Research has demonstrated that early assessment of autism spectrum disorder (ASD) in toddlers enables earlier intervention, which enhances developmental gains for this population (Rogers et al., 2012; Vernon, Koegel, Dauterman, & Stolen, 2012). Eye-tracking technology has proven to be a particularly promising avenue for assessment of atypical social looking behaviors in young children with ASD, due to its non-intrusive yet rigorous method of data collection (Oakes, 2012). A growing body of eye-tracking literature demonstrates that when compared to typically developing controls, toddlers on the spectrum exhibit decreased attention towards socially and emotionally salient stimuli and increased fixation on non-social stimuli (Pierce et al., 2011; Shic, Bradshaw, Klin, Scassellati, & Chawarska, 2011). This ability of eye-tracking methodology to detect deviation away from a typical gaze pattern when presenting social and non-social stimuli, and even in doing so predict an autism diagnosis, justifies its increasing use as a method of autism assessment. However, there is currently little known regarding how looking preferences in specific eye-tracking paradigms relate to severity of autism social affect symptoms, or how they may relate to preferential gaze during naturalistic interactions with caregivers. The current study seeks to firstly assess replication of ASD preferential non-social looking behaviors with the use of an eye-tracking paradigm. This project also seeks to enhance the understanding of social attention in toddlers on the spectrum by investigating relationships between non-social eye-tracking looking behaviors and severity of autism symptoms, as well as with social gaze during naturalistic dyadic interactions.