
So far, studies have confirmed that overall intense earliest interventions result in slight communication improvement for the autism affected child, but does not diminish autistic features for the majority. More study is needed to determine overall lifetime benefits of earliest interventions.
Overall, the prognosis for the majority of autism affected individuals who have a pattern of plateau, or no loss and no plateau - is that of less severity in autism symptoms as they become older, regardless of interventions. Earliest detection for the slight minority within the autism spectrum who experience regressive autism seems to remain out of reach.
As an aside, my autistic daughter had a rather hard to identify slide into autism.
While she did not gravitate toward typical language development, as early as one year old we heard her imitate the evil queen’s laugh from an animated movie. She also imitated Darla in a Little Rascals movie by saying, “No, no, no way.” We made a game out of counting as Steve held up each finger, “One, two, three, four, five.” We even made a game out of saying, “Daddy.” While fully capable of repeating those words, she did not seem to equate their meaning; or if she did understand their meaning, she did not apply it. She seemed to utilize the words as a form of verbal stimulatory behavior, and not as a form of actual communication. Since every child’s development is different, we hoped she just needed more time. (Hello, Dr. Wells)