上周,在IJAC优先在线发表的几篇论文中,麻省理工美国人文与科学院院士Tomaso Poggio的一篇有关深度学习的综述成为一大亮点。Poggio教授在这篇文章中阐述了有关神经网络深度学习的一个基本问题:为什么深层网络比浅层网络更好? 文章内容延续了Poggio教授在2016年8月在中国人工智能大会(2016CCAI)上的演讲报告《The Science and the Engineering ofIntelligence》。 图1 来源于2016 CCAI 演讲PPT 图2 来源于2016 CCAI 演讲PPT “ 深度学习架构和机器学习模式的搭建,来自于神经学方面的研究进展,换句话说,同样的架构是存在于大脑皮质当中的。 关于深度学习,已经有成千上万的研究者在不同领域进行这方面的研究,比如无人驾驶、语音识别等等。可是我们还不清楚,为什么深度学习在这些工程应用中会起作用,深度学习的机理是什么? 我们对这个问题很感兴趣另外一个原因是:探讨深度学习的机理也将有助于我们理解‘为什么大脑皮质会存在一些不同的层次?’ ” Poggio 教授在这篇文章中,将为您解读深度学习的关键理论、最新成果和开放式研究问题。 同时这篇文章也是IJAC即将发表的 Special Issue on Human Inspired Computing 中的一篇文章。该专题其他热文将陆续优先在线发表,敬请期待。 一点点题外话:小编在去年的CCAI大会上有幸拜访了Poggio教授,教授博学、谦逊而富有亲和力的形象给小编也留下深刻印象。他曾提到:期望能帮助年轻人更好的了解神经科学、理解机器学习。如果要在智能方面走得远,不能只靠计算机,还需要与人类本身的研究相互结合,才能碰撞出更多的东西。 接下来,小编将为您奉上这篇文章的具体信息,以及IJAC近期在线发表的其他文章链接,欢迎下载阅读 【 Title 】 Why and when can deep-but not shallow-networksavoid the curse of dimensionality: A review 【 Author 】 Tomaso Poggio, Hrushikesh Mhaskar, LorenzoRosasco, Brando Miranda, Qianli Liao 【 Abstract 】 The paper reviews and extends an emerging bodyof theoretical results on deep learning including the conditions under which itcan be exponentially better than shallow learning. A class of deepconvolutional networks represent an important special case of these conditions,though weight sharing is not the main reason for their exponential advantage.Implications of a few key theorems are discussed, together with new results,open problems and conjectures. 【 Keywords 】 Machine learning, neural networks, deep andshallow networks, convolutional neural networks, function approximation, deeplearning 【 Full Text 】 https://link.springer.com/article/10.1007/s11633-017-1054-2 【 Publish date 】 Published online March 14, 2017 The other recentlypublished online papers include: 【 Title 】 Improvement of wired drill pipe data qualityvia data validation and reconciliatio 【 Author 】 Dan Sui, Olha Sukhoboka, Bernt Sigve Aadnøy 【 Keywords 】 Data quality, wired drill pipe (WDP), datavalidation and reconciliation (DVR), drilling models 【 Full Text 】 https://link.springer.com/article/10.1007/s11633-017-1068-9 【 Publish date 】 Published online March 4, 2017 【 Title 】 Reaction torque control of redundantfree-floating space robot 【 Author 】 Ming-He Jin, Cheng Zhou, Ye-Chao Liu, Zi-QiLiu, Hong Liu 【 Keywords 】 Redundant space robot, reaction torque,reactionless control, base disturbance minimization, Linux/real timeapplication interface (RTAI) 【 Full Text 】 https://link.springer.com/article/10.1007/s11633-017-1067-x 【 Publish date 】 Published online March 4, 2017 【 Title 】 A piecewise switched linear approach fortraffic flow modeling 【 Author 】 Abdelhafid Zeroual,Nadhir Messai, SihemKechida, Fatiha Hamdi 【 Keywords 】 Switched systems, modeling, macroscopic,traffic flow, data calibration 【 Full Text 】 https://link.springer.com/article/10.1007/s11633-017-1060-4 【 Publish date 】 Published online March 4, 2017 【 Title 】 Navigation of non-holonomic mobile robot usingneuro-fuzzy logic with integrated safe boundary algorithm 【 Author 】 A. Mallikarjuna Rao, K. Ramji, B. S. K.Sundara Siva Rao, V. Vasu, C. Puneeth 【 Keywords 】 Robotics, autonomous mobile robot (AMR),navigation, fuzzy logic, neural networks, adaptive neuro-fuzzy inference system(ANFIS), safe boundary algorithm 【 Full Text 】 https://link.springer.com/article/10.1007/s11633-016-1042-y 【 Publish date 】 Published online March 4, 2017 【 Title 】 Method for visual localization of oil and gaswellhead based on distance function of projected features 【 Author 】 Ying Xie, Xiang-Dong Yang, Zhi Liu, Shu-NanRen, Ken Chen 【 Keywords 】 Robot vision, visual localization, 3D objectlocalization, model based pose estimation, distance function of projectedfeatures, nonlinear least squares, random sample consensus (RANSAC) 【 Full Text 】 https://link.springer.com/article/10.1007/s11633-017-1063-1 【 Publish date 】 Published online March 4, 2017 【 Title 】 Virtual plate based controlling strategy oftoy play for robots communication development in JA space 【 Author 】 Wei Wang, Xiao-Dan Huang 【 Keywords 】 Human robot cooperation, joint attention (JA)space, reachable space, toy play ability, a virtual plate 【 Full Text 】 https://link.springer.com/article/10.1007/s11633-016-1022-2 【 Publish date 】 Published online February 21, 2017 阅读更多IJAC优先在线出版论文: https://link.springer.com/journal/11633
