ABSTRACT Repetitive behaviors comprise one of the two core domains of symptoms for an autism spectrum disorder diagnosis (hereafter ‘autism’), and the lived experience of autistic individuals suggests that these behaviors serve important functions that are largely misunderstood and poorly addressed clinically. Autistic individuals report engaging in repetitive behaviors for reasons related to enjoyment and reasons related to distress (i.e., to decrease stress and anxiety). The field lacks a validated model that encompasses these seemingly polarized lived experiences, which negatively impacts our ability to provide appropriate supports, and educate the public to better accommodate autistic individuals. The guiding scientific premise of this R21 is that autistic individuals are driven to engage in repetitive behaviors via both approach (enjoyment) and avoidance (avoiding/easing exposure to negative stimuli) mechanisms. The approach component of the model is supported by a burgeoning literature and our preliminary work demonstrating that measures from the Research Domain Criteria’s Positive Valence System are implicated for special interest repetitive behaviors. The avoidance component of the model is supported by a robust literature and our preliminary data associating sensory reactivity and insistence on sameness repetitive behaviors with measures from the Research Domain Criteria’s Negative Valence System. The objective of this project is to take the first step in evaluating the approach-avoidance model of repetitive behaviors in a study with autistic adolescents. Adolescence is a developmental period when many autistic youth can reflect on the drivers of their repetitive behaviors. Aim 1 will recruit 250 autistic adolescents between the ages of 13 and 17 and a caregiver to provide ratings on the type of repetitive behaviors they engage in, approach and/or avoidance motivations for engaging in the behavior(s), and the intensity of the emotional valence associated with each repetitive behavior. We will derive latent profiles of repetitive behaviors as a novel method for parsing the heterogeneous clinical presentations of repetitive behaviors. Aim 2 will recruit 60 teens from Aim 1 to complete neuroimaging and behavioral protocols associated with the Research Domain Criteria’s Positive Valence (approach), and Negative Valence systems (avoidance); the goal is to link these dimensional assessments of approach and avoidance mechanisms with self-report of valence direction and intensity of repetitive behaviors. We hypothesize that Aim 1 will yield 3 unique latent profiles—Predominantly Approach, Predominantly Avoidant, and Mixed Approach/Avoidant—and that Aim 2’s canonical correlation analyses will show the Positive Valence System to be associated with approach repetitive behaviors and Negative Valence System to be associated with avoidance repetitive behaviors. This line of research has the potential to redefine models of repetitive behavior to align ...