The effects of five factors on the sensitivity of 36 women to infant cries were studied. The variables were listener experience (none, one month, four months), age of crying infant (one month, four months), type of cry (tired, angry, hungry), cry typicality (prototypical, moderately typical) and cry length (fixed, variable). A five-way analysis of variance was performed after scores were converted to the signal detection measure of sensitivity, d′. In addition to several main effects, significant age of infant × typicality × cry type, typicality × listener experience, and typicality × cry length interactions were found. The finding of an influence of cry length provided a possible explanation for the discrepancy in results in the infant cry literature, since the better discriminated single cries of variable length were typical of those cry signals used in studies reporting successful discrimination, and the poorly discriminated 15 sec fixed-length cry sequences were typical of those used in the studies failing to find discrimination of infant cries.

Mothers talked to their infants via a video system designed so that each partner saw a full-face, life-size image of the other on a screen before them, perfect eye-to-eye contact being possible. They were presented either with live, real-time video sequences of their infants, where communication was therefore potentially mutually responsive, or else with the same sequence replayed some minutes later. The mother's babytalk differed consistently between live and replay sequences. Since the infants' acts were identical in form in the two conditions, the only difference being the infant's potential responsiveness to the mother, these results are evidence for the infant's active role in contributing to the nature of interactions with adults.

Research is reported which investigated the ability of 4-month-old hearing infants to discriminate between gestures, derived from American Sign Language (ASL), that differ solely in terms of contrasts along underlying movement dimensions. Using an habituation–dishabituation task, it is shown that young infants can make such discriminations for global movement contrasts as well as contrasts along a single movement dimension. Thus, at four months, infants possess certain prerequisite perceptual abilities for the analysis of the signs of a visually specified language. Partial failure to dishabituate to another movement contrast, however, suggests that readiness to utilize the information in gestural events may vary depending on the movement in question. Results are contrasted to infant speech perception findings and implications for further research in the development of sign perception are discussed.

Young children's productions of novel words serving as names of objects and actions were compared. Although the object and action names were matched according to consonant and syllable structure, accurate production of newly emerging consonants was greater for the object words for each of the measures used: spontaneous production and unsolicited imitation during the word exposure sessions, and responses to production probes administered during and following the exposure sessions. These findings are attributed to the fact that action words have greater semantic complexity than object words, and that the increased processing requirements involved in referring to actions reduce the capacity to use newly developing sounds in these words.

Twenty-month-old children learned to recognize nonsense labels for five novel object concepts and were tested on generalization to variants of these concepts. Children were presented with either one or three examples of each object type during learning sessions. Results showed that receptive learning of names for object concepts was significantly related to a number of possible manipulations specific to each object type and to labelling by children. Children's generalization choices were consistent with adults' ranking of similarity of variants to concept prototypes. Children who learned less well were more likely to generalize to new instances of an object concept and to a greater number of variants if they had been exposed to three rather than one example during training sessions. Results also support the hypothesis that differentiation of objects in interaction is important to the formation of an object concept at this age.

Details of the linguistic modifications in speech to children are provided for the Mayan language Quiché. Quiché input is evaluated with respect to 17 features listed in Ferguson (1978). Eight additional features are noted for Quiché speech to children: whispering, initial-syllable deletion, BT forms for verbs, a verbal suffix that appears exclusively in speech to children, a relatively fixed word order with relatively fewer overt noun phrases, more imperatives, and a special interpretive routine. Quiché parents have a special register for speaking to young children. However, Quiché speech to children has only five of the features that Ferguson cites: repetition, BT forms for qualities, compound verbs, diminutives, and special sounds. This suggests that the features of speech to children are not universal, but are determined by the conventions for interacting with children in each community. Functional explanations of such features will have to take this degree of cultural variation into consideration.

This study compared the effects of three types of adult feedback (acceptance, correction with joint labelling, and correction with explanation) on young children's inappropriate word usage. Four children were visited in their homes twice a week, from the time they were between 1;1 and 1;3 until they were approximately 1;7. Differential feedback was applied by the experimenter whenever the children extended a term to referents that were inappropriate, but similar to appropriate ones. Comprehension and production probes were administered prior to, during, and following the feedback trials to assess changes in word usage over time. Results indicated that the three types of feedback varied in their relative effectiveness in facilitating a positive change in word usage. Correction with explanation was more effective than correction with joint labelling, which in turn was more effective than simple acceptance.

The acquisition of colour terms in pre-school-aged children was investigated in two studies. The first study explored the role of conceptual factors in 2-, 3- and 4-year-old children's performance on two tasks. Comprehension and production of basic colour words were tested using stimuli that varied in focality. The results confirmed earlier research demonstrating the important and early influence of focality on children's colour concepts and developing colour lexicon. However, only moderate support was obtained for the order-of-acquisition hypothesis proposed by Berlin & Kay (1969). Quite large discrepancies between comprehension and production were found in this study, with no clear relationship between these two performances. In the second study the role of maternal input on children's learning of colour words was investigated, using the spontaneous speech transcripts from Adam, Eve and Sarah. Significant correlations between mothers' and children's uses of specific colour words were found across all three subjects. The findings from both studies confirm that both conceptual and environmental factors are important in shaping the child's developing colour concepts and colour lexicon, particularly for mapping out the boundaries of the basic colour spaces, which are culturally, rather than innately, determined.

This study examined young children's responses to adult contingent queries. Each of 22 children in Language Stages II–V conversed alone with their mother and alone with an adult experimenter. The adults queried the child's multi-word utterances with either a neutral or a specific query. Children at all stages responded differently to the two types of query. In response to the neutral query children tended to repeat their entire utterance, whereas in response to the specific query they most often replied with only the asked-for information. Some children found it easier to differentiate the query types when their mother was the listener. These findings suggest that very young children can comprehend the linguistic structure of specific queries and that they can make pragmatically appropriate responses.

In deciding whether to use a definite or indefinite referring expression a speaker must apply a pragmatic rule: the definite article is usually appropriate if the listener is already familiar with the referent, and the indefinite article if not. Several studies have investigated at what age this rule is mastered by children. The most satisfactory procedure so far adopted is a task in which the subject must narrate a pictorially-presented story to another child who cannot see the pictures. It has been found that 5-year-old children already follow the correct rule quite well, except that they make ‘egocentric errors’ (i.e. using the definite article when first mentioning a referent) in around 15–35% of instances. Experiment 1 here confirms that this pattern of results is also obtained using Italian children. A problem with this experimental design is then raised: it does not exclude the possibility that subjects might produce the correct response distribution by following a rule based on the SPEAKER'S familiarity with the referent, not the listener's. To check this possibility, a modified design was used (Experiment 2) in which the subject had to narrate the story to two listeners, one after the other. Significantly more egocentric errors were made on the second narration of the story than on the first narration.