Introduction
Stuttering is a speech disorder that occurs with frequent and abnormal disruptions in speech, such as sound repetition, sound prolongation, and sound or airflow blockage. Environmental, movement, cognitive, and linguistic factors play a role in the development of stuttering [1]. Despite these various factors proposed for the etiology of stuttering, its cause has not yet been entirely determined. In understanding the etiology of stuttering, we can refer to cognitive and linguistic theories, such as the covert repair hypothesis (CRH) and the vicious circle hypothesis (VCH). In the CRH, it is proposed that stuttering based on the model of speech production and its monitoring [2] is a phenomenon of Covert Repair. According to this theory, the cause of disfluency is a defect in phonological coding, which causes errors during speech planning, and as a result of repeating these errors, repair occurs, and stuttering is its side effect [3].
The VCH also is the increase in speech disfluency due to the hypervigilance to speech, which causes warnings to be sent during coding errors, and the primary disfluency is highlighted. When the child becomes aware of his disfluency, it ultimately leads to a vicious cycle that causes compensation. According to this theory, when a person with stuttering pays attention to other activities while speaking, there is a reduction in abnormal disfluency (especially seizures) [4]. Also, in the hypothesis of disfluency as a result of explicit attention, dyspraxia is the result of a defect in execution, and the reason for that is much attention to speech and high self-awareness; this theory considers an essential distinction in information processing between explicit and implicit memory. Implicit memory is more for performing motor skills, and explicit memory is involved in conscious activities. Implicit memory does not require working memory (WM), is less prone to confusion, and is, therefore, suitable for good management of motor skills and acquisition of executive patterns. The pressure or demand causes one to consciously focus on internal movements and makes explicit memory to be involved in controlling movements. As a result, confusion occurs in the automaticity of movements [5].
 Also, regarding cognitive aspects, we can mention the role of WM in the language performance of children with stuttering. WM is a system that temporarily actively stores information in mind to perform verbal and non-verbal tasks and make them available for further processing. According to the Baddeley model, WM consists of three primary components. The core of this system is the central executive, which is responsible for information coordination and acts as a short-term attention system. The next component is the phonological loop, which is related to speech encodings, and phonological information is stored in this reservoir for a short time. The third component is the visuospatial sketchpad, which is dedicated to maintaining and manipulating visuospatial images [6]. The relationship between WM and stuttering has been evaluated, and most studies believe that people with stuttering have deficits in WM [7, 8, 9, 10, 11, 12, 13, 14, 15, 16]. Sakhaei et al. found that children before school age had a weaker performance in repeating non-words than nonstuttering children of the same age, which indicates problems in their phonological WM [17]. In this regard, it is possible to investigate cognitive effects on the severity of stuttering using dual tasks in people with stuttering verbally (phonological), non-verbally (visuospatial), or using attentional conflict (central executive). The meaning of dual tasks is that during the speech, a second task is presented to the person, which engages his attention resources, and his attention is directed to a subject other than his speech so that monitoring of speech is suppressed [5] and the effects of reducing attention and WN on the severity of stuttering should be measured. However, there are contradictions to dual tasks and stuttering. Based on the theories mentioned above, people with stuttering experience a reduction in their disfluency when they are exposed to distracting stimuli (dual tasks) [2, 18, 19, 20 ,21, 22]. For example, in 2020, 26 people with stuttering and 24 nonstuttering people aged 19-38 years were investigated for the severity of stuttering when people have to remember the target words while reading the text. The results indicated that the frequency of disfluency in these people decreased significantly [19].
