**5. Blindness duration and cortical reorganization**

The lack of vision seems to interfere with the development of space representation, so an interesting question is: what happens when the subject loses visual input later in life? Late blindness is a condition worthy of investigation concerning this issue because spatial hearing of late blind subjects is shaped by unique combination of visual calibration in childhood and prolonged blindness in adulthood. As well as for early blind individuals, research on late blind individuals shows contrasting results at both the behavioral and cortical levels. For example, scientific evidence shows that, late blind individuals are better compared to sighted people in using spectral cues when they localize sound position in peripheral regions [47, 48]. Similar to early blind participants, they also show auditory and tactile recruitment of occipital regions [49–51]. Voss and colleagues [43] investigated the effect of blindness on brain activity by using positron emission tomography (PET) during one binaural and one monaural sound source discrimination task (SSDT) in early and late-onset blind individuals. In their study, Voss et al. observed that no difference was present between groups for the binaural task. Contrarily, during the monaural condition, early blind individuals performed significantly better than all the other groups (in agreement with the behavioral study of Lessard and colleagues [10]). Late blind subjects are more similar to sighted individuals concerning other skills too, such as absolute auditory distance estimation [27], locational judgments after a perspective change in small-scale space [52], audio shape recognition and navigation tasks [53]. In a recent study from our group [54], late blind individuals were involved to allow the study to investigate how blindness duration (BD) affects auditory spatial bisection skills and neural correlates. In late blind individuals, we replicated the same behavioral and EEG experiment previously performed among early blind people (see section above, [46]). We observed that the early (50–90 ms) ERP response, previously observed in sighted [45] and not in early

**291**

*Audio Cortical Processing in Blind Individuals DOI: http://dx.doi.org/10.5772/intechopen.88826*

longer period of blindness.

**Figure 5.**

in the brain [5, 30, 31, 53, 55].

**6. Time to infer space in blindness**

blind [46] individuals, is dependent on the amount of time spent without vision (i.e., BD, [54]). In particular, we observed that a shorter period of blindness links to stronger contralateral activation in the visual cortex (see **Figure 5**) and better performance during spatial bisection tasks. Contrarily, we observed non-lateralized visual activation and lower performance in individuals who had experienced a

*Results of linear regression analyses in late blind individuals. Years of blindness duration (BD) negatively correlate with lateralized (i.e., contralateral-ipsilateral to S2 position) ERP amplitude in a 50- to 90-ms time* 

*window after S2 for the spatial bisection task. With permission from Amadeo et al. [54].*

Time spent without vision seems to gradually impact neural circuits underlying the construction of space representation in late blind participants. On the one hand, duration of blindness directly impacts both neural and behavioral correlates of late blind individuals during auditory spatial processing similarity between neural circuits and competences of late blind individuals with short blindness duration and, confirming there is indeed a key relationship between visual deprivation and auditory spatial abilities in humans. On the other hand, the sighted people suggest that an early visual experience is necessary and sufficient to fully develop neural areas involved in complex representations of space. These results agree with previous works in animals showing visual information during the first years of life is essential toward calibrating auditory space representation

Almost 100 years ago, Piaget [56] stated that the temporal metric is strictly related to spatial metric development: "Space is a still of time, while time is space in motion" [57]. What Piaget did not discuss is the role of different sensory modalities in this link. Visual experience is important for the development of spatial metric representations, such as for bisecting sounds. Starting from Piaget's idea, one might hypothesize that when vision is unavailable, such as in the case of blindness, temporal representation of events can set spatial representation. Indeed, while early and late blind individuals with long blindness duration show strong deficits in terms of spatial bisection tasks, they show performance and cortical activations similar to sighted individuals in the time domain, such as in a temporal bisection task [45]. In support

*Audio Cortical Processing in Blind Individuals DOI: http://dx.doi.org/10.5772/intechopen.88826*

#### **Figure 5.**

*Visual Impairment and Blindness - What We Know and What We Have to Know*

seems to impact the development of this processing and underlying neural circuits, thereby impairing understanding of Euclidean relationships, such as those involved in solving a spatial bisection task. These findings agree with our previous behavioral results [30], at the same time revealing that the neural correlates of the audio space bisection deficit reported in blind individuals might correspond to reduction of early occipital contralateral activation. We speculate that cortical activation underlying the C1 ERP component (usually elicited by visual stimuli) plays a fundamental role in the construction of metrics in the spatial domain independently of the involved sensory modality. Moreover, the construction of spatial metrics may

