SPEECH DURING
FLUENCY FLUCTUATIONS THE MENSTRUAL CYCLE ELLEN-MARIE SILVERMAN
Marquette University, Mihvaukee, Wisconsin CATHERINE H. ZIMMER
University of Wisconsin-Milwaukee, Milwaukee, Wisconsin This study sought to determine whether changes in disfluency characteristics accompany biological and emotional changes during the menstrual cycle. Four extemporaneous speech samples were collected from each of 12 women, one at ovulation (when the average woman experiences her greatest feelings of self-esteem and selfconfidence) and one at premenstruation (when she experiences a significant increase in anxiety level) for two consecutive cycles. The mean total frequency of the women's speech disfluencies produced at premenstruation was significantly greater than the mean total of their speech disfluencies produced at ovulation. The disfluency type that accounted for much of the difference in total frequency of disfluency between the two cycle points was revision-incomplete phrase. The tendency for these nonstutterers to produce more speech disfluencies at premenstruation than at ovulation is similar to a tendency we have observed in stutterers. Nonstutterers' speech fluency has been hypothesized to vary under certain conditions (for example, Rochester, 1973). One such condition is affective state. The more negative emotion a speaker experiences, the less fluent he presumably will become (Dibner, 1958; Kasl and Mahl, 1958; Goldman-Eisler, 1961). Conversely, the more positive emotion he experiences the more fluent he can be expected to become (Brutten and Shoemaker, 1967). If affective state influences nonstutterers' speech fluency, then one might expect a systematic variation in the speech fluency of some nonstutterers during the menstrual cycle. It has been well documented that the typical woman not taking oral contraceptives, or taking sequential-type oral contraceptives, experiences a peak in positive emotion at ovulation and a peak in negative emotion at premenstruation. At ovulation (that is, midcycle) when estrogen level is at a maximum, the average woman experiences her greatest feelings of self-esteem and self-confidence (Benedek and Rubenstein, 1942; Ivey and Bardwick, 1968). At premenstruation (that is, two to three days prior to the onset of menses) when both estrogen and progesterone levels drop markedly, she exhibits a significant increase in anxiety level compared to that observed at ovulation (Benedek and Rubenstein, 1942; Coppen and Kessel, 1963; Dalton, 1964; Ivey and Bardwiek, 202
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
1968; Moos et al., 1969; Paige, 1971). One might expect, therefore, that women not taking oral contraceptives, or taking sequential-type oral contraceptives, would be less fluent at premenstruation than at ovulation. The purpose of this study was to determine whether such women produce more speech disfluencies at premenstruation than at ovulation and, if they do, to determine which types of speech disfluency they produce more frequently at premenstruation. METHOD
Subjects The subjects were 12 Caucasian university students ranging in age from 17 to 22 years. None was using oral contraceptives. All experienced regular cycles. None had a history of a stuttering problem.
Procedures Menstrual Data. Subject participation involved four three-minute speech samples, one at ovulation and one at premenstruation for two consecutive cycles. Participation began either at ovulation or premenstruation depending on the sampling sequence to which each subject had been randomly assigned. The samples at ovulation (midcycle) were scheduled for 14 to 16 days following the onset of menses, in the majority of the cases. The premenstruation samples were scheduled for two to three days prior to expected menstruation. All subjects recorded basal body temperature daily. A low point preceding a rise in temperature was taken to indicate ovulation. These data were used to check whether the "ovulation" samples had been collected, in fact, at ovulation. No independent measures of emotional state were made since recurrent, predictable alterations in mood during the menstrual cycle have been well established by others (Benedek and Rubenstein, 1942; Coppen and Kessel, 1963; Dalton, 1964; Ivey and Bardwick, 1968; Moos et al., 1969; Paige, 1971). Speech Samples. Each subject's speech was tape recorded in a sound-treated booth. The subject sat at a table facing a wall, the experimenter at another behind and to the right of the subject. When the subject indicated she was ready to begin, she was given a card with a topic written on it. After a 15second waiting period, she was signaled to begin speaking. The subjects spoke on one of four different topics each session. The order in which they spoke on the four topics was random. Random assignment of topics avoided any "loading" of a particular topic with a particular time in the cycle. The four topics were (1) women's liberation, (2) your family, (3) probable issues for the 1972 presidential campaign, and (4) a memorable life experience. We thought these would be topics about which all subjects could speak extemporaneously. Disfluency Analysis. Verbatim transcripts were prepared from the recordings. The total number of words produced during each sample was determined using SILVER.MAN,ZEM'MER:Speech Fluency Fluctuations 203
