🧬 biology

From Semiotic Fitting to Semiotic Collapse: A Multi-Level Process Model of Extinction

This paper proposes and empirically validates a multi-level process model of extinction as the gradual collapse of semiotic systems, demonstrating that a measurable index of semiotic overlap (SO) across six biological kingdoms serves as a superior predictive indicator of population decline compared to classical demographic or variance-based early-warning signals.

Original authors: Amir Mohammad Falak Aflaki

Published 2026-07-06

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Amir Mohammad Falak Aflaki

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

The Big Idea: Extinction is a Broken Conversation, Not Just a Empty Room

Usually, when scientists talk about extinction, they look at a counting problem. They watch a population of animals get smaller and smaller until there are none left. It's like watching a theater empty out; when the last person leaves, the show is over.

This paper argues that extinction actually starts much earlier. It's not just about the numbers dropping; it's about the conversation breaking down.

Imagine every animal has a "language" it uses to understand the world. A bird hears a specific song (a cue) that tells it, "It's spring, time to lay eggs." A hare sees the days getting shorter (a cue) that tells it, "Time to grow a white coat."

The author, Amir Mohammad Falak Aflaki, suggests that extinction happens when the animal stops understanding its environment. The "language" becomes gibberish. The bird hears the song, but it no longer means "spring" anymore. The animal is still there, but it is lost. The paper calls this process "Semiotic Collapse."

The Three Parts of the "Connection Score"

To measure how well an animal is "listening" to its world, the author created a score called Semiotic Overlap (SO). Think of this score like a "Wi-Fi signal strength" for an animal's relationship with nature. It has three parts:

The Signal (SO_I): Does the clue still match the result?
- Analogy: Imagine a weather app that used to be 100% accurate. If it says "sunny" and it rains every time, the app is broken. In nature, if a bird hears "warm weather" but the food (caterpillars) isn't there yet, the signal is broken.
The Variety (SO_B): Are the animals all doing the exact same thing?
- Analogy: Imagine a crowd of people. If everyone is wearing the exact same outfit and walking in the exact same direction, they are vulnerable. If they are dressed differently and walking in different directions, the group is safer. If a whole population of hares all turn white at the exact same time, but the snow doesn't come, they all starve. They need "variety" to survive.
The Network (SO_C): Do the different clues agree with each other?
- Analogy: Imagine you are trying to navigate a city. If the street signs, the GPS, and the landmarks all point the same way, you are good. If the street sign says "North" but the GPS says "South," the map is broken. Nature uses many clues (temperature, day length, snow); if they start contradicting each other, the animal gets confused.

The Five Rules of Collapse

The paper tests five ideas (propositions) using data from nine different animals and plants (from bacteria to bears). Here is what they found:

1. The "Canary in the Coal Mine" Rule

The Claim: The "connection score" (SO) drops before the population numbers drop.
The Finding: In the Winter Moth, the connection score started falling a full year before the moth population crashed. It was a warning sign. However, in the Great Tit (a bird), the score didn't drop first because the bird population is very good at "buffering" (absorbing the shock) for a while.

2. The "Clustering" Danger Zone

The Claim: If animals stop being diverse and start acting exactly the same (synchronizing), they are in trouble.
The Finding: The Snowshoe Hare was a perfect example of this. They all turned white at the exact same time (low variety, high synchronization). Because they were all doing the same thing, when the snow didn't come, they were all exposed and vulnerable. In contrast, the Great Crested Newt had high variety and didn't synchronize, which kept them safe.

3. The "Rigidity vs. Speed" Rule

The Claim: If the environment changes too fast, animals that are too rigid (stuck in their ways) will fail.
The Finding:
- Sea Turtles: Their sex is determined by sand temperature. If the sand gets too hot too fast, they can't adapt, and their "connection" breaks.
- Boechera stricta (a plant): This plant is very flexible. It changes its flowering time quickly to match the weather. It stayed healthy because it could "fit" the new reality faster than the environment changed.

4. The "Ghost of the Past" (Hysteresis)

The Claim: Even if a population grows back in numbers, it might still be "broken" inside.
The Finding: The Eurasian Otter was hunted to near extinction and then recovered. Their numbers are up, but their genetic diversity is still low. It's like a house that has been rebuilt with the same old, weak foundation. They look fine on the outside (numbers), but they are still fragile on the inside (genetics).

5. The "Better Warning System"

The Claim: This new "connection score" is a better warning sign than the old ways of measuring extinction.
The Finding: In the Winter Moth, the old methods (looking at how much the population numbers wiggle up and down) didn't warn them until it was too late. The new "connection score" warned them a year earlier. It found the problem before the population even started to shrink.

The Two Extreme Stories

The paper uses two specific examples to show the difference between "natural fitting" and "forced selection":

The Plant (Boechera stricta): Nature works like a dance. The plant "chooses" how to react to the weather (fitting), and then evolution catches up later. The plant is dancing with the music.
The Salmon (Cedar River Sockeye): Humans stepped in and forced the salmon to change. They bred them in hatcheries to spawn earlier. This was like forcing the salmon to dance to a different beat. The result? The salmon lost their natural connection to the river. They stopped "fitting" the environment, and their numbers crashed from 213,000 to 45,000. The paper says this is what happens when you force an animal to ignore its own instincts.

The Bottom Line

This paper suggests that we shouldn't just count how many animals are left. We need to check if they still "speak the language" of their environment.

If the signal is broken (the weather doesn't match the food),
If the variety is gone (everyone is doing the same thing),
Or if the map is confusing (clues contradict each other),

...then the animal is on the path to extinction, even if there are still a lot of them around. The "conversation" has ended, and the silence is coming.

