musical backgrounds, listening habits, and aesthetic · pdf filemusical backgrounds, listening...

J Am Acad Audiol 11 : 390-406 (2000)

Musical Backgrounds, Listening Habits, and Aesthetic Enjoyment of Adult Cochlear Implant Recipients Kate Gfeller*t$ Aaron Christ* John F. Knutson§ Shelley Wittt Kathleen T. Murray§ Richard S. Tylert

Abstract

This paper describes the listening habits and musical enjoyment of postlingually deafened adults who use cochlear implants . Sixty-five implant recipients (35 females, 30 males) par-ticipated in a survey containing questions about musical background, prior involvement in music, and audiologic success with the implant in various listening circumstances . Responses were correlated with measures of cognition and speech recognition . Sixty-seven implant recipients completed daily diaries (7 consecutive days) in which they reported hours spent in specific music activities . Results indicate a wide range of success with music. In general, people enjoy music less postimplantation than prior to hearing loss . Musical enjoyment is influenced by the listening environment (e.g ., a quiet room) and features of the music.

Key Words : Cochlear implants, listening habits, music enjoyment, timbre

Abbreviations: HIGH = high school, POST = adult education, PRIM = primary school, SCT = Sequence Completion Test, SE = standard error, TERT = tertiary, VMT = Visual Monitoring Task, VMT1 = Visual Monitoring Task one per second rate, VMT2 = Visual Monitoring Task two per second rate .

T

he cochlear implant is an assistive hear-ing device that has been designed pri-marily for persons who are profoundly

deaf and who receive little or no benefit from tra-ditional hearing aids . Over the past two decades, considerable information has been amassed about the extent to which the various device types or coding strategies facilitate verbal com-munication . Because the device has been designed primarily for speech perception, it is not particularly surprising that little research has

*School of Music, 'Department of Speech Language Pathology and Audiology, $Department of Otolaryngology, "Department of Statistics and Actuarial Sciences, and §Department of Psychology, University of Iowa, Iowa City, Iowa

Reprint requests : Kate Gfeller, Department of Otolaryn-gology, University of Iowa, Iowa City, IA 52242

been conducted to date regarding the extent to which music is accurately perceived or enjoyed by adult implant recipients . However, music is a pervasive art form and environmental sound. Consequently, music perception and enjoyment seem reasonable topics for systematic inquiry in the context of evaluating the impact of implants .

Existing research on music perception and enjoyment of adult implant recipients suggests that some features of music are more effectively transmitted via the implant. For example, stud-ies investigating the rhythmic perception of adult implant recipients indicate that they are less accurate than normal-hearing adults on temporal perception tasks presented at rates that push the auditory system (Gfeller et al, 1997). However, these same recipients demon-strate accuracy of discrimination similar to that of normal-hearing adults when simple rhythm patterns are presented at moderate tempi (Gfeller and Lansing, 1991, 1992; Gfeller et al,

390

1997) . Thus, with regard to typical music lis-tening experiences, implant recipients are able to perceive with reasonable accuracy much of the basic meter (beat) and rhythmic patterns .

Whereas rhythmic patterns are transmit-ted fairly faithfully through the present gener-ation of implants, that is not the case for pitch sequences (melodies) . Implant recipients are sig-nificantly less accurate than are normal-hearing adults for perception of simple melodic patterns (Gfeller and Lansing, 1991, 1992 ; Gfeller et al, 1997 ; Pijl, 1997), and accuracy for pitch percepts varies considerably from one implant recipient to the next (Dorman et al, 1991) . Implant recip-ients also differ significantly from normal-hear-ing adults with regard to timbre (tone quality of instruments) recognition and enjoyment (Gfeller et al, 1998a) . More specifically, implant recipients are less accurate in identifying from sound alone commonly heard orchestral instruments (i .e ., violin, piano, clarinet, and trumpet) . Further, their ratings of the sound quality are generally lower and more restrictive in range than are the ratings by normal-hearing adults .

Because pitch and timbre are such impor-tant structural features of music, these research findings imply that musical enjoyment may well be an unrealistic outcome for implant recipi-ents and that implant recipients might avoid or forego those situations that involve consider-able music listening. In fact, some recipients implanted at The University of Iowa have expressed disappointment or frustration about how music sounds through their implant. How-ever, one cannot assume that personal enjoyment of music is a function only of accuracy of per-ception. Anecdotal reports from other implant recipients, as well as articles in implant user magazines, indicate that some implant recipients do enjoy music, despite the technical features of the present generation of devices.

Anecdotal reports of music enjoyment, although interesting, should not be assumed to represent the general population of implant users with regard to perceptual accuracy or appreciation of music postimplantation . There-fore, the purpose of this study was to system-atically gather information from a representative sample of implant recipients that would accu-rately reflect the experiences of implant recipi-ents with regard to musical involvement and enjoyment postimplantation . More specifically, the study was designed to determine whether cochlear recipients enjoy and engage in musical activities and whether their backgrounds, prior involvement in music, audiologic success with the

Music and Cochlear Implants/Gfeller et al

implant, and preimplant attributes are associ-ated with reports of music participation .

In order to answer these questions, two studies were conducted: study 1 administered a questionnaire to experienced adults who received a cochlear implant at the Iowa Cochlear Implant Clinical Research Center, and study 2 was designed to assess participation in music-related activities by recent implant recipients and long-term implant users with a daily diary following the recommendations of Tye-Murray et al (1993) .

STUDY 1

Method

Participants

Potential participants included 70 consec-utively implanted adults who had received a multichannel cochlear implant at The Univer-sity of Iowa as part of the National Institute on Deafness and Other Communication Disorders (NIDCD) grant-supported Iowa Cochlear Implant Clinical Research Center. To be eligible for an implant in this program, participants were required to be at least 18 years of age, have a postlingually acquired bilateral profound deafness (minimum 95 dB HL), and receive no measurable benefit from hearing aids . Addi-tionally, they participated in annual follow-up assessments that provided audiologic outcome measures . It was in the context of the routine annual follow-up assessments that participants were contacted by letter and invited to partici-pate in the current study, all who are active participants in our center's research program . Sixty-five implant recipients (94% response rate) completed and returned the questionnaire . This group included 35 females and 30 males rang-ing in age from 29 to 80 years of age (mean = 57.63, SD = 14.99) . Length of profound hearing loss ranged from less than 1 year to 58 years (mean = 11.82, SD = 12.63) . There were 29 Clar-ion, 17 Nucleus, 11 MED-EL, and 8 Ineraid users ; length of implant use ranged from 33 months to 142 months (mean = 85.69, SD = 33.21) .

