Dissociable Neural Mechanisms for Human Inference Processing Predicted by Static and Contextual Language Models.

Language models (LMs) continue to reveal non-trivial relations to human language performance and the underlying neurophysiology. Recent research has characterized how word embeddings from an LM can be used to generate integrated discourse representations in order to perform inference on events. The current research investigates how such event knowledge may be coded in distinct manners in different classes of LMs and how this maps onto different forms of human inference processing. To do so, we investigate inference on events using two well-documented human experimental protocols from Metusalem et al. (2012) and McKoon and Ratcliff (1986), compared with two protocols for simpler semantic processing. Interestingly, this reveals a dissociation in the relation between local semantics versus event-inference depending on the LM. In a series of experiments, we observed that for the static LMs (word2vec/GloVe), there was a clear dissociation in the relation between semantics and inference for the two inference tasks. In contrast, for the contextual LMs (BERT/RoBERTa), we observed a correlation between semantic and inference processing for both inference tasks. The experimental results suggest that inference as measured by Metusalem and McKoon rely on dissociable processes. While the static models are able to perform Metusalem inference, only the contextual models succeed in McKoon inference. Interestingly, these dissociable processes may be linked to well-characterized automatic versus strategic inference processes in the psychological literature. This allows us to make predictions about dissociable neurophysiological markers that should be found during human inference processing with these tasks.

Opinion attribution improves motivation to exchange subjective opinions with humanoid robots.

In recent years, the development of robots that can engage in non-task-oriented dialogue with people, such as chat, has received increasing attention. This study aims to clarify the factors that improve the user's willingness to talk with robots in non-task oriented dialogues (e.g., chat). A previous study reported that exchanging subjective opinions makes such dialogue enjoyable and enthusiastic. In some cases, however, the robot's subjective opinions are not realistic, i.e., the user believes the robot does not have opinions, thus we cannot attribute the opinion to the robot. For example, if a robot says that alcohol tastes good, it may be difficult to imagine the robot having such an opinion. In this case, the user's motivation to exchange opinions may decrease. In this study, we hypothesize that regardless of the type of robot, opinion attribution affects the user's motivation to exchange opinions with humanoid robots. We examined the effect by preparing various opinions of two kinds of humanoid robots. The experimental result suggests that not only the users' interest in the topic but also the attribution of the subjective opinions to them influence their motivation to exchange opinions. Another analysis revealed that the android significantly increased the motivation when they are interested in the topic and do not attribute opinions, while the small robot significantly increased it when not interested and attributed opinions. In situations where there are opinions that cannot be attributed to humanoid robots, the result that androids are more motivating when users have the interests even if opinions are not attributed can indicate the usefulness of androids.

A study of interactive robot architecture through the practical implementation of conversational android.

This study shows an autonomous android robot that can have a natural daily dialogue with humans. The dialogue system for daily dialogue is different from a task-oriented dialogue system in that it is not given a clear purpose or the necessary information. That is, it needs to generate an utterance in a situation where there is no clear request from humans. Therefore, to continue a dialogue with a consistent content, it is necessary to essentially change the design policy of dialogue management compared with the existing dialogue system. The purpose of our study is to constructively find out the dialogue system architecture for realizing daily dialogue through implementing an autonomous dialogue robot capable of daily natural dialogue. We defined the android's desire necessary for daily dialogue and the dialogue management system in which the android changes its internal (mental) states in accordance to the desire and partner's behavior and chooses a dialogue topic suitable for the current situation. The developed android could continue daily dialogue for about 10 min in the scene where the robot and partner met for the first time in the experiment. Moreover, a multimodal Turing test has shown that half of the participants had felt that the android was remotely controlled to some degree, that is, the android's behavior was humanlike. This result suggests that the system construction method assumed in this study is an effective approach to realize daily dialogue, and the study discusses the system architecture for daily dialogue.

A Model of Online Temporal-Spatial Integration for Immediacy and Overrule in Discourse Comprehension.

During discourse comprehension, information from prior processing is integrated and appears to be immediately accessible. This was remarkably demonstrated by an N400 for "salted" and not "in love" in response to "The peanut was salted/in love." Discourse overrule was induced by prior discourse featuring the peanut as an animate agent. Immediate discourse overrule requires a model that integrates information at two timescales. One is over the lifetime and includes event knowledge and word semantics. The second is over the discourse in an event context. We propose a model where both are accounted for by temporal-to-spatial integration of experience into distributed spatial representations, providing immediate access to experience accumulated over different timescales. For lexical semantics, this is modeled by a word embedding system trained by sequential exposure to the entire Wikipedia corpus. For discourse, this is modeled by a recurrent reservoir network trained to generate a discourse vector for input sequences of words. The N400 is modeled as the difference between the instantaneous discourse vector and the target word. We predict this model can account for semantic immediacy and discourse overrule. The model simulates lexical priming and discourse overrule in the "Peanut in love" discourse, and it demonstrates that an unexpected word elicits reduced N400 if it is generally related to the event described in prior discourse, and that this effect disappears when the discourse context is removed. This neurocomputational model is the first to simulate immediacy and overrule in discourse-modulated N400, and contributes to characterization of online integration processes in discourse.

