Brown, Joe R.

Less than half of a series of 168 patients with multiple sclerosis presented significant speech deviations. The most frequent speech deviations observed were impaired loudness control and harshness; less frequently occurring deviations were defective articulation, restricted use of vocal variations for emphasis, poor pitch control, hypernasality, inappropriate pitch level, and breathiness. “Scanning speech” was not a prominent characteristic. Speech deviations were not solely attributable to cerebellar involvement. Their incidence was positively related to severity of neurologic involvement, becoming more prominent as additional motor systems were implicated.

Four groups of 10 patients, each group having a different neurogenic disorder of communication, were tested for impairment in 10 language categories. Five categories—auditory comprehension, adequacy of response, arithmetic, syntax, and naming—although impaired to some degree in all groups, did not sharply differentiate them. The most strongly differentiating disabilities in the four groups were aphasia—auditory retention span and fluency; apraxia of speech—fluency; confused language—relevance, reading comprehension, and writing of words to dictation; and general intellectual impairment—reading comprehension and auditory retention span (with preservation of relevance). The groups differed also in the onset and duration of the communication difficulty, and in the nature and locus of the neurologic problem.

Thirty-second speech samples were studied of at least 30 patients in each of 7 discrete neurologic groups, each patient unequivocally diagnosed as being a representative of his diagnostic group. Three judges independently rated each of these samples on each of 38 dimensions of speech and voice using a 7-point scale of severity. Computer analysis based on the means of the three ratings on each patient on each dimension yielded results leading to these conclusions: (1) Speech indeed follows neuroanatomy and neurophysiology. There are multiple types or patterns of dysarthria, each mirroring a different kind of abnormality of motor functioning. (2) These patterns of dysarthria can be differentiated; they sound different. They consist of definitive groupings of certain dimensions of speech and voice, deviant to distinctive degrees. (3) Five types of dysarthria were delineated: flaccid dysarthria (in bulbar palsy), spastic dysarthria (in pseudobulbar palsy), ataxic dysarthria (in cerebellar disorders), hypokinetic dysarthria (in parkinsonism), and hyperkinetic dysarthria (in dystonia and chorea). Also, a mixed dysarthria combining elements of flaccid and spastic dysarthrias was identified in amyotrophic lateral sclerosis. (4) Observed occurrence of a single dimension uniquely in a given neurologic disease and distinctive co-occurrence of several dimensions can aid diagnostically in identification of neurologic disorders.

Correlation matrices were used to demonstrate co-occurrence of deviant speech dimensions observed. Application of this technique led to the emergence of eight distinctive clusters of dysfunction. Each of seven neurologic disorders studied had its own unique group or pattern of clusters. However, any one cluster emerged in and was shared by more than one disorder. Inspection of the dimensions present in a cluster permitted giving the cluster a logically determined name based usually on the defective physiology responsible for the cluster. Knowledge of the neuromuscular characteristics of each disorder led to deductions concerning the neuromuscular substrate for each cluster. Intercluster correlations yielded clues concerning co-occurrence of certain neuromuscular defects. Further inspection led to identification of the probable neuromuscular bases of individual deviant speech dimensions. These conclusions may serve as hypotheses for more accurate physiologic and neurophysiologic measurements to further delineate the problems of dysarthria.