EAT OR BE EATEN:
An Open-Source Multispecies Simulation of Sensory Reality, Ecological Behavior, and Perspective
DOI: [to be assigned]
John Swygert
September 3, 2026
Abstract
Animals occupy the same physical world without necessarily inhabiting the same perceptual world. Differences in visual spectra, olfaction, audition, mechanoreception, electroreception, magnetoreception, thermoreception, pressure sensitivity, hydrodynamic sensing, and internal physiological feedback create species-specific informational environments. A landscape that appears visually empty to a human may contain magnetic, chemical, electrical, acoustic, thermal, or hydrodynamic structure that is behaviorally significant to another organism.
This paper proposes Eat or Be Eaten, an open-source scientific simulation and multiplayer video-game architecture in which all organisms occupy a common physical environment while each player receives a different representation of that environment according to the sensory capabilities of the organism being controlled. The central innovation is not merely the ability to play different species. It is the deliberate absence of a universal perceptual representation of the game world.
A salmon, shark, bat, spider, bee, snake, human, and other playable organisms would therefore encounter the same simulated physical events through fundamentally different informational interfaces. Sensory representations would be constrained wherever possible by empirical measurements of receptor sensitivity, sensory range, temporal resolution, spectral response, behavioral experiments, and known ecological function.
The system would simultaneously function as entertainment, education, scientific visualization, hypothesis-generation infrastructure, and potentially an experimental environment for studying animal navigation and behavior. The project is proposed as open source so that biologists, neuroscientists, ecologists, game developers, educators, students, and citizen scientists can progressively improve individual species models.
Its underlying proposition is simple:
Physical reality may be shared while accessible reality is species-dependent.
The player is therefore not merely asked to control another organism.
The player is asked to learn how to inhabit another organism's informational world.
Would you like to play?
1. Introduction
Video games ordinarily provide different characters with different abilities while preserving essentially the same representation of reality for every player. A character may run faster, see farther, breathe underwater, fly, or possess different weapons, yet the visual and informational architecture through which the player encounters the game remains fundamentally human.
Eat or Be Eaten reverses this assumption.
Every player occupies the same underlying simulated physical environment, but the environment presented to the player is transformed according to the sensory architecture of the selected organism.
Consequently, two players may occupy exactly the same coordinates while experiencing substantially different informational environments.
A human player may see a dark body of water.
A shark may receive information corresponding to nearby electrical activity and hydrodynamic disturbances.
A salmon may encounter information associated with temperature, water chemistry, flow, depth, and geomagnetic position.
A bat may construct spatial information primarily from emitted sound and returning echoes.
A bee may have access to ultraviolet and polarization information unavailable to ordinary human vision.
A snake possessing infrared-sensitive pit organs may receive thermal information that dramatically alters the perceptual prominence of nearby organisms.
A spider may obtain information from vibrations propagated through a web or substrate.
The game therefore becomes an experiment in comparative sensory reality.
This approach draws upon the biological concept of the Umwelt: the meaningful perceptual environment available to an organism. The objective is not to claim that a computer simulation can reproduce the subjective consciousness of another species. That claim would exceed available scientific evidence.
Instead, the scientifically defensible objective is to reconstruct, translate, and simulate portions of the information available to an organism.
This distinction is essential.
The simulation does not claim:
This is what being a salmon feels like.
It asks:
What does the physical world contain for a salmon that it does not contain for us perceptually, and what behavior emerges when an agent must operate using that information?
2. One Physical World, Multiple Accessible Realities
The architecture begins with a common environmental state.
Let the physical environment at time t be represented as
[ E(t) ]
containing measurable variables such as:
[ E(t)={L,S,T,P,C,M,F,V,A,\ldots} ]
where, for example:
L = electromagnetic radiation,
S = sound,
T = temperature,
P = pressure,
C = chemical composition,
M = magnetic field,
F = fluid movement,
V = vibration,
A = electrical activity.
No organism has unrestricted access to E(t).
Instead, species i possesses a sensory transformation:
[ R_i(t)=\Phi_i[E(t)] ]
where \Phi_i represents the sensory capabilities, detection thresholds, spatial resolution, temporal resolution, receptor distributions, and neural preprocessing available to that organism.
Thus,
[ R_{\text{human}}(t) \neq R_{\text{salmon}}(t) ]
even though both organisms occupy the same E(t).