Technical Summary: From Semiotic Fitting to Semiotic Collapse

Problem Statement
Conventional extinction analysis treats disappearance as a terminal demographic event, relying on population viability analysis and early-warning signals (EWS) derived from dynamical variables (abundance, range, variance). This framework lacks the conceptual vocabulary to address the informational dimension of decline: the gradual erosion of the relationship between an organism's interpretive system and the environmental cues it relies on. While biosemiotics (specifically Kalevi Kull's theory of semiotic fitting or Einpassung) posits that organisms establish recognition-based bonds with their environment distinct from slow Darwinian adaptation (Anpassung), these concepts have remained qualitative. There is no operational bridge to measure the decay of these bonds using quantitative ecological time series, leaving phenomena like phenological mismatch, ecological traps, and artificial selection maladaptation ununified under a single theoretical machinery.

Methodology
The study proposes a multi-level process model operationalizing Kull's semiotic fitting into a measurable index: Semiotic Overlap (SO). SO quantifies the structural alignment between an organism's sign-interpreting system and its ecological environment. It is decomposed into three components, all computable from open-access ecological data:

SO_I (Informational Alignment): The absolute Pearson correlation ( $|r|$ ) between an environmental cue and the biological outcome it predicts (e.g., spring temperature vs. caterpillar peak), measured over a moving 5-year window.
SO_B (Behavioural Dispersion): A measure of within-population diversity in response to cues, calculated via standard deviation (normalized) or Shannon entropy ( $H(B)$ ) of behavioural distributions.
SO_C (Structural Coherence): The internal consistency of the cue network, measured either by the mean pairwise correlation of cues across spatial sites or by the residual coherence of a cue–outcome regression.

The composite index is $SO = (SO_I + SO_B + SO_C) / 3$ .

The framework was tested against nine independent ecological systems spanning six biological kingdoms (prokaryotes, plants, insects, fish, amphibians, reptiles, birds, mammals) and three continents. The study derived five formal propositions regarding SO as a leading indicator, behavioural clustering thresholds, rate-by-rigidity interactions, hysteresis in recovery, and the predictive superiority of SO over classical variance-based EWS. Statistical tests included Granger-style F-tests, lagged correlations, Welch's t-tests, and entropy calculations.

Key Contributions

Operationalization of Biosemiotics: The paper translates Kull's qualitative theory of Einpassung (fitting) into a quantitative, falsifiable framework, bridging the gap between biosemiotic theory and quantitative ecology.
The Semiotic Overlap (SO) Index: Introduces a composite metric that captures the informational integrity of ecological systems, distinct from demographic or dynamical metrics.
Phase-Dependent Extinction Model: Proposes that extinction is a staged process (Silent Divergence $\to$ Threshold Crossing $\to$ Structural Collapse $\to$ Hysteresis) where the predictive utility of SO depends on the system's phase (e.g., buffered vs. collapsing).
Distinction between Fitting and Adaptation: Provides empirical evidence supporting the claim that "choice comes first, selection may work afterwards," demonstrating that rapid plastic fitting often outweighs slow evolutionary adaptation in maintaining cue–outcome relationships.

Results

Leading Indicator Capability: In the winter moth (Operophtera brumata), SO acted as a significant leading indicator of population decline (Granger $F = 4.54, p = 0.049, \Delta R^2 = +0.108$ ), predicting population state one year in advance. Conversely, in the demographically buffered great tit (Parus major), SO did not lead, supporting the phase-dependence hypothesis.
Clustering Threshold: The snowshoe hare (Lepus americanus) exhibited behavioural entropy ( $H(B)$ ) well below the proposed 1.5-bit threshold (0.27–0.63 bits) with high inter-individual synchronisation ( $\bar{r} = +0.86$ to $+0.99$ ), correlating with population collapse. In contrast, the great crested newt (Triturus cristatus) showed high entropy and low synchrony, shielding the metapopulation.
Rate-by-Rigidity Interaction: Systems with rigid cue–response architectures (e.g., sea turtles with temperature-dependent sex determination) showed reduced SO in high-warming environments. The Cedar River sockeye salmon (Oncorhynchus nerka) demonstrated that artificial selection (hatchery propagation) can override natural fitting, causing a 62% weakening of cue–outcome coupling and a demographic crash.
Hysteresis in Recovery: The Eurasian otter (Lutra lutra) showed demographic recovery (spatial expansion) but persistent genetic/interpretive lag, evidenced by a sustained gap between expected and observed heterozygosity ( $H_{exp} - H_{obs} \approx 0.125$ ) after two decades.
Superiority over Classical EWS: In the winter moth, SO provided predictive information ( $\Delta R^2 = +0.109$ ) that classical variance-based EWS did not ( $\Delta R^2 = +0.015$ ), indicating the informational dimension is not redundant with dynamical dimensions.
Cross-Taxonomic Ordering: Systems were ordered by SO value (ranging from 0.35 for the snowshoe hare to 0.68 for Boechera stricta) rather than taxonomy, with the same fish species appearing in different collapse phases depending on management regimes.

Significance and Claims
The paper modestly claims to demonstrate that semiotic fitting is a measurable quantity that behaves as predicted across taxonomically diverse systems. It argues that the informational dimension of ecological systems is not redundant with their dynamical or demographic dimensions. The study does not claim to have established a final, unified theory of extinction but offers a "preliminary empirical anchoring" and an invitation to test these operationalized biosemiotic concepts against the vast existing base of long-term ecological monitoring data. The primary significance lies in shifting the focus of extinction analysis from the disappearance of individuals to the dissolution of coded relations (recognition-based bonds) that precede demographic collapse.