Perception, 2011, volume 40, pages 739 ^ 742
doi:10.1068/p6953
SHORT AND SWEET
Individual differences in visual search: relationship to autistic traits, discrimination thresholds, and speed of processing Jon Brockô, Jing Y Xu½, Kevin R Brooks½
ARC Centre of Excellence in Cognition and its Disorders (ô and Macquarie Centre for Cognitive Science; ½ Department of Psychology), Macquarie University, Sydney, NSW 2109, Australia; e-mail:
[email protected] Received 7 March 2011, in revised form 22 June 2011
Abstract. Enhanced visual search is widely reported in autism. Here we note a similar advantage for university students self-reporting higher levels of autism-like traits. Contrary to prevailing theories of autism, performance was not associated with perceptual-discrimination thresholds for the same stimuli, but was associated with inspection-time thresholdöa measure of speed of perceptual processing. Enhanced visual search in autism may, therefore, at least partially be explained by faster speed of processing.
One of the many intriguing aspects of autism is that, alongside the defining impairments in social interaction, communication, and behavioural flexibility, autistic individuals often demonstrate atypical visual perception (Simmons et al 2009). Of particular note, numerous studies have shown that individuals with an autism diagnosis are quicker to detect targets in visual-search tasks compared with appropriately matched control subjects (eg Plaisted et al 1998). O'Riordan and Plaisted (2001) attributed this performance advantage to an enhanced ability to discriminate between perceptually similar objects. In support of this hypothesis, they reported that autistic superiority was greatest in the most difficult visual-search tasks with highest levels of target ^ distractor similarity. However, targets and distractors were still easily discriminable, and other direct investigations of visual discrimination have failed to find an autistic advantage (Franklin et al 2008). An alternative hypothesis is suggested by evidence that autistic individuals perform relatively well on inspection-time tasks that index speed of perceptual processing (Sheuffgen et al 2000). Faster speed of processing may confer an advantage on visual-search tasks that would be magnified as the processing demands increased (cf Joseph et al 2009). As a first step towards investigating these two competing hypotheses, we tested thirty-eight undergraduate students (fourteen males; mean age 22 years, range 18 to 48 years) on two visual-search tasks in which the target was defined by a conjunction of orientation and spatial frequency (figure 1). Orthogonal task manipulations included display size, target presence or absence, and task difficulty (target ^ distractor similarity). Participants also completed the Autism Quotient (AQ öBaron-Cohen et al 2001), a 50-item self-administered questionnaire targeting sub-clinical autism-like traits (mean 17:0; range 6 ^ 34). Visual-search response times were then subjected to ANCOVA with z-transformed AQ score as the covariate. Individuals with higher AQ scores were faster overall on the visual-search task (F1, 36 7:27, p 0:011). They also demonstrated smaller effects of display size (F1:2, 42:6 5:22, p 0:022) and a marginal reduction in the effect of target ^ distractor similarity (F1, 36 3:13, p 0:085) (see figure 2). Thus, the autistic advantage on visual-search tasks extends to individuals in the general population selfreporting high levels of autistic traits (see also Almeida et al 2010).
740
J Brock, J Y Xu, K R Brooks
Target
Distractors
Easy
908, 2 cycles degÿ1
308, 2 cycles degÿ1
908, 6.73 cycles degÿ1
908, 2 cycles degÿ1
758, 2 cycles degÿ1
908, 3.36 cycles degÿ1
Difficult
Figure 1. Visual search task. Stimuli occupied approximately 2.2 deg of visual angle at a viewing distance of 57 cm. The target object was used as a fixation point at the start of each trial. Easy (low similarity)
2500
Difficult (high similarity) Low AQ, target present Low AQ, target absent High AQ, target present High AQ, target absent
Mean reaction time=ms
2000
1500
1000
500
0 4
8 Display size
16
4
8 Display size
16
Figure 2. Visual-search performance: reaction times for correct responses are displayed with (for illustrative purposes) a median split based on AQ scores. Error bars represent 1 SE.
To test the hypothesised link between visual search and discrimination of perceptually similar objects, we measured discrimination thresholds for the relevant stimulus parameters (figure 3a). Thresholds were estimated separately for orientation and spatial frequency via a `2 down, 1 up' staircase method, with the final five reversals averaged (spatial frequency: mean 0:16 cycle degÿ1 , SD 0:08 cycle degÿ1 ; orientation: mean 2:448, SD 2:458). Composite z-transformed discrimination threshold (treated as the covariate) was not predictive of visual-search response times (F1, 36 1:38, p 0:247) and there were no significant two-way interactions with task manipulations (Fs 5 1). Thus, we found no evidence linking performance on visual-search tasks to the ability to make fine-grained discriminations along the stimulus dimensions that differentiated targets from distractors.
