MIT's 9.35 Perception course, taught by Josh McDermott, examines how the auditory system reconstructs distinct sound sources from a single incoming waveform. McDermott covers the core problem of auditory scene analysis: the ear receives one mixed signal, yet listeners routinely separate it into speech, music, footsteps, or other overlapping sources. The lecture works through the cues the brain uses to group and segregate acoustic energy, such as onset timing, harmonicity, and spatial location, and connects these grouping principles to classic demonstrations in the perception literature. Runtime runs roughly eighty minutes, consistent with a full lecture session, and the material builds on earlier sessions in the course sequence. This is lecture five of the Spring 2024 offering, part of MIT OpenCourseWare's published recordings for the course.