In this work, in order to improve spatial recognition abilities for the long-term operation tasks of the assistant robot for the elderly, a novel approach of semantic region estimation is proposed. We define a novel g...In this work, in order to improve spatial recognition abilities for the long-term operation tasks of the assistant robot for the elderly, a novel approach of semantic region estimation is proposed. We define a novel graphbased semantic region descriptions, which are estimated in a dynamically fashion. We propose a two-level update algorithm, namely, Symbols update level and Regions update level. The algorithm firstly adopts particle filter to update weights of the symbols, and then use the Viterbi algorithm to estimate the region the robot stays in based on those weights, optimally. Experimental results demonstrate that our proposed approach can solve problems of the long-term operation and kidnapped robot problem.展开更多
This study examined the influence of semantic and phonological priming on L2 speech planning,as well as the difference between native and non-native speakers of English in terms of lemma activation.Two potential expla...This study examined the influence of semantic and phonological priming on L2 speech planning,as well as the difference between native and non-native speakers of English in terms of lemma activation.Two potential explanations for the contrast between the performance of L2 speakers and native controls were considered.The first of which was that L2 speakers’ phonological forms are activated before selection of syntactic frame occurred,whereas the reverse is true for native speakers.The second explanation posits that the organisation of the speech production procedure is fundamentally similar in native and L2 speakers,and the disparity in performance arises from difference in the levels of activation of stored items.The results of the present experiment suggest that lemma selection is indeed what drives syntactic frame selection.However,lemmas in L2 speakers can be primed through a chain of connections demonstrated as:L2 phonological form →L1 phonological form → L1 lemma.展开更多
BACKGROUND： Dyslexic children exhibit reading ability unmatched to age, although they possess normal intelligence and are well educated. OBJECTIVE： To examine the performance of dyslexic children in Chinese charac...BACKGROUND： Dyslexic children exhibit reading ability unmatched to age, although they possess normal intelligence and are well educated. OBJECTIVE： To examine the performance of dyslexic children in Chinese characters visual recognition tasks and to investigate the relationship between priming effect in character recognition and dyslexia. DESIGN, TIME AND SETTING： A case-control study was performed at the Department of Children and Adolescent Health and Maternal Care, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology between March and June 2007. PARTICIPANTS： A total of 75 primary school students in grades 3 and 5 were selected from two primary schools in Wuhan City, Hubei province, China, and were assigned to three groups. （1） Reading disability （RD, n = 25）; （2） chronological age （CA） group （n = 25 normal readers that were intelligence quotient and age-matched to the RD group）; （3） reading level （RL） group （n = 25 normal readers that were intelligence quotient and RL-matched to the RD group）. All children were right-handed and had normal or corrected-to-normal vision. METHODS： Recognition of target characters was performed in each child using a masked prime paradigm. Recognition speed and accuracy of graphic, phonological, and semantic characters were examined. Simultaneously, data, with respect to response time for each target character and error rate, were recorded to calculate facilitation values （unrelated RT-related RT）. MAIN OUTCOME MEASURES： Response time, facilitation, and error rate in Chinese character recognition task were calculated. RESULTS： The baseline-adjusted facilitation of graphic, phonological, and semantic priming for dyslexic children was -0.010, -0.010, and 0.001, respectively. Dyslexic children displayed inhibition in graphic and phonological prime conditions. Facilitations under the three prime conditions were 0.026, 0.026, and 0.022 for the CA group. In the RL group, results were 0.062, 0.058, and 0.031 respectively. The differences of baseline-adjusted facilitation across the three groups were significant [F （2, 216） = 17.91, P 〈 0.01 ], whereas the main effect of prime condition IF （2, 216） = 0.49, P 〉 0.05] and the interaction of group x prime condition [F （4, 216） = 0.91, P 〉 0.05] were not significant. The error rate under the three prime conditions was 0.066, 0.077, and 0.079 for the RD group. As for the CA group, the results were 0.057, 0.071, and 0.074 accordingly, and in the RL group, the results were 0.119, 0.111, and 0.121, respectively. The difference of error rate across the three groups was significant （F = 6.61, P = 0.002）. The error rate was significantly greater in the RL group, compared with the CA and RD groups, while no significant difference existed between the RD and CA groups. CONCLUSION： Dyslexic children were slower to recognize target characters than normal readers, and did not exhibit significant prime effects in graphic and phonological primer condition. These results suggested that dyslexic children exhibited general deficits in phonological and graphic processing, and presented a different pattern in character recognition. These results could be interpreted by parallel-distributed processing models.展开更多
基金supported by the National Natural Science Foundation of China (61305103 and 61473103)the Foundation for Innovative Research Groups of the National Natural Science Foundation of China (51521003 )+1 种基金the Natural Science Foundation of Heilongjiang Province, China (QC2014C072 and F2015010)SelfPlanned Task (SKLRS201609B and SKLRS201411B) of State Key Laboratory of Robotics and System (HIT)
文摘In this work, in order to improve spatial recognition abilities for the long-term operation tasks of the assistant robot for the elderly, a novel approach of semantic region estimation is proposed. We define a novel graphbased semantic region descriptions, which are estimated in a dynamically fashion. We propose a two-level update algorithm, namely, Symbols update level and Regions update level. The algorithm firstly adopts particle filter to update weights of the symbols, and then use the Viterbi algorithm to estimate the region the robot stays in based on those weights, optimally. Experimental results demonstrate that our proposed approach can solve problems of the long-term operation and kidnapped robot problem.
