International Journal of Speech-Language Pathology
ISSN: 1754-9507 (Print) 1754-9515 (Online) Journal homepage: http://www.tandfonline.com/loi/iasl20
Development of Trivia Game for speech understanding in background noise Kathryn Schwartz, Stacie I. Ringleb, Hilary Sandberg, Anastasia Raymer & Ginger S. Watson To cite this article: Kathryn Schwartz, Stacie I. Ringleb, Hilary Sandberg, Anastasia Raymer & Ginger S. Watson (2015) Development of Trivia Game for speech understanding in background noise, International Journal of Speech-Language Pathology, 17:4, 357-366, DOI: 10.3109/17549507.2014.979875 To link to this article: http://dx.doi.org/10.3109/17549507.2014.979875
Published online: 24 Nov 2014.
Submit your article to this journal
Article views: 60
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=iasl20 Download by: [University of California Santa Barbara]
Date: 05 November 2015, At: 17:17
International Journal of Speech-Language Pathology, 2015; 17(4): 357–366
Development of Trivia Game for speech understanding in background noise
KATHRYN SCHWARTZ1, STACIE I. RINGLEB2, HILARY SANDBERG1, ANASTASIA RAYMER1 & GINGER S. WATSON3 Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
1Communication
Disorders & Special Education, 2Mechanical & Aerospace Engineering, 3Instructional Design & Technology, Old Dominion University, Norfolk,VA, USA
Abstract Purpose: Listening in noise is an everyday activity and poses a challenge for many people. To improve the ability to understand speech in noise, a computerized auditory rehabilitation game was developed. In Trivia Game players are challenged to answer trivia questions spoken aloud. As players progress through the game, the level of background noise increases. A study using Trivia Game was conducted as a proof-of-concept investigation in healthy participants. Method: College students with normal hearing were randomly assigned to a control (n 13) or a treatment (n 14) group. Treatment participants played Trivia Game 12 times over a 4-week period. All participants completed objective (auditoryonly and audiovisual formats) and subjective listening in noise measures at baseline and 4 weeks later. Result: There were no statistical differences between the groups at baseline. At post-test, the treatment group significantly improved their overall speech understanding in noise in the audiovisual condition and reported significant benefits in their functional listening abilities. Conclusion: Playing Trivia Game improved speech understanding in noise in healthy listeners. Significant findings for the audiovisual condition suggest that participants improved face-reading abilities. Trivia Game may be a platform for investigating changes in speech understanding in individuals with sensory, linguistic and cognitive impairments.
Keywords: Auditory processing, auditory comprehension, adults
Introduction Successful understanding of speech is affected by a number of internal and external factors in listening situations (Sweetow & Palmer, 2005). Intensity of the speech signal, complexity of the incoming message, rate of speech input, integrity of the incoming speech signal (dialect, articulatory precision) and acoustic environment are external influences on the ability of a listener to process an incoming signal. Likewise, factors internal to the listener, such as sensory, linguistic and cognitive abilities, also impact accurate and efficient processing of speech signals. While all of these factors can influence and challenge speech processing even in healthy listeners, difficulties can be especially exacerbated for listeners with disabilities affecting sensory, linguistic and other cognitive abilities. One of the most over-riding complaints heard from healthy and impaired listeners is difficulty when listening in environments with background noise, such as a reverberant setting or competing speakers (Assmann & Summerfield, 2004; Neff & Green, 1987).
Individuals with neurologic damage, such as stroke or brain injury, often experience difficulties understanding speech (Davis, 2007). Individuals with aphasia have impairments of linguistic comprehension and have significantly more difficulty than healthy controls when processing speech in the presence of background noise (Basili, Diggs, & Rao, 1980; Healy, Moser, Morrow-Odom, Hall, & Fridriksson, 2007; Kittredge, Davis, & Blumstein, 2006). Mild forms of traumatic brain injury and blast exposure can disrupt auditory and visual sensory processes (Oleksiak, Smith, St. Andre, Caughlan, & Steiner, 2012; Saunders & Echt, 2012) as well as cognitive processes (e.g., attention, memory) critical for speech processing (Belanger, Kretzmer, YoashGantz, Pickett, & Tupler, 2009; Cornis-Pop, Mashima, Rothi, MacLennan, Picon, Hammond, et al., 2012). Recent evidence shows that hearing loss is also common in populations with these disorders (Lew, Garvert, Pogoda, Hsu, Devine, White, et al., 2009). Moreover, individuals with mild forms of traumatic brain injury commonly voice complaints
Correspondence: Dr Stacie I. Ringleb, Old Dominion University, Mechanical & Aerospace Engineering, Kaufman Hall 238C, Norfolk, VA 23529, USA. E-mail: [email protected] ISSN 1754-9507 print/ISSN 1754-9515 online © 2014 The Speech Pathology Association of Australia Limited Published by Informa UK, Ltd. DOI: 10.3109/17549507.2014.979875
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
358
K. Schwartz et al.
of exacerbated auditory processing problems in the presence of background noise, although this finding has not been systematically examined (Fausti, Wilmington, Gallun, Myers, & Henry, 2009; Musiek, Baran, & Shinn, 2004). Individuals with certain developmental conditions also can have inordinate difficulty when listening to speech in noise, such as children with central auditory processing disorders (Bamiou, Campbell, & Sirimanna, 2006). Individuals with hearing loss complain frequently of increased difficulties when listening in noise (e.g., Hannaford, Simpson, Bisset, Davis, McKerrow, & Mills, 2005). The primary means to address hearing loss is the use of hearing aids. Individuals with hearing loss fitted with hearing aids may still be distressed by speech understanding difficulties in noisy situations (Kochkin, 2005; Takahashi, Martinez, Beamer, Bridges, Noffsinger, Sugiura, et al., 2007), thereby leading to reluctance to participate in noisy life situations or discontinued use of the hearing aids altogether. Clearly, speech processing in individuals with a variety of developmental and acquired sensory, cognitive and linguistic impairments is undermined by background noise. Even though these conditions have different underlying causes for decreased speech understanding in noise, all experience difficulties processing speech in noise at either the peripheral or central processing level (Grant & Walden, 2013). Auditory training is a common way to address these difficulties. Auditory training teaches individuals to maximize all available auditory and visual cues to improve speech understanding (Schow & Nerbonne, 2013). Computerized auditory training: Listening in noise Given the numerous groups that experience difficulty listening in noise, several methods of auditory training have been developed to address these issues. Sweetow and Palmer (2005) refer to some of these training approaches as analytic, as they incorporate tasks that target bottom-up processing of acoustic properties of speech signals, often at the syllable or single word level. Other components of auditory training centre on synthetic or top-down processing, such as teaching compensatory strategies to enhance auditory processing (e.g., reducing background noise, increasing reliance on visual input) (Chisolm, Abrams, & McArdle, 2004). Auditory training that incorporates both analytic and synthetic tasks have been implemented with some success in a group setting (Hawkins, 2005). However, application of auditory training in individualized sessions, while providing some positive outcomes, can be quite costly and may not be reimbursed in the current healthcare setting (Sweetow & Palmer, 2005). Recently, several computer-administered programs have been developed to provide direct, individualized auditory intervention for individuals with
hearing problems (Pallarito, 2011; Sweetow & Sabes, 2006). Computerized programs emphasize individualized auditory training that is cost-effective, fun and rewarding, accessible, verifiable and with necessary feedback (Sweetow & Palmer, 2005). Several programs are available that focus on various elements of auditory processing: Computer-Assisted Speech Perception Training-CASPER-Sent (Boothroyd, 2008), Earobics (Cognitive Concepts, 1997), FastForWord (Scientific Learning Corporation, 1998), Listening and Communication Enhancement program (LACE) (Sweetow & Sabes, 2004), ReadMyQuips (Levitt, Oden, Simon, Noack, & Lotze, 2011), and Seeing and Hearing Speech (Ingvalson, Lee, Fiebig, & Wong, 2013). Improvements in speech understanding in noise have been demonstrated utilizing computerized auditory training programs (Olson, Preminger, & Shinn, 2013; Song, Skoe, Banai, & Kraus, 2012; Sweetow & Sabes, 2004, 2006). Although benefits are evident for existing computerized auditory training tools, the programs are not without their limitations. Few existing tools provide visual face information along with listening in noise tasks, a critical factor to enhance top-down speech processing (Chisholm et al., 2004; Sweetow & Palmer, 2005). Further, the engagement and interest level for the computerized training tasks is somewhat limited, thereby reducing compliance (Sweetow & Sabes, 2010). Given these shortcomings, Trivia Game, a computerized auditory training program based on more interactive gaming technology, was developed. Development of trivia game Trivia Game is played independently where players view a man on the computer screen who speaks a trivia question presented in the presence of multitalker babble. The multi-talker babble is looped with no silent intervals. The purpose of the game is to hear the trivia questions in the presence of background noise and answer them correctly. Trivia questions are divided into 12 categories representing familiar topics such as sports, music and foods, with 40–50 questions per category, allowing listeners to use a unique category each playing session. Players receive points for correctly and quickly answering the question by selecting the written answer on the screen. The primary score is the percentage correct and is based on responses to the original trivia question. A series of hints are provided to guide the listeners who did not answer correctly on the first attempt. Hints are provided to keep the player engaged even if they have little knowledge of the trivia category. Hints provide bonus points and are not included in the percentage correct for the purposes of adjusting the background noise levels. Figure 1 illustrates a sample screen from Trivia Game.
Stuttering in school children in Egypt
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Figure 1. Trivia Game screen shot. Male speaker is verbally saying the phrase, “What is the fastest land animal?”
A total of 23 signal-to-noise ratio (SNR) levels are included in the game. The SNR levels range from 27 dB (level 1) to SNR of 39 dB (level 23). SNR is defined as the difference between the sound level of the spoken dialog and the competing noise, where the lower the SNR, the more difficult it is to hear the speaker. An adaptive algorithm was used in which the background noise systematically increases (becomes more difficult) with success in the game (defined as at least 80% correct across five successive questions) or reduces with difficulty (defined as less than 40% correct across five consecutive questions). The game starts at an SNR of 27 dB. When 80% is achieved the SNR level decreases by 3 dB (making speech understanding more difficult) and continues to decrease in 3 dB intervals as long as the 80% is sustained. If the percentage correct is between 40– 80%, the SNR remains unchanged. If the percentage correct drops to 40% or less, the SNR level increases by 3 dB (making speech understanding easier). This adaptive branching is a form of levelling used to deliver more challenge to the player who performs well or less challenge to the player who performs poorly. Providing the appropriate amount of challenge to match performance is a key factor in players’ willingness to play a game repeatedly across long periods of time (Kiili, 2005). To account for player knowledge of a particular trivia topic, a final screen is displayed that asks if the player knew the answer to the trivia question prior to the hints. If the player answers “yes” indicating they knew the answer, the percentage correct will change accordingly. If the player answers “no” the percentage correct remains unchanged. This ensures that the difficulty of the game adjusts based on what the player heard and not just the knowledge of a trivia category. If a player knew the answer to the trivia question but answered incorrectly due to the difficult SNR level (they did not hear the question), the percentage correct would decrease. If the player knew the answer to the trivia question and answered correctly, it is assumed that they heard the question and, therefore, the percentage correct increased.
