Procedia - Social and Behavioral Sciences 193 ( 2015 ) th Oxford Dysfluency Conference, ODC 2014, July, 2014, Oxford, United Kingdom

Transcription

1 Available online at ScienceDirect Procedia - Social and Behavioral Sciences 193 ( 2015 ) th Oxford Dysfluency Conference, ODC 2014, July, 2014, Oxford, United Kingdom Comparison of acoustic startle response in school-age children who stutter and their fluent peers Brent Andrew Gregg, Ph.D., CCC-SLP* Megan Scott, M.S., CCC-SLP University of Central Arkansas, Conway, Arkansas, 72034, U.S.A. Abstract It is theorized that stuttering emerges as the result of an interaction between constitutional and environmental factors (Van Riper, 1982; Bloodstein, 1995), and that constitutional factors in persistent stuttering may include an emotionally reactive temperament (sometimes referred to as a sensitive temperament) (Brutten & Trotter, 1986; Brutten & Shoemaker, 1967; Conture, 1991; Guitar, 1998, 2000). Additionally, it has been proposed that children who stutter (CWS) may be inherently inclined to have a sensitive temperament compared to their normally fluent peers, which may contribute to their vulnerability in beginning, maintaining, or recovering from stuttering (Conture, 1991, 2001; Guitar, 1998; Zebrowski & Conture, 1998, Karass, Walden, Conture, Graham, Arnold, & Hartfield, 2006; Eggers, De Nil, & Van den Bergh, 2010; Walden, Buhr, Johnson, Conture & Karass 2012). The purpose of this research is to examine the reactivity/sensitivity of school-age CWS, as evidenced by the acoustic startle response and scores on a standardized temperament scale. Acoustic startle response, determined by electromyography (EMG), measures the amplitude of eyeblink response to a brief pulse of white noise. This neurophysiological assessment of emotional reactivity has been widely used in psychological research (Vrana, Spence, & Lang, 1988, p.487). This physiological measure will be paired with a parent-report measure to assess emotional sensitivity The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license 2015 The Authors. Published by Elsevier Ltd. ( Peer-review Peer-review under under responsibility responsibility of the of the Scientific Scientific Committee Committee of ODC of ODC Keywords:stuttering; school-age; acoustic startle response; temperament *Corresponding author. Tel.: ; Fax: address: bgregg@uca.edu The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( Peer-review under responsibility of the Scientific Committee of ODC doi: /j.sbspro

2 116 Brent Andrew Gregg and Megan Scott / Procedia - Social and Behavioral Sciences 193 ( 2015 ) Temperament and Stuttering The influence of temperament and emotional factors on the development of stuttering has been a topic of discussion for several decades (e.g. Bender, 1939; Brown & Hull, 1942; Glauber, 1958; Murphy, 1953; Murphy & Fitzsimons, 1960). Research in this area is prevalent today in terms of theoretical, empirical, and clinical examinations of psychological dimension as they relate to childhood stuttering disorder (e.g. Alm, 2004; Conture et al., 2006; Eggers, De Nil & Van den Berg, 2010; Peters & Hulstijn, 1984; Weber & Smith, 1990; Yairi, 1997). As previously stated, research suggests that children who stutter demonstrate increased emotional reactivity/sensitivity compared to their normally fluent peers (Fowlie & Cooper, 1978; Glasner, 1949; Schwenk, Conture, & Walden, 2007). Studies have shown that children who stutter may react more negatively to environmental stimuli (Fowlie & Cooper, 1978; Johnson et al., 2010; Karass et al., 2006) and may demonstrate higher levels of impulsivity and activity (Embrechts, Ebben, Franke, & van de Poel, 2000). Evidence also suggests children who stutter, as a group, demonstrate less emotional regulation (Karass et al., 2006; Anderson et al., 2003, Karass et al., 2003), adaptability, and inhibitory control (Anderson et al., 2003; Embrechts et al., 2000). The review of literature appears to suggest that the notion of links between stuttering and temperament should be further pursued, as most studies are in agreement that children who stutter exhibit distinguishing temperamental characteristics relative to their normally fluent peers. Specific data regarding the influence of temperament on the onset, development, and maintenance of stuttering is difficult to obtain as emotional manifestations of temperament are highly variable in the presence of different environmental stimuli. Temperament is a