Some stats:
-V5 --vbr-new with and without --athaa-sensitivity 1
when difference is from: -1.0 to 0.4 point (70 samples) +2,84 kbps in average with --athaa..1
when difference is from: 0.5 to 0.9 point (29 samples) +4,17 kbps in average with --athaa..1
when difference is from: 1.0 to 1.4 points (23 samples) +7,70 kbps in average with --athaa..1
when difference is from: 1.5 to 1.9 points (13 samples) +8,69 kbps in average with --athaa..1
when difference is from: 2.0 to 2.4 points (09 samples) +14,44 kbps in average with --athaa..1
when difference is from: 2.5 points and more (05 samples) +19,00 kbps in average with --athaa..1
CONCLUSION: the perceptual quality improvment of --ahtaa-sensitivity 1 has a direct relation with the amount of bit added by this command line. When --ahtaa-sensitivity 1 has no impact on bitrate it has no impact on quality; when --ahtaa-sensitivity 1 "bloats" the bitrate it also "bloats" the quality.
10% of the tested samples reveal a tendency to "bitrate bloat" (+15 kbps or more). For these 14 samples, the average notation is going from 2.3 points (without --ahtaa-sensitivity 1) to 4.3 points (with --ahtaa-sensitivity 1).
____
Appendix:
RATIO
[average notation diff in points] / [average bitrate diff in kbps]
-0.01 / 2.84 =-0.0020
0.53 / 4.17 = 0.1264
1.03 / 4.70 = 0.1333
1.56 / 8.69 = 0.1796
2.00 / 14.44 = 0.1384
2.70 / 19.00 = 0.142
1