Who Is Responsible for a Dialogue Breakdown? An Error Recovery Strategy That Promotes Cooperative Intentions From Humans by Mutual Attribution of Responsibility in Human-Robot Dialogues.

We propose a strategy with which conversational android robots can handle dialogue breakdowns. For smooth human-robot conversations, we must not only improve a robot's dialogue capability but also elicit cooperative intentions from users for avoiding and recovering from dialogue breakdowns. A cooperative intention can be encouraged if users recognize their own responsibility for breakdowns. If the robot always blames users, however, they will quickly become less cooperative and lose their motivation to continue a discussion. This paper hypothesizes that for smooth dialogues, the robot and the users must share the responsibility based on psychological reciprocity. In other words, the robot should alternately attribute the responsibility to itself and to the users. We proposed a dialogue strategy for recovering from dialogue breakdowns based on the hypothesis and experimentally verified it with an android. The experimental result shows that the proposed method made the participants aware of their share of the responsibility of the dialogue breakdowns without reducing their motivation, even though the number of dialogue breakdowns was not statistically reduced compared with a control condition. This suggests that the proposed method effectively elicited cooperative intentions from users during dialogues.

[Intrapulmonary Sewing Needle].

Intrapulmonary aberrant needles are rarely encountered in clinical practice. A 82-year-old woman, though she was asymptomatic, was referred to our department due to an abnormal shadow on a chest X-ray. Chest X-ray and chest computed tomography showed a foreign body suspected to be a sewing needle in the left upper lobe. The needle was successfully removed by video-assisted thoracoscopic surgery.

[Analysis of 76 patients with urticaria and angioedema induced by non-steroidal anti-inflammatory drugs (NSAIDs) in Japan].

BACKGROUND: The pathogenesis of urticaria and angioedema induced by non-steroidal anti-inflammatory drugs (NSAIDs) is still obscure. We analyzed the clinical characteristics of patients with NSAIDs-induced urticaria and angioedema without asthma in Japan. METHODS: We retrospectively collected the cases of NSAIDs-induced urticaria and angioedema from Japanese medical journals in 2000-2009. RESULTS: Seventy-six patients were analyzed. The male/female ratio was 1:2.5 and the mean age was 38.1 years. Urticaria was most frequent clinical manifestation in 3 groups; urticaria alone, urticaria and angioedema, and angioedema alone. Time interval from drug administration to onset was 5 minutes to 48 hours by aspirin at a dose of 25-1000 mg. Skin prick test was performed with aspirin in 33 patients, and the results were negative in all patients. Meloxicam, a selective cyclooxygenase-2 (COX-2) inhibitor, and celecoxib, a new selective COX-2 inhibitor, were administered safely in 4 of 6 patients and in 2 of 3 patients with NSAIDs-induced urticaria, respectively. These drugs were administered safely in all administered patients with NSAIDs-induced angioedema. Tiaramidehydrochroride (a basic COX-1 inhibitor) was safely used in 23 administered patients with NSAIDs-induced angioedema. Leukotriene receptor antagonists were effective in 2 of 5 patients administered, but aggravated symptoms in the others. CONCLUSION: Diversity of NSAIDs-induced urticaria and angioedema was shown in this study. Pathogenesis of NSAIDs-induced urticaria and angioedema without asthma seems to be different from that of NSAIDs-induced asthma.

[A case of latex allergy suspected drug allergy].

A 30-year-old female with paroxysmal supraventricular tachycardia (PSVT) underwent catheter ablation. About 30 minutes later, urticaria and dyspnea occurred suddenly. Blood pressure decreased to 62/41 mmHg, and she fell into the state of anaphylactic shock. She recovered within one hour following treatment. We initially suspected the onset of anaphylaxis was caused by either the local anesthetic or the intravenous antibiotic administered. Following thorough investigation (skin tests and challenge tests), we concluded that the anaphylaxis was not drug induced. Subsequently, we suspected latex allergy. Skin prick test showed a positive reaction to rubber gloves. The specific test for IgE antibody against latex was positive at 10.8 UA/ml. From these results, anaphylactic shock caused by latex (probably medical gloves) was diagnosed. Doctors should take preventive measures against latex allergy not only in operating rooms but also during minor treatments. It is possible that latex allergy is responsible for some cases of anaphylaxis of unknown origin.