The Self-Reported Roles of Executive Functions, Behavior Experiences and Locus of Control in Adults Who Stutter

Purpose: Myriad studies have endeavored to determine why people stutter. It has been suggested that prior experiences, self-perceptions, specific cognition abilities (i.e., executive functions), and/or whether a person perceives their stuttering to be controlled internally or externally can provide insight into stuttering behaviors. What is not known is whether a relationship exists between these variables has more influence on persons who stutter than another. This preliminary study used the Locus of Control of Behavior scale (LCB; Craig et al., 1984), Overall Assessment of the Speaker’s Experience of Stuttering; (OASES; Yaruss & Quesal, 2006, Yaruss & Quesal, 2008), and the Comprehensive Executive Function Inventory, (CEFI-A-SR; Naglieri & Goldstein, 2013) to better understand what relationships, if any, exist between locus of control of behavior, stuttering experiences, and executive functions in persons with fluency disorders. Method: A total of 116 adults who stutter completed online surveys which included demographic questions, the LCB, OASES, and CEFI-A-SR. Results: Self-monitoring was the lone executive function scale score demonstrating a relationship to LCB and OASES scores. Conclusions: An internal control of behavior and positive outlook on stuttering experiences as well as a sense of effective self-monitoring may be required for a person to manage stuttering behaviors effectively.


Introduction
Stuttering is a disorder of speech fluency that affects approximately 1% of adults (Craig et al., 2002).
Although it can often be diagnosed early (i.e., around two years of age), stuttering may persist in as much as 20 percent of this population (Andrews et al., 1983;Bloodstein, 1995). For years, researchers have examined behaviors, perceptions, and cognitive functions of persons with fluency disorders. These investigations seek critical information, as adults who stutter report higher levels of stress, anxiety, depression, and social disorders than persons who do not stutter (Blood et al., 2001;Iverach et al., 2009;Tran et al., 2011). Persons who stutter also report feeling less control over life stressors than persons who do not stutter (Craig et al., 2009).
Research has investigated several theories about why stuttering occurs. From a behavioral standpoint, it has been speculated that a relationship exists between Locus of Control of Behavior (LCB; Craig et al., 1984) and long-term stuttering exists (DeNil & Kroll, 1995). Other research has endeavored to determine what impact stuttering experiences have on stuttering (OASES; Yaruss & Quesal, 2006, Yaruss & Quesal, 2008. Some researchers have reported functional brain differences may exist in persons who stutter compared to those who do not stutter (Gordon, 2002;Chang et al., 2009;Sowman et al., 2017).
More specifically, abnormal attention and executive functioning network differences have been noted in children who stutter (Chang et al., 2017). These areas are discussed in depth in the following sections.

Locus of Control of Behavior and Stuttering
It has been suggested that the locus of control of behavior (i.e., whether persons feel their conditions arise due to internal vs. external factors) can impact long-term outcomes for individuals receiving behavioral therapy. In some studies, having an internal locus of control of behavior has been implicated in sustaining long-term outcomes in persons who stutter (Andres & Craig, 1988;De Nil & Kroll, 1995). In other works, locus of control has not been identified as a psychological predictor of stuttering (Ginsberg, 2010). Locus of control of behavior has also been shown to differ between persons who stutter and persons who do not stutter (Kumbhar & Gupta,201). In a study aimed to determine the impact of self-reporting measures in persons who stutter, self-reporting of stuttering events were shown to be beneficial in the assessment and treatment of stuttering as they provided information about when a loss of control was perceived by the person who stutters. (Guntupalli, Kalinowski, & Saltklaroglu, 2010). Thus, there is the need to further examine the impact locus of control has on persons who stutter.

Experiences of Persons Who Stutter
In addition to the LCB, experiences and reactions of persons who stutter have been identified as factors influencing not only stuttering treatment outcomes but also quality of life in persons who stutter (Klompas & Ross, 2004;Nang et al., 2018). Negatuive self-perceptions of stuttering have been associated with stuttering impairments (Nang et al., 2018), and attitudes have been identified as influencing the social interactions of persons who stutter (Leith et al., 1993). Fortunately, understanding the environment, attitudes, and social experiences of persons who stutter has been shown to help shape self-perceptions and personal narratives, which can improve stuttering outcomes (O'Dwyer et al., 2018).