stable, trait-like characteristic that can manifest in a variety of emotional, state-like, ways. Therefore assessing the superficial emotions can lead to misrepresentations of temperament. Temperament is stable and interacts with stuttering as an attribute domain and not an ability domain (such as language or phonology). Temperament influences the speech disfluency as this may be a function of one s reaction or response to sensory stimuli. Additionally, manifestations of stuttering are highly variable in the presence of environmental stimuli. Previous research has failed to control for the variable nature of both stuttering and temperament-driven emotional manifestations, as it has relied heavily on parent-report and clinical observations. In order to accurately assess temperament at a constitutional level in children who stutter, neuropsychological measures must be used to assess components of temperament such as reactivity, hyper vigilance, stress response, and emotional regulation. The aforementioned investigations have addressed various aspects of temperament in children who stutter as well as children who do not stutter. However, their methods largely failed to include biological measures and assessment procedures. To best evaluate the relationship between stuttering and temperament, children should be examined using neurophysiological measures of temperament paired with parental observations, instead of parentor self-report alone. One such biological measure that has proven to be highly replicable throughout research literature is the acoustic startle response. This reflex has not been shown to be reliable over the past several years, but it also is an excellent means for addressing dimensions of temperament and neuropsychology from the perspective of the underlying mechanisms that modulate expressions of these two domains. Dawson, Schell, and Bohmelt (1999) deem the startle reflex, an exceptional tool for the study of emotion and psychopathology. The startle reflex provides hard data in domains of psychology, psychopathology, behavioral, and emotional research, which historically have relied on soft data. Additionally, this reflex is extremely similar with consistent patterns across animals and humans, allowing for investigations of attentional and emotional processing, as well as their underlying information processing mechanism. Recently, the startle response has been used in a variety of emotion-related studies including studies of fear and phobia (Hamm, Cuthbert, Globisch, & Vaitl, 1997), schizophrenia (Schlenker, Cohen, & Hopman, 1995), affect deficits and neurological impairments (Morris, Bradley, Bowers, Lang, & Heilman, 1991), anxiety disorders (Grillon, Ameli, Goddard, Woods, & Davis, 1994; Cuthbert, Straus, Drobes, Patrick, Bradley, & Lang, 1997), individual differences in emotionality (Cook, Hawk, Davis, & Stevenson, 1991; Grillon, Ameli, Foot, & Davis, 1993; Blumenthal, Chapman, & Muse, 1995; Collins, Hale, & Loomis, 1995; Corr, Wilson, Fotiadou, Kumari, Gray, N.S., Checkley, & Gray, J.A. 1995), as well as emotional development (Balaban, 1995; McManis, Bradley Cuthbert, & Lang, 1997). A large body of literature exists regarding the influence of temperament on children who stutter; however, a physiological study examining sensitivity/reactivity in school-age children who stutter using the startle response

3 Brent Andrew Gregg and Megan Scott / Procedia - Social and Behavioral Sciences 193 ( 2015 ) has not been conducted to date. Previous research indicates a number of temperamental factors may be influencing the onset, development, and maintenance of childhood stuttering disorders. These factors include heightened sensitivity/reactivity, anxiety, nervousness, helplessness, hyperactivity, hypervigilance, as well as heightened susceptibility to stress. Additionally, negative communication attitudes, poor social interactions, behavioral inhibition, and social withdrawal have been well-documented among school-age children who stutter. These studies, however, rely heavily on self- report or parent-report and fail to include physiological measures of temperamental reactivity/sensitivity. In order to accurately establish a neurophysiologic basis in a field that relies heavily on self-report and questionnaire, there is a need to include physiological measures when investigating the relationship between reactive temperament and persistent stuttering. The purpose of this research was to examine the sensitive, or reactive, temperament characteristic of children who stutter, as evidenced by the acoustic startle response and scores on subscales of a temperament measure. EMG surface electrodes were used to measure the eye blink response to white noise in children who stutter as compared to their normally fluent peers. This study differed from previous research because participants were school-age children who stutter as well as their fluent peers. Additionally, data from the EMG portion of the study were normalized in order to provide accurate and meaningful interpretation of results. 