Different Weighting of Auditory Memory and Attention as Cognitive Resources in Bilingual’s Speech Perception in Noise

Abstract

Understanding speech in noisy environments is a challenging task that requires sensory and cognitive functions, including memory and auditory attention. Bilinguals and monolinguals have different scores of these abilities. This study aims to investigate the relationship between these cognitive skills and compare Turkish-Persian bilinguals with Persian monolinguals regarding speech-in-noise scores. 45 Turkish-Persian bilinguals and 45 Persian monolinguals, aged 18–25 (mean age 21.67), with normal hearing, participated in a speech in noise, auditory working memory, and auditory attention tests. The bilingual group performed significantly worse than the monolingual group in the quick speech-in-noise and n-back auditory working memory tests (p < 0.05). However, their score on the auditory attention test was better (p < 0.05). Bilinguals showed a significant correlation between auditory working memory, auditory attention, and speech perception in noise. A linear regression validated the proposed model, predicting speech perception scores in noise based on auditory attention and auditory working memory in bilinguals (r2: 0.20, P: 0.008). Research suggests that various cognitive factors impact an individual’s ability to perceive speech in noisy environments. Specifically, auditory attention and working memory have different levels of influence on this skill for bilingual individuals.

Introduction

Bilingualism and multilingualism are widespread across the globe, with almost every country having a population that speaks two or more languages [1]. However, the impact of bilingualism on cognitive abilities and auditory processing is a matter of debate. Cognitive hearing science is based on the idea that cognitive functions play a pivotal role in auditory processing, spanning from the subcortical stage to interactions across the memory system at the cortical levels of listening, language comprehension, and discourse [2]. One specific auditory processing that is affected by bilingualism is speech perception in noise, particularly in the second language [3]. Therefore, it is necessary to investigate the effect of cognitive factors such as working memory and auditory attention on this ability. Additionally, the structure and linguistic information of the second language can influence auditory-cognitive processing [4], so it is therefore preferable to assess this ability among different languages. Our study aims to explore the correlation between speech perception in noise and working memory and auditory attention in Turkish-Persian bilinguals and Persian monolinguals aged 18 to 25 years to better understand the mechanisms of noise perception in bilinguals. This article provides an overview of bilingualism and its connection with cognitive factors before delving into the mechanisms involved in speech perception in noise. Finally, we briefly state the purpose of our study.

Bilingualism refers to the ability to use two languages someone who has this ability being called a “bilingual.“. Bilingualism is divided into two types based on the time of beginning education in the second language. People who learned the second language before the age of 6 are called, early bilinguals, and people who learned the second language after the critical age or puberty are called late bilinguals [5].

Bilingualism has both positive and negative aspects due to the complex functioning of the mental system in bilinguals [6]. Various theories have discussed the relationship between bilingualism and the human mind. Based on the balance theory, the capacity of the human mind is limited, so it considers bilingualism a negative phenomenon and a factor that leads to a decrease in the ability of both languages. For instance, when speech is processed, bilingual individuals have to handle more than one language and work with two linguistic systems, which makes them manage extra competing information compared to native monolingual speakers. When perceiving speech, both languages are simultaneously available and they affect each other bidirectionally. This co-activation may weaken predictive processing and efficient use of contextual cues, affecting the sorting out of distracting stimuli [7]. However, threshold theory considers bilingualism as a factor of cognitive superiority over monolingualism if a person is fully capable in two languages [8].

Bilingualism affects two important cognitive abilities: attention and working memory. Paying attention to a specific speech source plays a crucial role in speech perception, especially in challenging auditory conditions. Neurological studies indicate that attention modulates nerve stimulation to speech sounds, which helps to facilitate the perception of target sounds [9]. Previous research has shown that bilinguals have better attention compared to monolinguals [10].

