The purpose of this study was to examine the relationship between language skills and prosocial behavior in 37 children with language impairment (LI) and 37 typically developing peers matched for age (ranging from 6;11 to 11;1 years). The influence of gender on this relationship was also considered. Three different subtests of the Comprehensive Assessment of Spoken Language (Carrow-Woolfolk, 1999) were used to evaluate language ability in the areas of language comprehension, language production and pragmatic judgment skills. The Teacher Behavior Rating Scale (C. H. Hart & Robinson, 1996) was used to evaluate prosocial behavior. The current study replicated previous research by documenting that children with LI demonstrate significantly poorer prosocial behavior skills than do typically developing peers. Children with LI also performed significantly more poorly on the three language subtests of paragraph comprehension, syntactic construction, and pragmatic judgment skills compared to typical peers. No significant gender differences were noted on any of the comparisons. Multiple regression analyses were used to evaluate the relationship between the three language subtests and prosocial behavior in the group with LI compared to the typical group. Results for both groups indicated that paragraph comprehension, syntactic construction, and pragmatic judgment skills were not significant predictors of prosocial behavior when used in combination or independently. Results suggest that language alone cannot predict prosocial behavior in children with LI or typically developing children.

Previous studies have shown the importance of cricothyroid muscle activation in altering fundamental frequency in the human voice. Other studies have investigated the non-linear properties of vocal fold tissue and the impact of this non-linearity on frequency response. Several physical models of the vocal folds have been made for research purposes. However, all have been isotropic in nature with linear stress-strain properties. The purpose of this study was to create a physical model with non-linear stress-strain properties to investigate the frequency response of the model as cricothyroid muscle activation was simulated (in other words, as the vocal folds were stretched in an anterior-posterior dimension). In this study the physical models of the vocal folds were stretched in 1 mm increments and the fundamental frequency (F0) was recorded at each position. Subglottal pressure was also monitored and phonation threshold pressures were recorded for each adjustment in length and vocal fold tension, because this can influence F0. Results were obtained for models with and without non-linear properties for comparison. Tensile tests were also conducted for the linear and non-linear synthetic vocal folds. Results indicate that non-linear models demonstrated a more substantial frequency response than linear vocal fold models and a more predictable F0 increase with respect to increasing vocal fold length. Phonation threshold pressures also increased with increasing vocal fold length for non-linear vocal fold models. This trend was reversed for linear vocal fold models, with phonation threshold pressures decreasing with increasing vocal fold length. These results indicate that the non-linear vocal fold models more accurately represent the human vocal folds than do linear models. This study serves as the foundation for future research to quantify the impact of non-linear tissue properties versus active tensioning (through antagonistic thyroarytenoid muscle activation) on F0 response and phonation threshold pressure.

The purpose of this study was to determine whether 20 listeners could identify the gender of 10 preadolescent children from speech samples. An additional aim was to evaluate whether listeners identified children more accurately when listening to speech samples when more linguistic context was available. The listeners were presented with a total of 190 speech samples in four different categories of linguistic context: segments, words, sentences, and discourse. The listeners were instructed to listen to each speech sample and decide whether the speaker was a male or female. In addition, the listeners were instructed to rate their level of confidence in their decision on a 1-10 scale. Results showed listeners identified the gender of the speakers with a high degree of accuracy, ranging from 86% to 95%. In addition, statistical analysis showed significant differences in the accuracy of listener judgments among the four levels of linguistic context, with segments having the lowest (83%) and discourse the highest accuracy (99%). At the segmental level, the listeners' ability to identify the each speaker's gender from a speech sample was greater for vowels than for fricatives, with both types of phoneme being identified at a rate well above chance. Significant differences in identification were found between the /s/ and /ʃ/ fricatives, but not between the four corner vowels. The perception of gender is likely multifactorial, with listeners possibly using phonetic, prosodic, or stylistic speech cues to determine a speaker's gender.

The purpose of this study was to examine the prosodic cues used by 7 to 11 year-old children to signal questions and declarative statements in terms of changes in fundamental frequency (F0), duration, and intensity. Additional aims were to evaluate how children's use of prosody changes as a function of gender and method of elicitation. A group of 16 children participated in three different types of elicitation tasks (imitative, reading, and naturalistic). An acoustic analysis revealed that the participants produced the different sentence types using a variety of acoustic cues. Not only do children vary the mean of F0 and intensity at the end of the sentences, but they also seemed to use relative differences in peak intensity and F0. Differences between sentence types were also found in the F0 and intensity slope in the terminal portion of sentences. In addition, the way in which the participants signaled sentence type changed as a function of speaker gender and elicitation method for a limited number of acoustic measures. Although the present study found acoustic differences in how the participants' produced the sentence types, additional research is needed to determine the perceptual impact of such differences.