International Workshop on Computational, Cognitive and Behavioral Social Science (CCB) 计算认知与行为社会科学国际学术研讨会 台湾政治大学 December 6-8, 2013 Call for Papers This workshop is motivated by the constant efforts made to search for a unified framework for social sciences. Among several different threads which have been developed over the last decade, a major one is the use of the agent-based model as a unified framework to study the complexity of social dynamics. This development is built upon a view that Stephen Wolfram (Wolfram, 2002; Zenil, 2013) coined as computational equivalence (CE), i.e., social processes, characterized as interactions of heterogeneous agents, can be regarded as an equivalent of computation, or even more, universal computation. A similar view is widely shared among social scientists (Axelrod, 1997; Albin and Foley, 1998; Velupillai, 2000, 2010; Miller and Page, 2007; Mirowski, 2007). While the historical root of CE can be traced all the way back to Alan Turing and other pioneers before or around the 1950s, it is the extensive use and studies of cellular automata, such as John Conway's Game of Life initiated in the late 1960s or Stephen Wolfram's Elementary Cellular Automata in the early 1980s, that enhances the general awareness of the computational nature of social sciences and facilitates the emergence of this new discipline, known as computational social science (CSS). Thomas Schelling's work on the segregation model (Schelling, 1971) is considered as one of earliest and classic work in this CSS. This work is an enlightenment for many physicists who later become interested in social sciences (Buchanan, 2007). A milestone of the development of computational social science is the recent publication of the 4-volume collections of 66 influential articles written over the last four decades, edited by Nigel Gilbert (Gilbert, 2010). This collection not only clearly indicates what computational social science was and is, but also feature its possible becoming. This CCB workshop can be considered as a part of this long endeavor. In addition to the computational social science, we also notice another unified efforts emanating from cognitive science and psychology, which is known as the cognitive and behavioral social science. The attempt to have a cognitive social science has an even longer history than computational social science. Herbert Simon had a great influence on initiating this development. In his academic lifetime, he constantly called for the conversation between cognitive science and social sciences. The endeavor was followed by Daniel Kahneman, Amos Tversky, Paul Slovic, Richard Thaler, Reinhard Selten, Gerd Gingerenzer, Vela Velupillai, to name a few. This long series of efforts prompts the advent of the name cognitive social science (Turner, 2001; Sun, 2012). This line of development has extended into a blend of various methodologies, some using laboratory approach (partially accompanied by the recent progresses in neuroscience), some relying on computation modeling and simulation approach. For the former, the success of experimental economics has gradually spread to other social disciplines, such as political sciences, sociology, and management; as to the later, the early mathematical psychological model of reinforcement learning (Bush and Mosteller, 1955) has been further extended by behavioral economists into generalized reinforcement learning model (Camerer and Ho, 1999), not to mention the fusion of various heuristics borrowed from computer scientists