2.0 Method 2.1 Participants Five school-age children who stutter (ranging in age from 8:0 to 14:0, with a mean of 10.1) and five school-age children who do not stutter (ranging in age from 8:0 to 14:1, with a mean of 10.3) participated in this study. There were 4 males and 1 female in each group. 2.2 Procedures Surface electrodes were fixed to the participants (as described below), and headphones were fitted for the startle stimulus. Participants then were asked to sit silently and gaze at a spot on the wall, so that baseline data could be collected relative to the individual s natural eye-blink. Approximately 20 eye blinks were recorded under natural conditions, without auditory stimuli. Participants then were informed that they would hear a series of white noise bursts (at this point the sound was imitated by the researcher) separated by randomly chosen intervals between 20 and 30 seconds (Berg & Balaban, 1999; Guitar, 2003). Participants were not informed how many noise bursts were in the series. The series of noise burst consisted of approximately 20 bursts of white noise, in order to elicit 20 eye blinks (Guitar, 2003). 2.3 Startle Apparatus and Stimuli Baseline eye blink data were collected using a handheld trigger and open/close switch. This push-button trigger was integrated with LabView software to mark blinks based on visual observation by the investigator. Acoustic stimuli consisted of a 95-dB burst of white noise presented for 50ms with a 10-ms rise and fall time. These parameters were based on long-standing protocol for startle reflex analysis (Berg & Balaban, 1999), as well as procedures followed in previous research (Guitar, 2003). This burst was presented binaurally through Beyerdynamic DT 48 A.00 headphones. Stimuli were presented and responses were collected using a LabView template that allows for the administration of white noise bursts as well as the display of EMG and trigger data. A National Instruments USB data acquisition device as well as a Grass amplifier was used to collect electromyographic data as well. Startle responses were detected electromyographically using miniature 2cm silver-silver chloride electrodes placed on the periorbital area on the skin below the right eye. Gereonics electrodes were used in this study. Electrodes were trimmed, collared, and gelled. This is in accordance with procedures for orbicularis occuli placement given by Guitar (2003) as well as Fridlund and Cacioppo (1986). The electrodes were placed exactly 2 cm apart, with a reference electrode stationed on the forehead of the individual. The electromyographic signals then were filtered between 30 and 1000 Hz. Responses were detected during a 200-ms window that began at the instant of the startle

4 118 Brent Andrew Gregg and Megan Scott / Procedia - Social and Behavioral Sciences 193 ( 2015 ) stimulus presentation. Auditory acoustic stimuli consisted of approximately 95dB burst of white noise presented for 50ms. This burst was presented binaurally through Beyerdynamic DT 48 A.00 headphones. These procedures are consistent with procedures recommended by researchers in the field of psychology for use of the startle paradigm (Berg & Balaban, 1999). Inter-electrode distance was carefully controlled for each participant in order to prevent this from influencing startle amplitudes. Currently, research on children between the ages of 6 and 12 suggests that body size or muscularity does not influence startle amplitudes (Guitar, 2003). In order to control for a large range of inherent differences in individual muscle size and tissue distribution, as lipose acts as a low-pass filter, all EMG startle responses were normalized in this study. These differences make it difficult to discern meaningful comparisons across individuals. Additionally, reflexes can be highly variable across individuals. Normalization is a process that is frequently used in EMG research in order to