• Korean Journal of Cognitive Science
• /
• v.33 no.2
• /
• pp.109-133
• /
• 2022

Episodic memory consists of a core event and the associated contexts. Although the role of the hippocampus and its neighboring regions in contextual representations during encoding has become increasingly evident, it remains unclear how these regions handle various context-specific information other than spatio-temporal contexts. Using high-resolution functional MRI, we explored the patterns of the medial temporal lobe (MTL) and cortical regions' involvement during the encoding of various types of contextual information (i.e., journalism principle 5W1H): "Who did it?," "Why did it happen?," "What happened?," "When did it happen?," "Where did it happen?," and "How did it happen?" Participants answered six different contextual questions while looking at simple experimental events consisting of two faces with one object on the screen. The MTL was divided to sub-regions by hierarchical clustering from resting-state data. General linear model analyses revealed a stronger activation of MTL sub-regions, the prefrontal lobe (PFC), and the inferior parietal lobule (IPL) during social (Who), behavioral (How), and intentional (Why) contextual processing when compared with spatio-temporal (Where/When) contextual processing. To further investigate the functional networks involved in contextual encoding dissociation, a multivariate pattern analysis was conducted with features selected as the task-based connectivity links between the hippocampal subfields and PFC/IPL. Each social, behavioral, and intentional contextual processing was individually and successfully classified from spatio-temporal contextual processing, respectively. Thus, specific contexts in episodic memory, namely social, behavior, and intention, involve distinct functional connectivity patterns that are distinct from those for spatio-temporal contextual memory.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.74-79
• /
• 2009

Fiducial markers have been used frequently in augmented reality applications. However, they represent only ID information to identify a marker, and thus they have limitations when we have multiple types of markers. In this paper, we propose a new marker design for augmented reality applications. We embed a marker’s context, such as type, size, orientation, and ID, in the barcode that has been used just for storing an ID. BY embedding the context information, it is possible to recognize multiple types of markers at once and to render corresponding contents with proper scale. Our marker design can be used in a mobile augmented reality environment where many unknown types of markers may exist.

• Annual Conference on Human and Language Technology
• /
• 1992.10a
• /
• pp.413-421
• /
• 1992

두 개의 실험을 통하여, 시간경과에 따른 한글 다의 단어의 의미처리과정을 알아보고자 하였다. 실험1에서는 의미를 편향시키는 맥락이 없는 상황에서 다의어의 의미처리를 알아보고자 하였는데, 결과는 사용빈도가 높은 의미의 활성화 촉진의 정도가, 빈도가 낮은 의미에 비하여 크고 오래 지속됨을 보여주었다. 실험 2에서는 다의어의 의미를 하나의 의미로 편향시키는 맥락을 사용하였는데, 맥락에 부합하는 의미의 반응시간이 부합하지 않는 의미에 비하여 빨랐다. 그리고 처음에는(SOA가 짧을 때) 일차 의미와 이차 의미의 활성화가 동시에 이루어지지만, 시간이 경과할수록 일차 의미의 활성화촉진은 이차의미에 비하여 크고 오랫동안 유지됨을 보여주었다.

• Korean Journal of Cognitive Science
• /
• v.18 no.4
• /
• pp.351-370
• /
• 2007

This study investigated whether top-down information influenced the filing-in of the blind spot. Two potential stimuli, which could fill in the blind spot, were presented at the location of the blind spot. When the two stimuli competed against each other to fill in the blind spot with equal probability, the depth-context of one of the stimuli was manipulated. We used an overlapping cue as our depth-context. Participants' task was to report the relative depth of a target presented in the blind spot. We found that the overlapping rue influenced the frequency of reported depth. These results, however, were nor observed in the off blind spot. These results suggest that top-down information such as an overlapping cue influences the filing-in of the blind spot.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.2
• /
• pp.79-86
• /
• 2024

In this paper, a novel visual tracking method which can utilize the context of object regions is presented. Conventional methods have the inherent problem of treating all candidate regions independently, where the tracker could not successfully discriminate regions with similar appearances. This was due to lack of contextual modeling in a given scene, where all candidate object regions should be taken into consideration when choosing a single region. The goal of the proposed method is to encourage feature exchange between candidate regions to improve the discriminability between similar regions. It improves upon conventional methods that only consider a single region, and is implemented by employing the MLP-Mixer model for enhanced feature exchange between regions. By implementing channel-wise, inter-region interaction operation between candidate features, contextual information of regions can be embedded into the individual feature representations. To evaluate the performance of the proposed tracker, the large-scale LaSOT dataset is used, and the experimental results show a competitive AUC performance of 0.560 while running at a real-time speed of 65 fps.

