Target tracking and temporal patterning of sonar calls

Ninad Kothari

Insectivorous bat species rely on echolocation to navigate and capture moving prey in the dark (Griffin, 1958). Echolocating bats produce ultrasonic signals, which travel through air, reflect back from objects in the immediate environment, and provide snapshots of information to build an acoustic scene. Echolocation is not only an active sensing system, but an adaptive system, in which bats modify sonar call parameters such as pulse duration (PD), pulse intensity, pulse interval (PI) and frequency content of calls, in response to information gathered from the environment and the task at hand. The bat's adaptive changes in call parameters directly influence the information contained in the returning echoes (Moss and Surlykke, 2010).


Figure 1

In addition to the above adaptive behaviors, bats have also been observed to temporally pattern their sonar calls into “Sonar Sound Groups” (SSGs), which are clusters of sonar calls with relatively stable intervals, flanked by calls at longer pulse intervals. Bats produce SSGs in both lab conditions as well as in the field (Kothari et al., 2014; Moss and Surlykke, 2001; Petrites et al., 2009; Surlykke and Moss, 2000).

It has been shown that in complex situations, which require greater spatio–temporal resolution of targets under focus, bats produce sonar sound groups (Falk et al., 2014; Falk et al., 2015; Kothari et al., 2014; Moss et al., 2006; Petrites et al., 2009; Sändig et al., 2014; Surlykke et al., 2009). In addition to challenging tasks which require greater spatio–temporal resolution, in their natural environment, echolocating bats also track and intercept free–flying prey, including those that have evolved ears (Roeder, 1962; Roeder, 1967). In some insect species, the temporal ultrasonic vocalization pattern of attacking echolocating bats can provide cues to trigger evasive maneuvers, like diving and unpredictable erratic trajectories (Corcoran et al., 2009; Ghose et al., 2009; Roeder, 1962; Triblehorn and Yager, 2005). An important aspect of erratic insect flight trajectories is that it makes spatio–temporal localization of the target a challenge for echolocating bats. It has been shown (although not quantitatively analyzed) that while tracking insects with unpredictable trajectories bats produce sonar sound groups (Ghose et al., 2009; Triblehorn et al., 2008). It is hypothesized that such temporal patterning of sonar sounds, which allows less time for hearing insects to react to ultrasound produced by attacking bats might be an effective strategy used by bats to become successful predators (Triblehorn et al., 2008).

In this project we set out to explore the question of whether temporal patterning of sonar sounds is a strategy employed by bats to track unpredictably moving targets. We designed an experiment in which we could control the relative bat–target position, while engaging the bat in its natural target tracking behavior and also allow us to systematically control the predictability in target trajectory over successive trials.

Experimental Design


Figure 2

We trained bats to rest on a platform and track a tethered target whose trajectory could be controlled using a software controlled stepper motor (Figure 2a). This paradigm allowed us to manipulate the predictability in target trajectory in a controlled manner, while allowing repeatability in the bat's naturalistic echolocation tracking behavior. Bats are initially trained on a Simple Motion (SM) task. The displacement and velocity with respect to the stationary bat are shown in Figure 2b (blue trace). Once the bat reliably tracked the tethered insect following the smooth motion (SM) trajectory, two novel types of target motion trajectories were introduced to the bat on the day of the experiment. We refer to these target motion trajectories as Complex Motions 1 & 2 (CM1 and CM2, respectively). In the novel complex motion trajectories the target first moved towards the bat, after which it oscillated back and forth, before finally reaching the bat. The target displacement and velocities relative to the stationary bats are shown in Figure 1b (Complex 1 – red, Complex 2 – black). The simple and two complex motion trajectories were designed with the following criteria. The initial phase (starting 1400 ms) of target motion was comparable in each motion trajectory. This phase is marked by the two green lines in Figure 2c. After this identical initial phase, the target motion paths diverge and follow pre–determined trajectories. The goal of keeping the initial trajectories comparable was to present comparable target position information to the bats at the beginning of each trial. In order to introduce target motion uncertainty the trials (CM1, CM2 and SM) were randomized. This is marked as the block of random trials, the first box/step, in the flow chart in Figure 2c. After the presentation of random SM and CM trials, the bat was presented with a sequence of SM trials, as shown in Figure 2c.

Results

We observe that as target uncertainty reduces (repeated presentation of SM) bats reduce the production of sonar sound groups (Figure 3).


Figure 3

Copyright@2015 Batlab, Johns Hopkins University
Questions and comments to wxian1@jhu.edu

Template by OS Templates