Working memory is the ability to store and process information simultaneously, and is another important cognitive skill that affects speech perception in bilingual people; However, previous studies suggest that working memory is weaker in bilinguals compared to monolinguals [11, 12]. Two hypotheses attempt to explain the relationship between speech perception in noisy environments and working memory. The Capacity hypothesis suggests that working memory has a limited capacity that gets divided among all cognitive tasks. When the working memory capacity gets maxed out, access to other cognitive resources decreases. On the other hand, the Ease of Language Understanding (ELU) hypothesis stipulates that the degree of interference of cognitive function in speech recognition is contingent upon the hearing conditions [13]. Furthermore, individual cognitive ability can predict the amount of speech recognition in adverse hearing conditions. According to the ELU model, working memory plays a crucial role in explaining the perception of speech in noisy environments in people with hearing impairment. However, there are exceptions for people with normal hearing. some studies have shown that hearing-impaired and normal-hearing people perform equally on a cognitive factor as a predictor of speech perception in noise performance. Additionally, the reliance on working memory is the same across different signal distortion conditions when comparing between those different groups [2].

Bilingualism may impact the ability to understand speech in noisy environments, which relies on sensory and cognitive skills [10]. Since speech is often heard amid competing sounds, it’s essential to understand the mechanisms of speech perception in noisy environments. These mechanisms can be divided into five categories, including neural encoding and decoding [14], pitch perception [15], asymmetric sampling in time [16], centrifugal pathways (Pickles 1998), and cognitive skills such as attention and working memory [17]. Previous research has shown that bilingualism can affect these cognitive abilities, suggesting that changes in speech perception abilities among bilingual individuals may be partially explained by changes in these cognitive skills.

Cognitive and auditory processes differ between bilinguals and monolinguals. These differences are influenced by various factors such as age, the central processing evaluated, and the difficulty level of the test materials. Also, individual differences among bilinguals, including the various proficiency in their non-native language, can cause differences in the scores of this group in their auditory processing such as speech perception in noisy environments in their second language [18]. It has been found that working memory and auditory attention have an impact on speech perception in noise. There are differences in these cognitive characteristics between bilingual and monolingual individuals. Therefore, conducting a study to investigate the relationship between speech perception in noise, working memory, and auditory attention can provide a more complete understanding of this challenging ability. Additionally, it could help revise evaluation and intervention programs for bilingual and monolingual individuals. It is worth noting that none of the published studies have shown the simultaneous effects of working memory and selective attention on speech perception in noise in bilinguals compared to monolinguals.

Materials and Methods

After obtaining consent from the participants, each participant was given a history questionnaire to provide information on their neurological problems and musical skills due to its proven effects on cognitive processing [19],. To homogenize the participants and prevent the possible influence of cerebral lateral superiority, participants completed the Edinburgh Handedness Inventory, and right-brain people were selected (Right-handed individuals were selected because the probability of dominance of the language hemisphere is higher in right-handed people). Following this, an audiometry test (Interacoustics AC33, Denmark) and a tympanometry test (Interacoustics AT235, Denmark) were conducted. Those who showed normal results were then evaluated for speech perception in noise, working memory, and auditory attention. These tests were conducted in random order in both ears and on an individual basis (Fig. 1).

Fig. 1.

Concise set of steps of study

Participants

45 Turkish-Persian bilinguals (24 women and 21 men) and 45 Persian monolinguals (26 women and 19 men) 18 to 25 years old from Shahrekord University of Medical Sciences participated in this study. Bilinguals were early Turkish-Persian bilingual, they were born in Turkish-speaking cities, and became familiar with Turkish at home before starting school. They had only heard Persian on television and began to learn it in preschool. However, their formal education was primarily in Persian after they enrolled in school. Additionally, a questionnaire was given to bilingual individuals, and those who scored high were selected to participate in the study. they could read fluently and write in both languages. and used both languages in their daily life in written and oral form.

Procedure and Assessments

In order to initiate the investigations, the participants of this study, comprising Turkish-Persian bilinguals and Persian monolinguals, were invited to the acoustic hearing center on specific days and hours (9 AM to 12 PM). The objectives, the test procedure, and the conditions of entering and exiting the study were explained beforehand. The study consisted of four stages, which were examined in detail.

Basic Evaluations of Audiology

During the audiological evaluations, basic assessments were conducted, including otoscopy and pure tone and speech audiometry. Pure tone audiometry assessed air conduction at frequencies between 250 and 8000 Hz, followed by bone conduction at frequencies between 250 and 4000 Hz. Individuals with air-bone conduction thresholds better than 25 dB and air-bone gap less than 10 dB in both ears were invited to participate in auditory processing and cognitive evaluations.

