Elizabeth Pellicano, Lisa Gibson, Murray Maybery,
Kevin Durkin, and David R. Badcock
School of Psychology, University of Western Australia

Individuals with autism often demonstrate preserved or superior skills in various areas, such as enhanced rote memory, preoccupation with parts of objects, and savant abilities. Psychological theories tend to focus on the impairments that characterise autism—the difficulties in socialisation, communication, and flexible behaviour— but struggle to explain some of the non-social abilities. Uta Frith (1989) proposed that the pattern of social and non-social assets and deficits in autism can be explained by a particular style of processing information: weak central coherence. She describes normal information processing as a preference for perceiving global information in favour of the individual parts of a stimulus, and a tendency to strive for meaning. Frith asserts that this normal drive for coherence is much weaker in individuals with autism, and is characterised by a preference for processing parts over wholes; an overall lack of influence from context.

Milne and colleagues (2002) first examined how this preference for parts in autism might manifest itself in the brain. Specifically, they looked to see whether children with autism had abnormal processes in one of the pathways in the visual system, the magnocellular pathway (which was named after the big size of the brain cells that make up the system). They presented children with a task tapping magnocellular functioning, a global dot motion task, in which they were required to detect the direction of moving dots. They found that high-functioning children with autism obtained significantly higher global motion thresholds (lower sensitivity) than typically developing children of similar age and ability; that is, children with autism required a higher percentage of the dots to be moving in one direction in order to detect the overall direction of movement. Milne et al. (2002) suggested that these findings demonstrate that children with autism might have a general impairment in the magnocellular visual pathway. They speculated that this imbalance could lead to a different way of seeing things and might explain how children with autism have difficulty integrating parts into wholes.

We further investigated Milne and colleague’s (2002) findings. We wanted to examine whether children with autism had a generalised abnormality in the magnocellular pathway or whether it was present in a more specific area. Twenty children with high-functioning autism, and 20 typically developing children matched on age, gender, handedness, and nonverbal ability were required to complete two visual tasks tapping magnocellular functioning. To test this, we had children perform 2 visual tasks on the computer – both looked at the functioning of the magnocellular pathway. The first task examined early functioning in this pathway, and tested one’s ability to detect a flickering stimulus. A second task looked at more complex processes further along the visual pathway, and required children to detect the direction of movement of coherently moving dots (similar to the task used by Milne et al.). We also administered the Children Embedded Figures Test (CEFT), a common test measuring central coherence.

We reasoned that if the high global motion thresholds are due to a general impairment in the magnocellular pathway, then children with autism should perform worse than typically developing children in both visual processing tasks. Alternatively, if the elevated motion coherence thresholds are a result of higher-level magnocellular functioning, then children with autism should not show poor sensitivity in detecting flicker in the lower-level visual task.

Consistent with previous research (e.g. Milne et al., 2002; Spencer et al., 2000), children with autism performed worse than typically developing children on the higher-level magnocellular task tapping children’s ability to perceive global motion. However, we found no difference between groups on the low-level flicker contrast sensitivity task, suggesting that children with autism do not exhibit generalised impairments in the magnocellular visual pathway. Additionally, children with autism were much faster at finding the hidden figures in the CEFT than their typical peers, and performance on the CEFT was inversely related to global motion thresholds in the autism group (that is, children who were fast at finding the hidden figures, were also worse at perceiving global motion). These results suggest that children with autism have a specific difficulty in integrating pieces of information into coherent wholes, a difficulty which seems to manifest at higher levels in the magnocellular visual pathway.

We would like to thank all of the children and parents who participated in this project, Alana Maley who interviewed parents regarding diagnostic information, and the Apex Foundation Trust for Autism who provided financial support for this work. For more information regarding this study, and about other studies that we are currently conducting, please contact:

Liz Pellicano
Child Study Centre
School of Psychology
University of Western Australia
Phone: 9380 3575
liz@psy.uwa.edu.au

Frith, U. (1989). Autism: Explaining the Enigma. Oxford, UK: Blackwell Publishers

Happé, F. (1999). Autism: Cognitive deficit or cognitive style? Trends in Cognitive Sciences, 3, 216-222.

Milne, E., Swettenham, J., Hansen, P., Campbell, R., Jeffries, H., & Plaisted, K. (2002). High motion coherence thresholds in children with autism. Journal of Child Psychology and Psychiatry, 43, 255-263.

Spencer, O’Brien, J., Riggs, K., Braddick, O., Atkinson, J., & Wattam-Bell, J. (2000). Motion processing in autism: evidence for a dorsal stream deficiency. NeuroReport, 11, 2765-2767.