Date of Award
Bachelor of Science (BS)
Department of Communication Sciences and Disorders
Specific language impairment (SLI) is one of the most commonly occurring communication disorders (Castrogiovanni (2008)). SLI is an impairment in the language of children, adolescents and adults who show no other impairment such as mental, emotional or physical problems. Individuals with SLI demonstrate normal intelligence as shown by nonverbal Intelligence Quotient (IQ) scores; however, their language skills appear to be worse than their normally achieving (NA) peers. There has been an issue in identifying individuals with SLI. One factor contributing to the identification issues of SLI is the lack of a clinical marker. Although no clinical marker currently exists, tense marking, specifically past tense marking, seems to be a grammatical limitation for children with SLI and implies possible clinical identification (Rice & Wexler, 1996). The finding of a clinical marker would aid in the diagnosis of SLI.
In the search for a potential clinical marker, researchers have conducted multiple studies on children with SLI. Since there is no established clinical marker, there is an issue with the misdiagnosis and under identification of individuals with SLI. According to previous studies, tense marking may be a potential clinical marker that would aid in the diagnosis of SLI. Researchers have found that children with SLI are aware of tense marking, but they may find it optional or are unable to deal with the complexity. As concluded by Rice and Wexler (1996), verb morphology, specifically past tense marking, seems to be an issue in children with SLI. In order to test verb morphology, the researchers in the Rice and Wexler study obtained language samples from a group of SLI and NA children between the ages of 4-6. Spontaneous language samples were taken in order to test all different tenses and grammatical forms in utterances. In the task, the children are given picture cards and they are asked to describe what is occurring the picture or what the object in the picture is doing. In the Leonard, Finneran and Miller study (2009) with children ages 8-10, the child is given a model picture and is told what is going on in the picture and the child must repeat the sentence. After that, the child is shown a second picture and must describe what is happening without a model. This method enables the child to use different syntactic structures. The research conducted on children with SLI has helped to identify verb morphology as a potential clinical marker.
Although researchers have studied the verb morphology of children with SLI, they have neglected to study the adolescent SLI population. As observed in children, verb morphology appears to be a weakness in adolescents. A longitudinal study conducted by Rice (2009) using the participants of the Rice et al. (1998) study examined the long-term grammatical acquisition of children with SLI as they grew older. SLI children and their NA peers were given a list of grammatical and ungrammatical sentences that they had to mark as “good” or “not so good”. The results demonstrated that SLI adolescents do not catch up to their NA peers and they continue to present lower language and grammar skills. Similarly, in the Leonard et al. (2009) and Miller et al. (2008) study, adolescents at age 16 were tested on their grammaticality judgment and the results yielded that SLI adolescents had particular trouble identifying past tense (-ed) verbs in comparison to their NA peers. Adolescents with SLI identified (–ed) omissions only about 80% of the time, while their NA peers identified these same errors 89% of the time. An extraordinary difference does not exist between the two groups, however it cannot be assumed that SLI adolescents have caught up to their NA peers in terms of verb morphology and tense marking. It is important to note that SLI adolescents are still struggling with tense marking, thus making it a higher possibility to be identified as a clinical marker of impairment. Although it appears that both children and adolescents appear to struggle with verb morphology, it can be noted that there has been a lack of consistency with the various language-sampling methods that have been used amongst the child and adolescent SLI population.
The methods that have been used in SLI research include spontaneous language sampling (Rice & Wexler, 1996), the model method (Leonard &Miller, 2008), and the grammatical judgment tasks (Rice, 2009) (Leonard, 2009). Lastly, in the study conducted by Evernden and Reed et al. (2000), children from the ages 8-12 were asked to perform a narrative elicitation task. In this task, the children are given a wordless picture book entitled Frog, Where Are You by Mercer Mayer. They are then asked to look through the book and then tell a story that goes along with the pictures in the book. A story-telling example demonstrates necessary verb command as stories are often told in the past tense. Where verb morphology is of interest, narratives tend to elicit use of past tense because they typically involve recounts of events that have already occurred (Reed, 2011). It is possible, however, that different methods of eliciting narratives may yield different patterns of verb use. To date, only the Frog, Where Are You story recount has been used. Since adolescents have only been given the easy task of grammaticality judgment, they need to be challenged with something more difficult, such as eliciting a narrative. Due to the lack of research with adolescents with SLI, the only results and conclusions about adolescents are to be judged off of the grammaticality tests. If adolescents are given a different, more difficult language sampling task, the differences between SLI and NA adolescents may be shown to be greater than proven. As stated previously, individuals with SLI never fully catch up to their NA peers, specifically in terms of tense marking. If presented with the Narrative Elicitation task, SLI adolescents will be forced to give a narrative similar to that of their NA age-matched peer, and the results may produce greater differences in past tense marking than observed in the grammaticality tests. If greater differences are observed, then the Narrative Elicitation Task may prove to be more valid tool than for example probes or the model method for determining verb performances of adolescents. In my research, I plan on comparing the Frog, Where Are You verb performance results to the results obtained from a different narrative elicitation task in order to determine if different results emerge. Reed completed a similar study with SLI and NA adolescents who were 14-16 years old. Their results indicated that the narrative using Frog, Where Are You elicited more past tense verbs than the narrative task for which adolescents generated a story from seeing a picture. I will replicate this study with a group of younger SLI and NA adolescents to determine if the two narrative elicitation methods yield similar results with younger teens.
