The findings suggest that, after controlling for the effect of hearing loss, there are a number of test measures in which performance of elderly and young listeners differs. Regarding lateral asymmetry, a disproportionate decline in auditory processing in the left ear of elderly individuals has been demonstrated. The major clinical conclusion is that, when testing an elderly group's performance regarding any given auditory function, the influence of pure-tone threshold elevations, no matter how mild, cannot be disregarded.

A group of 29 elderly subjects between 60.0 and 83.7 years of age at the beginning of the study, and whose hearing loss was not greater than moderate, was tested twice, an average of 5.27 years apart. The tests measured pure-tone thresholds, word recognition in quiet, and a set of tests of understanding speech with various types of distortion (low-pass filtering, time compression) or interference (single speaker, babble noise, reverberation). Performance declined consistently and significantly between the two testing phases. In addition, the variability of speech understanding measures increased significantly between testing phases, though the variability of audiometric measurements did not. A right-ear superiority was observed but this lateral asymmetry did not increase between testing phases. Using a group of young subjects with normal hearing to establish a baseline, we found that the decline of speech understanding measures accelerated significantly relative to the decline in audiometric measures in the seventh to ninth decades of life. On the assumption that speech understanding depends linearly on age and audiometric variables, there is evidence that this linear relationship changes with age, suggesting that not only the accuracy but also the nature of speech understanding evolves as we mature.

Factor analysis of test measures obtained from a group of elderly individuals with normal hearing or mild-to-moderate hearing loss led to two main findings. First, it portrayed hearing loss as a component of different factors rather than as a factor on its own. Second, the independence of measures of speech understanding in babble or reverberation from other measures suggests that such tests should become an integral part of audiological test batteries designed to assess auditory functions in aging.

The findings suggest that, to determine the relationship between audiological/auditory test results of an elderly population, it is important to remove the effects of hearing loss through appropriate statistical methods.

Coren (1989) noticed that the various air-conduction pure-tone thresholds obtained from one ear ofan individual in quiet tend to display a high correlation. In addition, Coren and Hakstian (1990) In two recent articles, Coren (1989) and Coren and Hakstian (1990) noted that pure-tone thresholds in one ear of a given individual (measured at 6 audiometric frequencies covering the five-octave range between 250 and 8,000 Hz), as well as pure-tone threshold estimates in the two ears of the same individual, exhibit a high degree of correlation. On the basis of this observation, these authors propose to replace, in group data, (1) the pure-tone thresholds obtained at the 6 frequencies for each ear of an individual with a single number representing their average, and (2) the pure-tone thresholds obtained in the two ears of each individual with a single number representing either the average threshold of the two ears or the average threshold in the better ear. In other words, these authors would like to substitute a single number for either one or both of the left and right audiograms every time the object of study is a group of individuals, rather than a single individual. Although it is undeniable that pure-tone audiometric thresholds obtained at different frequencies in one or both ears of an individual often show high positive zeroorder correlations, 1 there are several valid reasons for not collapsing the (usually 12) threshold values into one number. The following paragraphs represent an attempt to bringThe authors gratefully acknowledge comments by James Jerger and two anonymous reviewers on an earlier version of the manuscript. On Discounting Ears Headfirst: A Case Against "Counting Heads, Not Ears" Audiological research, as a matter of principle, is aimed at issues related to the hearing status, rather than that of the ears, of individuals. The measure that serves as the starting point (and, quite often, the only point) to determine an individual's hearing status is the audiogram, that is, estimation of pure-tone audibility thresholds in quiet at various frequencies (usually six). The overall hearing level and the shape of the right and left audiograms, as well as the difference between the two audiograms, provide information on the basis of which the audiologist can establish the hearing status of the individual. The audiograms and the assessment of hearing for each ear of the same individual can, and often do, differ. It is the audiological significance of differences between left and right thresholds at each frequency that constitutes the most powerful argument against Coren and Hakstian's (1990) suggestion-namely, to consider, in group data, only one threshold estimate per frequency (i.e., left-right average or one of the ears) instead of two. There are, however, additional reasons for arguing in favor of treating the two audiograms as separate data, even if thresholds measured at all frequencies in the two ears of each individual in a group were found to be correlated, and even if such a correlation had be...