Cochlear Implant Electrode
Cochlear Implant Electrode

Cybernetic senses have been the subject of science fiction for decades. The idea of using sophisticated technology to repair damaged bodies, or even to enhance normal ones, has a tremendous appeal – but how far have we progressed towards that goal?

In some ways, we’ve gotten amazingly far. Cochlear implants are now a normal – if controversial – treatment for deafness. They substitute for damaged or missing portions of the inner ear, gathering and processing sound. The first generation of cochlear implants provided only a distant approximation of sound, making them of limited usefulness, particularly for understanding speech sounds. Even the more sophisticated models of today have yet to approach the functionality of a normal ear, though they are far more useful than their predecessors.

While those dealing with cybernetic hearing seem to have decided upon their basic approach, dealing with lost vision is a different ball game. Several different research groups, using various methods, are attempting to produce cybernetic vision. Some, like cochlear implants, seek to replace a malfunctioning part. Others are attempting to produce something entirely new that will nonetheless function as vision. One of the projects that is furthest along is using exactly that kind of substitution. Rather than attempting to somehow re-engineer the eye, the vOICe system (Oh, I see) is using sound to bypass the eyes altogether and substituting the ears in their stead.

The vOICe system was developed by Dr. Peter Meijer – a senior researcher with Philip Research Laboratories (a Netherlands based company). It uses a computer program, a pair of video-sunglasses (sunglasses with a small video camera on the bridge), and a pair of stereo earphones to provide an auditory image of the world. Scanning left-to-right, once a second, the program translates the images seen by the camera into coded sound for the user to interpret. The format is fairly simple – louder sounds mean brighter, higher pitch means that something is higher up in the visual field, and so forth.

Learning to translate the sounds into visual meaning is another task altogether. Users of the device liken it to learning a foreign language. As with a foreign language, the more the vOICe system is used, the more quickly the user gains facility. Most users appear to start the learning process with a set of images on a computer screen (available as a free download off the Internet), and then progress to using the mobile system. The amount of information that can be distinguished even by a novice user is fairly impressive. Within a week of starting use, at least one congenitally blind woman was reporting being able to distinguish walls, stairs, and windows in her house, as well as whether the lights were on or not. The vOICe website reports that a trained user can distinguish approximately 1000 to 4000 pixels per 1 second scan. Comparatively speaking, an average sighted person can identify visual objects in an image of 32×32 pixels – or about 1024 pixels.

In what may be the most interesting part of the vOICe system, its constant use seems to cause a sort of induced synesthesia in the user – a cross-wiring of the senses, where input from one sense is perceived in another. Brain plasticity – the ability of the brain to rearrange itself in response to demand – seems to come into play, as the brain sorts auditory input into visual data. Some previously sighted vOICe users report consistent visual responses comparable to blurry or foggy vision, while their awareness of the sounds themselves recedes into the background. Users who have been blind from birth obviously cannot compare their experiences to a previous experience of vision, but they too seem to rapidly stop processing the vOICe input as auditory data.

32 x 32 pixel image
32 x 32 pixel image

While the results so far are exciting, there are a few downsides to the vOICe system. Since distinguishing the auditory landscapes requires good hearing, the system is not going to work well, if at all, for someone with any sort of hearing impairment. Additionally there is some concern that the headphones and the sounds produced by the system could interfere with normal hearing function while the system is in use. However, the difference in how the brain processes the vOICe data seems to enable less interference between the competing sounds than one might imagine, with some users reporting being able to use the system while sewing, listening to TV, or even listening to music.

Slow scanning speed is likely to be the most difficult problem to improve. While retinal or brain implants (neither yet available for general use, but under investigation) allow scanning the visual field between four and eight times a second, the vOICe scan is only once a second – a fairly slow pace for interacting with the world while moving, but dictated by the need for the user to process the sounds. The brain seems to be able to compensate, though, and the vOICe users have few complaints about slow updating speed.

The last major limitation is that of current technology. While the system is portable, the need to carry a laptop, and wear a video camera and earphones does use up some carrying capacity. The limitations of the laptop battery also make venturing forth for long periods of time problematic. These problems will ease with time though. Even in the short time the vOICe has been available, new and smaller devices have become available for each of the system components. A major plus for the system is the ease of upgrading as newer technology comes on-line. Since the system is physically separate from the user, upgrading is as simple as buying the new equipment and integrating it. Upgrading a retinal implant, let alone a chip in the brain, is a much trickier proposition.

As promising as it is, the vOICe system is far from the last word in cybernetically enhanced vision. Brain implants, retinal implants, and devices that translate visual imagery to touch are all under active investigation. Thus far vOICe seems to have a clear lead by utility, by its lack of invasiveness, and even by cost, but who’s to say which of the other contenders may surpass it tomorrow?

28 Comments