359
Individuals can experience difficulty listening in noise for a variety of sensory, cognitive or linguistic difficulties. Thus, they may require varied auditory rehabilitation treatments to address those diverse disturbances. Because the game is conducted using specific semantic categories, players can constrain their expectations of words to be processed while listening to the trivia in noise. This factor could be especially beneficial to individuals with language impairments such as aphasia. Further, the visual information provided by reading the waiter’s face as he presents trivia questions and clues provides support to the listener as the signal degrades with increased background noise. Thus, the game has the potential to enhance top-down processing for individuals with a variety of sensory, cognitive or language disorders that undermine bottom-up processing. Trivia Game was developed as a potential tool for use by a variety of populations that experience auditory difficulties in noise. Prior to implementing the game in one of those populations, a proofof-concept study was conducted to determine if Trivia Game was effective in improving speech understanding in an unimpaired population. The resulting study examined the effects of Trivia Game in healthy, college-age listeners compared to untrained controls. The specific research questions for this study were: (1) Does the treatment group perform better than the control group on objective listening-in-noise measures? (2) Is there a difference in performance gain for the treatment group in an auditory-only vs audiovisual condition?
Methods Participants Forty-three college students were recruited for this proof-of-concept study. Participants were randomly assigned to a treatment or control group using a random number generator that determined group membership prior to the study. Of the 27 participants who completed all phases of the experiment, 13 were in the treatment group (six males, seven females) with a mean age of 21.6 years (SD 2.1) and 14 participants served as controls (seven males, seven females) with a mean age of 21.4 years (SD 2.3). Of those who did not complete the study, six were originally assigned to the control group and 10 to the treatment group. Participants who failed to complete the study reported scheduling conflicts, lack of interest and technical difficulties running the game as the reasons for withdrawal. All participants had hearing thresholds 25 dB HL for pure-tone frequencies from 250–8000 Hz, bilaterally. In addition, all participants were native speakers of American English and passed an auditory processing screening as determined by the SCAN-3 (Keith,
360
K. Schwartz et al.
2009). The researchers recruited participants through fliers placed around Old Dominion University’s campus and through announcements made in classes. The study was approved by the Old Dominion University Institutional Review Board (IRB). All participants provided written informed consent to participate in this study.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Test measures Quick Speech-In-Noise (Quick SIN). The QuickSIN (Killion, Niquette, Gudmundsend, Revit, & Banerjee, 2004) test was used as an objective measure of the amount of speech that is understood in the presence of background noise. The test presents sentences to a listener who verbally repeats the sentence. Scores are calculated on the accuracy of each word repeated. QuickSIN is composed of 18 lists with six sentences per list. Each sentence contains five key words. The sentences are spoken by a female talker in the presence of four-talker babble (one male and three females). In each list, the sentences are presented at progressively lower signal-to-noise ratios (SNR). The first sentence is presented at a SNR of 25 dB and subsequent sentences in the list decrease SNR by 5 dB decrements, with the last sentence in the list being presented at 0 dB SNR. QuickSIN was administered in the standard auditory-only format and in an experimental audiovisual format in which the female speaker was both heard and viewed on a computer screen. This version was provided by the Killion group. Synchronization of the auditory and visual signals was verified daily. The audiovisual presentation of QuickSIN was used as a secondary measure of audiovisual benefit. The primary findings of this study are based upon the results of the standard QuickSIN and the subjective rating scales. Speech, Spatial and Qualities of Hearing Scale (SSQ). In addition to the objective measurement of listening in noise, all participants completed the SSQ version 5.6 (Gatehouse & Noble, 2004) during the initial test session to ensure that subjective auditory experiences were similar for the treatment and control groups. This questionnaire was chosen to reflect how well participants function in the real world when communicating in difficult listening situations. The SSQ is a 50-item subjective questionnaire with three sub-scales which evaluates a variety of listening conditions encountered in daily life: speech hearing, spatial hearing and qualities of hearing. Each question describes a communication scenario and asks a question relative to a participant’s performance ability in that listening situation. For example, question six states “You are in a group of about five people in a busy restaurant. You CANNOT see everyone else in the group. Can you follow the conversation?” Participants rate their ability on a 10-point visual rating
scale ranging from 0–10 to indicate the ease of hearing in the scenarios posed by each item. Zero indicates “not at all” and 10 indicates “perfectly”. Therefore, higher scores indicate greater ease of listening in everyday scenarios. Participants were instructed to answer the questions based on their everyday experiences. Each question included a “not applicable” option that the participant could select if the situation did not apply. Responses that were marked “not applicable” were not scored. The SSQ-B (B for Benefit) version 5.6 (Jensen, Akeroyd, Noble, & Naylor, 2009) was completed by all participants at the completion of the study. This instrument was administered in order to measure the subjective impact of game-playing on the treatment participant’s ability to understand speech in difficult listening situations. The SSQ-B measures benefit of aided hearing aid conditions when compared to unaided conditions. The instrument consists of all items on the SSQ with identical instructions and scoring criteria, but uses an absolute rating scale that compares aided experiences to those without hearing aids. The SSQ-B is scored on a relative scale ranged from 5 to 5, where 5 is “much worse”, 0 is “unchanged” and 5 is “much better”. The SSQ-B wording was modified by replacing “comparing your ability now with your ability before getting your hearing aid/s” with “comparing your ability now with your ability before study participation”.
Procedures All participants completed a pre-treatment laboratory test battery of pure tone testing, auditory processing screening, QuickSIN and SSQ questionnaire. Four weeks after initial laboratory testing, all participants returned for post-testing with the QuickSIN and the SSQ-B. The QuickSIN, SSQ and SSQ-B were scored by two independent assessors. These assessors were not involved in the development of Trivia Game and did not have any financial conflict of interest. All participants were compensated for their time at the end of the study; control participants received $20 and treatment participants were compensated $100.
Outcome measures For the QuickSIN test, participants were tested bilaterally in two conditions: auditory-only and audiovisual. For the auditory-only condition, the QuickSIN stimulus was played at 70 dB HL on a CD player routed through an audiometer (GSI-61) to TDH-50 headphones. For the auditory-visual condition, the same auditory set-up was used with the addition of a computer monitor. The computer monitor displayed an image of a female speaking the stimulus sentences as the participant listened through the
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Stuttering in school children in Egypt
headphones. A practice list of six sentences was given to each participant to acquaint them with the task for both the auditory-only and audiovisual conditions. All testing was administered in a double-walled sound booth. Each participant was given 16 lists during each testing session (eight lists in the auditory-only condition and eight lists in the audiovisual condition). Listening conditions and list order were counterbalanced and randomized across participants. This ensured that no lists were repeated in a single session and that participants did not hear the same list in the same listening condition for either session. QuickSIN was scored as the number of key words correctly repeated. In clinical settings, QuickSIN is scored in words correctly repeated, but the score is converted to a score of “SNR loss”. The SNR loss quantifies the extent to which the hearing aid user’s speech understanding in noise is worse than that of normalhearing listeners. Because the purpose of this test was to determine speech perception benefit in normal hearing listeners following Trivia Game participation, the SNR loss was not calculated for the current study. Two versions of the SSQ were administered (SSQ and SSQ-B). Participants completed the SSQ surveys using a paper and pencil format. During pre-testing, all participants completed the SSQ to characterize their impressions of listening experiences and difficulties in daily life. In post-testing, all participants completed the SSQ-B where they compared their listening experience after study participation to listening experiences prior to study participation. Because Trivia Game focuses on hearing speech in noise, only the Speech Hearing sub-scale of the SSQ and SSQ-B was scored.
361
Results Training logs Review of the training logs indicated that participants progressed through the game. Twelve of the 13 participants recorded their game level for every session and results are displayed in Figure 2. The remaining participant was inconsistent in recording and results are not displayed. The most difficult SNR in the game was 39 dB, which was reached by 58% of participants. Of the remaining participants, four reached an SNR level of 36 dB and one reached an SNR level of 12. QuickSIN The objective listening in noise measure used in the experiment was QuickSIN. The total number of words correct for each condition (auditory-only and audiovisual) for the pre-test and post-test sessions were calculated. A one-way ANOVA was conducted to compare the pre-test scores of the treatment and control groups for each QuickSIN condition. Following that, a one-way ANCOVA with pre-test scores as a covariate was conducted to assess the effects of the training. Use of a one-way ANCOVA was chosen to avoid errors associated with repeatedmeasures ANOVA for treatment and sessions (Dimitrov & Rumrill, 2003). Auditory-only QuickSIN. A one-way ANOVA was conducted to evaluate the equivalence of the treatment and control groups on the number of words correct on the QuickSIN pre-test score for the auditory-only condition. The test was not significant, F(1,25) 0.02, p 0.88, indicating that there was not a significant difference between the treatment (M 194.62, SD 9.88) and control
Trivia Game play Treatment group participants were introduced to Trivia Game during the pre-test laboratory session and were given verbal and written instructions on how to play the game. They were instructed to set the computer volume at a comfortable level at the beginning of each session and not to change the volume after the session had begun. Participants were further instructed to play in a quiet environment at the same time each day. They were loaned Sennheiser PC 151 headphones to use with their personal computer and were instructed to play the game four times a week over a 3-week period, for a total of 12 training sessions. They completed a playing log where they recorded their game level at the end of each session. Each session lasted ∼ 20 minutes, with participants playing one trivia category per session. Contacts were made on a weekly basis with participants to answer questions or troubleshoot equipment problems.
