Bone conduction thresholds poorer than air

In an article by Barry (1994), he addresses the issue of whether bone conduction thresholds can be poorer than air conduction, and if so, is it alright to record BC thresholds at the AC threshold. 

Audiologists are taught that BC test results cannot be lower than AC thresholds.  AC measures the sensitivity of the entire hearing mechanism while BC measures only a part of that mechanism. The logic here stems from the fact that a loss in part of the system cannot exceed the loss in the entire system, hence BC can not be poorer than AC.  Although what we have been taught does make sense, Audiologists need to know that it may not always hold true. 

According to Barry (1994), the reasoning above fails to consider both the ossicular and meatal contributions to BC sensitivity, and it fails to take into account the significant differences which exist within and across individuals in the efficiency of signals transmission through the AC and BC pathways.  Barry (1994) then goes to argue that AC and BC reference thresholds as indicated in standards, represent average values with a normal sampling distribution (otherwise known as a bell curve).  Most people therefore will have their results fall under average values where AC and BC will be clinically agreeable.  It is possible that both AC and BC results can fall on either extreme of the distribution, thus producing a better or poorer agreement with average values, but nonetheless still clinically agreeable.  

Average values:

Purely as a result of statistics, it is also possible for the AC thresholds to be on one extreme tail of the normal distribution while BC thresholds to be on the opposite.  When this happens a apparently false air bone gap can be found.  The reverse can also occur where BC thresholds fall on the less sensitive tail of the distribution while AC thresholds fall on the more sensitive side.  When this happens, BC thresholds can appear to be poorer than AC.  As stated in ANSI S3.42 – 1992,” exact equivalence of air- and bone-conduction thresholds cannot be expected.”

Having said this however, it is highly unlikely that major differences in AC and BC thresholds can be attributable to statistical variation by itself.  A 5 or 10 dB difference where BC thresholds appear poor than AC may not be startling, but a 15 dB or more difference is subject to greater examination. 

Note:  This site is best viewed using IE 4.X or higher under 800x600 resolution