Discovery algorithms for hierarchical relations

Two algorithms based on a latent class model are presented for discovering hierarchical relations that exist among a set ofK dichotomous items. The two algorithms, stepwise forward selection and backward elimination, incorporate statistical criteria for selecting (or deleting) 0–1 response pattern vectors to form the subset of the total possible 2 ^k vectors that uniquely describe the hierarchy. The performances of the algorithms are compared, using computer-constructed data, with those of three competing deterministic approaches based on ordering theory and the calculation of Phi/Phi-max coefficients. The discovery algorithms are also demonstrated on real data sets investigated in the literature.     

Embodied AI, Creation, and Cog

This is a reply to comments on my paper Cog, a Humanoid Robot, and the Questions of the Image of God ; one was written by Mary Gerhart and Allan Melvin Russell, and another one by Helmut Reich. I will start with the suggested analogy of the relationship between God and us and the one between us and the humanoid robot Cog and will show why this analogy is not helpful for the dialogue between theology and artificial intelligence (AI). Such a dialogue can succeed only if both our fascination for humanoids and our fear of them are equally accepted. Any avoidance of these emotions, as well as any rejection of the possibility that Cog might one day be humanlike, destroy the dialogue. The interpretation of both scientific theories and religious metaphors as stories replaces seemingly "rational" arguments with the confession of the respective commitments to a body of stories and opens up a space for exchange and friendship between AI-researchers and theologians—an option that usually remains closed.     

Scientific societies and whistleblowers: the relationship between the community and the individual

Formalizing shared ethical standards is an activity of scientific societies designed to achieve a collective goal of promoting ethical conduct. A scientist who is faced with the choice of becoming a “whistleblower” by exposing misconduct does so in the context of these ethical standards. Examination of ethics policies of scientific societies which are members of the Council of Scientific Society Presidents (CSSP) shows a breadth of purpose and scope in these policies. Among the CSSP member societies, some ethics policies chiefly present the ethical culture of the community in an educational context and do not have enforcement procedures. Other policies are more comprehensive and include standards for certification, procedures for addressing ethical issues, and established sanctions. Of the 36 member societies of CSSP that have developed a code or adopted a code of another professional society, 18 specifically identified a responsibility to expose ethical misconduct, demonstrating an acknowledgment of the possible critical role of the whistleblower in addressing ethical issues. Scientific societies may revise their ethics codes based upon experience gained in addressing cases of ethical misconduct. In most cases, the action of a whistleblower is the initial step in addressing an ethics violation; the whistleblower may either be in the position of an observer or a victim, such as in the case of someone who discovers that his or her own work has been plagiarized. The ethics committee of a scientific society is one of several possible outlets through which the whistleblower can voice a complaint or concern. Ethical violations can include falsification, fabrication, plagiarism and other authorship disputes, conflict of interest and other serious violations. Commonly, some of these violations may involve publication in the scientific literature. Thus addressing ethical issues may be intertwined with a scientific society’s role in the dissemination of new scientific results. For a journal published by a scientific society, the editor can refer at some point to the ethics committee of the society. Whereas, in the case of a journal published by a commercial publisher, the editor may be without direct support of the associated scientific community in handling the case. The association of a journal with a scientific society may thus direct a whistleblower towards addressing the issue within the scientific community rather than involving the press or talking to colleagues who may gossip. A formal procedure for handling ethics cases may also discourage false accusers. Another advantage of handling complaints through ethics committees is that decisions to contact home institutions or funding agencies can be made by the ethics committee and are not the responsibility of the whistleblower or the editor of the journal. The general assessment is that the establishment of ethics policies, especially policies covering publication in society journals, will promote a culture supportive of whistleblowers and discouraging to false accusers. An earlier version of this paper was presented at the symposium entitled “Damned If You Do, Damned If You Don’t: What the Scientific Community Can Do About Whistleblowing” held during the Annual Meeting of the American Association for the Advancement of Science, Seattle, Washington, 15 February, 1997.     

Richard Rorty's realism

An examination of late Rorty shows that he does not abandon belief in an external world about which we can, and indeed must, acquire knowledge. His disapproval of the correspondence theory of truth does not involve the idea that anything other than local weather, for example, could falsify remarks about local weather. It is just that once we get done looking out the window or, if we are outside, feeling the right kind of drops make contact with our skin, there is nothing else we can do, nothing better, to make ourselves more certain, more cognitively secure. One can see this in the detailed work that enables scientific progress. Science improves itself by doing more of the same. G. E. Moore's famous open question stays open only as a reminder that our fallibility never disappears and that our cognitive security is never better than pro tem. Rorty, as a faithful pragmatist and undogmatic meliorist, thinks this is perfectly O.K.     

科学发明情境中问题提出的脑机制再探

以高生态学效度的科学发明问题情境作为实验材料, 采用静息态功能磁共振成像技术, 基于低频振幅(ALFF)和静息态功能连接(RSFC)的分析方法, 探讨创造性科学问题提出的脑机制。结果发现, 在控制了被试性别、年龄后, 提出新颖有效性问题的比率越高, 左内侧前额叶(Left media prefrontal cortex, L-mPFC)和右小脑前叶(Right cerebellum)的ALFF值越高。进一步功能连接分析发现, 提出新颖有效性问题的比率与mPFC和楔叶(Cuneus)之间的功能连接强度呈显著正相关。结果强调mPFC对于科学发明情境中问题提出的重要作用, 且更高比率的新颖有效性问题的提出是通过mPFC与其它脑区的协同联结来实现的。     

Do you believe what you have been told? Morality and scientific literacy as predictors of pseudoscience susceptibility

The aim of the current study was to examine the antecedents of susceptibility to pseudoscientific information. Participants were asked to assess their perceived probability of and familiarity with statements containing popular science misconceptions and pseudoscientific claims, which were fabricated for the study. Measures of scientific literacy, moral inclinations, need for closure, and a set of additional variables were collected. Analysis of the results showed that susceptibility to known pseudoscientific theories predicted falling for novel unwarranted claims. Scientific literacy strongly predicted lower pseudoscience susceptibility of all types. Finally, participants' moral inclinations were found to be a strong predictor of different types of pseudoscientific statements.