Cognitive Skills and Stuttering: The Role of Executive Functions and Self-Reports
Another area that has been explored in persons who stutter involves cognition. The act of stuttering has been suggested to involve certain cognitive breakdowns (Karniol, 1995). More specifically, persons who stutter have demonstrated deficits in attention (Heitmann et al., 2004;Ofoe et al., 2018;Singer, et al., 2020), self-monitoring deficits (Goldiamond, 1965;James, 1981;Lickley et al., 2005;Vasic & Wijnen, 2005), and inhibitory control (Eggers et al., 2013). These specific cognitive processes fall under the umbrella term of "executive functions", or "EFs".
There are many definitions of EFs. EFs have been named as cognitive control activities that manage all other cognitive processes (Denckla & Reader, 1993). Baggetta and Alexander (2016) compared researcher definitions of EFs and found that most individuals identify EFs as a group of cognitive functions that provide guidance for learning and executing activities.
Because EFs include several processes, (e.g., working memory, cognitive flexibility, inhibitory control) learning which process is involved and to what extent in which activity can be challenging. Tests of EFs exist to help parse out which EFs may be stronger or more involved in determining task outcomes. In persons who stutter, brain-based differences have been reported by researchers (Etchell & Sowman, 2014). Additionally, attention to tasks that can impact comprehension have been shown to differ in persons who stutter compared to persons who do not stutter (Kamhi & McOsker, 1982).
As the use of self-reporting has often been used by researchers and clinicians working with persons who stutter, this study investigated self-reported experiences, self-reported feelings of control, and self-reported EFs of persons who stutter. No research to date has directly examined these potential relationships. Therefore, gaining a better understanding about locus of control, life experiences and cognitive interactions of adults with persist stuttering can provide critical insights to drive support strategies and identify possible treatment options.
For this study, an examination of EFs, self-perceptions and opinions of control were selected for investigation in persons who study as they reflect the outcomes of prior research. Although EFs may relate to factors impacting stuttering events in adults, no research to date has directly examined the potential relationship between EFs, self-perceptions and locus on control to determine the interactions between these variables in adults who stutter. www.scholink.org/ojs/index.php/csm Communication, Society and Media Vol. 4, No. 2, 2021 50 Published by SCHOLINK INC.

Purpose of the Study
Few studies, if any, have investigated the relationships between the self-reported locus of behavior control, self-perceptions and cognitive functions (i.e., EFs) in adults who stutter. As locus of behavior control, experiences, and EFs have been suggested to influence stuttering behaviors and outcomes, the strength of each of these variables and the relationships between them in adults stutter can provide valuable insight into this population. For example, if strong relationships exist between certain variables, or if certain EFs are identified as strengths or weaknesses by persons who stutter, treatment could be designed accordingly. Therefore, in this study, the self-reports of behavioral control, self-perceptions and self-reported EFs in persons who stutter were collected and analyzed. It is hypothesized that EFs (e.g., attention, working memory, self-monitoring) may play a more significant role in the act of stuttering than previously determined. It is further hypothesized that, self-perceptions, locus of control, and EFs will demonstrate relationships with each other. Descriptive statistics, correlations and regressions will be used to analyze the results received.

Participants
This study investigated the self-reported EFs, behaviors and perceptions of persons who stutter who responded to an online invitation to participate. Upon receiving approval from the Institutional Review Board (IRB) at Old Dominion University, an announcement explaining the study purpose and invitation to participate was posted on the National Stuttering Association (NSA) website. Interested participants who met inclusion criteria (i.e., adults 18+ years of age who stutter) received further instructions were emailed links for completion.

Procedure
Data were collected anonymously, via online assessments. The invitation to participate in the assessments was posted for three months. Participants who volunteered for this study and met inclusion criteria received email access to the survey. The collection of information occurred online via self-reported responses to the materials listed below.

Materials
Following questions regarding demographic information and stuttering history, participants completed what happens to them is the result of external circumstances). Higher scores indicate more externality. In studies of persons who stutter and persons who do not stutter, significant differences were noted in the LCB scale scores of the two groups (Craig et al., 1984;Kumbhar & Gupta, 2016;Nil & Kroll, 1995), with persons who stutter demonstrating higher scores, thus greater externality. Kumbhar and Gupta (2016) suggest that the LCB scale may provide valuable insight to guide assessment and treatment of persons who stutter. mild-moderate (impact score of 1.50-2.24); moderate (impact score of 2.25-2.99); moderate-severe (impact score of 3.00-3.74); or severe (impact score of 3.75-5.00). Thus, the higher the impact score, the more severe the person who stutters feels their stuttering impacts each category.

