Robyn M. Cox

An international comparison of long‐term average speech spectra

The long‐term average speech spectrum (LTASS) and some dynamic characteristics of speech were determined for 12 languages: English (several dialects), Swedish, Danish, German, French (Canadian), Japanese, Cantonese, Mandarin, Russian, Welsh, Singhalese, and Vietnamese. The LTASS only was also measured for Arabic. Speech samples (18) were recorded, using standardized equipment and procedures, in 15 localities for (usually) ten male and ten female talkers. All analyses were conducted at the National Acoustic Laboratories, Sydney. The LTASS was similar for all languages although there were many statistically significant differences. Such differences were small and not always consistent for male and female samples of the same language. For one‐third octave bands of speech, the maximum short‐term rms level was 10 dB above the maximum long‐term rms level, consistent across languages and frequency. A ‘‘universal’’ LTASS is suggested as being applicable, across languages, for many purposes including use in hearing aid prescription procedures and in the Articulation Index.

Development of the Connected Speech Test (cst)

This paper describes the first phase in the development of the Connected Speech Test (CST). This test of intelligibility of everyday speech has been developed primarily for use as a criterion measure in investigations of hearing aid benefit. The test consists of 48 passages of conversationally produced connected speech. Each passage contains 25 key words for scoring. All passages are of equal intelligibility for the average normal hearer. Key words vary in intelligibility within a passage but span the same intelligibility range in all passages. Several passages are administered, and the results averaged, to yield a single intelligibility score. For pairs of scores, each based on mean performance across 4 randomly-chosen passages, the 95% critical difference is estimated to be about 14 rationalized arcsine units (rau). The performance-intensity function for the CST has a slope of 12 rau/dB signal-to-babble ratio. Investigations of the test are continuing with hearing-impaired listeners.

Use of the connected speech test (CST) with hearing impaired listeners

Two studies were performed in which hearing-impaired subjects responded to the Connected Speech Test (CST). In experiment 1, 40 subjects, divided into four groups according to extent and configuration of hearing loss, responded to the CST version 1 (CSTv1). This version of the test consisted of 57 passages of connected speech: 48 test passages and 9 practice passages. It was developed on the basis of data for normal-hearing listeners. Performance of hearing-impaired listeners for the CSTv1 revealed that, although the passages were equal in average intelligibility for normal hearers, they were not equally intelligible for hearing-impaired persons. Based on results of data analyses, the 57 passages were reconstituted into 28 pairs of passages: 24 test pairs and 4 practice pairs. The pairs were equal in average intelligibility for both normal and hearing-impaired listeners. This form of the test was named the CST version 2 (CSTv2). In experiment 2, an additional 23 hearing-impaired subjects responded to the CSTv2. Critical differences and the slope of the signal to babble ratio (SBR) function were determined for the CSTv2 for hearing-impaired listeners. When two CSTv2 pairs were used per score, the 95% critical difference for hearing-impaired subjects was about 15.5 rationalized arcsine units (rau). The mean SBR function slope for hearing-impaired listeners was 8.5 rau/dB. Comparing the critical difference with the SBR function slope, it may be seen that, for hearing-impaired listeners, differences in intelligibility equivalent to a 2 dB change in SBR can be detected with CST scores based on mean performance across two passage pairs.

The Contour Test of Loudness Perception

Objective: This article presents the underlying rationale, normative data, and reliability data for a test of loudness perception (the Contour Test) that was devised for use in clinical hearing aid fitting. The Contour Test yields data describing the sound level required for each of seven categories of loudness ranging from very soft to uncomfortably loud. Design: Two experiments are described. Experiment 1 yielded norms for the test. The subjects were 23 male and 22 female normal‐hearing listeners. Test stimuli included warble tones at six frequencies and broad band speech. Experiment 2 assessed the reliability of the test results. Ten hearing‐impaired listeners responded to the test at two frequencies on two occasions separated by several days. Both experiments also evaluated the effect of using different stimulus increment sizes on the measured levels of loudness categories. Results: Based on the data from experiment 1, norms for each category of each stimulus are reported in terms of mean level and typical between‐subject variation in responses. Data are provided in HA‐1 2 cm3 coupler levels as well as in hearing levels (dB HL). The shape of the loudness growth function for warble tones was somewhat different from that for speech. When data were expressed in HL, there were no differences in mean loudness category levels across warble tone test frequencies. Thus, test frequencies were combined and equations were generated to describe the upper and lower limits of typical normal performance for warble tone stimuli. These equations can be used to construct a template for clinical comparison of normative values to patient loudness growth curves. Experiment 2 provided information about the test‐retest variability of data yielded by the Contour Test. Reliability appears to be similar to that of the few other category scaling tests described in the literature. Most test‐retest differences were 6 dB or less. Although a moderate variation in test increment size did not significantly affect the loudness category levels for young normal‐hearing listeners, levels corresponding to loudness categories were significantly higher when larger increments were used with elderly hearing‐impaired listeners. Conclusions: Evidence from this and other research indicates that standardized measurement of loudness perception is an achievable goal for clinical practice. The Contour Test appears to offer a viable approach to clinical measurement of loudness perception: It has good patient acceptance and combines fairly rapid administration with acceptable reliability. Details of test procedures and scoring sheets for manual administration can be down‐loaded from the Internet at www.ausp.memphis.edu/harl. However, it is important to keep in mind that the application of loudness perception data for narrowband stimuli (such as warble tones) to hearing aid prescription is complicated by the need to account for the effects of loudness summation across bandwidth. There is a need for additional research to establish an empirical link between clinically measured loudness perception and optimal amplification characteristics.