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
the protocol outlined by Johnson, Darley, and Spriestersbach (1963) for identifying words produced during spontaneous speech. Instances of speech disfluency were identified from the recordings and their location marked on the transcripts. The following disfluent behaviors were identified: interjection of sound or syllable, part-word repetition, whole-word repetition, phrase repetition, revision-incomplete phrase, disrhythmic phonation, and tense pause. (See Johnson, 1959, for a description of the first five behaviors and Williams, Silverman, and Kools, 1968, for a description of the remaining two.) Disfluencies always were counted singly. When numerous repetitions occurred within an instance, it was considered one disfluency. The frequency (per 100 words spoke,n) of each disfluent behavior and of all types combined was computed for each sample from each subject. To increase the reliability of the scores for individual subjects, the disfluency frequencies computed for each sample were averaged for each subject to yield one score at ovulation and one score at premenstruation. Reliability of Disfluency Identification. The reliability of the experimenters' disfluency identification was estimated by computation of an agreement index (Sander, 1961). Four samples were randomly selected for reanalysis. The total number of times the two experimenters agreed as well as the total number of times they disagreed identifying individual instances of disfluency was tabulated. The resultant index of 0.90 was interpreted as refecting adequate reliability for purposes of this study. Statistical Analysis. The t test for related measures was used to evaluate the results obtained here against the null hypothesis that there was no difference between the ovulation and premenstruation samples in total frequency of disfuency. A one-tailed test was used since there was no reason to believe that the women would produce more disfluencies at ovulation than at premenstruation. The alpha level was 0.05 in each instance cited below.
RESULTS The basal body temperature data indicated that the ovulation samples were taken either at ovulation or one to two days preceding or following ovulation. Since the estrogen level for the average woman not taking oral contraceptives is approximately the same one to two days preceding or succeeding ovulation as it is at ovulation (Bardwick, 1972), the samples obtained just prior to and immediately following ovulation were considered to be taken at ovulation. All premenstrual samples were collected two to three days prior to the onset of menses. All 12 women produced speech disfluencies each session. The maiority (nine) produced more speech disfluencies at premenstruation than at ovulation. The mean total frequency of disfluency for the group was 7.20 at premenstruation and 6.00 at ovulation. The difference between these means was significant (t = 2.22; df = 11). Of the three subjects who produced more disfluencies at 204 1ournal o[ Speech and Hearing Research
18 202-206 1975
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
ovulation, two had only a small difference between their ovulation and premenstruation frequencies, that is, less than 0.65. The disfluency types that accounted for most of the difference between mean disfluency frequencies at premenstruation and at ovulation were interjection and revision-incomplete phrase (see Table 1). The difference between the TABLE 1. Mean frequency (per 100 words spoken) of each of seven categories of speech disfluency produced by 12 nonstutterers at ovulation and at premenstruation.
Disfluency Category
Ovulation
Interjection Part-word repetition Whole-word repetition Phrase repetition Revision-incomplete phrase Disrhythmic phonation Tense pause
2.80 0.20 0.50 0.20 1.44 0.50 0.20
Premenstruation
*
3.50 0.20 0.00 0.25 1.81 0.70 0.15