Comprehensive Executive Function
Inventory-Adult -Self-Report (CEFI-A-SR). Participants also completed the Comprehensive Executive Function Inventory-Adult -Self-Report (CEFI-A-SR). This Likert-like scale has been clinically proven to provide insight into an individual's assessment of the various components of their own EFs. This report assesses EF-related behaviors (e.g., inhibitory control, working memory), and identifies an individual's EF strengths and weaknesses. Information captured by the CEFI-A-SR is useful to help identify a person's specific EF skills based on the setting. The CEFI-A-SR provides a comprehensive look at overall EFs via standard scores and individualized subtest standard scores on 9 EF areas (i.e., attention, emotion regulation, flexibility, inhibitory control, initiation, organization, planning, self-monitoring, and working memory). The higher the scores, the stronger the individuals perceive their ability in that EF area.

Data Analysis
This study was a correlation and a stepwise regression research design and included an analysis of descriptive statistics involving several variables. Data were imported into SPSS Version 25 for data management and analysis. LCB, OASES, and CEFI-A-SR scale and subscale scores were calculated according to publication manual guidelines. Analysis began by obtaining study sample characteristics.
Frequencies and percentages were reported for all variables. Independent samples t-tests and one-way analysis of variance (ANOVA) were used to compare LCB, OASES, and CEFI-A-SR total score and subscale scores across demographic variables. Post hoc pairwise comparisons with Tukey's correction for multiple comparisons was conducted for one-way ANOVA tests that yielded significant results.
Pearson correlations were used to assess the correlation between LCB, OASES, and CEFI-A-SR total score and subscale scores. All assumptions for ANOVA and Pearson correlations were satisfied. A p-value < 0.05 was used to determine statistical significance.

Results
Initially, 171 individuals responded to the survey invitation. Among those individuals, 55 did not complete the survey in full and were excluded from the analytic sample. The final population consisted of 116 individuals. Table 1 displays the sample characteristics obtained from the nine demographic questions in the survey (see Appendix A). determined that they had fluency disorder, while 44% participants reported to be school age. Almost 90% of participants reported they were not currently seeking treatment for their disorder; however, 73.3% were/are involved with a stuttering self-help group (53% public and 20% private). There were 40% of participants who reported they had a family history of stuttering.

LCB, OASES, and CEFI-A-SR Scores by Race
The Hispanics, Whites, and mixed race on the CEFI-A-SR OG subscale (all p < 0.02). Asians scored less than mixed race participants on the CEFI-A-SR OG subscale (p = 0.02).
African Americans and Asians scored less than Hispanics, Whites, and mixed-race participants on the CEFI-A-SR Planning (PL) subscale (all p < 0.03). African Americans scored less than Hispanics, Whites and mixed-race participants on the CEFI-A-SR SM subscale (all p < 0.04). Asians scored less than Whites on the CEFI-A-SR subscale (p = 0.03). African Americans and Asians scored less than mixed race participants on the CEFI-A-SR Working Memory (WM) subscale (p = 0.02 and p = 0.02 respectively). African Americans and Asians had lower CEFI-A-SR total scores than Hispanics, Whites, and mixed-race participants (all p < 0.05). Post hoc pairwise comparisons of CEFI-A-SR Inhibitory Control (IC) subscale scores between racial/ethnic categories that corrected for multiple comparisons did not yield statistically significant results, despite the statistically significant one-way omnibus ANOVA test. OASES QOL impact scores were lower in participants 55 to 64 years of age than participants 18 to 24 years of age and participants 35 to 44 years of age (p = 0.03). The OASES overall raw and impact score was less in participants 35 to 44 years of age than participants 55 to 64 years of age (p = 0.04).

LCB, OASES, and CEFI-A-SR Scores by Age
Participants 18 to 24 years of age scored lower than participants older than 45 years of age on the CEFI-A-SR Emotional Regulation subscale (p = 0.001, p = 0.02, and p = 0.03 respectively). CEFI-A-SR Planning subscale scores were higher among 45 to 54 years of age than participants 18 to 24 years of age (p = 0.03).

