During a German air raid over Oslo in 1941, shrapnel struck a 30-year-old native Norwegian woman on the left side of her head. The damage caused right-sided hemiplegia, Broca's type aphasia, and a seizure disorder. Within a year, her language improved but her speech had an altered rhythm and melody suggesting a foreign accent (although she had never traveled outside of the country). The respected neurologist Monrad-Krohn described this incident in a detailed 1947 case report. The study is one of the most frequently referenced cases of what is now referred to as "foreign accent syndrome" (FAS). Since this famous case, approximately 40 individuals with FAS have been described in world literature. Reported accent changes include Japanese to Korean, British-English to French, American-English to British-English, and even Spanish to Hungarian.
The concept that a foreign-sounding accent may emerge after brain damage has puzzled both the public and the medical community. A few high-profile news reports have recently aired on American TV, but there remains a "gee-whiz" quality to this rare disorder. FAS is only briefly mentioned in motor speech textbooks and is not recognized in the Diagnostic and Statistical Manual of Mental Disorders, IV. As one of our recent patients remarked, "No one knows about it, so no one helps."
FAS is often caused by a stroke, although traumatic brain injury or multiple lesions can also lead to this disorder. FAS can be of unknown etiology, and multiple sclerosis has been listed as a possible factor in a reported case. Stroke-related damage usually occurs within a network of brain structures associated with speech production, including the left-hemisphere Broca's area, pre-motor and motor areas, and the basal ganglia. However, at least two reports describe FAS associated damage to the right-hemisphere, a part of the brain known to play a role in speech prosody.
As the name suggests, FAS is the sudden and unexpected appearance of a seemingly "foreign" accent. Although some features of FAS speech may resemble those of other neurogenic disorders (e.g. aphasia, dysarthria, and apraxia of speech), patients with FAS do not usually sound pathological to the average listener. Instead, they are commonly perceived as non-native English speakers. For instance, one of our patients, a native New Yorker, was so often mistaken for being Swedish that her doctor jokingly suggested she call herself "Olga."
Each case of FAS presents uniquely, but most cases share a core set of prosodic and segmental changes (see Moen, 2000, for details). For each patient, errors seem fairly predictable, giving the sense of a consistent "accent." The typical patient has good comprehension, but notably altered speech production. Prosody is impaired at the word and sentence level, and there is usually a tendency toward "isochrony" (equal and excessive stress at the syllable level). However, there are reports of more fluent individuals with FAS who present with relatively preserved global prosody (sentence level intonation).
Segmental changes are those that affect individual consonants and vowels. For individuals with FAS, consonants may be altered, substituted, or deleted. Voicing changes also occur. Complex clusters may present particular difficulty. For American-English speakers with FAS, a hallmark symptom appears to be difficulty realizing the alveolar tap/flap following a stressed syllable. Instead, patients may produce equally stressed syllables with an intervening stop (e.g., "Betty").
Vowel changes include distortion, prolongations, and substitutions. FAS vowel substitutions have particularly puzzled researchers. For instance, one of the subjects we have observed frequently substitutes the low back vowel /α/ for the target vowel /ε/, a mid-front vowel. Thus, "yeah" is pronounced "yah" (/jα/), contributing to the impression of a Swedish accent. This type of sound substitution is quite different from those of individuals with aphasia or apraxia of speech (in which a substituted sound is typically one phonetic feature away from its intended target). Nevertheless, individuals with FAS also insert epenthetic vowels into blends (e.g., realizing "strike" as ), a pattern also observed in apraxia of speech (AOS).
Adding to the mysterious nature of the disorder, evidence suggests that although some cases evolve, many do not. A few documented FAS cases have completely resolved within two months without therapeutic intervention. The evolving quality prompted Berthier and colleagues (1991) to follow recovery in four cases with varied lesion size and location. Of the four subjects studied, two with relatively small lesions in the premotor cortex, an area known for its role in speech prosody, showed gradual improvement over two months. The other two subjects had substantial damage to the precentral gyrus, an articulation area. These individuals had cases that persisted more than a year. These data suggest a potential effect of lesion size. Functional brain imaging will play an important future role in delineating both the brain states responsible for FAS and for its recovery.
Assessment & Treatment
A team of professionals should collaborate in the diagnosis of suspected FAS. An ideal team would include a speech-language pathologist, a neurologist for medical management, a radiologist to interpret brain computed tomography or magnetic resonance imaging, a neuropsychologist to evaluate cognitive function, and a clinical psychologist to aid in vocational, social, and familial adjustments.
In our lab, we are working closely with four suspected cases of FAS. Our assessment includes a complete case history (language exposure, education, family history, medical history); an oral mechanism evaluation; and standardized aphasia, dysarthria, and AOS examinations. An important first task is to rule out psychological or psychiatric involvement. While some mood disorders can affect speech quality, FAS accent changes do not result from psychological or psychiatric problems. It is critical to attend to this issue for accurate diagnosis. For example, two of our patients have been evaluated for Conversion Disorder, a mind/body deficit in which an individual's unconscious psychological state leads to physical changes. In typical conversion cases, the patient derives a type of "secondary gain" from the disorder, such as finding an unconscious solution for a difficult life situation. However, the patients we evaluated could not easily change their "foreign" accent and were quite distressed about having it. There appeared to be no secondary gain. Also, the accents persisted even under conditions of delayed auditory feedback (DAF), which reduced the likelihood that the accents were maintained through conscious effort.
Patients undergo rigorous speech testing. We record the "Grandfather Passage" over multiple meetings and gather both spontaneous and repeated speech samples. The samples are reviewed for errors in consonant and vowel productions, and analyzed for error consistency. We have also assembled a prosody battery that includes contrastive stress placement in compound nouns), tonic stress in sentences ("The horse jumped the fence" vs. "The horse jumped the fence"), noun/verb contrast in sentences ("Let's present the girl with her present"), probes for sentence level intonation ("I have a red pencil box," "Is today Thursday?"), and materials that assess affective (happy/angry/sad) speech prosody. Because of the prevalence of flap/tap errors in American -English speakers with FAS, we also include a list of polysyllabic words with alveolar stops in a post-stressed environment ("daddy, "patio").
The benefit of speech treatment in FAS remains largely unknown. Anecdotal reports from patients in our laboratory and in others (e.g., Blumstein et al., 1987) suggest speech therapy is beneficial. Two of our patients have received services from SLPs. One patient reported benefit from stress and timing work, using a pacing board to improve word- and sentence-level prosody. Another reported using a mirror to help guide lip and jaw placement during speech.
We are currently completing a single-subject, multiple-baseline treatment study of vowel and consonant articulation with the help of an SLP with experience in accent modification services. The preliminary results indicate improvement on the targeted behaviors. The data also suggest that minimal pair work on selected consonant and vowel targets can increase awareness of speech production and improve articulatory accuracy. We tentatively conclude that techniques from accent modification work are feasible for use in speech-directed services with patients with FAS.
Although FAS is a little-understood adult neurogenic speech disorder, the syndrome may hold the promise of future understanding concerning the functional and neural organization of speech production. Many questions remain, including: Is FAS a true disorder in its own right or a phenomenon in the ear of the listener? Do subtypes of this disorder exist? Is the perceived accent primarily a prosodic deficit? What are its neural bases?
As SLPs, we should recognize that we are an important resource in diagnosing and treating this unusual disorder. If one encounters suspected FAS, it is important to be flexible in applying a variety of speech remediation techniques, and to be responsive to the individual needs of these intriguing cases.