رابطه بین اضطراب و لکنت زبان: یک رویکرد چند بعدی

The relationship between anxiety and stuttering is equivocal from both clinical and empirical perspectives. This study examined the relationship within the framework of the multidimensional interaction model of anxiety that includes an approach to general anxiety in specific situations [J. Pers. Soc. Psychol. 60 (1991) 919]. Ninety-four males aged 18–43, half disfluent speakers and half fluent speakers completed two questionnaires: The Trait Anxiety Inventory [C.D. Spielberger, R.L. Gorsuch, R.E. Lushene, Manual for the State-Trait Anxiety Inventory (Self Evaluation Questionnaire), Consulting Psychologists Press, Palo Alto, CA, 1970] and the Speech Situation Checklist [G.J. Brutten, Neurolinguistic Approaches to Stuttering, Mouton, The Hague; G.J. Brutten, Stuttering: A Second Symposium, Harper and Row, New York, 1973; G.J. Brutten, P. Janssen, Proceedings 18th Congress of the International Association of Logopedics and Phoniatrists, Washington, DC, 1975; M. Vanryckeghem, Proceedings of the XXIVth Congress of the International Association of Logopedists and Phoniatrists, Nijmegen University Press, Nijmegen, 1981]. In addition, after performing speech and non-speech tasks, participants evaluated their level of anxiety on a subjective scale, labeled Task-Related Anxiety—TRA. The stuttering group also evaluated the level of severity of their stuttering. Findings indicate that trait anxiety is higher among people who stutter compared to fluent speakers, thus indicating that anxiety is a personality trait of people who stutter. State anxiety in social communication is higher among severe stutterers as compared to mild stutterers and fluent speakers. Thus, state anxiety is related to stuttering severity. The results are discussed in the frame of the multidimensional model of anxiety.

Anxiety is a complex psychological construct said to involve three components, namely verbal-cognitive, behavioral, and physiological (Lazarus & Opton, 1966). The reactions of anxiety to stressful situations and their determinants have received much attention during the last decade (e.g., Endler, Edwards, & Vitelli, 1991). Two types of anxiety have been recognized: Trait anxiety refers to a person’s inherent level of anxiety and state anxiety, referring to a condition or situation-specific anxiety (Spielberger, Gorsuch, & Lushene, 1970). Spielberger (1979) suggests that the way a person perceives a threat determines his/her reaction. A situation will be non-stressful when the person does not perceive it as dangerous, or he has the capacity or experience to cope with this situation. The state-trait anxiety entails a consideration of both personality and situational factors in predicting changes in level anxiety. Each of these two major types of anxiety has been conceptualized as rather uniform in nature and as a single construct. This view can mask the possibility to see particular components of trait or state anxiety (Spielberger et al., 1970). Contrary to the traditional approach, a model developed in the last decade holds that both state and trait anxiety are multidimensional (Endler, Edwards, et al., 1991; Endler, Parker, Bagby, & Cox, 1991). The model suggests four types of situationally determined trait anxiety: social evaluation, physical danger, ambiguity, and daily routines. These components are seen to be largely independent (Endler, Parker, et al., 1991). The multidimensionality of trait anxiety implies that individuals differ in their predisposition to experience anxiety. For example, individuals with a high disposition to experience trait anxiety in the social evaluation sphere will exhibit increased state anxiety when confronting social evaluation but not necessarily when confronting physical danger. Endler and his colleagues demonstrate the effectiveness of the multidimensional concept of trait anxiety in predicting increases in state anxiety. They show the complex interplay between person variables and situation variables. For example, individuals who were high on the physical danger type of situation-determined trait anxiety exhibited increased state anxiety when confronted with the missile attack during the 1991 Gulf War in Israel (Lobel, Gilat, & Endler, 1993). Anxiety has been assigned various roles in theories of stuttering. Whereas some view anxiety as the main cause of the disorder (Sheehan, 1970 and Wischner, 1952), others treat it as a mediating variable be it precipitating, perpetuating, or aggravating factors (Brutten & Shoemaker, 1967; Gregory, 1991 and Van Riper, 1973). Others still, view anxiety as a by-product of stuttering (Perkins, 1979 and Ryan, 1974). In addition, several researchers and clinicians who noted emotional reactions typical to people who stutter, view anxiety as a general stress trait (Barbara, 1960; Caruso, Chodzko-Zajko, Bidinger, & Sommers, 1994; Craig, 1990; Craig, Hancock, Tran, & Craig, 2003) while others regard it as a state condition related to communication in general and to speech communication in particular (Kraaimaat, Janssen, & Van Dam-Baggen, 1991; Miller & Watson, 1992; Peters, 1987). Craig (1990) claims