The Perception of Integrated Events in Autism Spectrum Disorders: The Role of Semantic Relatedness and Timing

AbstractAutism Spectrum Disorders (ASD) has been associated with impaired multisensory processing, however research on the topic has been inconclusive. For instance, research on the synchrony perception of complex stimuli has shown that children with ASD have impaired integration of audiovisual speech (e.g., a woman counting or telling a story) but normal non speech integration (e.g., a ball moving through a series of plastic ramps and cliffs; Bebko et al., 2006), while research on adult ASD patients has shown impaired integration for both speech (e.g., syllables) and non speech events (e.g., flash-beeps, handclap; de Boer- Schellekens et al., 2013). Studies utilizing simple stimuli and illusory paradigms such as the double-flash illusion have suggested that individuals with ASD exhibit an extended temporal integration window as compared to healthy participants (Foss-Feig et al., 2010; Kwakye et al., 2011), while others have shown no such effect (e.g., Van der Smagt et al., 2007). It is as yet unclear, therefore, whether or not individuals with ASD have impaired integration mechanisms and whether this impairment is due to the stimuli presented (simple vs. complex; social vs. non social), the population used (adult vs. children, severity of symptoms), and/or the tasks utilized (e.g., preferential looking paradigm vs. temporal order judgments). Additionally, it is as yet unclear whether individuals with ASD are impaired in terms of timing or binding per se (e.g., Freeman et al., 2013). In order to elucidate this issue, we aim to further examine the nature of these deficits through a well-formed group of children with similar symptom severity and two types of tasks. Initially, using a reaction time (RT) task, we will assess the audiovisual integration capabilities of children with ASD as compared to typically developing (TD) children without the involvement of timing (i.e., no timing differences will be introduced). According to the ‘unity effect’, a multisensory event is perceived as an integrated multisensory event (rather than multiple unimodal events) when signals are present close in time and space and due to other factors (e.g., informational relatedness; e.g., Vatakis & Spence, 2007). In the RT experiment, therefore, we will not modulate space and time, but informational relatedness. Specifically, participants will be asked to complete speeded detection of two targets. The targets will be audiovisual, visual, and auditory and for the audiovisual cases the steams will be presented in congruent and incongruent format.Subsequently, the same group of individuals will be tested in a simultaneity judgment (SJ) task where the temporal window of integration will be assessed. The SJ task will target the evaluation of temporal processing. For both tasks, three types of stimuli will be used: a) simple stimuli in order to minimize the meaningful context (Bien et al., 2013), b) stimuli with emotional context depicted through human faces in order to assess ASD processing of facial social stimuli, and c) stimuli with emotional context using body expressions (instead of faces). The use of the RT and SJ tasks we allow the evaluation of the interaction of multisensory integration and synchrony perception in ASD as well as the role of stimulus type (semantic vs. non semantic, social vs. non social) in multisensory processing