Also, in a study published in 2019, the role of dual tasks in stuttering was discussed. In this study, 19 people with stuttering and 20 nonstuttering people ranging in age from 18 to 35 years were investigated. The participants had to have 12 one-minute speech samples during one session, some of these tasks were single, and some were dual (verbal and non-verbal). The results of this study indicated that when the attention was divided between two phenomena and the participants were simultaneously doing a task with the computer (the tasks were in the form of mathematical calculation and following an arrow) that involved their attention, the disfluency reduced. Moreover, this happened when the demands of the second task during the speech were minimal and simple [23]. Also, in another study, 11 people with stuttering and 11 nonstuttering people with an average age of 26 years were examined to follow the movement of a dot on the computer screen during a spontaneous speech in a session that lasted 30 to 40 minutes. Then, under these conditions, the severity of their stuttering was examined. The results indicated that people’s stuttering is reduced in these conditions [18].
 However, some studies believe that disfluency increases under the conditions of dual tasks [24, 25, 26, 27]. It is also mentioned in the overload theory that stuttering is a problem in the motor control of speech, and the motor problems of people with disfluency are that their speech is not automatic enough. As a result, it is assumed that during dual tasks, the second task will demand more attention on the central attention system because the speech of people with stuttering requires an unusual amount of attention control, and at the same time, more attention demands are needed during dual tasks, and the result will be an increase in their stuttering. Therefore, stuttering is the immediate effect of saturation of the capacity of the central attention reservoir [18]. Metten et al. assessed stuttering and dual tasks. In part of their work, they examined 17 people with stuttering, whose average age was 36 years, in a 90-minute session. The participants told a story from the six presented images and had to simultaneously place the cards on the table according to the instructions. The findings indicated that this condition increases the disfluency of people [27]. In a study conducted in 1982, subjects with stuttering were examined during dual conditions. In this condition, they should read aloud and perform movement activities (sitting and standing up); these people showed increased disfluency [26].
 The only study found by researchers in the field of dual tasks and stuttering in Iran was that conducted by Bahrami et al. in 2006. In one session, they investigated the dependence of speech on attention in children and adolescents with developmental stuttering. In this study, 30 children ranging in age from 8 to 13 years did single (text reading) and double (text reading with finger tapping) tasks, and the severity of their stuttering was investigated. The results indicated that people’s disfluency increases in dual tasks [28].
According to the mentioned issues and the contradictions obtained from the studies, the question arises whether performing dual tasks causes a change in stuttering severity. Looking at past studies, we can see that this issue has been of interest to researchers for a long time (since 1986). In the older studies on dual tasks, WM and its performance in people with stuttering have been most discussed, but as the studies become more recent they have mainly investigated the effects of dual tasks on the severity of stuttering. Also, in these studies, contradictory results have been obtained due to the various manipulations that have been done on the tasks. Despite the attractiveness of the topic and considering that it is still of interest to researchers, the current study is one of the few studies that have been conducted on children in the field of dual tasks and the proposal of studying in this field in previous studies has been proposed [5]. Also, both verbal and non-verbal tasks have been used in this study, and despite numerous international studies in the field of double tasks, very little has been done in Iran (the researcher only looked at one similar study example in Iran). While considering the cultural differences, the environment, and the different experiences of people with stuttering, localizing this study is necessary. Finally, considering that the current study measured the effect of dual tasks immediately after presenting the dual tasks individually, we decided to measure the effects of dual tasks on the severity of stuttering. Therefore, this study investigated the effect of dual tasks on the severity of stuttering in school-age children.
Materials and Methods
This research was a cross-sectional observational study. Considering the information obtained from Bahrami’s article [28], the standard deviation of the stuttering severity scores before and during the performance of the skill was about 3.4 (s=3.4), and considering the potency of 80% and the confidence level of 95% and reaching a difference of at least 2 (d=2), the sample size was determined to be 39 people using the following Equation 1:


A sampling of eligible people considering the inclusion criteria was done from March 2019 to November 2019 by convenience sampling from primary schools in districts four and eight and speech therapy clinics in Tehran. The onset of stuttering before elementary school and not having any accompanying disorders (physical, mental, and psychological) were the inclusion of criteria for the study. Also, if the child could not perform dual tasks, or the subject and his parents expressed displeasure to continue cooperation at each stage of the study, he was excluded.
 Forty-four children aged 8-12 years with developmental stuttering (five girls and 39 boys) participated in this study, of whom 39 children met the inclusion criteria and remained in the study, and five children were excluded.
At first, the questionnaire on personal information was completed by parents. Then, a prototype of verbal and non-verbal tasks was provided to the examinees as a preparation period of 5-10 minutes, and they were allowed to familiarize themselves with the tasks during this time. The children who performed the test sample (verbal and non-verbal tasks alone) successfully entered the next stage, and others were excluded from the study. Four speech samples of one minute were taken from the selected children. In the first and fourth speech samples, the subject was asked to speak for one minute about a topic of interest from the list (speech topics were selected in consultation with people who were proficient in working with children). In the second and third speech samples, the subject had to perform one of the WM tasks (verbal and non-verbal) randomly presented within one minute, in addition to talking about the topic of interest. In the verbal memory program, a numerical calculation was used. The subject had to add or subtract three numbers that appeared on the screen at an interval of 3 seconds and the result was recorded as a number using the keyboard. In the non-verbal memory program, orientation was used, in which the subject had to follow three arrows that were presented at an interval of 3 seconds, and in the end, he determined the position of the point inside the square that had four houses. The keyboard recorded the correct answer as a number (1 to 4). On average, the sampling process of each of the subjects was done in one session that lasted about 30 to 40 minutes. After collecting the data, the severity of stuttering was obtained and calculated based on the percentage of stuttered syllables (SS%) from four speech samples. The data were entered into SPSS and analyzed using statistical tests. This work was done using a Sony VGN-CR353 laptop.
The task paradigm used in the designed software was similar to the tasks used in the article by Eichorn [5]. In order to check the face validity, the desired program was given to five speech and language pathologists and ten children who were in the age range of 8 to 12 years. Specialists were asked to express their opinion regarding the appropriateness of the designed program and its compliance with the purpose of the study. Also, the children did the homework and were asked to comment on the ease of implementing the program.
The experts determined that the implementation of this program covers the intended study goals, and their opinions and suggestions were considered and applied in the final design of the designed software. Also, the tested children stated that implementing the program is easily possible.
Also, a speech and language pathologist was asked to randomly obtain the percentage of stuttered syllables in 15 speech samples selected from the single elementary task to check inter-rater reliability. Then, using Spearman’s correlation coefficient, the correlation between the severity of stuttering determined by the two raters was estimated at 0.968 with a significance level of 0.000, which indicated a very high correlation between the results of the two raters.
Results
In the present study, 39 children with stuttering with an average age of 10.25 years and a standard deviation of 1.35 were investigated, of whom five cases were girls (12.8%) and 34 cases (87.2%) were boys. The minimum and maximum age of the subjects were 8 and 12 years, respectively. Also, 12-year-old children showed the highest frequency (n=12), and 8-year-old children showed the lowest frequency (n=5). In this model, the effect of gender and age was first adjusted, but because it was not significant (P>0.05), it was removed from the model. The information on the subjects according to age and gender is shown in Table 1.


