← BACK TO PITCH DECK

Open Book

MIGUEL · Screenplay-Level Tonal Analysis with a Neuroscience-Validated Methodology

This document explains, step by step, how tonal proximity between MIGUEL and its comparables was measured at the screenplay level — before a single frame is shot. The methodology is calibrated against a public functional neuroimaging dataset, and every source is independently verifiable.

CONTENTS
  1. Overview
  2. The problem of comparing films at the screenplay level
  3. The three Lettieri dimensions
  4. Pulp Fiction as scientific control
  5. The three similarity metrics
  6. Results: the four matrices
  7. Dramaturgical interpretation
  8. Technical reproducibility
  9. Downloadable dataset
  10. Bibliography and sources

1. Overview

The analysis positions MIGUEL relative to three tonal anchors (Memories of Murder, Marshland, Black Bread) and one scientific control (Pulp Fiction). Proximity is computed across three emotional dimensions that neuroscience has validated as universal: polarity, complexity, and intensity (Lettieri et al., Nature Communications, 2019).

KEY FINDING
Working from the five screenplays, MIGUEL forms a compact tonal cluster with its three anchors (Wasserstein similarity 0.84–0.85) and separates clearly from the scientific control, Pulp Fiction (0.77). The system discriminates real dramaturgical signatures, not surface resemblances.

The choice of Pulp Fiction is not aesthetic. It is the only film among the five for which a measured fMRI brain response exists, recorded from 86 subjects watching it in full [1, 2]. That makes it a uniquely qualified control: the system can be calibrated against publicly available physiological data.

2. The problem of comparing films at the screenplay level

When a producer says "my film has the tone of X," they usually mean it as a felt impression: they have seen X, read the screenplay, and intuit the resemblance. That intuition has value, but it is not verifiable. A system that aims to measure tonal proximity between films — at the script level, before a single frame is shot — has to solve two problems.

PROBLEM 1 — WHAT TO MEASURE

It is not enough to tag generic emotions ("sad," "tense"). Such labels depend on the observer. Affective neuroscience has shown that three psychological dimensions account for 85% of the emotional variance when subjects watch cinema naturalistically, and that those dimensions predict brain activity in independent subjects.

PROBLEM 2 — HOW TO COMPARE TWO CURVES

Two films can produce emotional curves with the same temporal shape but live in different regions of the affective space. Classical correlation (Pearson) is blind to magnitude. It must be combined with a metric that measures the distribution itself, not just the synchrony of its oscillations.

3. The three Lettieri dimensions

Lettieri and colleagues (IMT School for Advanced Studies, Lucca) ran the following experiment. Twelve Italian subjects continuously rated their emotional experience while watching Forrest Gump; in parallel, fourteen independent German subjects watched the same film inside an fMRI scanner. Principal component analysis on the Italian behavioral ratings revealed three orthogonal dimensions that predicted brain activity in the German cohort [3]. This independence between behavioral source and neural prediction is what validates these three dimensions as universal rather than film-specific.

POLARITY
45%of variance
Range −10 to +10. Pure emotional valence: how negative or positive the experience is. The negative pole runs from unease to terror; the positive pole, from calm to full joy.
COMPLEXITY
24%of variance
Range −10 to +10. How much cognitive processing the scene demands. Positive = ambivalence, moral dilemma, dramatic irony. Negative = a primitive, unmediated response: visceral fear, revulsion, flight.
INTENSITY
16%of variance
Range 0 to 10. The absolute force of the experience, regardless of sign. A cathartic climax and an act of extreme violence can share peak intensity despite opposite polarities.

These three dimensions map onto spatially distinct zones of the right temporo-parietal cortex, organized as parallel gradients in much the way the visual cortex organizes retinal position. The authors propose the term emotionotopy for this topographic organization of emotion [3]. Although neither experiment used a screenplay-based stimulus, the resulting three-dimensional taxonomy is general: any time-varying emotional signal — including the beat-level emotional content of a screenplay — can be projected onto these three axes.

4. Pulp Fiction as scientific control

The Naturalistic Neuroimaging Database (NNDb), maintained by the LAB Lab at University College London, published a dataset in 2020 that recorded fMRI brain activity from 86 subjects watching ten complete commercial feature films (between 91 and 154 minutes each) [1, 2]. Pulp Fiction (Quentin Tarantino, 1994, 148 minutes) is one of those ten.

That makes Pulp Fiction a uniquely qualified control: it is the only film among the five analyzed here for which measured brain response data is publicly available. If the system discriminates well, it should separate Pulp Fiction from MIGUEL's cluster of moral-thriller anchors, because Pulp Fiction is structurally very different: non-linear narrative, no guilt-redemption axis, violence and humor woven together.

