The research related to this paper started with an interest in aphasia. From there it developed into a specific type of treatment for aphasia. I ultimately chose to dive deeper into the research of melodic intonation therapy (MIT). This seems to be an emerging topic in the field of speech-language pathology. When conducting a PubMed search of ‘aphasia’ there were 16,021 hits, but when searching ‘melodic intonation therapy’ there were only 46 hits. I also utilized resources such as Google Scholar and the Michigan State University library to find more research. I was not able to find any systematic reviews or meta-analyses related to melodic intonation therapy.
MIT is a type of therapy that uses different intonations and rhythms to help restore language (Van Der Meulen, Van De Sandt-Koenderman, Heijenbrok, Visch-Brink, & Ribbers, 2016). Several multi-syllable words and phrases have been separated into three different levels of treatment, each level becoming more difficult and using different amounts of therapist support. Each syllable receives a different pitch depending on the syllable’s stress. In addition, the patient will tap their left hand for each present syllable. Research throughout history has suggested that patients with damage to the left-hemisphere are capable of singing words even if they cannot speak them. This is related to centers in the brain that can process music that are left preserved in the right-hemisphere. MIT is typically used for patients with severe, non-fluent aphasia (Norton, Zipse, Marchina, & Schlaug, 2009). Broca’s aphasia falls into this category of aphasia and will be seen as one of the inclusion criteria for several research studies related to MIT (Aphasia Definitions, n.d.).
Van Der Meulen, et al. (2016) conducted a randomized control trial on 17 patients that had been diagnosed with chronic aphasia. This type of study would be of the highest study design strength due to the manipulation of the independent variable and the randomization of patients into an experimental or control group. There is one independent variable with two levels involved in this study, the time of when the therapy was received. The type of therapy received is broken down into receiving six weeks of MIT at a dosage of five hours per week(experimental) and receiving no individual therapy for six weeks (control). The dependent variables of this study focused on improvement of trained items (operationally defined as the performance on the MIT repetition task for trained items), improvement on non-trained items (operationally defined as performance on the MIT repetition task for untrained items and performance on the Aachen Aphasia Test (AAT) repetition subtest), and generalization to functional language use (operationally defined as performance on the AAT subtest of naming, performance on the Amsterdam-Nijmegen Everyday Language Test (ANELT), and the Sabadel story retell task).
New measurements were then taken for each group after the second six weeks in which the experimental group did not receive any individual therapy and the control group did receive MIT. The results of this study showed that in the first six weeks of the study the experimental group had a statistically significant improvement in the MIT tasks of trained (p<.01) and untrained items (p=.03). In addition, after the experimental and control group both received MIT, they both had a statistically significant increase of scores for the MIT tasks of trained items (p<.01) and untrained items (p<.01) (Van Der Meulen, et al., 2016). I chose this research article because of the fact that it was a randomized control trial with thought out variables. Due to the small sample size, it would be difficult to generalize to the population of all patients with aphasia, but I believe it could be the foreground for several future research studies.
Zumbansen, Peretz, & Hébert (2014) conducted a research study to observe whether rhythm and pitch both play a vital role in the delivery of MIT. Three participants with chronic Broca’s aphasia caused by a stroke were involved. There were two independent variables present, one multivalent and one bivalent. The first independent variable of therapy type was broken down into the levels of melodic therapy, rhythm therapy, and normally spoken therapy. Melodic therapy had aspects that involved both pitch and rhythm, rhythm therapy only involved rhythm, and the normally spoken therapy had zero melodic qualities. The second independent variable was the time of measurements, pre- and post-treatment. The dependent variables to be measured were the MIT task of non-trained stimuli, the MIT task of trained stimuli, connected speech, motor-speech ability, and mood. The MIT task of trained stimuli and non-trained stimuli were operationally defined as the number of correct syllables in trained and non-trained sentences. The trained sentences were used every session, whereas the untrained sentences were new at every session. Connected speech was operationally defined as discourse informativeness as measured by collecting speech samples of participants describes stick figure characters carrying out everyday activities. Motor-speech ability was measured by the patients diadochokinetic rate subtest of the Apraxia Battery for Adults (ABA-2). Finally, the participant mood was measured by the Visual Analog Mood Scales (VAMS).