and psychologists, such as case-based decisions, decision trees (fast-and-frugal heuristics), and many other computational intelligence tools. They together provide social sciences a foundation from brain to mind and further to decision-making, which enhance our understandings of preference and expectation formation, choice and judgement making, learning with experiences, and risk management. Computational social science has the interactions of heterogeneous agents following different rules as the main theme, but it may not pay much attention to the fine details on who these agents are, their cognitive constraints, and their behavioral rules. Nonetheless, in the empirical-oriented, agent-based models, data from human-subject experiments are, however, employed to design reasonable artificial agents. This development is convincing enough that there will be more fruitful collaborations between computational social science and cognitive and behavioral social science in the future. In fact, both the upward causation and downward causation of agent-based modeling may involve genes, neurons, personality, and culture as part of the mechanisms. to put it alternatively, these fine details are the subroutines or modules to be seen everywhere in the interaction processes. While it is not necessary for all agent-based models to have genes or neurons as their elementary units, knowledge of cognitive and behavioral social science can help us decide, for example (just as a convenient metaphor), among the 256 rules in Wolfram’s one-dimensional cellular automata, which ones are more human-like? On the other hand, it is also desirable to know how these fine details can amplify them in the social emergence; for example, how amygdala can help herding behavior and enhance instability of financial markets. Can we design a treatment that people can easily develop trust relation, which in turn beef up the team production, GDP and happiness index? The exemplar questions may not be rigorously shaped, but they give the idea and the flavor. Being capable of addressing the question with that depth of individuals (foundation) and breath of the society (aggregation) is the purpose of bringing together these two different but closely related treads. We believe that the cross-fertilization of these two unified social sciences is the next step of each of the two, and we believe that computational, cognitive and behavioral social science is the future of an integrated social science. The uniqueness of this workshop is to bring together the scholars from both strands and begin the constructive and fascinating conversation. We welcome submissions addressing various social dynamics, such as, but definitely not limited to, voting, identity, segregation, social exclusion, discrimination, financial instability, urban dynamics, social networks, leadership, congestion, disease transmission, gossip and mass media, culture and social norms, interpersonal relations, pro-social behavior, using (agent-based) computational, cognitive and behavioral modeling, laboratory and field experiments, learning, decision-making under risky environment, gambling, crime, national security, etc. We sincerely welcome participants from different sister disciplines, such as economics, ecology, political science, public health, sociology, social anthropology, ethnology, geography, psychology, communication, law, management science, linguistics, religion, cultural studies, biology, physics, computer sciences, mathematics, neuroscience and genetics,…etc. Organizers Shu-Heng Chen , National Chengchi University, Taiwan Ming Hsu , University of California at Berkeley, USA Akira Namatmae , National Defense Academy, Japan Leonid Perlovsky , Harvard University, USA David Sallach, University of Chicago, USA John Staddon , Duke University, USA Paul Wang , Duke University, USA Yingxu Wang , University of Calgary, Canada International