The mean and standard deviation of the stuttering intensity of the participants in four speech samples was obtained in terms of the SS%, which is shown in Table 2.


The lowest mean of the stuttering intensity in 39 subjects in all four tasks performed was related to the dual non-verbal task (14.85). Moreover, the maximum mean intensity of stuttering was related to the primary task (17.11), the minimum intensity of stuttering recorded in all speech samples was zero, and the maximum intensity was 60.86.
Using the Shapiro-Wilk test, it was determined that the variables did not follow the normal distribution. As a result, due to the non-normality of the data distribution, using the generalized linear model with the generalized estimating equations (GEE) method, it was found that the performed tasks had a significant effect on the intensity (Table 3).


Finally, the Bonferroni test was used for pairwise comparison of mean intensity and its results showed that the only different mean values in stuttering intensity that had a significant difference were related to the initial and final tasks (P>0.05) (Table 4).


In the side-by-side comparison of other stuttering severity mean values, no significant difference was observed (P>0.05), although the probability value was close to the significance level of 0.05 in some comparisons.
Discussion
The results of this study showed that performing dual tasks by children with stuttering reduced the severity of their stuttering. In other words, directing the attention of the children studied in this study to verbal and non-verbal tasks while they were speaking led to a reduction in their disfluency, and the effects of this decrease in intensity continued even after the dual tasks in the final task. In general, the mean of all stuttering intensities obtained (in the second, third, and fourth tasks) was lower than the initial stuttering intensity. As seen in Table 2, in the pairwise comparison, the intensity difference in the final speech sample was significantly less than the initial sample (P>0.05).
In the first investigation, it was observed that the mean intensity of stuttering in the dual verbal task was lower than the mean intensity of stuttering in the speech sample of the primary task, which is in line with the findings of Eichorn et al. [5]. In their study, which was held in one session, a rest time was considered for the participants. The severity of stuttering decreased during the dual verbal task and the reason for this similarity is that in both studies, the verbal task had a minor difficulty and the tasks were automatic. It should be noted that automatic performances require minimal attention and are not affected by continuous mental activities. When a task is well practiced, and we have much information about it, continuous efforts are no longer needed [5]. Also, in this regard, it is possible to refer to a similar study conducted by Bosshardt et al. on the effect of dual verbal tasks on the severity of stuttering. They concluded that the participants’ disfluency in the dual verbal tasks decreases when the minimum demands of mental storage and recovery are applied, but when demands for short-term storage and recovery increase and the task becomes more complex, disfluency also increases [20]. In a different study from the present study, Metten et al. stated that the disfluency in dual verbal tasks (word classification) increases during the implementation of the Camperdown treatment program. They assumed that during dual tasks, especially when the second task was language, the neural processing load of people increases, and the participants have a hard time focusing on extended speech in the presence of the second task, and as a result, more disturbances occur in their speech [27]. It can be concluded that while performing these dual tasks, people had to focus on learning the treatment program using obvious attention. As a result, a double load was imposed on the central executive system and the phonological loop, which increased disfluency.
 In the second comparison, which was between the primary and dual non-verbal tasks, the intensity of stuttering decreased in the dual task, but this difference was not significant. In this regard, Eichorn et al. [5] also reached similar results. The reason for this similarity can be attributed to the reduction of explicit attention on speech as well as the simplicity and automaticity of the second task. We can also mention the study by Arends et al., in which the severity of stuttering decreased during dual non-verbal tasks. They believed that stuttering comes from the person’s efforts during the speech, which interferes with the automaticity of speech, and more demands on the attention system during dual tasks reduce the interference and, as a result, lead to a reduction in disfluency [18].