Who wants a hearing aid? Personality profiles of hearing aid seekers

Objective: Evidence indicates that elderly hearing-impaired people who use amplification live happier, healthier, and longer lives than those who do not. Nevertheless, only a small fraction (approximately 23%) of hearing-impaired adults actually seek and use hearing aids. This study explored the personalities of hearing aid seekers in an attempt to determine whether those who seek hearing aids are systematically different from the general population. Design: In this cross-sectional survey, self-report data were obtained from 230 older adults with bilateral, symmetrical, sensorineural, mild to moderately severe hearing impairment. Subjects were representative of patients served either in a publicly funded hospital-based system (VA) or in a free-standing private practice system (PP). All subjects were seeking new hearing aids. Subjects completed a comprehensive personality questionnaire (NEO-Five-Factor Inventory) as well as questionnaires determining locus of control and preferred coping strategies. Results: Individuals who seek amplification are not simply a random sample of the general population and presumably not a random sample of the hearing-impaired population. Compared with the typical adult, hearing aid seekers tended to be more pragmatic and routine-oriented and probably less imaginative in coming up with novel approaches to dealing with a complex problem such as hearing impairment. These individuals also were found to feel relatively more personally powerful in dealing with life’s challenges. Further, hearing aid seekers reported using social support coping strategies less frequently than their non–hearing-impaired peers. In addition, there were significant differences in personality patterns between hearing aid seekers in the PP and VA systems. Differences noted in the personality traits of Openness and Neuroticism might be a useful guide to selecting treatment approaches and expectations for patients in each setting. Additional differences in Agreeableness imply that patients in the private practice system were more trusting than those in the general population, whereas this was not seen for patients in the public health VA system. One interpretation of this finding is that hearing-impaired individuals who are more suspicious and cynical are reluctant to try amplification in the PP system. This observation underscores the need to improve the public image of hearing health care to increase the uptake of hearing aids in general. Conclusions: Although individual hearing aid seekers display personality characteristics within the range of normal, this study suggested that they are not simply a random sample of the general population. Possible explanations are offered for significant personality differences, and potential clinical relevance is noted for some effects. The data also point to a need to improve the public image of hearing health care in the PP system.

Maturation of hearing aid benefit: objective and subjective measurements.

The goals of this investigation were to determine whether hearing aid benefit improved significantly over the first 10 weeks of hearing aid use and whether time-related changes in benefit (if any) were affected by the type of benefit measurement (i.e., objective or subjective). A total of 17 hearing-impaired subjects participated, with different subjects completing different phases of the study. Benefit was measured soon after the hearing aid fitting and again after 10 weeks of adjustment to hearing aid use. Objective benefit data were determined using the Connected Speech Test. No significant changes in objective benefit were noted in noisy or reverberant listening environments when visual cues were available. However, in a low-noise setting and in a noisy setting without visual cues, improvements in objective benefit were seen over time. Subjective benefit data were derived from responses to the Profile of Hearing Aid Benefit. These data indicated significant benefit improvement over time in all five types of daily life situations assessed, although the improvement was small in reverberant and noisy environments. Significant, but modest, correlations were found between objective and subjective data for low-noise and reverberant listening environments. Comparison of experienced and novice hearing aid wearers suggested that although experienced wearers obtain more benefit than novice wearers, they evidence similar time-related changes in benefit during the first 10 weeks of new hearing aid use.

A review of past research on changes in hearing aid benefit over time.

&NA; Hearing aid benefit refers to a relative change in performance on a particular measure between aided and unaided listening conditions. A number of studies in recent years have investigated the hypothesis that hearing aid benefit increases over time after the initial fitting of the aid. Both objective (speech recognition) and subjective (questionnaire) measures have been used to measure hearing aid benefit. Some studies have reported a positive increase over time in group mean benefit, and some have reported no change in benefit, whereas none have reported a group mean negative change in benefit. However, individual subjects in these studies can show changes in benefit in either a positive or negative direction. The variability across subjects in each study has been large in comparison with the observed amount of benefit increase. In this review of the literature, it is argued that the studies present essentially similar results and the range of values across subjects in the various studies shows considerable overlap. Although there does appear to be a tendency for hearing aid benefit to increase over time, there are other, much stronger, factors influencing changes in hearing aid benefit that make it impossible at present to predict which patients will show a reliable increase (or decrease) in hearing aid benefit over time. (Ear & Hearing 1996;17;14S‐28S)

