Marcus Anderson

Research notes, papers, software, and archived music work.

New invite code j7Uh4vGg4UrF Try my probabilistic music-generation studio at thesoundinator.com

Research

Experience-Dependent Differences in the Neural Representation of Food: An EEG Study of Long-Term Dietary Habits
Master's thesis (MPhil). An EEG and multivariate-pattern study finding that long-term vegetarians and vegans show enhanced neural discrimination of meat under deliberate viewing — a broad expansion of the food representational space that runs counter to exposure-based models of perceptual learning. Empirical Neuroscience EEG Perception
Empathy as a Signal of Type: A Verification-Horizon Model of Trust and Strategic Mimicry
Develops a signalling-game model of trust under limited verification, showing when costly prosocial action can distinguish integrated from strategic types and how that evidence weakens as the verification gap grows. Theory Game theory AI alignment Moral agency
Musical Repetition as Vocal Self-Simulation
A mechanistic theory arguing that precise repetition of affective vocalisation is the core process that turns sound into musical experience. Theory Neuroscience Music
Inverse-Square Distance Pruning: A Spatial Initialisation Method for Clustered Small-World Graphs
Defines the foundational random-spatial graph construction behind the notebook series and later pruning studies, with density-matched evidence for its clustering, small-world, and visible hierarchical properties. Theory Empirical Machine learning Networks
Inverse-Square Distance Priors Improve Network Pruning at Extreme Sparsity
Tests a biologically inspired 1/d^2 wiring prior for sparse neural networks, showing improved ResNet-18 performance at 98-99% sparsity. Empirical Machine learning Neuroscience Sparsity
Bandwidth and Capacity Systematically Shape the Topology of Inverse-Square Distance Pruning
Characterizes how inverse-square pruning changes network topology across input resolution and model size, highlighting bandwidth-dependent clustering and stronger 2D embeddings. Empirical Machine learning Networks

Software

Soundinator — a probabilistic, physically-modelled music studio and research instrument

A full browser-based music studio, designed and built end-to-end. It generates music from weighted probability distributions across several musical timescales, renders it through a physical-modelling synthesis engine, and arranges it on a multitrack producer timeline — while doubling as a reproducible instrument for researching why music sounds good.

Soundinator is now live at thesoundinator.com — in early testing, so access is invite-only. The first 30 people can create an account with the invite code j7Uh4vGg4UrF. Found a problem? There's a “Report a problem” button in the account menu — every report helps.

Soundinator's generative view: recent history to the left of the playhead, and a probability field of possible futures to the right, across melody, rhythm, surprise and motif-memory lanes.

Generative core — melody, rhythm, register, articulation and surprise are all driven by editable probability distributions, with motif memory that repeats, drifts and mutates into a growing repertoire.
Physical-modelling synthesis — every note is built from an excitor → resonator → body → effects → space chain: additive and formant voices, inharmonicity, material damping, vibrato and per-note envelopes rather than samples.
Per-layer effects rack — eleven skeuomorphic, plugin-style effects across EQ, drive, modulation, delay and character, each with its own DSP, presets and expand-to-overlay face.
Binaural spatialisation — position each source around the listener with a head / room / air model fitted from measured KEMAR HRTFs, with per-track motion through time.
Microtonal Scale Lab — N-EDO systems, world tunings, per-degree cents, sub-scales and roots, with Scala import / export.
Producer / DAW — arrange instruments on a multitrack timeline with seed-reproducible takes, a bake-to-piano-roll note editor, a channel-strip mixer, and offline WAV mixdown.
Community library — invite-gated sharing of instruments, sequences and behaviours with ratings and profiles, plus an auditioner that layers shared patches into a live jam.
Research-ready — every play, rating and edit is reproducible from a seed and can be logged (opt-in only) with expectation / surprise metrics, so appeal can be modelled directly against the mechanisms that generated the music.

The Macro Probability panel for melody: walk and arpeggio generation modes, interval and accuracy sliders, and a per-scale-degree probability histogram mixing generation, accuracy and surprise. — The distributions behind the field — every choice the generator makes is an editable probability: here the melody walker's interval shape, accuracy and per-scale-degree histogram, with generation, accuracy and surprise stacked in one view.

Soundinator's sub-note designer: the excitor, resonator, body, effects and space stages above a binaural stage view, with an Ears panel showing what reaches each ear. — The sub-note designer — every note is built from an excitor → resonator → body → effects → space chain and placed in one shared binaural room; the Ears panel shows what actually arrives at each ear, when, and how it's coloured.

