Note: Most Internet Explorer 8 users encounter issues playing the presentation videos. Please update your browser or use a different one if available.

Tonic and Phasic Saccadic Reaction Time Differences in Infants At High Risk for ASD

Saturday, 4 May 2013: 09:00-13:00
Banquet Hall (Kursaal Centre)
E. J. Jones1, S. Wass1, T. Charman2, M. H. Johnson3 and *. the BASIS Team1, (1)Birkbeck, University of London, London, United Kingdom, (2)Centre for Research in Autism & Education, Institute of Education, London, United Kingdom, (3)Centre for Brain & Cognitive Development, Birkbeck, University of London, London, United Kingdom

Visual attention is the primary means of environmental exploration for young infants.  Orienting to stimuli is a key component of visual attention that has been linked to the development of self-regulation (e.g. Posner & Rothbart, 1998), and early atypicalities in visual orienting may therefore have pervasive effects on later development.  Studies of visual orienting in high-risk infants have revealed that such atypicalities may be a risk factor for the development of autism (Elsabbagh et al., submitted; Zwaigenbaum et al., 2005).  This work has traditionally examined mean reaction times derived from a series of trials. However, recent work has shown that examining intra-individual variability can provide additional insight into mean-based effects (Milne et al., 2011; Dinstein et al., 2012).  To further characterize visual orienting atypicalities in the early development of autism, in this study we explored intra-individual variability across an attention-shifting paradigm in infants with older siblings with autism.


To examine within-task variability in attention-shifting in 7- and 14-month-old infants at familial high-risk for autism.


Participants were infants with older siblings with autism (‘high-risk’; n=53) or typical development (‘low-risk’; n=27).  At 8- and 14-months, infants completed the gap-overlap task, a computer-based test of visual attention shifting.  Current analyses focused on latency to saccade measured in two conditions: Gap, in which the central stimulus disappears 200ms before peripheral stimulus onset; and Baseline, in which the central stimulus disappears simultaneously with peripheral stimulus onset. We examined i) mean reaction times; ii) intra-individual variability using ex-Gaussian modelling; iii) change in performance across the four blocks of the task.


  1. Mean comparisons:  Mean baseline saccadic reaction times were significantly faster in high-risk than low-risk infants at 8 but not 14 months; mean saccadic reaction times in gap trials did not significantly differ by risk group at either age. 
  2. Intra-individual variability: Ex-Gaussian modeling of individual reaction times indicated that that the reaction-time distribution had a longer tail in the low-risk group; there were more trials with reaction times >1200ms in the low-risk group at both 8 and 14 months.
  3. Changes with time: At both 8m and 14m, group differences in baseline RTs were most pronounced in the last block of the paradigm. At 14m, this was driven by slower reaction times in the second vs first half of the paradigm in low-risk but not high-risk infants. In gap trials administered later in the paradigm, at 14m high-risk infants produced greater numbers of very rapid (<200ms) saccades than low-risk infants.


High-risk infants showed shorter overall saccadic reaction times to the appearance of stimuli in non-competition conditions and fewer prolonged reaction times associated with ‘zoning out’ at both 8 and 14m; and at 14m in later blocks showed no slowing of baseline reaction times and more very rapid saccades in the gap condition.  This pattern of findings is consistent with the suggestion that high-risk infants maintain a heightened vigilance to their environment.   Talk 3 will present further data consistent with this proposal.

| More