Understanding Time Estimation in Autism Spectrum Disorders

Background: Evidence that persons with autism spectrum disorders (ASD) are less able to accurately estimate temporal durations  is interpreted as indicating a widespread deficit in time perception (Allman et al. 2011; Maister & Plaisted-Grant, 2011; Martin et al. 2010). However, this evidence is only from tasks involving longer durations (>1000ms). Conversely, in an initial study with shorter durations, Mostofsky et al. (2000) found no estimation difficulties among high-functioning individuals with ASD. In order to extend Mostofsky et al.’s  findings, we studied time perception of a short duration among low- functioning children with ASD in relation to mental-age (MA) matched children and adolescents (youths) with Down syndrome (DS) and typically developing (TD) children.

Objectives: We extended the study of temporal perception among persons with ASD by examining the estimation of short durations in functioning children with ASD as compared to that of MA-matched youths with DS and TD children.

Methods: Currently, the groups (ASD, DS, TD) are each comprised of 10 participants and are matched on an approximate MA of 6 years.  Group matching was based on scores from the Leiter-R International Performance Scale Revised (Leiter-R; Roid & Miller, 1997) with p-values for group differences greater than 0.10 for all comparisons. The groups differed in mean age (DS = 15 years; ASD = 10 years; TD = 6 years). All participants completed a temporal bisection task using auditory stimuli. Standard durations of 200 and 800 ms were compared to a range of comparison durations from 200 to 800 ms.  After familiarization with the standard durations, participants were asked if each comparison tone was more like the short or the long tone over a series of trials. The primary dependent variable was the proportion of long responses made for each comparison duration condition.

Results: Preliminary analyses yielded a significant Group by Duration interaction, reflecting different patterns of responding across the groups. Participants with ASD reported more “long” responses at the shorter durations and fewer “long” responses at the longer durations than the other groups. The patterns of the DS and TD groups were similar, although youths with DS produced slightly fewer long responses in the longer duration condition. The shallow slopes of the functions produced by the children with ASD indicate a lack of sensitivity to duration differences. Psychophysical measures of sensitivity have been acquired (e.g., Weber fractions).

Conclusions: Contrary to Mostofsky et al.’s (2000) findings with high functioning persons with ASD, we found poorer sensitivity to the variability of durations under 1000ms among a group of low-functioning children with ASD. The ASD group, with an average MA of 5.9 years, produced bisection functions that suggest poorer sensitivity to time at these durations than was reported for TD children at 5 years of age (McCormack et al., 1999). However, the DS and TD groups produced functions consistent with their developmental level. This is preliminary evidence of an ASD specific deficit in time estimation.