Continuous Wavelet Transform (Scalograms)

Description

Localises frequency content in time by sliding scaled wavelets across a signal, revealing when bursts, transitions and slow drifts occur in behaviour.

Where Fourier reveals global rhythm, wavelets reveal local story. The continuous wavelet transform correlates the signal with a wavelet at many scales and positions, producing a time-frequency scalogram. In the Behavioral Spectral Framework, short-scale bursts are read as spikes of cognitive load, surprise or momentary confusion, long-scale drifts as slow learning, sustained focus, fatigue or attention decay, and coexisting scales as multitasking or conflicting goals. Recommended adaptations include adaptive scales, wavelet energy envelopes to estimate task load and recovery time after failures, and scalograms aligned with qualitative annotations. The synthesis method extracts dominant bands with FFT, then localises them in time with CWT or STFT and aligns them with coded moments. In the fintech case, Frustration-coded moments consistently coincided with transient high-frequency spikes in the wavelet energy map.

Attributes

Loss type
Unmigrated
Stage
Post-launch
Effort
Days

Steps

  1. Choose a mother wavelet and a range of scales matched to the behaviours of interest.
  2. Compute the CWT of the behavioural signal and plot the scalogram.
  3. Compute wavelet energy envelopes to estimate load and recovery time after failure events.
  4. Detect bursts and drifts and export their timestamps.
  5. Align bursts with qualitative annotations (video, transcript, sentiment) in a Spectral Annotation Matrix.
  6. Report scalograms with annotation overlays.

Sources

  • Vivek Gopinath (2025), From Stars to Signals: Translating Matteo Paz's Fourier-Wavelet Model into Behavioral and UX Research Methodology
  • Matteo Paz (2024), A Sub-Millisecond Fourier and Wavelet Based Model to Extract Variable Candidates from the NEOWISE Single-Exposure Database

Also known as

  • CWT
  • Wavelet analysis
  • Engagement scalograms

Categories

  • Information Theory, Physics & Mathematics