Understanding Some girls have bags is difficult for preschoolers, because they may not properly calculate scalar implica-ture. In this study, we examined whether preschoolers and adults guess scalar information without scale markers (some,all) using negative/positive Japanese sentences in picture selection task. This task consisted of three cards, e.g., SOMEcard illustrated two persons with a bag and four persons without a bag. The results were that the positive expression,Baggu wo motte-iru hito ga imasu (there are persons with a bag), was guessed as meaning of ALL (all persons have a bag)by children, while guessed as SOME by adults. Interestingly, the negative expression, Baggu wo motte-inai hito ga imasu(there are persons without a bag), was guessed as NONE by children, but guessed as SOME by adults. The results suggestchildren may not utilize the combination of existence and agents state information to guess scalar implicature.

An influential claim about co-speech gestures is that they are used as a supplement to unsaid meaning in speech. However, when the gesture is fully synchronized with speech, the supplementary role appears unnecessary. The present study examined whether people use gestures differently when they produce Japanese noun phrases that contain verbs. This study compared an ambiguous noun phrase “Rakka-shiteiru (falling) otoko-no (man’s) keitai (cell-phone),” which can be interpreted following the left branching (LB) structure as “The falling man’s cell-phone ({{falling, man}, cell-phone})” or the right branching (RB) structure as “The man’s cell-phone, which is falling ({falling, {man, cell-phone}}).” This study predicted that the onset of the first gesture would be delayed with the RB structure, as the important chunk {man, cell-phone} is produced later in the utterance than in the LB structure. The results supported the prediction, indicating that the onset of the first gesture tended to be delayed when RB was produced. This finding suggests that people may disambiguate syntactically ambiguous linguistic structures through gesture use.

This study aimed to investigate whether people produce spontaneous co-speech gestures that reflect the underlying linguistic structures and additional information when speech is restricted by prompt phrases. Participants were asked to convey information about an animated movie using a three-word phrase in Japanese that could be interpreted in two different ways depending on their underlying deep structures. The animated movie included or did not include important information that was not described by the prompt phrases. The results showed that most participants produced gestures while uttering the phrase, and the onset of the nouns reflected the underlying linguistic structures. A time-series analysis revealed that the occurrence of object-action gestures that depicted a noun’s movement tended to reflect the associated linguistic structures. People spontaneously produce gestures, which may syntactically and semantically help to disambiguate ambiguous phrases.

Conveying referential intention is essentially important to cooperate with others. It is reported that even adults some-times take ego-centric perspective (i.e., perspective that is based on one’s own perspective ignoring other’s perspective) incomprehending others utterances. In the present study we used a modified version of Keysars paradigm of 4x4 grid, andexamined whether the interpretation of the instruction by the addressee was affected by the directors use of two social-pragmatics aspects; demonstratives and gestures. Results showed if the director did not use a demonstrative and handpointing, the addressees interpreted the object from ego-centric perspective. In contrast, if the director used a demon-strative and hand pointing, the addressees correctly interpreted the referred object showing their use of the directorsperspective. The result suggested that demonstratives and hand pointing may promote the addressees interpretation basedon the directors perspectives.

Spoken language often includes ambiguity in meaning.Compounds such as “green teacup” can be interpreted with twodifferent meanings: “green colored teacup” and “cup for greentea.” We can assume there are two different underlyingsyntactic structures. Phonetic aspects have been studied in thedisambiguation process of such ambiguous phrases, but theroles of nonlinguistic information such as gestures have notbeen explored yet. We investigated whether people usegestures differently when they were asked to describe themeanings of Japanese compounds that can be interpreted astwo different meanings. We found that the timing of gesturesin relation to the target words of accompanying speech wasdifferent between right branching compounds and leftbranching compounds. Gestures seem to be used to suggestupcoming two words (adjective and noun) as a unit inbranching. Gestures can be a useful means to disambiguate themeanings of compounds.

The pointing gesture is regarded as indicating an object or location in the environment. People sometimes point to invisible objects, but the inferential mechanism is not known. This study examined comprehension of pointing with a bent index finger at an invisible object behind a wall. The experimenter pointed at an object using either typical pointing or “enforced pointing” behind a wall that was either opaque or transparent. In enforced pointing, the experimenter moved his arm in an arc movement. The participants guessed which object was being denoted. The wall was also either relatively high or relatively low. When the participants looked at typical pointing, they thought that objects both in front of the wall and behind the wall were being denoted. However, when they looked at enforced pointing, they more frequently thought that objects behind the wall were denoted. People seemed to use pragmatic knowledge on this “enforced” pointing gesture.