OASES Scores
3.5.1 OASES scores by gender. LCB, OASES, and CEFI-A-SR scale and subscale scores did not differ by gender or whether they were currently involved in a self-help stuttering group.
3.5.2 OASES scores by age when fluency disorder was determined. OASES general information raw score, OASES reactions to stuttering raw and impact scores, OASES communication in daily situations impact score, and OASES overall impact score differed by what age it was determined a participant had a fluency disorder. Participants for whom it was determined that they have a fluency disorder at preschool age and school age had higher OASES general information raw scores than those for who it was determined that they have a fluency disorder as an adult (both p = 0.01). Post hoc pairwise comparisons of OASES reactions to stuttering raw and impact scores, OASES communication in daily situations impact score, and OASES overall impact score between age at diagnosis categories that corrected for multiple comparisons did not yield statistically significant results, despite the statistically significant one-way omnibus ANOVA test.

OASES scores and present involvement in treatment for stuttering. OASES general information
raw and impact score differed by whether they were currently seeking treatment for stuttering, with those not seeking treatment scoring higher on both subscales than those seeking treatment (both p < 0.02).

LCB, OASES, and CEFI-A-SR Scores and Other Diagnosed Conditions
Locus of control, OASES communication in daily situations impact score, OASES QOL impact score, OASES overall impact score, CEFI-A-SR Attention subscale score, CEFI-A-SR Inhibitory Control

LCB, OASES, and CEFI-A-SR Scores and Family Stuttering History
OASES reactions to stuttering impact score, OASES QOL impact score, OASES overall and impact scores, and CEFI-A-SR Attention subscale scores differed by family history of stuttering (all p < 0.05).
Specifically, those who had a family history of stuttering had lower OASES reactions to stuttering impact score, OASES QOL impact score, and OASES overall and impact scores and higher CEFI-A-SR Attention subscale scores than those without a family history of stuttering.

Discussion
As the need for additional research and avenues into other potential means of managing fluency disorders in adults is warranted, this research was designed to help persons who stutter, as well as clinicians and therapists working with persons who stutter. The demographic questions, LCB, OASES, CEFI-A-SR self-report in this survey provide insights into the background, experiences, and perceptions of adults who stutter.

Demographic Differences
Differences in scores were reported occurred based on region. Persons who stutter from the Southeastern and Western parts of the United States felt more organized, reported greater self-monitoring, and overall scored higher on their self-assessment of executive function via the CEFI-A-SR than others throughout the country and those who live outside United States. This may suggest the need for increased supports for individuals who stutter based on where they live, how much interaction they have with others, and regional and demographic parameters.
CEFI-A-SR scores overall differed significantly by race, with Asians and African American participants reporting lower levels than Whites on several subscales (e.g., Flexibility, Organization). These results reflect the work of others who caution generalizing information about persons who stutter to other cultures (Shames, 1989) and encourage further research into stuttering treatment that considers multicultural identification (Finn & Cordes, 1997).
Age was a significant factor in the reported scores of all scales. Overall, the younger the person who participated in our survey was, the less they felt that they had control over their stuttering behaviors, and the less emotional control and planning they felt they had. Younger participants also felt that www.scholink.org/ojs/index.php/csm Communication, Society and Media Vol. 4, No. 2, 2021 60 Published by SCHOLINK INC. stuttering had a greater overall impact on their lives. Interestingly, studies of children who stutter have often reported little to no significant difference in the stuttering events of African American children and White children (Olsen et al. (1999). Similarly, when queried about their own confidence and social acceptance, young children who stutter reported no difference when compared to children who did not stutter (Hertsberg & Zebrowksi, 2016). Likewise, there were no significant difference in the classroom interactions of students who stutter and students who do not stutter (Adriaensens et al., 2017). All studies, including the present investigation, underline the importance of understanding the individual and unique nature of stuttering.

Perceptions of Stuttering
The present study supports the work of other investigations that report fluency disorders and/or the perception of fluency change over time for the person who stutters. Researchers have found that changes occur for individuals who stutter throughout the life span (Peters & Starkweather, 1989). In the present study, self-consciousness related to stuttering becomes more apparent with age.
Self-consciousness also increased the longer a person had been identified or self-identified as a person who stutters. Specifically, participants in this study who stuttered as preschool and school age had higher OASES general information raw scores than those for whom it was determined had a fluency disorder as an adult. For persons working with individuals who stutter, the importance of knowing how long that person had been dealing with stuttering is important to consider, as time since onset or diagnosis may impact perceptions and outcomes in therapy.
Participants who were seeking treatment had lower OASES general information raw and impact scores than those not seeking treatment. These findings support the work of other researchers who have identified the importance of support groups for persons who stutter (Bradberry, 1997;Krall, 2001;Ramig, 1993;Yaruss et al., 2002). The current study suggests support groups may help persons better internalize control over their stuttering and reduce the overall impact of stuttering in their lives. These results suggest participation in support groups should be encouraged for persons who stutter.