Figure 2. Participant SNR (signal-to-noise ratio) level at end of each training session. Each line represents an individual participant. Lower scores indicate more difficult listening situations.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
362
K. Schwartz et al.
(M 195.14, SD 8.26) groups on the pre-test score. The eta2 index indicated that less than 1% of the variance in the QuickSIN auditory-only pre-test score was accounted for by whether a participant was assigned to the treatment or control group. A one-way analysis of covariance (ANCOVA) was conducted. The independent variable, group, included two levels: treatment and control. The dependent variable was the gain in number of words correct auditory-only condition calculated between the post-test and pre-test sessions. The covariate was the number of words correct in the pre-test session. A preliminary analysis evaluating the homogeneity-of-slopes assumption indicated that the relationship between the pretest audio-only score and the gain in the audio-only score did not differ significantly as a function of the group, F(1,23) 0.02, MSE 26.92, p 0.89, partial eta2 0.001. The ANCOVA was not significant, F(1,24) 3.86, MSE 25.85, p 0.06. The gain in words correct for the audio-only condition adjusted for pre-test scores was nearly significant. The treatment group demonstrated higher gain (M 10.59), while the control group showed a lower gain (M 6.74), as illustrated in Figure 3. The strength of the relationship between groups and gain in words correct for the audio-only score was assessed by a partial eta2 with the group factor accounting for 14% of the variance in the gain in words correct in the audio-only condition, holding constant the number of words correct in the pre-test audio-only condition. Audiovisual QuickSIN. A one-way ANOVA was conducted to evaluate the equivalence of the treatment and control groups on the number of words correct on the QuickSIN pre-test score for the audiovisual condition. The test was not significant, F(1,25) 1.17, p 0.29, indicating that there was not a significant difference between the treatment (M 208.00, SD 9.39) and control (M 211.79, SD 8.77) groups on the number of words correct pre-test audiovisual score. The eta 2 index indicated that less than 5% of the variance in the QuickSIN audiovisual pre-test score was
Figure 3. Total QuickSIN Words Correct for the Auditory-Only condition. Error bars represent one standard deviation.
accounted for by whether a participant was assigned to the treatment or control group. A one-way analysis of covariance (ANCOVA) was conducted to assess the independent variable, group (treatment and control) on the dependent variable gain in number of words correct audiovisual condition calculated between the post-test and pre-test sessions. The covariate was the number of words correct in the pre-test session. A preliminary analysis evaluating the homogeneity-of-slopes assumption indicated that the relationship between the pre-test audiovisual score and the gain in the audiovisual score did not differ significantly as a function of the group, F(1,23) 1.73, MSE 36.43, p 0.20, partial eta2 0.07. The ANCOVA was significant, F(1,24) 8.15, MSE 37.54, p 0.01. The gain in words correct for the audiovisual condition adjusted for pre-test scores showed significantly higher gains for the treatment group (M 9.72) than the control group (M 2.83), as shown in Figure 4. The strength of the relationship between groups and gain in words correct for the audiovisual score was assessed by a partial eta2 with the group factor accounting for 25% of the variance in the gain in words correct in the audiovisual condition, holding constant the number of words correct in the pre-test audiovisual condition. SSQ The Speech and Hearing sub-scale of the SSQ questionnaire was administered at pre-test session (max score 140). The scores on the SSQ were compared for the two groups using an independent samples t-test. The analysis was completed on 24 participants as three participants (one control and two treatment) failed to complete the instrument. There were no significant differences between the treatment (M 109.97, SD 17.88) and control (M 106.07, SD 17.51) groups at pre-test, t 0.54, p 0.79. A Cronbach’s alpha internal reliability coefficient of 0.92 was calculated for these scores.
Figure 4. Total QuickSIN Words Correct for the Audiovisual condition. Error bars represent one standard deviation. * p 0.05 ** p 0.01 for Gain in Words Correct.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Stuttering in school children in Egypt
The SSQ-Benefit sub-scale (SSQ-B) was administered post-test. The analysis was completed on 26 participants as one treatment participant failed to complete the instrument. One participant in the control group marked all items as “not applicable” and one in the treatment group marked four items as “not applicable”. These items were not included in scoring. Independent samples t-test were conducted to compare the mean benefit for the treatment and control groups. The treatment group (M 1.23, SD 0.90) reported significantly greater benefit than the control group (M 0.25, SD 0.53), t 3.37, p 0.003. A Cronbach’s alpha internal reliability coefficient of 0.89 was calculated for these scores. As presented in Figure 5, the treatment group reported benefits in all listening situations. Table I lists the exact wording for each listening situation. The perceived benefits for the treatment group were significantly greater than the control group in conditions where the signal was in competition with background noise: single-talker conversation with TV on (item 1), group conversation in a restaurant (item 4), individual conversation in the presence of background noise (item 5), group conversation in a restaurant but cannot see talker (item 6), following television news with single-talker simultaneously (item 10) and single-talker conversation with multiple conversations in the background such as a cocktail party (item 11). Table II summarizes the descriptive statistics for each item for each group. Discussion Understanding speech in the presence of background noise is problematic for a variety of populations, such as those with aphasia, presbycusis, traumatic brain
363
injury and developmental auditory processing problems (Belanger et al., 2009; Kochkin, 2005; Takahashi et al., 2007). Despite the use of aural rehabilitation and/or hearing aids to address sensory impairments, the difficulty understanding speech often persists. The current study served as proof of concept for the efficacy of Trivia Game, a computer program created to improve speech understanding in noise for a variety of populations. Results suggest that Trivia Game improves speech understanding in noise in an unimpaired population. Game performance The overall design of the game and implementation of the training program (12 sessions over 4 weeks) was adequate to observe improvement in both the subjective and objective measures of listening in noise. Furthermore, game logs indicate that all treatment participants improved their speech understanding during game play, ending their last session at a more difficult SNR level than where they began. A total of 70% of the treatment participants reached the most difficult SNR level (39 dB) in the game. Due to the difficulty of this SNR level (39 dB), it suggests that the visual cues were utilized during game play. Objective speech understanding measures The speech understanding in noise improved for the treatment group. Objective testing was measured using QuickSIN presented in two conditions: auditory-only and audiovisual. No significant improvements were found for the control group in either condition. The treatment group reached near significance in the audio-only condition. Previous
Figure 5. Mean SSQ-B item ratings at post-testing for the Speech-Hearing sub-scale. Positive scores represent perceived benefit in each listening condition. (SSQ-B - Speech, Spatial and Qualities of Hearing Benefit questionnaire). See Table I for definitions of the 14 listening situations.
364
K. Schwartz et al.
Table I. SSQ-B (Speech, Spatial and Qualities of Hearing Benefit) questionnaire wording for the Speech Hearing sub-scale (Jensen et al., 2009). Item
Wording
1.
You are talking with one other person and there is a TV on in the same room. Without turning the TV down, can you follow what the person you’re talking to says? You are talking with one other person in a quiet, carpeted lounge-room. Can you follow what the other person says? You are in a group of about five people, sitting round a table. It is an otherwise quiet place. You can see everyone else in the group. Can you follow the conversation? You are in a group of about five people in a busy restaurant. You can see everyone else in the group. Can you follow the conversation? You are talking with one other person. There is continuous background noise, such as a fan or running water. Can you follow what the person says? You are in a group of five people in a busy restaurant. You CANNOT see everyone else in the group. Can you follow the conversation? You are talking to someone in a place where there are a lot echoes, such as a church or railway terminus building. Can you follow what the other person says? Can you have a conversation with someone when another person is speaking whose voice is the same pitch as the person you’re talking to? Can you have a conversation with someone when another person is speaking whose voice is different in pitch from the person you’re talking to? You are listening to someone talking to you, while at the same time trying to follow the news on TV. Can you follow what both people are saying? You are in conversation with one person in a room where there are many other people talking. Can you follow what the person you are talking to is saying? You are with a group and the conversation switches from one person to another. Can you easily follow the conversation without missing the start of what each new speaker is saying? Can you easily have a conversation on the telephone? You are listening to someone on the telephone and someone next to you starts talking. Can you follow what’s being said by both speakers?
2. 3. 4. 5. 6.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
7. 8. 9. 10. 11. 12. 13. 14.
studies have found significant improvements for QuickSIN auditory-only tests following computerized auditory training programs (Sabes & Sweetow, 2007; Song et al., 2012). The near-significant findings for the treatment group in the auditory-only condition are conservative given that participants did not have any impairment. Their ability to benefit from this training is much more limited than that of impaired peers. Therefore, the near significant finding is important as it demonstrates that the Trivia Game training improves speech understanding-in-noise. Table II. SSQ-B (Speech, Spatial and Qualities of Hearing Benefit questionnaire) mean for the Speech-Hearing sub-scale reported by training and control group participants. Control group (n 14) Item No.
Mean
SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14
0.08 0.75 0.75 0.15 0.15 0.69 0.23 0.45 0.69 0.04 0.08 0.54 0.75 0.15
1.19 1.65 1.37 0.69 0.38 1.70 0.73 1.40 1.25 1.33 1.44 1.13 1.52 1.21
Training group (n 12) Mean 1.42 1.05 1.21 1.67 1.83 0.71 1.00 1.21 1.38 1.00 1.64 0.93 1.21 0.66
*The indicates significance, where p0.05.
The treatment group demonstrated significant improvements in the audiovisual condition following the training sessions. A common auditory rehabilitation strategy when listening in difficult situations involves focusing on a speaker’s face. Trivia Game provides practice for this strategy. A unique feature of Trivia Game, that sets it apart from other auditory training software programs, is the inclusion of a visual stimulus presented simultaneously with the speech-in-noise stimulus. Given that Trivia Game focuses on practicing the integration of audiovisual cues, the significant findings for the treatment group on the audiovisual QuickSIN suggest the effectiveness of Trivia Game and could also reflect an improvement in face-reading abilities. Subjective speech understanding measures
SD
p-value
1.17 1.63 1.31 1.30 1.40 0.96 1.13 1.37 1.52 0.95 0.67 1.12 1.52 1.01
0.009* 0.656 0.406 0.000* 0.000* 0.020* 0.052 0.183 0.232 0.036* 0.003* 0.408 0.465 0.090
In addition to changes on the objective measure of speech understanding in noise, significant improvement in perceived communication function in the treatment group when compared to the control group were also documented. Because the treatment group consisted of unimpaired listeners, improvements in subjective performance were not anticipated. Even though both groups consisted of unimpaired listeners, at pre-testing participants still reported difficulties in certain situations such as talking with a person in the presence of continuous background noise as reported by the SSQ. These findings are similar to those reported by Demeester, Topsakal, Hendrickx, Fransen, van Laer, Van Camp, et al. (2012). Following 12 sessions of game play,
Stuttering in school children in Egypt
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
treatment participants noted significant improvements in their communication function in daily life as measured by the SSQ-B. The most substantial improvements were in the following situations: having a conversation in a group in the presence of background noise where the speakers were visible, speaking with one person in the presence of background noise and following a single conversation when in the midst of a group of conversations. All of these situations involve speech understanding in a variety of background noise sources. Trivia Game provided direct practice with this skill. The improvements documented by the SSQ-B may generalize to improvement in listening in real-world situations. Limitations There were several limitations with this study. One limitation was the number of participants who failed to complete the study. Of the 43 participants who started the study, 16 dropped out before study completion. This finding was similar to drop-out reports in previous studies using computerized auditory training programs (Sabes & Sweetow, 2007). The majority of drop-out participants reported class demands and scheduling conflicts as reasons for their inability to complete the training sessions. Although not reported, it is also feasible that these healthy listeners were not motivated to complete the game, given their lack of impairment. Another limitation was that the participants were normal hearing, college-age students. These results may not replicate to individuals who have auditory processing impairments. Educated young adults do not represent the typical population and results may not generalize to older, less educated populations with acquired sensory, cognitive and linguistic impairments. Training was conducted by the participants at home. While the participants reported that they completed the protocol given to them, it is not known if they actually completed it as specified. For instance, it is not known whether they adjusted volume within a session, played for the length of time specified, on the days reported or the categories prescribed. Players who follow the protocol as specified may see greater benefits from the game. In this study design, the control group did not complete a listening activity in the 4-week period between pre- and post-testing. Future studies should implement an activity in both participant groups to strengthen the experimental design. The goal of Trivia Game is to be engaging; however, several players anecdotally reported frustration with the game. As participants reached more difficult SNR levels in game play, some reported frustration at their inability to progress beyond the difficult levels. Formal engagement and usability of the game were not measured in this study, but are recommended in future research.
