DeutschClear Cookie - decide language by browser settings
Heinrich, Melina (2015): Aspects of spatiotemporal integration in bat sonar. Dissertation, LMU München: Graduate School of Systemic Neurosciences (GSN)



Bat sonar is an active sense that is based on the common mammalian auditory system. Bats emit echolocation calls in the high frequency range and extract information about their surroundings by listening to the returning echoes. These echoes carry information, like spatial cues, about object location in the three-dimensional space (azimuth, elevation, and distance). Distance information, for example, is obtained from temporal cues as the interval between the emission of an echolocation call and the returning echo (echo delay). But echoes also carry information about spatial object properties like shape, orientation, or size (in terms of its height, width, and depth). To achieve a reliable internal representation of the environment, bats need to integrate spatial and temporal echo information. In this cumulative thesis different aspects of spatiotemporal integration in bat sonar were addressed, beginning with the perception and neural encoding of object size. Object width as size relevant dimension is encoded by the intensity of its echo. Additionally, the sonar aperture (the naturally co-varying spread of angles of incidence from which the echoes impinge on the ears) co-varies proportionally. In the first study, using a combined psychophysical and electrophysical approach (including the presentation of virtual objects), it was investigated which of both acoustic cues echolocating bats (Phyllostomus discolor) employ for the estimation of object width. Interestingly, the results showed that bats can discriminate object width by only using sonar-aperture information. This was reflected in the responses of a population of units in the auditory midbrain and cortex that responded strongest to echoes from objects with a specific sonar aperture, independent of variations in echo intensity. The study revealed that the sonar aperture is a behaviorally relevant and reliably encoded spatial perceptual cue for object size. It furthermore supported the theory that the mammalian central nervous system is principally aiming to find modality independent representation of spatial object properties. We therefore suggested that the sonar aperture, as an echo acoustic equivalent of the visual aperture (also referred to as the visual angle), could be one of these object properties. In the visual system object size is encoded by the visual aperture as the extent of the image on the retina. It depends on object distance that is not explicitly encoded. Thus, for reliable size perception at different distances, higher computational mechanisms are needed. This phenomenon is termed ‘size constancy’ or ‘size-distance invariance’ and is assumed to reflect an automatic re-scaling of visual aperture with perceived object distance. But in echolocating bats object width (sonar aperture) and object distance (echo delay) are accurately perceived and explicitly neurally encoded. In the second study we investigated whether bats show the ability to spontaneously combine spatial and temporal cues to determine absolute width information in terms of sonar size constancy (SSC). This was addressed by using the same setup and species as in the psychophysical approach of the first study. As a result SSC could not be verified as an important feature of sonar perception in bats. This lack of SSC could result from the bats relying on different modalities to extract size information at different distances. Alternatively, it is thinkable that familiarity with a behaviorally relevant, conspicuous object is required, as it was discussed for visual size constancy. But size constancy is found in many sensory modalities and more importantly, SSC was recently found in a blind human echolocator. It was discussed to be based on the same spatial and temporal cues as presented in our study. Thus, this topic should be readdressed in bats in a more natural context as size constancy could be a general mechanism for object normalization. As the spatiotemporal layout of the environment and the objects within changes with locomotion, in the third study the spatiotemporal integration in bat biosonar in a natural and naturalistic context was addressed. Trawling bats species hunt above water and capture fish or insects directly from or close to the surface. Here water acts as an acoustic mirror that can reduce clutter by reflecting sonar emissions away from the bats. However, objects on the water lead to echo enhancement. In a combined laboratory and field study we tested and quantified the effect of different surface types with different reflection properties (smooth and clutter surface) and object height on object detection and discrimination in the trawling bat species, Myotis daubentonii. The bats had to detect a mealworm presented above these different surfaces and discriminate it from an inedible PVC disk. At low heights above the clutter surface, the bats’ detection performance was worse than above a smooth surface. At a height of 50 cm, the surface structure had no influence on target detection. Above the clutter surface, object discrimination decreased with decreasing height. The study revealed different perceptual strategies that could allow efficient object detection and discrimination. When approaching objects above clutter, echolocation calls showed a significantly higher peak frequency, eventually suggesting a strategy for temporal separation of object echoes from clutter. Flight-path reconstruction showed that the bats attacked objects from below over water but from above over clutter. These results are consistent with the hypothesis that trawling bats exploit an echo-acoustic ground effect, in terms of a spatiotemporal integration of direct object reflections with indirect reflections from the water surface. It could lead to optimized prey-detection and discrimination not only for prey on the water but also above. Additionally, the bats could employ a precedence-like strategy to avoid misleading spatial cues that signal the wrong object elevation by using only the first and therewith direct echo for object localization.