This was a within-subject designed study where each participant engaged in each level of the independent variables. The research design would be considered a small-N experiment due to its small sample size and manipulation/randomization of the independent variable. This causes the design to be high strength, but with skepticism. The subjects received each therapy for two weeks at a time over a six-week period. These therapies were delivered for one hour, three times a week. Each of the three participants were assigned their therapy order randomly. Each individual was scored four times, before and after each new therapy. All three participants only showed a statistically significant advancement after receiving melodic therapy, including rhythm and pitch. Participant I with a p-value of .036, Participant II with a p-value of .044, and Participant III with a value of .024 (Zumbansen, et al., 2014). I chose this study because I found the small sample size to be interesting. Although it would be impossible to generalize these results to any population, there is some strong evidence suggesting that using pitch and rhythm together could be beneficial for clients with Broca’s aphasia. This study could be a great stepping stone for future research.
Tabei, et al. (2016) conducted a single-subject experiment in which one patient received a version of MIT that was modified to be used with Japanese. All of the qualities of MIT were present, but some of the tonal qualities were changed due to the nature of the language. The subject of this study was one 48-year-old man that had been receiving therapy with no success for several years in regard to his chronic non-fluent aphasia. The independent variables of this study were the reception of MIT and measurements pre- and post- treatment. This man received a rigorous nine-day therapy block where he received 45 minutes of therapy every day. Measurements and a functional Magnetic Resonance Imaging (fMRI) were taken at the end of this nine-day period. The dependent variables of this study included language output, auditory comprehension, picture naming response time, and fMRI images. Language output and auditory comprehension were measured by performance on the Western Aphasia Battery (WAB), picture naming response time was measured by naming pictures on a computer screen with words chosen from the Aphasia Quotient (AQ), the fMRI images were taken during the picture naming task. There were notable improvements across all WAB fields in regard to language output and auditory comprehension. The result with statistical significance (p=.049) was the decrease in picture naming response time. Although a qualitative measure, there were also changes in the fMRI images. During the picture naming response, the images showed less activation in the right hemisphere of the brain when naming the pictures incorrectly. The researchers suggest that these results could mean that the client was able to process these pictures more efficiently following MIT. This study could be considered to be of high strength due to the manipulation of an independent variable, but would be impossible to generalize to a population. I chose this study because of its use of fMRI images. This study would be incredibly expensive to perform with several subjects, but I would be interested to see if these results could be replicated.
Many studies use MIT for patients with chronic non-fluent aphasia. Morrow-Odom & Swann (2013) were curious to see if the success of MIT could be applied to patients with global aphasia after experiencing a stroke in the right hemisphere. This was a single-subject experiment involving a 65-year-old woman who suffered from a hemorrhagic stroke in the right hemisphere of her brain. This woman would receive an intensive treatment regimen where she went to therapy for seven successive weeks, five times a week, and for two and a half hour sessions. The independent variables of this study were the reception of treatment and measurements pre- and post-treatment. The dependent variables measured included communication abilities, functional communication, and quality of life. Communication abilities were measured by performance on the Aphasia Diagnostic Profiles (ADP). Functional communication was measured by performance on the American Speech-Language-Hearing Association Functional Assessment of Communication Scale (ASHA-FACS) and quality of life was measured by the Stroke and Aphasia Quality of Life Scale-39 (SAQOL-39), both of which were administered at home by the client’s spouse. Results showed that the client improved over several of the measurement subtests. Over the seven-week process MIT scores increases and there was an improvement in language processing, mean length of utterance, use of gestures, etc. The at-home assessments also showed improvements of functional communication and quality of life, but the possible bias in these assessments should be interpreted with caution. Overall, the researchers suggest that use of MIT could be a viable treatment approach for patients with global aphasia that meet specific criteria. I chose this study because of its reach beyond non-fluent Broca’s aphasia. I found it interesting to compare and contrast this study to the study conducted by Tabet, et al. The fact that it a single subject study gives it a high strength, but it is impossible to generalize to a population.