*Significant at the 0.05 level of confidence (t test for related measures). mean number of instances of the latter type at the two cycle points was significant (t = 1.80; df = 11). The other five types (all of which occurred less than once per 100 words spoken) had approximately the same frequencies of occurrence at premenstruation as at ovulation. It was considered desirable to ascertain whether any subjects had surmised the exact purpose of the experiment. Such knowledge on the part of the maiority of subjects could have biased the results systematically (Rosenthal, 1969). At the conclusion of their participation, therefore, subjects were asked to write down what they thought the aim of the experiment had been. Two indicated that they thought the purpose had been to study the relationship between the menstrual cycle and speech fluency. Both were more disfluent premenstrually. The disfluency frequencies of these two women did not differ from that of the group. One woman produced 1.43 more disfluencies at premenstruation than at ovulation. The other produced 0.74 more disfluencies at premenstruation. DISCUSSION The findings reported here indicate that the speech of some women is more fuent at ovulation, when feelings of self-esteem are highest, and less fluent at premenstruation, when anxiety level is highest. This suggests that as self-esteem and anxiety may have varied during the menstrual cycle, disfluency attributes could be considered to have correlated with those changes in emotional state. The tendency for nonstutterers to be more disfluent at premenstruation than at ovulation is similar to a tendency we have observed in stutterers (Silverman, Zimmer, and Silverman, 1974)..The four stutterers whose spontaneous SILVERMAN, ZI'M'MER:Speech Fluency Fluctuations 205
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
speech we recorded at ovulation and at premenstruation for two consecutive cycles produced more speech disfluencies at premenstruation than at ovulation. Thus, observations of stutterers" and nonstutterers" speech disfluency at selected points in the menstrual cycle suggest that fluency level is related to affective state similarly in both stutterers and nonstutterers. ACKNOWLEDGMENT Requests for reprints should be addressed to Ellen-Marie Silverman, Communicative Disorders Research Laboratory, Marquette University, 619 North 16th Street, Milwaukee, Wisconsin 53233. REFERENCES BARDWICK,J., Her body, the battleground. Psychol. Today, 5, 50-56 ( 1972 ). BSNED~.K, T., and RUBENSTEXN,B., The sexual cycle in women: The relation between ovarian function and psychodynamic processes. Psychosom. Med Mongr., 3, 245-270 ( 1942 ). BaVTTEN, E., and SHOEMXXEa, D., The Modification of Stuttering. Englewood Cliffs, N.J.: Prentice-Hall ( 1967 ). COPI"EN, A., and KESSF.L,N., Menstruation and personality. Brit. J. Psychiat., 109, 711-721 (1963). DXLTON, K., The Premenstrual Syndrome. Springfield, Ill.: Charles C Thomas ( 1964 ). DXBNEa, A., Ambiguity and anxiety. J. abnorm. (soc.) Psychol., 56, 165-174 (1958). GOLDMAN-EISLEa, F., A comparative study of two hesitation phenomena. Lang. Speech, 4, 18-26 ( 1961 ). IVEy, M., and B.*mDWICK,J., Patterns of affective fluctuation in the menstrual cycle. Psychosore. Med., 30, 336-345 ( 1968 ). JOHNSON, W., The Onset of Stuttering. Minneapolis: Univ. of Minnesota Press (1959). JOHNSON, W., D~aLEY, F., and SPaIESTEBSBACH,D., Diagnostic Methods in Speech Pathology. Evanston, Ill.: Harper and Row ( 1963 ). KASL, F., and MAHL, G., Experimentally induced anxiety and speech disturbances. Amer. Psychol., 13, 349 ( 1958 ). Moos, R., KOPELL, B., MELGER,G., YALON,I., LUNDE, D., CLAYTON,K., and HAMBURT,D., Fluctuations in symptoms and moods during the menstrual cycle. 1. psychosora. Res., 13, 37-44 ( 1969 ). P~acE, K., The effects of oral contraceptives on affective fluctuations associated with the menstrual cycle. Psychosom. Med., 33, 515-537 ( 1971 ). ROCHESTEa, S., The significance of pauses in spontaneous speech. J. psycholing. Res., 1, 5181 (1973). ROSENTKaL, R., Interpersonal expectations: Effects of the experimenter's hypothesis. In R. Rosenthal and R. Rosnow (Eds.), Artifacts in Behavioral Research. New York: Academic ( 1969 ). SXNDER, E., The reliability of the Iowa speech disflueney test. In W. Johnson (Ed.), Studies of the Speech Di~uency and Rate of Stutterers and Nonstutterers. J. Speech Hearing Dis. Monogr. Suppl. 7, 21-30 ( 1961 ). SmVEaMAN, E.-M., ZrMMER, C., and SmVEaMAN, F., Variability in stutterers' speech disfluency: The menstrual cycle. Percept. mot. Skills, 38, 1037-1038 ( 1974 ). W X L ~ S , D., SILVEnMAN, F., and KOOLS, J., Disfluency behavior of elementary school stutterers and nonstutterers: The adaptation effect. J. Speech Hearing Res., 11, 622-630 ( 1968 ). Received March 19, 1974. Accepted July 4, 1974.
