In the dark rivers of South America, a small fish called Eigenmannia fills the water around it with a steady electric hum and reads the faint distortions that objects make in that field. When a neighbor arrives humming at nearly the same pitch, the two fields collide and the fish’s picture of the world blurs. The fish’s answer, discovered in 1963 and studied for six decades since, is simple to state: each one nudges its own frequency away from the other until the interference is easy to ignore [1][2]. Neuroscientists call it the jamming avoidance response, and it is among the most thoroughly studied computations in neuroscience.
Satellite navigation has the opposite problem. In 2026, interference with GPS and its sister systems is described as a persistent feature of conflict zones on several continents [6], and European aviation regulators are on their fourth revision of a safety bulletin about it [7]. This article asks whether the fish’s trick can be borrowed. My finding is a similar pattern with an important difference. The fish solves a real problem, separating your own signal from a neighbor’s, and an engineering team has already built a working light-based copy of it [3][4]. But the fish can move its frequency, and a navigation receiver cannot move a satellite’s. The part of the fish’s solution that transfers cleanly to navigation is not the frequency shift at all.
Scientific Foundation
The jamming avoidance response was found by Akira Watanabe and Kimihisa Takeda in 1963, in an Eigenmannia discharging at about 300 hertz. Stimuli within a few hertz of the fish’s own frequency made it shift its discharge, by up to about 6.5 hertz. Theodore Bullock and colleagues named the behavior in 1972, and in 1975 Walter Heiligenberg found it in an unrelated African fish, Gymnarchus, which shows that evolution arrived at the same solution twice [2]. Related behavior appears in other lineages. One bat species studied by Nachum Ulanovsky and colleagues shifts its echolocation call frequencies when flying with others, while another bat species they tested showed no such response [10].
The computation is the interesting part. When two fish have similar frequencies, their fields add to produce a slow wobble in amplitude, at a rate equal to the frequency difference. The fish must work out whether its neighbor is higher or lower than itself, because the correct escape is to move in the opposite direction. As summarized by Stamper and colleagues, the fish does this using amplitude and phase information gathered across its body, and it does so without an internal copy of its own discharge [1]. Plotted as a loop in the amplitude and phase plane, the signal turns clockwise for one sign of the difference and counterclockwise for the other, and that direction of rotation predicts which way the fish will shift [1]. Experiments in which the fish’s own electric organ was silenced showed that the response depends on the external stimulus frequency compared with the fish’s own discharge as sensed by its electroreceptors [2]. In the midbrain, sign-selective neurons respond to only one sign of the difference, and their output reaches the pacemaker that sets the discharge rate [2].
Two honest corrections belong here. First, “jamming” in this literature means accidental interference between neighbors, not hostile action. Second, the textbook response is not the whole story. Stamper and colleagues showed that the fish also respond to the slow envelope created when three or more fish are present, moving that envelope into a particular low-frequency range, and they note that the loop is closed: the fish’s own shift changes the input it then senses [1].
Now the navigation side. Jamming overwhelms satellite signals so that a receiver loses service; spoofing feeds a receiver false signals so that it keeps working but reports a wrong position [6][7]. The numbers have grown quickly. Aircraft interference cases in Poland rose from 1,908 in October 2024 to 2,732 in January 2025. During the June 2025 Iran–Israel war, jamming was estimated to affect about 970 ships a day in the Strait of Hormuz [6]. The International Air Transport Association reports a 220 percent rise in GPS signal-loss events between 2021 and 2024, and EUROCONTROL estimates that up to 38 percent of European en-route traffic passes through regularly affected regions [7]. In June 2026 an experimental satellite from the company Xona Space Systems produced what is described as the first space-based map of the problem, with signal strength at its altitude falling from about 40 decibels to as low as 10 decibels in the worst areas, from France to the Pakistan border. Xona’s co-founder cautions that the altitude means the map need not match what ground users experience [8].
