Beyond just a torus? Re-contextualising the AGN type paradigm using data-driven host galaxy quenching histories
The orientation of a dusty torus within a unified model explains the presence (Type-1) or absence (Type-2) of broad emission lines in Active Galactic Nuclei (AGN) spectra. This orientation-based unified model, however, fails to explain differences in host galaxy properties, such as star formation rates, of Type-1 and Type-2 AGNs. Star-formation histories (SFHs) can shed light on the connection between AGN type and host galaxy properties, but are challenging to extract from Type-1 AGN that outshine their host galaxies. We use mean-field independent component analysis (MFICA) spectrum decomposition to extract the stellar populations of $\sim$3000 optically selected Type-1 and Type-2 AGN host galaxies from the seventh data release (DR7) of the Sloan Digital Sky Survey (SDSS), matched in redshift ($0.16 \leq z \leq 0.4$) and AGN luminosity ($6.5<\log_{10} [L_{[OIII]}/L_\odot] <9.5$). Using MFICA, we find that only $\sim$10 per cent of AGN host galaxies have recently ($\lesssim1$\,Gyr) and rapidly quenched star formation following a starburst, with Type-2 AGN host galaxies having done so at most 100\,Myrs more recently than Type-1 AGN host galaxies. For most ($\sim$90 per cent) AGN hosts with normal recent star formation histories, Type-2 AGNs are associated with slightly younger stellar populations than Type-1 AGNs. The younger stellar populations of Type-2 AGN hosts can be explained by a link between dusty torus covering factors and stellar populations, by host-galaxy obscuration of AGN, or by an evolution in AGN type without major bursts or rapid quenching of star formation.