control for these differences. The electromyogram is the sum of the motor unit activity within a specific contraction at a given electrode location. This activity is then expressed in millivolts by the data collection instruments used in this study. EMG normalization expresses the millivolts of activity as a percentage. The percentage is representative of that muscle's activity during a test contraction relative to baseline contraction measures (reference voluntary contractions). Therefore, in this study the startle was expressed as a percent of contraction relative to an individual s normal (baseline) eye blink. Normalization is critical as it controls for variables such as electrode application and placement, temperature, perspiration, muscle fatigue, contraction velocity, muscle shape and length, crosstalk from neighbouring muscles, fat tissue thickness, and slight variations in task executions. It would be impossible to control all of these variables of EMG amplitude in a clinical setting. Normalization controls for the aforementioned variables and facilitates the comparison of EMG signals in a more accurate manner. Expressing the neural activity (EMG amplitude) as a percentage makes interpretation of the signal more meaningful and significant. All previous studies (Guitar, 2003; Alm, 2005; Alm & Riseburg, 2007) concerning the startle reflex in individuals who stutter failed to include data normalization procedures within the methodology. This poor EMG technique can easily lead to misinterpretations of data. For example, a startle response of 148mV (averaged across 10 trials) across two individuals (Participant A and Participant B) would be interpreted as the same response according to the methodology employed by Guitar (2003), Alm (2005), and Alm & Riseburg (2007). However, upon further examination, Participant A s baseline eye blink might have been 100mV (mean) while Participant B s baseline eye blink might have been 140mV (mean). Normalization procedures would have revealed Participant A s startle response to be 150% greater than his/her normal eye blink, while Participant B s startle is exactly the same as his/her normal eye blink. In this example, normalizing the data reveals a significant difference between the startle responses of the two individuals. Without normalization procedures, this data is easily misinterpreted and meaningful analysis cannot occur. All data is arbitrary and does not allow for individual characteristics affecting signal amplitude, such as fatty tissue, muscle distribution, and reflex variation. Thus resulting conclusions are rendered irrelevant as well. 2.4 Reliability In order to ensure reliability in EMG procedures several variables were kept consistent throughout these procedures. The sampling rate remained at 2000Hz, contributing to the reliability of the EMG signal. Additionally, electrode placement and skin preparations were carefully monitored in order to ensure consistency. Skin preparation consists of cleaning the skin above the orbicularis occuli with an alcohol swab in order to remove any dirt or dead skin cell particles that may interfere with the surface electrode signal. Multiple baseline blinks (10) as well as multiple startle responses (10) were recorded and analyzed in order to derive response means as well as standard deviations for each participant. Outliers were defined as responses beyond two standard deviations of the mean. In this way, reliable muscle activation data was collected. Additionally, all responses were detected during a single session for each individual. This contributed to the reliability of this study. 2.5 Temperament Scales The Temperament in Middle Childhood Questionnaire (TMCQ: Simonds & Rothbart, 2004) and the Early Adolescence Temperament Questionnaire-Revised, Parent Report (EATQ-R: Ellis & Rothbart, 1999) was

