How do the various theories of stuttering facilatate our therapeutic approach?

Abstract

Considering the current ascendency of learning theories in the treatment of behavior disorders, I believe that the question posed might be rephrased by many as “Are theories of stuttering really necessary?” More specifically, the question might be framed as “What evidence exists that stuttering behaviors differ from other behaviors that have been demonstrated to be under operant control?” To suit those favoring a two-factor theory of learning, the last question could be extended to include control by classical conditioning or desensitization. In these forms the question at issue is not whether theory plays a guiding role in therapy, but what sort of theory is appropriate in the treatment of stuttering. The latter is an interesting question while the former is a trivial one. Almost certainly theory of some sort, however unsystematic or to whatever extent implicit, guides therapeutic practice. Pure empiricism is a myth in treatment and research. The issue of current interest is whether the guiding theory should be one of the general theories of learning or should be some special theory that distinguishes stuttering behaviors from other behaviors and gives specific therapeutic guidance beyond that issuing from general theories of learning. The answer that I propose to offer is that, at the present time, the treatment of stuttering cannot be approached adequately without resort to special theory. This answer should not be construed to imply that stuttering will never be embraced by general learning theory, but is simply a recognition that the facts of stuttering cannot be accounted for by present learning theories. Studies of normal speakers have shown that speech disfluencies come under operant control with the attachment of rewarding and aversive contingencies (Hill, 1954; Savoye, 1959; Flanagan et al., 1959; Stassi, 1961; Stevens, 1963; Siegel and Martin, 1965a, b, 1966, 1967, 1968; Brookshire and Martin, 1967; Martin and Siegel, 1969; Siegel et al., 1969). When one turns to stuttering behaviors, however, several studies have failed to demonstrate a reduction in stuttering in the presence of negative contingent stimulation (Frick, 1951; Stevens, 1963; Daly, 1967, 1968; Timmons, 1966). Other studies have shown that when stuttering behaviors are differentiated, some behaviors will increase while others will decrease with negative contingent reinforcement (Martin et al., 1964; Quist, 1966; Webster, 1968; Starkweather, 1970). Such findings suggest that