In the third comparison of stuttering intensity between verbal and non-verbal dual tasks, we concluded that the mean stuttering intensity of the non-verbal dual task was lower than the mean intensity in the dual verbal task, which can be attributed to the processing load of the second task. We assume that in non-verbal tasks, the processing interference of two tasks is less and causes the central nervous system to perform two tasks with less pressure, which is in line with the results obtained by Bosshardt et al. who pointed out that people with stuttering while doing competitive tasks that involve similar neurological resources, they are more prone to speech dissociation [29]. These findings are in line with the multi-component view of WM proposed by Baddelly and Hitch who believed that exercises that include different neural areas are often performed without their function interfering [30], or Leclercq who believed that while doing processes that require similar neural areas, the results of tasks interfere with each other [31].
In the fourth comparison, which was made between the initial and final single tasks, it was observed that the severity of stuttering in the single final speech was less than in the initial single speech. In other words, it can be said that the impacts of the previous tasks on the participants reduced the severity of stuttering. This result means that the use of dual tasks during this study reduced participants’ attention to their speech, and this impact remained even after the removal of the second task, reducing the conscious effort that interferes with the automaticity of speech and subsequently, the severity of the disfluency will also decrease. These findings are in line with the Vicious Cycle hypothesis and CRH. Unfortunately, the researchers could not find a similar study, in which the stuttering intensity was compared in two single samples, the initial and the final. In short, our study showed that the involvement of attention and WM during speech would reduce stuttering, and when the modalities for doing tasks are not the same, its impacts on reducing disfluency are more evident. The obtained results also support the CRH. Therefore, it is expected that WM affects the integrity of speech monitoring and increases speech errors. As a result, when the WM is busy with another task, this monitoring and stuttering are reduced. Also, according to the hypothesis of disfluency as a result of explicit attention, movement improvement is observed during dual tasks involving the WM. What was obtained was in line with the assumption that people relying on implicit memory, experience easier speech through dual tasks, leading to reduced disfluency. These results will also confirm the Vicious Cycle hypothesis; when a person with stuttering engages in attention-demanding requests while speaking, we see a reduction in disfluencies.
Finally, it can be mentioned that in studies on dual tasks, factors, such as the type of task (verbal or non-verbal), the task’s difficulty level, and whether the second task is automatic or non-automatic have an impact on the results of dual studies. For example, in some studies, only verbal or non-verbal task was used, which may have different results depending on the nature of the task. Alternatively, the level of difficulty of tasks affects how people perform them. For example, by increasing the difficulty level of the second task, the amount of processing load that the tasks have on the nervous system increases and produces different results. Also, the degree of automaticity or non-automaticity of the tasks influenced the results obtained. For example, Metten et al. mentioned that stuttering intensity increased when the newly learned pattern was implemented during dual tasks, which was unusual and non-automatic for the person. These three factors or other factors can be mentioned in justifying the contradictory results. In this regard, it is suggested that factors, such as difficulty level, type of tasks, and the degree of automaticity of tasks for the participant should be included in the articles in order to better understand the nature and type of influence of dual tasks on changes in the severity of dysfluency. It is also suggested that more studies should be done on children, considering that most studies have been conducted in the adult age range.
Conclusion
 This study showed that doing verbal or non-verbal tasks while speaking reduced the disfluency of children participating. Therefore, it is expected that when we direct the attention of a person with stuttering from speech to another subject, the severity of his disfluency will decrease.
In this study, all 31 codes of ethics and guidelines for vulnerable groups were observed, written consent was obtained from the participants, and full information about the study process was given to all. The obtained information, both personal information and the results of the tests, were confidential and were only with the senior researcher of this study, and no personal information was allowed to be published in any way. Also, if the participants at any study stage wish not to continue the cooperation, they could withdraw from the study. It can be safely noted that this study did not have any adverse effects on the participants and their treatment process. This study was free, and the participants were charged no fees. Also, reporting the study results guarantees the material and intellectual rights of all persons related to the study.