Questionnaire

The questionnaire used for this study, the Iowa Musical Background Questionnaire (Gfeller et al, 1998b), was a modified and expanded ver-sion of a questionnaire used in prior research by Gfeller and Lansing (1991, 1992). Additional

391

Journal of the American Academy of Audiology/Volume 11, Number 7, July/August 2000

questions were added regarding formal musical training, the enjoyment of specific musical instruments and musical styles, and particular environmental circumstances that may influ-ence the quality of musical sound. These addi-tional questions were developed after interviewing implant recipients (N = 35) in our own clinic and through a survey published in CONTACT, an international magazine expressly for implant recipients . The survey consisted of open-ended questions regarding musical involve-ment and enjoyment. The responses were coded for prominent themes and objective multiple-choice questions relevant to the purposes of this study were developed (Gfeller, 1996,1998; Gfeller et al, 1997). Other portions of the questionnaire focused on attitude or personal preferences regarding particular types of musical sounds . Attitudes were measured using Likert-type scales, visual analog scales, and open-ended questions.

A draft of the expanded questionnaire was given to three audiologists and three implant recipients, who evaluated the document for clarity, appropriateness, and comprehensive-ness of the items. A second draft of the ques-tionnaire was completed and was then reviewed once again by two audiologists, three adult implant recipients, and four professionals involved in the research and design of cochlear implants. The final version of the questionnaire consisted of 21 items including multiple-choice questions, Likert-type rating scales, visual ana-log scales, and open-ended questions.

Speech and Cognitive Measures

the possible impact of missing data that would have occurred by using data points from one particular testing appointment.

Because both timbral recognition and speech perception by implant recipients were predicted by a number of experimental cognitive mea-sures administered prior to implantation (see Gfeller et al, 1998; Knutson et al, 1991), scores from those measures were correlated with selected music questionnaire scores . The cogni-tive measures used were originally included to assess several specific attentional and cogni-tive abilities . These tests consist of a Visual Monitoring Task (VMT) and a Sequence Com-pletion Test (SCT). The VMT, as a test of atten-tion, reaction time, and working memory, requires that the subject observe a computer screen as numbers are displayed at either a one per second rate (VMT1) or a two per second rate (VMT2). When the displayed number reflects an even-odd-even pattern, the subject is required to strike a key on the computer keyboard . The score from the VMT is a standardized signal detection score that incorporates accurate responding (true positives and true negatives) and inaccurate responding (false positives and false negatives) . The score from the SCT is based on the work of Simon and Kotovsky (1963) and reflects the number of sequences correctly iden-tified in a fixed time period. As a measure of asso-ciative memory, participants are required to correctly identify the numbers that are associ-ated with 12 two-dimensional geometric forms, with the score reflecting the number of trials to mastery and the mean number of correct iden-tifications across 10 trials .

To determine whether music listening and participation were related to the demographic characteristics of the participants or their audi-ologic performance with the implant, survey data were correlated with the following mea-sures: age, length of profound deafness, length of implant use, and speech perception measures. The speech perception measures included audi-tion-only versions of the Consonant Test with and without noise, the Iowa Sentence Test (Tyler et al, 1986), the Vowel Recognition Test (Tyler et al, 1986), and the Northwestern University Audi-tory Test No . 6 (Tillman and Carhart, 1966). Because most of the implant recipients have multiple measures of speech perception over time, we selected the speech perception data as follows: all audiologic tests for as long as 4 years prior to participating in the survey were aver-aged to reduce test variation. This also reduced

Results

Musical Background and Listening Habits

Musical Training Prior to Hearing Loss. Six questionnaire items focused on formal musi-cal training prior to hearing loss, including par-ticipation in music lessons, musical ensembles (e.g ., band, choir, and orchestra), and music classes (e .g ., general music classes, music appre-ciation or theory, etc.) . Respondents were asked to list the number of years of training for each category of musical training (e.g ., music class, orchestra, etc.) . In addition, respondents were to choose one of five descriptions that best matched their self-assessment of exposure to and experience with music. The five response options ranged from no formal music training or

392


knowledge about music to extensive formal training and considerable knowledge about music.

A total score for formal musical training prior to hearing loss was determined as follows: to reduce colinearity among the many training variables, the years of training were summed into four time categories : (primary school) PRIM, (high school) HIGH, (tertiary) TERT, and (adult education) POST. Two individuals, however, reported an extremely high number of years training for the TERT and POST variables (12 and 20 years) . Since most respondents listed zero years in these variables, the values for these two individuals were truncated to a level beyond the range of the other participants (2 years in the case of TERT variables and 3 years in the case of POST variables) . This allowed a reflection of years of training noticeably higher than any other respondent without creating wildly outlying and highly influential values for the overall score. Factor analysis revealed two significant factors, namely, PRIM and an aver-age of HIGH, TERT, and POST. The catego-rization of musical training into these two factors (PRIM and the combined score of HIGH, TERT, and POST) seems quite appropriate from a cur-ricular standpoint in that primary school music is often general in nature and emphasizes expo-sure to music and enjoyment of music, whereas high school, tertiary, and adult education are more likely to include participation in musical ensembles that required advanced musical train-ing and skill development. Musical involvement in high school and beyond is more likely to be an elective activity (thus, self-selection becomes a stronger factor in participation) than is par-ticipation in elementary school music classes.

The range of scores for formal musical train-ing during the elementary school years ranged from 0 to 27 (mean = 6.15, SD = 6.22), with higher scores indicating several years of expe-rience in multiple musical activities (e .g ., par-ticipation in band, general music, and choir throughout elementary school). The range of scores for formal musical training in high school, college, or after college was 0 to 4.67 (mean = .77, SD = 1 .14) . Table 1 presents the breakdown of global assessment of musical training, knowl-edge, and experience . Seventy-seven percent indicated that they listened to or were involved in musical activities prior to hearing loss .

Listening Habits . A Likert-type scale (1 point for strongly disagree to 4 points for strongly agree) was used to ascertain to what extent

Table 1 Self-assessment of Musical Training, Knowledge, and Experience

Category Percentile

No formal training, little knowledge about 23.0 music, and little experience in listening to music

No formal training or knowledge about music, 40.0 but informal listening experience

Self-taught musician who participates in 1 .5 musical activities

Some musical training, basic knowledge 23.0 of musical terms, and participation in music classes or ensembles in elementary or high school

Several years of musical training, knowledge 7 .7 about music, and involvement in music groups

these implant recipients would describe them-selves as persons who often chose to listen to music. One item solicited their interest in music prior to hearing loss and another item their musical interest postimplantation . Next, they were to indicate how much they typically listened to music prior to hearing loss and then after becoming accustomed to their implant. They indicated the amount by selecting one of four cat-egories of listening time per week (e.g ., 1 point = 0-2 hours weekly, 4 points = 9 or more hours of listening weekly). These two items (attitude and listening habits) were combined to create a score for music listening habits prior to hearing loss (range of 2-8 points) and another score for postimplantation (range of 2-8 points). The mean score for listening habits prior to hearing loss was 5.29 (SD = 2.02) . The mean score for lis-tening habits postimplantation was lower, at 4 .11 (SD = 1 .94) .