Schlaug, Marchina, & Norton (2008) conducted a small-n experiment involving two patients with severe non-fluent aphasia following a left-hemisphere stroke. Both patients were diagnosed with Broca’s aphasia. The independent variables of this study were the type of therapy being received and measurements pre- (at session 1 and 4), middle (after session 40), and post-treatment (after session 75). The patient in the experimental group received MIT and the patient in the control group received Speech Reception Threshold (SRT) therapy through the first 40 sessions and then received MIT for the 35 remaining sessions. The patients received therapy one and a half hours a day, five days a week, for 75 sessions (15 weeks). The dependent variable of this study was spontaneous speech, confrontational picture naming tasks, and fMRI imaging. This variable was measured by a conversational interview and descriptions of complex pictures. The conversational interview touched on topics such as biographical data, daily activities, etc. The descriptions of complex pictures were measured by calculating the number of correct information units (CIU) per minute and the number of syllables per phrase. The confrontational picture naming tasks were measured by performance on the Boston Naming Test (BNT) and Snodgrass-Vanderwart color pictures. Results showed that the patient in the experimental group started to show improvement in spontaneous speech and confrontational picture naming after 40 sessions and continued to show significant improvement after 75 sessions. This patient also showed prominent right-hemisphere activation, as well as activation at several points in the left-hemisphere in their fMRI. The patient in the control group showed improvement across speech scores through the 40 sessions, but the scores were not as high as the MIT patient. The control fMRI showed more prominent left-hemisphere activation. Following these 40 sessions the control patient was then enrolled into a 35 session MIT program. This program showed speech output and picture naming continued to increase. The researchers in this study suggest that MIT may target structures in the right-hemispheres that can compensate for damage/lesions caused by left-hemisphere strokes. I chose this article to compare its methods and results with other small-n studies done related to MIT and the study’s use of fMRI. This study was a small-N experiment that used randomization and manipulation of a variable. This study has high strength evidence, but would be difficult to generalize to a population due to its small subject size.
Schlaug, Marchina, & Norton (2009) conducted a comparative study in order to observe changes in the white-matter tracks of individuals with severe, chronic Broca’s aphasia caused by a left hemisphere stroke. This study included six patients that were chosen from previous MIT studies. The independent variable of this study included the reception of intensive MIT. The dependent variable was the images created by using a Magnetic Resonance Imaging (MRI) with diffusion tensor imaging (DTI). DTI allows the MRI to capture images of the size, shape, and orientation of the white matter tracts in the brain (Alexander, Lee, Lazar, & Field, 2007). The white matter tract that was of particular interest in this study was the arcuate fasciculus (AF), the tract that connects auditory and speech structures in the frontal, temporal, and parietal lobes. The imaging can be operationally defined as the number of AF fibers and the volume of AF fibers. Imaging was taken before and after 75 daily therapy sessions. Due to the nature of all of the patients’ strokes the imaging data analysis was confined to the right-sided AF. Results showed a statistically significant increase in the absolute number of AF fibers in all six patients (p=.04). Researchers suggest that the use of MIT may have an effect on the brain’s white matter tract and its ability to compensate for trauma within the brain using its own plasticity (Schlaug, et al, 2009). I chose this study because it had a larger number of patients than the many single subject and small-n studies related to MIT. In addition, I find studies that utilize brain scans intriguing. This study was a comparative study due to its lack of randomization causing it to have a high strength of evidence. It would be interesting to see a similar study replicated on a larger number of individuals to help with generalizability.
Research related to MIT has been present for years. While collecting articles it appears that there has been a fluctuation in more published research and the use of MRI to track the brain structures related to MIT. It is almost impossible to generalize any of the research above to a population due to the number of small-N and single subject experiments, but results seem to be consistent across the board. Many researchers suggest that the use of MIT can help patients that fit a specific criterion of severe, non-fluent aphasia improve their expressive language. It will be interesting to see future research and where the development of MIT will go from here.