Ethical Considerations
Compliance with ethical guidelines
This study was approved by the Ethics Committee of the University of Social Welfare and Rehabilitation Sciences (Code: 1398.187IR.USWR.REC).

Funding
The article was extracted from the master’s thesis of the first, Department of Speech Therapy, Faculty of Rehabilitation, University of Social Welfare and Rehabilitation Sciences.

Authors' contributions
Conceptualization and supervision; Akbar Darouie, Fahime Keyhani; Methodology: Akbar Darouie, Samane Hosseinzade, Fahime Keyhani; Programming: Mohammad Keyhani; Investigation: Morteza Farazi, Akbar Darouie; Writing: Fahime Keyhani; Review and Aditing: All authors.

Conflict of interest
The authors report no conflicts of interest.

Acknowledgments
The authors consider it necessary to thank the officials of the General Administration and Regions four and eight of Tehran, Hasheminejad, Ibn Sina, Muslim, Jahantarbit, Quds, Fakhami, and Soroush schools, the children and their families participating in this study, the colleagues who participated in the introduction of the samples, Ms. Kausar Salemi and other dignitaries who helped in some way to carry out this study. 


References

1. Guitar B. Stuttering: An integrated approach to its nature and treatment. Philadelphia: Lippincott Williams & Wilkins; 2013. [Link]
2. Levelt WJ, Roelofs A, Meyer AS. A theory of lexical access in speech production. Behavioral and Brain Sciences. 1999; 22(1):1-38. [DOI:10.1017/S0140525X99001776] [PMID]
3. Postma A, Kolk H. The covert repair hypothesis: Prearticulatory repair processes in normal and stuttered disfluencies. Journal of Speech, Language, and Hearing Research. 1993; 36(3):472-87. [DOI:10.1044/jshr.3603.472]
4. Vasic N, Wijnen F. Stuttering as a monitoring deficit. In: Hartsuiker RJ, Bastiaanse R, Postma A, Wijnen F, editors. Phonological encoding and monitoring in normal and pathological speech. London: Psychology Press; 2005. [Link]
5. Eichorn N, Marton K, Schwartz RG, Melara RD, Pirutinsky S. Does working memory enhance or interfere with speech fluency in adults who do and do not stutter? Evidence from a dual-task paradigm. Journal of Speech, Language, and Hearing Research. 2016; 59(3):415-29. [DOI:10.1044/2015_JSLHR-S-15-0249] [PMID]
6. Baddeley AD, Hitch G. Working memory. Psychology of Learning and Motivation. 1974; 8:47-89. [DOI:10.1016/S0079-7421(08)60452-1]
7. Amini M, Shafiei B, Abedi A. [Investigation of the effectiveness of phonological working memory intervention on decreasing of severity of stuttering in preschool children: A single-subject research (Persian)]. Journal of Research in Rehabilitation Sciences. 2015; 11(3):220-7. [DOI:10.22122/JRRS.V11I3.2320]
8. Barikroo A, Tohidast SA, Mansuri B, Yadegarfar G. [Comparing phonological working memory in preschool children with and without stuttering (Persian)]. Journal of Research in Rehabilitation Sciences. 2011; 7(3):368-80. [DOI:10.22122/JRRS.V7I3.243]
9. Dhatri S, Kumar UA, Santosh M. Comparison of working memory abilities in adults who do and do not stutter. Journal of Indian Speech Language & Hearing Association. 2017; 31(2):42. [DOI:10.4103/jisha.JISHA_5_17]
10. Farazi M, Kamkary K, Hassanzade Noghani A. [A survey on cognitive functions of students with stuttering (Persian)]. Archives of Rehabilitation. 2018; 19(2):160-7. [DOI:10.32598/rj.19.2.160]
11. Kaganovich N, Wray AH, Weber-Fox C. Non-linguistic auditory processing and working memory update in pre-school children who stutter: An electrophysiological study. Developmental Neuropsychology. 2010; 35(6):712-36. [DOI:10.1080/87565641.2010.508549] [PMID] [PMCID]
12. Mansuri B, Tohidast SA. [Working memory impairments in children with stuttering (Persian)]. 2013; 8(5):977-87. [DOI:10.22122/JRRS.V8I5.52]
13. Pyasik M, Kozlovskiy S, Vartanov A, Glozman J.  Visual working memory in people with stuttering: ERP study. Paper presented at: Conference Abstract: ACNS-2013 Australasian Cognitive Neuroscience Society Conference. 28 November 2013; Melbourne, Australia. [DOI:10.3389/conf.fnhum.2013.212.00174]
