Significance of suppressive kernels is tested by bootstrap randomization conserving spike timing statistics.

Evidence

To determine statistical significance of each weighted vector, we re-computed h 2 from 20 different spike:train randomizations, conserving the first-order inter-spike-interval distribution. Based on this bootstrap distribution, weights were expressed as z-scores, and any vector with |z| > 3 and rank ≤ 20 was considered significant. — Dijk, Pim van, Physiology, Psychoacoustics and…, 3848

Categorization

• XREF: Connects to the reader’s interest in systems-identification methods for sensory neural coding.
• corroborated: 2 independent passages support this claim — Dijk, Pim van, Physiology, Psychoacoustics and…, 3848