No Difference in Spatial Contrast Sensitivity Function (sCSF) Between Adolescents with ASD and Typically-Developing (TD) Controls

Background:

Individuals with Autism Spectrum Disorders (ASD) have been shown to exhibit enhanced processing of local aspects of visual stimuli (Plaisted et al, 1999). As local features of visual stimuli are mainly conveyed by high spatial frequency (HSF) information, and global configurations by low spatial frequency (LSF) information, it was speculated that individuals with ASD may possess relatively greater sensitivities to HSF versus LSF visual mechanisms (Deruelle et al, 2004).  Consistent with this hypothesis, in a study that measured visual acuity as the smallest gap in a C-shaped stimulus that could be correctly located, it was reported that adults with ASD exhibit higher visual acuity, i.e. greater sensitivity to HSF, than typically-developing (TD) controls (Ashwin et al, 2008). However, another study that measured contrast sensitivities for a range of low to high spatial frequencies using Vistech contrast sensitivity charts (Vistech Consultants, 1988), reported no significant differences between participants with ASD and TD controls (De Jonge et al, 2007). Thus, it is still debatable whether there exists differential spatial frequency sensitivity in ASD individuals. 

Objectives:

In the current study, we investigated differences in LSF versus HSF sensitivities between ASD and TD individuals using a more sensitive measure of spatial contrast sensitivity.  This involved using psychophysical techniques to map out the entire spatial contrast sensitivity function (sCSF), allowing group comparisons of peak spatial frequency (the spatial frequency yielding highest contrast sensitivity), the peak sensitivity (the highest contrast sensitivity) and visual acuity (the highest spatial frequency detectable).

Methods:

Data were collected from 10 adolescents with ASD and 25 TD adolescents, matched on chronological age (mean=185 months) and non-verbal IQ (mean=107). The stimuli were static horizontal Gabor patches, formed by convolving luminance gratings (subtending 3.1x3.1°) with a Gaussian envelope (S.D.=0.5°), presented at 7 different SFs (0.5cpd, 2cpd, 4cpd, 8cpd, 12cpd, 16cpd, 20cpd).  To obtain contrast sensitivities, on each trial, participants reported in which of two temporal intervals the stimulus was presented (centrally), and contrast was varied in a staircase paradigm. Double exponential curves were fit to the data for each participant to obtain their sCSF. Three variables of interest were extracted from the individual sCSFs: peak spatial frequency (peakSF), peak contrast sensitivity (peakCS), and highest spatial frequency detectable i.e. visual acuity (VA).

Results:

There were no significant group differences in peakSF (t(df=33)=1.08, p=0.286), peakCS (t(df=33)=0.211, p=0.834) or VA (t(df=33)=0.749, p=0.459).

Conclusions:

The results revealed typical processing of low and high spatial frequency information in adolescents with ASD. Differences between our results (typical VA in ASD) and those of Ashwin et al. 2008 (enhanced VA in ASD) might be explained by the different stimuli/tasks used in the two studies (gratings versus letters) or by the different ages (adolescence versus adults).