5 Brent Andrew Gregg and Megan Scott / Procedia - Social and Behavioral Sciences 193 ( 2015 ) administered in order to measure individual temperament. These measures were used to assess the reactivity of participants in this study. They were paper and pencil parent questionnaires. It is necessary to use both measures in order to accommodate the 7:0-14:0 age range of participants in this study. These measures were developed at the University of Oregon by Rothbart and are statistically significantly correlated with one another. These measures are the best-fit for this current study because they explicitly test the aforementioned three dimension of temperament: surgency/extroversion, negative affect, and effortful control. The subscales for the TMCQ and EATQ-R that were used for the purposes of this examination are: anger/frustration, fear, inhibitory control, and shyness. The anger/frustration and fear subscales are designed to specifically reflect components of negative affect. The inhibitory control subscale is designed to specifically reflect aspects of effortful control within the child s temperament. The shyness subscale speaks directly to surgency/extroversion. These aspects of temperament are complex and must be finely-differentiated. These measures were developed by Mary Rothbart, whose research identified, defined, and subsequently developed accurate assessment measures for specifically examining these dimensions of temperament (Rothbart, 2004, 2007). For the purposes of this study, data from the following subscales, (which are common to both the EATQ-R and TMCQ), were analyzed: anger/frustration, fear, inhibitory control, and shyness. A five-point rating scale was used by the parent to answer each item on the test. Parents were asked to describe how true or false a statement is by circling a 1 for almost always untrue of your child, a 2 for usually untrue of your child, a 3 for sometimes true, sometimes untrue of your child, a 4 for usually true of your child, and a 5 for almost always true of your child. Each dimension received an average rating (1-5) indicating the mean response given by the parent for each item within a specific dimension. These responses were calculated and compared between the two groups of participants. 3.0 Data Analysis Data was reported in terms of amplitude of eye-blink response, which is given in analog-to-digital units. Independent t tests were used to test for the presence of significant differences between startle amplitude between the two groups. A one-tailed t test was used to compare the differences between the first response and the 10 th response within each individual to examine habituation differences. The means and standard deviations for the EATQ-R and TMCQ subscales were reported for the stuttering and non-stuttering group. A multivariate analysis of variance (MANOVA) was used to examine group differences. Pearson-produce-moment correlations tested for significant correlation between the sub scales and the mean amplitude of the response for individuals within stuttering and non-stuttering group. A discriminate analysis was preformed to investigate the extent to which the startle response measure (in terms of amplitude of response) and the scores on the EATQ-R and TMCQ subscales would discriminate between the stuttering and non-stuttering groups. 4.0 Results 4.1 Differences in Startle Response As a measure of startle response, EMG waves were analyzed and the following summary variables were provided to compare across groups. To determine the difference in startle response between groups mean amplitude and latency and standard deviations were analyzed. Shapiro Wilk test statistics and Levene s test statistics were calculated for each of these comparisons to confirm normality and homogeneity of variance, respectively, between groups. Non-significant p values for these tests indicated if the distribution of data was sufficiently normal and if variances in mean amplitude, mean latency, and habituation rate, of acoustic startle responses were comparable between groups. It was expected that these tests would indicate that a one-way multivariate analysis of variance (MANOVA) test statistic is appropriate to use for independent group comparisons (Howell, Davis, Patel, Cunife, Downing-Wilson, Au-Yeung, & Williams, 2004). In the MANOVA, participant group served as a single fixed factor. Any resulting p values that are significant in the overall MANOVA were confirmed in light of multiple comparisons and were further analyzed using Tukey s honestly significant difference test to correct for inflated