that trait anxiety develops from the social communicative disturbance resulting from the fluency problem. Empirical studies of anxiety and stuttering have mostly supported a positive relationship (Blood, Blood, Bennett, Simpson, & Susman, 1994; Craig, 1990; Fitzgerald, Djurdjic, & Maguinet, 1992; Fowlie & Cooper, 1978; Kraaimaat, Janssen, & Brutten, 1988; Kraaimaat, Vanryckeghem, & Van Dam-Baggen, 2002; Mahr & Torosian, 1999; Miller & Watson, 1992). In addition, a positive relation between the severity of stuttering and level of anxiety has been reported (Caruso et al., 1994 and Fitzgerald et al., 1992; Weber & Smith, 1990). However, there have been studies that failed to find that people who stutter are significantly more anxious than normally fluent speakers (Gray & Brutten, 1965; Janssen & Kraaimaat, 1980; Peters & Hulstijn, 1984). In spite of positive findings in a good number of studies, reviewers of the literature have tended to reject the idea of a meaningful relationship between anxiety and stuttering (Andrews et al., 1983, Bloodstein, 1995 and Ingham, 1984). Menzies, Onslow, and Packman (1999) reviewed the literature on the role of anxiety in stuttering and asked “why, despite considerable evidence of a relationship between anxiety and stuttering, the nature of that relationship has not been identified” (p. 5). They suggested that the reason may be related to the construct of anxiety and how it is defined, and recommended that it be construed as multidimensional, and that self-reports and behavioral measures of anxiety be incorporated in future research. They further state “It would appear essential to include expectancy measures of social threat in studies of the role of anxiety in this population” (p. 6). Research of the verbal-cognitive aspect of anxiety is typically carried out through self-report instruments. Historically, the first questionnaires used with adults who stutter were early personality inventories, such as the Woodworth-Matthews and Woodworth-Cady (McDowell, 1928), Woodworth-House Mental Hygiene Inventory (Johnson, 1932), and the Bernreuter Personality Inventory (Bender, 1939). In the 1950s, the use of psychological tests, such as The Minnesota Multiphasic Personality Inventory (MMPI), gained a notable influence in personality research (Boland, 1953; Dahlstrom & Craven, 1952; Pizzat, 1951). On the whole, the finding indicated that people who stutter showed a tendency towards a slightly less favorable adjustment than normally fluent speakers, but the scores fell well within the normal range (Boland, 1953; Fowlie & Cooper, 1978; Pizzat, 1951; Sermas & Cox, 1982). The use of general psychological tests in stuttering research has been continued with tests, such as—State-Trait Anxiety Index (STAI), becoming increasingly more popular (Blood et al., 1994 and Craig, 1990; Miller & Watson, 1992). During the long period covered in the above review, several self-report questionnaires have been developed specifically for adults who stutter. These instruments contributed to the study of personality as they are concerned with perceptions and evaluations of the self. They included the Speech Attitude Scale (Knower, 1938), the Iowa Scale of Attitude toward Stuttering (Ammons & Johnson, 1944), the Communication Attitude Scale (Erickson, 1969), the Speech Situation Checklist (Brutten, 1973 and Brutten, 1975; Brutten & Janssen, 1981; Vanryckeghem, 1999), and the Inventory of Communication Attitude (Watson, 1995). Generally speaking, the method of self-report has the advantage of providing information about the individual’s evaluation of the subjective experience, especially when the measure controls motivational distortions, such as social desirability (Derogatis, 1982). In our study we elected, therefore, to employ several measures of self-reports to quantify general anxiety and social communicational anxiety. Subjective state anxiety was measured by two dimensions: general state anxiety that the situation evokes and the second specific state anxiety that the task evokes. As stated earlier, past research dealing with the relationship between stuttering and anxiety has been limited and inconclusive, resulting perhaps from widely differing methods and flaws in experimental controls. Not less important, however, may be the limitations of the traditional anxiety constructs. Whereas anxiety is a term with many meanings, much of the research has targeted a single level of that construct. Recent scholars have conceptualized anxiety using a multidimensional model that implies that those who score high on trait anxiety manifest greater state anxiety when situational stress is congruent with one of their trait anxiety levels. So far no research in the stuttering population has pursued this multidimensional approach. This may be one reason why fine tendencies in the stuttering-anxiety connection have not been detected. Therefore the purposes of this study were to re-examine the relationship between stuttering and trait and state anxiety in general and after performing a specific task and to explore whether the multidimensional interaction model facilitates new perspectives of this connection.