General Assessment of Satisfaction with the Device for Music Listening

An overall score for satisfaction regarding how music sounds postimplantation, as com-pared with recollection of how music sounded prior to hearing loss, was obtained from a sin-gle seven-choice questionnaire item . Possible responses ranged from no interest or enjoyment before or after implantation to considerable enjoyment after implantation . The seven items were combined into three general categories : (a) 23 percent indicated little satisfaction in music listening prior to or after implantation, (b) 43 percent reported that the sound of music is improving over time or is better than no music at all (although less pleasant than before hearing

393


loss), and (c) 23 percent noted that music now sounds as pleasant as before hearing loss, or more so . The general listening satisfaction scores were then correlated with musical background, listening habits, age, hearing history, and speech and cognitive measures. Significant correlations were found between general enjoyment of music and the following measures : postimplant lis-tening habits (r = .73, p<_ .00) and age (r = -.42, p S .00). No significant correlations were obtained for musical background, preimplant listening habits, length of profound deafness, length of implant use, or any of the speech measures .

Figure 1 Mean score for adjective descriptors of music through the implant .

Adjective Descriptors of Music

The overall quality of musical sound heard through the implant was assessed using seven 100-mm visual analog scales, each anchored with bipolar adjectives . The bipolar adjectives were selected in response to empirical data on adjective descriptors for musical sounds (Von Bismarck, 1974 ; Pratt and Doak, 1976), as well as prior feedback from implant recipients regard-ing the quality of musical sounds through the implant. The adjective pairs were as follows: like-dislike, pleasant-unpleasant, natural-mechanical, clear-fuzzy, sounds like music-does not sound like music, easy to follow-difficult to follow, and simple-complex. Figure 1 shows the mean values obtained for the seven adjective pairs.

Scores on the positive pole (maximum of 100) of the adjective pairs were then correlated with age, length of profound deafness, length of implant use, musical background, listening habits, and speech and cognitive perception

measures . Table 2 provides a matrix of correla-tions for those measures that result in a large proportion of significant correlations . As can be seen, age, listening habits (especially postimplant listening), and speech perception measures were those significantly correlated with the largest number of positive adjective descriptors. There were no significant correlations for length of profound deafness and length of implant use and few isolated significant correlations with musi-cal training and cognitive measures . The over-all pattern of the latter correlations seems most consistent with a type I error.

Enjoyment of Musical Styles

A number of implant recipients have reported informally that particular styles of music sound better than others with the implant or have noted changes in style preference postim-plantation (Gfeller et al, 1996b; Gfeller, 1998).

Table 2 Correlation Matrix for Adjective Descriptors, Age, Preimplant Listening Habits, Postimplant Listening Habits, and Selected Speech Perception Measures

Pleasa nt Like N atural Clear

Sounds Like Music

Easy to

Follow Simp le

r p n r p n r p n r p n r p n r p n r p n

Age -.500 *** 65 -.523 *** 62 -.610 *** 62 -.519 *** 62 -.462 *** 64 -.536 *** 61 -.124 57 Preimplant habits 273 * 63 265 * 61 .193 60 .324 * 61 .315 * 63 .311 * 60 012 56 Postimplant habits .712 *** 62 737 *** 60 .693 *** 60 .648 *** 60 .736 *** 61 738 *** 60 .328 * 55 IA Consonant Test 255 59 338 * 56 168 56 .342 * 56 303 * 58 .328 * 55 -.087 51

Noise IA Consonant Test .216 59 297 * 56 .090 56 285 * 56 292 * 58 .274 * 55 -.107 51

Sound Sentence Test .411 * 59 427 ** 56 .262 56 .425 ** 56 .475 ** 58 .384 ** 55 -.022 51 NU-6 words 244 58 291 * 55 .272 * 56 .351 * 56 .339 * 57 .331 * 55 -.118 51

* <05, **<001, ***.0001 . r = correlation, p = p value, n = number of participants responding .

394


Therefore, the questionnaire included pre- and postimplantation preferences for 11 specific and commonly heard musical genre (e .g., country and western, pop, classical ; Gfeller et al, 1996a ; Hoffer, 1992) that tend to have particular and typical structural characteristics with regard to rhythm, harmonic structure, instrumenta-tion, and melody. For example, country and western often includes a simple melody line sung by a vocalist (consequently, the individual may be able to extract words), a clear and pre-dictable beat, and a relatively simple, repetitive, and predictable harmonic structure . In contrast, many classical selections are played by large orchestras, have no vocal parts, and tend to have complex harmonic, melodic, and rhythmic structures .

In analyzing possible shifts in music style preference, a simple chi-square analysis of mar-ginal distributions is inappropriate due to an inherent association between the pre- and postimplant responses . Because the music pref-erence responses had intrinsic ordering (strongly dislike, dislike, like, strongly like), the use of a proportional odds cumulative logit' analysis is appropriate for testing such a preference shift . The association between pre-and postimplant responses for any particular music type was predicted to be strong . That is, if a participant

had a strong preference for a particular style of music prior to implantation, then that partici-pant would be much more likely to have simi-

lar preference after implantation rather than changing to the other extreme . This assumption is easily justified because of the influence of family, friends, personal habits, and enduring opportunities for exposure . Due to this predicted strong association between pre- and postim-plant responses, a linear-by-linear association model was fit to each music style . This choice, in effect, assumes that most of the responses are going to be along the diagonal (i.e ., a participant's preferences pre- and postimplant would be close to the same), whereas the corners opposite the diagonal will have few, if any, counts . Although this association structure is important for the

'Basically, a proportional odds cumulative logit analy-sis describes the ratio : (odds of being less than or equal to state j postimplant) / (odds of being less than or equal to state j preimplant, where j = strongly dislike, dislike, like, strongly like) . For parsimony, this ratio was assumed to

be constant . Also, participants not responding to either a pre- or postoperative question for a particular music style were not included in the analysis of that music style (Agresti, 1990).

Table 3 Musical Style Preference, Pre- and Postimplantation

Music Type Odds

Proportion 95% Confidence

Interval

Pop 2.57 (1.17 5 .65) Rock 1 .99 (1 .11 3 .58) Hard rock 2.25 (1 .26 4 .02) Rap 1 .26 (0 .67 2 .35) Blues 1 .90 (0 .70 5 .19) Jazz 1 .64 (0 .74 3 .61) Country 1 .60 (0 .97 2 .65) Show tunes 3.12 (1 .65 5.89) Easy listening 3.67 (1 .78 7.58) Religious 2.19 (1 .34 3.56) Classical 1 .59 (0 .78 3.27)

An odds proportion greater than 1 denotes a shift downward in the appreciation of a given musical style from pre- to postimplant. As the odds proportion increases, the magnitude of the shift rises accordingly . A confidence interval that contains 1 reflects the lack of a statistically significant shift in appreciation from pre- to postimplant

As an example of interpreting the above table, we would say that for pop music, the odds of a level j or lower response

postimplant is 2 .57 (1 .17 to 5 .65) times higher than the odds of a

level j or lower response prelmplant, for j = 1 (strongly dislike), 2 (dislike), 3 (like) . Note that since there are only four levels,

looking at level four or lower would include all respondents .

estimation of model parameters, there is greater interest in assessing shifts up or down in the

marginal distributions . Therefore, an unre-stricted association was also fit to the data, giv-ing similar but slightly deflated results in regard

to the marginal shifts measured . Table 3 contains results for each individual style .2 Participant responses fit the association and marginal com-ponents reasonably well . Although about half of the odds proportions were not significantly dif-ferent from 1, the fact that they were all posi-tive provides some evidence that there was a consistent shift downward in appreciation of musical styles from pre- to postimplant . Although the fitted model can account for possible strong associations between pre- and postimplant responses for a single style, there is no account for associations between styles . For example, if participants have different interpretations of the rating scale, participants common to any pair of styles changes would inflate the standard

'Since a third to a half of participants did not respond to most music types, the tables or responses were rather sparse . To increase stability in the estimation, a small con-stant of 0 .001 was added to each cell count. This constant was chosen through an empirical sensitivity analysis in regards to effectiveness of stabilization coupled with neg-ligible parameter influence.