Program Committee Yuji Aruka , Chuo University, Japan Shih-Fen Cheng , Singapore Management University, Singapore Siew Ann Cheong , Nanyang Technological University, Singapore Ya-Chi Huang , Lunghwa University of Science and Technology, Taiwan Kai Pui Lam , Chinese University of Hong Kong, Hong Kong Honggang Li , Beijing Normal University, China Sai-Ping Li , Academia Sinica, Taiwan Chia-Yang Lin, National Chengchi University, Taiwan Wen-Jong Ma , National Chengchi University, Taiwan Akira Namatame , National Defense Academy, Japan Heping Pan , Chongqing Institute of Finance, China Da Ren, Tianjin University, China Kwok Yip Szeto , Hong Kong University of Science and Technoloygy, Hong Kong Chung-Ching Tai , Tunghai University, Taiwan Bing-Hong Wang , University of Science and Technology of China, China Guocheng Wang , Chinese Academy of Social Sciences, China Zhijian Wang , Zhejiang University, China Zhongyu Wang , Harbin Institute of Technology, China Haijun Yang , Beijing University of Aeronautics and Astronautics, China Chia-Hsuan Yeh , Yuan Ze University, Taiwan Nai-Shing Yen , National Chengchi University, Taiwan Tongkui Yu , Southwest University, China Wei Zhang , Tianjin University, China Yu Zhang , St. John's University, USA Wei-Xing Zhou , East China University of Science and Technology, China Chia-Ling Chang, National Chengchi University, Taiwan Paper Submissions All papers are to be submitted electronically through the TAAI 2013 submission website: https://www.easychair.org/conferences/?conf=taai2013 . Each submission should be regarded as an undertaking that, if the paper is accepted, at least one of the authors must attend the conference to present the work. Templates · 8.5 x 11 Word Template (ZIP) · 8.5 x 11 double-column LaTex Package (ZIP) Important Dates Call for papers 15 March 2013 Manuscripts due 23 August 2013 Acceptance notification 4 October 2013 Camera-ready versions due 14 October 2013 Paper author registrations due 14 October 2013 Early-bird registration due 28 October 2013 Conference dates 6-8 December 2013 Sponsors New Mathematics and Natural Computation AI-ECON Research Center, NCCU
近年来试图做一些神经网络进化,发育和可塑性建模方面的研究,最近发表的部分论文如下。如对某篇论文感兴趣,欢迎索取并批评指正。 1. J. Chrol-Cannon, A. Gruning, Y. Jin . The emergence of polychronous groups under varying input. IJCNN 2012 2. J. Yin, Y. Meng and Y. Jin . A developmental approach to structural self-organization in reservoir computing. IEEE Transactions on Autonomous Mental Development , 2012 3. D. Bush and Y. Jin . Calcium control of triphasic hippocampal STDP. Journal of Computational Neuroscience . 2012 4. B. Inden, Y. Jin , R. Haschke, H. Ritter. Evolving neural fields for problems with large input and output spaces . Neural Networks , 28:24-39, 2012 5. L. Schramm, Y. Jin , and B. Sendhoff. Evolution and analysis of genetic networks for stable cellular growth and regeneration . Artificial Life , 2012 6. Y. Meng, Y. Jin and J. Yin. Modelling activity-dependent plasticity in BCM spiking neural networks with application to human behaviour recognition. IEEE Transactions on Neural Networks , 22(12):1952-1966, 2011 7. L. Schramm, Y. Jin , B. Sendhoff. Emerged coupling of motor control and morphological development in evolution of multi-cellular animates. 10th European Conference on Artificial Life , Budapest, September 2009 8. B. Jones, Y. Jin , B. Sendhoff, and X. Yao. Emerged optimal distribution of computational workload in the evolution of an undulatory animat. The 11th International Conference on Simulation of Adaptive Behaviors (SAB 2010), August 24-28, 2010 9. B. Jones, Y. Jin , B. Sendhoff, and X. Yao. The effect of proprioceptive feedback on the distribution of sensory information in a model of an undulating organism. 10th European Conference on Artificial Life , Budapest, September 2009 10. B. Jones, Y. Jin , B. Sendhoff, and X. Yao. Evolving functional symmetry in a three dimensional model of an elongated organism. Artificial Life XI , pp.305-312, 2008 11. Y. Jin , L. Schramm, and B. Sendhoff. A gene regulatory model for the development of primitive nervous systems. INNS-NNN Symposia on Modeling the Brain and Nervous Systems , November 2008, Auckland, New Zealand, LNCS 5506, pp.48-55, 2009
