Methodological norms are seen as rules defining a competitive game, and it is argued that rational recognition-seeking scientists can reach a collective agreement about which specific norms serve better their individual interests, especially if the choice is made `under a veil of ignorance', i.e. , before knowing what theory will be proposed by each scientist. Norms for theory assessment are distinguished from norms for theory choice (or inference rules), and it is argued that pursuit of recognition only affects this (...) second type of rule. An inference rule similar to `eliminative induction' is defended on the basis of such a possible agreement. According to this contractarian approach, both the explanation and the justification of scientific norms only need to refer to the preferences of individual scientists, without assuming the existence of `collective' points of view. (shrink)

Do accounts of scientific theory formation and revision have implications for theories of everyday cognition? We maintain that failing to distinguish between importantly different types of theories of scientific inference has led to fundamental misunderstandings of the relationship between science and everyday cognition. In this article, we focus on one influential manifestation of this phenomenon which is found in Fodor's well-known critique of theories of cognitive architecture. We argue that in developing his critique, Fodor confounds a variety (...) of distinct claims about the holistic nature of scientific inference. Having done so, we outline more promising relations that hold between theories of scientific inference and ordinary cognition. (shrink)

This short paper serves as an introduction to a debate between representatives of two fundamentally different points of view regarding the nature of scientific inference. Colin Howson and Peter Urbach represent a Bayesian point of view and Deborah Mayo represents a version of classical statistics called error statistics. The paper begins by reviewing earlier versions of the same two points of view due to Rudolf Carnap and Hans Reichenbach, respectively. After a few remarks about philosophical approaches to understanding (...) scientific reasoning between 1960 and 1980, I turn to substantive differences between the two approaches. (shrink)

The sciences use a wide range of visual devices, practices, and imaging technologies. This diversity points to an important repertoire of visual methods that scientists use to adapt representations to meet the varied demands that their work places on cognitive processes. This paper identifies key features of the use of visualization in a range of scientific domains and considers the implications of this repertoire for understanding scientists as cognitive agents.

To learn about real world phenomena, scientists have traditionally used models with clearly interpretable elements. However, modern machine learning (ML) models, while powerful predictors, lack this direct elementwise interpretability (e.g. neural network weights). Interpretable machine learning (IML) offers a solution by analyzing models holistically to derive interpretations. Yet, current IML research is focused on auditing ML models rather than leveraging them for scientific inference. Our work bridges this gap, presenting a framework for designing IML methods—termed ’property descriptors’—that illuminate (...) not just the model, but also the phenomenon it represents. We demonstrate that property descriptors, grounded in statistical learning theory, can effectively reveal relevant properties of the joint probability distribution of the observational data. We identify existing IML methods suited for scientific inference and provide a guide for developing new descriptors with quantified epistemic uncertainty. Our framework empowers scientists to harness ML models for inference, and provides directions for future IML research to support scientific understanding. (shrink)

A scientific theory is originally based on a particular set of observations. How can it be extended to apply outside this original range of cases? This question, which is fundamental to natural philosophy, is considered in detail in this book, which was originally published in 1931, and first published as this third edition in 1973. Sir Harold begins with the principle that 'it is possible to learn from experience and to make inferences from beyond the data directly known to (...) sensation'. He goes on to analyse this principle, discuss its status and investigate its logical consequences. The result is a book of importance to anyone interested in the foundations of modern scientific method. His thesis provides a consistent account of how the theories proposed by physicists have been derived from, and are supported by, experimental data. (shrink)

Paradoxes are poems of science and philosophy that collectively allow us to address broad multidisciplinary issues within a microcosm. A true paradox is a source of creativity and a concise expression that delivers a profound idea and provokes a wild and endless imagination. The study of paradoxes leads to ultimate clarity and, at the same time, indisputably challenges your mind. Paradoxes in Scientific Inference analyzes paradoxes from many different perspectives: statistics, mathematics, philosophy, science, artificial intelligence, and more. The (...) book elaborates on findings and reaches new and exciting conclusions. It challenges your knowledge, intuition, and conventional wisdom, compelling you to adjust your way of thinking. Ultimately, you will learn effective scientific inference through studying the paradoxes. (shrink)

Inductive inference and "experimental error"; A population sample model: local inference; Expansion of the model: inference in the large; Expansion of the model: inference in the large; Interpretation of results; Random variables and distributions; Variance and related topics; Problems of sampling in physical situations; Randomization; Restricted randomization and experimental designs; Regression or curve fitting.

