Effects of Weak Central Coherence on Resistance to Distractor Inhibition for Children with Autism

  

Effects of Weak Central Coherence on Resistance to Distractor Inhibition for Children with Autism

Background: Inhibition tasks requiring resistance to distracting stimuli, such as flanker tasks, reveal a deficit in children with autism (Christ et al., 2007).  However, for tasks requiring suppressing a prepotent response, children with autism appear to have intact inhibition (Bryson, 1983).  These data may reflect a distinction between prepotent response and resistance to distractor inhibition (Friedman & Miyake, 2004), which may be differentially impaired in autism.  Children with autism are also known to experience a weak bias to central coherence (Best, Moffat, Power, Owens, & Johnstone, 2007).  This could be related to resistance to distractor inhibition performance of children with autism.

Objectives: The current study investigated the possibility that children with autism may be differentially impaired in prepotent response and resistance to distractor inhibition and investigated effects of weak central coherence in autism on the flanker task.

Methods: Seventeen children with autism, with learning disabilities, and typically developing children were assessed.  Groups were matched using the Raven’s Coloured Progressive Matrices Test and were given the Embedded Figures Test.  Experimental tasks included one task of prepotent response inhibition (modified stop-signal task) and of resistance to distractor inhibition (flanker task).  During the stop-signal task, children categorized pictures as animals or non-animals except when they heard a beep, which occurred at three time intervals based on each participant’s average reaction time, where they were instructed to withhold response.  For the flanker task, children identified the direction of a central arrow varying in size and space from distractors across trials, which appeared flanked with congruently or incongruently pointing arrows. 

Results: On the stop-signal task, there was no group effect on reaction time, F(2, 44) = 0.55, p =.58.  Although beeps appearing at interval three were the most difficult, F(2, 41) = 4.02, p = .02, there were no group differences, F(2, 42) = 1.46, p = .24, and no sign of group effect on prepotent errors made, F(4, 84) = 0.47, p = .76.   Results from the flanker task showed a four way interaction, essentially showing that while the groups made comparable errors on the classic flanker condition, F(2, 44) = 2.37,p = .12, children with autism maintained interference on modified (easier) trials , F(2, 44) = 7.87,p = .001, while the group with learning disabilities did not, t(44)=3.07, p = .002.

Conclusions: Individuals with autism seem to show intact prepotent response inhibition.  On the flanker task, while the group with autism performed comparably to controls on the classic condition, they showed more error interference than the group with learning disabilities across modified conditions. This can be explained in terms of weak central coherence in autism, which makes competitor items in the global display less distracting and serves as an advantage for those with autism on the classic flanker.  However, when advantages from a weak central coherence in autism were accounted for they no longer experienced an advantage in comparison with controls, making inhibitory difficulties for those with autism clear.