Defenses exist and are layered. The European aviation bulletin recommends combining satellite navigation with inertial systems, ground beacons, radar and air traffic control, and notes that receivers using several frequencies and constellations help but can be hit across nearby bands. Galileo’s message authentication, operational since July 2025, helps verify that navigation data is genuine, which addresses spoofing but not jamming [7]. The bulletin’s most useful line is about trust: a missing position is easier to handle than a convincing but false one [7]. Military and some commercial users also fit controlled reception pattern antennas, arrays that apply adjustable phase weights to each element so that signals from a jammer’s direction cancel while satellite signals are kept. The cost is that each extra null can remove part of the sky and, with it, some satellites [5].
Cross-Domain Connection
Think of two guitarists tuning by ear. When their strings are nearly in tune, the sound swells and fades in slow pulses; that pulsing is the beat, and it is the sound of interference. The fish’s response is to detune deliberately, turning a slow, confusing throb into a fast buzz that its brain can sort out, and its real skill is knowing which direction to detune.
That skill has already been copied. A team led by Mable Fok at the University of Georgia built a photonic version of the fish’s response from an off-the-shelf semiconductor optical amplifier [3][4]. It detects whether a nearby signal could cause jamming, then shifts its own emitted signal up or down, away from the interferer and without crossing it. In tests with microwave-band jamming signals, the shift began when a jammer approached and stopped when it left [4]. The group describes the approach as workable from hundreds of megahertz to tens of gigahertz [3]. This is a real engineering kinship with the fish, with one caveat I discuss below about what the sources do and do not show.
Now to GNSS. The fish adapts by moving its own transmitter, and a navigation receiver has no transmitter to move. Satellites broadcast on fixed, internationally allocated frequencies, so a receiver cannot slide away from a jammer the way the fish slides away from its neighbor. The fish’s most celebrated move is therefore not available to the people who need it most. It is available, though, to systems that both transmit and receive, such as radar, wireless links and data networks, which is where the photonic copy points [3][4].
What does transfer is the way the fish decides. The fish extracts a direction, higher or lower, from the geometry of a signal sampled at different points on its body [1]. A controlled reception pattern antenna does something with a similar spirit: it extracts the direction of an interferer from the phase relationships across an array and then acts on it [5]. This is my synthesis, not a claim made in any source I found, and the two systems are far apart in mechanism, since one is a neural computation on frequency differences and the other is linear algebra on spatial weights. But both turn a muddy sum of signals into a directional answer by comparing samples taken from different places. That is the shared idea.
Two differences limit the analogy. The first is intent. The fish’s neighbor is not trying to blind it, and in the biological response both animals can shift. A hostile jammer can follow a target’s frequency, and I found no source testing a fish-like rule against an adaptive adversary. The second is that spoofing is a different problem from jamming. The fish’s response handles interference, not forgery, and I found no equivalent of message authentication in the electric-fish literature I retrieved. A spatial method also has a blind spot: its basic assumption is that interference and satellite signals arrive from different directions [5], so a spoofer lying in the same direction as a satellite would be hard to separate by direction. That last point is my inference from the stated assumption, not a limit stated by the source.
What Remains Undemonstrated
The photonic copy is the best evidence in the field, and it is thinner than its reputation suggests. The project page I retrieved gives no performance figures [3], and trade-press coverage of the work reports a demonstration with microwave-band signals and a plan to handle multiple jammers [4]. I did not retrieve the underlying paper’s data, so I cannot say how fast it reacts, how many simultaneous interferers it tolerates, or how it fares against a jammer that deliberately pursues it. Claims about helping a spectrum shortage or protecting military radar are expectations in the coverage, not results [4].
On the biological side, the neural wiring has been mapped in considerable detail in Eigenmannia [2], but even the mathematical description has open questions. The authors of one study note that the clean mathematical transform used in textbook models cannot be what the brain does in real time, and that rectification and filtering are more likely, without identifying where in the circuit this happens [1].
For navigation, I found no work applying a fish-style frequency-shifting rule to GNSS itself, and given the fixed-frequency problem above, none is expected. The newer low Earth orbit navigation constellations are a more promising case, because their designers can choose signals. Xona describes a planned 300-satellite constellation with signals up to 100 times stronger than GPS and cryptographic authentication, and its co-founder estimates that existing jammers would affect only about 5 percent of the area they disrupt today [8][9]. These are company statements, the 100-times figure is a stated maximum with no test conditions, and the source I retrieved does not say the jam resistance has been demonstrated in orbit [9]. None of this involves frequency agility, so it is a parallel in goals, not in method.