6 120 Brent Andrew Gregg and Megan Scott / Procedia - Social and Behavioral Sciences 193 ( 2015 ) family-wise error rate. 4.2 Differences in Temperament Scores The means and standard deviations for the EATQ-R and TMCQ subscales were reported for the stuttering and non-stuttering group. A multivariate analysis of variance (MANOVA) was used to examine group differences. Pearson-produce-moment correlations tested for significant correlation between the sub scales and the mean amplitude of the response for individuals within stuttering and non-stuttering group. A discriminate analysis was performed to investigate the extent to which the startle response measure (in terms of amplitude of response) and the scores on the EATQ-R and TMCQ subscales discriminated between the stuttering and non-stuttering groups. 5.0 Discussion The findings indicate that CWS may have a more intense startle response than CWNS. Their average "peak" (amplitude) during their response is higher (185.9) than the average "peak" of a CWNS (95.91), with amplitude referring to how high above baseline the response is for the designated response window (mean normalized RMS startle amplitude). Additionally, there was a statistically significant correlation between scores on the anger/frustration, fear, inhibitory control, and shyness subscales of the EATQ-R and TMCQ, and amplitude (or extent of muscle activity) of startle response (correlation coefficient =.092). More specifically, when originally presented, the EMG data is displayed positively and negatively (above and below the 0 line). Calculating the RMS inverts all negative to positive and places all data above the 0 line. This creates the "house" image or "peak" image that can be analyzed for amplitude and latency. The latency is the length of the response, or the time it takes for a child to startle and then calm to baseline. Again, the data must be normalized as a result of the variability in individual s eye blinks. Therefore, for the purposes of this investigation, the average amplitude of a child's normal eye blink and the average amplitude of their startle were combined, and a percentage was derived. In other words, a child's startle may be 200% what their normal eye blink reaction is. The percentages then were compared to make a conclusion. Preliminary results on these 5 children indicate that CWS have a significantly higher level of physiological reactivity, as measured by mean normalized root mean squared (RMS) amplitude of startle response scores, compared to their fluent peers. Additionally, there was a statistically significant correlation between scores on the anger/frustration, fear, inhibitory control, and shyness subscales of the EATQ-R and TMCQ, and amplitude (or extent of muscle activity) of startle response. The findings indicate that CWS may have a more intense startle response than CWNS. Their average "peak" (amplitude) during their response is higher than the average "peak" of a CWNS., with amplitude referring to how high above baseline the response is, for the designated response window. More specifically, when originally presented, the EMG data is displayed positively and negatively (above and below the 0 line). Calculating the RMS inverts all negative to positive and places all data above the 0 line. This creates the "house" image or "peak" image that can be analyzed for amplitude and latency. The latency is the length of the response, or the time it takes for a child to startle and then calm to baseline. Again, the data must be normalized as a result of the variability in individual s eye blinks. Therefore, for the purposes of this investigation, the average amplitude of a child's normal eye blink and the average amplitude of their startle were combined, and a percentage was derived. In other words, a child's startle may be 200% what their normal eye blink reaction is. The percentages then were compared to make a conclusion. Pairing these two measures, one direct and one indirect, provided insight into the link between temperamental factors and fluency disorders within the pediatric population. However, these two methods are rarely combined effectively within the design of a singular study, resulting in less-than-comprehensive results relative to temperament. Additionally, it bears repeating that the data reported here represent a small sample, as this is a preliminary study. While differences were exhibited, it remains to be determined if certain temperamental differences are inherent to individuals who stutter, or perhaps a result of the stuttering itself. For example, counterintuitive findings by Alm (2004, 2005) showed individuals who stutter to have a reduction in heart rate and blood pressure when compared to their normally fluent peers. Alm postulated this could be due to a co-activation of autonomic nervous system branches that are responding to communication induced stress and anxiety, indicating

