Fine-grained anechoic measurement statistics can near-perfectly predict listener preference rankings

Evidence

Earlier in the book it was shown that anechoic frequency-response data can be used to predict steady-state room curves above the transition frequency. We know that measuring a room curve with a microphone is not the same as perceptions by the human binaural hearing system, but the information used by human listeners is hidden within the room curve. Separating the individual direct and early reflected sounds in a steady-state room curve after the fact is not possible. However, the reverse is possible. The anechoic loudspeaker measurements and the post processing for spinorama presentation was an attempt to evaluate the direct and reflected sounds arriving at a listener’s ears in a typical domestic room. — Toole, Floyd E, Sound reproduction- the acoustics and…, 1224

Employing the much more comprehensive, higher-resolution data generated routinely in the Harman International tests, Olive identified a model that described the subjective preference ratings in a near-perfect manner, as shown in Figure 11.1(b). — Toole, Floyd E, Sound reproduction- the acoustics and…, 1205

Categorization

• XREF: Directly relevant to how spinorama presentation and room curve prediction are built and applied.
• QUESTION: Which specific spinorama statistics carried the predictive weight in Olive’s model? Worth tracing to the Olive 2004a/b papers.
• corroborated: 2 independent passages support this claim — Toole, Floyd E, Sound reproduction- the acoustics and…, 1224; Toole, Floyd E, Sound reproduction- the acoustics and…, 1205