Evaluation of Speech Perception in Noise

Speech perception in noise with quick speech in noise (Q-SIN) test in an acoustic place with calibrated Headphones (MOYAN MB-H1) and laptop (Lenovo Ideapad 130-15IKB) at listening comfort level (70dB HL or 40dB SL) was evaluated. In this study, two equivalent lists of the Persian version of the Q-SIN test with confirmed validity and reliability were used (40). Each list contains six sentences and each sentence has 5 keywords, and the sentences are presented in four-talker babble noise with a signal-to-noise ratio of + 25 to 0. Separate aligned lists were randomly used for each ear to evaluate the speech perception in noise score. The scoring method of this test is that the number of words correctly repeated in each list is recorded and subtracted from 27.5 to determine SNR [20].

Assessment of Auditory Working Memory

The n-back test was used to evaluate the working memory capacity. The test consists of six sets of nine random numbers. The examiner read out one set completely each time, and participants were required to repeat the certain stimulus before the final digit, which could be 1, 2, or 3. The number of items that the participants correctly repeated was their working memory score. The test was conducted in an acoustic environment [21].

Evaluation of Auditory Attention

The Persian version of the Bergen dichotic listening test was used to evaluate selective auditory attention. This test consisted of 36 items, each made up of a repetition of six various syllables in a specific consonant-vowel pattern. The different syllables were presented simultaneously to both ears of the participants. They were instructed to concentrate on the syllable being played in one ear while disregarding the one being played on the other side. The number of syllables that were correctly selected and repeated according to one ear was recorded as the score of auditory attention of that ear, so scores were calculated out of 36 [22]. The test was initiated randomly between ears and was conducted in an acoustics room at the most comfortable level, with calibrated headphones and a laptop.

Data Analysis

The data was analyzed using SPSS statistics 23 in this study. The results section reported descriptive statistics such as frequency, mean ± standard deviation, minimum, and maximum. The Shapiro-Wilk test was used to assess the normality of the variables. Since, the data distribution was non-normal (p < 0.05), the relationship between the variables was explored using Spearman’s correlation coefficient. The linear regression model was utilized to determine the predictive power of the independent variables for the dependent variable. The Mann-Whitney test was employed to compare quantitative variables between groups, a probability value less than 0.05 was considered statistically significant.

Results

The Effect of Gender

As the data did not follow a normal distribution, the Mann-Whitney test was used. The results of the Q-SIN, n-Back, and Bergen auditory attention tests showed no significant difference between men and women in any of the monolingual and bilingual groups (p > 0.05).

The Effect of the Group (Monolingual and Bilingual)

The test results indicate that there is a statistical difference (p < 0.05) in the Q-SIN test scores of both the right and left ears, the n-back working memory, and the Bergen auditory attention between monolinguals and bilinguals. The scores of speech perception in noise and working memory in the bilingual group are significantly lower than those in the monolingual group. However, the auditory attention in the bilingual group is significantly higher than that in the monolingual group (Table 1) (Given that the score of speech perception in noise in the Q-SIN test is based on the score of SNR Loss, a lower numerical value indicates a better performance of speech perception in noise) (Figs. 2 and 3).

Table 1.

Scores of Q-SIN, n-back, and Bergen auditory attention tests in two groups and investigating difference of them between the two groups

Test	Group	Min, Max	Mean	P.value
Right Q-SIN	Monolingual Bilingual	(-2.5,1.5) (-2.5,4.5)	-0.87 ± 0.93 0.07 ± 1.51	0.001
Left Q-SIN	Monolingual Bilingual	(-2.5,1.5) (-2.5,3.5)	-0.85 ± 0.98 -0.01 ± 1.27	0.001
n-back	Monolingual Bilingual	(4,6) (2,6)	5.35 ± 0.71 4.68 ± 1.10	0.003
Right Bergen	Monolingual Bilingual	(19,33) (20,34)	24.60 ± 3.42 26.68 ± 3.23	0.005
Left Bergen	Monolingual Bilingual	(19,33) (20,33)	23.97 ± 3.53 25.22 ± 2.99	0.050

Fig. 2.

