Real-time Frequency Visualization
Animated representation of sound waves in the forest canopy. Humans hear the high-power, low-frequency rumbles (red). Animals hear the low-power, high-frequency ultrasonic chirps (blue).
The Power vs. Info Paradox
Sound Power: Low frequencies carry immense physical energy and travel great distances (wind, thunder, large animals). Humans are well-adapted to hear this raw power. Ultrasonic frequencies attenuate (fade) very quickly in the air.
Information: Bandwidth equals information capacity. The ultrasonic spectrum is vast. While humans hear ~20 kHz of bandwidth, a bat has access to over 150 kHz of bandwidth. This allows for incredibly fast, highly detailed, and complex communication and echolocation that we are completely deaf to.
Ambient Sound Power (dB SPL)
Estimated average acoustic energy in a forest environment.
Relative Information Capacity (Bandwidth)
Wider hearing ranges allow for processing vastly more ecological data per second.
Ray-Traced Acoustic Simulator: Direction & Reflection
Physics of Echolocation: Sound waves only reflect off objects larger than their wavelength. This ray-traced simulation models hundreds of acoustic paths, showing exactly how power scatters in different directions upon reflection, and how low frequencies diffract (pass through) small obstacles.
Experiment Designer: Target Sizing & Mathematical Acoustics
Because acoustic energy drops by extreme orders of magnitude via spherical spreading and Rayleigh scattering, we must use a Decibel (dB) scale to properly compare returning power. Target placed at 1.5 meters (3m round trip). Max possible reflection = 0 dB.
Calculated via target surface area interception and Rayleigh scattering limits.
Remaining energy integrated against human hearing sensitivity thresholds.
Select a size
Power Spectrum (Logarithmic dB)
Notice how high frequencies drop off heavily due to atmospheric attenuation ($0.0003 \times f^2$ dB/m). Low frequencies drop off on small targets due to Rayleigh scattering.
🔬 Advanced Spectrographic Analysis: Avoidance of Self-Deafening
This 3-panel spectrograph models absolute physical energy (-60 to 0 dB) and displays three temporal phases:
1. Source (t≈0) | 2. Received Echo (t≈10ms) | 3. Perceived Echo (t≈18ms): Filtered by the subject's brain (H(f)).
Observe the biological mastery: The Bat CW emits a tone at 74 kHz, but its brain is tuned to an 80 kHz "Acoustic Fovea". This decoupling prevents the loud emission from deafening the bat, relying on the Doppler-shift of flight to perfectly land the echo into the 80 kHz fovea! The Bat FM emits a 100-25 kHz sweep, but its brain ignores the heavily attenuated highest frequencies, peaking its sensitivity near 45 kHz.
Field Study: Subject Echolocation & Distance Attenuation
An expert bio-acoustician and experimental psychologist collaborated to measure 5 human subjects. They recorded the spectrographic profiles of their palatal clicks reflecting off 4 targets (5cm, 10cm, 20cm, 40cm) at a fixed 1m distance. Concurrently, they mapped the subjects' reported auditory detection rates as the targets were moved further away (1m to 4m), demonstrating the harsh realities of the acoustic inverse-square law ($1/r^4$ radar equation) and Forward Temporal Masking.
Hypothetical AI Simulation Notice: This section represents a predictive, AI-driven model used to inform the design of a planned real-world empirical study. By simulating these variables first, we optimize our wooden target sizes and distance intervals before testing human subjects in the forest.
- The Mathematical Sweet Spot: Calculated analytically by mapping the intersection of the acoustic radar equation (echo power decaying rapidly at $1/r^4$) against the physiological forward masking recovery function (inner-ear sensitivity recovering logarithmically over ~10-15ms after the initial loud click). The calculus predicts the maximum perceivable Signal-to-Masking Ratio should occur around 1.8m to 2.2m.
- The Empirical Sweet Spot: Will be determined in the real experiment by charting actual subject detection success rates across hundreds of blindfolded trials at randomized distance intervals.
- Expected Real-World Results: Real-world empirical results are expected to validate this AI simulation's general macro-trends (the dip at 1m, peak at 2m, and fade by 4m). However, we anticipate significant individual variance in the precise location of the "sweet spot" based on each subject's unique click duration, frequency spectrum, and neural recovery speed.
Measured Clicks & Echoes (Target at 1m)
Notice the 5cm target reflects almost zero energy due to severe Rayleigh diffraction. The 10cm target reflects the high-frequencies, but loses the low.
Subject Detection Rate vs Distance
With Masking: At 1m, the echo returns in just ~5.8ms. The loud outgoing click temporarily "deafens" the human ear (Forward Masking), causing detection rates to drop at close range! The 2m mark acts as a perceptual "sweet spot".