One of the most interesting behavioral phenomena in neurology is the astounding ability of some people with severe aphasia to sing words in song before they can speak them normally. Aphasia is the language impairment that results from acquired damage to the “zone of language,” typically located in the left cerebral hemisphere, within the distribution of the middle cerebral artery, which feeds portions of the frontal, parietal, and temporal lobes. Depending on the location of the damage within the language zones, different types of aphasia can occur.
Most aphasia types can be categorized as fluent or nonfluent, according to the relative ability of the person to produce words and sentences. A severe type of nonfluent aphasia was first described by Paul Broca, whose patient could produce only the utterance “tan-tan” despite having the ability to understand the speech of others. This type of aphasia is associated with damage to Broca’s area, the left lateral frontal, pre-Rolandic, suprasylvian area of the brain, with extension of the lesion into subcortical white-matter pathways. Not all people with Broca’s aphasia have a limited stereotypic expression such as tan-tan, but those with severe Broca’s aphasia often do.
In their book, “Gabby: A Story of Courage and Hope,” former U.S. congresswoman Gabrielle Giffords and her husband, retired astronaut Mark Kelly, relate that early on in her recovery from the bullet wound that traversed her language zone, the congresswoman perseverated on “chicken-chicken.” One of the first language therapy methods used with Ms. Giffords was melodic intonation therapy, described below. One of the reasons this method was employed is that, early on, Giffords’ clinicians discovered their patient could sing many of the words of well known songs, including the lyrics to “The Sun Will Come out Tomorrow” from the musical “Annie,” in which she starred in the fifth grade.
The first published observation that a person with severe nonfluent aphasia may be able to produce words in song was published by Swedish nobleman Olof von Dalin in 1745. Von Dalin described the case of a man who, at age 33, was struck with a violent illness that left him with severe right hemiplegia and a complete loss of the ability to speak. He eventually went to a mineral spring for a “cure,” and there he was invited to stay with the local vicar and his family. The man’s gait became steadier, but he continued to be unable to communicate verbally except for the word “ja” (“yes”). In church, however, he could easily sing words of hymns. This startling ability led the vicar and villagers to question whether he truly could not produce purposeful speech, or if he was perpetuating a hoax to receive free room and board. Eventually, however, it became clear that the man, indeed, could only produce words in song.
In his landmark paper on the treatment of aphasia in 1904, American neurologist Charles Mills stated that because some people with aphasia can sing before they can talk, it might be beneficial to play the piano and encourage these individuals to sing popular songs. Clinical experience subsequently showed that although singing well known songs may beneﬁt the person psychologically, there was little evidence that this activity resulted in improved propositional speech skills.
In 1945, Ollie Backus suggested that a better method might be to present useful words and phrases to people with aphasia in a rhythmic, unison fashion. It was not until 1972, however, that formal study of such an approach was undertaken. I had just begun working at the renowned Aphasia Unit at the Boston Veterans Administration Hospital. The first person with aphasia assigned to me for evaluation was a 46-year-old former army nurse who had sustained a left-hemisphere stroke, resulting in right hemiplegia and severe nonfluent aphasia. On every speech task, her verbal output was completely limited to the stereotypic, nonsense utterance “nee-nee-nah-nah.” She produced this stereotypy effortlessly, often adding prosodic contours and emphasis to convey intent. For example, she used intonation to indicate questioning (“nee-nee-nah-NAH?”), or disagreeing (“nee-nee-NAH-nah!”). Although her auditory comprehension was relatively good, she showed no recognition of her production problem, and therefore never tried to inhibit it.
Shortly after the exam, I observed the former army nurse singing along with a volunteer piano player, producing most of the words of well known songs. Being struck by this dichotomy between her speaking and singing, I fetched my colleagues Martin Albert, a behavioral neurologist, and Robert Sparks, the senior speech-language pathologist, and proposed that there had to be a way that we could use this ability to produce words in song as a springboard for treating propositional speech.
We then began a series of experimental trials to develop a method based on musical intonation. The resulting method, melodic intonation therapy, or MIT, is a hierarchically structured program that is divided into three levels. In the ﬁrst two levels, multisyllabic words and short, high-probability phrases are musically intoned. The third level introduces longer or more phonologically complex sentences that first are intoned, then produced with exaggerated speech prosody, and, ﬁnally, spoken normally. On all intoned phrases, the clinician assists the person with aphasia in tapping their left hand once for each syllable. Items are intoned slowly, with continuous voicing, using simple high-note/low-note patterns based on the normal speech prosody of the phrases.
In 1973, my aforementioned colleagues and I published the ﬁrst description of the effects of MIT on three people with severe nonfluent aphasia in Archives of Neurology. In 1974, we published a study in Cortex describing the effects of MIT on eight individuals who had shown no improvement in verbal expression despite other forms of therapy. The popular press picked up on MIT, and this led to its widespread use and misuse. In 1994, the American Academy of Neurology subcommittee for assessing therapeutic and technological methods recruited a panel of experts to review the safety and effectiveness of MIT. Among the conclusions of the panel were that MIT is best administered by a speech-language pathologist experienced with the method; that it should be given in short, frequent sessions during a limited time span of three to six weeks; and that MIT is appropriate for some individuals with severe Broca’s aphasia but not for individuals with other forms of aphasia.
In considering the mechanisms for improvement with MIT, we posited in our Cortex paper that “it is unlikely that the undamaged right hemisphere, which is the minor one in our subjects, is suddenly starting to speak for itself.” Instead, we conjectured that the right-hemispheric dominance for melodic aspects of speech facilitates recovery of residual left-hemisphere speech skills. Since then, several studies have used functional neuroimaging techniques to examine the brain dynamics associated with MIT treatment.
The results of these studies have been mixed, probably depending on the techniques used and variations in the people studied. Although it is of interest from a scientific standpoint to identify the brain dynamics associated with rehabilitation of intentional speech with MIT, the clinical importance lies in the positive outcomes it may have with people with severely restricted verbal output due to nonfluent aphasia.
Nancy Helm-Estabrooks, Sc.D., has an ASHA Certificate of Clinical Competence for Speech-Language Pathologists and ANCDS Board Certification in Adult Neurologic Communication Disorders. She is the professor emeritus in the Department of Communication Sciences and Disorders at Western Carolina University. She is one of the three developers of melodic intonation therapy.