365
Future directions Our study supports the use of cognitive-based auditory training programs to improve speech understanding in everyday listening environments for normal hearing young adults. The results of this study provide evidence that Trivia Game improved speech understanding in noise. Speech understanding in noise is a communication difficulty experienced by a variety of populations. The improvement of speech understanding in noise through computerized auditory training programs may have broad implications for many populations such as people who have hearing loss, excessive noise exposure, traumatic brain injury (Cornis-Pop et al., 2012), blastrelated injuries (Gallun, Lewis, Folmer, Diedesch, Kubli, McDermott, et al., 2012) and aphasia. Studies should be conducted to test the efficacy of Trivia Game with these populations. This study tested 12 trials within a 4-week period. It is not clear whether the number of trials and the number of weeks are the most effective combination to ensure maximum benefits and should be addressed in future work. Future studies should also be conducted to test the long-term benefits of the Trivia Game several months after completion of training. Conclusions Understanding speech in noise is critical to daily communication. Difficulties communicating in the presence of background noise are experienced by a variety of populations. A tool to address this pervasive difficulty would be a valuable addition to quality-of-life. Trivia Game has been shown to benefit the understanding of speech in noise and perceived communication function in an unimpaired sample. Future studies should assess the benefits of Trivia Game in impaired populations. Declaration of interest: The researchers are collaborators in the development of Trivia Game and have a financial conflict of interest with A2T2, VRR, Inc. located in Clermont, FL. This research was funded by a grant from the Office of the Secretary of Defense with a subproject to Old Dominion University (OSD-0082-H02-3163). References Assmann, P., & Summerfield, Q. (2004). The perception of speech under adverse conditions. New York: Springer. Bamiou, D. E., Campbell, N., & Sirimanna, T. (2006). Management of auditory processing disorders. Audiological Medicine, 4, 46–56. Basili, A. G., Diggs, C., & Rao, P. (1980). Auditory processing of brain-damaged adults under competitive listening conditions. Brain & Language, 9, 362–371. Belanger, H. G., Kretzmer, T., Yoash-Gantz, R., Pickett, T., & Tupler, L. A. (2009). Cognitive sequelae of blast-related versus other mechanisms of brain trauma. Journal of the International Neuropsychological Society, 15, 1–8.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
366
K. Schwartz et al.
Boothroyd, A. (2008). CasperSent: a program for computerassisted speech perception testing and training at the sentence level. Journal of the Academy of Rehabilitative Audiology, 41, 31–52. Chisolm, T. H., Abrams, H. B., & McArdle, R. (2004). Short- and long-term outcomes of adult audiological rehabilitation. Ear & Hearing, 25, 464–477. Cognitive Concepts. (1997). Earobics: Auditory development and phonics program [Computer software]. Cambridge, MA: Cognitive Concepts. Cornis-Pop, M., Mashima, P. A., Rothi, C. R., MacLennan, D. L., Picon, L. M., Hammond, C. S., et al. (2012). Cognitivecommunication rehabilitation for combat-related mild traumatic brain injury. Journal of Rehabilitation Research and Development, 49, xi–xxxii. Davis, G. (2007). Aphasiology: Disorders and Clinical Practice (2nd ed.). Boston, MA: Pearson. Demeester, K., Topsakal, V., Hendrickx, J.–J., Fransen, E., van Laer, L., Van Camp, G., et al. (2012). Hearing disability measured by the Speech, Spatial, and Qualities of Hearing Scale in clinically normal-hearing and hearing-impaired middle-aged persons, and disability screening by means of a reduced SSQ (the SSQ5). Ear & Hearing, 33, 615–626. Dimitrov, D. M., & Rumrill, P. D. (2003). Pretest-posttest designs and measurement of change. Work: Journal of Prevention, Assessment and Rehabilitation, 20, 159–165. Fausti, S. A., Wilmington, D. J., Gallun, F. J., Myers, P. J., & Henry, J. A. (2009). Auditory and vestibular dysfunction associated with blast-related traumatic brain injury. Journal of Rehabilitation Research & Development, 46, 797–809. Gallun, F. J., Lewis, M. S., Folmer, R. L., Diedesch, A. C., Kubli, L. R., McDermott, D. J., et al. (2012). Implications of blast exposure for central auditory function: A review. Journal of Rehabilitation Research and Development, 49, 1059–1074. Gatehouse, S., & Noble, W. (2004). The speech, spatial and qualities of hearing scale (SSQ). International Journal of Audiology, 43, 85–99. Grant, K. W., & Walden, T. C. (2013). Understanding excessive SNR in hearing-impaired listeners. Journal of the American Academy of Audiology, 24, 258–273. Hannaford, P. C., Simpson, J. A., Bisset, A. F., Davis, A., McKerrow, W., & Mills, R. (2005). The prevalence of ear, nose and throat problems in the community: Results from a national cross-sectional postal survey in Scotland. Family Practice Advance Access, 22, 227–233. Hawkins, D. (2005). Effectiveness of counselling-based adults group aural rehabilitation programs: A systematic review of the evidence. Journal of the American Academy of Audiology, 16, 485–493. Healy, E. W., Moser, D. C., Morrow-Odom, K. L., Hall, D. A., & Fridriksson, J. (2007). Speech perception in MRI scanner noise by persons with aphasia. Journal of Speech, Language, Hearing Research, 50, 323–334. Ingvalson, E. M., Lee, B., Fiebig, P., & Wong, P. C. M. (2013). The effects of short-term computerized speech-in-noise training on postlingually deafened adult cochlear implant recipients. Journal of Speech, Language, and Hearing Research, 56, 81–88. Jensen, N., Akeroyd, M., Noble, W., Naylor, G. (2009). The Speech, Spatial and Qualities of hearing scale (SSQ) as a benefit measure. NCRAR Conference on “The Ear-Brain System: Approaches to the Study and Treatment of Hearing Loss”, Portland. Held on October 2009. Keith, R. W. (2009). SCAN-3 for Adolescents and Adults: Tests for Auditory Processing Disorders. San Antonio, TX: Pearson. Kiili, K. (2005). Digital game-based learning: Towards an experiential gaming model. Internet and Higher Education, 8, 13–24. Killion, M., Niquette, P., Gudmundsend, G., Revit, L., & Banerjee, S. (2004). Development of a quick speech in noise
test for measuring signal to noise ratio loss in normal-hearing and hearing-impaired listeners. Journal of the Acoustical Society of America, 116, 2395–2405. Kittredge, A., Davis, L., & Blumstein, S. E. (2006). Effects of nonlinguistic auditory variations on lexical processing in Broca’s aphasia. Brain & Language, 97, 25–40. Kochkin, S. (2005). MarkeTrak VII: Customer satisfaction with hearing instruments in the digital age. The Hearing Journal, 58, 30–39. Levitt, H., Oden, C., Simon, H., Noack, C., & Lotze, A. (2011). Entertainment overcomes barriers of auditory training. The Hearing Journal, 64, 40–42. Lew, H. L., Garvert, D. W., Pogoda, T. K., Hsu, P. T., Devine, J. M., White, D. K., et al. (2009). Auditory and visual impairments in patients with blast-related traumatic brain injury: Effect of dual sensory impairment on Functional Independence Measure. Journal of Rehabilitation Research & Development, 46, 819–826. Musiek, F. I., Baran, J. A., & Shinn, J. (2004). Assessment and remediation of an auditory processing disorder associated with head trauma. Journal of the American Academy of Audiology, 15, 117–132. Neff , D., & Green, D. (1987). Masking produced by spectral uncertainty with multicomponent maskers. Perception of Psychophysics, 41, 409–415. Oleksiak, M., Smith, B. M., St. Andre, J. R., Caughlan, C. M., & Steiner, M. (2012). Audiological issues and hearing loss among veterans with mild traumatic brain injury. Journal of Rehabilitation Research and Development, 49, 995–1004. Olson, A. D., Preminger, J. E., & Shinn, J. B. (2013). The effect of LACE DVD training in new and experienced hearing aid users. Journal of the American Academy of Audiology, 24, 214–230. Pallarito, K. (2011). Retraining the brain when hearing aids aren’t enough. The Hearing Journal, 64, 25–34. Sabes, J. H., & Sweetow, R. W. (2007). Variables predicting outcomes on listening and communication enhancement (LACE) training. International Journal of Audiology, 46, 374–383. Schow, R. L., & Nerbonne, M. A. (2013). Introduction to Audiologic Rehabilitation (6th ed.). Boston, MA: Pearson Education. Saunders, G. H., & Echt, K. V. (2012). Blast exposure and dual sensory impairment: An evidence review and integrated rehabilitation approach. Journal of Rehabilitation Research and Development, 49, 1043–1058. Scientific Learning Corporation. (1998). Fast ForWord [Computer software]. Berkeley, CA: Scientific Learning Corporation. Song, J. H., Skoe, E., Banai, K., & Kraus, N. (2012). Training to improve hearing speech in noise: Biological mechanisms. Cerebral Cortex, 22, 1180–1190. Sweetow, R. W., & Palmer, C. V. (2005). Efficacy of individual auditory training in adults: A systematic review of the evidence. Journal of the American Academy of Audiology, 16, 494–504. Sweetow, R. W., & Sabes, J. H. (2004). The case for LACE: Listening and auditory communication enhancement training. Hearing Journal, 57, 32–40. Sweetow, R. W., & Sabes, J. H. (2006). The need for and development of an adaptive listening and communication enhancement (LACE) program. Journal of the American Academy of Audiology, 17, 538–558. Sweetow, R. W., & Sabes, J. H. (2010). Auditory training and challenges associated with participation and compliance. Journal of the American Academy of Audiology, 21, 586–593. Takahashi, G., Martinez, C. D., Beamer, S., Bridges, J., Noffsinger, D., Sugiura, K., et al. (2007). Subjective measures of hearing aid benefit and satisfaction in the NIDCD/VA follow up study. Journal of the American Academy of Audiology, 18, 323–349.