Visual search and autistic traits
741
fixation s 500 m
stimuli 500 ms
response
(a)
Spatial frequency
Orientation
fixation 500 ms
stimuli
ble
varia
mask/ response
(b)
Spatial frequency
Orientation
Figure 3. (a) Discrimination and (b) inspection-time tasks. Participants viewed a display containing the visual-search target and a single distractor with either higher spatial frequency or oblique orientation. They then indicated by keypress whether the target was on the left or right of the display. Practice trials reinforcing the target identity were completed prior to each task.
To test the alternative hypothesis that visual search is linked to speed of processing, participants completed two inspection-time tasks (figure 3b) that were similar to the corresponding versions of the discrimination task but with the identity of the distractors held constant at the stimulus parameters of the easy visual-search task. Stimuli were backward-masked with a composite of the target and distractor, and inspection time was varied, with thresholds determined by the staircase method described above (spatial frequency: mean 33:2 ms, SD 25:7 ms; orientation: mean 79:9 ms, SD 40:3 ms). ANCOVA revealed that participants with shorter inspection times (average of z-transformed thresholds) gave quicker visual-search responses (F1, 36 6:94, p 0:012) and demonstrated a marginally reduced effect of display size (F1:2, 42:9 3:56, p 0:059).(1) (1) As noted by one reviewer, in the orientation version of the task the stimulus ^ mask transition could result in perception of apparent rotation. However, this additional cue should weaken any association between inspection-time thresholds and other measures (Mackenzie and Bingham 1985). Indeed, when inspection-time threshold was measured with the use of the spatial-frequency version alone, the main effect of inspection time (F1, 36 12:88, p 0:001) and the inspection time by display size interaction (F1:2, 43:4 5:03, p 0:009) were both strengthened.
742
J Brock, J Y Xu, K R Brooks
These findings may have important implications for the understanding of visual perception in autism. In particular, the conventional practice of controlling for IQ or mental age in autism research leaves open the possibility of confounding differences in speed of processing that could explain superior visual search. Speed of processing may not, however, be the whole story. Further analyses revealed that AQ scores were not directly correlated with inspection-time thresholds (r38 ÿ0:114, p 0:495) or indeed with discrimination thresholds (r38 ÿ0:053, p 0:754). Thus, the link between visual search and sub-clinical autism traits appears to be mediated by some other cognitive variable that was not measured in the current study. To the extent that AQ scores are a proxy for autism, speed of processing provides only a partial explanation for group differences in autism studies. Nonetheless, in the light of the current results, future studies of autism should consider the extent to which variation in visual-search performance is determined by individual differences in speed of processing, before seeking an explanation for any residual autistic advantage. References Almeida R A, Dickinson J E, Maybery M T, Badcock J C, Badcock D R, 2010 ``A new step towards understanding Embedded Figures Test performance in the autism spectrum: the radial frequency search task'' Neuropsychologia 48 374 ^ 381 Baron-Cohen S, Wheelwright S, Skinner R, Martin J, Clubley E, 2001 ``The Autism-Spectrum Quotient (AQ): Evidence from Asperger Syndrome/high-functioning autism, males and females, scientist and mathematicians'' Journal of Autism and Developmental Disorders 31 5 ^ 17 Franklin A, Sowden P, Burley R, Notman L, Alder E, 2008 ``Color perception in children with autism'' Journal of Autism and Developmental Disorders 38 1837 ^ 1847 Joseph R M, Keehn B, Connolly C, Wolfe J W, Horowitz T S, 2009 ``Why is visual search superior in autism spectrum disorder?'' Developmental Science 12 1083 ^ 1096 Mackenzie B, Bingham E, 1985 ``IQ, inspection time and response strategies in a university population'' Australian Journal of Psychology 37 257 ^ 268 O'Riordan M, Plaisted K, 2001 ``Enhanced discrimination in autism'' Quarterly Journal of Experimental Psychology A 54 961 ^ 979 Plaisted K, O'Riordan M, Baron-Cohen S, 1998 ``Enhanced visual search for a conjunctive target in autism'' Journal of Child Psychology and Psychiatry 39 777 ^ 783 Scheuffgen K, Happe¨ F, Anderson M, Frith U, 2000 ``High intelligence, low IQ? Speed of processing and measured IQ in children with autism'' Development and Psychopathology 12 83 ^ 90 Simmons D R, Robertson A E, McKay L S, Toal E, McAleer P, Pollick F E, 2009 ``Vision in autism spectrum disorders'' Vision Research 49 2705 ^ 2739
ß 2011 a Pion publication
ISSN 0301-0066 (print)
ISSN 1468-4233 (electronic)
www.perceptionweb.com
Conditions of use. This article may be downloaded from the Perception website for personal research by members of subscribing organisations. Authors are entitled to distribute their own article (in printed form or by e-mail) to up to 50 people. This PDF may not be placed on any website (or other online distribution system) without permission of the publisher.