Many individuals living with autism spectrum disorders are non-verbal. That is, they cannot functionally communicate with others using their voice. Despite this substantial fraction, we still know very little about these individuals, their abilities, and their needs. Non-verbal individuals are nearly always presumed to have low intellectual capabilities; however, in some of these cases, the ability to produce language may be quite separate from the ability to understand and perceive language. In fact, there are examples of non-verbal individuals who have demonstrated their capacity for language comprehension by learning to type. The investigators in this proposal will use cognitive event-related brain potentials (ERPs) recorded in a structured protocol to evaluate cognitive function in non-verbal individuals with autism, including individuals with autism who use alternate means of communication. These methods were originally developed for assessing brain-injured people who have received diagnoses of "vegetative state" and "locked-in" syndrome, and are expected to provide a rigorous means of demonstrating speech comprehension at different levels of sophistication and related cognitive functions. What this means for people with autism: This project is designed to learn about the cognitive capabilities of nonverbal people with autism. If successful, this study will provide a method for the individualized assessment of language comprehension ability in non-verbal individuals with autism. This assessment can help direct educational and treatment resources so non-verbal individuals will receive targeted interventions to better help them learn to communicate and achieve the best outcome.