Self-Monitoring
The CEFI-A-SR Self-Monitoring subscale was the lone variable that demonstrated a positive correlation between LCB scores and the OASES general information impact score. As previously noted, the CEFI-A-SR Self-Monitoring subscale is defined as the ability to evaluate behaviors internally and make changes as needed to fix or amend mistakes. This suggests that participants in this study feel that when their self-monitoring is not effective, their feelings of internal control diminish, and their experiences about stuttering are also negatively affected. Introspection or self-monitoring has long been examined to support long-term therapeutic changes in persons who stutter (Goldiamond, 1965;James, 1981;Lickley et al., 2005;Martin & Haroldson, 1982;Vasic & Wijnen, 2005 The concept of self-monitoring is straightforward, with the impetus of the control following on the person who stutters. But therein may lay the challenge. If a person who stutters feels that their ability to self-monitor is determined by their ability to manage their fluency, anything other than a stutter-free event may create a negative feeling about their internal control abilities and that entire experience. This may be what occurred for the persons in this study. It could be that persons who usually feel they possess control of their situations internally and overall have a positive life experiences related to stuttering feel when a stutter event happens, their ineffective or poorer self-monitoring is to blame. Both traditional means of stuttering management (i.e., fluency-shaping and stuttering management techniques) incorporate the use of self-monitoring. Therapists and clinicians should understand the importance of self-monitoring and how it influences perceptions of t abilities and even reality for persons who stutter. For example, the work of Lickley et al. (2005) reported that persons who stutter are often hypervigilant about their speech and may even assume or identify dysfluencies even when they do not occur. If self-monitoring of speech in all persons is likely rooted in a person's inner speech as suggested by Levelt (1983), persons who stutter may be underestimating their self-monitoring abilities their own internal control of fluency events that negatively impacts their stuttering experience.

Conclusion
Findings in this study provide insight into the behaviors, perceptions and cognitive functions of persons who stutter. Clinicians and therapists must be knowledgeable of effective ways to provide guidance and therapy to persons who stutter. The results strongly suggest there is the potential need for increased studies involving the influence and perceptions of self-monitoring in persons who stutter. As many individuals in this study have suggested, internal control of behavior and perceptions of their experiences with stuttering is not enough to override feelings of limited or ineffective self-monitoring in persons who stutter.

Limitations
Since the design of this research involved online data collection, it is possible that internal validity was influenced by instrumentation, selecting and administration. Posting the invitation to participate solely on the National Stuttering Association webpage limits the randomness of this sample. Instrumentation limits internal validity in that the survey was designed and validated by the researchers of this study via a pilot sampling of 10 individuals. Although all instructions were provided in a written format, the influence of a prior question response on a subsequent answer, especially if survey fatigue set in cannot be ruled out.
As the survey was administered online, the potential for someone other than the NSA member to complete it was a present, albeit small danger. External validity threats are few as it can be presumed the persons who responded to the invitation to participate were a representative sample of the population of persons with fluency disorders. Thus, the findings of this research may be widely applied to assist persons who stutter.
The possibility that participating individuals were either highly motivated for or against the topic being questioned cannot be overlooked in this online format. Accordingly, the motivation for responding and participating may have some inherent outcome bias. The persons who stutter and chose to participate in this survey may not represent the feelings or comfort level of all persons who stutter.

Future Directions
To strengthen the validity of this study's findings, future studies should include direct assessments (i.e., not self-reports) to determine whether self-monitoring skills is an EF difficulty consistently present in persons who stutter. Additionally, these direct assessments can help identify other EF challenges that persons who stutter may encounter and thus guide treatment, management, and behavior modification.
These studies could then be compared to the self-reported EF results to determine whether persons who stutter accurately portray their actual EF abilities. As many studies focus on children who stutter, an increased focus on adults who stutter and persons who have are known to be lifelong stutters would provide insight into the changes in the perceptions, behaviors, abilities, and needs of persons who stutter. This data would also provide valuable insight into the needs of persons who stutter over time.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.