【文献导读】 雷帕霉素(西罗莫斯)是从吸水链霉菌中分离出来的一种抗生素,因最初来自复活节岛(Rapa Nui)而得名。雷帕霉素具有很强的免疫抑制作用,已于1999年被美国FDA批准为器官移植手术后的抗免疫排斥药。近年来发现,雷帕霉素还能抗癌、治疗类风湿性关节炎和延缓衰老。 6月29日Science Daily撰文介绍了美国德克萨斯大学科学家取得的一项新成果,他们将雷帕霉素添加到小鼠食物中,结果发现幼鼠的认知力(学习能力和记忆力)有很大提高,老龄鼠的认知力也有极大改善。在服用雷帕霉素后,那种常见的随年龄增大学习能力下降和记忆力减退的现象似乎消失不见,而且焦虑和抑郁等行为也变得较为罕见。 雷帕霉素的抗焦虑和抗抑郁作用在不同月龄的小鼠中都能观察到,短的为4个月(相当于人类上大学的年龄),较长的有12个月(中年),最长的达到25个月(老年)。更令人惊讶的是,在饲喂雷帕霉素的小鼠中脑内,血清素、多巴胺、去甲肾上腺素等“幸福感”神经递质的含量显著升高,这大概就是雷帕霉素让小鼠“快活”的重要原因。 尽管雷帕霉素抗衰老已经在酵母、线虫、果蝇、小鼠中证实,但至今尚未在人体上得到确认。不过,估计世界上目前至少有10几家实验室正在开展密集的攻关研究,很快就会有雷帕霉素对人类衰老及其相关疾病作用的研究结果发表,请大家拭目以待。 两年前,美国普林斯顿大学的科学家在线虫中研究发现,胰岛素信号转导与限食对学习能力和记忆力随年龄增长而下降的影响不同。胰岛素/胰岛素样生长因子-1受体突变体daf-2能改善成长早期记忆力并在老年维持良好的学习能力,但不能维持长期记忆力。相反,限食模型的eat-2突变体造成年幼时长期记忆力损害,但能将此记忆力水平保持得更长久( http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1000372 )。 假如你不是一位刻意追求“长生不老”的人,但雷帕霉素提高学习能力和增强记忆力的奇特效果肯定对你有吸引力。不过,雷帕霉素尚未批准用于抗免疫排斥以外的用途,它也不是可以在药店随意购买的非处方药,而且服用它还要以牺牲部分免疫力为代价,因此在使用之前你必须认真掂量掂量。 本文于6月28日在线发表于《神经科学》(Neuroscience)杂志上:Chronic Inhibition of mTOR by Rapamycin Modulates Cognitive and Non-cognitive Components of Behavior Throughout Lifespan in Mice ( http://www.sciencedirect.com/science/article/pii/S0306452212006720?v=s5 ) Science Daily的评述:Easter Island Drug Raises Cognition Throughout Life Span in Mice ( http://www.sciencedaily.com/releases/2012/06/120629211902.htm )
语料库进入认知语言学、隐喻研究领域,这是人类研究语言发展的一种大势所趋。 近段时间在读一些有关语料库与系统功能语言学的书籍,一些文章很有趣。比如,Stubbs在其文章 “Corpus analysis: the state of the art and three types of unanswered questions” 中提出: 在由 Firth , Halliday 和 Sinclair 为代表的传统英国语言学领域中,文本与语料库分析是中心。在这一传统中,非常重视常规短语学这一概念,在语言使用的创造性与常规性之间达成一种平衡。如, Firth (1953)谈到词与短语的习惯性、约定俗成性与典型性; Halliday(1978) 指出:大部分语篇或多或少是常规性的,我们“总是一而再再而三地表达同一种观点”; Sinclair (1991) 提出“两种解释原则”,一种是习语原则,一种是开放性选择原则,认为半固定的短语是“非常普遍的”。 Stubbs 尝试运用语料库来研究英语短语学的范围与优势,并将探讨语料库方法可以为语义的认知模型研究提供大量语料 , 有助于解决认知和社会理论中长期以来悬而未决的问题 , 即语言如何与认知和社会系统相联系。他提出 他的中心议题:语料库方法能有助于解决认知和社会理论中长期以来悬而未决的问题(大致分成三类:比较简单的,比较困难的,不可能的),即: Easier descriptive questions concern how we can make generalizations about phraseology across the lexicon. More difficult questions concern whether different models of phrasal units can be related to each other. The deepest—maybe impossible—questions concern whether linguistic, cognitive and social patterns can be related. 简单的描述性问题:主要探讨我们是如何将短语学从词汇层面加以一般化(泛化); 稍难的问题:不同的短语单位模式能否彼此相关联; 最深层次问题(有可能是无能为力的问题):语言学、认知与社会模式能否彼此相关联。 从第二个层面的问题来看,语料库数据能为语义关系提供经验性证据( empirical evidence ),因此也就能够为心理词汇是如何组织的提供证据; 从第三个层面的问题来看,语料库数据能为说话者经常谈论的事物提供证据,因此也就能为社会性意义突出的词汇范围提供证据,如“金钱” (money) 和“人群” (group of people) 。 从第一个层面,即浅层次的描写型层面来看,通过那些文化关键词汇的经验性描述,给生活一种全新的阐释。 Popper(1963:125) 指出,社会理论的任务就是解释我们的目的与行为是如何引出那些无意识的后果 / 结果 。 我们试图通过抱怨门不能推移,或表达对某人延年益寿的羡慕,用语言来传达自己的意义。但是,我们并没有有意识去复制那些典型的英语短语。很明显,人们对语言的使用都是重复性的,但要解释跨社区之间的语言重复或要解释在功能系统中,什么层次的重复是最优化的,要做到这一点,可不是容易的事情。 也只有通过这些问题,语料库研究才能从描述性发展到解释性。 Stubbs通过大量例证,来说明语料库在研究人类认知活动中的功用,如: A second example of such an area is the ways in which people are classified and talked about. Two concepts which are encoded in a large number of the words and phrases we use to talk about social life are ‘groups of people’ and ‘the passing of time’ : The large number of approximate synonyms for ‘groups of people’ is not surprising, since the different ways in which people can be grouped is of inherent social interest. Here are just a few examples: - band, bunch, crew; family, flock, gang, group, jury, rabble, team - crowd, horde, mob; angry mob, lynch mob, barbarian horde - relative, friends, acquaintances, neighbours, strangers - cults, extremists, fanatics, fundamentalists, militants - anarchy, riot; concert, demonstration, applause, cheer, fame, scandal - infant, baby, child, adolescent, teenager, youth, adult - chilhood, schooldays, youthful, middle-aged, elderly, old, senile - age group, age bracket, age of consent, come of age - in my younger days, in his/her day, in his/her heyday, {cut down in} in his/her prime, thirty something; over the hill, burnt out, past it, twilight years, ripe old age 【备注】 这只是本人读书的点滴收获。还有很多话题等待挖掘。让我感兴趣的是,既然人类通过语言来表达自己的观点、感受,生活中的喜怒哀乐,政治上的价值取向,那么,同一社区的人,是否有共同的认知模式或隐喻方式?如有,这种模式是如何构成的?不同社区的人,又是什么样的情况呢? 语料库、语言、文本、语篇、认知、隐喻、社会生活......