Not since Ernest Nagel’s 1939 monograph on the theory of probability has there been a comprehensive elementary survey of the philosophical problems of probablity and induction. This is an authoritative and up-to-date treatment of the subject, and yet it is relatively brief and nontechnical. Hume’s skeptical arguments regarding the justification of induction are taken as a point of departure, and a variety of traditional and contemporary ways of dealing with this problem are considered. The author then sets forth his own (...) criteria of adequacy for interpretations of probability. Utilizing these criteria he analyzes contemporary theories of probability, as well as the older classical and subjective interpretations. (shrink)

The Akaike Information Criterion can be a valuable tool of scientific inference. This statistic, or any other statistical method for that matter, cannot, however, be the whole of scientific methodology. In this paper some of the limitations of Akaikean statistical methods are discussed. It is argued that the full import of empirical evidence is realized only by adopting a richer ideal of empirical success than predictive accuracy, and that the ability of a theory to turn phenomena into (...) accurate, agreeing measurements of causally relevant parameters contributes to the evidential support of the theory. This is illustrated by Newton's argument from orbital phenomena to the inverse-square law of gravitation. (shrink)

This book would be very important indeed if Mr. Spencer Brown had substantiated his claims “that the concept of probability used in statistical science is meaningless in its own terms” , and that confirming this is the only significance of experiments in psychical research. The six short introductory chapters need not be discussed here. It is in Chapters VII to IX that the author develops his thesis that the concept of randomness is self–contradictory, and the statistician's concept of probability consequently (...) meaningless. I shall examine what I take to be the central argument leading to this conclusion. This is developed from a distinction between “primary chance or randomness” and “secondary chance or randomness” . The former concept is to be applicable only to individual events and is to depend upon their “unexpectedness or unpredictability”; the latter concept, applicable only to a series as such, is denned as “possessing no discernible pattern” . The definition of “primary randomness” is amplified, but not clarified, on page 49: “An event is primarily random in so far as... one cannot be sure of its occurrence... The only relevant criterion is that we are able to guess ”. We are then told that primary randomness “admits of analysis in subjective terms”, since the same event may be predictable by one person but unpredictable by another; whereas secondary randomness is “a more objective concept”. (shrink)

This work has been selected by scholars as being culturally important and is part of the knowledge base of civilization as we know it. This work is in the public domain in the United States of America, and possibly other nations. Within the United States, you may freely copy and distribute this work, as no entity has a copyright on the body of the work. Scholars believe, and we concur, that this work is important enough to be preserved, reproduced, and (...) made generally available to the public. To ensure a quality reading experience, this work has been proofread and republished using a format that seamlessly blends the original graphical elements with text in an easy-to-read typeface. We appreciate your support of the preservation process, and thank you for being an important part of keeping this knowledge alive and relevant. (shrink)

In spite of doubts in many quarters there seems to be considerable confidence in the benefits of health examinations. In my opinion it is important to analyze the structure and the purpose of the examinations in order to elucidate the practical thinking and logical quality of OHS. In this article I will conceive health examination as an information process and offer types of inference, i.e. prediction, abduction and induction, feasible for the analysis. I will study the logical conditions and (...) possibilities to draw conclusions using these inference methods and try, in a preliminary way, to show how the qualitative or quantitative formulations of presuppositions affect the formulation of the conclusion. Finally I will explicate the implicit purposes of the health examinations stipulated by Finnish legislation with the aid of the above inference methods. I find the methods feasible for analyzing the information process, and will set forth some conditions for presuppositions used with inference and point out the alternative purposes of Finnish occupational health examinations. (shrink)

Originally published as a long essay in Mind and Cosmos, Volume II of the University of Pittsburgh series in the philosophy of science, this study admirably fills the need for an elementary survey of problems in the area of probability and induction. But it is more than an introduction. The author is working on the general thesis that Bayes' theorem of the probability calculus holds the key to the understanding of scientific inference. Guided by this idea he attempts (...) to salvage what he thinks is of value in various current theories of probability and scientific inference, particularly, the frequency theory of probability as defended by Reichenbach, Popper's views on the falsifiability of theories, and Hanson's work on the logic of discovery. These attempts to develop a theory in the latter part of the book follow discussions of Hume's critique of induction, the probability calculus and five past and present interpretations of the calculus. This is a well-written and challenging introduction to the field.—R. H. K. (shrink)

Para entender por qué la interacción de la ciencia y el derecho puede ser tan problemática, no basta con apuntar vagamente hacia un supuesto contraste entre los “modos de pensamiento” científico y jurídico. Es necesario considerar, en cambio, las consecuencias de los distintos objetivos que tanto la ciencia como el derecho persiguen, así como las limitaciones bajo las cuales dichos objetivos son perseguidos, y las diferentes culturas que involucran a ambas empresas. Desde esa perspectiva, es posible observar no sólo por (...) qué el derecho en algunas ocasiones pide más de lo que la ciencia puede dar en tanto que en otras recibe menos de lo que la ciencia podría dar, sino también por qué la simple dicotomía “inferencia científica” vs. “razonamiento jurídico” es en realidad más engañosa que útil. (shrink)

