An ancient evolutionary calculus for attention… — Plain-Language Explanation

Imagine you are at a concert. You aren't just hearing notes; you are watching a performer juggle three invisible balls at the same time: Control, Energy, and Surprise.

This paper proposes that our brains have been wired since the dawn of animal life to pay attention to these three specific things. The authors, Gregory Babbitt and Ernest Fokoue, call this the "Calculus of Attention." They argue that music (and animal songs) isn't just art; it's a biological signal that tells us, "Look at me! I am fit, smart, and in control."

Here is the breakdown of their idea using simple analogies:

1. The Three Invisible Balls (CES)

The authors break down every sound into three mathematical components, which they call CES:

Control (The Tightrope): This is how steady the performer is. Imagine a tightrope walker. If they wobble too much, they fall. If they are too stiff, they look robotic. Good "Control" means hitting the right notes at the right time without slipping.
- In math terms: It's like the first derivative of a function ( $f(x)$ ). It's the "position."
Energy (The Rocket): This is how loud, fast, or intense the sound is. It's the fuel. A whisper has low energy; a drum solo has high energy.
- In math terms: It's the rate of change ( $f'(x)$ ). It's the "speed."
Surprise (The Magic Trick): This is the element of novelty. If a song is predictable, you get bored. If it's random noise, you get confused. "Surprise" is the perfect mix of doing something new that still makes sense.
- In math terms: It's the change in direction ( $f''(x)$ ). It's the "acceleration."

2. The "Stability" Test

The paper's big discovery is about Stability.

Imagine you are drawing a line on a piece of paper.

Brown Noise (Static): If you close your eyes and scribble randomly, your line goes everywhere. It's chaotic. The paper calls this 0% stability.
A Novice Singer: They try to hit the notes but wobble a bit. Their line is a bit shaky.
A Pro Musician: Their line is smooth, intentional, and controlled. They move their "pen" (their voice) with purpose. They don't just wander; they dance within the lines.

The authors built a software tool called POPSTAR (which sounds like a pop star, but is actually a "Point of Performance Signaling and Tracking Analysis Routine"). This software takes a song, breaks it down into those three balls (Control, Energy, Surprise), and plots them on a triangle.

The Result: Professional musicians (and very talented animals like Nightingales and Lyrebirds) keep their "line" incredibly stable. They stay in a specific zone of the triangle. Novices and random noise wander all over the place.

3. Why Do We Care? (The Evolutionary Story)

Why do we love a good singer? The authors suggest this goes back 541 million years to the time when animals first started moving fast and hunting each other (the Ediacaran-Cambrian explosion).

The Ancient Brain: Early animals needed to track fast-moving prey or predators. They needed to calculate: Where is it? (Control), How fast is it going? (Energy), and Is it going to change direction suddenly? (Surprise).
The Modern Connection: Our brains still use this same "calculator" today. When we hear music, we are subconsciously checking: Is this person in control? Do they have the energy? Are they clever enough to surprise me?

If a performer can hold these three balls steady while juggling them, our brains say, "This individual is high quality! They are fit! They are smart!" This is why we feel a connection to a great singer or a complex bird song. It's an honest signal of fitness.

4. Cool Experiments They Did

The team tested this theory on a bunch of different things:

Humans vs. Animals: They found that professional opera singers and jazz musicians have "stability" scores just as high as (or even higher than) the most complex bird songs.
Live vs. Studio: They looked at the singer Björk. In the studio, she could experiment wildly and change her style a lot. But when she performed live in front of a crowd, her "stability" went up. The audience forced her to be more consistent. The crowd acts like a filter, demanding a higher level of "fitness."
Experts vs. Newbies: Expert singers (and even adult canaries) are much more stable than beginners. Beginners are like a toddler learning to walk—lots of wobbles. Experts are like Olympic gymnasts—smooth and precise.
Concertos vs. Background Music: They compared difficult piano concertos (where the soloist is the star) to "ambient" piano music (like furniture music meant to be background). The concertos had much higher "stability." The music was demanding your attention; the background music was letting you drift off.

5. The Big Picture

The paper ends with a fascinating thought: This "Calculus of Attention" isn't just for music. It might be the blueprint for how we understand everything.

Sports: A basketball player is showing Control (dribbling), Energy (running), and Surprise (a fake-out move).
Politics: A politician uses Control (steady voice), Energy (passion), and Surprise (a new idea) to hold your attention.

The authors warn that in the age of AI and social media, algorithms are learning to hack this ancient "Calculus." They can create content that hits those three buttons perfectly to keep us glued to our screens, sometimes even better than a real human can.

In short: Music is a biological test. When you listen to a song, your brain is running a math equation to see if the performer is "fit." The best performers are the ones who can keep the math balanced, stable, and surprising, just like our ancestors needed to do to survive in the wild.

1. Problem Statement

The paper addresses a gap in evolutionary biology and musicology regarding the proximate mechanisms by which musical performance signals individual fitness (quality). While the ultimate evolutionary drivers of music (e.g., social bonding, sexual selection) are well-studied, the specific acoustic features that honestly signal a performer's physical and cognitive quality to an audience remain undefined.

The authors hypothesize that attention signaling is an ancient evolutionary trait rooted in the Ediacaran-Cambrian transition (~541 million years ago), where organisms evolved to detect swift, directed movement. They propose that modern music and animal vocalizations retain a "calculus of attention" based on three core components:

Control (Position): Motor precision.
Energy (Change in Position): Kinetic/Potential energy and velocity.
Surprise (Change in Direction): Cognitive complexity and novelty.