The experiment I would propose has three parts. First, test a sign-aware frequency-shift controller on a system that can transmit, such as a data link or a ground-based navigation beacon, against both a fixed jammer and one that tracks it. Second, compare how many interferers an array-based method and a fish-style comparison across sensor positions can separate as the number of jammers grows. Third, publish the results with the evaluation protocol, since a rule that works against a polite neighbor may fail against a pursuer.
Why It Matters
For engineers, the lesson is where to look. The fish does not solve the problem by being stronger. It solves it by knowing which way to move, and by being able to move. Navigation designers who can choose new signals, as the low orbit constellations can, may be able to give receivers something to latch onto beyond raw power.
For pilots, shipping companies and anyone relying on a map app, the practical point is the one in the European bulletin: a navigation system that fails loudly is safer than one that lies quietly [7]. Jamming and spoofing deserve different defenses, and a technique borrowed from biology should be judged against the one it actually fits.
For readers of science news, this is also a caution about the phrase “inspired by nature.” The inspiration here is real and has been built, but the sources describe a lab demonstration for transmitters, not a fix for the navigation crisis.
Human Dimension
There is something quietly civilized about the electric fish. Two animals meet in the dark; each registers that the other’s signal is a nuisance; and each makes a small adjustment so that both can keep working. Nothing in the biology requires a treaty.
The GNSS problem is almost the reverse. The people jamming signals are not accidental neighbors, and the response so far has been diplomatic protests at international bodies, regulatory bulletins and technical workarounds [6][7]. It is a humbling comparison. A fish brain solved its interference problem long ago, and we are still debating who should be allowed to shout in the shared channel.
Sources
- Journal of Experimental Biology, Stamper, Madhav, Cowan and Fortune, “Beyond the Jamming Avoidance Response: weakly electric fish respond to the envelope of social electrosensory signals” (2012), https://limbs.lcsr.jhu.edu/wp-content/uploads/2013/05/Stamperbeyond2012.pdf
- Wikipedia, “Jamming avoidance response,” https://en.wikipedia.org/wiki/Jamming_avoidance_response
- University of Georgia WAVE Lab, “Bio-inspired jamming avoidance response” (project page), https://wave.engr.uga.edu/portfolio/bio-inspired-jamming-avoidance-response/
- eeNews Europe, “Fish show how to use limited bandwidth more efficiently” (2018), https://www.eenewseurope.com/en/fish-show-how-to-use-limited-bandwidth-more-efficiently/
- GPS World, “Anti-jam technology: Demystifying the CRPA,” https://www.gpsworld.com/anti-jam-technology-demystifying-the-crpa/
- Inside GNSS, “GNSS Interference Now a Constant of Modern Conflict, SWF Annual Report Finds” (April 2026), https://insidegnss.com/gnss-interference-now-a-constant-of-modern-conflict-swf-annual-report-finds/
- Geoawesome, “EASA Updates GNSS Guidance as Jamming and Spoofing Surge,” https://geoawesome.com/easa-updates-gnss-guidance-as-jamming-and-spoofing-surge-what-geospatial-professionals-need-to-know/
- Space.com, Pultarova, “‘It’s quite a bit more than we expected’: Satellite reveals immense scale of GPS signal tampering” (June 2026), https://www.space.com/space-exploration/satellites/its-quite-a-bit-more-than-we-expected-satellite-reveals-immense-scale-of-gps-signal-tampering
- GPS World, “Xona Space Systems Pulsar-0 satellite begins testing for commercial LEO navigation” (2025), https://www.gpsworld.com/xona-space-systems-pulsar-0-satellite-begins-testing-for-commercial-leo-navigation/
- Proceedings of the Royal Society B, Ulanovsky, Fenton, Tsoar and Korine, “Dynamics of jamming avoidance in echolocating bats” (2004), https://pmc.ncbi.nlm.nih.gov/articles/PMC1691745
Idea originated at artificialideas.org. Article researched and written by Claude Sonnet 5.5. Published at artificialideas.org.