The simulation therefore maintains a distinction among:
[ \text{Physical Reality} ]
[ \downarrow ]
[ \text{Species-Specific Detectable Information} ]
[ \downarrow ]
[ \text{Integrated Operational Representation} ]
[ \downarrow ]
[ \text{Behavior} ]
The player's screen represents an experimentally informed human translation of the second and third layers.
It is not intended to represent consciousness itself.
3. The Central Game Mechanic
The defining mechanic of Eat or Be Eaten is straightforward:
There is no universal game screen.
The server maintains one physical simulation.
Each client renders a species-specific interpretation.
Consider two players separated by ten meters.
One controls a small mammal.
Another controls a snake.
The mammal player may see vegetation, shadows, movement, and visual landmarks.
The snake player may receive an altered representation emphasizing thermal contrasts relevant to its infrared-sensitive sensory system.
Neither player receives an objectively superior representation.
Each receives different information.
This produces asymmetrical gameplay without artificially assigning arbitrary powers.
Biology itself generates the asymmetry.
4. Sensory Channels
Each playable species would be defined through a modular Sensory Profile.
Possible channels include:
Photoreception
Visible wavelengths, ultraviolet sensitivity, spectral discrimination, polarization sensitivity, motion sensitivity, low-light performance, and spatial resolution.
Audition
Frequency range, directional localization, amplitude sensitivity, and temporal resolution.
Echolocation
Emission characteristics, echo delay, frequency information, object reflectivity, movement, and Doppler-related information where applicable.
Olfaction and Chemoreception
Chemical identity, concentration gradients, temporal persistence, waterborne or airborne transport, and learned chemical signatures.
Mechanoreception
Touch, vibration, substrate movement, lateral-line information, web vibration, and hydrodynamic disturbances.
Thermoreception
Ambient temperature and, in organisms possessing specialized systems, directional thermal information.
Electroreception
Biological electrical fields and environmental electrical information.
Magnetoreception
Directional and/or positional information associated with Earth's magnetic field where supported by evidence.
Pressure and Depth
Hydrostatic pressure and pressure changes.
Proprioception and Vestibular Information
Orientation, acceleration, limb position, and body movement.
These modules can be combined differently for each organism.
5. The Salmon Model
Salmon provide an especially useful demonstration because their navigation appears to integrate information across multiple spatial scales.
Evidence supports the importance of olfactory imprinting in recognizing natal streams, while research also supports a role for geomagnetic information in long-distance navigation. Reviews of ocean navigation further emphasize combinations of magnetic, chemical, and hydrodynamic information rather than a single universal navigational cue.
A simplified salmon model might therefore contain:
[ R_s = f(M,C,T,F,P,L,I) ]
where:
M = magnetic information,
C = chemical information,
T = temperature,
F = water movement,
P = pressure/depth,
L = available light,
I = internal physiological state.
The game could consequently represent the ocean not as an undifferentiated blue space but as a complicated three-dimensional informational landscape.
6. The Moving-Highway Hypothesis
The game architecture also provides an experimental framework for a particularly interesting navigational possibility.
Ocean currents can be conceptualized as a changing network of vertically distributed moving highways.
At different depths, water masses may move:
in different directions,
at different velocities,
at different temperatures,
with different salinities,
with different oxygen concentrations,
and with different chemical or biological characteristics.
An organism capable of inexpensive vertical movement might therefore achieve substantial horizontal transportation by selecting among current layers.
Instead of:
[ \text{swim horizontally toward destination} ]
a strategy could sometimes resemble:
[ \text{detect state} \rightarrow \text{change depth} \rightarrow \text{enter advantageous current} \rightarrow \text{passive transport} \rightarrow \text{reassess} ]
This is presented here as a testable simulation hypothesis, not as a universal explanation of marine migration.
The animal might respond to temperature alone, geography alone, or—more plausibly for some species—to combinations of variables.
For example:
[ D = f(G,T,F,I) ]
where D is depth-selection behavior, G represents geographic/navigation information, T temperature, F current characteristics, and I internal state.
An organism might therefore possess a geographical migratory objective while simultaneously selecting environmentally favorable transportation layers.
7. Equilibrium-Seeking Behavior
A second testable model concerns physiological equilibrium.
Suppose an organism has a preferred environmental state:
[ Q^* ]
and experiences the current state:
[ Q(t) ]
The deviation can be represented conceptually as
[ \Delta Q = Q(t)-Q^* ]
Behavior can then alter the organism's location until |\Delta Q| decreases.