395


Table 4 Musical Style Preference, Pre- and Post-Implantation

Show Tunes Easy Listening Difference 95% Confidence Interval Difference 95% Confidence Interval

Pop 1 .21 (0.70 2 .10) 1 .43 (0 .62 3.31) Rock 1 .57 (0.99 2 .47) 1 .84 (0 .94 3.63) Hard rock 1 .39 (0.73 2.64) 1 .63 (0 .68 3.93) Rap 2.48 (1 .13 5.47) 2.92 (1 .07 7.99) Blues 1 .64 (0 .79 3.38) 1 .93 (0 .81 4 .62) Jazz 1 .90 (1 .08 3.37) 2.24 (1 .18 4.27) Country 1 .95 (1 .06 3.59) 2.29 (1 .15 4.57) Religious 1 .42 (0 .74 2.74) 1 .68 (0 .90 3 .14) Classical 1 .96 (1 .11 3.46) 2.30 (1 .19 4.47)

A difference greater than 1 denotes a proportionally greater shift downward in the appreciation of show tunes or easy listening from pre- to postimplant as compared to the downward shift in the appreciation of the various music types listed in the left column . As the difference increases, the shift in the appreciation of show tunes or easy listening becomes proportionally greater. A confidence interval that contains 1 reflects the lack of a statistically significant difference between the shifts in appreciation from pre- to postimplant.

An example of reading this table would be for rap music, the odds proportion (as defined in the previous table) for show tunes is 2.48 (1 .13 to 5.47) times larger than that for rap music. That is, if for rap music, the odds of responding level j or lower postimplant is two times higher than the odds of responding level j or lower preimplant (for j = 1 [strongly dislike], 2 [dislike], 3 [like]), then for show tunes, the odds of responding level j or lower postimplant is 2.2 .48 = 4.96 times higher than the odds of responding level j or lower preimplant (for j = 1 [strongly dislike], 2 [dislike], 3 [like]) .

error (SE) of the obtained difference in appre-ciation between the two tested styles . However, jackknife SE estimates based only on those par-ticipants responding pre- and postimplant for each pair of styles were remarkably close when calculated on a style-wise basis. Thus, the jack-knife SE estimates lend confidence to the accu-racy of the preference change when looking at pairwise differences as well. Interestingly, "show tunes" (see Table 3) and "easy listening" (see Table 3) were the only two that showed any sig-nificantly different shifts in the proportional odds . Table 4 expresses these differences and gives an approximate 95 percent confidence interval using the jackknife SE estimates. Again, it should be noted that although only about half of the styles showed significant differences, the overall trend of the point estimates provides extra evidence that the two styles show more of a drop than the other forms.

senting different principles of sound production and different frequency ranges . They include string family (violin and cello), woodwind family (flute, clarinet, and saxophone), brass family (trumpet and trombone), percussion (piano, pitched percussion ; drum, nonpitched percus-sion), and organ. The latter was included because of its familiarity and common use and because it was often specifically noted in the open-ended sur-vey of implant recipients (Gfeller, 1998).

Appraisal or degree of liking of those same 10 instruments was measured using 100-mm visual analog scales with bipolar anchors of "dis-like very much" (score = 0) to "like very much" (score = 100) . Respondents were asked to indi-cate how much they enjoyed the sound quality of each instrument that they recall having heard postimplantation . Figure 2 lists the mean score

Enjoyment of Different Instrumental Timbres

Most normal-hearing people have personal preference for some instruments over others (e.g ., may prefer a trumpet to a violin, etc.) . Prior research indicates a significant difference between normal-hearing adults and implanted adults regarding appraisal of selected musical instru-ments (Gfeller et al, 1998a) . As a result, the ques-tionnaire included items to assess reported liking of 10 commonly heard musical instruments rep-resenting different instrumental families repre-

Drums

Piano

Trumpet

Saxophone

Flute

Cello

Organ

Violin

Trombone

Clarinet

iffiffiRlsoffin

D Dislike

6D Ve ry

''Much Mean Score

Figure 2 Mean appraisal scores for instrumental timbre .

100 like very Much

396


Table 5 Correlation Matrix : Instrumental Timbre Appraisal x Age, Postimplant Listening Habits, Vowel Test, VMT1, VMT2, and SCT

Age

r p n

Postimplant Habits

r p n

Vo

r

wel

p

Test

n r

VMT 1

P n r

VMT2

P n r

SCT

P n

Violin -.518 ** 42 360 * 40 394 37 401 * 35 406 * 36 370 * 39 Cello -.590 * 744 ** 17 406 16 578 * 14 561 * 14 812 *** 15

Flute -.555 ** 29 620 ** 28 524 ** 25 591 ** 26 616 ** 27 539 ** 27 Clarinet -.391 * 28 306 25 540 26 538 * 22 519 * 23 313 26

Saxophone -.457 * 29 574 * 27 508 26 288 25 330 26 508 * 27 Trumpet -.377 * 34 518 * 32 537 30 235 29 233 30 349 * 32 Trombone -.294 28 345 26 365 25 593 * 24 439 * 25 682 *** 26 Piano -.355 * 550 *** 46 332 43 178 42 182 40 256 41

Drum -.371 * 45 421 * 42 135 40 398 * 39 244 40 274 41

Organ -.082 49 304 * 46 086 44 157 43 338 * 44 118 45

* <05, **<001 . *** 0001 . r = correlation, p = p value, n = number of participants responding .

for appraisal arranged by instrumental family groups . Although evaluations were generally in the upper half of the range, there was consid-erable variance among respondents and among instruments. The range of mean scores across the 12 instruments, however, is relatively narrow, indicating no obviously "better" timbres for implant recipients as a whole.

