Perception of the Sound Induced Flash Illusion in Adolescents with Autism Spectrum Disorders

Background:

Autism Spectrum Disorders (ASD) are complex neurodevelopmental disorders characterized by the presence of restricted or repetitive interests and impairments in communication and social or reciprocal behaviors.  In addition to deficits in this triad of domains, there have been numerous reports of sensory deficits spanning multiple modalities in ASD. These reports have prompted investigations into better characterizing the nature and role of sensory and multisensory processing deficits in autism.

Objectives:  

There has been a growing literature demonstrating atypical multisensory processing in individuals with ASD. One effective way of measuring multisensory integration is by using cross-modal illusions, such as the sound induced flash illusion (SIFI). In this illusion, when a single visual flash is presented in close temporal proximity to multiple auditory beeps, individuals often perceive multiple flashes. The basis of this illusion is the “binding” of the visual and auditory signals into a unified percept. The goals of the current study were to determine how ASD and typically-developing (TD) individuals perceive this illusion to probe for potential deficits in multisensory processing. Our hypotheses were that individuals with ASD would be less susceptible to this illusion due to deficits in multisensory binding processes, and that these differences would not be a consequence of changes in unisensory (i.e., visual alone, auditory alone) processing.  

Methods:  

32 TD individuals and 32 high functioning 6-18 year old individuals with ASD (2-subtest IQ: 115.7 (12.1) & 111.1 (16.1); p > 0.05) completed a sound induced flash illusion task where one visual flash was presented along with 0-4 auditory beeps. Control trials included 1-4 visual flashes presented with no auditory beeps. Participants were asked to ignore the beeps and report how many flashes they perceived. Individuals also completed unisensory auditory and visual temporal order judgment (TOJ) tasks in which they report which auditory or visual stimulus occurred first.

Results:  

Perception of the illusion differed between groups. For illusory conditions, individuals with ASD were less likely to report multiple flashes than TD individuals (p < 0.008 for 2, 3 & 4 auditory beep conditions), suggesting a difference in the magnitude of multisensory binding. In contrast, there were no significant differences between groups for the control trials (p > 0.2 for 0 auditory beep conditions), or for performance on the unisensory auditory and visual TOJ tasks (p > 0.05).

Conclusions:  

These results demonstrate that individuals with ASD perceive the sound induced flash illusion less often than TD individuals. This was not explained by differences in understanding or following directions or due to a bias to report a greater number of perceived flashes. This effect was specific to multisensory processing, since the unisensory tasks produced no significant differences. These preferential changes in multisensory processing for low-level visual and auditory stimuli suggest changes in neural networks responsible for multisensory binding in ASD. Furthermore, these results also suggest that deficits in early multisensory processes and networks may play an important contributory role in the higher-order domains known to be impacted in autism (e.g., communication, social interactions).