206 Journal of Speech and Hearing Research
18 202-206
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
1975
FLUENCY FLUCTUATIONS THE MENSTRUAL CYCLE ELLEN-MARIE SILVERMAN
Marquette University, Mihvaukee, Wisconsin CATHERINE H. ZIMMER
University of Wisconsin-Milwaukee, Milwaukee, Wisconsin This study sought to determine whether changes in disfluency characteristics accompany biological and emotional changes during the menstrual cycle. Four extemporaneous speech samples were collected from each of 12 women, one at ovulation (when the average woman experiences her greatest feelings of self-esteem and selfconfidence) and one at premenstruation (when she experiences a significant increase in anxiety level) for two consecutive cycles. The mean total frequency of the women's speech disfluencies produced at premenstruation was significantly greater than the mean total of their speech disfluencies produced at ovulation. The disfluency type that accounted for much of the difference in total frequency of disfluency between the two cycle points was revision-incomplete phrase. The tendency for these nonstutterers to produce more speech disfluencies at premenstruation than at ovulation is similar to a tendency we have observed in stutterers. Nonstutterers' speech fluency has been hypothesized to vary under certain conditions (for example, Rochester, 1973). One such condition is affective state. The more negative emotion a speaker experiences, the less fluent he presumably will become (Dibner, 1958; Kasl and Mahl, 1958; Goldman-Eisler, 1961). Conversely, the more positive emotion he experiences the more fluent he can be expected to become (Brutten and Shoemaker, 1967). If affective state influences nonstutterers' speech fluency, then one might expect a systematic variation in the speech fluency of some nonstutterers during the menstrual cycle. It has been well documented that the typical woman not taking oral contraceptives, or taking sequential-type oral contraceptives, experiences a peak in positive emotion at ovulation and a peak in negative emotion at premenstruation. At ovulation (that is, midcycle) when estrogen level is at a maximum, the average woman experiences her greatest feelings of self-esteem and self-confidence (Benedek and Rubenstein, 1942; Ivey and Bardwick, 1968). At premenstruation (that is, two to three days prior to the onset of menses) when both estrogen and progesterone levels drop markedly, she exhibits a significant increase in anxiety level compared to that observed at ovulation (Benedek and Rubenstein, 1942; Coppen and Kessel, 1963; Dalton, 1964; Ivey and Bardwiek, 202
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
1968; Moos et al., 1969; Paige, 1971). One might expect, therefore, that women not taking oral contraceptives, or taking sequential-type oral contraceptives, would be less fluent at premenstruation than at ovulation. The purpose of this study was to determine whether such women produce more speech disfluencies at premenstruation than at ovulation and, if they do, to determine which types of speech disfluency they produce more frequently at premenstruation. METHOD
Subjects The subjects were 12 Caucasian university students ranging in age from 17 to 22 years. None was using oral contraceptives. All experienced regular cycles. None had a history of a stuttering problem.
Procedures Menstrual Data. Subject participation involved four three-minute speech samples, one at ovulation and one at premenstruation for two consecutive cycles. Participation began either at ovulation or premenstruation depending on the sampling sequence to which each subject had been randomly assigned. The samples at ovulation (midcycle) were scheduled for 14 to 16 days following the onset of menses, in the majority of the cases. The premenstruation samples were scheduled for two to three days prior to expected menstruation. All subjects recorded basal body temperature daily. A low point preceding a rise in temperature was taken to indicate ovulation. These data were used to check whether the "ovulation" samples had been collected, in fact, at ovulation. No independent measures of emotional state were made since recurrent, predictable alterations in mood during the menstrual cycle have been well established by others (Benedek and Rubenstein, 1942; Coppen and Kessel, 1963; Dalton, 1964; Ivey and Bardwick, 1968; Moos et al., 1969; Paige, 1971). Speech Samples. Each subject's speech was tape recorded in a sound-treated booth. The subject sat at a table facing a wall, the experimenter at another behind and to the right of the subject. When the subject indicated she was ready to begin, she was given a card with a topic written on it. After a 15second waiting period, she was signaled to begin speaking. The subjects spoke on one of four different topics each session. The order in which they spoke on the four topics was random. Random assignment of topics avoided any "loading" of a particular topic with a particular time in the cycle. The four topics were (1) women's liberation, (2) your family, (3) probable issues for the 1972 presidential campaign, and (4) a memorable life experience. We thought these would be topics about which all subjects could speak extemporaneously. Disfluency Analysis. Verbatim transcripts were prepared from the recordings. The total number of words produced during each sample was determined using SILVER.MAN,ZEM'MER:Speech Fluency Fluctuations 203