Correlations between self-reported enjoy-ment of the various instrumental timbres were calculated with age, length of profound deaf-ness, age of implant use, listening habits, speech perception, and cognitive perception measures (Table 5) . Those factors significantly correlated with the ratings of the greatest number of instru-ments include age and postimplant listening habits . The VMT, at both the one and two per second rate, correlated significantly with the ratings of violin, cello, flute, clarinet, and trom-bone but did not correlate with ratings of trum-pet, saxophone, and piano. The VMT at the one per second rate correlated with the ratings of drums, and at the two per second rate, the VMT correlated with the ratings of organ. The SCT correlated with ratings of violin, cello, flute, saxophone, trumpet, and trombone but did not correlate with ratings of clarinet, piano, drum, and organ. Taken together, these data suggest that preimplant abilities to rapidly identify sequentially arrayed stimuli may relate to enjoy-ment of particular aspects of music stimuli (e.g., tone quality) postimplantation . Neither length of profound deafness nor length of implant use was significantly correlated with any of the instrument ratings. Musical training prior to deafness was significantly correlated with only two of the instrument ratings (trombone and

flute) . Few significant correlations (one to three instrument ratings per speech perception mea-sure) were obtained between speech perception and timbre appraisal measures, with the excep-tion of the vowel test, which was correlated sig-nificantly with 6 of the 10 instruments.

Environmental/Listening Circumstances That Enhance or Impede Music Listening

Sixteen specific questions provided an oppor-tunity for the respondents to rate enjoyment associated with various musical environments or types of sound equipment on a 4-point Likert-type scale (with a score of 4 being most enjoy-able) . Figure 3 shows the percentage of respondents who reported having had experience listening in each of the situations, along with the mean scores for particular listening situations . Although few reported playing a musical instru-ment, with the exception of using direct input, the majority of implant recipients had had musi-cal experience in most of the settings assessed . Importantly, however, the mean ratings of the musical experiences suggest that they were not highly enjoyable.

The listening environments included in the questionnaire are complex listening situations that often require the listener to organize and comprehend a mixture of music and speech, to integrate the signal with prior knowledge about the music, or to extract the musical sound from background noise. Music varies considerably in its structural complexity and may be presented with or without visual cues . Therefore, we exam-ined the extent to which satisfaction in music lis-tening in these listening circumstances was

397


Response to "Since my Implant, I usually enjoy listening to music In the following situations."

No difference! No answer

Background on the TV -

at a movie music at social event -

on aPA-

tn the car On record, over speakers tape or CD

direct input

Making music

singing

instrument

0 10 15 20

Number of responses

Strongly Agree M Disagree Agree M Strongly Disagree

Figure 3 Extent to which the respondent enjoys music listening in particular listening settings .

correlated with age, length of profound deaf-ness, length of implant use, musical background, listening habits, speech outcome measures, and preimplant cognitive measures . Those factors yielding the strongest correlations include age (negatively correlated with satisfaction in all 16 listening situations except one and signifi-cantly correlated for 8 of those), preimplant lis-tening (11 of the 16 situations), and postimplant listening habits (13 of the 16 listening situa-tions) . Listening to music while in the car, either from the radio or from a cassette recording, is significantly correlated with a number of speech

21

17

10

11

33

28 35

14

24 18

20

13

19

40

49

31

30

measures (Table 6) . The VMT (regardless of stimulus presentation rate) correlated signifi-cantly and positively with ratings of enjoyment of the radio or recorded music played in the car, recorded music through speakers, singing, indoor and outdoor concerts, and music on TV The SCT correlated with ratings of enjoyment in those same settings, except outdoor concerts (only approached significance) . The SCT and the VMT (one per second rate) also correlated sig-nificantly with enjoying music over public address systems in restaurants and stores. Taken together, these correlations suggest that

398


Table 6 Correlation Matrix : Listening Situations x Age, Training, Pre and Post Listening Habits, All Speech and Cognitive Measures

Organ Ch oir Radio/ Car

Ra Di In

dio/ rect put

Ra W

or H

dio/ ork ome

Li Con Indo

ve Li cert Con ors Out

ve cert door

Back Mu ground sic/ TV

r p n r p n r p n r p n r p n r p n r p n r p n

Age -.028 64 047 61 -.525 *** 64 -.109 58 -.486 *** 63 -.179 62-120 60 -.489 *** 63

Length of use 071 64 121 61 - .158 64 -.118 58 - .031 63 094 62-024 60 .016 63 Strategy used -.125 64 -.168 61 - .262 * 64 -.315 * 58 - .258 * 63 -.129 62-049 60 -.113 63 Primary school 273 * 64 261 * 61 219 * 64 .093 58 .130 63 199 62 .037 60 .237 63 Secondary 242 64 232 61 133 64 .111 58 .105 63 284 * 62 .266 * 60 .245 63 and university

Preimplant 142 62 123 59 321 * 62 .140 58 .324 * 62 320 * 60 .332 * 59 .307 * 61 listening habits

Postimplant 207 61 272 * 59 689 *** 61 310 * 55 .706 *** 60 412 * 59 .344 * 58 .638 *** 60 listening habits

Consonant 113 58 -.018 56 419 * 58 .159 52 .358 * 57 061 56 .098 55 .243 58 test in noise average

Consonant 094 58 -.036 56 353 * 58 .226 52 .313 * 57 073 56 .064 55 .206 58 sound only average

Sentence test 275 * 58 187 56 605 *** 58 .302 * 52 .536 *** 57 168 56 .077 55 .360 * 58 sound only average NU-6 word 189 57 039 55 542 *** 57 .232 51 446 ** 56 066 55 .038 54 .246 57 recognition test average VMT1 140 55 131 52 273 * 55 .034 50 .233 55 313 * 53 .308 * 52 .326 * 54 VMT2 211 56 093 53 237 56 .016 51 190 56 302 * 54 .248 53 .279 * 55

Sequence 055 58 048 55 299 * 58 .068 52 .252 57 323 * 56 .227 54 .299 * 57 completion test

* <05, **<001, ***< .0001 . r = correlation, p = p value, n = number of participants responding .

those implant recipients who are better at iden-tifying rapidly changing and sequentially arrayed stimuli find musical experiences under less than ideal conditions to be more enjoyable than implant recipients who do not manifest those cognitive and attentional skills . No sig-nificant correlations were found for length of pro-found deafness or length of implant use .

In addition to using a Likert-type-scale, we devised a list of 20 factors that implant recipi-ents had reported as affecting the quality of their musical experiences. Some factors are related to nonmusical cues and past listening experiences (e.g., watching the performer, length of cochlear implant use) ; other factors were related to the properties of the music itself (e .g., structural simplicity, volume, etc. ) . Still other fac-tors were related to the acoustic properties of the listening environment, whether it was ambient sound or through headphones (e.g ., quiet envi-ronment, direct input, etc.) . In response to each item, the respondent was to mark a "+" if that

factor enhanced music listening, a "0" next to items that had no perceptible influence, and a

next to factors that seemed to impede musi-cal enjoyment . As Figure 4 illustrates, certain factors, such as prior familiarity with the music or a quiet listening environment, are especially helpful in an enjoyable music listening experi-ence . A large proportion of respondents indi-cated that loud volume or a noisy room detracts from listening enjoyment . Figure 4 provides the relative ratings of each item for positive or neg-ative impact . No specific implant device was identified with particularly positive or negative music listening experiences on any of the factors evaluated in this study.

