The influence of two different types of language therapy was investigated by assigning 32 patients to three groups: 10 patients were treated with a Systematic Therapy program for Auditory Comprehension Disorders (group 1), 11 patients received conventional stimulation therapy (group 2), and the other 11 patients received no treatment at all (group 3). Language recovery in the three groups was examined with a test battery consisting of two parts: Part I consisted of 3 subtests for auditory comprehension, which were also used as practice materials in the treatment program given to the patients of group 1; Part II contained 8 tests for auditory comprehension, reading comprehension, and oral expression, the items of which were not used as practice materials in either of the two treated groups. Although the influence of age, time post onset, and severity of the receptive disorder were controlled by multiple regression analysis, it was found that the differences between the three treatment conditions were far from significant on almost all evaluation tests of Part II, as well as on the evaluation tests of Part I. On the basis of these results it was concluded that neither the Systematic Therapy program for Auditory Comprehension Disorders nor the conventional stimulation therapy had any clear effect on the recovery process of the treated patients in groups 1 and 2. Some possible reasons for the failure to demonstrate significant differences in language gains between the three therapeutic conditions are discussed.

In the experiment reported here, subjects heard the beginnings of spoken words, followed by increasingly larger segments of word-onset information until the words could be correctly identified. The likelihood of word identification at any given point was found to be an inverse power function of the number of words in a dictionary count that began with the same sounds as the stimulus. Subjects' prerecognition responses consisted of words drawn from a wide range of word frequencies. The variance in frequency decreased with increasing amounts of word-onset information. Results are discussed in terms of word-initial phonology as a trigger for response activation.

The method of sorting is used to compare sets of adjectival contexts. Contexts of directly antonymous adjectives are found to be highly discriminable, both with sentential and phrasal contexts. These results are used to argue that words with different meanings normally appear in discriminably different contexts, and that the cue for learning to associate direct antonyms is not their substitutability, but rather their relatively frequent co-occurrence in the same sentence.