Validation of the SADL Questionnaire

OBJECTIVE To cross-validate the psychometric characteristics of the Satisfaction with Amplification in Daily Life (SADL) questionnaire (Cox & Alexander, 1999), and to explore the SADLs construct validity. DESIGN Thirteen private practice Audiology clinics each distributed SADL questionnaires, by mail, to 20 adults who had recently obtained hearing aids. The completed questionnaires were returned to a central site and subject anonymity was assured. There were 196 usable responses. RESULTS Psychometric characteristics of the items were found to be very similar to those reported previously. Thus, the internal validity of the instrument was strongly supported. The assumption that the SADL quantifies satisfaction by assessing its components was evaluated by examining the relationship between SADL scores and scores on a traditional single-item satisfaction measure. A logical and statistically significant relationship was seen between the two measures, thereby supporting the construct validity of both types of data. For private-pay clients, satisfaction scores were very similar to the interim norms published by Cox and Alexander (1999). However, clients whose hearing aids were partly or fully purchased by insurance or benefits programs tended to be more satisfied than interim norms for third-party pay clients derived 5 yr ago. For most types of clients, there was a tendency toward more satisfaction in the Negative Features subscale than observed in our previous research. CONCLUSIONS Both construct and internal validity of the SADL questionnaire were supported by this research. The previously published interim norms appear to be mostly appropriate for private-pay clients, but might require adjustment in the Negative Features subscale. Further research is needed to explore the relationship between satisfaction and device purchase issues (third-party versus private pay).

Personality, Hearing Problems, and Amplification Characteristics: Contributions to Self-report Hearing Aid Outcomes

Objective: When we evaluate the success of a hearing aid fitting, or the effectiveness of new amplification technology, self-report data occupy a position of critical importance. Unless patients report that our efforts are helpful, it is difficult to justify a conclusion that the intervention has been successful. Although it is generally assumed that subjective reports primarily reflect the excellence of the fitted hearing aid(s) within the context of the patient’s everyday circumstances, there is relatively little research that assesses the validity of this assumption. In previous work, we have reported some contributions of the service delivery setting (private practice versus public health) to self-report outcomes. The purpose of the present investigation was to assess the relative contributions of patient variables (such as personality and hearing problems) and amplification variables (such as soft sound audibility, gain and maximum output) to self-reports of hearing aid fitting outcomes. Design: A cross-sectional survey of 205 patients was conducted with cooperation of eleven Audiology clinics. All subjects were recruited when they were seeking new hearing aids. Before the hearing aid fitting, measurements of personality and response bias were made, as well as measures of hearing problems and expectations about amplification. At the fitting, traditional verification data were measured including sound field thresholds, preferred gain for conversation, and maximum output. Six months after the fitting, a set of 12 standardized self-report outcomes was completed. Analyses concerned: (1) the associations among personality, response bias, and self-reports about hearing problems that are available before the hearing aid fitting, and (2) the associations of these precursor variables, and fitting verification data, with self-report data assessing the outcome of hearing aid provision. Results: Self-reports of hearing problems, sound aversiveness, and hearing aid expectations obtained before the fitting were found to be more closely related to the strength of certain personality traits than to audiometric hearing loss. Response bias also was associated with personality variables. Analyses of the collection of outcome measures produced a set of three components that were interpreted as a Device component, a Success component, and an Acceptance component. The Device component was construed as reflecting characteristics of the hearing aid whereas the two other components were construed as reflecting attributes of the wearer. The Success and Acceptance components were each significantly associated with several personality traits, but the Device component was not associated with personality. Variables available before the fitting accounted for 20 to 30% of each outcome component whereas amplification variables measured to verify the fitting accounted for only 10% on only one component. Conclusions: As reported in previous research, personality is associated with self-report outcome data. However, if practitioners utilize existing measures of hearing problems at the prefitting stage, separate personality data will not yield additional leverage in prediction of long-term fitting outcomes. Traditional fitting verification data as measured in this study, proved minimally useful in prediction of long-term outcomes of the fitting. A large proportion of variance in self-report fitting outcomes has yet to be accounted for. Finally, it appears that certain types of questionnaires might be more appropriate for research evaluating new amplification devices, whereas a different questionnaire approach might be optimal for evaluating intervention effectiveness in a clinical context.

Assessment of subjective outcome of hearing aid fitting: getting the client's point of view

This paper provides an overview of the measurement of hearing aid fitting outcomes in real life using self-report methods. Three topics are addressed: (1) why we should measure real-life outcomes; (2) seven different types of self-report outcome data; and (3) issues to consider in self-report outcome measurement- It is stressed that self-report data provide a unique view of the way that clients function and feel in their daily lives with regard to their hearing health. Self-report outcome data are multidimensional, and the different domains are only moderately correlated. When these types of data are gathered, it is essential to control relevant variables to safeguard validity and reliability.