Age Related Changes in Conjunctive Visual Search Among Children with and without ASD

Background:  There is evidence of enhanced visual search performance in people with ASD including toddlers (Kaldy et al. 2011), school age children (O’Riordan at al., 2001), and adults (O’Riordan, 2004). However, previous studies are limited by small sample sizes, using fewer trials than are typically used for visual search tasks and calculating reaction times (RT) based on means instead of medians (McDonald, 2010).

Objectives:  The objective of this study was to investigate age related changes in conjunctive visual search performance in a large sample of high functioning children and adolescents with ASD as compared to their age and IQ matched TD peers.

Methods:  Participants were between the ages of 7-20 years, and consisted of an ASD (n=49) and a TD (n=42) group. Participants with ASD were high functioning (IQ >80), and matched on IQ and age with the TD group.  A conjunctive visual search task was employed; the goal was to find the target (a red ‘X’) in a multi-element display with red ‘T’ and green ‘X’ distracters. It included two blocks of 60 trials (120 total) with two crossed factors: display size (5, 15, 25) and probe (target present or absent). Participants were instructed to press the ‘z’ key if the target was present, and the ‘.’ key if it was absent.

Results:  Analyses were conducted on median RTs of each participant at each display size. A repeated-measures ANOVA was conducted with a between-subject factor of group, and within-subject factors of target presence and set size. Contrary to expectations, there was no difference between the ASD and TD groups on either error or RT (all p>.05). The groups were analyzed separately using correlations, with Bonferroni correction. Slope was calculated for each participant using their median RT for each set size, and intercept was calculated using the x-intercept resulting from the slope calculation. Slope is a measure of efficient search strategy and intercept in a measure of processing speed. Age was significantly related to RT performance (measured by the intercept) in both the target present and absent conditions in both groups (p<.01); in both groups as participants got older they got significantly faster. Age was not related to slope in target absent or present condition. IQ was not related to any of the performance measures in either group.

Conclusions: Contrary to previous studies, we did not find any differences in performance on a conjunctive visual search task. This may be because this study included a much larger, more heterogeneous sample of individuals with ASD than previously reported studies. We found that similar factors contributed to visual search performance in both groups. Specifically, both groups showed faster search performance with age yet their search strategy did not change with age. The findings indicate that conjunctive visual search strategy and speed develop similarly in individuals with and without ASD and whereas visual search strategy is mature at age 7 years, the efficiency of search continues to improve until age 20 years.