More on Autism and Faces

Less than a week after I had to correct myself on autism and face perception, I read another article on the subject that has me skeptical.

Let's see if we can apply what we learned before. The conclusions from my earlier piece were a couple: 1) people with autism do not seem to possess a global deficit in face perception although they may evaluate faces differently and 2) the evidence with respect to fMRI is mixed.

Here is the new work (unfortunately not available as a paper yet):

In a report to be presented May 5 at the International Meeting for Autism Research in Seatlle, researchers from UCLA will show that children with autism can't do this. They hear and they see, of course, but the areas of the brain that normally respond to such visual cues simply do not respond.

Led by Mari Davies, a UCLA graduate student in psychology, and Susan Bookheimer, a professor of psychiatry and biobehavioral sciences at the Semel Institute for Neuroscience and Human Behavior at UCLA, the research compared brain activity between 16 typically developing children and 16 high-functioning children with autism. While undergoing functional magnetic resonance imaging (fMRI), both groups were shown a series of faces depicting angry, fearful, happy and neutral expressions. In half the faces, the eyes were averted; with the other half, the faces stared back at the children.

With the typically developing group, the researchers found significant differences in activity in a part of the brain called the ventrolateral prefrontal cortex (VLPFC), which is known to play a role in evaluating emotions. While these children looked at the direct-gaze faces, the VLPFC became active; with the averted-gaze pictures, it quieted down. In contrast, the autistic children showed no activity in this region of the brain whether they were looking at faces with a direct or an indirect gaze.

"This part of the brain helps us discern the meaning and significance of what another person is thinking," Davies said. "When responding to someone looking straight at you, as compared to someone who's looking away, the brain discerns a difference. When the other person looks away, the brain quiets down."

For instance, with angry expressions, the brain may quiet down, because when a negative gaze is averted, it is no longer seen as a direct threat. "Gaze has a huge impact on our brains because it conveys part of the meaning of that expression to the individual. It cues the individual to what is significant," Davies said.

While the results show the key role of eye gaze in signaling communicative intent, it also shows that autistic children, even when gazing directly into someone's eyes, don't recognize visual cues and don't process that information. That may be why children diagnosed with autism have varying degrees of impairment in communication skills and social interactions and display restricted, repetitive and stereotyped patterns of behavior.

As you can read, they make the same argument that I had been making before correcting myself. Basically the line they are taking with this research is that the failure to activate certain face-specific processing areas in the brain may be responsible for some of the social deficits in autism.

So let's apply what we learned. Before I say anything let me just note that this paper is not published, so my skepticism may not be justified. However, I am skeptical for the following reasons:

  • 1) Have the experimenters controlled for attention? The patients with autism may not have been paying as much attention to the faces as the controls, and this may explain the differences in activation.
  • 2) Have the experimenters controlled for gaze? This is sort of related to the attention issue, but we know that people with autism focus on different features when they look at faces. Focusing on different features could also account for the differences in activation. (Incidentally, the gaze that they are talking about in the article is the gaze in the pictures. I am talking here about where the experimental subjects are looking. Did they look at the picture straight in the eyes or off to the side?)
  • 3) Have the experimenters controlled for verbal fluency? No reason that anyone would know this, but it was mentioned in the review from my earlier piece that the ability to evaluate faces can be related to verbal fluency. It is important to control this factor because if you picked a bunch of people with autism that had low fluency, you may be creating an effect that wouldn't otherwise be there.

Again, I don't really know until I read it whether these factors were controlled for, but given my latest run-in with this field I feel like a little skepticism is in order. We don't want to leap to conclusions about what people with autism can and cannot do without making sure we controlled for these factors.


More like this

Toddlers learn new words at an astonishing rate—an average, according to Steven Pinker, of over a word every two hours. Yet attempts to drill children to improve vocabulary are often frustrating. Kids seem to learn words better through observing the environment than they do by rote. So what exactly…
"Instead of trying to produce a programme to simulate the adult mind, why not rather try to produce one which simulates the child's?" - Alan Turing (Computing Machinery, p456) One of the defining features of childhood cognition is "behaving without thinking." Not surprisingly, developmental…
Normal children - and adult patients with frontal damage - frequently have difficulty changing their responses to stimuli when the correct response changes. This difficulty is often considered an inability to switch between rules, but might result not so much from an inability to switch as from an…
What was your 6th birthday party like? If you successfully retrieved that memory, you may now be ever so slightly less able to remember your other childhood birthdays. A variety of behavioral evidence has shown that such "retrieval induced forgetting" of strongly competing memories is…

I didn't go to that presentation at IMFAR (there was a huge number of face studies, as sometimes happens at IMFAR), but you can find the abstract here

1. They controlled for attention by using fixation crosses (this is a good idea). But it's unclear whether they had some kind of task as well.

2. They used eye-tracking to show the participants were fixating on facial features. But--my guess (for lots of reasons) is that they didn't do their eye tracking in the magnet.

The abstract can't answer what you raise in #3.

I wonder which face image set they used. One set I know of that meets the description in the abstract is the Karolinska set, which features very exaggerated, obviously fake facial affect.

Autistic kids cleverly fail to be fooled by fake distress, while typical kid are fooled every time. Are there limits to the extent to which autistics will respond to obviously fake emotions?

Another thing to consider is that autistics, starting very early in life, orient in atypical ways to visual stimuli, whether those stimuli are social or non-social. This would be in real-life situations (as opposed to looking at images while immobilized in a magnet). We've proposed that this atypical orienting is adaptive, considering what we know about atypical visual perception in autism (Mottron et al., 2007). So it's possible that an "averted" gaze doesn't mean the same thing for autistics and non-autistics.

But it's hard to guess without a full study to read (and without having seen the presentation at IMFAR). Critical thinking is always good though, and essential in anything to do with autism.

I am also wondering at the set of pictures used---my son has often had a lot of difficulty identifying emotions and facial features and if the photos used were black and white, that might also have confused him.