科学松鼠会 发表于 2011-03-03 07:26 作者:赵承渊 媳妇有孕以来,每晚临睡前“胎教”成了我义不容辞的功课。至今《红楼》已读完,《西游》读了半部,目标是争取在娃娃降生之前遍阅四大名著。话说这睡前夜读,名曰“胎教”,实则是给孕妇大人解闷催眠耳。往往我还在口水飞溅地念书,那厢孕妇已经呼呼入梦乡了。结果便是第二天还得从前一天晚她娘意识朦胧处重读,颇费工夫,我稍有微词,孕妇便祭起傲娇大招:我孕傻嘛,记不住昨晚念了啥,重念吧!我便只有遵命的份了。 虽说孕傻这说法流传甚广,但我毕竟有些心怀不甘——明明我亲眼看见是睡着了啊!倒不是为了逃避劳动,只是这事儿本来也值得较较真,既然大家经常这么说,那么看看有没有科学依据也是好的。在松鼠会搜索“孕傻”,果然蹦出瘦驼去年的一篇文章,文中提到不少文献,关于人类孕傻的调查似乎有很多支持的证据,不过动物实验却与此相反:怀孕、哺乳和经产大鼠在精神状态和迷宫中的表现都要优于“处女”大鼠。结论似乎仍处于模棱两可之中。然而不管怎么说,以动物实验的结论来推翻基于人类对象的调查研究,似乎仍嫌牵强,“孕傻”仍被作为一种现象写进了许多科学著作,并被媒体,流行书籍等采信并转载,传播越来越广。 正如瘦驼文中所说,“孕傻”是个跨文化的流行词汇,英文称为baby brain或placenta brain(胎盘脑?!)。国内的说法不知从何而起,英文文献笔者最早追溯到1968年Kane FJ Jr.等在Br J Psychiatry上发表的一篇题为Emotional and cognitive disturbance in the early puerperium(产褥早期情感和认知障碍)的文章。自那以后,关于孕产妇甚至做母亲的女性发生认知障碍,记忆力下降的文献报道就层出不穷,包括一些交叉研究和纵向对比研究。这些报道影响力巨大,甚至波及产后女性就业:某些用工单位会因此拒绝继续雇佣产后女性,怀孕往往意味着女性职业抱负的破碎。看来 “孕傻”可不仅仅是一个简单的调侃话题。 面对“孕傻”的观测结果,人们进行了多种推测,从体内激素变化到脑容量变化,科学家们尝试为“怀孕会导致认知功能受损”这一假说寻找依据。然而动物实验中大鼠们的表现使人们再次陷入疑惑之中。尽管衡量大鼠和人类认知功能的方法大相径庭,但是“绝大多数哺乳动物具有类似的母性行为,这些行为在人类和大鼠很可能为相同区域的脑组织所控制”(most mammals share similar maternal behaviours, which are probably controlled by the same brain regions in both humans and rats),这一认识使人们不能忽视动物实验的观测结果。 于是问题就归结在,人和大鼠相比,谁的实验结果更可靠?从1968年算起,孕傻的问题纠结了40多年,最终人们还是要重新审视这些声称观察到“孕傻”现象的文献报道,以便和不擅撒谎的大鼠们做个比照。结果发现,之前所引的文献,还是有很多毛病可挑:不是存在样本选择上的偏倚,就是缺乏必要的孕前认知功能的评估资料,要么就是缺乏远期随访的结果。总之毛病多多。难道“孕傻”的结果,都是人们观测偏倚的错觉吗? 2010年Helen Christensen等发表在Br J Psychiatry上的一篇题为Cognition in pregnancy and motherhood: prospective cohort study(关于孕期和母亲时代认知功能的前瞻性队列研究)的重磅文章似乎终于要将“孕傻”问题画一个句号了。这项始于1999年的研究对象是2404名 20-24岁的青年女性,共有1241名对象通过当年的基础评估,随后在2003年和2007年分别进行了两次随访,共有1058名对象完成了整个实验,历时八年。在实验过程中,有188名受试者成为母亲,有76名受试者在随访时正值孕期。只选择初产妇作为研究对象,全面评估了受试者认知速度,工作记忆,瞬时和延迟回忆等四方面能力并进行认知变化评分,同时注意一些控制变量如受教育水平、抑郁、焦虑、服药、睡眠情况等。经过统计学分析,结论是:无论孕中孕后,短期或是长期,女性的认知功能均无明显变化。只有一项测试表明在孕晚期女性的认知速度有所下降,但关于瞬时回忆,延迟回忆和工作记忆的测试均未见恶化迹象。 这项研究结果是否推翻了“孕傻”的结论?事情好像还不是那么简单。有些学者认为,孕妇常常会出现的注意力难以集中的问题,在实验室之中往往表现并不明显。这项实验和之前所作的其他实验一样,未能就某些潜在的基于动机和 气质的认知缺陷予以探讨。另外,此实验是基于三次面试的结果,未能就某一具体的怀孕时段或产后时段,比如产后32周认知情况如何进行评价,难以反映认知变化的动态规律——如果有的话。值得一提的是,在动物实验上观察到产后大鼠认知能力提高的现象并未在人类试验中观察到,还需进一步分析深层次原因,是社会条件、文化因素制约了女性发挥聪明才智呢,还是动物实验本身还有瑕疵?这些都是未来需要进一步研究的方向。 但是总体看来,之前盛传的女性在孕期和产后发生记忆力下降,认知功能障碍等结论并不严谨,“孕傻”看上去更像是一个以讹传讹的谣言。不过饶是如此,“不孕傻”的女性毫无疑问地仍是家庭中地位最高的人,身为人母的她们在工作能力上也不应被区别对待。 致谢: 瘦驼@科学松鼠会 《怀孕傻三年?》 2010-08-07 参考文献 Helen Christensen, Liana S. Leach, Andrew Mackinnon. Cognition in pregnancy and motherhood: prospective cohort study. The British Journal of Psychiatry 2010;196:126-132 Kane FJ Jr, Harman WJ Jr, Keeler MH. Emotional and cognitive disturbance in the early puerperium. Br J Psychiatry 1968 ;114(506):99-102. Kinsley CH, Lambert KG. The maternal brain. Sci Am 2006; 294: 72-9. 您也可能喜欢: 怀孕傻三年? 当科普的科幻尝起来是文学的 电视科普----想说爱你不容易 舌头能告诉我们什么? 从单基因糖尿病的故事看科普的意义 无觅
这是一本新书,由World Scientific出版公司于出版。但是,要说明的是,这本书是该出版公司出版的关于 信息研究的丛书 的第二部。目前该丛书已经预告到了第三部。具体说来,第一部的书名叫《信息理论——基础性、多样性和统一性》(Thoery of Information: Fundamentality,Deversity and Unification),第二部就是我现在要介绍的 《信息与计算——对信息和计算基础理解的科学与哲学文集》 (Information and Computation:Essays on Scientific and Philosophical Understanding of Foundation of Information and Computation)而第三部这是《涌现的信息:统一信息论架构大纲》(An Outline Unifified Theory of Information Framework)。 第一、三两部都是单人的专著,均可自成一说,可是信息这个问题是极其复杂的一个概念,如何理解信息可谓见仁见智。而第二部这体现出包容的内容。首先,它是一部文集,其中所选的17篇文章都是编者们(Gordana Dodig-Crnkovic Mark Burgin )精心挑选的,因而比较有代表性;其次,这些文章的作者也都是相关领域的领军人物,因而可以相对代表各个方面的研究现状。第三,编者本人也是在这个领域中极其活跃的人物。由于信息与计算目前无论是在国际还是在国内都属于“显学”,所以关注这些活跃人物的工作也无疑是重要的。 具体来说,信息是世界的基础结构,而计算就是动态地处理信息变化的过程。那么本书所提供的就是这个领域中,世界前沿的的成果。