Discussion

A wide range of musical involvement and enjoyment on the part of adult cochlear implant recipients was reported . Because musical tastes

399


Table 6 Correlation Matrix : Listening Situations x Age, Training, Pre and Post Listening Habits, All Speech and Cognitive Measures (continued)

Back- ground at Movies

Back- ground at Event

B grou Sy

ack- nd/PA stem

Recording/ Car

Reco Di In

rding/ rect put

Reco O

Spe

rding ver akers

Pla a

Instru

ying n ment Sin gin g

r p n r p n r p n r p n r p n r p n r p n r p n

Age -.406 * 61 - .302 * 63 -.266 * 58 -.571 *** 63 -.141 55 -.522 *** 61 -.084 56 -.114 57 Length of use 063 61 132 63 034 58 -.122 63 -.195 55 -.072 61 -.003 56 .031 57 Strategy used -.030 61 - .096 63 -.052 58 -.129 63 -.162 55 -.049 61 .032 56 -.181 57 Primary school 245 61 308 * 63 219 * 58 .184 63 .032 55 044 61 .319 * 56 .305 * 57 Secondary and 187 61 223 63 209 58 .103 63 .050 55 107 61 .331 * 56 .335 * 57 university Preimplant 194 59 182 61 296 * 57 .381 * 61 279 * 55 .328 * 60 .301 * 55 .328 * 56 listening habits

Postimplant 557 *** 59 511 *** 60 .546 *** 55 .762 *** 60 .434 * 53 817 *** 58 .134 54 .178 55 listening habits

Consonant 234 56 194 57 .113 53 .411 * 57 .143 50 293 * 56 .290 * 53 .177 54 test in noise average

Consonant 207 56 167 57 108 53 .383 * 57 .225 50 275 * 56 .294 * 53 .185 54 sound only average

Sentence 374 * 56 388 * 57 .192 53 .483 *** 57 .144 50 317 * 56 .226 53 .208 54 test sound only average NU-6 word 253 55 233 56 .106 52 .399 * 56 .093 49 258 55 .252 52 .304 * 53 recognition test average VMT1 .254 52 213 54 351 * 49 .295 * 54 .117 47 248 53 .165 49 .343 * 49 VMT2 151 53 137 55 251 50 .296 * 55 .052 48 245 54 .219 50 .372 * 50 Sequence 134 55 199 57 297 * 52 .412 * 57 .203 49 311 * 55 .199 50 .372 * 51 completion test

* <.05, **<001, ***<0001 . r = correlation, p = p value, n = number of participants responding .

and involvement vary considerably within the normal-hearing population, as well as among persons with assistive hearing devices, it is important to consider postimplantation responses in contrast to reports of musical lives prior to hearing loss in order to more clearly determine the impact of the implant on music involvement. Not surprisingly, global self-assess-ment of music listening prior to deafness and postimplantation indicates that implant recip-ients generally report spending less time lis-tening to music after implantation than prior to sustaining a profound hearing loss .

The extent to which music listening is sat-isfactory after implantation varies considerably among recipients . Approximately one-quarter of the sample indicated little satisfaction in music listening prior to or after implantation . However, approximately two-thirds of the sam-ple were optimistic because (a) music was becom-ing more pleasant over time, (b) some music

was better than none, or (c) music was actually considered to be more pleasant than what was recalled prior to the hearing loss . Several per-sons who reported greater enjoyment of music postimplantation had lost their hearing over a lengthy time period ; in their last few years of hearing aid use, hearing aids did little more than transmit basic rhythmic pulses. Conse-quently, the implants, although far from pro-viding faithful representation of music, were reported to offer more input with regard to pitch-based and timbral percepts than did traditional hearing aids . Thus, a number of implant recip-ients indicated that hearing music through an implant, although not perfect, was preferable to not hearing music at all.

If ratings of instruments reflect something about the benefit one receives from an implant, the correlates of instrument ratings provide an interesting picture of predictors of one dimension of implant benefit. Although it is not surprising

400


Non-musical Cues and Listening Experience

Follow musical score, words

Watch performer

Length of CI use

Change over time

Practice, music listening

Coding stragegy

Features of the Music

Simple musical structure

Familiar music

Know song title

Clear rythm, beat

Quality recording

Soft volume

Experience playing the music Loud volume

Listening:. Environment

Quiet environment

Good accoustics

Optimal seating

Direct audio Input

Echoey room

Noisy room

M= I

Number of responses hinders

Figure 4 Factors that enhance or impede musical enjoyment.

that appraisal of musical instruments correlated with age and postimplant listening habits, it is somewhat surprising that duration of profound deafness and duration of implant use were not predictive of ratings of instruments . Both of those factors have been related to other indices of implant benefit (Tyler and Summerfield, 1996). The fact that the preimplant cognitive measures correlated with the ratings of musical instru-

ments is conceptually consistent with the recent work by Gfeller et al (1998a), where those preim-

plant cognitive measures predicted the ability to recognize timbral differences in instruments . The present data and the earlier work strongly implicate some cognitive abilities in determining a range of implant benefits well beyond speech perception . Additionally, the modest correlations between speech perception and the ratings of

Number of responses helps

16 19

33

43 38

56

16 14

16 21

24 24

49 32

10 16 22 40

18 11

sound quality of instruments in the present study are also consistent with the lack of an associa-tion between accuracy in the identification of instrumental timbre and speech perception in the Gfeller et al (1998a) study. Importantly, these data do not support the idea that particular musical instruments are inherently more pleas-ant for implant recipients .

The possible importance of specialized cog-nitive abilities in the outcome of implants was also implicated in the correlations between rat-ings of musical stimuli in various listening envi-

ronments . That is, the VMT has been shown to predict speech perception (e.g ., Knutson et al,

1991) and timbral recognition (Gfeller et al, 1998a), and in the present study it predicts the appreciation of music in some very difficult lis-tening environments . Importantly, since the

401


VMT is not a self-report measure, the correla-tions between the VMT and the various indices of musical listening activity and enjoyment can-not be attributed to some shared method vari-ance . Although the speech perception measures did not correlate well with the instrument rat-ings, strong correlations between speech recog-nition tasks and reported participation in complex music listening tasks probably reflect the role of speech perception in those activities where there is likely to be a combination of both spoken and musical information. The fact that various aspects of music listening are differen-tially correlated with different types of speech recognition and cognitive tests indicates that music listening may require different acoustic processing than speech and reinforces the idea that the term music reflects a wide range of acoustic sounds, listening situations, and social circumstances. In short, one global indicator for music perception or enjoyment does not ade-quately reflect the diverse universe of musical sounds and listening experiences any more than one test could appropriately reflect all aspects of spoken communication.

Not surprisingly, general enjoyment of music was positively correlated with time spent in lis-tening to music postimplantation . However, it is not clear whether more practice or exposure to music enhances music listening over time or whether those who receive a relatively pleasant signal soon after hook-up are more likely to spend time in music listening. Further testing is required to determine the effects of listening experience . It is noteworthy that age was neg-atively correlated with almost every indicator of musical enjoyment, including general satisfac-tion . Perhaps the changes in perceptual pro-cessing associated with aging have a negative impact on an older implant recipient's ability to use complex sequential information such as music. It is also noteworthy that there are no obvious differences among the responses of implant recipients who use devices based on very different coding strategies .

