The accuracy of auditory spatial judgments in the visually impaired is dependent on sound source distance

Abstract

Blindness leads to substantial enhancements in many auditory abilities, and deficits in others. It is unknown how severe visual losses need to be before changes in auditory abilities occur, or whether the relationship between severity of visual loss and changes in auditory abilities is proportional and systematic. Here we show that greater severity of visual loss is associated with increased auditory judgments of distance and room size. on average participants with severe visual losses perceived sounds to be twice as far away, and rooms to be three times larger, than sighted controls. Distance estimates for sighted controls were most accurate for closer sounds and least accurate for farther sounds. As the severity of visual impairment increased, accuracy decreased for closer sounds and increased for farther sounds. However, it is for closer sounds that accurate judgments are needed to guide rapid motor responses to auditory events, e.g. planning a safe path through a busy street to avoid collisions with other people, and falls. Interestingly, greater visual impairment severity was associated with more accurate room size estimates. the results support a new hypothesis that crossmodal calibration of audition by vision depends on the severity of visual loss.The World Health Organization (WHO) estimated that globally 188.5 million people have mild visual loss, 217 million have moderate to severe losses, and 36 million are blind 1 . It has now been well established that full blindness (total visual loss or light perception only) can result in enhancement of certain auditory spatial abilities and worsening of others (for reviews, see 2-7 ). For example, blindness often results in dramatic improvements in echolocation skills 4,8 and the ability to locate sounds in azimuth (left-front-right judgments) 9,10 , but leads to significantly poorer ability to judge the vertical position of sounds 11,12 , or judge sound position with respect to external acoustic landmarks 13 . It has been suggested that the changes underlying enhanced performance are fundamentally related to adaptations within the occipital cortex, where visual areas of the brain are recruited to process auditory inputs in the event of visual loss 5,14,15 . However, the underlying principles of what drives changes in auditory abilities following visual loss are not well understood. It is not yet known how severe the visual loss needs to be before significant alterations in auditory abilities are observed, or whether the relationship between severity of visual loss and changes in auditory abilities is systematic. If it is systematic, then people with more modest visual losses should exhibit smaller changes in auditory abilities than those with more severe visual losses.Previous work showed that compared to sighted controls, individuals with total vision loss estimate near sound sources to be farther away and estimate farther sound sources to be closer 16,17 , demonstrating the critical role that vision plays in calibrating auditory space 18 . A number of studies ...

A framework to account for the effects of visual loss on human auditory abilities.

“…Després, Candas, and Dufour (2005c) showed that near-sighted and amblyopic participants performed better in a self-positioning task than normally sighted controls. Kolarik et al (2020) investigated the effect of severity of visual loss on auditory distance judgments using stimuli with simulated distances from 1.2 to 13.8 m. Sighted controls and participants with a range of visual losses (groups with mild, mid-range, and severe loss) were tested in simulated anechoic and reverberant environments using speech, music and noise stimuli.…”

“…For localization in azimuth, blindness in one eye is associated with enhancement (Hoover, et al, 2012), while central visual loss in both eyes is associated with degradation (Lessard, et al, 1998). Severe visual loss is associated with reduced accuracy in judging the distance of closer sounds and increased accuracy for farther sounds, reflecting systematic changes in the mapping between simulated and perceived distance (Kolarik, et al, 2020). Further studies are needed to clarify the effects of the type of visual loss on hearing, such as monocular blindness with one unimpaired eye or central or peripheral visual loss.…”

Partial visual loss disrupts the relationship between judged room size and sound source distance

“…For statistical analysis, data were grouped into near (1.22-1.72 m), middle (2.44-4.88 m), and far (6.90-13.79 m) distances. This procedure was similar to that used by Kolarik et al (2020), which followed the methods utilized by Voss et al (2015).…”

Factors Affecting Auditory Estimates of Virtual Room Size: Effects of Stimulus, Level, and Reverberation

“…These advances have enabled the use of virtualization techniques to explore a range of issues, including how reverberation affects speech understanding (Ellis & Zahorik, 2019), and how the availability of visual depth information increases the accuracy of auditory estimates of distance within a room (Anderson & Zahorik, 2014). How audition provides spatial information for judgments of the distance of sound sources (Bidart & Lavandier, 2016; Kolarik, Cirstea, & Pardhan, 2013a, 2013b; Zahorik, 2002) and room size (Kolarik, Pardhan, et al., 2013; Kolarik et al., 2020) has also been investigated using virtualization techniques. Audition provides valuable spatial information when vision is unavailable and is critical for spatial awareness and navigation by blind people.…”

“…The simulation methods have been described in our previous studies investigating auditory 171 judgments of distance (Kolarik et al 2013a, b;Kolarik et al 2013c;Kolarik et al 2017a) and room size (Kolarik et al 2013d;Kolarik et al 2020). In the current study, the virtualization was made more realistic by convolving sound reflections with the appropriate head-related transfer functions (HRTFs) in addition to doing this for the direct sound component (Culling et al 2013;Culling 2013;Moore et al 2016).…”