IMFAR 2011: Characterizing Cognition in Nonverbal Individuals With Autism: Innovative Assessment & Treatment

This session was introduced by Geri Dawson, and included a number of presenters, whose work I am only summarizing briefly. Again, any omissions or errors are my doing. -SR

Chair: Geri Dawson, Chief Science Officer for Autism Speaks

Why a specific focus on nonverbal kids with autism? Because they’re often not included in research — such kids can be difficult to test, especially when so many evaluation paradigms include verbal testing. As a result, little is known about why some kids with autism don’t develop spoken language, or even what best approaches are.

Biggest challenge: Evaluating cognitive abilities! Lots of kids who appear to have intellectual disability do not, at least not to degree perceived.

In 2008, Autism Speaks launched a special initiative focused on nonverbal autism, and in April 2010, NIH sponsored a workshop to address what we do know, and gaps in knowledge.

We need understand that nonverbal individuals with autism are a variable population. We need to be thoughtful about what “nonverbal” means. There is no single set of defining characteristics for a nonverbal individual with autism, but there are three main types:

  1. Preverbal – young, newly diagnosed, or pre-intervention
  2. Non-speaking (AAC users)
  3. Non-communicative — does not have functional communication.

Need innovative approach to assess cognitive and communication abilities.

Other thoughts:

  • What percent of kids with ASD are nonverbal? Estimated: 30%.
  • Age 5 is not a gatekeeper age for future verbal ability.
  • Connie Kasari: We need to figure out better ways to assess the kids who have higher functioning skills that have gone undetected.
  • Nicole Gage: What if your sensory system is somehow atypical, discordant — and you can’t tune them out? If this was happening with your visual system as well? Can’t you posit that these differences can get in way of the profound developmental changes that occur in infancy and early childhood, and interfere with development in language ability, interfere with the right signals getting to the right places in the brain?
  • Geri Dawson: Nonverbal is a hard thing to define, many distinctions are arbitrary, people define it differently. Recommends seeing NIH’s 2010 Workshop on Nonverbal School-Aged Children with Autism.


April Benasich, Rutgers University

Is the Frog Green?

Note: study is ongoing, still seeing kids, still analyzing data. We’re trying to figure out what these kids know! Many don’t even sign.

Do kids who are nonspeaking have capcity to understand spoken and written language? Lack of expressive language does not indicate lack of receptive language!

Tried to apply info from assessing nonverbal indivduals (babies) to nonverbal individuals with autism.

We need to look at “higher functioning” autistic kids compared to nonverbal autistic kids.

John Connolly, McMasters University

Innovations in Assessing Cognition in Nonverbal Children With Autism

His previous research has been with individuals with traumatic brain injuries, in “vegetative” state, meaning primarily, no evidence of awareness of self or environment at any time.

He is amazed by how much nonverbal autistic people get treated like people in vegetative state, in terms of assumptions regarding lack of response to external input.

Many subjects whose testing indicate low IQ before testing were able to demonstrate greater comprehension using his much more rigorous and constrained methodologies and approaches.


Nicole Gage, University of California, Irvine

Developing Biomarkers of Language Impairment in Nonverbal Children With Autism: A MEG Investigation.

She’s a neurolinguist, interested in why language didn’t blossom in these kids as with typical peers.

How can we characterize auditory and language function in kids who are minimally verbal, and possibly less compliant as a result?

Case study approach: Case-by-case basis, also included siblings discordant for autism including twins. Only included kids with idiopathic autism, not kids with PDD-NOS or Asperger’s.

Multidisciplinary investigation of the auditory system: looking at cortical structures, as relates to language outcome and brainstem pathways.

MEG: Magnetoencephalography. MEG gets improved source localization compared to EEG. It is considered a “direct” sign of neural activity. MEG is also silent, compared to MRI, and non-invasive — parent/caregiver can be in room with child as long as child doesn’t move. Can do time, freequency domain analyses, etc. — MEG is a very rich signal.

Have to think about synchrony, about what’s going on in the brain. Speech is the most complex signal our brain needs to decode, lots of disruption and interfering environmental stimuli. The audtiory system’s highlight is its sensitvity to decoding this signal — but this is not working as well in kids with autism, who can have disruption and asynchrony.

Goal: use our brain measures to find and predict brain-based biomarkers of language impairment.


Connie Kasari, UCLA

CCNIA Intervention: Spoken and Augmented Means of Communication

Problem: we have significant number of  kids with autism without verbal communication ability despite years of intervention. Very concerning to parents and professionals. These kids get excluded from research as mentioned previously so we know very little about them.

Question: How do you intervene with kids older than age 5 who are still not using spoken and written language?

Study merged two evidence-based interventions for younger language-learning kids

  • JASPER: Joint attention, symbolic play, engagement and regulation
  • EMT: Enhanced milieu treatment
  • Added in speech generating device

Primary aim: To increase expressive language in a brief (6 mos) intervention (spoken or augmented)

Randomized trial of 96 kids age 5 – 8 diagnosed with autism via ADOS, across three sites (Vanderbilt, UCLA, Kennedy Krieger):

  • Above 24 months in terms of nonverbal cognition
  • At least two years of intensive intervention, minimally verbal (<20 non-scipted words)

Most kids came in with very poor play skills, didn’t want to engage or have others folks play with their toys.

75% of the kids were responders in first three months/first phase. What does first-phase responder means? Novel grouped words, Socially communicative utterances.

We’re hoping to figure out why kids do and do not respond. The kids who don’t respond tend to have higher rates of compulsive behavior.

The treatment can work, in a very short time frame: Three months.

Likely important moderators for success haven’t been ID’d just yet, but repetitive behaviors seem to be one factor. We’ll know more in a year, hopefully.