*Average source activity within the selected time window (50–90 ms) compared between sighted and blind subjects. Left and right panels of the figure report the conditions in which S2 was presented from either the left (i.e., −4.5°, narrow first distance) or the right side (i.e., +4.5°, wide first distance), respectively. We report results of paired two tailed* t *tests with the scale in terms of t-statistic. We also display significant values of t statistic: reddish and bluish colors indicate stronger activations in sighted and early blind subjects, respectively, while intensity indicates magnitude of t (i.e., strength of difference). Only t values corresponding to* p *< 0.0001* 

The lack of vision seems to interfere with the development of space representa-

tion, so an interesting question is: what happens when the subject loses visual input later in life? Late blindness is a condition worthy of investigation concerning this issue because spatial hearing of late blind subjects is shaped by unique combination of visual calibration in childhood and prolonged blindness in adulthood. As well as for early blind individuals, research on late blind individuals shows contrasting results at both the behavioral and cortical levels. For example, scientific evidence shows that, late blind individuals are better compared to sighted people in using spectral cues when they localize sound position in peripheral regions [47, 48]. Similar to early blind participants, they also show auditory and tactile recruitment of occipital regions [49–51]. Voss and colleagues [43] investigated the effect of blindness on brain activity by using positron emission tomography (PET) during one binaural and one monaural sound source discrimination task (SSDT) in early and late-onset blind individuals. In their study, Voss et al. observed that no difference was present between groups for the binaural task. Contrarily, during the monaural condition, early blind individuals performed significantly better than all the other groups (in agreement with the behavioral study of Lessard and colleagues [10]). Late blind subjects are more similar to sighted individuals concerning other skills too, such as absolute auditory distance estimation [27], locational judgments after a perspective change in small-scale space [52], audio shape recognition and navigation tasks [53]. In a recent study from our group [54], late blind individuals were involved to allow the study to investigate how blindness duration (BD) affects auditory spatial bisection skills and neural correlates. In late blind individuals, we replicated the same behavioral and EEG experiment previously performed among early blind people (see section above, [46]). We observed that the early (50–90 ms) ERP response, previously observed in sighted [45] and not in early

**290**

depend on visual experience.

**Figure 4.**

**5. Blindness duration and cortical reorganization**

*after FDR correction appear. Adapted with permission from Campus et al. [46].*

*Results of linear regression analyses in late blind individuals. Years of blindness duration (BD) negatively correlate with lateralized (i.e., contralateral-ipsilateral to S2 position) ERP amplitude in a 50- to 90-ms time window after S2 for the spatial bisection task. With permission from Amadeo et al. [54].*

blind [46] individuals, is dependent on the amount of time spent without vision (i.e., BD, [54]). In particular, we observed that a shorter period of blindness links to stronger contralateral activation in the visual cortex (see **Figure 5**) and better performance during spatial bisection tasks. Contrarily, we observed non-lateralized visual activation and lower performance in individuals who had experienced a longer period of blindness.

Time spent without vision seems to gradually impact neural circuits underlying the construction of space representation in late blind participants. On the one hand, duration of blindness directly impacts both neural and behavioral correlates of late blind individuals during auditory spatial processing similarity between neural circuits and competences of late blind individuals with short blindness duration and, confirming there is indeed a key relationship between visual deprivation and auditory spatial abilities in humans. On the other hand, the sighted people suggest that an early visual experience is necessary and sufficient to fully develop neural areas involved in complex representations of space. These results agree with previous works in animals showing visual information during the first years of life is essential toward calibrating auditory space representation in the brain [5, 30, 31, 53, 55].

#### **6. Time to infer space in blindness**

Almost 100 years ago, Piaget [56] stated that the temporal metric is strictly related to spatial metric development: "Space is a still of time, while time is space in motion" [57]. What Piaget did not discuss is the role of different sensory modalities in this link. Visual experience is important for the development of spatial metric representations, such as for bisecting sounds. Starting from Piaget's idea, one might hypothesize that when vision is unavailable, such as in the case of blindness, temporal representation of events can set spatial representation. Indeed, while early and late blind individuals with long blindness duration show strong deficits in terms of spatial bisection tasks, they show performance and cortical activations similar to sighted individuals in the time domain, such as in a temporal bisection task [45]. In support