According to these data, there are no par-ticular styles or genre of music that implanted adults report as preferable to others as a func-tion of the implant. Average scores for each musical style showed some decline in appraisal from recollection of enjoyment prior to deafness compared with present enjoyment postimplant. However, those declines are relatively modest. Only in the case of two genres, "easy listening" and "show tunes," do statistical analyses indi-cate a significant drop in reported enjoyment of

that style postimplantation. The structural char-acteristics of the easy listening genre may in part explain this decline. Although many implant recipients find that a clear and prominent beat facilitates understanding, many "easy listen-ing" selections do not include a clear and obvi-ous rhythmic pulse. In addition, some implant recipients find it helpful to follow along with familiar lyrics while listening. Many "easy lis-tening" selections include only instrumental music, such as an orchestra of string instru-ments. Often, "easy listening" music is played over public address systems at places of business . This is a listening situation that many of these respondents characterized as less than ideal. Finally, perhaps the term "easy listening" seems ironic to implant recipients, who find it difficult to follow along with music (see Fig. 1). The genre of "show tunes" often includes lyrics, which may or may not be intelligible to implant recipients . However, there is often a full orchestra and sometimes a chorus that accompanies the solo voice. Perhaps this very complex multitude of instruments and voices playing many notes at one time results in a signal that is too complex for implant recipients using the present gener-ation of devices.

STUDY2

T he questionnaire data from study 1 indicate that implant recipients can describe a range

of musical experiences subsequent to their receipt of a cochlear implant. Moreover, the data indicate that there may be considerable vari-ability among implant recipients with respect to the degree to which they find musical experiences enjoyable. Although the questionnaire data pro-vide a reliable and efficient means of collecting information from implant recipients, the results reflect the participants' integration of the sum total of their listening experiences over a long period of time . To determine whether musical lis-tening experiences varied as a function of implant use, a daily diary was used to determine the number of hours in a day that implant recip-ients actually devoted to musical listening.

Method

Participants

Sixty-seven postlingually deafened adult implant users drawn from the same population as study 1 participated. Fifty-two of the implant recipients from study 1 also participated in

402


study 2. Because the diary protocol was intro-duced after many implant recipients had had years of experience with their cochlear implants, the sample of 67 participants included Ineraid and Nucleus users with up to 7 years of experi-ence with an implant, as well as Clarion users who had only 1 year of implant use.

Diary Forms. The diary forms included true-false items with respect to engaging in various hearing activities and items that permitted the respondents to indicate the number of hours during 1 day devoted to a number of specific lis-tening contexts . For the purposes of the pre-sent study, five items related specifically to musical activities were analyzed .

0.5

R

0.0

Procedure

Approximately 2 to 3 weeks prior to annual follow-up visits to the Iowa Cochlear Implant Clinical Research Center, research participants were mailed a packet of daily diary forms and seven self-addressed postage-paid return envelopes. Participants were instructed to com-plete one diary form each day for 7 consecutive days . To ensure that participants actually com-pleted the diaries on a daily schedule, they were told to mail a diary form as soon as it was com-pleted . Research assistants could then use post-marks to assess compliance with the daily protocol .

Results

To determine whether time engaged in musi-cal activities was affected by the duration of implant use, the mean number of hours per day devoted to the use of radio, recorded music, con-cert attendance, and the playing of musical instruments was contrasted between three groups of participants . Long-term users (n = 15) had 60 months or more of implant experience, midterm users (n = 21) had at least 36 months of implant experience but less than 60 months, and short-term users (n = 31) had 24 months or less experience with the device . All of the long-

term users were recipients of the Ineraid or Nucleus devices, and all of the short-term users were recipients of the Clarion device .

Figure 5 shows the mean number of hours of each of the various musical activities engaged in by the three groups . Although the overall means suggest that the short-term users engaged

Radio Recorded Concerts Instruments

Figure 5 Mean hours of musical participation reported on daily diaries at follow-up .

in somewhat more daily musical participation, the variability within groups was as great as the variability across groups, and group differences did not meet contemporary standards of statis-tical significance .

As an alternative analytic strategy, a within-subjects analysis of musical participation was completed with a subgroup of participants

(n = 29) who completed diaries at annual inter-vals across 3 years . Much like the between-sub-jects analyses, the data did not suggest that musical participation changed appreciably as a function of implant experience . Importantly, however, based on both the within- and between-subjects analyses, the diary data do indicate that some cochlear implant recipients may devote considerable time to musical listening . The overall mean scores, however, did not ade-quately reflect the fact that some implant recip-ients actively participated in a large amount of musical activities and others did not .

Although analyses of speech perception out-come data do not suggest that there are signif-icant differences between the audiologic outcomes of the Ineraid and Nucleus implants, design differences between the two devices sug-gest that there could be differences in respond-ing to musical stimuli . Additionally, because the Clarion device uses yet another processing strat-egy that could affect musical listening, the per-centage of recipients of each device who participated in various musical listening activ-ities was assessed . For this analysis, the per-centage of recipients with each device who engaged in musical activity at follow-ups from 12 months to 120 months is shown in Figure 6 . To complete this analysis, there had to be at least six participants within the device group who provided data . Clearly, the long-term data are primarily based on Ineraid and Nucleus

403


3 4 5 6 7

Years Postimplantation

and the playing of musical instruments, a major-ity of implant users actively engage in some musical listening.

Discussion

The daily diaries of study 2 provide a sys-tematic replication and extension of the results of study 1. Most importantly, for both long- and short-term implant users, participation in a broad range of musical activities was reflected in the daily diaries. There was, however, con-siderable variability among recipients with respect to the type and amount of musical par-ticipation, regardless of the implant device used and the duration of implant experience . The lack of systematic change in diary reports of music participation as a function of implant experience in both between- and within-subject analyses is somewhat surprising . Although the detection of change could be compromised by the variance, it may be that implant recipients are not actively pursuing musical listening oppor-tunities and testing their limits . Thus, although the diary data, uncompromised by long-term retrospective recall, indicate that musical lis-tening is one important aspect of implant out-come that is worthy of consideration in the evaluation of the current generation of cochlear implants, the data do not suggest that musical listening has been fully explored by implant recipients .

GENERAL DISCUSSION

Years Postimplantation

Figure 6 Percentage of implant users who reported A, listening to music on the radio ; 8, using recorded music ; and C, playing a musical instrument in daily diaries at annual intervals .

recipients and the short-term data are primar-ily based on Clarion recipients .

Three patterns are most apparent in Figure 6. First, there are no obvious differences among implant types with respect to engaging in daily musical activity. Second, there is no clear asso-ciation between musical activity and amount of experience with the implant. Finally, except for attendance at concerts (which are not typically a daily occurrence regardless of hearing status)

T he data from both study 1 and study 2 indi-cate that musical enjoyment varies not only from one recipient to the next but that there are often intraindividual differences that occur depending on factors such as the listening sit-uation (e.g., quiet room, good sound equipment, prior familiarity with the music being heard) or structural features of the music (e.g., a clear rhythmic beat). Therefore, studies regarding perception and enjoyment of music by implant recipients need to take into account the wide range of structural features and factors that have an impact on music satisfaction when eval-uating success of the implant in music percep-tion and enjoyment. Furthermore, it appears that most implant recipients can enhance their music enjoyment to some extent by controlling environmental conditions . Implant recipients may also take initiative in following visual cues, such as following the lyrics of a song, in order to more easily understand the signal . These

404

practical listening strategies can reduce the impact of the limitations of the present gener-ation of devices .