• Korean Journal of Cognitive Science
• /
• v.20 no.4
• /
• pp.555-571
• /
• 2009

This study investigated whether contexts could influence the speed of mental rotation. Experiment 1 investigated whether the angle of neighboring letters influenced the recognition of the target letter. Reaction time of target recognition increased linearly, as angular differences between the angle of the target letter and that of neighboring letters increased. Moreover, this trend was more pronounced when the target and neighboring letters made a word as compared to when they did not. Experiment 2 examined the effects of practice difficulty on mental rotation. Reaction time again increased linearly with the degree of angular rotation. However, there were no significant differences between the easy and the difficult conditions. These results suggest that contextual information is important in mental rotation.

• Korean Journal of Cognitive Science
• /
• v.22 no.2
• /
• pp.173-192
• /
• 2011

The visual system applies previously learned contextual knowledge to facilitate the perception and encoding of scenes. When extrapolation following the employment of contextual information occurs, it sometimes leads to scene memory distortion: people report as if they saw more than they actually had seen. This phenomenon is called the "boundary extension" effect (Intraub & Richardson, 1989). The present study aimed to clarify the effects of contextual information on boundary extension in a more systematic way. Based on the assumption that it is harder to extract contextual information from inverted scenes compared to intact scenes, we presented inverted scenes either during encoding or retrieval to manipulate the level of contextual information and compared the magnitude of boundary extension effect for upright versus inverted scenes. In a series of experiments, we found that scene inversion during encoding, but not during retrieval, significantly reduced boundary extension. Showing reduced memory distortion for inverted scenes, the current study directly demonstrated that access to contextual information is a critical component of scene extrapolation process.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.173-177
• /
• 2009

With the advent of light-weight daily physiological signal monitoring sensors, intelligent inference and analysis method for physiological signal monitoring application, commercialized products and services are released. However, practical constraints still remain for daily physiological signal monitoring. Most devices provide rough health check function and analyze with randomly sampled measurements. In this work, we propose the probabilistic modeling of physiological signal analysis. This model represent the relationship between previous user measurement (history), other group`s type, model and current observation. From the experiment, we found that the personalized analysis with long term regular data shows reliable result and reduces the analyzing errors. In addition, participants agree that the personalized analysis shows reliable and adaptive information than other standard analysis method.

• Annual Conference on Human and Language Technology
• /
• 1992.10a
• /
• pp.435-443
• /
• 1992

본 연구에서는 이해능력수준에 따른 억압기제의 효율성 차이를 대용어 참조 과정을 통해 검증하였다. 실험 1에서는 단어재인과제를 사용하여 이해능력에 따른 가능한 참조어의 활성화 차이를 살펴 보았다. 그 결과 낮은 수준의 이해자는 높은 수준의 이해자에 비해 가능한 참조어 중 문장맥락에 맞는 적절한 참조어와 함께 맥락에 맞지 않는 부적절한 참조어의 활성화도 유지하고 있는 경향이 있었다. 실험 2에서는 검사단어의 맥락적절성 판단과제를 실시하였는데 낮은 수준의 이해자는 높은 수준의 이해자에 비해 부적절한 참조어를 부정하는데 반응시간이 오래 걸렸다. 이 결과들은 낮은 수준의 이해자가 덜 효율적인 억압기제를 가졌다는 가설을 지지하는 결과로 논의되었다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.404-408
• /
• 2023

음성 사용자 인터페이스(Voice User Interface)에 대한 수요가 증가함에 따라 음성 합성(Speech Synthesis) 시스템에서 자연스러운 음성 발화를 모방하기 위해 적절한 위치에 휴지를 삽입하는 것이 주된 과업으로 자리잡았다. 대화의 연속성을 고려했을 때, 자연스러운 음성 기반 인터페이스를 구성하기 위해서는 대화의 맥락을 이해하고 적절한 위치에 휴지를 삽입하는 것이 필수적이다. 이에 따라 본 연구는 대화 맥락에 기반하여 적절한 위치에 휴지를 삽입하는 Long-Input Transformer 기반 휴지 예측 모델을 제안하고 한국어 대화 데이터셋에서 검증한 결과를 보인다.