II. Tentative Outline
My goal is to determine if different narratives will or will not yield different patterns of verb morphology between SLI and NA adolescents. In order to do this, I will compare language samples taken from a sample of 12- to 14- year old adolescents, both Normally Achieving and SLI when they tell narratives with different elicitation approaches. My advisor and mentor, Dr. Vicki Reed, already has the access to the samples from 36 adolescents, aged 12-14 (12 SLI, 12 NA) and has provided me with existing database. For the data comparison, I will be analyzing the 36 language samples and I will be comparing the verb morphology between the SLI and NA adolescents. I will determine the difference between the two groups and then I will compare that difference to the Frog, Where Are You narrative that Dr. Reed and colleagues have already collected. My goal is to determine if the methods produce the same gap between SLI and NA adolescents or if there is a difference. This is important in order to determine which method of language sampling works better in order to demonstrate and diagnose the patterns and difficulties individuals with SLI possess.
III. Methodology and Timeline
This semester, Spring 2014, I have begun practice with analyzing and classifying the verbs from existing language samples taken by my advisor, Dr. Vicki Reed. Before I can work with real language samples, I must demonstrate at least a 95% reliability and accuracy in my classification skills. In order to check my skills, I have completed multiple practice samples and I have checked these against my language proficient partner, Ray Varner. We completed numerous practice samples until we reached 95% agreement rate. After checking my scoring against Ray, we checked our scoring against Dr. Reed in order to determine our agreement rate validity. After 8 weeks, we reached our reliability rating and Dr. Reed has certified that I am calibrated enough to be trusted with the true language narrative samples.
Beginning this semester and continuing into next semester, I will be classifying verbs and analyzing data from the real narrative language samples. As mentioned earlier, Dr. Reed has an existing database from which I will draw my samples. I will take 36 samples from 36 difference adolescents, all between the ages of 12 and 14, and I will listen to each of their language samples and classify their verb usage depending on type and grammatical accuracy. There will be 18 Normally Achieving (NA) adolescents and 18 SLI adolescents. Each SLI adolescent will be paired based upon gender and language ability as determined by the results from the Matrices subtest from the Kaufman Brief Intelligence Test (Kaufman (1983)). Adolescents will also be matched based upon economic upbringing, in order to minimize external differences between subjects being compared. The original researcher used one of two pictures in order to extract a narrative from each adolescent. One picture showed two people and a cat sitting in a tree house, hiding a water balloon and the other picture was of two girls sitting on a couch crying and reading a letter. The adolescents will be asked to provide a story about one of the pictures shown. Each of the language samples was recorded for purposes of transcribing and eventually classifying the verbs from each sample. I will continue the research by examining each of these 36 language samples and classifying each verb into a set category from the standard template provided by Dr. Reed. The template was previously used in the Huber and Reed (2011) study.
I will then mark whether the verb was used correctly or incorrectly based on grammatical rules for the particular context in which it was used. I will then analyze the tense patterns amongst the SLI adolescents and compare them to the tense patterns amongst the NA adolescents. After analyzing and classifying the verbs, I will repeat the process multiple times to enhance accuracy and eliminate bias. As previously demonstrated with the practice samples, I will be checked by intra-rater and inter-rater reliability from my language proficient peer, Ray Varner and my mentor, Dr. Reed. During reliability coding, all three of us will be blind to one another’s coding. These measures will enhance the confidence in the accuracy of the study. During the fall, I will also write the introduction and methods chapter of my thesis. At the end of Fall Semester 2014, I will have scored all 36 language samples and I will compare each SLI adolescents verb command to their Language Matched (LM) peer. I will compare each verb type and the percent of each, including the percent of correct and incorrect use of each type of verb. After I finish scoring, I will lay out the data in matching tables and draw conclusions from the data presented. During Fall Semester 2014, I will then compare the results of the Huber and Reed study to my findings. During Spring Semester 2015, I will have all of my data and my analyses and conclusions will be drawn out. I will complete the writing of my thesis paper in order to present the findings of my research.
Castrogiovanni, A. (2008). Incidence and Prevalence of Communication Disorders and Hearing Loss in Children-2008 Edition. from http://www.asha.org/research/reports/children.htm
Evernden, A., Reed, V. . (2000). An Exploration Into the Verb Morphology of Older Children with Specific Language Impairment.
Finneran, D., Leonard, L., Miller, C. . (2009). Speech Disruptions in the Sentence Formulation of School-Age Children w./ SLI.
Kaufman, A. S., Kaufman, N.L. (1983). Kaufman Brief Intelligence Test (K-BIT).
Reed, V., Huber, M. . (2011). The Effects of Narrative Elicitation Task on the Verb Morphological Patterns of Older Adolescents with and without Specific Language Impairment. James Madison University
Rice, M., Wexler, K. . (1996). Toward Tense as a Clinical Marker of Specific Language Impairment in English-Speaking Children.
Taliaferro, Megan T., "The Effects of Narrative Elicitation Task on the Verb Morphological Patterns of Younger Adolescents with and without Specific Language Impairment" (2015). Senior Honors Projects, 2010-current. 10.