ISSN: 1754-9507 (Print) 1754-9515 (Online) Journal homepage: http://www.tandfonline.com/loi/iasl20
Development of Trivia Game for speech understanding in background noise Kathryn Schwartz, Stacie I. Ringleb, Hilary Sandberg, Anastasia Raymer & Ginger S. Watson To cite this article: Kathryn Schwartz, Stacie I. Ringleb, Hilary Sandberg, Anastasia Raymer & Ginger S. Watson (2015) Development of Trivia Game for speech understanding in background noise, International Journal of Speech-Language Pathology, 17:4, 357-366, DOI: 10.3109/17549507.2014.979875 To link to this article: http://dx.doi.org/10.3109/17549507.2014.979875
Published online: 24 Nov 2014.
Submit your article to this journal
Article views: 60
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=iasl20 Download by: [University of California Santa Barbara]
Date: 05 November 2015, At: 17:17
International Journal of Speech-Language Pathology, 2015; 17(4): 357–366
Development of Trivia Game for speech understanding in background noise
KATHRYN SCHWARTZ1, STACIE I. RINGLEB2, HILARY SANDBERG1, ANASTASIA RAYMER1 & GINGER S. WATSON3 Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
1Communication
Disorders & Special Education, 2Mechanical & Aerospace Engineering, 3Instructional Design & Technology, Old Dominion University, Norfolk,VA, USA
Abstract Purpose: Listening in noise is an everyday activity and poses a challenge for many people. To improve the ability to understand speech in noise, a computerized auditory rehabilitation game was developed. In Trivia Game players are challenged to answer trivia questions spoken aloud. As players progress through the game, the level of background noise increases. A study using Trivia Game was conducted as a proof-of-concept investigation in healthy participants. Method: College students with normal hearing were randomly assigned to a control (n 13) or a treatment (n 14) group. Treatment participants played Trivia Game 12 times over a 4-week period. All participants completed objective (auditoryonly and audiovisual formats) and subjective listening in noise measures at baseline and 4 weeks later. Result: There were no statistical differences between the groups at baseline. At post-test, the treatment group significantly improved their overall speech understanding in noise in the audiovisual condition and reported significant benefits in their functional listening abilities. Conclusion: Playing Trivia Game improved speech understanding in noise in healthy listeners. Significant findings for the audiovisual condition suggest that participants improved face-reading abilities. Trivia Game may be a platform for investigating changes in speech understanding in individuals with sensory, linguistic and cognitive impairments.
Keywords: Auditory processing, auditory comprehension, adults
Introduction Successful understanding of speech is affected by a number of internal and external factors in listening situations (Sweetow & Palmer, 2005). Intensity of the speech signal, complexity of the incoming message, rate of speech input, integrity of the incoming speech signal (dialect, articulatory precision) and acoustic environment are external influences on the ability of a listener to process an incoming signal. Likewise, factors internal to the listener, such as sensory, linguistic and cognitive abilities, also impact accurate and efficient processing of speech signals. While all of these factors can influence and challenge speech processing even in healthy listeners, difficulties can be especially exacerbated for listeners with disabilities affecting sensory, linguistic and other cognitive abilities. One of the most over-riding complaints heard from healthy and impaired listeners is difficulty when listening in environments with background noise, such as a reverberant setting or competing speakers (Assmann & Summerfield, 2004; Neff & Green, 1987).
Individuals with neurologic damage, such as stroke or brain injury, often experience difficulties understanding speech (Davis, 2007). Individuals with aphasia have impairments of linguistic comprehension and have significantly more difficulty than healthy controls when processing speech in the presence of background noise (Basili, Diggs, & Rao, 1980; Healy, Moser, Morrow-Odom, Hall, & Fridriksson, 2007; Kittredge, Davis, & Blumstein, 2006). Mild forms of traumatic brain injury and blast exposure can disrupt auditory and visual sensory processes (Oleksiak, Smith, St. Andre, Caughlan, & Steiner, 2012; Saunders & Echt, 2012) as well as cognitive processes (e.g., attention, memory) critical for speech processing (Belanger, Kretzmer, YoashGantz, Pickett, & Tupler, 2009; Cornis-Pop, Mashima, Rothi, MacLennan, Picon, Hammond, et al., 2012). Recent evidence shows that hearing loss is also common in populations with these disorders (Lew, Garvert, Pogoda, Hsu, Devine, White, et al., 2009). Moreover, individuals with mild forms of traumatic brain injury commonly voice complaints
Correspondence: Dr Stacie I. Ringleb, Old Dominion University, Mechanical & Aerospace Engineering, Kaufman Hall 238C, Norfolk, VA 23529, USA. E-mail: [email protected] ISSN 1754-9507 print/ISSN 1754-9515 online © 2014 The Speech Pathology Association of Australia Limited Published by Informa UK, Ltd. DOI: 10.3109/17549507.2014.979875
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
358
K. Schwartz et al.
of exacerbated auditory processing problems in the presence of background noise, although this finding has not been systematically examined (Fausti, Wilmington, Gallun, Myers, & Henry, 2009; Musiek, Baran, & Shinn, 2004). Individuals with certain developmental conditions also can have inordinate difficulty when listening to speech in noise, such as children with central auditory processing disorders (Bamiou, Campbell, & Sirimanna, 2006). Individuals with hearing loss complain frequently of increased difficulties when listening in noise (e.g., Hannaford, Simpson, Bisset, Davis, McKerrow, & Mills, 2005). The primary means to address hearing loss is the use of hearing aids. Individuals with hearing loss fitted with hearing aids may still be distressed by speech understanding difficulties in noisy situations (Kochkin, 2005; Takahashi, Martinez, Beamer, Bridges, Noffsinger, Sugiura, et al., 2007), thereby leading to reluctance to participate in noisy life situations or discontinued use of the hearing aids altogether. Clearly, speech processing in individuals with a variety of developmental and acquired sensory, cognitive and linguistic impairments is undermined by background noise. Even though these conditions have different underlying causes for decreased speech understanding in noise, all experience difficulties processing speech in noise at either the peripheral or central processing level (Grant & Walden, 2013). Auditory training is a common way to address these difficulties. Auditory training teaches individuals to maximize all available auditory and visual cues to improve speech understanding (Schow & Nerbonne, 2013). Computerized auditory training: Listening in noise Given the numerous groups that experience difficulty listening in noise, several methods of auditory training have been developed to address these issues. Sweetow and Palmer (2005) refer to some of these training approaches as analytic, as they incorporate tasks that target bottom-up processing of acoustic properties of speech signals, often at the syllable or single word level. Other components of auditory training centre on synthetic or top-down processing, such as teaching compensatory strategies to enhance auditory processing (e.g., reducing background noise, increasing reliance on visual input) (Chisolm, Abrams, & McArdle, 2004). Auditory training that incorporates both analytic and synthetic tasks have been implemented with some success in a group setting (Hawkins, 2005). However, application of auditory training in individualized sessions, while providing some positive outcomes, can be quite costly and may not be reimbursed in the current healthcare setting (Sweetow & Palmer, 2005). Recently, several computer-administered programs have been developed to provide direct, individualized auditory intervention for individuals with
hearing problems (Pallarito, 2011; Sweetow & Sabes, 2006). Computerized programs emphasize individualized auditory training that is cost-effective, fun and rewarding, accessible, verifiable and with necessary feedback (Sweetow & Palmer, 2005). Several programs are available that focus on various elements of auditory processing: Computer-Assisted Speech Perception Training-CASPER-Sent (Boothroyd, 2008), Earobics (Cognitive Concepts, 1997), FastForWord (Scientific Learning Corporation, 1998), Listening and Communication Enhancement program (LACE) (Sweetow & Sabes, 2004), ReadMyQuips (Levitt, Oden, Simon, Noack, & Lotze, 2011), and Seeing and Hearing Speech (Ingvalson, Lee, Fiebig, & Wong, 2013). Improvements in speech understanding in noise have been demonstrated utilizing computerized auditory training programs (Olson, Preminger, & Shinn, 2013; Song, Skoe, Banai, & Kraus, 2012; Sweetow & Sabes, 2004, 2006). Although benefits are evident for existing computerized auditory training tools, the programs are not without their limitations. Few existing tools provide visual face information along with listening in noise tasks, a critical factor to enhance top-down speech processing (Chisholm et al., 2004; Sweetow & Palmer, 2005). Further, the engagement and interest level for the computerized training tasks is somewhat limited, thereby reducing compliance (Sweetow & Sabes, 2010). Given these shortcomings, Trivia Game, a computerized auditory training program based on more interactive gaming technology, was developed. Development of trivia game Trivia Game is played independently where players view a man on the computer screen who speaks a trivia question presented in the presence of multitalker babble. The multi-talker babble is looped with no silent intervals. The purpose of the game is to hear the trivia questions in the presence of background noise and answer them correctly. Trivia questions are divided into 12 categories representing familiar topics such as sports, music and foods, with 40–50 questions per category, allowing listeners to use a unique category each playing session. Players receive points for correctly and quickly answering the question by selecting the written answer on the screen. The primary score is the percentage correct and is based on responses to the original trivia question. A series of hints are provided to guide the listeners who did not answer correctly on the first attempt. Hints are provided to keep the player engaged even if they have little knowledge of the trivia category. Hints provide bonus points and are not included in the percentage correct for the purposes of adjusting the background noise levels. Figure 1 illustrates a sample screen from Trivia Game.
Stuttering in school children in Egypt
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Figure 1. Trivia Game screen shot. Male speaker is verbally saying the phrase, “What is the fastest land animal?”
A total of 23 signal-to-noise ratio (SNR) levels are included in the game. The SNR levels range from 27 dB (level 1) to SNR of 39 dB (level 23). SNR is defined as the difference between the sound level of the spoken dialog and the competing noise, where the lower the SNR, the more difficult it is to hear the speaker. An adaptive algorithm was used in which the background noise systematically increases (becomes more difficult) with success in the game (defined as at least 80% correct across five successive questions) or reduces with difficulty (defined as less than 40% correct across five consecutive questions). The game starts at an SNR of 27 dB. When 80% is achieved the SNR level decreases by 3 dB (making speech understanding more difficult) and continues to decrease in 3 dB intervals as long as the 80% is sustained. If the percentage correct is between 40– 80%, the SNR remains unchanged. If the percentage correct drops to 40% or less, the SNR level increases by 3 dB (making speech understanding easier). This adaptive branching is a form of levelling used to deliver more challenge to the player who performs well or less challenge to the player who performs poorly. Providing the appropriate amount of challenge to match performance is a key factor in players’ willingness to play a game repeatedly across long periods of time (Kiili, 2005). To account for player knowledge of a particular trivia topic, a final screen is displayed that asks if the player knew the answer to the trivia question prior to the hints. If the player answers “yes” indicating they knew the answer, the percentage correct will change accordingly. If the player answers “no” the percentage correct remains unchanged. This ensures that the difficulty of the game adjusts based on what the player heard and not just the knowledge of a trivia category. If a player knew the answer to the trivia question but answered incorrectly due to the difficult SNR level (they did not hear the question), the percentage correct would decrease. If the player knew the answer to the trivia question and answered correctly, it is assumed that they heard the question and, therefore, the percentage correct increased.