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
the protocol outlined by Johnson, Darley, and Spriestersbach (1963) for identifying words produced during spontaneous speech. Instances of speech disfluency were identified from the recordings and their location marked on the transcripts. The following disfluent behaviors were identified: interjection of sound or syllable, part-word repetition, whole-word repetition, phrase repetition, revision-incomplete phrase, disrhythmic phonation, and tense pause. (See Johnson, 1959, for a description of the first five behaviors and Williams, Silverman, and Kools, 1968, for a description of the remaining two.) Disfluencies always were counted singly. When numerous repetitions occurred within an instance, it was considered one disfluency. The frequency (per 100 words spoke,n) of each disfluent behavior and of all types combined was computed for each sample from each subject. To increase the reliability of the scores for individual subjects, the disfluency frequencies computed for each sample were averaged for each subject to yield one score at ovulation and one score at premenstruation. Reliability of Disfluency Identification. The reliability of the experimenters' disfluency identification was estimated by computation of an agreement index (Sander, 1961). Four samples were randomly selected for reanalysis. The total number of times the two experimenters agreed as well as the total number of times they disagreed identifying individual instances of disfluency was tabulated. The resultant index of 0.90 was interpreted as refecting adequate reliability for purposes of this study. Statistical Analysis. The t test for related measures was used to evaluate the results obtained here against the null hypothesis that there was no difference between the ovulation and premenstruation samples in total frequency of disfuency. A one-tailed test was used since there was no reason to believe that the women would produce more disfluencies at ovulation than at premenstruation. The alpha level was 0.05 in each instance cited below.
RESULTS The basal body temperature data indicated that the ovulation samples were taken either at ovulation or one to two days preceding or following ovulation. Since the estrogen level for the average woman not taking oral contraceptives is approximately the same one to two days preceding or succeeding ovulation as it is at ovulation (Bardwick, 1972), the samples obtained just prior to and immediately following ovulation were considered to be taken at ovulation. All premenstrual samples were collected two to three days prior to the onset of menses. All 12 women produced speech disfluencies each session. The maiority (nine) produced more speech disfluencies at premenstruation than at ovulation. The mean total frequency of disfluency for the group was 7.20 at premenstruation and 6.00 at ovulation. The difference between these means was significant (t = 2.22; df = 11). Of the three subjects who produced more disfluencies at 204 1ournal o[ Speech and Hearing Research
18 202-206 1975
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
ovulation, two had only a small difference between their ovulation and premenstruation frequencies, that is, less than 0.65. The disfluency types that accounted for most of the difference between mean disfluency frequencies at premenstruation and at ovulation were interjection and revision-incomplete phrase (see Table 1). The difference between the TABLE 1. Mean frequency (per 100 words spoken) of each of seven categories of speech disfluency produced by 12 nonstutterers at ovulation and at premenstruation.
Disfluency Category
Ovulation
Interjection Part-word repetition Whole-word repetition Phrase repetition Revision-incomplete phrase Disrhythmic phonation Tense pause
2.80 0.20 0.50 0.20 1.44 0.50 0.20
Premenstruation
*
3.50 0.20 0.00 0.25 1.81 0.70 0.15