The correlation between right and left Bergen and the average of SNR loss scores is evaluated with the QSIN test in monolingual and bilingual individuals

Fig. 3.

The correlation between n-back and the average of SNR loss scores is evaluated with the QSIN test in monolingual and bilingual individuals

Investigating the Correlation Between Speech Perception in Noise with Working Memory and Auditory Attention

To assess any discrepancies between the scores of the left and right ears during the Q-SIN test, the Wilcoxon test was utilized. Nevertheless, as there was no noteworthy variation found (p = 0.977 for bilinguals, p = 0.983 for monolinguals), the mean score of both ears was considered for evaluating the correlation between this test and the Bergen auditory attention and n-back tests. The scores from these tests were analyzed using Spearman’s test, which revealed a significant correlation between speech perception in noise in the Q-SIN test and both left and right-sided auditory attention in the Bergen test, as well as working memory capacity in the n-back test, for bilinguals. However, no such correlation was observed for monolinguals (Table 1).

Regression Model of Speech Perception in Noise Based on Cognitive Parameters of Working Memory and Auditory Attention

The results of the study showed that there was no significant correlation between the n-back working memory scores and Bergen auditory attention with QSIN results in the monolingual group. As a result, the regression model was not able to predict the scores of speech perception in noise in this group (P: 0.322). However, in bilinguals, the regression model was significant (P: 0.008), and the coefficient related to auditory working memory (n-back test results) was 0.34, while the coefficient related to auditory attention (Bergen test results) was 0.11.

Discussion

The objective of our study was to investigate the impact of working memory and auditory attention on speech perception in noisy surroundings. Furthermore, we sought to compare the speech perception abilities of bilinguals and monolinguals aged between 18 and 25 under these cognitive circumstances. In the subsequent section, we will elaborate on the outcomes of our study.

Is Gender an Influencing Factor in Q-SIN, n-back, and Bergen Auditory Attention Tests?

According to a recent study, there appears to be no notable variance in speech perception in noisy environments between males and females. similar findings are found in evaluating speech perception in noise with the Q-SIN test on Kurdish-Persian bilinguals and Persian monolinguals, which shows does not affect speech perception in noise [23]. Additionally, Calais and his colleagues, conducted a study using filtered white noise at high and low-frequency ranges and concluded that there is no significant contrast between males and females in terms of speech in noise ability [24].

Furthermore, working memory scores were found to be similar between males and females in both monolingual and bilingual contexts, aligning with the outcomes of previous studies that explored working memory tests [25].

There was no notable difference in auditory attention scores between men and women in either group, according to the data. This is in agreement with a previous study that utilized a similar iteration of the Bergen test on a younger demographic. Additionally, these findings align with the results of the analysis of the test database, which was conducted on almost a thousand participants [22].

Are the Scores of Speech Perception in Noise and Working Memory and Auditory Attention Influenced by Bilingualism?

Table 1 compares the Q-SIN test scores of bilinguals and monolinguals in both ears. The data suggests that monolinguals tend to perform better than bilinguals. This finding is consistent with previous studies that have reported poorer speech perception in noise at the sentence level among bilinguals [23, 26].

Moreover, research has shown that bilinguals and trilinguals exhibit lower performance than monolinguals in speech perception as the noise level increases. However, they perform similarly to monolinguals in silence [27]. These findings indicate that bilingualism can impact the comprehension of sentences in noisy environments, despite having normal hearing and fluency in the second language. Furthermore, According to a study conducted in 2022, bilingual individuals scored lower than monolinguals in understanding speech in noisy environments, not only in their second language but also in their first language [7].

In this study, the auditory working memory of both monolingual and bilingual individuals was examined through the n-back test. The results revealed that monolinguals possessed significantly higher working memory than bilinguals, as displayed in Table 1. This finding aligns with earlier research that also demonstrated weaker working memory in bilingual individuals [11, 12]. Nonetheless, this study contradicts a previous investigation that found dichotic verbal auditory memory to be superior in bilinguals than in monolinguals. It should be noted that the sample size in this mentioned study was limited (30 in each group), and the bilingual group was not homogenous in terms of the second language type. Furthermore, the dichotic auditory-verbal memory test utilized in the other study involved selective auditory attention, which implies that its outcomes do not solely rely on auditory memory [28].

