Search

Basic level advantage effect in human cognition suggested there was a more readily accessible cognitive status to the human mind than others during information retrieval, but the phenomenon of advantage effect switching to superordinate level suggested basic level would not always a sufficient way. Processing demands of tasks might explain the discrepancy between basic and superordinate precedence. In this study, we used word-picture matching task and picture-word matching task with same materials to investigate the neural systems of basic level advantage and its switching. Results showed that more activation in the area of fusiform and lingural gyrus in word-picture matching task, suggesting a visual perceptual processing loaded more in this task; while more activation in left inferior frontal gyrus in picture-word matching task, suggesting a semantic memory relied more in this task. These contrast analysis revealed different processing strategies across the two tasks which led to different advantage effect. Furthermore, the inferior parietal lobe played an important role for the advantage effect during information retrieval, with weakest deactivations appeared in the superordinate level in word-picture matching task and in the intermediate level in picture-word matching task, which might served as a control or centralized system to integrate all kinds of cognitive resources. [ABSTRACT FROM AUTHOR]

Problem representation is one of the key factors in problem solving. According to previous studies, PPC (Posterior Parietal Cortex) is critical for problem representation. Whether does problem expression form affect problem representation? What are the cognitive role of PPC in representation? In order to answer these questions, a fMRI experiment was performed in this study to examine the role of PPC in problem solving. It was a 2 × 2 designed experiment with two 2-level factors: task complexity (one-step and two-steps) and expression form (digits and symbols). In a digital task, 4 digits are provided in the initial grids, while 4 symbols of poker are provided in a symbolic task. In a task of one-step, participants only need one time rules retrieving to get the target answer, while in a task of two-steps, participants need two times rules retrieving to get the answer of target after getting the answer of a bridging location. The results of fMRI show that PPC activated significantly. The further analysis shows that there is a positive correlation between the activation intensity of PPC and task complexity, but the correlation between the activation intensity of PPC and task expression is not significant. According these results, we infer that PPC plays an important role in problem representation, maybe this representation is a high level abstraction. [ABSTRACT FROM AUTHOR]

One of the major problems for search at Web scale is that the search results on the large scale data might be huge and the users have to browse to find the most relevant ones. Plus, due to the reason for the context, user requirement may diverse although the input query may be the same. In this paper, we try to achieve scalability for Web search through social relation diversity of different users. Namely, we utilize one of the major context for users, social relations, to help refining the search process. Social network based group interest models are developed according to collaborative networks, and is designed to be used in more wider range of Web scale search tasks. The experiments are based on the SwetoDBLP dataset, and we can conclude that proposed method is potentially effective to help users find most relevant search results in the Web environment. [ABSTRACT FROM AUTHOR]

In this study, we explore the approach using Support Vector Machines (SVM) to predict the high-level cognitive states based on fMRI data. On the base of taking voxels in the brain regions related to problem solving as the features, we compare two feature extraction methods, one is based on the cumulative changes of blood oxygen level dependent (BOLD) signal, and the other is based on the values at each time point in the BOLD signal time course of each trial. We collected the fMRI data while participants were performing a simplified 4*4 Sudoku problems, and predicted the complexity (easy vs. complex) or the steps (1-step vs. 2-steps) of the problem from fMRI data using these two feature extraction methods, respectively. Both methods can produce quite high accuracy, and the performance of the latter method is better than the former. The results indicate that SVM can be used to predict high-level cognitive states from fMRI data. Moreover, the feature extraction based on serial signal change of BOLD effect can predict cognitive states better because it can use abundant and typical information kept in BOLD effect data. [ABSTRACT FROM AUTHOR]

In this paper, we present an investigation on heuristics retrieval in human problem solving by combining the computational cognitive model ACT-R (Adaptive Control of Thought-Rational) and advanced fMRI (functional Magnetic Resonance Imaging) brain imaging technique. As a new paradigm, 4*4 Sudoku is developed to facilitate this study, in which seven heuristics that can be classified into 3 groups are designed to solve two types of tasks: simple and complex ones. The cognitive processes of the two types of 4*4 Sudoku tasks are explored based on the outputs of ACT-R model. This study shows that several key elements take important roles in the retrieval of heuristics, including the ways of problem presentation, complexity of heuristics and status of goal. The fitness of model prediction to real participants΄ data on behavior and BOLD (Blood Oxygenation Level-Dependent) response in five predefined brain regions illustrates that our hypotheses and results are acceptable. This work is a significant step towards tackling the puzzle of the heuristics retrieval in human brain. [ABSTRACT FROM AUTHOR]

With the advancement both in the Web (e.g., semantic Web and human-level wisdom-Web computing) and in Brain Informatics (BI) (e.g., advanced information technologies for brain science and non-invasive neuroimaging technologies, such as functional magnetic resonance imaging (fMRI)), several lines of BI research have been developed directly or indirectly related to Web Intelligence (WI). Some of them can be treated as the extension to the Web research of the tradition BI research, such as computational cognitive modeling like ACT-R. ACT-R is a theory and model of computational cognitive architecture which consists of functional modules, such as declarative knowledge module, procedural knowledge module, goal module and input (visual, aural), output (motor, verbal) modules. Information can be proposed parallel inside and among the modules, but has to be sequentially if it needs procedural module to coordinate the behavior across modules. At the International WIC Institute (WICI), we are trying to introduce this kind of architecture and the mechanism of activation of the units in declarative knowledge module into our Web information system. Based on or related to ACT-R, theories and models that are with very close relation to WI have also been developed, such as threaded cognition for concurrent multitasking, cognitive agents, human-Web interaction (e.g., SNIT-ACT (Scentbased navigation and information foraging in the ACT cognitive architecture). At the WICI, we are also working on the user behavior and reasoning on the Web by eye-tracker and fMRI. Some of other BI studies, however, have been developed directly by the requirement of WI research. For example, to meet the requirement of the development of Granular Reasoning (GrR) technologies in WI research, people at the WICI have been checking how human can perceive the real world under many levels of granularity (i.e., abstraction) and can also easily switch among granularities. By focusing on different levels of granularity, one can obtain different levels of knowledge, as well as in-depth understanding of the inherent knowledge structure. The interaction between human intelligence inspired WI methodology research and WI stimulated BI principle research will benefit both BI and WI researches greatly. [ABSTRACT FROM AUTHOR]