To reconstruct Pulp Fiction beat by beat we downloaded the official NNDb CSV containing all 16,155 spoken words of the film, each with a validated sub-second timestamp [4]. Those words were then grouped into 140 temporal beats using silence gaps greater than 4 seconds as a proxy for scene transitions. Each beat preserves its real start and end times in the film, so every annotation is directly alignable with the published fMRI activity.

5. The three similarity metrics

Each film produces three curves — one per Lettieri dimension — over its real duration in seconds. Comparing two curves with a single metric is insufficient: each metric captures a different aspect of similarity. The system combines three.

PEARSON
"Do they rise and fall together?"
Linear correlation between two curves resampled to equal length. Measures synchrony of temporal shape. Blind to absolute magnitude: two curves living in opposite regions of the emotional space can correlate highly if their oscillations align.
DTW
"Are the peaks at the same minutes?"
Dynamic Time Warping. Aligns both curves allowing small temporal displacements and measures the resulting minimum distance. Robust to differences in duration. Captures temporal structure.
WASSERSTEIN
"Do they live in the same emotional neighborhood?"
Earth Mover's Distance. Measures how much "mass" has to be moved to transform one distribution into another. Independent of temporal shape: captures where each curve lives in the affective space, not how it varies over time.

The three metrics are combined via geometric mean. The choice is deliberate: the geometric mean penalizes disparity. A genuinely robust similarity requires a high score in all three metrics. A single weak metric drags the combined score down, preventing an extreme value in one dimension from masking weakness in another.

Technical note. Within each metric, the Lettieri dimensions are weighted by the variance they explain in the original paper: polarity 52.9%, complexity 28.2%, intensity 18.8% (proportional to 45/24/16). This reflects the relative contribution of each dimension to total emotional variance.

6. Results: the four matrices

The four symmetric similarity matrices between the five films follow. Green diagonal = self-identity (1.00). Amber = tonal cluster of MIGUEL + 3 anchors. Cells involving Pulp Fiction are shaded to mark its role as control.

Pearson Matrix (Temporal Shape)

Pulp FictionMiguelMemoriesMarshlandBlack Bread
Pulp Fiction1.000.440.550.450.47
Miguel0.441.000.530.520.56
Memories of Murder0.550.531.000.520.55
Marshland0.450.520.521.000.66
Black Bread0.470.560.550.661.00

Five films with markedly different temporal signatures (all in 0.44–0.66). Pearson confirms that no film "copies" another; each has its own emotional rhythm. The MIGUEL–Pulp Fiction similarity is the lowest in the matrix (0.44).

DTW Matrix (Temporal Alignment)

Pulp FictionMiguelMemoriesMarshlandBlack Bread
Pulp Fiction1.000.960.940.920.94
Miguel0.961.000.950.940.94
Memories of Murder0.940.951.000.980.98
Marshland0.920.940.981.000.98
Black Bread0.940.940.980.981.00

All five films have a similar peak structure (all alignable with DTW > 0.92). This metric is less discriminative on this sample because all five share similar variability. DTW contributes robustness but does not separate the films on its own.

Wasserstein Matrix (Distribution in Emotional Space)

Pulp FictionMiguelMemoriesMarshlandBlack Bread
Pulp Fiction1.000.910.790.800.77
Miguel0.911.000.850.850.84
Memories of Murder0.790.851.000.940.97
Marshland0.800.850.941.000.96
Black Bread0.770.840.970.961.00

Here two clear clusters emerge. Memories of Murder + Marshland + Black Bread form a tight cluster (0.94–0.97). MIGUEL sits in a hinge position: it shares intensity range with Pulp Fiction (0.91) and qualitative emotional signature with the moral anchors (0.84–0.85). Pulp Fiction remains separated from the moral cluster (0.77–0.80).

Combined Matrix (Geometric Mean)

Pulp FictionMiguelMemoriesMarshlandBlack Bread
Pulp Fiction1.000.730.740.690.70
Miguel0.731.000.760.750.76
Memories of Murder0.740.761.000.780.81
Marshland0.690.750.781.000.85
Black Bread0.700.760.810.851.00

The combined (conservative) metric confirms the tonal cluster of MIGUEL with its anchors and the separation from the control. The result is stable and does not depend on any single metric.

7. Dramaturgical interpretation

What appears numerically aligns with what an experienced festival programmer or buyer would recognize on viewing the five films:

MIGUEL's hinge position is not a weakness — it is the asset. It allows the project to converse with three award-winning anchors (10 Goyas, 9 Goyas, international festival recognition) without being a replica of any of them.

