Colour Perception in Autism and Williams Syndrome

Background:  Recent studies report a varied visual function profile in Autism Spectrum Disorders (ASD; Simmons et al., 2009), with relative strengths and weaknesses found in dorsal and ventral stream functions. The genetic disorder Williams Syndrome (WS) has been described as illustrating a contrasting social profile to ASD, but socio-communicative deficits overlap in the two conditions (Klein-Tasman et al., 2009) as do sensory processing difficulties, with profound deficits in dorsal stream visuo-spatial processing in WS. Given the increasing importance of understanding sensory processing difficulties and sensory reactivities in both conditions, there is a need to better differentiate between the two sensory profiles.  The ventral stream primarily processes visual properties which underlie key aspects of socio-communicative ability (e.g. face/object recognition). Colour perception (a key function of the ventral stream) is relatively understudied in ASD and WS. Despite this, there is an increasing use of colour in behavioural interventions for ASD (e.g. sensory rooms, coloured overlays; Ludlow et al., 2006).

Objectives:  To assess colour perception as a measure of ventral stream visual function in ASD and WS, relative to typically developing children (TD), at different levels including chromatic-discrimination (sensory level) and colour preference (cognitive/affective level).

Methods:  15 individuals with ASD (7-18y), 26 with WS (7-20y), matched individually on the Ravens Coloured Progressive Matrices to TD (4-8y), took part in the study. Two tasks were used to examine different aspects of colour perception. First, a chromatic-discrimination task measured chromatic-discrimination thresholds along three colour directions, the cone-opponent (“red-green”, “blue-yellow”) and luminance colour axes. The task required participants to identify the direction of a briefly flashed arrow of variable contrast against a grey background (staircase procedure). To measure colour preference, a two-alternative-forced choice task was used in which participants chose the preferred colour from each of all possible pairs of nineteen colours varying systematically in hue, lightness and saturation. 

Results:  Performance patterns differed between ASD and WS groups relative to the TD group. On the chromatic-discrimination task, significant interaction between colour axis and group was found between ASD and TD (p<0.01), with ASD “Blue-Yellow” axis performance significantly worse relative to TD (p<0.01). No significant differences were found between the WS and TD groups (p=0.39). For colour preference,   patterns differed between groups. TD and ASD groups show similar preference patterns across hues, but whereas the TD group show increased preference for lighter-vs-darker colours, the ASD group do not (p<0.01).  The WS group displayed greater inter-individual variation. 

Conclusions: These findings provide evidence for atypical colour perception in both ASD and WS but with condition-specific patterns. Relative to mental-age TD controls, in ASD “low-level” chromatic-discrimination is impaired, and colour preference depends on hue but lacks dependence on lightness.  In WS, chromatic-discrimination is unimpaired, but colour preference shows high inter-individual variation. These results suggest that in behavioural interventions for ASD, it is more important to consider affective responses to hue than to lightness of colours, whereas in WS, colours should be tailored to the individual. Future research will further inform better use of colour in behavioural interventions.