LEA👧 - Linearized Encoding Analysis

This MATLAB🏔️ package is to perform linearized encoding analysis on time series data (see Kim, 2022, Frontiers in Neuroscience).

This version (v0.0.1-alpha-20250329) is for a new preprint: Kim. (2025). Reverse Double-Dipping: When Data Dips You, Twice—Stimulus-Driven Information Leakage in Naturalistic Neuroimaging. [bioRxiv]

Please see /README.mlx on MATLAB to learn more about it.

If you don't have access to a MATLAB license, please know that you can still use MATLAB Online Basic for an educational purpose (free 30 hours/month) at matlab.mathworks.com.

For the methodological background of the analysis, please see the tutorial (mostly English).

(CC-BY-4.0) 2025-03-29, seung-goo.kim@ae.mpg.de

Old notes:

• [2024-09-02] This version (v0.0.0-alpha-20240902) is for a tutorial at KSMPC (Korean Society for Music Perception and Cognition) Summer School 2024 [only Korean]. Currently, only nested leave-one-stimulus-out cross-validation is implemented. Each response unit is optimized separately via grid search, but a single lambda regularizes all stimulus features.