A general recommendation therefore might be to simply repeat the analyses with other reference lags if researchers suspect that there might be significant structure at both l ̄ and − l ̄ , or to fix the reference lag to be, e.g., l * = ± ( l m a x + 1 ) , with l m a x the maximum lag tested for, for all lags considered.
← all excerpts
Cell assemblies at multiple time scales with arbitrary lag constellations.
1
—
—