文摘This study examined the influence of semantic and phonological priming on L2 speech planning,as well as the difference between native and non-native speakers of English in terms of lemma activation.Two potential explanations for the contrast between the performance of L2 speakers and native controls were considered.The first of which was that L2 speakers’ phonological forms are activated before selection of syntactic frame occurred,whereas the reverse is true for native speakers.The second explanation posits that the organisation of the speech production procedure is fundamentally similar in native and L2 speakers,and the disparity in performance arises from difference in the levels of activation of stored items.The results of the present experiment suggest that lemma selection is indeed what drives syntactic frame selection.However,lemmas in L2 speakers can be primed through a chain of connections demonstrated as:L2 phonological form →L1 phonological form → L1 lemma.
基金the National Natural Science Foundation of China,No.30872132
文摘BACKGROUND： Dyslexic children exhibit reading ability unmatched to age, although they possess normal intelligence and are well educated. OBJECTIVE： To examine the performance of dyslexic children in Chinese characters visual recognition tasks and to investigate the relationship between priming effect in character recognition and dyslexia. DESIGN, TIME AND SETTING： A case-control study was performed at the Department of Children and Adolescent Health and Maternal Care, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology between March and June 2007. PARTICIPANTS： A total of 75 primary school students in grades 3 and 5 were selected from two primary schools in Wuhan City, Hubei province, China, and were assigned to three groups. （1） Reading disability （RD, n = 25）; （2） chronological age （CA） group （n = 25 normal readers that were intelligence quotient and age-matched to the RD group）; （3） reading level （RL） group （n = 25 normal readers that were intelligence quotient and RL-matched to the RD group）. All children were right-handed and had normal or corrected-to-normal vision. METHODS： Recognition of target characters was performed in each child using a masked prime paradigm. Recognition speed and accuracy of graphic, phonological, and semantic characters were examined. Simultaneously, data, with respect to response time for each target character and error rate, were recorded to calculate facilitation values （unrelated RT-related RT）. MAIN OUTCOME MEASURES： Response time, facilitation, and error rate in Chinese character recognition task were calculated. RESULTS： The baseline-adjusted facilitation of graphic, phonological, and semantic priming for dyslexic children was -0.010, -0.010, and 0.001, respectively. Dyslexic children displayed inhibition in graphic and phonological prime conditions. Facilitations under the three prime conditions were 0.026, 0.026, and 0.022 for the CA group. In the RL group, results were 0.062, 0.058, and 0.031 respectively. The differences of baseline-adjusted facilitation across the three groups were significant [F （2, 216） = 17.91, P 〈 0.01 ], whereas the main effect of prime condition IF （2, 216） = 0.49, P 〉 0.05] and the interaction of group x prime condition [F （4, 216） = 0.91, P 〉 0.05] were not significant. The error rate under the three prime conditions was 0.066, 0.077, and 0.079 for the RD group. As for the CA group, the results were 0.057, 0.071, and 0.074 accordingly, and in the RL group, the results were 0.119, 0.111, and 0.121, respectively. The difference of error rate across the three groups was significant （F = 6.61, P = 0.002）. The error rate was significantly greater in the RL group, compared with the CA and RD groups, while no significant difference existed between the RD and CA groups. CONCLUSION： Dyslexic children were slower to recognize target characters than normal readers, and did not exhibit significant prime effects in graphic and phonological primer condition. These results suggested that dyslexic children exhibited general deficits in phonological and graphic processing, and presented a different pattern in character recognition. These results could be interpreted by parallel-distributed processing models.
