Yes I agree. Some researchers hypothesize that we may have low dopamine levels. There are some ideas for this. For example, excessive anxiety in stuttering will lower dopamine levels. [This](https://www.reddit.com/r/Stutter/comments/13ceqpk/tips_to_improve_stuttering_interventions_for/) research study (which I summarized/reviewed) by Mark Onslow, states the following about anxiety and stuttering: * Adults who stutter (AWS) have displayed high levels of anxiety on a variety of instruments including trait and state measures (Craig, 1990), social evaluative anxiety measures (Kraaimaat, Janssen, & Van Dam-Baggen, 1991), and projective tests (Bender, 1942). * In a study PWS were significantly more anxious than the control subjects, although significantly less anxious than the social phobics (Kraaimaat et al., 1991) * A recent study indicated that 87% of PWS and 97% of the speech-language pathologists who responded believed that anxiety is involved in the disorder. Further, 65% of the speech-language pathologists who treated stuttering reported that they regularly used anxiety management strategies with their clients. This is intriguing because training programs in Australia typically focus on direct modifications to speech behavior rather than on emotional, family, or associated issues (Attanasio, Onslow, & Menzies, 1996) * Hanson, Rice, and Gronhovd (1981) successfully identified stutterers from nonstutterers on the basis of modified Speech Situation Checklist (SSC) scores (on the basis of emotional response scores alone) * Adults who stutter may experience unusual levels of anxiety independent of speech to the extent that they appear anxious even when distant from speaking tasks (Despert, 1946). During high general stress, persons who stutter (PWS) show greater increases in salivary cortisol than do control subjects (Blood, Blood, Bennett, Simpson, & Susman, 1994) * **Conclusions**: * We agree with Bloodstein (1987, 1995), Ingham (1984), and Andrews et al. (1983) that, on balance, the literature does not identify a systematic relationship between stuttering and anxiety. It has not been clearly established that people who stutter are more anxious than those who do not, and the efficacy of anxiety management in the treatment of stuttering has not been demonstrated unequivocally. Many of the studies that have obtained positive findings have shortcomings in design and have been contradicted by at least one failed replication. A variety of design-related features have biased research against the identification of anxiety in subjects who stutter. The most prominent of these are related to the limited definition of anxiety adopted (Mark Onslow) * The likely outcome of reliance on the physiological system as an index of anxiety is understatement of the prevalence of anxiety in stuttering. Unrelated influences on arousal are likely to lead to an underestimation of any mediating role for anxiety in stuttering (Onslow) * Modern conceptions of anxiety have emphasized the role of “expectancies of harm” as being central to the construct (see Beck & Emery, 1985) * It is now widely accepted that anxiety is associated with, and mediated by, the expectancy of social or physical “danger.” Individuals do not appear to become anxious in the absence of expectancy of negative evaluation that is perceived to be hurtful in some way. Given demonstrations of the negative evaluation of stutterers by teachers, potential employers, and significant others * Poulton and Andrews (1994) have shown that anxiety during a speaking task is highly related to expectancies of negative social evaluation at any given time * PWS tend to have a negative evaluation from others (e.g., being misunderstood, being asked to repeat an answer, apologizing, refuting a criticism, being interviewed for a job, trying to get across a point of view, talking to teachers) * 23 of 24 subjects who stuttered had higher pulse volumes across a variety of tasks than the control subjects. (Peters and Hulstijn, 1984) That is, at 6 points of measurement for anticipation of speaking, 12 points of measurement during the task, and 6 points of measurement in the minutes following recovery, the mean pulse rate increase was higher for the stuttering subjects than for the controls