The present study measured the accuracy of automated grammatical tagging software as compared to manual tagging in Spanish-speaking children's personal and fictional event narrative language samples. Studies have identified articles, clitic (contracted with a verb) pronouns, and verbs as clinical markers for language impairment in Spanish-speaking children. Automated grammatical tagging software may aid in the rapid identification of these grammatical markers. Grammatical morphemes of 30 first and fourth grade children's personal and fictional event narrative samples were tagged and compared with their respective manually tagged samples. The accuracy of word-level coding averaged 91%, and similar accuracy was found for clinically significant tags. Automated grammatical analysis has the potential to accurately identify clinically relevant grammatical forms in samples from children who speak Spanish.

The purpose of this study was to evaluate the use of a robot (Troy) in a low-dose treatment protocol for two children with Autism Spectrum disorder. The efficacy of intervention was measured by comparing social engagement in two contexts: interactions with two adults and interaction with an unfamiliar adult. During the treatment, a robot was included in pseudo-triadic interactions with the clinician and the child. The robot was programmed to perform simple actions that imitated the actions of the clinician in an effort to engage the child. These pseudo-triadic interactions were incorporated into intervention over the course of a 16-session intervention program. Each child's social engagement was assessed pre-intervention as well as post-intervention in the context of symbolic play in a triad and interaction with an unfamiliar adult. These pre-and post-intervention levels of social engagement were compared in order to detect significant gains after the intervention program with the robot. An increase in both participants' engaged behaviors during post-assessment measures was observed. One participant demonstrated significantly more engaged behaviors than the other.

Prosody plays an important role in speech communication. Many individuals with motor speech disorders have decreased prosodic control and thus lower overall intelligibility. Few studies have examined the effect of a flattened prosodic contour on the intelligibility of dysarthric speech, and little is known about the role that listener gender plays in understanding disordered speech. The purpose of this study was to quantify the impact of artificial prosodic manipulation on the intelligibility of dysarthric speech as a function of the extent of fundamental frequency (F0) contour flattening. A further goal was to examine the influence of listener gender on intelligibility. Speech recordings from two speakers (one with mild dysarthria and one with severe dysarthria) were synthetically altered by reducing F0 variability by 50%, 75%, and 100%. Fifty listeners transcribed the sentences and rated the perceived difficulty of the task. Results of the study indicated that a flattened F0 contour led to decreases in the intelligibility of both speakers with dysarthria, both in terms of transcription accuracy and ratings of listener confidence. All altered conditions resulted in poorer intelligibility than the unaltered utterances. For the mild speaker, scores and ratings decreased predictably in proportion to the extent of F0 flattening, whereas for the severe speaker, there was not a steady decrease in intelligibility as the F0 was progressively flattened. The utterances were more intelligible to female than male listeners.

This study examined the use of a humanoid robot to engage two children with Autism Spectrum Disorder (ASD) on responding language behaviors including language, affect, imitation, and eye contact. The robot was integrated into each child's regular intervention in low-doses (10 min of a 50 min session). The goal was to increase responding language behaviors in the children with their conversational partners. The two children participated in pre and post assessment sessions as well as 16 intervention sessions. The data from these sessions were coded into two main categories including how the children interacted (Initiating Engagement, Responding to Engagement, and Non-Engagement) and who the children interacted with (Robot Only and Both). Both children improved in response to the intervention indicating a relation between improved behavior and intervention with the robot.

This study focused on intervention using a humanoid robot to facilitate social engagement and joint attention in four children with autism. Intervention was conducted over a three month period, with each child receiving pre-testing, intervention, and post-testing. Intervention was based on the SCERTS model (Prizant, Wetherby, Rubin, & Laurent, 2003). Pre- and post-testing involved interactions with a parent, a familiar adult interaction, a less-familiar adult interaction, and a triadic interaction. This study focuses on the baseline and follow-up testing from the interaction with the familiar adult. Following a period of traditional intervention, sessions involving a humanoid robot (named Troy) were conducted. The robot was integrated into the therapy in a low dose model, meaning that during a 50-minute therapy session, approximately 10 minutes were designated to interactions using the robot to facilitate the interaction. Pre- and post-intervention assessments were recorded, analyzed, and coded for social engagement behaviors. Results comparing baseline to follow-up assessments of the interactions with the familiar adult indicated that the most notable changes were observed in reciprocal action and eye contact. Little change was noted in initiation of social engagement, symbolic play, and language. The implications of these results are discussed and recommendations for future research are provided.

This study examined a sentence-initial one-second sound prolongation as a possible fluency-inducing condition in people who stutter. The effects of this prolongation technique on the single sentence utterances of five people who stutter (PWS) and five age- and gender-matched controls were investigated. Variables tested included stuttering percentages, speaking rate, duration of phonated intervals, and correlation between upper lip and lower lip/jaw. Results showed a non-significant trend for less stuttering to occur when participants used the prolongation technique. Significant findings included longer durations of phonated intervals and more negatively correlated upper- and lower-lip movements during the prolongation condition. Rate of speech was not affected. These findings suggest that the prolongation technique caused measurable changes in speech motor control, possibly leading to greater fluency for PWS.