Stretched upside-down maps of auditory space in the optic tectum of blind-reared owls; Acoustic basis and behavioral correlates

Vision during early life plays an important role in calibrating sound localization behavior. This study investigates the effects of visual deprivation on sound localization and on the neural representation of auditory space. Nine barn owls were raised with eyelids sutured closed; one owl was congenitally anophthalmic. Data from these birds were compared with data from owls raised with normal visual experience. Sound localization behavior was significantly less precise in blind-reared owls than in normal owls. The scatter of localization errors was particularly large in elevation, though it was abnormally large in both dimensions. However, there was no systematic bias to the localization errors measured over a range of source locations. This indicates that the representation of auditory space is degraded in some way for blind-reared owls, but on average is properly calibrated. The spatial tuning of auditory neurons in the optic tectum was studied in seven of the blind-reared owls to assess the effects of early visual deprivation on the neural representation of auditory space. In normal owls, units in the optic tectum are sharply tuned for sound source location and are organized systematically according to the locations of their receptive fields to form a map of auditory space. In blind-reared owls, the following auditory properties were abnormal: (1) auditory tuning for source elevation was abnormally broad, (2) the progression of the azimuths and elevations of auditory receptive fields across the tectum was erratic, and (3) in five of the seven owls, the auditory representation of elevation was systematically stretched, and in the two others large portions of the representation of elevation were flipped upside down. The following unit properties were apparently unaffected by blind rearing: (1) the sharpness of tuning for sound source azimuth, (2) the orientation of the auditory representation of azimuth, and (3) the mutual alignment of the auditory and visual receptive fields in the region of the tectum representing the area of space directly in front of the animal. The data demonstrate that the brain is capable of generating an auditory map of space without vision, but that the normal precision and topography of the map depend on visual experience. The space map results from the tuning of tectal units for interaural intensity differences (IIDs) and interaural time differences (ITDs; Olsen et al., 1989). To explore parameters of these cues that might account for the abnormalities in the auditory space map, we measured these cues across the frontal hemifield and determined their spatial pattern (pattern of change in cue values across space), frequency dependence (variation in cue value across frequency for each location), and individual variability (variability of cue values across individuals). Comparison of these localization cue parameters with unit spatial tuning from blind-reared owls revealed the following: (1) an upside-down representation of elevation predicts an upside-down mapping of IID in the tectum - this prediction was confirmed by direct measurement of unit IID tuning; (2) the prevalence of large visual-auditory misalignments in elevation correlates strongly with the frequency dependence of IID cues produced by a sound source at the location of a unit's visual receptive field; and (3) the prevalence of large visual-auditory misalignments in azimuth correlates with the frequency dependence of both ITD and IID cues associated with the location of a unit's visual receptive field. Thus, vision is most important for the auditory representation of those peripheral locations where localization cue values are highly frequency dependent and is substantially less important for the representation of frontal locations where auditory cue values are relatively constant across frequency and predictable across individuals. The altered topography of the auditory space map in blind-reared owls did not result from an adaptation to an abnormal motor map. The tectal map of head movement was assessed using microstimulation of physiologically defined sites. Although the topographies of both the motor map and the auditory map were abnormal, the abnormalities were different in each case and therefore represent separate effects of early blindness on the two maps.