8. Technical reproducibility

Annotation modelClaude Sonnet 4.5 (Anthropic). Unified prompt applied identically to all five films. Output sanitization to prevent parse errors (multiple JSON blocks, code fences).
Beats per screenplayMIGUEL 119 · Pulp Fiction 140 · Memories of Murder 81 · Marshland 85 · Black Bread 83. Total: 508 annotated beats.
Real durationsMIGUEL 105 min · Pulp Fiction 148 min (NNDb) · Memories of Murder 131 min (MoMA, Criterion) · Marshland 105 min · Black Bread 108 min.
Temporal axisReal time in seconds, not normalized to 100 points. Resolution: 1 sample every 10 seconds. Curve construction: linear interpolation between beats (gaps not filled with zeros).
MetricsPearson (with resampling), Dynamic Time Warping with a proportional window, Wasserstein-1 over the distributions. Combined metric: geometric mean.
Dimension weightsPolarity 52.9% · Complexity 28.2% · Intensity 18.8%. Proportional to the variance explained reported in Lettieri 2019.

Re-annotating the same screenplays with a different LLM should yield matrices with the same qualitative topology, since the Lettieri dimensions are external to the annotation process and validated by neuroscience.

9. Downloadable dataset

All files are publicly available in the Open Book's data/ folder and can be downloaded directly from the project's portal. No registration or credentials required.

FileContents
miguel_unified.json119 MIGUEL beats annotated with Save the Cat + Lettieri (polarity, complexity, intensity)
pulp_fiction_reannotated.json140 beats with real NNDb timestamps · Tarantino, 1994
memories_of_murder_unified.json81 beats · Bong Joon-ho, 2003
marshland_unified.json85 beats · Alberto Rodríguez, 2014
black_bread_unified.json83 beats · Agustí Villaronga, 2010
matrix_final_pearson.csvPearson matrix 5×5
matrix_final_dtw.csvDTW matrix 5×5
matrix_final_wasserstein.csvWasserstein matrix 5×5
matrix_final_combined.csvCombined matrix (geometric mean) 5×5
three_metric_comparison.pngFull visualization of the four matrices
Visual comparison of the four similarity matrices
Visual comparison of the four metrics. The two most informative are Pearson (rules out formal replication) and Wasserstein (reveals the real tonal clusters). DTW confirms that no film is structurally atypical. The combined metric stabilizes the result.

Any advisor reviewing the project can audit the numerical foundation with these ten files.

10. Bibliography and sources

  1. Aliko, S., Huang, J., Gheorghiu, F., Meliss, S., & Skipper, J. I. (2020). A Naturalistic Neuroimaging Database for understanding the brain using ecological stimuli. Scientific Data 7(1):347.
    DOI: 10.1038/s41597-020-00680-2
    OpenNeuro dataset: 10.18112/openneuro.ds002837.v2.0.0
    Project portal: naturalistic-neuroimaging-database.org
  2. Naturalistic Neuroimaging Database — Movie Annotations. LAB Lab, University College London. Index of all word and face annotations (CC-BY).
    naturalistic-neuroimaging-database.org/annotations
  3. Lettieri, G., Handjaras, G., Ricciardi, E., Leo, A., Papale, P., Betta, M., Pietrini, P., & Cecchetti, L. (2019). Emotionotopy in the human right temporo-parietal cortex. Nature Communications 10:5568.
    DOI: 10.1038/s41467-019-13599-z
    Data and code (CC-BY 4.0): osf.io/tzpdf
  4. Pulp Fiction — Word Annotations. Official CSV with 16,155 words of the film and validated timestamps, NNDb.
    naturalistic-neuroimaging-database.org/pages/pulp_fiction_words.csv
  5. Lettieri, G., Handjaras, G., Ricciardi, E., Pietrini, P., & Cecchetti, L. (2021). Chronotopic encoding of emotional dimensions in the human brain assessed by fMRI. European Psychiatry 64(S1):S129.
    DOI: 10.1192/j.eurpsy.2021.361
  6. Lettieri, G., et al. Italian behavioral ratings from the Emotionotopy study. DataLad dataset of perceived emotion annotations on Forrest Gump used in [3].
    github.com/psychoinformatics-de/studyforrest-data-perceivedemotions
  7. Mittal, T., Mathur, P., Bera, A., & Manocha, D. (2021). Affect2MM: Affective Analysis of Multimedia Content Using Emotion Causality. CVPR 2021.
    arXiv: arxiv.org/abs/2103.06541
  8. Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology 39(6):1161–1178.
    DOI: 10.1037/h0077714