359
Individuals can experience difficulty listening in noise for a variety of sensory, cognitive or linguistic difficulties. Thus, they may require varied auditory rehabilitation treatments to address those diverse disturbances. Because the game is conducted using specific semantic categories, players can constrain their expectations of words to be processed while listening to the trivia in noise. This factor could be especially beneficial to individuals with language impairments such as aphasia. Further, the visual information provided by reading the waiter’s face as he presents trivia questions and clues provides support to the listener as the signal degrades with increased background noise. Thus, the game has the potential to enhance top-down processing for individuals with a variety of sensory, cognitive or language disorders that undermine bottom-up processing. Trivia Game was developed as a potential tool for use by a variety of populations that experience auditory difficulties in noise. Prior to implementing the game in one of those populations, a proofof-concept study was conducted to determine if Trivia Game was effective in improving speech understanding in an unimpaired population. The resulting study examined the effects of Trivia Game in healthy, college-age listeners compared to untrained controls. The specific research questions for this study were: (1) Does the treatment group perform better than the control group on objective listening-in-noise measures? (2) Is there a difference in performance gain for the treatment group in an auditory-only vs audiovisual condition?
Methods Participants Forty-three college students were recruited for this proof-of-concept study. Participants were randomly assigned to a treatment or control group using a random number generator that determined group membership prior to the study. Of the 27 participants who completed all phases of the experiment, 13 were in the treatment group (six males, seven females) with a mean age of 21.6 years (SD 2.1) and 14 participants served as controls (seven males, seven females) with a mean age of 21.4 years (SD 2.3). Of those who did not complete the study, six were originally assigned to the control group and 10 to the treatment group. Participants who failed to complete the study reported scheduling conflicts, lack of interest and technical difficulties running the game as the reasons for withdrawal. All participants had hearing thresholds 25 dB HL for pure-tone frequencies from 250–8000 Hz, bilaterally. In addition, all participants were native speakers of American English and passed an auditory processing screening as determined by the SCAN-3 (Keith,
360
K. Schwartz et al.
2009). The researchers recruited participants through fliers placed around Old Dominion University’s campus and through announcements made in classes. The study was approved by the Old Dominion University Institutional Review Board (IRB). All participants provided written informed consent to participate in this study.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Test measures Quick Speech-In-Noise (Quick SIN). The QuickSIN (Killion, Niquette, Gudmundsend, Revit, & Banerjee, 2004) test was used as an objective measure of the amount of speech that is understood in the presence of background noise. The test presents sentences to a listener who verbally repeats the sentence. Scores are calculated on the accuracy of each word repeated. QuickSIN is composed of 18 lists with six sentences per list. Each sentence contains five key words. The sentences are spoken by a female talker in the presence of four-talker babble (one male and three females). In each list, the sentences are presented at progressively lower signal-to-noise ratios (SNR). The first sentence is presented at a SNR of 25 dB and subsequent sentences in the list decrease SNR by 5 dB decrements, with the last sentence in the list being presented at 0 dB SNR. QuickSIN was administered in the standard auditory-only format and in an experimental audiovisual format in which the female speaker was both heard and viewed on a computer screen. This version was provided by the Killion group. Synchronization of the auditory and visual signals was verified daily. The audiovisual presentation of QuickSIN was used as a secondary measure of audiovisual benefit. The primary findings of this study are based upon the results of the standard QuickSIN and the subjective rating scales. Speech, Spatial and Qualities of Hearing Scale (SSQ). In addition to the objective measurement of listening in noise, all participants completed the SSQ version 5.6 (Gatehouse & Noble, 2004) during the initial test session to ensure that subjective auditory experiences were similar for the treatment and control groups. This questionnaire was chosen to reflect how well participants function in the real world when communicating in difficult listening situations. The SSQ is a 50-item subjective questionnaire with three sub-scales which evaluates a variety of listening conditions encountered in daily life: speech hearing, spatial hearing and qualities of hearing. Each question describes a communication scenario and asks a question relative to a participant’s performance ability in that listening situation. For example, question six states “You are in a group of about five people in a busy restaurant. You CANNOT see everyone else in the group. Can you follow the conversation?” Participants rate their ability on a 10-point visual rating
scale ranging from 0–10 to indicate the ease of hearing in the scenarios posed by each item. Zero indicates “not at all” and 10 indicates “perfectly”. Therefore, higher scores indicate greater ease of listening in everyday scenarios. Participants were instructed to answer the questions based on their everyday experiences. Each question included a “not applicable” option that the participant could select if the situation did not apply. Responses that were marked “not applicable” were not scored. The SSQ-B (B for Benefit) version 5.6 (Jensen, Akeroyd, Noble, & Naylor, 2009) was completed by all participants at the completion of the study. This instrument was administered in order to measure the subjective impact of game-playing on the treatment participant’s ability to understand speech in difficult listening situations. The SSQ-B measures benefit of aided hearing aid conditions when compared to unaided conditions. The instrument consists of all items on the SSQ with identical instructions and scoring criteria, but uses an absolute rating scale that compares aided experiences to those without hearing aids. The SSQ-B is scored on a relative scale ranged from 5 to 5, where 5 is “much worse”, 0 is “unchanged” and 5 is “much better”. The SSQ-B wording was modified by replacing “comparing your ability now with your ability before getting your hearing aid/s” with “comparing your ability now with your ability before study participation”.
Procedures All participants completed a pre-treatment laboratory test battery of pure tone testing, auditory processing screening, QuickSIN and SSQ questionnaire. Four weeks after initial laboratory testing, all participants returned for post-testing with the QuickSIN and the SSQ-B. The QuickSIN, SSQ and SSQ-B were scored by two independent assessors. These assessors were not involved in the development of Trivia Game and did not have any financial conflict of interest. All participants were compensated for their time at the end of the study; control participants received $20 and treatment participants were compensated $100.
Outcome measures For the QuickSIN test, participants were tested bilaterally in two conditions: auditory-only and audiovisual. For the auditory-only condition, the QuickSIN stimulus was played at 70 dB HL on a CD player routed through an audiometer (GSI-61) to TDH-50 headphones. For the auditory-visual condition, the same auditory set-up was used with the addition of a computer monitor. The computer monitor displayed an image of a female speaking the stimulus sentences as the participant listened through the
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Stuttering in school children in Egypt
headphones. A practice list of six sentences was given to each participant to acquaint them with the task for both the auditory-only and audiovisual conditions. All testing was administered in a double-walled sound booth. Each participant was given 16 lists during each testing session (eight lists in the auditory-only condition and eight lists in the audiovisual condition). Listening conditions and list order were counterbalanced and randomized across participants. This ensured that no lists were repeated in a single session and that participants did not hear the same list in the same listening condition for either session. QuickSIN was scored as the number of key words correctly repeated. In clinical settings, QuickSIN is scored in words correctly repeated, but the score is converted to a score of “SNR loss”. The SNR loss quantifies the extent to which the hearing aid user’s speech understanding in noise is worse than that of normalhearing listeners. Because the purpose of this test was to determine speech perception benefit in normal hearing listeners following Trivia Game participation, the SNR loss was not calculated for the current study. Two versions of the SSQ were administered (SSQ and SSQ-B). Participants completed the SSQ surveys using a paper and pencil format. During pre-testing, all participants completed the SSQ to characterize their impressions of listening experiences and difficulties in daily life. In post-testing, all participants completed the SSQ-B where they compared their listening experience after study participation to listening experiences prior to study participation. Because Trivia Game focuses on hearing speech in noise, only the Speech Hearing sub-scale of the SSQ and SSQ-B was scored.
361
Results Training logs Review of the training logs indicated that participants progressed through the game. Twelve of the 13 participants recorded their game level for every session and results are displayed in Figure 2. The remaining participant was inconsistent in recording and results are not displayed. The most difficult SNR in the game was 39 dB, which was reached by 58% of participants. Of the remaining participants, four reached an SNR level of 36 dB and one reached an SNR level of 12. QuickSIN The objective listening in noise measure used in the experiment was QuickSIN. The total number of words correct for each condition (auditory-only and audiovisual) for the pre-test and post-test sessions were calculated. A one-way ANOVA was conducted to compare the pre-test scores of the treatment and control groups for each QuickSIN condition. Following that, a one-way ANCOVA with pre-test scores as a covariate was conducted to assess the effects of the training. Use of a one-way ANCOVA was chosen to avoid errors associated with repeatedmeasures ANOVA for treatment and sessions (Dimitrov & Rumrill, 2003). Auditory-only QuickSIN. A one-way ANOVA was conducted to evaluate the equivalence of the treatment and control groups on the number of words correct on the QuickSIN pre-test score for the auditory-only condition. The test was not significant, F(1,25) 0.02, p 0.88, indicating that there was not a significant difference between the treatment (M 194.62, SD 9.88) and control
Trivia Game play Treatment group participants were introduced to Trivia Game during the pre-test laboratory session and were given verbal and written instructions on how to play the game. They were instructed to set the computer volume at a comfortable level at the beginning of each session and not to change the volume after the session had begun. Participants were further instructed to play in a quiet environment at the same time each day. They were loaned Sennheiser PC 151 headphones to use with their personal computer and were instructed to play the game four times a week over a 3-week period, for a total of 12 training sessions. They completed a playing log where they recorded their game level at the end of each session. Each session lasted ∼ 20 minutes, with participants playing one trivia category per session. Contacts were made on a weekly basis with participants to answer questions or troubleshoot equipment problems.
