EARSHOT: A Minimal Neural Network Model of Incremental Human Speech Recognition

Despite the lack of invariance problem (the many-to-many mapping between acoustics and percepts), human listeners experience phonetic constancy and typically perceive what a speaker intends. Most models of human speech recognition (HSR) have side-stepped this problem, working with abstract, idealized inputs and deferring the challenge of working with real speech. In contrast, carefully engineered deep learning networks allow robust, real-world automatic speech recognition (ASR). However, the complexities of deep learning architectures and training regimens make it difficult to use them to provide direct insights into mechanisms that may support HSR. In this brief article, we report preliminary results from a two-layer network that borrows one element from ASR, long short-term memory nodes, which provide dynamic memory for a range of temporal spans. This allows the model to learn to map real speech from multiple talkers to semantic targets with high accuracy, with human-like timecourse of lexical access and phonological competition. Internal representations emerge that resemble phonetically organized responses in human superior temporal gyrus, suggesting that the model develops a distributed phonological code despite no explicit training on phonetic or phonemic targets. The ability to work with real speech is a major advance for cognitive models of HSR.     

小学低段汉字识别和听写的发展轨迹：语素意识的预测作用

汉字识别和听写是字词水平上两项重要的读写技能。对127名一年级儿童进行了两年4个时间点的追踪调查, 通过测查儿童的一般认知能力、语音意识、正字法意识和语素意识(包括同音语素意识、复合语素意识), 探讨影响儿童汉字识别和听写发展的因素。利用潜变量增长模型, 结果发现：(1)汉字识别呈线性发展趋势, 听写呈先快后慢的非线性发展趋势, 二者均存在补偿效应; (2)语素意识显著预测了汉字识别的起始水平和发展速度, 但语素意识对听写的起始水平和发展速度均不具有显著的预测作用。研究结果表明儿童入学初始的汉字识别和听写水平并不决定后期的发展速度; 语素意识对儿童汉字识别和听写发展的作用不同。     

Disentangling the far transfer of language comprehension gains using latent mediation models

While we know that interventions targeting oral language can be effective, little is known about what drives these effects. In this study, we examine whether gains in transfer measures are mediated through the specific words that are trained in a language intervention. Based on a large‐scale randomized controlled trial of language intervention in 4‐ to 5‐year‐old children, latent mediation models were used to disentangle oral language gains in transfer measures. The results first showed that the effects of the language intervention and the transfer effects are generated through expressive rather than receptive measures of language. Second, we found that the effects of the intervention on intermediate transfer measures of language were mediated through the ability to define the trained words. Third, and critically, for far transfer measures that did not contain any of the trained words, the effects were mediated through the trained words. The findings relate to theories of transfer and support the idea that far transfer is possible, at least within the same domain. In addition, it seems that effects on receptive language skills are difficult to obtain and that what is improved is instead the children's ability to express themselves and use procedures to explain words. Thus, to optimize intervention effects, future studies should focus on expressive language.     

面孔识别的自我群体偏向

面孔认知研究表明, 人们识别和再认自己所属群体(如种族、性别、年龄)面孔的成绩显著好于识别其他群体面孔的成绩。近年来围绕面孔识别的这种自我群体偏向进行了实验研究, 研究者提出了知觉经验说和社会认知论两种理论来解释其认知加工机制, 而类别化-个性化模型和双路径模型则是将两种理论分别加以整合提出的新解释。研究者还对面孔识别自我群体偏向的神经机制进行了研究, 探讨了评价者、评价对象和评价任务等因素对自我群体偏向的影响。提升面孔识别自我群体偏向研究的生态效度, 构建整合性理论模型以及加强跨文化和本土研究是未来研究的重要方向。     

隐藏情绪分析与识别方法

隐藏情绪识别对公共安全防范与预警具有重要的意义。微表情是揭示隐藏情绪的一条重要通道。但目前隐藏情绪研究较少且微表情因其细微幅度与快速出现等特性难以识别, 其研究尚未在实际中广泛应用。因为, 隐藏情绪的认知与表达机理亟需系统的研究,采集实际场景中的微表情数据, 并以脑电信号辅助微表情的精确标注是提高微表情标注效率的有效途径。深入研究微表情识别方法, 并辅以人脸颜色、注视估计和非接触生理信号等多通道数据, 以检测与识别隐藏情绪。社会公共安全是隐藏情绪分析和识别的典型场景。面向精神疾病患者两害行为(即危害自身或他人的危险行为)风险评估和服刑人员会见场景隐藏情绪检测, 可以有效地对相应系统和方法进行验证和修正。     

藏语母语者汉语词汇识别的眼动研究

采用眼动技术,通过两个实验考察了藏语母语者在不同语境中阅读汉语句子时,字形、语音信息在词汇识别中分别发挥的作用以及词频效应。结果发现：（1）在高限制性句子语境中,字形和语音共同作用;（2）在低限制性句子语境中,语音作用显著;（3）词频效应出现在高限制性语境的晚期,以及低限制性语境的中期和晚期。该结果表明,在藏语母语者阅读汉语的过程中,句子语境影响词汇识别过程中字形、语音的作用及作用的时间进程,藏语母语者汉语词汇识别符合双通道理论。     

Parts and wholes in person recognition: developmental trends

This study's aim is the investigation of short-term visual person recognition in 8- and 10-year-olds and adults, within the part-whole paradigm introduced by. Natural unfamiliar whole persons were contrasted with natural unfamiliar faces to test for differences between person processing and face processing. Two experiments showed advantages of whole face recognition over isolated face feature recognition. Also, these was a complete over part probe advantage (CPA, ) for person recognition in all age groups. Thus, recognition became more accurate between 8 years and adulthood, but no developmental shift in visual information processing was observable with face and whole person recognition. I conclude that person recognition does not rely on processes completely different from those of face recognition and that this holds for 8- and 10-year-olds as well as for adults.     

Crossmodal temporal order and processing acuity in developmentally dyslexic young adults

We investigated crossmodal temporal performance in processing rapid sequential nonlinguistic events in developmentally dyslexic young adults (ages 20-36 years) and an age- and IQ-matched control group in audiotactile, visuotactile, and audiovisual combinations. Two methods were used for estimating 84% correct temporal acuity thresholds: temporal order judgment (TOJ) and temporal processing acuity (TPA). TPA requires phase difference detection: the judgment of simultaneity/nonsimultaneity of brief stimuli in two parallel, spatially separate triplets. The dyslexic readers' average temporal performance was somewhat poorer in all six comparisons; in audiovisual comparisons the group differences were not statistically significant, however. A principal component analysis indicated that temporal acuity and phonological awareness are related in dyslexic readers. The impairment of temporal input processing seems to be a general correlative feature of dyslexia in children and adults, but the overlap in performance between dyslexic and normal readers suggests that it is not a sufficient reason for developmental reading difficulties.     

Sex differences in judgement of facial affect: a multivariate analysis of recognition errors

The present paper investigated recognition errors in affective judgement of facial emotional expressions. Twenty-eight females and sixteen males participated in the study. The results showed that in both males and females emotional displays could be correctly classified, but females had a higher rate of correct classification; males were more likely to have difficulty distinguishing one emotion from another. Females rated emotions identically regardless of whether the emotion was displayed by a male or female face. Furthermore, the two-factor structure of emotion, based on a valence and an arousal dimension, was only present for female subjects. These results further extend our knowledge about gender differences in affective information processing.