For persons who once enjoyed music but

who find the present signal to be unacceptable, music can be a frustrating experience . "I hear

it as tinny," writes one person. "I love music and miss it very much . Most music is not very pleasant to listen to . It is all pretty confusing," responded another. Yet another recipient writes, "I think that listening to music is a wonderful pastime, giving hours of pleasure and enjoy-

ment . I wish I could really enjoy it again . . ." Still another implant recipient describes music as "just noise."

However, a perusal of open-ended comments by implanted recipients emphasizes the socio-cultural component of music listening, that is,

musical enjoyment is influenced by more than the acoustic signal . A number of implant recip-ients describe ways in which music listening has enhanced the quality of their life with regard to particular life events (e .g ., music as a part of funerals or holiday celebrations) or social struc-tures . One recipient states that he attends all concerts for his area concert association : "I get very little out of them . It is still better than sit-ting at home doing nothing ." Another writes, "Getting music back [that is postimplantation]

filled in holes for me . It's the little stuff, even the music they play on the Cable Weather Channel, the music that accompanies skaters in compe-titions-with the cochlear implant, I realize how much more interesting it is to watch when I can hear the music and relate it to their chore-ography . . . I think that those of us who were intimately connected to music before our hear-ing losses notice the little ways the music is part of the warp and woof of life, and relish our recovery of something priceless ." Although these implant users acknowledge openly that music does not sound "normal" or completely satis-factory, there are some users who emphasize the significant role that having "some music" can play in helping them to feel connected to the hearing world . For example, one person writes, "This last Christmas, on PBS, I listened to the Moorhead College Choir of Atlanta . The singing was great . It was captioned, which helped me

with the words . I really enjoyed the presentation, so much so that tears literally streamed down my cheeks the whole time . No, music is not the same as before, but it's good . I can experience it,

feel it ." Another person wrote about the impor-

tant expressive role that music played at the time

of his father's funeral : "Taps from the military


rites [played at the funeral] played over and over in my mind, which is OK-that is a part of grieving . [After the funeral] I put in a CD called Classical Jukebox, curled up on the couch with a warm blanket, and listened to that music in a quiet house for hours . . . Music opens up ways to enjoy and cope with life ." These comments emphasize the importance of considering the sociocultural and acoustic aspects of the listen-ing experience when considering music percep-tion and enjoyment via the cochlear implant .

Acknowledgment. This study was supported by grant

P50 DC00242 from the NIDCD, NIH; grant RR00059 from the General Clinical Research Centers Program, NCRR, NIH, and the Iowa Lions Foundation . Thanks are due to Ms . Maureen Mehr in her assistance with preparation of this manuscript .

REFERENCES

Agresti A. (1990) . Categorical Data Analysis . New York : Wiley, 322-324.

Dorman M, Basham K, McCandless G, Dove H. (1991) . Speech understanding and music appreciation with the Ineraid cochlear implant. Hear J 44:32-37 .

Gfeller K. (1996) . Appreciating music with the implant. CONTACT 11(2):36-37 .

Gfeller K. (1998) . Music appreciation from the perspec-tive of implant recipients . CONTACT 12(3):24-25 .

Gfeller K, Lansing C. (1991) . Melodic, rhythmic, and tim-bral perception of adult cochlear implant users. J Speech Hear Res 34:916-920 .

Gfeller K, Lansing C. (1992) . Musical perception of cochlear implant users as measured by the Primary Measures of Music Audiation: an item analysis . J Music Ther 29(1):18-39 .

Gfeller KE, Hanson-Abromeit D, Knutson J, Witt S. (1996a, November). The Development of a Popular Music Appraisal Battery for Cochlear Implant Recipients . Poster presented at the National Association for Music Therapy, Nashville, Tennessee.

Gfeller K, Knutson JF, Woodworth G, Witt S, DeBus B. (1998a). Timbral recognition and appraisal by adult cochlear implant users and normal-hearing adults . JAm Acad Audiol 9:1-19.

Gfeller K, Witt S, Knutson J, Coffman D, Woodworth G. (1996b, April) . The Effects of Familiarity and Complexity on Music Perception of Cochlear Implant Recipients . Poster presented at the Music Educators National Conference, Kansas City, MO .

Gfeller K, Woodworth G, Robin D, Witt S, Knutson JF. (1997) . Perception of rhythmic and sequential patterns by normally hearing adults and adult cochlear implant users. Ear Hear 18:252-260 .

Gfeller KE, Witt S, Kim KH, Knutson J . (1998b) . The Iowa Music Perception and Appraisal Battery. Iowa City,

405


IA : The University of Iowa Hospitals, Department of Otolaryngology-Head and Neck Surgery.

Hoffer C . (1992) . Sociology and music education . In : Colwell R, ed . Handbook of Research on Music Teaching and Learning. New York : Schirmer, 713-723 .

Knutson JF, Hinrichs JV, Tyler RS, Gantz BJ, Schartz HA, Woodworth G. (1991) . Psychological predictors of audiological outcomes of multichannel cochlear implants : preliminary findings . Ann Otol Rhinol Laryngol 100: 817-822.

Pijl S. (1997) . Labeling of musical interval size by cochlear implant patients and normally hearing subjects . Ear Hear 18:364-372 .

Pratt RL, Doak PE . (1976). A subjective rating scale for timbre . J Sound Vibration 45:317-328.

Simon HA, Kotovsky K. (1963) . Human acquisition of concepts for sequential patterns . Psychol Rev 70:543-546.

Tillman TW Carhart R. (1966) . An Expanded Test for Speech Discrimination Utilizing CNC Monosyllabic Words: Northwestern University Auditory Test No . 6. Technical Report No . SAM-TR-66-55 . Brooks Air Force Base, TX: USAF School of Aerospace Medicine .

Tye-Murray N, Knutson JF, Lemke JH . (1993) . Assess-ments of communication strategies use: questionnaires and daily diaries Semin Hear 14:338-353 .

Tyler RS, Preece JP, Tye-Murray N. (1986). Iowa Phoneme and Sentence Test [laser video disc] . Iowa City, IA : The University of Iowa, Department of Otolaryngology.

Tyler RS, Summerfield AQ . (1996) . Cochlear implanta-tion : relationships with research on auditory deprivation and acclimatization . Ear Hear 17:385-505 .

Von Bismarck G . (1974) . Timbre of steady sounds : a fac-torial investigation of its verbal attributes. Acustica 30 : 146-172 .

musical backgrounds, listening habits, and aesthetic · pdf filemusical backgrounds, listening...

Documents