*Significant at the 0.05 level of confidence (t test for related measures). mean number of instances of the latter type at the two cycle points was significant (t = 1.80; df = 11). The other five types (all of which occurred less than once per 100 words spoken) had approximately the same frequencies of occurrence at premenstruation as at ovulation. It was considered desirable to ascertain whether any subjects had surmised the exact purpose of the experiment. Such knowledge on the part of the maiority of subjects could have biased the results systematically (Rosenthal, 1969). At the conclusion of their participation, therefore, subjects were asked to write down what they thought the aim of the experiment had been. Two indicated that they thought the purpose had been to study the relationship between the menstrual cycle and speech fluency. Both were more disfluent premenstrually. The disfluency frequencies of these two women did not differ from that of the group. One woman produced 1.43 more disfluencies at premenstruation than at ovulation. The other produced 0.74 more disfluencies at premenstruation. DISCUSSION The findings reported here indicate that the speech of some women is more fuent at ovulation, when feelings of self-esteem are highest, and less fluent at premenstruation, when anxiety level is highest. This suggests that as self-esteem and anxiety may have varied during the menstrual cycle, disfluency attributes could be considered to have correlated with those changes in emotional state. The tendency for nonstutterers to be more disfluent at premenstruation than at ovulation is similar to a tendency we have observed in stutterers (Silverman, Zimmer, and Silverman, 1974)..The four stutterers whose spontaneous SILVERMAN, ZI'M'MER:Speech Fluency Fluctuations 205
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
speech we recorded at ovulation and at premenstruation for two consecutive cycles produced more speech disfluencies at premenstruation than at ovulation. Thus, observations of stutterers" and nonstutterers" speech disfluency at selected points in the menstrual cycle suggest that fluency level is related to affective state similarly in both stutterers and nonstutterers. ACKNOWLEDGMENT Requests for reprints should be addressed to Ellen-Marie Silverman, Communicative Disorders Research Laboratory, Marquette University, 619 North 16th Street, Milwaukee, Wisconsin 53233. REFERENCES BARDWICK,J., Her body, the battleground. Psychol. Today, 5, 50-56 ( 1972 ). BSNED~.K, T., and RUBENSTEXN,B., The sexual cycle in women: The relation between ovarian function and psychodynamic processes. Psychosom. Med Mongr., 3, 245-270 ( 1942 ). BaVTTEN, E., and SHOEMXXEa, D., The Modification of Stuttering. Englewood Cliffs, N.J.: Prentice-Hall ( 1967 ). COPI"EN, A., and KESSF.L,N., Menstruation and personality. Brit. J. Psychiat., 109, 711-721 (1963). DXLTON, K., The Premenstrual Syndrome. Springfield, Ill.: Charles C Thomas ( 1964 ). DXBNEa, A., Ambiguity and anxiety. J. abnorm. (soc.) Psychol., 56, 165-174 (1958). GOLDMAN-EISLEa, F., A comparative study of two hesitation phenomena. Lang. Speech, 4, 18-26 ( 1961 ). IVEy, M., and B.*mDWICK,J., Patterns of affective fluctuation in the menstrual cycle. Psychosore. Med., 30, 336-345 ( 1968 ). JOHNSON, W., The Onset of Stuttering. Minneapolis: Univ. of Minnesota Press (1959). JOHNSON, W., D~aLEY, F., and SPaIESTEBSBACH,D., Diagnostic Methods in Speech Pathology. Evanston, Ill.: Harper and Row ( 1963 ). KASL, F., and MAHL, G., Experimentally induced anxiety and speech disturbances. Amer. Psychol., 13, 349 ( 1958 ). Moos, R., KOPELL, B., MELGER,G., YALON,I., LUNDE, D., CLAYTON,K., and HAMBURT,D., Fluctuations in symptoms and moods during the menstrual cycle. 1. psychosora. Res., 13, 37-44 ( 1969 ). P~acE, K., The effects of oral contraceptives on affective fluctuations associated with the menstrual cycle. Psychosom. Med., 33, 515-537 ( 1971 ). ROCHESTEa, S., The significance of pauses in spontaneous speech. J. psycholing. Res., 1, 5181 (1973). ROSENTKaL, R., Interpersonal expectations: Effects of the experimenter's hypothesis. In R. Rosenthal and R. Rosnow (Eds.), Artifacts in Behavioral Research. New York: Academic ( 1969 ). SXNDER, E., The reliability of the Iowa speech disflueney test. In W. Johnson (Ed.), Studies of the Speech Di~uency and Rate of Stutterers and Nonstutterers. J. Speech Hearing Dis. Monogr. Suppl. 7, 21-30 ( 1961 ). SmVEaMAN, E.-M., ZrMMER, C., and SmVEaMAN, F., Variability in stutterers' speech disfluency: The menstrual cycle. Percept. mot. Skills, 38, 1037-1038 ( 1974 ). W X L ~ S , D., SILVEnMAN, F., and KOOLS, J., Disfluency behavior of elementary school stutterers and nonstutterers: The adaptation effect. J. Speech Hearing Res., 11, 622-630 ( 1968 ). Received March 19, 1974. Accepted July 4, 1974.
206 Journal of Speech and Hearing Research
18 202-206
Downloaded From: https://jslhr.pubs.asha.org/ by a University Library Utrecht User on 03/29/2018 Terms of Use: https://pubs.asha.org/ss/rights_and_permissions.aspx
1975
