2019ApJ...882..141M

Research over the past decade has shown diminishing evidence for major galaxy mergers being a dominant mechanism for the growth of supermassive black holes (BHs) in galaxies and the triggering of optically or X-ray-selected active galactic nuclei (AGNs). For the first time we test whether such a connection exists at least in the most "plausible" part of parameter space: the highest specific accretion rate broad-line AGNs at the peak epoch of BH activity around z = 2. To that end we analyze two samples-21 AGNs with L/L_edd > 0.7 and 92 stellar mass- and redshift-matched inactive galaxies-observed with HST/WFC3. We remove the AGN point sources from their host galaxies and avoid bias in visual classification by adding and then subtracting mock point sources to and from the comparison galaxies, producing matched residual structures for both sets. The resulting samples are joined and visually ranked according to distortion strength by 10 experts. The ensuing individual rankings are combined into a consensus sequence and from this we derive the merger fractions for both samples. With the merger fractions f_m,agn = 0.24 ± 0.09 for the AGN host galaxy sample and f_m,ina = 0.19 ± 0.04 for the inactive galaxies, we find no significant difference between the samples. This finding is consistent with previous studies for different AGN populations, and we conclude that even BH growth at the highest specific accretion rates and at the peak of cosmic AGN activity is not predominantly caused by major mergers.