This paper describes in some detail a pattern of justification which seems to be part of common sense logic and also part of the logic of scientific investigations. Calling this pattern “abduction,” the paper lays out an “abduction-prediction” model of scientific inference as an update to the traditional hypothetico-deductive model. According to this newer model, scientific theories receive their claims for acceptance and belief from the abductive arguments that support them, and the processes of scientific (...) discovery aim to develop theories with strong abductive support. It is suggested that the study of diagnosis presents a good opportunity for studying abduction under somewhat simpler and more reproducible conditions than occur in scientific discovery. A computer-based diagnostic system is described which provides a small-scale validation of the abduc-tion-prediction model by showing that a version of it can be made precise enough to be implemented and to perform correctly for diagnosis. (shrink)

In this work, we explore the epistemic import of the value-ladenness of Neyman-Pearson’s Theory of Testing Hypotheses by reconstructing and extending Daniel Steel’s argument for the legitimate influence of pragmatic values on scientific inference. We focus on how to properly understand N-P’s pragmatic value-ladenness and the epistemic reliability of N-P. We develop an account of the twofold influence of pragmatic values on N-P’s epistemic reliability and replicability. We refer to these two distinguished aspects as “direct” and “indirect”. We (...) discuss the replicability of experiments in terms of the indirect aspect and the replicability of outcomes in terms of the direct aspect. We argue that the influence of pragmatic values is beneficial to N-P’s epistemic reliability and replicability indirectly. We show that while the direct influence of pragmatic values can be beneficial, its negative effects on reliability and replicability are also unavoidable in some cases, with the direct and indirect aspects possibly being incongruent. (shrink)

In the Principles of Philosophy, Descartes explains several observable phenomena showing that they are caused by special arrangements of unobservable microparticles. Despite these microparticles being unobservable, many passages suggest that he was very confident that these explanations were correct. In other passages, however, Descartes points out that these explanations merely hold the status of “suppositions” or “conjectures” that could be wrong. My main goal in this chapter is to clarify this apparent conflict. I argue first that for Descartes it was (...) indeed possible to have knowledge of unobservable particles and structures, and that the possibility of natural explanations being wrong should be understood as these explanations not being absolutely certain, but only morally certain. I use the debate in contemporary philosophy of science between scientific realism and antirealism as a framework to understand how Cartesian explanations work. Especifically, I argue that Cartesian explanations rely on what Ernan McMullin calls retroduction, which is a mode of inference that justifies beliefs in concrete unobservable entities and processes, based on considerations such as simplicity, coherence, etc. This kind of justification is of common use among scientific realists. Thus, another goal of this chapter is to highlight the relevance of Descartes’ ideas about explanation in the contemporary debate on scientific realism. (shrink)

This paper addresses the extent to which quotidian cognition is like scientific inference by focusing on Jerry Fodor's famous analogy. I specifically consider and rebut a recent attempt made by Tim Fuller and Richard Samuels to deny the usefulness of Fodor's analogy. In so doing, I reveal some subtleties of Fodor's arguments overlooked by Fuller and Samuels and others. Recognizing these subtleties provides a richer appreciation of the analogy, allowing us to gain better traction on the issue concerning (...) the extent to which everyday cognition is like scientific inference. In the end, I propose that quotidian cognition is indeed like scientific inference, but not precisely in the way Fodor claims it is. (shrink)

One persistent challenge in scientific practice is that the structure of the world can be unstable: changes in the broader context can alter which model of a phenomenon is preferred, all without any overt signal. Scientific discovery becomes much harder when we have a moving target, and the resulting incorrect understandings of relationships in the world can have significant real-world and practical consequences. In this paper, we argue that it is common (in certain sciences) to have changes of (...) context that lead to changes in the relationships under study, but that standard normative accounts of scientific inquiry have assumed away this problem. At the same time, we show that inference and discovery methods can “protect” themselves in various ways against this possibility by using methods with the novel methodological virtue of “diligence.” Unfortunately, this desirable virtue provably is incompatible with other desirable methodological virtues that are central to reliable inquiry. No scientific method can provide every virtue that we might want. (shrink)

Erratum to: Synthese DOI 10.1007/s11229-014-0408-3Appendix 1: NotationLet \(X\) represent a sequence of data, and let \(X_B^t\) represent an i.i.d. subsequence of length \(t\) of data generated from distribution \(B\).We conjecture that the i.i.d. assumption could be eliminated by defining probability distributions over sequences of arbitrary length, though this complication would not add conceptual clarity. Let \(\mathbf{F}\) be a framework (in this case, a set of probability distributions or densities).Let any \(P(\,)\) functions be either a probability distribution function or probability density (...) function, as appropriate. Let \(M_\mathbf{F}\) be a method that takes a data sequence \(X\) as input and outputs a distribution. (shrink)

This paper derives a mathematical expression giving the development of the probability of a scientific hypothesis with the number of confirming tests, as determined by Bayes's theorem, in a special case in which all the tests are "independent" of one another. The simple expression obtained shows clearly how the various factors influence the growth of the probability. The result is used to set a numerical lower bound on the probabilities representing the a priori beliefs of humans in generalizations that (...) become accepted. By making a comparison with the predictions of a "logical atomic" model in the case of physical laws, it is argued that humans have significant a priori "knowledge" in a weak sense. (shrink)