Loudness functions with air and bone conduction stimulation in normal-hearing subjects using a categorical loudness scaling procedure

• The loudness function was steeper when stimulation was by BC than by AC.
• The difference between the loudness function slopes was greatest at low frequencies.
• The average ratios between AC and BC loudness function slopes were between 0.88 and 0.92.
• The difference between the AC and BC loudness functions were found at both low and high intensities.

In a previous study (Stenfelt and Håkansson, 2002) a loudness balance test between bone conducted (BC) sound and air conducted (AC) sound was performed at frequencies between 0.25 and 4 kHz and at levels corresponding to 30–80 dB HL. The main outcome of that study was that for maintaining equal loudness, the level increase of sound with BC stimulation was less than that of AC stimulation with a ratio between 0.8 and 0.93 dB/dB. However, because it was shown that AC and BC tone cancellation was independent of the stimulation level, the loudness level difference did not originate in differences in basilar membrane stimulation. Therefore, it was speculated that the result could be due to the loudness estimation procedure. To investigate this further, another loudness estimation method (adaptive categorical loudness scaling) was here employed in 20 normal-hearing subjects.The loudness of a low-frequency and a high-frequency noise burst was estimated using the adaptive categorical loudness scaling technique when the stimulation was bilaterally by AC or BC. The sounds where rated on an 11-point scale between inaudible and too loud. The total dynamic range for these sounds was over 80 dB when presented by AC (between inaudible and too loud) and the loudness functions were similar for the low and the high-frequency stimulation. When the stimulation was by BC the loudness functions were steeper and the ratios between the slopes of the AC and BC loudness functions were 0.88 for the low-frequency sound and 0.92 for the high-frequency sound. These results were almost equal to the previous published results using the equal loudness estimation procedure, and it was unlikely that the outcome stems from the loudness estimation procedure itself. One possible mechanism for the result was loudness integration of multi-sensory input. However, no conclusive evidence for such a mechanism could be given by the present study.This article is part of a Special Issue entitled “MEMRO 2012”.