14. Oyoun HA, El Dessouky H, Shohdi S, Fawzy A. Assessment of working memory in normal children and children who stutter. Journal of American Science. 2010; 6(11):562-6. [Link]
15. Reilly J. Verbal working memory skills of children who stutter: A preliminary investigation. Contemporary Issues in Communication Science and Disorders. 2005; 32:38-42. [DOI:10.1044/cicsd_32_S_38]
16. Saifpanahi S, Sobhani Rad D, Afzali M, Izanloo S, Mardani N, Gholamian M. [An investigation of the correlation between phonological and visual working memory with severity of stuttering in 6-12 years-old children (Persian)]. Journal of Paramedical Sciences & Rehabilitation. 2015; 4(4):20-6. [DOI:10.22038/JPSR.2015.5344]
17. Sakhai F, Darouie A, Anderson JD, Dastjerdi-Kazemi M, Golmohammadi G, Bakhshi E. A comparison of the performance of Persian speaking children who do and do not stutter on three nonwords repetition tasks. Journal of Fluency Disorders. 2021; 67:105825. [DOI:10.1016/j.jfludis.2020.105825] [PMID]
18. Arends N, Povel DJ, Kolk H. Stuttering as an attentional phenomenon. Journal of Fluency Disorders. 1988; 13(2):141-51. [DOI:10.1016/0094-730X(88)90035-6]
19. Arongna, Sakai N, Yasu K, Mori K. Disfluencies and strategies used by people who stutter during a working memory task. Journal of Speech, Language, and Hearing Research. 2020; 63(3):688-701. [DOI:10.1044/2019_JSLHR-19-00393] [PMID]
20. Bosshardt HG. Effects of concurrent mental calculation on stuttering, inhalation and speech timing. Journal of Fluency Disorders. 1999; 24(1):43-72. [DOI:10.1016/S0094-730X(98)00022-9]
21. D’Ambrosio M, Bracco F, Benso F. Assessing fluency in persons with stuttering by complex automa-tized and non-automatized dual-task conditions. Giornale Italiano di Neuroscienze, Psicologia e Riabilitazione. 2014; 1(5):104-8. [Link]
22. Eichorn N, Marton K. When less can be more: Dual task effects on speech fluency. Paper presented at: The 37th Annual Meeting of the Cognitive Science Society. 22-25 July 2015; California, USA. [Link]
23. Eichorn N, Pirutinsky S, Marton K. Effects of different attention tasks on concurrent speech in adults who stutter and fluent controls. Journal of Fluency Disorders. 2019; 61:105714. [DOI:10.1016/j.jfludis.2019.105714] [PMID]
24. Bosshardt HG. Effects of concurrent cognitive processing on the fluency of word repetition: Comparison between persons who do and do not stutter. Journal of Fluency Disorders. 2002; 27(2):93-114. [DOI:10.1016/S0094-730X(02)00113-4]
25. Caruso AJ, Chodzko-Zajko WJ, Bidinger DA, Sommers RK. Adults who stutter: Responses to cognitive stress. Journal of Speech, Language, and Hearing Research. 1994; 37(4):746-54. [DOI:10.1044/jshr.3704.746] [PMID]
26. Kamhi AG, McOsker TG. Attention and stuttering: Do stutterers think too much about speech? Journal of Fluency Disorders. 1982; 7(3):309-21. [DOI:10.1016/0094-730X(82)90025-0]

27. Metten C, Bosshardt H-G, Jones M, Eisenhuth J, Block S, Carey B, et al. Dual tasking and stuttering: From the laboratory to the clinic. Disability and Rehabilitation. 2011; 33(11):933-44. [DOI:10.3109/09638288.2010.515701] [PMID]
28. Bahrami H, Nejati V, Pouretemad H. [Attentional demand of speech in children and adolescents with developmental stuttering (Persian)]. Journal of Fasa University of Medical Sciences. 2012; 2(2):96-100. [Link]
29. Bosshardt HG. Cognitive processing load as a determinant of stuttering: Summary of a research programme. Clinical Linguistics & Phonetics. 2006; 20(5):371-85. [DOI:10.1080/02699200500074321] [PMID]
30. Cocchini G, Logie RH, Della Sala S, MacPherson SE, Baddeley AD. Concurrent performance of two memory tasks: Evidence for domain-specific working memory systems. Memory & Cognition. 2002; 30(7):1086-95. [DOI:10.3758/BF03194326] [PMID]
31. Leclercq M. Theoretical aspects of the main components and functions of attention. Applied neuropsychology of attention: Theory, diagnosis and rehabilitation. London: Psychology Press; 2002. [Link]