A small team of independent researchers is trying to fill in the long-missing "olfactory link" for virtual reality. However, the path they chose is not to emit smells in the air, but to "write" smell perception directly in the brain. The prototype device developed by the team uses focused ultrasound to accurately stimulate the olfactory bulb responsible for processing olfactory information without the need for any chemicals, odor boxes or injection devices. If subsequent verification is feasible, it is expected to bring a new immersive experience to virtual and augmented reality.
The current immersive system mainly revolves around vision and hearing, and there has been some progress in tactile feedback. However, the sense of smell, which is most closely related to memory and emotion, is still almost absent. Biologically speaking, olfactory signals will directly enter the limbic system, including the hippocampus, without first going through higher-level cortical processing. This special pathway is considered to be the key mechanism of "smell brings back memories" and is also an effect that is difficult to replicate with existing VR.
Research team members Lev Chizhov, Albert Yan-Huang, Thomas Ribeiro and Aayush Gupta decided to abandon the traditional route of reconstructing smells in the air and instead directly stimulate the olfactory bulb area in the brain through ultrasound. They said that there were few previous attempts to use ultrasound to directly induce odor perception in living bodies, even in animal models, so this direction is quite exploratory in terms of technology.
The olfactory bulb is located above the nasal cavity, is deeply located, and is wrapped by bones and soft tissue. It is not easy to access from outside the head. At the same time, ultrasound propagates poorly in the air, which makes precise targeting more challenging. To do this, the researchers fixed the ultrasound transducer on the forehead, using what they describe as a "solid, jelly-like pad" to provide support and comfort, and then tilted the ultrasound beam downward toward the target area.
The team used magnetic resonance imaging (MRI) data from one of the researchers to estimate the approximate coordinates and depth of the olfactory bulb to determine the location of the ultrasound focus. On this basis, they repeatedly adjusted the ultrasound frequency and pulse timing to find a parameter combination that could both penetrate the skull and focus energy at the target depth to obtain relatively stable subjective feedback.
During the experiment, subjects reported a range of experiences that ranged between clear smells and vague sensations, including fresh air, ozone, burning wood, and decaying organic matter. The researchers noticed that there is a relatively clear difference between "smell" and "feeling": the former has a clearer outline and seems to have a specific source point, as if the direction can be locked by "sniffing"; the latter is weaker and slower, and is mostly described as a vague impression rather than an identifiable specific smell.
Some subjects also reported mild physical sensations, such as a subtle itching or tingling sensation on the face, suggesting that ultrasound stimulation not only acts on the olfactory pathway but may also involve peripheral sensations. How you breathe also affects the intensity of the experience: a gentle inhale tends to amplify this sense of smell or perception, so participants were asked to "sniff" slightly while holding the device to their forehead.
In some trials, the associated sensations accumulated gradually over several breaths, while in others they appeared almost suddenly. When some subjects experienced a smell similar to rotting garbage, they subconsciously regarded it as a real smell in the environment, showing an intuitive reaction similar to "mistaking virtuality for reality."
From an engineering perspective, this is still an early prototype: the device barely qualifies as a "head-mounted" device, but currently it must be fixed on the forehead by hand. To achieve practical applications, the device needs to be further miniaturized and deeply integrated with wearable hardware such as VR/AR headsets to meet long-term wear, mobile use and safety requirements.
The potential implications of this research may go beyond "virtual smells." It points to a broader direction: "writing signals to the brain" through non-invasive technology without the need for craniotomy or implanted electrodes, rather than just reading passive information such as brain electricity or blood flow changes. This prospect remains highly speculative at this time, but in theory, similar approaches might be extended to other senses and perceptual pathways beyond smell.
In terms of foreseeable short- and medium-term applications, immersive media is the most direct landing place: if the headset can generate an "in-brain sense of smell" without relying on consumable chemicals and odor cartridges, it will eliminate a long-standing limitation in virtual scene design. Of course, to truly enter the consumer market, we still face multiple engineering challenges such as cost control, volume and weight, safety supervision, and experience consistency. Therefore, it is more likely to be implemented on enterprise-level training, professional simulation, and scientific research platforms first.
Looking at a deeper level, this path that bypasses physical air and chemical molecules and directly reaches the olfactory center has changed people's traditional imagination of "digital smell". It does not recreate various fragrances or odors in the real space, but attempts to trigger the brain's subjective perception of "smelling a certain smell" at the neural level. Once this idea matures, it may open up a new technical route in the fields of perceptual computing and human-computer interaction.