Soundinator's resonator partial field: individual partials plotted by frequency and level, with decay time, wobble and cross-partial coupling on a pinned partial's inspector. — Inside the resonator — a field of individual partials, each with its own level, decay time, wobble and coupling to other partials; focus filters isolate the odd, even, coupled or long-ringing ones.

The Sculpt EQ, a skeuomorphic program equaliser, open in Soundinator's per-layer effects rack ahead of a tube drive and wavefolder. — The effects rack — plugin-style effects with their own DSP and presets; here a program equaliser sits ahead of a tube drive and a wavefolder in one layer's chain.

Soundinator's producer with the piano-roll note editor open beneath a multitrack timeline of seed-reproducible regions. — The producer — arrange seed-reproducible takes on a multitrack timeline (with a global scale and a global space that evolve over time), and bake any region open into a piano-roll note editor.

Soundinator's Global Space lane in the producer: per-track distance curves evolving across the arrangement above the multitrack timeline. — Global Space in the producer — the whole arrangement shares one room, and each track's position through it is drawn over time, so voices drift nearer and farther as the piece plays.

Soundinator's Scale Lab: a microtonal tuning wheel with just-ratio and EDO grids, tuning presets and Scala import and export. — The Scale Lab — design microtonal scales on a tuning wheel: N-EDO grids, just-ratio intervals, per-degree roles and cents, with Scala import / export.

Soundinator's community library: shared instrument cards with ratings, and the auditioner bar layering two shared patches into a live jam. — The community library — share instruments, sequences and behaviours, rate what others post, and stack shared patches in the auditioner to hear them jam together live.

Vanilla JavaScript + Web Audio in the browser, backed by a Python (NumPy / SciPy) engine for reproducible offline rendering. · Source and full write-up on GitHub →

Biologically Inspired Graph Networks — a self-teaching notebook course on network topology, from graph theory to developmental pruning

A seven-notebook course that starts from graph-theory basics and builds towards the pruning-based network construction behind my sparse-network papers. It's written as a way into thinking about neural networks the way biology does: not wired from scratch, but grown dense and then pruned back. Every notebook motivates its maths with the brain — neurons as nodes, synapses as weighted edges — and ends in runnable code you can re-generate, change, and measure.

Four network-generation methods drawn side by side — Erdős-Rényi, Watts-Strogatz, Barabási-Albert, and a spatial inverse-square model — each labelled with its clustering coefficient.

The through-line is one question: what makes a network look like a brain? The course answers it by measuring candidate networks against biological criteria, then showing that pruning a dense weighted graph reproduces those criteria more faithfully than the standard random models.

Starts from the vocabulary — directed, weighted and simple graphs; degree, paths, cycles and connectivity — each introduced through its neural counterpart.
Measures what matters — clustering coefficient, average path length and the small-world coefficient: the metrics used to decide whether a generated network is biologically plausible.
Surveys the classic generators — Erdős-Rényi, Watts-Strogatz, Barabási-Albert and spatial / inverse-square models — and shows where each falls short of real cortical structure.
Builds networks by pruning — deterministic, probabilistic and distance-weighted pruning, then multiple developmental cycles of weight decay, Pareto-principle reinforcement and elimination.
Ends with functional analysis — spectral clustering, Louvain community detection and weighted modularity to read out the modular structure that pruning leaves behind.

Three small labelled graphs — undirected, directed, and weighted — side by side. — The vocabulary the course starts from — undirected, directed and weighted graphs — introduced through their neural counterparts: synapses that point one way, and connections that vary in strength.

The same four-node graph shown four times, each highlighting one node and its clustering coefficient, from 1.00 down to 0.00. — Clustering coefficient, one node at a time — the fraction of a node's neighbours that are also connected to each other. It's the first quantitative handle on whether a network has brain-like local structure.

Left: a dense hairball of 1,770 edges. Right: the same nodes reduced to a sparse, clearly structured 177-edge network after keeping only the strongest connections. — The pruning philosophy in one picture — a dense 1,770-edge graph reduced to 177 edges by keeping only the strongest connections. The brain builds sparse networks the same way: start dense, then prune. This is the process the papers build on.

Python notebooks (NetworkX / NumPy / Matplotlib / scikit-learn), runnable in Jupyter or Colab. · Notebooks and full course on GitHub →

Marcus Anderson

Research

Software

Archive