Display functions of dinosaur proto-wings before powered flight

This study demonstrates through computer animations and neural response modeling that early pennaceous feathers on non-volant dinosaur proto-wings and tails functioned primarily to enhance the efficiency of visual motion displays before being exapted for powered flight.

The Gist

The Big Question: Why Did Dinosaurs Grow Feathers?

For a long time, scientists have argued about why early dinosaurs (the ancestors of birds) grew feathers. The obvious answer is flight. But here's the problem: the earliest dinosaurs with feathers, like Caudipteryx, had tiny, stiff wings that were way too small to lift their heavy bodies into the air. They couldn't fly.

So, if they couldn't fly, what were those feathers for?

This paper proposes a new theory: The feathers were originally "flashy signs" used to scare prey. Think of them less like airplane wings and more like a magician's cape or a bullfighter's red cloth.

The Experiment: The "Dinosaur Robot" vs. The "Locust Alarm"

Since we can't ask a 100-million-year-old dinosaur what it was thinking, the researchers used a clever workaround. They treated the dinosaur's prey (insects) as the test subjects.

1. The Model: They created computer animations of a small, feathered dinosaur. They made two versions: one with fluffy proto-wings and a feathered tail, and one that was just a naked, scaly lizard.
2. The Action: They animated these dinosaurs doing a specific move called a "flush-display." This is a move seen in modern birds (like roadrunners) where they suddenly spread their wings and tail to startle a hiding insect, making it jump out so the bird can catch it.
3. The Receiver: They showed these animations to locusts (grasshoppers). Locusts have a super-fast "panic button" in their brains called the DCMD neuron. When a predator looks like it's about to pounce, this neuron fires, telling the locust to jump away.
4. The Measurement: The researchers didn't just watch if the locust jumped; they plugged electrodes into the locust's brain to measure exactly how hard that "panic button" was pressed.

The Results: Feathers = Super Scare

The results were clear and exciting:

• The Naked Lizard: When the naked dinosaur animation moved, the locusts' panic neurons fired a little bit.
• The Feathered Dinosaur: When the same dinosaur moved but had feathers, the panic neurons went crazy. The firing rate was much higher, and the locusts were much more likely to jump.

The Analogy: Imagine you are hiding in a closet.

• Scenario A: A person walks up to the door and opens it slowly. You might peek out.
• Scenario B: A person walks up, and suddenly throws a giant, bright red cape over the door, making a loud whoosh sound. You will definitely jump back in terror.

The feathers acted like that giant red cape. Even though the dinosaur couldn't fly, the feathers made its movements look much bigger, faster, and more threatening to its prey.

The "Double-Duty" Evolution

This is where the story gets really cool. The paper suggests that evolution is a bit of a "happy accident."

1. Step 1 (The Display): Dinosaurs grew feathers to look scary and flush out prey. The bigger and stiffer the feathers, the better the scare.
2. Step 2 (The Bonus): As they evolved bigger, stiffer feathers to be better at scaring prey, they accidentally got better at aerodynamics.
   • Analogy: Imagine a person wearing a giant, stiff raincoat to look bigger and scarier in a fight. Eventually, they realize that if they run fast enough, that same raincoat acts like a parachute, helping them glide. They didn't build the coat to fly; they built it to look tough, but flying became a useful side effect.

The Conclusion

The paper concludes that feathers didn't evolve for flight first. They evolved for communication and hunting.

• They were used to scare prey (flush-displays).
• They were likely used to show off to mates or fight rivals (like a peacock's tail).
• Only after these feathers became large and stiff enough to be great visual signals did nature "exapt" (co-opt) them for flight.

In short: The first birds didn't learn to fly because they wanted to; they learned to fly because they were already so good at waving their fancy feathered arms to scare bugs that the wind eventually caught them.

Technical Summary

1. Problem Statement

The evolutionary origin of pennaceous feathers (feathers with a stiff central rachis and vanes) in non-volant (non-flying) dinosaurs remains a subject of debate. While modern bird feathers are adapted for flight, early pennaceous feathers found in basal pennaraptoran dinosaurs (e.g., Caudipteryx, Protarcheopteryx) were likely too small and structurally unsuited for generating sufficient aerodynamic lift or thrust for powered flight.

Previous hypotheses suggested these proto-wings were used for:

• Aerodynamic functions (e.g., wing-assisted inclined running, drag-based maneuvering).
• Visual signaling (display functions).

However, the visual display hypothesis has lacked empirical testing because it concerns the behavior of extinct organisms. Specifically, it is unclear if the limited range of motion in early pennaraptoran forelimbs could generate visual stimuli strong enough to trigger escape responses in prey or conspecifics, thereby providing a selective advantage for the evolution of these structures.

2. Methodology

The authors employed a multi-disciplinary approach combining paleontological reconstruction, computer animation, and neurophysiology to test the efficacy of dinosaurian visual displays.