For temperature:
[ T_{\text{cold}} \rightarrow \text{vertical search} \rightarrow T_{\text{preferred}} \rightarrow \text{reduced search behavior} ]
If the preferred water mass is itself moving horizontally, apparently purposeful geographic movement can emerge partly from repeated local attempts to maintain physiological conditions.
The stronger model combines equilibrium with navigation:
[ B(t)=f(G,\Delta Q,E,H) ]
where:
G = geographic objective,
\Delta Q = physiological disequilibrium,
E = environmental information,
H = memory/history.
This allows the organism to make tradeoffs.
It might temporarily tolerate colder water because that current transports it toward a migratory destination.
Thus equilibrium need not mean instantaneous homeostasis.
Behavior may optimize future states.
8. Predation and the Meaning of the Title
The title Eat or Be Eaten describes the evolutionary pressure that gives sensory systems their importance.
A sensory ability has consequences.
Detecting prey sooner matters.
Detecting predators sooner matters.
Finding mates matters.
Finding food matters.
Locating shelter matters.
Navigating efficiently matters.
Avoiding unnecessary energy expenditure matters.
The game therefore avoids conventional balance whenever possible.
A prey species is not made artificially equivalent to a predator.
Instead, survival emerges from the interaction among different biological capabilities.
A predator may possess extraordinary detection abilities while prey possesses camouflage, group behavior, environmental familiarity, escape performance, or sensory systems optimized for detecting that predator.
Evolution becomes the game's balancing mechanism.
9. Multiplayer Perspective
The multiplayer implementation creates one of the project's most unusual possibilities.
Suppose six players occupy the same ecosystem.
Their displays need not resemble one another.
Player A might receive conventional visual information.
Player B may receive an echo-derived spatial representation.
Player C may have ultraviolet information.
Player D may perceive biologically relevant electrical fields.
Player E may have enhanced thermal information.
Player F may navigate using chemical and magnetic information.
They share:
[ E(t) ]
but receive:
[ R_A,R_B,R_C,R_D,R_E,R_F ]
The game therefore demonstrates experimentally and intuitively that:
[ E(t) \neq R_i(t) ]
The environment is not identical to any organism's representation of it.
10. Translating Nonhuman Senses for Humans
A fundamental design problem is that human players cannot directly experience magnetoreception, electroreception, ultraviolet vision outside our visible spectrum, or many other animal sensory capabilities.
The game must therefore perform sensory translation.
For example, magnetic information could be represented through persistent spatial geometry, texture, orientation, or other visual/auditory transformations.
Electrical fields might appear as localized spatial disturbances whose intensity corresponds to biologically meaningful signal strength.
Chemical gradients might be represented as spatially persistent fields whose intensity, diffusion, and movement follow the simulated chemistry.
Echolocation could temporarily construct environmental geometry following emitted pulses.
Importantly, these are not claims that animals phenomenologically experience colors, shapes, or sounds corresponding to the game's representations.
They are human interfaces to nonhuman information.
The scientific standard should therefore be:
preserve the informational relationships even when the phenomenology cannot be reproduced.
11. Scientific Mode
In addition to ordinary gameplay, Eat or Be Eaten should contain a scientific mode.
Researchers could enable or disable sensory channels.
For example:
[ \text{Salmon}_{\text{full}} ]
could be compared with:
[ \text{Salmon}_{-\text{magnetic}} ]
[ \text{Salmon}_{-\text{olfactory}} ]
[ \text{Salmon}_{-\text{temperature}} ]
or combinations thereof.
Researchers could then compare simulated trajectories.
The same principle applies to other organisms.
If removing a sensory channel causes simulated behavior to diverge substantially from observed biological behavior, that result can identify the importance of the missing information.
Conversely, if unexpectedly simple sensory rules reproduce complex observed behavior, the simulation may suggest hypotheses suitable for empirical testing.
Simulation would not prove that animals use the modeled mechanism.
It would determine whether the mechanism is sufficient to generate particular behavioral patterns under specified conditions.
That is an important scientific distinction.
12. Validation Against Reality
Species models should be evaluated against empirical data whenever available.
Relevant measurements might include:
migration tracks,
diving profiles,
orientation experiments,
sensory thresholds,
feeding behavior,
predator-prey encounters,
habitat selection,
seasonal movements,
laboratory cue-manipulation experiments,
biologging records,
acoustic recordings,
temperature-depth profiles,
ocean-current models,
geomagnetic measurements.
