This dissertation proposes perceptual engineering as a new solution to longstanding challenges in sensory interface design. Perceptual engineering is a computational approach that systematically modulates human sensory mechanisms (such as thermal receptors, trigeminal nerve endings, and taste buds) to alter perception in controlled and reproducible ways. As computing moves beyond screens and into the body, designers face persistent limitations in engaging the full range of human senses. Modalities like smell, taste, temperature, and subtle touch remain underexplored in interactive systems due to their high power requirements, difficulty of miniaturization, and lack of sensory specificity. Perceptual engineering addresses these issues by targeting the physiological roots of perception, offering precise, low-power methods to create embodied sensory experiences. As the first technical instantiation of this framework, I introduce chemical interfaces: a new class of wearable systems that produce targeted feedback by interacting directly with the body’s chemical pathways. These systems exploit the potency and selectivity of safe chemical compounds to evoke haptic and thermal sensations as well as modulate gustatory sensations with minimal hardware. I demonstrate how chemical interfaces enable diverse applications: from delivering temperature illusions through trigeminal stimulation, to creating multi-sensation haptic feedback with a single actuator, to selectively suppressing or transforming specific tastes to promote healthier eating. Because they operate efficiently and can be miniaturized, chemical interfaces offer a practical path for integrating rich sensory modalities into everyday and immersive technologies. Building on this foundation, I extend perceptual engineering to include non-chemical forms of stimulation that share the same principles of direct, physiological modulation. For instance, I developed a stereo-smell interface that uses electrical stimulation of the nasal septum to generate lateralized chemesthetic sensations (subset of smell sensations), allowing users to localize virtual odors. I also designed a system that applies thermal feedback to the nose during inhalation, creating the illusions of inhaling more or less air—without changing actual airflow. These systems support new applications ranging from gas localization for safety (e.g., detecting odorless carbon monoxide leaks) and smell-based navigation to anxiety regulation and improved mask comfort. Together, these contributions define perceptual engineering as a generalizable foundation for future sensory interfaces. Unlike traditional approaches that simulate perception through audiovisual substitution or mechanical actuation, perceptual engineering enables compact, energy-efficient systems that engage the sensory body directly. This work lays the groundwork for interfaces that not only render sensation but give users agency over it: systems that allow people to actively shape what they feel, taste, or smell in response to context, preference, or need. Ultimately, this dissertation advances a broader vision: a future in which modifying perception (whether to improve health, enhance comfort, or deepen immersion) is as seamless and accessible as adjusting a phone's settings.