of this hypothesis, we recently tested and verified that space representation of blind individuals is strongly influenced by the temporal representation of events [58]. We performed different versions of the spatial bisection task in sighted and blind individuals, in which we presented spatial and temporal independent, coherent, and conflicting information (**Figure 6** top panel). Similar to the original version of the bisection task [30], in one condition the temporal delay between the three sounds was always the same, and only spatial cues were relevant to compute the task (i.e., Equal bisection, **Figure 6A** top panel). In other conditions instead, we presented a spatio-temporal coherent or conflicting information. For example, in the coherent bisection, a longer spatial distance between the first and the second sound was associated with a longer temporal delay between the two sounds, and the reverse was the case for shorter distances (see **Figure 6B** top panel). In the opposite bisection, a longer spatial distance between the first and the second sound was associated with a shorter temporal delay between the two sounds, and the reverse for shorter distances (see **Figure 6C** top panel). Thanks to these two manipulations, it was possible to disentangle the role of spatial and temporal cues when it comes to the audio spatial bisection task. Our results show that these manipulations modified the performance of blind but not sighted participants. Indeed, in blind individuals, the spatial bisection deficit observed in the original version of the task disappeared when the study presented coherent temporal and spatial cues, and it increased in the conflicting condition. **Figure 6** (lower panels) plots the proportion of answer "second sound closer to the third sound" as a function of the position of the second sound for one blind (in red) and one age-matched sighted individual (in gray). In the equal bisection condition, we observed the same deficit observed previously [30], with random responses and no psychometric function for the blind subject. Interestingly, in the

#### **Figure 6.**

*Bisection tasks: coherent and conflicting manipulations of space and time. Results of the three conditions of the spatial bisection task for a typical blind participant (red symbols) and a typical sighted control (gray symbols). Subjects sat in front of an array of 23 speakers, which are illustrated by the sketches. (A) Equal spatial bisection. Top: the time interval between the first and the second sound (750 ms) was equal to the time interval between the second and the third sound. Bottom: proportion of trials judged "closer to the right sound source" plotted against the speaker position for the second sound. The size of the dots is proportional to trial number at that position. We fitted both sets of data with the Gaussian error function. (B) Coherent spatial bisection. Top: spatial distances and temporal intervals between the three sounds were directly proportional (e.g., long spatial distance and long temporal interval). Bottom: same as for (A). (C) Opposite spatial bisection. Top: spatial distances and temporal intervals between the three sounds were inversely proportional (e.g., long spatial distance and short temporal interval). Bottom: same as for (A) and (B). With permission from Gori et al. [58].*

**293**

*Audio Cortical Processing in Blind Individuals DOI: http://dx.doi.org/10.5772/intechopen.88826*

direction than expected).

**7. Space, time, and speed**

spatial bisection tasks.

abilities.

coherent bisection condition, the deficit disappeared and there was similar performance between the sighted and blind participants. More interestingly, in the opposite bisection condition (**Figure 6C**), while there was no effect of the manipulation that was evident in the sighted individual, in the blind individual, the response was inverted (i.e., the psychometric function was reversed and presented in the opposite

Performance of blind individuals reveals a strong temporal dominance for the spatial bisection task, suggesting that temporal cue is attracting the spatial auditory response [58]. A possible explanation is that, while the retinotopic organization of the visual cortex may support the reorganization underlying some enhanced audio spatial skills in blindness (such as the sound localization ability), it may be insufficient to guarantee the development of more complex spatial skills, such as those required for the audio spatial bisection task. Our results about the role of time in space representation suggest that temporal information can act as an alternative cue for reorganizing space representation subtending some more complex spatial

How can temporal information support space processing in blindness? It might be that, for some complex spatial representations, the visual system calibrates the auditory sense of space by processing the speed of the stimuli. Neurons that process speed information have been demonstrated for the visual modality in the visual cortex [59]. These neurons could be responsible for processing information during

In typical conditions, it may be that the visual system facilitates transfer of audio processing from a temporal to a spatial coordinate system. Indeed, audition is the most reliable sense to represent time information, and vision is the most reliable sense to represent space information. The mediator between auditory time and visual space could be velocity processing, which may represent a channel of communication between the two sensory systems. **Figure 7** reports a graphical description of how vision and audition may collaborate to estimate space and time starting from the speed properties of an object. Concerning space estimation in sighted individuals, given the higher weight of vision, it is independent of the temporal coordinates of the stimulus for both coherent (**Figure 7A**) and conflicting (**Figure 7B**) situations. On the other hand, when the visual information is unavailable, the spatial counterpart seems unable to develop and blind individuals seem to rely only on temporal coordinates to infer metric spatial information. One might then speculate that when the visual network is impaired, blind individuals internalize a statistical prior (i.e., a prior on the constant velocity of stimuli) derived from environmental statistics. This drives them to infer space from time. This idea is in agreement with the Imputed Velocity Theory [60], which asserts that humans intuitively attribute constant velocity to a single object moving through space over time. If we assume that blind individuals assume a prior of constant velocity of objects in space, they can use this information to extract space cues using time cues. This strategy would help blind people to overcome metric problems by using unimpaired temporal maps to decode spatial metrics. This may also facilitate their interaction with others (**Figure 7** left). This mechanism would be adaptive for blind individuals as it allows them to process spatial information correctly at the auditory level based on its temporal representation. On the other side, this mechanism could be maladaptive when conflicting spatial and temporal information is provided, as blind individuals can be deceived by the temporal