Magnetic navigation provides an especially useful example because experimental manipulation of magnetic fields can cause animals to respond as though geographically displaced. Research across salmon, sea turtles, birds, and other taxa supports the biological importance of magnetic positional information, although the underlying receptor mechanisms and integration of magnetic information remain areas of active investigation.
The simulation should preserve uncertainty rather than conceal it.
Each sensory module could therefore receive an evidence classification:
Established
Strong replicated evidence.
Supported
Substantial evidence but incomplete mechanism.
Provisional
Plausible interpretation with limited evidence.
Experimental
Hypothesis introduced specifically for simulation testing.
This prevents entertainment requirements from being mistaken for established biology.
13. Open-Source Architecture
The project should be developed as open-source infrastructure rather than as a closed biological database.
The core architecture could separate:
Physical Environment Engine
Simulates the shared world.
Species Model
Contains anatomy, movement, metabolism, and behavioral constraints.
Sensory Model
Determines what information from the physical environment is accessible.
Translation Layer
Converts nonhuman sensory information into a human-perceivable interface.
Behavioral Agent
Controls non-player organisms and enables scientific simulation.
Player Interface
Allows human control while restricting the player to species-appropriate information.
Scientific Instrumentation Layer
Records trajectories, sensory inputs, decisions, environmental states, and experimental interventions.
This modularity allows individual research communities to improve particular organisms without rewriting the entire game.
14. Community Species Modules
A scientific open-source project could eventually contain hundreds or thousands of species.
Each species package should contain:
taxonomic identity,
habitat,
sensory systems,
sensory evidence,
detection thresholds where known,
locomotor constraints,
metabolic parameters,
ecological relationships,
behavioral rules,
uncertainties,
references,
validation datasets.
Contributors could propose modifications through version-controlled review.
A marine biologist might improve salmon migration.
An arachnologist might improve spider mechanoreception.
An entomologist might refine bee vision.
A bat researcher might improve echolocation.
A neuroscientist might improve sensory integration.
A game developer might improve the translation of those models into playable experiences.
Thus the project becomes a collaborative library of operational animal perspectives.
15. Artificial Agents
Artificial intelligence introduces another possibility.
Instead of programming every behavioral response explicitly, an artificial agent could be placed inside a species-specific sensory environment and required to survive using only the information available to that organism.
Critically, the agent should not receive the simulator's omniscient state.
It receives:
[ R_i(t) ]
rather than:
[ E(t) ]
The agent therefore confronts the same fundamental problem as the biological organism:
act successfully using incomplete information.
Researchers could then investigate whether recognizable ecological behaviors emerge through learning.
16. The Perspective Experiment
The game's deepest educational feature may be its simplest.
Allow a player to pause the simulation and switch among species occupying the same location.
The underlying physical world does not change.
Only the sensory transformation changes.
Human.
Bee.
Spider.
Snake.
Bat.
Salmon.
Shark.
The forest, river, ocean, or night sky progressively becomes a series of radically different informational spaces.
This provides an intuitive demonstration of a profound biological principle:
an environment can contain information that exists physically but does not exist perceptually for a particular observer.
17. Implications for Consciousness Research
The project cannot determine subjective experience merely by reconstructing sensory input.
Nevertheless, it could provide useful infrastructure for consciousness research because discussions of animal consciousness frequently encounter a fundamental epistemological obstacle:
Humans naturally imagine another organism's world through human senses.
A scientifically constrained simulation can reduce that error.
Before asking:
What is it like to be this organism?
one can first ask:
What information is available to this organism?
The distinction separates an empirically approachable problem from the harder phenomenological problem.
Thus:
[ \text{Physical Environment} \rightarrow \text{Available Information} \rightarrow \text{Neural Integration} \rightarrow \text{Behavior} ]
can be investigated without pretending that:
[ \text{Available Information}
\text{Subjective Experience} ]
The simulation therefore becomes a bridge between sensory ecology and questions concerning animal perspective while maintaining an explicit epistemic boundary around consciousness itself.
18. Falsifiability and Scientific Limitations
The scientific usefulness of Eat or Be Eaten depends on avoiding an attractive but dangerous mistake: assuming that a compelling visualization is biologically correct.
Every visualization is a model.
Every sensory translation introduces human interpretation.
Some sensory mechanisms remain poorly understood.