它为我们勾勒出一幅对信息和计算的未来发展大致轮廓,还有就是信息与计算和认知、信息学、生物学人工智能和信息技术的相互作用。 本书的目录: 1 Cybersemiotics and the Question of Knowledge (S Brier) 2 Information Dynamics in a Categorical Setting (M Burgin) 3 Mathematics as Biological Process (G Chaitin) 4 Information, Computation, Measurement and Irreversibility (J Collier) 5 From Descartes to Turing: The Computational Content of Supervenience (B Cooper) 6 On the Algorithmic Nature of the World (J-P Delahaye H Zenil) 7 A Dialogue Concerning Two Possible World Systems (G Dodig-Crnkovic V Mueller) 8 Does Computing Embrace Self-Organization? (W Hofkirchner) 9 Analysis of Information and Computation in Physics Explains Cognitive Paradigms: From Full Cognition to Laplace Determinism to Statistical Determinism to Modern Approach (V Kreinovich R Araiza) 10 Bodies ? Both Informed and Transformed (B J MacLennan) 11 Computation on Information, Meaning and Representations, an Evolutionary Approach (C Menant) 12 Interior Grounding, Reflection, and Self-Consciousness (M Minsky) 13 Insights into the Biological Computing (W Riofrio) 14 Super-Recursive Features of Natural Evolvability Processes and the Models for Computational Evolution (D Roglic) 15 A Sketch of a Modeling View of Computing (O Shagrir) 16 What's Information, for an Organism or Intelligent Machine? How Can a Machine or Organism Mean? (A Sloman) 17 Inconsistent Information as a Natural Phenomenon (C N J de Vey Mestdagh J H Hoepman)
1999 年创刊的Spatial Cognition and Computation《空间认知和计算》,ISSN: 1387-5868,季刊,美国(TAYLOR FRANCIS INC, 325 CHESTNUT ST, SUITE 800, PHILADELPHIA, USA, PA, 19106)出版, 2009年入选 Web of Science的Social Sciences Citation Index,目前在SSCI数据库可以检索到该期刊2008年的第8卷第1-4期到2009年的第9卷第1-3期共27篇论文。 27 篇文章包括学术论文23篇、社论3篇、评论1篇。 27 篇文章的主要国家分布:德国12篇、美国7篇、澳大利亚6篇、英国2篇等。 该刊是一多学科期刊,主要刊载空间探测与认知、计算、模拟等领域的相互关系的研究论文,涉及认知心理学、地理学、神经心理学、人工智能等。 1999 年第1卷至2000年第2卷该刊是在Springer出版,2001年第3卷开始改为Taylor Francis出版。如果需要阅读Spatial Cognition and Computation《空间认知和计算》1999年第1卷至2000年第2卷的文章请到Springer出版社的Springerlink数据库,网址: http://www.springerlink.com/content/103795/ 2001 年第3卷以后的文章请到Taylor Francis出版社的informaworld数据库,Spatial Cognition and Computation《空间认知和计算》网址: http://www.informaworld.com/smpp/title~db=all~content=t775653698 Spatial Cognition and Computation作者指南: http://www.informaworld.com/smpp/title~db=all~content=t775653698~tab=submit~mode=paper_submission_instructions Spatial Cognition and Computation在线投稿: http://mc.manuscriptcentral.com/hspcc Spatial Cognition and Computation 所有最后手稿应符合APA美国心理学协会出版手册(第5版)(American Psychological Association 5th ed.),APA是一种英文论文文体指南,准备APA格式的手稿帮助,请访问: http://www.apastyle.org/elecref.html Editorial Board Editors Anthony G. Cohn University of Leeds, United Kingdom Daniel R. Montello University of California at Santa Barbara, United States Advisory Board Christian Freksa - Universit t Bremen Ben Kuipers- University of Michigan Lynn Nadel- University of Arizona Barbara Tversky - Stanford University Editorial Board Thomas Barkowsky- Universit t Bremen Kenny Coventry - Northumbria University Michel Denis- Universite de Paris-Sud Max Egenhofer- University of Maine Andrew Frank- Technical University of Vienna Alinda Friedman- University of Alberta Mark Gahegan- University of Auckland Antony Galton- University of Exeter Christopher Habel- University of Hamburg Mary Hegarty -University of California at Santa Barbara Stephen Hirtle - University of Pittsburgh Roberta Klatzky- Carnegie Mellon University Markus Knauff- University of Giessen Werner Kuhn- University of Mnster Barbara Landau- Johns Hopkins University Stephen Levinson- Max Plank Institute Lynn Liben- Pennsylvania State University Hanspeter Mallot- Eberhard-Karls-University Tbingen David Mark- University of Buffalo Tim McNamara -Vanderbilt University Nora Newcombe- Temple University Patrick Pruch - Universit de la Mditerrane Herb Pick- University of Minnesota Juval Portugali- Tel Aviv University Jochen Renz- Australian National University Barry Smith- University of Buffalo Elizabeth S. Spelke- Harvard University David Uttal- Northwestern University Stephan Winter- University of Melbourne Michael Worboys- University of Maine Wai Yeap- AUT University