Figure 2. Participant SNR (signal-to-noise ratio) level at end of each training session. Each line represents an individual participant. Lower scores indicate more difficult listening situations.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
362
K. Schwartz et al.
(M 195.14, SD 8.26) groups on the pre-test score. The eta2 index indicated that less than 1% of the variance in the QuickSIN auditory-only pre-test score was accounted for by whether a participant was assigned to the treatment or control group. A one-way analysis of covariance (ANCOVA) was conducted. The independent variable, group, included two levels: treatment and control. The dependent variable was the gain in number of words correct auditory-only condition calculated between the post-test and pre-test sessions. The covariate was the number of words correct in the pre-test session. A preliminary analysis evaluating the homogeneity-of-slopes assumption indicated that the relationship between the pretest audio-only score and the gain in the audio-only score did not differ significantly as a function of the group, F(1,23) 0.02, MSE 26.92, p 0.89, partial eta2 0.001. The ANCOVA was not significant, F(1,24) 3.86, MSE 25.85, p 0.06. The gain in words correct for the audio-only condition adjusted for pre-test scores was nearly significant. The treatment group demonstrated higher gain (M 10.59), while the control group showed a lower gain (M 6.74), as illustrated in Figure 3. The strength of the relationship between groups and gain in words correct for the audio-only score was assessed by a partial eta2 with the group factor accounting for 14% of the variance in the gain in words correct in the audio-only condition, holding constant the number of words correct in the pre-test audio-only condition. Audiovisual QuickSIN. A one-way ANOVA was conducted to evaluate the equivalence of the treatment and control groups on the number of words correct on the QuickSIN pre-test score for the audiovisual condition. The test was not significant, F(1,25) 1.17, p 0.29, indicating that there was not a significant difference between the treatment (M 208.00, SD 9.39) and control (M 211.79, SD 8.77) groups on the number of words correct pre-test audiovisual score. The eta 2 index indicated that less than 5% of the variance in the QuickSIN audiovisual pre-test score was
Figure 3. Total QuickSIN Words Correct for the Auditory-Only condition. Error bars represent one standard deviation.
accounted for by whether a participant was assigned to the treatment or control group. A one-way analysis of covariance (ANCOVA) was conducted to assess the independent variable, group (treatment and control) on the dependent variable gain in number of words correct audiovisual condition calculated between the post-test and pre-test sessions. The covariate was the number of words correct in the pre-test session. A preliminary analysis evaluating the homogeneity-of-slopes assumption indicated that the relationship between the pre-test audiovisual score and the gain in the audiovisual score did not differ significantly as a function of the group, F(1,23) 1.73, MSE 36.43, p 0.20, partial eta2 0.07. The ANCOVA was significant, F(1,24) 8.15, MSE 37.54, p 0.01. The gain in words correct for the audiovisual condition adjusted for pre-test scores showed significantly higher gains for the treatment group (M 9.72) than the control group (M 2.83), as shown in Figure 4. The strength of the relationship between groups and gain in words correct for the audiovisual score was assessed by a partial eta2 with the group factor accounting for 25% of the variance in the gain in words correct in the audiovisual condition, holding constant the number of words correct in the pre-test audiovisual condition. SSQ The Speech and Hearing sub-scale of the SSQ questionnaire was administered at pre-test session (max score 140). The scores on the SSQ were compared for the two groups using an independent samples t-test. The analysis was completed on 24 participants as three participants (one control and two treatment) failed to complete the instrument. There were no significant differences between the treatment (M 109.97, SD 17.88) and control (M 106.07, SD 17.51) groups at pre-test, t 0.54, p 0.79. A Cronbach’s alpha internal reliability coefficient of 0.92 was calculated for these scores.
Figure 4. Total QuickSIN Words Correct for the Audiovisual condition. Error bars represent one standard deviation. * p 0.05 ** p 0.01 for Gain in Words Correct.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
Stuttering in school children in Egypt
The SSQ-Benefit sub-scale (SSQ-B) was administered post-test. The analysis was completed on 26 participants as one treatment participant failed to complete the instrument. One participant in the control group marked all items as “not applicable” and one in the treatment group marked four items as “not applicable”. These items were not included in scoring. Independent samples t-test were conducted to compare the mean benefit for the treatment and control groups. The treatment group (M 1.23, SD 0.90) reported significantly greater benefit than the control group (M 0.25, SD 0.53), t 3.37, p 0.003. A Cronbach’s alpha internal reliability coefficient of 0.89 was calculated for these scores. As presented in Figure 5, the treatment group reported benefits in all listening situations. Table I lists the exact wording for each listening situation. The perceived benefits for the treatment group were significantly greater than the control group in conditions where the signal was in competition with background noise: single-talker conversation with TV on (item 1), group conversation in a restaurant (item 4), individual conversation in the presence of background noise (item 5), group conversation in a restaurant but cannot see talker (item 6), following television news with single-talker simultaneously (item 10) and single-talker conversation with multiple conversations in the background such as a cocktail party (item 11). Table II summarizes the descriptive statistics for each item for each group. Discussion Understanding speech in the presence of background noise is problematic for a variety of populations, such as those with aphasia, presbycusis, traumatic brain
363
injury and developmental auditory processing problems (Belanger et al., 2009; Kochkin, 2005; Takahashi et al., 2007). Despite the use of aural rehabilitation and/or hearing aids to address sensory impairments, the difficulty understanding speech often persists. The current study served as proof of concept for the efficacy of Trivia Game, a computer program created to improve speech understanding in noise for a variety of populations. Results suggest that Trivia Game improves speech understanding in noise in an unimpaired population. Game performance The overall design of the game and implementation of the training program (12 sessions over 4 weeks) was adequate to observe improvement in both the subjective and objective measures of listening in noise. Furthermore, game logs indicate that all treatment participants improved their speech understanding during game play, ending their last session at a more difficult SNR level than where they began. A total of 70% of the treatment participants reached the most difficult SNR level (39 dB) in the game. Due to the difficulty of this SNR level (39 dB), it suggests that the visual cues were utilized during game play. Objective speech understanding measures The speech understanding in noise improved for the treatment group. Objective testing was measured using QuickSIN presented in two conditions: auditory-only and audiovisual. No significant improvements were found for the control group in either condition. The treatment group reached near significance in the audio-only condition. Previous
Figure 5. Mean SSQ-B item ratings at post-testing for the Speech-Hearing sub-scale. Positive scores represent perceived benefit in each listening condition. (SSQ-B - Speech, Spatial and Qualities of Hearing Benefit questionnaire). See Table I for definitions of the 14 listening situations.
364
K. Schwartz et al.
Table I. SSQ-B (Speech, Spatial and Qualities of Hearing Benefit) questionnaire wording for the Speech Hearing sub-scale (Jensen et al., 2009). Item
Wording
1.
You are talking with one other person and there is a TV on in the same room. Without turning the TV down, can you follow what the person you’re talking to says? You are talking with one other person in a quiet, carpeted lounge-room. Can you follow what the other person says? You are in a group of about five people, sitting round a table. It is an otherwise quiet place. You can see everyone else in the group. Can you follow the conversation? You are in a group of about five people in a busy restaurant. You can see everyone else in the group. Can you follow the conversation? You are talking with one other person. There is continuous background noise, such as a fan or running water. Can you follow what the person says? You are in a group of five people in a busy restaurant. You CANNOT see everyone else in the group. Can you follow the conversation? You are talking to someone in a place where there are a lot echoes, such as a church or railway terminus building. Can you follow what the other person says? Can you have a conversation with someone when another person is speaking whose voice is the same pitch as the person you’re talking to? Can you have a conversation with someone when another person is speaking whose voice is different in pitch from the person you’re talking to? You are listening to someone talking to you, while at the same time trying to follow the news on TV. Can you follow what both people are saying? You are in conversation with one person in a room where there are many other people talking. Can you follow what the person you are talking to is saying? You are with a group and the conversation switches from one person to another. Can you easily follow the conversation without missing the start of what each new speaker is saying? Can you easily have a conversation on the telephone? You are listening to someone on the telephone and someone next to you starts talking. Can you follow what’s being said by both speakers?
2. 3. 4. 5. 6.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
7. 8. 9. 10. 11. 12. 13. 14.
studies have found significant improvements for QuickSIN auditory-only tests following computerized auditory training programs (Sabes & Sweetow, 2007; Song et al., 2012). The near-significant findings for the treatment group in the auditory-only condition are conservative given that participants did not have any impairment. Their ability to benefit from this training is much more limited than that of impaired peers. Therefore, the near significant finding is important as it demonstrates that the Trivia Game training improves speech understanding-in-noise. Table II. SSQ-B (Speech, Spatial and Qualities of Hearing Benefit questionnaire) mean for the Speech-Hearing sub-scale reported by training and control group participants. Control group (n 14) Item No.
Mean
SD
1 2 3 4 5 6 7 8 9 10 11 12 13 14
0.08 0.75 0.75 0.15 0.15 0.69 0.23 0.45 0.69 0.04 0.08 0.54 0.75 0.15
1.19 1.65 1.37 0.69 0.38 1.70 0.73 1.40 1.25 1.33 1.44 1.13 1.52 1.21
Training group (n 12) Mean 1.42 1.05 1.21 1.67 1.83 0.71 1.00 1.21 1.38 1.00 1.64 0.93 1.21 0.66
*The indicates significance, where p0.05.
The treatment group demonstrated significant improvements in the audiovisual condition following the training sessions. A common auditory rehabilitation strategy when listening in difficult situations involves focusing on a speaker’s face. Trivia Game provides practice for this strategy. A unique feature of Trivia Game, that sets it apart from other auditory training software programs, is the inclusion of a visual stimulus presented simultaneously with the speech-in-noise stimulus. Given that Trivia Game focuses on practicing the integration of audiovisual cues, the significant findings for the treatment group on the audiovisual QuickSIN suggest the effectiveness of Trivia Game and could also reflect an improvement in face-reading abilities. Subjective speech understanding measures
SD
p-value
1.17 1.63 1.31 1.30 1.40 0.96 1.13 1.37 1.52 0.95 0.67 1.12 1.52 1.01
0.009* 0.656 0.406 0.000* 0.000* 0.020* 0.052 0.183 0.232 0.036* 0.003* 0.408 0.465 0.090
In addition to changes on the objective measure of speech understanding in noise, significant improvement in perceived communication function in the treatment group when compared to the control group were also documented. Because the treatment group consisted of unimpaired listeners, improvements in subjective performance were not anticipated. Even though both groups consisted of unimpaired listeners, at pre-testing participants still reported difficulties in certain situations such as talking with a person in the presence of continuous background noise as reported by the SSQ. These findings are similar to those reported by Demeester, Topsakal, Hendrickx, Fransen, van Laer, Van Camp, et al. (2012). Following 12 sessions of game play,
Stuttering in school children in Egypt
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
treatment participants noted significant improvements in their communication function in daily life as measured by the SSQ-B. The most substantial improvements were in the following situations: having a conversation in a group in the presence of background noise where the speakers were visible, speaking with one person in the presence of background noise and following a single conversation when in the midst of a group of conversations. All of these situations involve speech understanding in a variety of background noise sources. Trivia Game provided direct practice with this skill. The improvements documented by the SSQ-B may generalize to improvement in listening in real-world situations. Limitations There were several limitations with this study. One limitation was the number of participants who failed to complete the study. Of the 43 participants who started the study, 16 dropped out before study completion. This finding was similar to drop-out reports in previous studies using computerized auditory training programs (Sabes & Sweetow, 2007). The majority of drop-out participants reported class demands and scheduling conflicts as reasons for their inability to complete the training sessions. Although not reported, it is also feasible that these healthy listeners were not motivated to complete the game, given their lack of impairment. Another limitation was that the participants were normal hearing, college-age students. These results may not replicate to individuals who have auditory processing impairments. Educated young adults do not represent the typical population and results may not generalize to older, less educated populations with acquired sensory, cognitive and linguistic impairments. Training was conducted by the participants at home. While the participants reported that they completed the protocol given to them, it is not known if they actually completed it as specified. For instance, it is not known whether they adjusted volume within a session, played for the length of time specified, on the days reported or the categories prescribed. Players who follow the protocol as specified may see greater benefits from the game. In this study design, the control group did not complete a listening activity in the 4-week period between pre- and post-testing. Future studies should implement an activity in both participant groups to strengthen the experimental design. The goal of Trivia Game is to be engaging; however, several players anecdotally reported frustration with the game. As participants reached more difficult SNR levels in game play, some reported frustration at their inability to progress beyond the difficult levels. Formal engagement and usability of the game were not measured in this study, but are recommended in future research.