#### *Audio Cortical Processing in Blind Individuals DOI: http://dx.doi.org/10.5772/intechopen.88826*

*Visual Impairment and Blindness - What We Know and What We Have to Know*

of this hypothesis, we recently tested and verified that space representation of blind individuals is strongly influenced by the temporal representation of events [58]. We performed different versions of the spatial bisection task in sighted and blind individuals, in which we presented spatial and temporal independent, coherent, and conflicting information (**Figure 6** top panel). Similar to the original version of the bisection task [30], in one condition the temporal delay between the three sounds was always the same, and only spatial cues were relevant to compute the task (i.e., Equal bisection, **Figure 6A** top panel). In other conditions instead, we presented a spatio-temporal coherent or conflicting information. For example, in the coherent bisection, a longer spatial distance between the first and the second sound was associated with a longer temporal delay between the two sounds, and the reverse was the case for shorter distances (see **Figure 6B** top panel). In the opposite bisection, a longer spatial distance between the first and the second sound was associated with a shorter temporal delay between the two sounds, and the reverse for shorter distances (see **Figure 6C** top panel). Thanks to these two manipulations, it was possible to disentangle the role of spatial and temporal cues when it comes to the audio spatial bisection task. Our results show that these manipulations modified the performance of blind but not sighted participants. Indeed, in blind individuals, the spatial bisection deficit observed in the original version of the task disappeared when the study presented coherent temporal and spatial cues, and it increased in the conflicting condition. **Figure 6** (lower panels) plots the proportion of answer "second sound closer to the third sound" as a function of the position of the second sound for one blind (in red) and one age-matched sighted individual (in gray). In the equal bisection condition, we observed the same deficit observed previously [30], with random responses and no psychometric function for the blind subject. Interestingly, in the

*Bisection tasks: coherent and conflicting manipulations of space and time. Results of the three conditions of the spatial bisection task for a typical blind participant (red symbols) and a typical sighted control (gray symbols). Subjects sat in front of an array of 23 speakers, which are illustrated by the sketches. (A) Equal spatial bisection. Top: the time interval between the first and the second sound (750 ms) was equal to the time interval between the second and the third sound. Bottom: proportion of trials judged "closer to the right sound source" plotted against the speaker position for the second sound. The size of the dots is proportional to trial number at that position. We fitted both sets of data with the Gaussian error function. (B) Coherent spatial bisection. Top: spatial distances and temporal intervals between the three sounds were directly proportional (e.g., long spatial distance and long temporal interval). Bottom: same as for (A). (C) Opposite spatial bisection. Top: spatial distances and temporal intervals between the three sounds were inversely proportional (e.g., long spatial distance and short temporal interval). Bottom: same as for (A) and (B). With permission from Gori et al. [58].*

**292**

**Figure 6.**

coherent bisection condition, the deficit disappeared and there was similar performance between the sighted and blind participants. More interestingly, in the opposite bisection condition (**Figure 6C**), while there was no effect of the manipulation that was evident in the sighted individual, in the blind individual, the response was inverted (i.e., the psychometric function was reversed and presented in the opposite direction than expected).

Performance of blind individuals reveals a strong temporal dominance for the spatial bisection task, suggesting that temporal cue is attracting the spatial auditory response [58]. A possible explanation is that, while the retinotopic organization of the visual cortex may support the reorganization underlying some enhanced audio spatial skills in blindness (such as the sound localization ability), it may be insufficient to guarantee the development of more complex spatial skills, such as those required for the audio spatial bisection task. Our results about the role of time in space representation suggest that temporal information can act as an alternative cue for reorganizing space representation subtending some more complex spatial abilities.

## **7. Space, time, and speed**

How can temporal information support space processing in blindness? It might be that, for some complex spatial representations, the visual system calibrates the auditory sense of space by processing the speed of the stimuli. Neurons that process speed information have been demonstrated for the visual modality in the visual cortex [59]. These neurons could be responsible for processing information during spatial bisection tasks.

