Executive Functioning in High Functioning Autism

A review of: GILBERT, S., BIRD, G., BRINDLEY, R., FRITH, C., BURGESS, P. (2008). Atypical recruitment of medial prefrontal cortex in autism spectrum disorders: An fMRI study of two executive function tasks. Neuropsychologia DOI: 10.1016/j.neuropsychologia.2008.03.025

Executive functioning is an umbrella term used in clinical neuropsychology and cognitive neurosciences to refer to a series of “higher-order” cognitive processes usually associated with frontal lobe functioning. These include planning, organizing, categorizing, response inhibition, monitoring, multitasking, etc. Research on executive functioning and autism has provided mixed results likely due to differences in the area of executive function measured. In this functional MRI study, the authors used a new test of executive function used to assess for stimulus dependent vs. stimulus independent thoughts. Stimulus dependent refers to cognitive processing (thinking) that is associated with, or dependent on, a specific stimulus that is presented. For example, I may ask you to press the b key when you see a blue square or the r key when you see a red square. For this task the stimulus dependent phase consisted of capital letters presented in alphabetical order. Once a letter was presented, the person had to press one key if the letter contained only straight lines (such as the letter A) and a different key if the letter contained curves (such as the letter B). In the stimulus independent task, the person was presented with one letter and ask to follow the same response pattern (one key if straight lines – another key if curves). However, the second letter presented was random and did not follow the alphabetical order, yet the person was asked to respond based on the next alphabetical letter. For example, assume the first letter was “C”, then the next letter presented was the letter “H” (random) yet the person was asked to respond to the next alphabetical letter starting from the first letter presented (C), thus the next response was based on the characteristics of the letter D, even though the person was seeing the letter H. This task therefore, requires the person to continue to “think” of the characteristics of the letters in alphabetical order, independent of the letters presented (which now are distractors).

In the study the authors examined 15 adults with high functioning autism and 18 typically developing adults that were matched for age (mean 38) and IQ (mean 119). The participants performed the task while undergoing a functional magnetic resonance imaging scan (fMRI). There was no difference in accuracy or response times between the groups. Both groups showed more activation of the lateral frontal and parietal cortex on the more difficult stimulus-independent task. However, the autism group had significantly more activation of specific areas of the medial prefrontal cortex during the easier stimulus dependent condition than the typically developing group (this was interpreted as failure to deactivate these areas). The authors argued that these results suggest an atypical brain organization in HFA with limited deactivation of the rostral prefrontal cortex during easier task compared to typically developing individuals. However, the equivalent performance (both groups did just as well on the task) suggests that the fMRI findings simply reflect different approaches to cognitive performance between the two groups.

Feeding difficulties in children with autism. Is it the autism or is he just a picky eater?

A review of: Martins, Y., Young, R.L., Robson, D.C. (2008). Feeding and Eating Behaviors in Children with Autism and Typically Developing Children. Journal of Autism and Developmental Disorders DOI: 10.1007/s10803-008-0583-5

In this study the researchers were interested in examining the rates of eating behavior problems among children with autism. Based on a parent-completed questionnaire, the researchers compared 41 children with autism spectrum disorders, 14 non-affected siblings, and 41 non-related typically developing children. These children were matched on communication, socialization, and daily living skills (based on the Vineland Adaptive Behavior Scale).

When compared to typically developing kids, children with ASD had more food avoidance behaviors (e.g. not eating something that has been touched), and food Neophobia (fear of unfamiliar foods). These ASD children were also more likely to control the feeding (feeding revolves around the child’s demands, such as timing, type of food, etc) and consequently less likely to have the parent control the feeding ('child must try the food before it is rejected', etc). Surprisingly, there were no differences in rates of ritualistic feeding behaviors, underscoring the commonly observed phenomena of ritualistic feeding behaviors among typically developing children.

However, when compared to their unaffected siblings, the ASD children continued to display higher rates of food avoidance behaviors and food Neophobia. Yet, both groups had the same level of control over feedings. This suggests that the findings concerning high levels of child control over feedings and low levels of parent control over feedings observed in the ASD group when compared to the typically developing group, could reflect parenting styles (since the parent seems to display the same behaviors towards the unaffected sibling) rather than something unique about the child with autism. It is also possible that the parenting behavior towards the unaffected sibling (such as allowing the child more control over the feeding situation) may simply be a management tool (whether conscious or not) used in order to standardize the feeding process in a household with a child with special needs (such as to not appear unfair to the unaffected sibling).

Two last comments. It is possible that the failure to see significant differences between the children with ASD and the unaffected siblings may be a pure artifact of the limitation of the statistical tools uses. The authors compared 14 matched pairs (N=14 per group) on 7 outcome variables via standard ANOVA. An argument can be made that the statistics used were not appropriate for the nature of the data due to limited power.

Finally, I initially had a concern about the fact that these groups had been matched on several functioning scales. I thought that the results would not be generalizable since the groups were purposely selected to be almost identical on many key factors. But the fact that they found significant problems with food avoidance and neophobia in the ASD group but not the others underscores how prevalent these eating behavior problems may be among children with ASD.