7 Brent Andrew Gregg and Megan Scott / Procedia - Social and Behavioral Sciences 193 ( 2015 ) that such a response was an attempt at self-regulation. Information about the interaction between temperamental characteristics and persistent stuttering in children is needed. Using the Dual Diathesis Stress Model of Stuttering (DD-S; Walden, Frankel, Buhr, Johnson, Conture, & Karass, 2012) framework, differential findings in the area of emotional reactivity in children on the persistent track of stuttering would suggest that temperament factors are playing influential roles in stuttering susceptibility, maintenance, and conversely recovery. From a clinical perspective, treating school-age children who stutter differs from treating pre-school children who stutter. The window for the phenomenon of natural recovery is narrowing. Treating school-age children who stutter focuses on a more comprehensive understanding of the child s experience of stuttering. A key contributing factor in this scenario is the child s reaction to both the impairment as well as the resulting participation limitations. Therefore, an understanding of the constitutional temperament driving these individual reactions is a critical component of both clinical assessment and treatment. The neurophysiological underpinnings of these reactions must be objectively identified, measured, and understood. References Alm, P. (2004). Stuttering and basal gangelia circuits: A critical review of possible relations. Journal of Communication Disorders, 37, Alm, P.A. (2005). On the causal mechanisms of stuttering. Doctoral thesis. Dept. of Clinical Neuroscience, Lund University, Sweden. Alm, P., & Riseberg, J. (2007). Stuttering in adults: The acoustic startle response, temperamental traits, and biological factors. Journal of Communication Disorders, 40, Anderson, J. D., Pellowski, M. W., Conture, E. G., & Kelly, E. M. (2003). Temperamental characteristics of young children who stutter. Journal of Speech, Language, and Hearing Research, 46, Balaban, M.T. (1995). Affective influences on startle in five-month old infants: Reactions to facial expressions of emotion. Child Development, 66, Bender, J.F. (1939). The personality structure of stuttering. Journal of American Psychology 14,189. Berg, W. K., & Balaban, M. T. (1999). Startle elicitation: Stimulus parameters, recording techniques, and quantification. In M. Dawson, A. Schell, & A. Bohmelt (Eds.), Startle modification: Implications for neuroscience, cognitive science, and clinical science. New York: Cambridge University. Blumenthal, T.D., Chapman, J.G., & Muse, K.B. (1995). Effects of social anxiety, attention, and extraversion on the acoustic startle eyeblink response. Personality & Individual Differences, 19, Brown, S. F., & Hull, H. C. (1942). A study of some social attitudes of a group of 59 stutterers. Journal of Speech Disorders, 7, Collins, D., Hale, B., Loomis, J. (1995). Differences in emotional responsivity and anger in athletes and non-athletes: Startle reflex modulation and attributional response. Journal of Sport & Exercise Psychology, 17, Conture, E. (2001). Stuttering: Its nature diagnosis and treatment. Boston: Allyn & Bacon. Conture, E. G., Walden, T., Arnold, H. S., Graham, C. G., Hartfield, K. N., & Karrass, J. (2006). A communication-emotional model of stuttering. In N. Bernstein-Ratner & J. Tetnowski (Eds.), Current issues in stuttering research and practice (pp ). Mahwah:Lawrence Erlbaum Associates. Cook, E.W., III, Hawk, L.W., Davis, T.L., & Stevenson V.E. (1991). Affective individual differences and startle reflex modulation. Journal of Abnormal Psychology, 100, Corr, J.P., Wilson, G.D., Fotiadou, M., Kumari, V., Gray N.S., Checkley, S., Gray, J.A. (1995). Personality and affective modulation of the startle reflex. Personality and Individual Differences, 19, Cuthbert, B.N., Strauss, C., Drobes, D., Patrick, C.J., Bradley, M.M., & Lang, P.J. (1997). Startle and the anxiety disorders. Manuscript submitted for publication. Dawson, M.E., Schell, A.M., & Bohmelt, A.H. (1999). Startle Modification; Implications for Neuroscience, Cognitive Science, and Clinical Science. New York, New York: Cambridge University Press. Eggers, K., De Nil, L. F., & Van den Bergh, B. R. H. (2010). Temperament dimensions in stuttering and typically developing children. Journal of Fluency Disorders, 35, Ellis, L. K., & Rothbart, M. K. (1999, 2002). Revision of the Early Adolescent Temperament Questionnaire. Manuscript in preparation. Embrechts, M., Ebben, H., Franke, P., & van de Poel, C. (2000). Temperament: A comparison between children who stutter and children who do not stutter. In H.-G. Bosshardt, J.S. Yaruss, & H.F.M. Peters (Eds.), Fluency Disorders: Theory, Research, Treatment and Self-Help. Proceedings of the Third World Congress onfluency Disorders in Nyborg, Denmark (pp ). Nijmegen: Nijmegen University