Behavior rarely follows from one sensory variable.
Learning, genetics, physiology, developmental history, social information, stochastic behavior, and environmental context can all contribute.
Consequently, the game must clearly distinguish:
[ \text{measured} ]
from
[ \text{modeled} ]
from
[ \text{inferred} ]
from
[ \text{speculative}. ]
This distinction transforms uncertainty from a weakness into part of the scientific architecture.
Unknown parameters become experimental variables.
19. From Game to Research Platform
The project could eventually support three overlapping modes.
Game Mode emphasizes survival, exploration, competition, reproduction, and ecosystem interaction.
Education Mode explains sensory systems and permits comparison among species.
Research Mode exposes model parameters, records data, supports controlled manipulations, and permits competing behavioral hypotheses.
The same underlying simulation serves all three.
A child could play as a salmon.
A biology class could examine why the salmon's representation differs from a human's.
A researcher could alter magnetic information and examine resulting navigation.
The transition from entertainment to scientific inquiry occurs without changing the fundamental architecture.
20. Conclusion
Eat or Be Eaten proposes an open-source multispecies simulation built around a deceptively simple observation:
Different organisms share the same physical universe without sharing the same access to it.
Traditional games change characters while preserving the player's world.
This project changes the world presented to the player because the character has changed.
The shark does not merely swim differently.
Its informational environment is different.
The bat does not merely fly.
Space itself is available through different signals.
The salmon does not merely travel through water.
Its navigational environment may contain magnetic, chemical, thermal, pressure, and hydrodynamic structure largely inaccessible to ordinary human perception.
The spider's web is not simply an object upon which it stands.
It can participate in the transmission of information.
The bee does not merely visit the same flower that a human sees.
The sensory information associated with that flower is different.
A scientifically grounded simulation of these differences would provide an unusual intersection of sensory ecology, neuroscience, ethology, movement ecology, evolutionary biology, artificial intelligence, education, consciousness research, and interactive entertainment.
More importantly, it would allow people to experience an approximation of an idea that scientific language alone can make difficult to appreciate:
The world an organism can use is not necessarily the world another organism can perceive.
There is one simulated Earth.
There are many informational worlds within it.
The objective is survival.
The method is perception.
The experiment is perspective.
EAT OR BE EATEN
Would you like to play?
References
Lohmann, K. J., Goforth, K. M., Mackiewicz, A. G., Lim, D. S., & Lohmann, C. M. F. (2022). Magnetic maps in animal navigation. Journal of Comparative Physiology A, 208, 41–67. DOI: 10.1007/s00359-021-01529-8.
Lohmann, K. J., & Lohmann, C. M. F. (2019). There and back again: natal homing by magnetic navigation in sea turtles and salmon. Journal of Experimental Biology, 222, jeb184077. DOI: 10.1242/jeb.184077.
Lohmann, K. J., Lohmann, C. M. F., & Endres, C. S. (2008). The sensory ecology of ocean navigation. Journal of Experimental Biology, 211, 1719–1728. DOI: 10.1242/jeb.015792.
Lohmann, K. J., Putman, N. F., & Lohmann, C. M. F. (2008). Geomagnetic imprinting: A unifying hypothesis of long-distance natal homing in salmon and sea turtles. Proceedings of the National Academy of Sciences.
Quinn, T. P. (1997). Homing in Pacific salmon: mechanisms and ecological basis. Journal of Experimental Biology, 200, 2287–2294.
Schneider, W. T., Holland, R. A., & Lindecke, O. (2023). Over 50 years of behavioural evidence on the magnetic sense in animals: what has been learnt and how? European Physical Journal Special Topics, 232, 269–278.
Walker, M. M., Dennis, T. E., & Kirschvink, J. L. (2002). The magnetic sense and its use in long-distance navigation by animals. Current Opinion in Neurobiology, 12(6), 735–744. DOI: 10.1016/S0959-4388(02)00389-6.
Wiltschko, R., & Wiltschko, W. (2021). The discovery of the use of magnetic navigational information. Journal of Comparative Physiology A.
John Swygert
September 3, 2026
Copyright © John Swygert 2026
TSTOEAO.com
IvoryTowerJournal.com
SecretarySuite.com
The TSTOEAO Room
https://chatgpt.com/g/g-6a6d017c73bc8191a6f5de01f7beab5d-the-tstoeao-room
Ivory Tower Publishing
No comments:
Post a Comment