MAYA RESEARCH SERIES · POST-SERIES CONSCIOUSNESS STUDY

From Representation to Experience

An mPCI-Based Empirical Test of Internal Affective State in a Neuromorphic SNN

Venkatesh Swaminathan · ORCID 0000-0002-3315-7907 · Nexus Learning Labs, Bengaluru

−0.0489
DELTA mPCI (aggregate)
Phase 3 vs Phase 1 across 3 seeds
2.05×
THRESHOLD RATIO
Delta exceeds 2× pooled SD criterion
3 seeds
REPLICATION
Seeds 42, 123, 7 — consistent shift
Partial
CONFOUND STATUS
Training depth partially confounds result
⚠ METHODOLOGY NOTE — READ BEFORE CITING

Three methodological controls were run after the main experiment. Control 1 (training depth) found that a model trained for 5 tasks without affective dimensions shows mPCI similar to Phase 3, partially confounding the affective state interpretation. Control 3 (shuffle) found that shuffled spike trains show larger differences between phases than structured ones. These findings are reported honestly in the paper. The mPCI shift is real and reproducible across 3 seeds. The interpretation requires nuance: training depth and affective state both contribute. Independent replication is needed to isolate the affective contribution cleanly.

MAYA RESEARCH SERIES · P1–P9 + POST-SERIES