8 122 Brent Andrew Gregg and Megan Scott / Procedia - Social and Behavioral Sciences 193 ( 2015 ) Press. Fowlie, G. M., & Cooper, E. B. (1978). Traits attributed to stuttering and nonstuttering children by their mothers. Journal of Fluency Disorders, 3, Fridlund, A.J., & Cacioppo, J.T. (1986). Guidelines for human electromyographic research. Psychophysiology 23, Glasner, P. (1949). Personality characteristics and emotional problems in stutterers under the age of five. Journal of Speech and Hearing Disorders, 14, Glauber, P.J., (1958). Stuttering and personality dynamics. In J. Eisenson (Ed.) Stuttering: a symposium (pp ), New York: Harper & Row. Grillon, C., Ameli, R., Goddard, A., Woods, S., & Davis M. (1994). Baseline and fear-potentiated startle in panic disorder patients. Biological Psychiatry, 35, Grillon, C., Ameli, R., Foot, M., & Davis, M. (1993). Fear-potentiated startle: Relationships to the level of state/trait anxiety in healthy subjects. Biological Psychiatry, 33, Guitar, B. (2003). Acoustic startle responses and temperament in individuals who stutter. Journal of Speech, Language,and Hearing Research, 46, Hamm, A.O., Cuthbert, B.N., Globisch, J., & Vaitl, D. (1997). Fear and startle reflex: Blink modulation and autonomic response patterns in animal and mutilation fearful subjects. Psychophysiology, 34, Howell, P., Davis, S., Patel, h., Cunife, P., Downing-Wilson, D., Au-Yeung, J., & Williams, R. (2004). Fluency development and temperament in fluent children and children who stutter. In: Packman A, Meltzer A, Peters HFM, eds. Theory, research and therapy in fluency disorders. Proceedings of the 4 th World Congress on Fluency Disorders. IFA. Montreal; Johnson, K., Walden, T., Conture, E.G., Karass, J. (2010). Spontaneous regulation of emotions in preschool children who stutter: Preliminary Findings. Journal of Speech, Language, and Hearing Research, 53, Karass, J., & Braungart-Rieker, J.M. (2003). Parenting and temperament as interacting agents in early language development. Parenting: Science and Practice, 3, Karass, J., Walden, T., Conture, E., Graham, C., Arnold, H., Hartfield, K., et al. (2006). Relation of emotional reactivity and regulation to childhood stuttering. Journal of CommunicationDisorders, 39, McManis, M.H., Bradley, M.M., Cuthbert, B.N., & Lang, P.J. (1997). Kids reactions to affective pictures: A 3-systems study. Manuscript submitted for publication. Morris, M., Bradley, M., Bowers, D., Lang, P., Heilman, K. (1991). Valence-specific hypoarousal following right temporal lobectomy. Paper presented at the Nineteenth Annual Meeting of the International Neuropsychological Society, San Antonio, Texas. Murphy, A. (1953). An electroencephalographic study of frustration in stutterers. Doctoral dissertation, University of Southern California. Los Angeles, CA. Peters, H. F., & Hulstijn, W. (1984). Stuttering and anxiety. Journal of Fluency Disorders, 9, Rothbart, M K (Oct 2004). Temperament and the pursuit of an integrated developmental psychology". Merrill-Palmer quarterly 50 (4): Rothbart, M.K. (2007). Temperament, development and personality. Current Directions in Psychological Science, 16, Schlenker, R., Cohen, R., Hopmann, G. (1995). Affective modulation of the startle reflex in schizophrenic patients. European Archives of Psychiatry and Clinical Neuroscience, 254, Schwenk, K., Conture, E.G., & Walden, T. (2007). Reaction to background stimulation of preschool children who do and do not stutter. Journal of Communication Disorders, 40, Simonds, J., & Rothbart, M. K. (2004). Temperament in Middle Childhood Questionnaire. Manuscript in Preparation. Walden, T., Frankel, C., Buhr, A., Johnson, K., Conture, E. G., Karass, J. (2012). Dual Diathesis-Stressor Model of Emotional and Linguistic Contributions to Developmental Stuttering. Journal of Abnormal Child Psychology 40, Weber, C. M., & Smith, A. (1990). Autonomic correlates of stuttering and speech assessed in a range of experimental tasks. Journal of Speech and Hearing Research, 33, Yairi, E. (1997). Epidemiological factors and stuttering research. In N.B. Ratner & C. E. Healey (Eds.), Stuttering research and practice: Bridging the gap (pp ). Mahwah: Lawrence Erlbaum Associates.