ScienceDaily (Aug. 1, 2009) Researchers at Georgia State University have found that diets high in fructose a type of sugar found in most processed foods and beverages impaired the spatial memory of adult rats. Amy Ross, a graduate student in the lab of Marise Parent, associate professor at Georgia State's Neuroscience Institute and Department of Psychology, fed a group of Sprague-Dawley rats a diet where fructose represented 60 percent of calories ingested during the day. She placed the rats in a pool of water to test their ability to learn to find a submerged platform, which allowed them to get out of the water. She then returned them to the pool two days later with no platform present to see if the rats could remember to swim to the platform's location. What we discovered is that the fructose diet doesn't affect their ability to learn, Parent said. But they can't seem to remember as well where the platform was when you take it away. They swam more randomly than rats fed a control diet. Fructose, unlike another sugar, glucose, is processed almost solely by the liver, and produces an excessive amount of triglycerides fat which get into the bloodstream. Triglycerides can interfere with insulin signaling in the brain, which plays a major role in brain cell survival and plasticity, or the ability for the brain to change based on new experiences. Results were similar in adolescent rats, but it is unclear whether the effects of high fructose consumption are permanent, she said. Parent's lab works with Timothy Bartness, Regents' Professor of Biology, and John Mielke of the University of Waterloo in Waterloo, Ontario, Canada to examine how diet influences brain function. Although humans do not eat fructose in levels as high as rats in the experiments, the consumption of foods sweetened with fructose which includes both common table sugar, fruit juice concentrates, as well as the much-maligned high fructose corn syrup has been increasing steadily. High intake of fructose is associated with numerous health problems, including insulin insensitivity, type II diabetes, obesity and cardiovascular disease. The bottom line is that we were meant to have an apple a day as our source of fructose, Parent said. And now, we have fructose in almost everything. Moderation is key, as well as exercise, she said. Exercise is a next step in ongoing research, and Parent's team will investigate whether exercise might mitigate the memory effects of high fructose intake. Her lab is also researching whether the intake of fish oil can prevent the increase of triglycerides and memory deficits. Results from that research will be presented by her graduate student Emily Bruggeman at the 2009 Society for Neuroscience meeting in Chicago this fall. 科学日报(20090716)乔治亚州大学的研究者们发现,饮食中果糖(存在于大多数加工食品或饮料中的一种糖)含量过高有损于成年大鼠的空间记忆能力。 Georgia 州神经研究所和心理学系的助理教授兼Marise Parent实验室的学者――Amy Ross,选用SD大鼠做实验,大鼠每天摄入能量的60%来自果糖。 她把实验大鼠放入特制的水池中,来检测大鼠空间识别能力。首先,在此水池中预先放入一被水淹没的平台,当大鼠爬上此平台后,大鼠就露出水平。二天后,她把实验大鼠再次放入平台已撤离的上述水池中,看看这些大鼠能否游到以前放置平台的位置。 我们发现,果糖饮食不会影响大鼠的学习能力。Parent实验室(发言人)说,但当我们撤离平台后,这些大鼠好象忘了平台放置的位置。与对照组大鼠相比,它们在水池中游地更无目的性 果糖与蔗糖(另一种糖)不一样,它只在肝脏降解,并大量产生一种能进入血流的脂类物质――甘油三酯。该脂类物质能干扰脑胰岛素的信号传导。脑胰岛素信号传导通路在脑细胞的存活与可塑性方面发挥着重要作用。或在脑接受新事物方面发挥重要作用。 她说,在成年大鼠的实验中,结果是相似的。但是,果糖的上述效应是否持久目前还不清楚。 研究饮食对脑功能影响的实验室除了Parent实验外,还有与之合作的生物学教授Regent的Timothy Bartness实验室、加拿大安大略省Waterloo大学的John Mielke实验室。 尽管人不会饮用象实验中大鼠摄入的(那么高水平的)果糖。但在人们摄入的食物当中,用果糖加甜食物(如both common table sugar,果汁浓缩液及对人有害的高果糖玉米糖浆)所占比例正逐渐增多。高摄入果糖可引起许多疾病,如胰岛素抵抗,II型糖尿病,肥胖及心血管疾病。 Parent说,我们推荐每天补充果糖的量以一个苹果为限。然而现在几乎每种食物都含有果糖。减少果糖的摄入与加强(自身体育)锻炼是最有效的办法。 体育锻炼(对降低果糖对机体的影响)是进一步研究的课题。Parent研究小组将研究体育锻炼是否能减轻高摄入果糖对记忆的影响。该实验室也着手研究,鱼油的摄入能否阻止(血中)甘油三酯水平的升高及对记忆力的损害。该实验结果将于2009年秋季的神经科学年会上由其学生Emily Bruggeman来陈述。