365
Future directions Our study supports the use of cognitive-based auditory training programs to improve speech understanding in everyday listening environments for normal hearing young adults. The results of this study provide evidence that Trivia Game improved speech understanding in noise. Speech understanding in noise is a communication difficulty experienced by a variety of populations. The improvement of speech understanding in noise through computerized auditory training programs may have broad implications for many populations such as people who have hearing loss, excessive noise exposure, traumatic brain injury (Cornis-Pop et al., 2012), blastrelated injuries (Gallun, Lewis, Folmer, Diedesch, Kubli, McDermott, et al., 2012) and aphasia. Studies should be conducted to test the efficacy of Trivia Game with these populations. This study tested 12 trials within a 4-week period. It is not clear whether the number of trials and the number of weeks are the most effective combination to ensure maximum benefits and should be addressed in future work. Future studies should also be conducted to test the long-term benefits of the Trivia Game several months after completion of training. Conclusions Understanding speech in noise is critical to daily communication. Difficulties communicating in the presence of background noise are experienced by a variety of populations. A tool to address this pervasive difficulty would be a valuable addition to quality-of-life. Trivia Game has been shown to benefit the understanding of speech in noise and perceived communication function in an unimpaired sample. Future studies should assess the benefits of Trivia Game in impaired populations. Declaration of interest: The researchers are collaborators in the development of Trivia Game and have a financial conflict of interest with A2T2, VRR, Inc. located in Clermont, FL. This research was funded by a grant from the Office of the Secretary of Defense with a subproject to Old Dominion University (OSD-0082-H02-3163). References Assmann, P., & Summerfield, Q. (2004). The perception of speech under adverse conditions. New York: Springer. Bamiou, D. E., Campbell, N., & Sirimanna, T. (2006). Management of auditory processing disorders. Audiological Medicine, 4, 46–56. Basili, A. G., Diggs, C., & Rao, P. (1980). Auditory processing of brain-damaged adults under competitive listening conditions. Brain & Language, 9, 362–371. Belanger, H. G., Kretzmer, T., Yoash-Gantz, R., Pickett, T., & Tupler, L. A. (2009). Cognitive sequelae of blast-related versus other mechanisms of brain trauma. Journal of the International Neuropsychological Society, 15, 1–8.
Downloaded by [University of California Santa Barbara] at 17:17 05 November 2015
366
K. Schwartz et al.
Boothroyd, A. (2008). CasperSent: a program for computerassisted speech perception testing and training at the sentence level. Journal of the Academy of Rehabilitative Audiology, 41, 31–52. Chisolm, T. H., Abrams, H. B., & McArdle, R. (2004). Short- and long-term outcomes of adult audiological rehabilitation. Ear & Hearing, 25, 464–477. Cognitive Concepts. (1997). Earobics: Auditory development and phonics program [Computer software]. Cambridge, MA: Cognitive Concepts. Cornis-Pop, M., Mashima, P. A., Rothi, C. R., MacLennan, D. L., Picon, L. M., Hammond, C. S., et al. (2012). Cognitivecommunication rehabilitation for combat-related mild traumatic brain injury. Journal of Rehabilitation Research and Development, 49, xi–xxxii. Davis, G. (2007). Aphasiology: Disorders and Clinical Practice (2nd ed.). Boston, MA: Pearson. Demeester, K., Topsakal, V., Hendrickx, J.–J., Fransen, E., van Laer, L., Van Camp, G., et al. (2012). Hearing disability measured by the Speech, Spatial, and Qualities of Hearing Scale in clinically normal-hearing and hearing-impaired middle-aged persons, and disability screening by means of a reduced SSQ (the SSQ5). Ear & Hearing, 33, 615–626. Dimitrov, D. M., & Rumrill, P. D. (2003). Pretest-posttest designs and measurement of change. Work: Journal of Prevention, Assessment and Rehabilitation, 20, 159–165. Fausti, S. A., Wilmington, D. J., Gallun, F. J., Myers, P. J., & Henry, J. A. (2009). Auditory and vestibular dysfunction associated with blast-related traumatic brain injury. Journal of Rehabilitation Research & Development, 46, 797–809. Gallun, F. J., Lewis, M. S., Folmer, R. L., Diedesch, A. C., Kubli, L. R., McDermott, D. J., et al. (2012). Implications of blast exposure for central auditory function: A review. Journal of Rehabilitation Research and Development, 49, 1059–1074. Gatehouse, S., & Noble, W. (2004). The speech, spatial and qualities of hearing scale (SSQ). International Journal of Audiology, 43, 85–99. Grant, K. W., & Walden, T. C. (2013). Understanding excessive SNR in hearing-impaired listeners. Journal of the American Academy of Audiology, 24, 258–273. Hannaford, P. C., Simpson, J. A., Bisset, A. F., Davis, A., McKerrow, W., & Mills, R. (2005). The prevalence of ear, nose and throat problems in the community: Results from a national cross-sectional postal survey in Scotland. Family Practice Advance Access, 22, 227–233. Hawkins, D. (2005). Effectiveness of counselling-based adults group aural rehabilitation programs: A systematic review of the evidence. Journal of the American Academy of Audiology, 16, 485–493. Healy, E. W., Moser, D. C., Morrow-Odom, K. L., Hall, D. A., & Fridriksson, J. (2007). Speech perception in MRI scanner noise by persons with aphasia. Journal of Speech, Language, Hearing Research, 50, 323–334. Ingvalson, E. M., Lee, B., Fiebig, P., & Wong, P. C. M. (2013). The effects of short-term computerized speech-in-noise training on postlingually deafened adult cochlear implant recipients. Journal of Speech, Language, and Hearing Research, 56, 81–88. Jensen, N., Akeroyd, M., Noble, W., Naylor, G. (2009). The Speech, Spatial and Qualities of hearing scale (SSQ) as a benefit measure. NCRAR Conference on “The Ear-Brain System: Approaches to the Study and Treatment of Hearing Loss”, Portland. Held on October 2009. Keith, R. W. (2009). SCAN-3 for Adolescents and Adults: Tests for Auditory Processing Disorders. San Antonio, TX: Pearson. Kiili, K. (2005). Digital game-based learning: Towards an experiential gaming model. Internet and Higher Education, 8, 13–24. Killion, M., Niquette, P., Gudmundsend, G., Revit, L., & Banerjee, S. (2004). Development of a quick speech in noise
test for measuring signal to noise ratio loss in normal-hearing and hearing-impaired listeners. Journal of the Acoustical Society of America, 116, 2395–2405. Kittredge, A., Davis, L., & Blumstein, S. E. (2006). Effects of nonlinguistic auditory variations on lexical processing in Broca’s aphasia. Brain & Language, 97, 25–40. Kochkin, S. (2005). MarkeTrak VII: Customer satisfaction with hearing instruments in the digital age. The Hearing Journal, 58, 30–39. Levitt, H., Oden, C., Simon, H., Noack, C., & Lotze, A. (2011). Entertainment overcomes barriers of auditory training. The Hearing Journal, 64, 40–42. Lew, H. L., Garvert, D. W., Pogoda, T. K., Hsu, P. T., Devine, J. M., White, D. K., et al. (2009). Auditory and visual impairments in patients with blast-related traumatic brain injury: Effect of dual sensory impairment on Functional Independence Measure. Journal of Rehabilitation Research & Development, 46, 819–826. Musiek, F. I., Baran, J. A., & Shinn, J. (2004). Assessment and remediation of an auditory processing disorder associated with head trauma. Journal of the American Academy of Audiology, 15, 117–132. Neff , D., & Green, D. (1987). Masking produced by spectral uncertainty with multicomponent maskers. Perception of Psychophysics, 41, 409–415. Oleksiak, M., Smith, B. M., St. Andre, J. R., Caughlan, C. M., & Steiner, M. (2012). Audiological issues and hearing loss among veterans with mild traumatic brain injury. Journal of Rehabilitation Research and Development, 49, 995–1004. Olson, A. D., Preminger, J. E., & Shinn, J. B. (2013). The effect of LACE DVD training in new and experienced hearing aid users. Journal of the American Academy of Audiology, 24, 214–230. Pallarito, K. (2011). Retraining the brain when hearing aids aren’t enough. The Hearing Journal, 64, 25–34. Sabes, J. H., & Sweetow, R. W. (2007). Variables predicting outcomes on listening and communication enhancement (LACE) training. International Journal of Audiology, 46, 374–383. Schow, R. L., & Nerbonne, M. A. (2013). Introduction to Audiologic Rehabilitation (6th ed.). Boston, MA: Pearson Education. Saunders, G. H., & Echt, K. V. (2012). Blast exposure and dual sensory impairment: An evidence review and integrated rehabilitation approach. Journal of Rehabilitation Research and Development, 49, 1043–1058. Scientific Learning Corporation. (1998). Fast ForWord [Computer software]. Berkeley, CA: Scientific Learning Corporation. Song, J. H., Skoe, E., Banai, K., & Kraus, N. (2012). Training to improve hearing speech in noise: Biological mechanisms. Cerebral Cortex, 22, 1180–1190. Sweetow, R. W., & Palmer, C. V. (2005). Efficacy of individual auditory training in adults: A systematic review of the evidence. Journal of the American Academy of Audiology, 16, 494–504. Sweetow, R. W., & Sabes, J. H. (2004). The case for LACE: Listening and auditory communication enhancement training. Hearing Journal, 57, 32–40. Sweetow, R. W., & Sabes, J. H. (2006). The need for and development of an adaptive listening and communication enhancement (LACE) program. Journal of the American Academy of Audiology, 17, 538–558. Sweetow, R. W., & Sabes, J. H. (2010). Auditory training and challenges associated with participation and compliance. Journal of the American Academy of Audiology, 21, 586–593. Takahashi, G., Martinez, C. D., Beamer, S., Bridges, J., Noffsinger, D., Sugiura, K., et al. (2007). Subjective measures of hearing aid benefit and satisfaction in the NIDCD/VA follow up study. Journal of the American Academy of Audiology, 18, 323–349.
