Computational prosodic markers for autism.

The team used free acoustic software to measure how kids with autism use pitch, loudness, and timing when they talk. They compared these computer numbers to those of typically developing peers during short speech tasks.

The design was quasi-experimental: no random groups, just careful lab recording and math tools that anyone can download.

Kids with autism did not simply sound 'flat' or 'monotone.' Instead, they mixed the three prosody cues in an unusual balance. The contrasts were still there, just weighted differently.

The software could spot this odd weighting and tell the groups apart, offering a cheap, objective flag for atypical prosody.

Fusaroli et al. (2017) later pooled many similar studies and found the same small but real pitch differences, confirming the 2010 pattern holds across labs.

Heaton (2005) seems to clash: that study showed autistic kids actually detect tiny pitch shifts better than peers. The gap closes when you see Pamela tested fine-grained hearing, while H et al. looked at natural speech production. Different tasks, different strengths.

Heaton et al. (2008) adds another layer: about ten percent of autistic kids have 'splinter skills' with pitch scores far above average. The group average in H et al. still shows an odd cue balance, but single cases could look stellar.

You no longer need pricey equipment to screen prosody. Download the acoustic tools, record a short story retell, and let the numbers show if cue balance is off. When you hear odd intonation in a client, check whether pitch, loudness, or timing is carrying too much of the load, then shape the missing piece in therapy drills.