In typical conditions, it may be that the visual system facilitates transfer of audio processing from a temporal to a spatial coordinate system. Indeed, audition is the most reliable sense to represent time information, and vision is the most reliable sense to represent space information. The mediator between auditory time and visual space could be velocity processing, which may represent a channel of communication between the two sensory systems. **Figure 7** reports a graphical description of how vision and audition may collaborate to estimate space and time starting from the speed properties of an object. Concerning space estimation in sighted individuals, given the higher weight of vision, it is independent of the temporal coordinates of the stimulus for both coherent (**Figure 7A**) and conflicting (**Figure 7B**) situations. On the other hand, when the visual information is unavailable, the spatial counterpart seems unable to develop and blind individuals seem to rely only on temporal coordinates to infer metric spatial information. One might then speculate that when the visual network is impaired, blind individuals internalize a statistical prior (i.e., a prior on the constant velocity of stimuli) derived from environmental statistics. This drives them to infer space from time. This idea is in agreement with the Imputed Velocity Theory [60], which asserts that humans intuitively attribute constant velocity to a single object moving through space over time. If we assume that blind individuals assume a prior of constant velocity of objects in space, they can use this information to extract space cues using time cues. This strategy would help blind people to overcome metric problems by using unimpaired temporal maps to decode spatial metrics. This may also facilitate their interaction with others (**Figure 7** left). This mechanism would be adaptive for blind individuals as it allows them to process spatial information correctly at the auditory level based on its temporal representation. On the other side, this mechanism could be maladaptive when conflicting spatial and temporal information is provided, as blind individuals can be deceived by the temporal

#### **Figure 7.**

*Graphical model of our theory. In sighted individuals, spatial estimation is independent of the temporal cue of the stimulus for both coherent (A) and conflicting (B) information. Blind individuals infer spatial information using temporal coordinates of the stimulus assuming constant velocity. When spatial-temporal coherent stimuli are present, the spatial estimation can be successfully extracted by the temporal cue (C). On the other hand, when conflicting spatial-temporal information (D) is provided, the temporal cue is wrongly used to drive the spatial sound position assuming constant velocity.*

cue in the spatial evaluation, perceiving an illusory spatial position of the sound based on its temporal coordinates (**Figure 7** right).

These findings support the cross-sensory calibration theory [5, 61], suggesting that visual information is necessary for normal development of auditory sense of space. In children younger than 12 years of age, there is visual dominance over audition in spatial bisection, and an auditory dominance over vision in temporal bisection [5]. The cross-sensory calibration of the visual system for the spatial bisection explains why blind subjects show a specific temporal response to the spatial bisection task, while also showing different processing to solve Euclidean, metric, relationships. We can speculate that these processes could be mediated in sighted but not in blind people by pathways involving the superior colliculus [30, 55, 62]. The present study adds new evidence, showing other possible interactions during development among sensory modalities, as well as spatial and temporal domains.

#### **8. Conclusion**

A lack of vision hampers strategies and neural circuits underlying complex spatial metrics, driving to multisensory interactions that bring to code space based

**295**

*Audio Cortical Processing in Blind Individuals DOI: http://dx.doi.org/10.5772/intechopen.88826*

to convey richer information.

**Acknowledgements**

**Conflict of interest**

**Author details**

Technology, Genova, Italy

Monica Gori1

The authors declare no conflicts of interest.

\*, Maria Bianca Amadeo1,2, Giulio Sandini3

Università degli Studi di Genova, Genova, Italy

\*Address all correspondence to: monica.gori@iit.it

provided the original work is properly cited.

1 Unit for Visually Impaired People, Italian Institute of Technology, Genova, Italy

2 Department of Informatics, Bioengineering, Robotics and Systems Engineering,

© 2019 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,

3 Robotics, Brain and Cognitive Sciences Department, Italian Institute of

and Claudio Campus1

reported.

on temporal instead of spatial coordinates. These findings open new opportunities for developing sensory substitution devices and rehabilitation technologies for blind people, where spatial and temporal cues could be simultaneously manipulated

The authors acknowledge the contribution of David Burr and Concetta Morrone in helping to shape the line of research that forms the basis of this chapter, including their continuous support and their participation in some of the experiments

*Audio Cortical Processing in Blind Individuals DOI: http://dx.doi.org/10.5772/intechopen.88826*

on temporal instead of spatial coordinates. These findings open new opportunities for developing sensory substitution devices and rehabilitation technologies for blind people, where spatial and temporal cues could be simultaneously manipulated to convey richer information.
