Fast-moving bat ears create informative Doppler shifts

Abstract

Many animals have evolved adept sensory systems that enable dexterous mobility in complex environments. Echolocating bats hunting in dense vegetation represent an extreme case of this, where all necessary information about the environment must pass through a parsimonious channel of pulsed, 1D echo signals. We have investigated whether certain bats (rhinolophids and hipposiderids) actively create Doppler shifts with their pinnae to encode additional sensory information. Our results show that the bats' active pinna motions are a source of Doppler shifts that have all attributes required for a functional relevance: (i) the Doppler shifts produced were several times larger than the reported perception threshold; (ii) the motions of the fastest moving pinna portions were oriented to maximize the Doppler shifts for echoes returning from the emission direction, indicating a possible evolutionary optimization; (iii) pinna motions coincided with echo reception; (iv) Doppler-shifted signals from the fast-moving pinna portion entered the ear canal of a biomimetic pinna model; and (v) the time-frequency Doppler shift signatures were found to encode target direction in an orderly fashion. These results indicate that instead of avoiding or suppressing all self-produced Doppler shifts, rhinolophid and hipposiderid bats actively create Doppler shifts with their own pinnae. These bats could hence make use of a previously unknown nonlinear mechanism for the encoding of sensory information, based on Doppler signatures. Such a mechanism could be a source for the discovery of sensing principles not only in sensory physiology but also in the engineering of sensory systems.

Keywords: Doppler shifts; biosonar; ear motions; nonlinear sensing; time-frequency signatures.

Fig. 1.

Pinna-tip speeds and maximum Doppler shifts for all three bat species studied. (A) Pinna-tip speeds calculated from reconstructed 3D trajectories of landmarks placed on the pinna tip. (B) Maximum Doppler shifts calculated under the assumption that the pinna moves in the direction of sound propagation (center line: median; box edges: 25th and 75th percentiles; whiskers: minimum and maximum values). Maximum Doppler shifts were calculated under the assumption that the direction of the maximum pinna-surface velocity is aligned with the radiation direction of the echoes.

Fig. 2.

Distribution of speed, directional cosine between surface velocity and radiation direction, and Doppler shift on the inner surface of the pinna. (A) Maximum surface speed during a pinna motion. (B) Directional cosine between surface velocity and the direction of sound propagation associated with the maximum speed. (C) Doppler-shift estimates based on A and B. Directional cosines were calculated under the assumption that the propagation direction of the echo corresponds to the direction associated with the maximum amplitude of the pulse.

Fig. 3.

Fast pinna motions and large Doppler shifts occur during echoes. (A) Pinna-surface speed calculated from the landmark on the tip of the pinna superimposed on the spectrogram of biosonar pulse from greater horseshoe bat (Rhinolophus ferrumequinum) and its echoes. In the shown example recording, the pulses coincided with a forward motion of the pinna (positive speeds). (B) Portion of pulses with maximum Doppler shift exceeding a certain threshold. Number of motion sequences/echoes analyzed: H. armiger, 39; H. pratti, 57; R. ferrumequinum, 36.

Fig. 4.

Direction-dependent Doppler-shift signatures received at the ear canal of a deforming biomimetic pinna. (A) Spectrogam of an example Doppler signature (azimuth, 60○; elevation, 0○) with a superposed prediction of the maximum Doppler shift based on pinna-surface velocity estimates (magenta line). The arrow indicates the start of the pinna motion. (B) Maximum number of direction that could be distinguished based on Doppler signatures as a function of the signal-to-noise ratio. (C) Clustering result showing an orderly breakup of the direction space based on the Doppler signatures. The different gray scale values and numbers denote the clusters that Doppler signatures for the respective direction were assigned to.