This paper describes a test of everyday speech reception, in which a listener's utilization of the linguisticsituational information of speech is assessed, and is compared with the utilization of acoustic-phonetic information. The test items are sentences which are presented in babble-type noise, and the listener response is the final word in the sentence (the key word) which is always a monosyllabic noun. Two types of sentences are used: high-predictability items for which the key word is somewhat predictable from the context, and low-predictability items for which the final word cannot be predicted from the context. Both types are included in several 50-item forms of the test, which are balanced for intelligibility, key-word familiarity and predictability, phonetic content, and length. Performance of normally heating listeners for various signal-to-noise ratios shows significantly different functions for low-and high-predictability items. The potential applications of this test, particularly in the assessment of speech reception in the hearing impaired, are discussed. . PACS numbers: 43.70.Ep, 43.50.Qp, 43.70.Ve tute the word, Some classes of sounds are more susceptible to masking by noise than others, s and consequently words containing these sounds are likely to be less intelligible than words containing sounds that are resistant to masking, In developing any test of speech intelligibility, therefore, care must be taken to select speech materials in which the phonetic content is properly balanced to reflect the distribution of speechsound classes that occur in the language, The acoustic attributes of sentences include not only the properties of phonetic units, but also prosodic parameters that signal the characteristics of larger units within the sentences. These consist of variations in the durations of sounds and in the fundamental frequency of voiced sounds. There is evidence that these prosodic 1337 : Test of speech intelligibility in noise parameters are used by a listener as cues for the understanding of sentences, since. they contribute information about stress and the grouping of words. B. Effect of sentence context The fact that in a noisy environment words in a sentence context are more intelligible than words spoken in isolation or without the benefit of sentence context has been demonstrated by Miller, Heise, and Lichten, 6 and by Miller. • These investigators argued that the sentence context imposes constraints on the set of alternative words that are available as responses at a particular location in a sentence, and noted that the intelligibility of words increases when the number of response alternatives decreases. This conclusion was supported and quantified further by Duffy and Giolas, 8 who examined the intelligibility of words in sentences in which the words had various demade to introduce some variety in the syntactic structure of the PH and PM sentences, subject to the various constraints noted above. For 95 words, two PH sentences were constructed.Since there was no obvious way to decide...

Performance of children aged 9 to 17 years on the SPIN test (Speech Perception in Noise) is described. The 11- and 13-year-olds performed significantly poorer than 15- and 17-year-olds, and this difference occurred primarily for high-predictability sentences presented at a O-dB signal-to-babble ratio. Performance of nine-year-olds was significantly poorer than performance of 11-year-olds. Possible reasons for these differences are discussed.

A four-alternative, forced choice adaptive procedure was used to measure the lowest intensity at which children could identify monosyllabic nouns that had been standardized to be understandable (at comfortable listening levels) to inner city, 3-year-old children. Results showed no age-related performance changes when the words were presented against a 12-talker babble or against filtered noise. In quit, however, performance improved between the ages of 5 and 10 years. Performance of children with learning problems was poorer than performance of children achieving normal school progress, even though clinical measures of auditory sensitivity showed no differences. Results are discussed in terms of "semantic closure" skills of children.

A forward-gating procedure, employing highly familiar monosyllabic words, was used in testing 5-7-year-old children, 15-17-year-old teenagers, and 70-85-year-old adults. Teenagers identified the words at shorter gate durations than either the children or older adults, whose identification performances were nearly identical. Teenagers gave meaningful guesses at shorter durations than children, who, in turn, gave meaningful guesses at shorter durations than adults. The oldest listeners provided the largest number of phonetic guesses, whereas teenagers gave almost none. Individual differences in auditory pure-tone sensitivity did not account for the results. It is hypothesized that both word frequency effects and temporal processing differences were responsible for the findings. 150In a study of speech perception of monosyllabic nouns, Elliott et al. (1979) found that 5-and 6-year-old children required higher intensity levels to identify words presented in quiet than did older children or adults, even though the stimuli were within the receptive vocabularies of 3-year-olds. Since an adaptive threshold-tracking procedure was used, stimuli were always presented at levels close to threshold. Elliott et al. suggested that listeners with higher levels of language skill might be more adept at identifying words from partial or limited acoustic information than subjects with less mature language development.The gating paradigm developed by Grosjean (1980) as a means of examining spoken-word recognition appears to offer a good means of further examining developmental changes in word recognition. In the gating paradigm, portions of words are presented, beginning at either the word onset (Cotton & Grosjean, 1984;Grosjean, 1980) or the word ending (Salasoo & Pisoni, 1985). For example, if 50-msec gates are used and forward gating is employed, one stimulus contains the initial 50 msec of a word, another contains the initial 100 msec, and so on. The listener's task is to identify the word, and this may require considerable guessing when the stimulus duration is brief. Grosjean (1980) used the gating paradigm to replicate influences of word frequency, word length, and sentence context on speech perception that were similar to those that had been reported by others.The gating paradigm provides one means of presenting limited acoustic information about a stimulus, since the first gates contain only the initial, incomplete portion of the word. Presenting speech stimuli at low-intensity

Two large groups of children—one progressing normally in school and the other exhibiting language-learning problems—were tested on a set of fine-grained auditory discrimination tasks that required responding to small acoustic differences. Discriminant analysis procedures, using only results for the auditory tasks, correctly classified nearly 80% of the 6- and 7-year-olds and nearly 65% of the 8- to ll-year-olds according to their school placements. Percentages of correct classifications increased to 87% and 75% when measures of receptive vocabulary (PPVT-R), receptive language (the Token Test for Children), and the Digit Span, Coding, and Block Design subtests of the WISC-R were also included in the discriminant functions. Results suggested that fine-grained auditory discrimination makes a major contribution to language learning, particularly in the early elementary school years.