The core problem is to mathematically define, measure, and validate whether "high-quality" performances (expert musicians, complex animal songs) exhibit a specific, non-random stability in these three dimensions compared to novices or non-musical sounds.

2. Methodology

The authors developed a computational framework and software tool called POPSTAR (Project for Observation of Performance Signaling Through Audio-visual Representation) to analyze audio files.

A. Feature Extraction (The CES Calculus)

The software segments audio into time windows (default 8s or beat-based) and extracts nine acoustic features grouped into three categories:

Control (C):
- Pitch Control (PC): Deviation of fundamental frequency ( $f_0$ ) from expected Western scale or cluster means.
- Harmonic Control (HC): Summed harmonic energy levels above $f_0$ .
- Timing Control (TC): Deviation of observed beat intervals from mean beat intervals.
Energy (E):
- Tempo (TP): Beats per minute.
- Reverberation (RV): Pearson correlation at lag 1 (autocorrelation peak).
- Amplitude (AM): Logarithmic sum of signal amplitude.
Surprise (S):
- Multi-scale Entropy (MSE): Complexity index using refined multi-scale entropy.
- Lempel-Ziv Complexity (LZC): Complexity of the binary representation of the audio.
- Note Variability Index (NVI): Spectral cross-correlation matrix to measure note diversity.

B. Data Representation and Analysis

Ternary Space Mapping: The normalized averages of C, E, and S for each time segment are plotted as a trajectory in a ternary space (summing to 1).
Dynamic Visualization: The software generates dynamic ternary plots and Chernoff faces (where facial features map to specific acoustic variables) to visualize high-dimensional data intuitively.
Stability Indicator ( $\eta$ ): The core metric is the stability of the trajectory. The authors use a permutation-based approach:
1. Calculate step distances ( $\delta_t$ ) between consecutive points in the observed CES trajectory.
2. Randomly shuffle the time order of points to create a null distribution of step distances ( $\delta_{\sigma}$ ).
3. Compare observed steps to shuffled steps. High stability is defined by a high percentage of observed steps being smaller than shuffled steps (indicating intentional, non-random movement).
Comparative Analysis:
- Functional Data Analysis (FDA): B-spline functions are fitted to trajectories and aligned using Fisher-Rao elastic registration to compare shapes.
- Network Analysis: Adjacency networks are built using L2 distances between song trajectories to map individuality and genre clustering.
- Machine Learning: Random Forest classifiers (500 trees) are used to distinguish between groups (e.g., Expert vs. Novice, Live vs. Studio) based on the 9 CES features.

3. Key Contributions

The CES Framework: Proposes a unified mathematical calculus (Control, Energy, Surprise) that bridges animal behavioral ecology and human music theory.
POPSTAR Software: An open-source tool that automates the extraction of CES features, generates dynamic visualizations (ternary plots/Chernoff faces), and calculates stability metrics.
Stability as a Fitness Signal: Demonstrates that "musicality" and "fitness" can be quantified as the non-random stability of a performer's trajectory in the CES ternary space.
Cross-Species Validation: Validates that this mechanism is not unique to humans but is shared across diverse taxa (birds, frogs, primates) and correlates with perceived quality.

4. Key Results

Stability Correlates with Musicality:
- Music vs. Noise: Professional music exhibits high CES stability (>60-70%), whereas brown noise has near 0% stability.
- Expert vs. Novice: Expert human singers and adult Yellow Canaries show significantly higher CES stability and lower "surprise" (randomness) than novices/immature birds. Random Forest classifiers achieved 99% accuracy in distinguishing experts from novices.
- Animal Vocalizations: Complex songbirds (Lyrebirds, Nightingales) and gibbons show CES stability comparable to or exceeding some human singers.
Genre and Performance Context:
- Live vs. Studio: Live performances (e.g., Björk) show higher control and lower surprise compared to studio recordings, suggesting audience feedback constrains the performer to a more stable, "honest" signal.
- Concertos vs. Ambient: Piano concertos (emphasizing soloist virtuosity) have significantly higher CES stability than ambient piano pieces (which emphasize listener environment), despite similar levels of C, E, and S.
- Genre Distinctiveness: Opera and Jazz show distinct CES signatures. Opera clusters closer to animal vocalizations (due to laryngeal similarities with anurans), while Pop music shows wider variance and relies more on technological augmentation.
Neurological Implications: The study suggests a link between the Dorsal Attention Network (processing external CES stimuli) and the Default Mode Network (internal self-awareness), proposing a geometric model where external performance balances internal cognitive states.

5. Significance

Evolutionary Continuity: The findings suggest that the mechanisms for attention signaling in music are not cultural artifacts but deep evolutionary adaptations originating from the Cambrian explosion, shared across species that evolved singing independently (convergent evolution).
Honest Signaling: The study provides a mathematical basis for "honest signaling" in performance. High stability in the CES space is difficult to fake without genuine motor control and cognitive reserve, explaining why audiences prefer live, unassisted performances.
Interdisciplinary Bridge: It connects behavioral ecology, acoustics, mathematics (calculus/chaos theory), and neuroscience.
Societal Application: The authors warn that understanding the "calculus of attention" is crucial in the age of Generative AI. If AI can manipulate C, E, and S to maximize attention, it could exploit these ancient neural pathways. Conversely, this framework could be used to design better educational tools for musicians or to detect AI-generated content that lacks the "stability" of genuine biological performance.

In summary, the paper posits that music is a sophisticated manipulation of an ancient biological calculus designed to capture attention, and that the "quality" of a performance is mathematically measurable by the stability and intentionality of its movement through the space of Control, Energy, and Surprise.

An ancient evolutionary calculus for attention signaling retained in modern music