During predation, big brown bats (Eptesicus fuscus) bats produce frequency modulated (FM) echolocation sweeps that exhibit rapid changes in duration, interval, and spectral features as they search for, track, and intercept prey (Schnitzler and Kalko, 2001 ; Simmons et al., 1979 ). Modeling how insectivorous bats modulate their echolocation calls with respect to the motion of their prey has been crucial for identifying the importance of target motion prediction and inter-call timing optimization during tracking (Erwin et. al 2001; Luo et. al 2017; Moss et al., 2004; Salles et. al 2020).

 

In this study, we apply an approach from control theory, known as system identification, to quantify the relationship between echolocation parameters and target distance. The bats are systematically exposed to prey targets that move back-and-forth over a range of oscillation frequencies. A broad range of motion frequencies is implemented to understand the way bats process different movement features (e.g., slow, steady motion or rapidly changing motion) through their echolocation parameters. We use these motions to create empirical response functions which map changes in target motion to changes in echolocation behavior.

 

To gain further intuition into the mapping between target motion and echolocation parameters, numerical transfer function models, such as delayed high-, low-, and band- pass filters, are fit to the observed response functions. These transfer functions provide insight into the aspects of target motion the bat prioritizes and quantify the processing delays that may exist during active sensing. Thus far, we have demonstrated that pulse duration and pulse interval mod