The objective of the research was to compare the selective auditory attention of bilingual and monolingual groups using the Persian version of the Bergen auditory attention test. The outcomes revealed that bilinguals performed better than monolinguals in this aspect, which is consistent with previous studies [29, 30]. Additionally, the study discovered that bilinguals outperformed monolinguals across all age groups, from youth to elderly [30].

However, a study conducted on 7-12-year-old children found no significant difference between bilingual and monolingual groups in auditory attention. This could be due to the age gap between the participants and the development of attention processing skills. Furthermore, the language disparity of the participants and the heterogeneousness of bilinguals in a second language type, as well as the use of different evaluation tests, may have contributed to this variation [31].

What is the Relationship Between Speech Perception in Noise with Working Memory and Auditory Attention?

A study conducted by Shokuhifar has revealed an interesting correlation between speech perception in noise and working memory capacity in Kurdish-Persian bilinguals [32]. Also, based on the electrophysiology research with a speech test auditory brainstem responses (speech ABR), Bilinguals have better-sustained attention and can encode the fundamental frequency more effectively, leading to an auditory attention advantage when the stimulus is presented with noise [29]. These findings might be based on differences in cortical cognitive processing in bilinguals, which can be traced through the auditory efferent system in the lower parts of the auditory pathway [33]. In the current study, there was no significant correlation between speech perception in noise and working memory and auditory attention in the monolingual group (Table 2), which is similar to the Magimairaj study [34] which seems to be due to less dependence on cognitive abilities for processing speech perception in noise in the mother tongue.

Table 2.

The correlation between the scores of the Q-SIN and n-back and the Bergen auditory attention test in both groups

Tests	Group	Correlation	P.value
Q-SIN & n-back	Monolingual	0.044	0.772
	Bilingual	0.414	0.005
Q-SIN & Right Bergen	Monolingual	0.219	0.149
	Bilingual	0.334	0.025
Q-SIN & Left Bergen	Monolingual	0.188	0.200
	Bilingual	0.305	0.042

Moreover, the study confirms that there is a direct correlation between speech perception in noise by the Q-SIN test and n-back auditory working memory and Bergen selective attention test in bilinguals. However, in the monolingual group, there was no significant correlation between speech perception in noise and working memory and auditory attention. This may be because monolinguals rely less on cognitive abilities for processing speech perception in noise in their mother tongue, which is consistent with the Magimairaj study [34].

Can Working Memory and Auditory Attention Predict Speech Perception in Noise in Bilinguals?

Based on the calculated values, it has been found that working memory and auditory attention play a significant role in speech perception in noisy environments for bilingual individuals. Additionally, the research indicates that working memory has a more pronounced effect on speech perception in noise in bilinguals than auditory attention. This sheds light on why bilingual individuals may struggle more with speech perception in noisy environments compared to monolingual individuals, despite having superior scores in auditory attention. These findings suggest that lower working memory scores in bilingual individuals may be a determining factor in their reduced speech perception abilities in noisy environments relative to monolingual individuals.

Conclusion

The current study appears to be the first of its kind, investigating the correlation between auditory working memory, auditory attention, and speech perception ability in noise in both bilinguals and monolinguals. It has been found that in speech comprehension, bilinguals require more inhibitory control [1], which can limit their cognitive processing capacity [35], It can lead to lower performance in speech perception in noisy environments. This supports the capacity hypothesis. Conversely, monolinguals rely more on their auditory abilities and require fewer inhibitory control processes when dealing with speech [34]. Although bilingualism comes with its own set of benefits, such as improved attention. However, speech perception in noisy environments is more challenging for bilingual individuals, as it relies on their auditory working memory. Even though bilinguals may have better attention skills, it cannot compensate for their struggles with speech perception. Therefore, it is crucial to focus on strengthening both auditory processing and cognitive evaluation, especially working memory, to help bilinguals with speech perception difficulties in noisy environments.

Funding

This research did not receive any grant from the funding agencies.

Declarations

Ethical Approval

The present paper is extracted from the M.Sc. thesis of NMD with the ethics code of IR.USWR.REC.1400.336 approved at the University of Rehabilitation and Social welfare.

Conflict of Interest

The authors declare that they have no conflict of interest.