• Model System: The study utilized the LGMD/DCMD pathway (Lobula Giant Movement Detector/Descending Contralateral Movement Detector) in the locust (Locusta migratoria). This neural circuit is a well-established model for detecting looming stimuli and accelerated edge translation, triggering innate escape jumps. The DCMD response is a quantifiable proxy for the probability of prey escape behavior.
• Stimulus Generation:
  • Researchers created 3D computer animations of a hypothetical early pennaraptoran (based on the morphology of Caudipteryx) performing "flush-displays."
  • These displays were modeled after three extant flush-pursuing birds: the Greater Roadrunner (Geococcyx californianus), the Rufous-tailed Scrub Robin (Cercotrichas galactotes), and the Northern Mockingbird (Mimus polyglottos).
  • Variables: Animations were generated with and without pennaceous feathers on the forelimbs (proto-wings) and tail. They were presented at two simulated distances (60 cm "close" and 120 cm "distant").
• Experimental Protocol:
  • Six female locusts were subjected to 16 recordings each (4 animations $\times$ 4 repetitions).
  • Extracellular recordings were taken from the right DCMD neuron while the left eye viewed the animations on a monitor.
  • Metrics: The study measured the peak spiking rate (spikes/20 ms) and the frequency of reaching a hypothetical escape threshold (5 spikes/20 ms, indicative of jump preparation).
• Data Analysis:
  • Visual stimulus properties were quantified, including angular size, angular velocity, and the acceleration of white-to-black pixel transitions (a key trigger for the DCMD).
  • Statistical analysis used Generalized Linear Mixed-Effects Models (GLMM) to assess the effects of feather presence, distance, and display phase on neural responses.

3. Key Contributions

• Empirical Validation of the Display Hypothesis: This is the first study to empirically test the visual display hypothesis for early pennaceous feathers using a neurophysiological model of prey response.
• Quantification of "Proto-Wing" Efficacy: The study demonstrates that even with the limited anatomical range of motion of early pennaraptorans, the presence of pennaceous feathers significantly enhances the visual stimulus intensity.
• Mechanism Identification: The research identifies acceleration of edge transitions (specifically white-to-black pixel changes) as the primary visual cue driving the enhanced neural response, rather than just static size or speed.
• Integration of Signaling and Aerodynamics: The paper proposes a framework where signaling functions (flush-displays) and non-signaling functions (drag/lift during pursuit) are intrinsically linked, driving the co-evolution of feathered surfaces and limb mobility.

4. Results

• Enhanced Neural Response: Across all three bird-model animations, the presence of feathers consistently increased the peak DCMD spiking rates and the probability of reaching the escape threshold compared to animations without feathers.
• Distance Effects:
  • In Roadrunner and Scrub Robin models, feathers increased response efficacy at both close and distant ranges.
  • In the Mockingbird model, feathers significantly increased peak spiking rates specifically at distant ranges, suggesting that feathers extend the effective range of the visual signal.
• Motion Dynamics: The strongest correlation with DCMD activity was found with the acceleration of white-to-black pixel transitions. The stiff rachis and broad vanes of pennaceous feathers maintained shape during rapid movement, creating sharper, more accelerated visual edges that effectively triggered the prey's escape circuit.
• Walking vs. Stationary: Adding a "walking" movement prior to the display slightly reduced peak spiking rates, but the presence of feathers still maintained a significant advantage over featherless models.
• Statistical Significance: Generalized linear models confirmed that "feather presence" was a highly significant predictor of both peak spiking rate and threshold crossing (P < 0.001 in most cases).

5. Significance and Implications

• Evolutionary Pathway: The findings support the hypothesis that pennaceous feathers initially evolved for visual signaling (specifically flush-displays to flush prey) rather than aerodynamics. The enhanced visual stimulus provided a direct survival advantage by increasing foraging success.
• Exaptation for Flight: The study suggests a mechanism for exaptation. As feathers evolved to be larger and stiffer for better visual signaling, they simultaneously improved aerodynamic properties (drag and lift). The "flush-display" context inherently links signaling with the physical act of pursuit, creating a reinforcing feedback loop where improvements in display efficacy also improved maneuverability.
• Multifunctionality: The paper argues that early pennaceous feathers were likely multifunctional, serving diverse signaling contexts (conspecifics, predators, prey) before being co-opted for powered flight in later avialans.
• Methodological Innovation: The use of modern neural circuitry (locust DCMD) as a proxy for extinct prey behavior offers a novel, quantifiable method for testing paleobiological hypotheses regarding the function of extinct anatomical structures.

In conclusion, the paper provides robust empirical evidence that early pennaceous proto-wings and tails were highly effective visual tools for flushing prey, offering a plausible selective pressure for their evolution prior to the origin of flight.