Abstract
Background
Worldwide 45 million people are blind. Sensory substitution may restore visual function by encoding visual information into a signal perceived by another still-functional sensory modality, such as audition. The vOICe device, for example, captures a video image and encodes vertical position as distinct frequencies, horizontal position as scan time (left-to-right), and the brightness of individual pixels as volume (Amedi, 2007)
Method
Behavioral tests were devised to monitor vision-like processing development during training on the vOICe device. Sighted subjects performed "visual" tasks while wearing glasses with small-embedded camera, attached earphones, and a connected portable computer.
fMRI was performed in a 3 Tesla magnet with pre-recorded vOICe signals.
Results
During the localization task, a white circle is placed in one of five locations on a black felt covered board and the subject points to the object’s center. During a training period of 4.5-6 hours four sighted subjects improved their reach inaccuracy from an average of 6.79 inches to an average of 3.85 inches. A forced-choice recognition task of 4-5 office objects showed an accuracy (average 95 %), which was well-above chance (average 21.25 %).
A study of mapping visual space to visual activation (via vOICe encoding) using fMRI has also been performed. Preliminary results (N=1) show bilateral activation of V3 (BA 19) when a visual dot in the left visual field is encoded in vOICe and localized by the subject and left V3 activation when the dot is on the right.
Discussion
Behavioral results indicate that visual processing skills can be acquired with vOICe device use. The (prelim.) fMRI results were also consistent in that the vOICe localization
training develops vision-style sensory processing from the auditory inputs.
Meeting abstract presented at VSS 2012