A comparison of speech training methods with deaf adolescents: spectrographic versus noninstrumental instruction

The effects of speech training with real-time spectrographic displays (SDs) were examined and compared to the effects of noninstrumental (NI) instruction (i.e., training without computerized displays of speech) for deaf adolescents. A single-case modified alternating-treatment experimental design with replication across subjects and speech targets was used to examine within-subject performance in establishing, maintaining, and generalizing target consonants. Comparisons between the two approaches were accomplished by determining how frequently each method resulted in improvement, maintenance of improvement, and generalization to untrained words. Each of the 4 subjects demonstrated improvement under both forms of instruction in a relatively short time. Maintenance of improvement was observed 6 weeks post-treatment for two NI-trained targets and one SD-trained target. Two subjects showed better generalization for their SD-trained target than their NI target. There was little difference in generalization scores for the remaining subjects. All subjects either regained their highest previous levels of acceptability or maintained high-level acceptability following brief, independent practice with SDs 10 weeks after training was discontinued. The expediency of independent practice with SDs was discussed.