9
$\begingroup$

I have been reading about the echolocation system in bats. papers state that bats tend to produce a sonar beam that has a beam angle (-3db) of about 30-40 degrees. This would be considered a wide beam given the fact the sonar sensors used by robots have beams 15-20 degrees or down to 10 degrees with multi-sonar.

As the beam width increases, the azimuth resolution decreases. Yet, bats have good navigational capabilities and good object discrimination in azimuth. How is that achieved? what is the technique used?

I searched hard but what I found is that they rely on the timing between two ears. however, given the complexity of the environment and the multiple echos that return, how is it possible to correlate which echo in the left ear belongs to which in the right ear?

$\endgroup$
3
$\begingroup$

I have been contemplating about this question for months and although I cannot provide you with a concise short answer, I do wish to share what I have so far.

First off, the acoustic beam that is sent is not really part of the echolocation resolution (the bat's echo-locating acuity). Consider the human field of view, which is nearly 180o wide in the horizontal plane. Covering your eyes and obscuring, say, half of the field will not substantially improve visual acuity, because it simply blocks half of the retinal surface from receiving light - it does not double the retinal surface devoted to the field of view.

Hence, I think it is not so much the bat's projected beam that matters, it is the incoming echo that matters. Note that objects such as a tree may scatter the sound that is reflected back, so the projected beam of 40o may be a dramatic underestimation of the chaotic signals bouncing back to the bat. What can they do to improve spatial detail in the received acoustic field of view?

Bats may increase the pulse repetition rate when they are approaching a target. Hence, they basically build a map by sampling the environment multiple times. Sampling frequencies may exceed 50 Hz (Grinnell and Griffin, 1985). For example, imagine you are blindfolded and allowed to scan an object with your fingers. Scanning it multiple times will definitely help you to mentally reconstruct the image. Indeed, bats can assemble information about echo delay changes over time, enabling them to temporally integrate information of the auditory scene. Moreover, they can change the spectral content of their calls that may help in echolocation (Moss & Surlykke, 2001).

Furthermore, to conclude with a personal thought - I think that bats must learn to decode complex echo profiles. For example, a chaotic huge blob of echo with low-echo efficiency (lots of sound is absorbed) that only changes when the bat moves is probably a tree. A big blob with a large echo-efficiency (a lot of sound comes back) with well-defined edges that only changes when the bat moves is likely a human structure or a rock. A pinpoint-echo that moves with respect to the bat is likely something tasty.

References
- Grinnell & Griffin, Biol Bull (1958): 10-22
- Moss & Surlykke, JASA; 110(4): 2207-26

$\endgroup$
1
$\begingroup$

I think you are forgetting bats can see as well. Bat's that use echo location are referred to as Microchiroptera. Since we are speaking about echolocation, we can disregard the megachiropteral class since there most prominent sense is their eyes. This doesn't mean though that Micro's don't use their eyes in conjunction with echolocation. In Handbuch Der Zoologie: Eine Naturgeschichte Der Stämme Des Tierreiches, Volume 2; Volume 5, the authors state

In experimental situations, when visual information and echolocation were competing, the bats relied on vision and ignored the acoustical cues [1].

From this we can surmise that even with echolocating bats, they still fall back on vision in ambiguous situations. Moreover, researchers have determined that bats have ultraviolet cone receptors [2].

To answer your question, Micro bats have good object discrimination because they use both their eyes and echolocation to discern objects in their environment instead of solely relying on echolocation.

$\endgroup$
  • 1
    $\begingroup$ yes sure they use vision for mapping in environment but many papers mention they can locate targets using sound only for example: link.springer.com/article/10.1007/BF00292562 bat located prey passively using its noise And here it says they are able to get the target location in 3D using difference in sound intensity but how do they correlate sounds at the two ears if many objects existed in the environment? sciencedirect.com/science/article/pii/S096098220500686X $\endgroup$ – Amr Morssy May 6 '15 at 10:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.