Kinds Of Inferences Research Articles

Response heterogeneity: Challenges for personalised medicine and big data approaches in psychiatry and chronic pain.

Response rates to available treatments for psychological and chronic pain disorders are poor, and there is a substantial burden of suffering and disability for patients, who often cycle through several rounds of ineffective treatment. As individuals presenting to the clinic with symptoms of these disorders are likely to be heterogeneous, there is considerable interest in the possibility that different constellations of signs could be used to identify subgroups of patients that might preferentially benefit from particular kinds of treatment. To this end, there has been a recent focus on the application of machine learning methods to attempt to identify sets of predictor variables (demographic, genetic, etc.) that could be used to target individuals towards treatments that are more likely to work for them in the first instance. Importantly, the training of such models generally relies on datasets where groups of individual predictor variables are labelled with a binary outcome category - usually 'responder' or 'non-responder' (to a particular treatment). However, as previously highlighted in other areas of medicine, there is a basic statistical problem in classifying individuals as 'responding' to a particular treatment on the basis of data from conventional randomized controlled trials. Specifically, insufficient information on the partition of variance components in individual symptom changes mean that it is inappropriate to consider data from the active treatment arm alone in this way. This may be particularly problematic in the case of psychiatric and chronic pain symptom data, where both within-subject variability and measurement error are likely to be high. Here, we outline some possible solutions to this problem in terms of dataset design and machine learning methodology, and conclude that it is important to carefully consider the kind of inferences that particular training data are able to afford, especially in arenas where the potential clinical benefit is so large.

Response heterogeneity: Challenges for personalised medicine and big data approaches in psychiatry and chronic pain

Response rates to available treatments for psychological and chronic pain disorders are poor, and there is a substantial burden of suffering and disability for patients, who often cycle through several rounds of ineffective treatment. As individuals presenting to the clinic with symptoms of these disorders are likely to be heterogeneous, there is considerable interest in the possibility that different constellations of signs could be used to identify subgroups of patients that might preferentially benefit from particular kinds of treatment. To this end, there has been a recent focus on the application of machine learning methods to attempt to identify sets of predictor variables (demographic, genetic, etc.) that could be used to target individuals towards treatments that are more likely to work for them in the first instance. Importantly, the training of such models generally relies on datasets where groups of individual predictor variables are labelled with a binary outcome category − usually ‘responder’ or ‘non-responder’ (to a particular treatment). However, as previously highlighted in other areas of medicine, there is a basic statistical problem in classifying individuals as ‘responding’ to a particular treatment on the basis of data from conventional randomized controlled trials. Specifically, insufficient information on the partition of variance components in individual symptom changes mean that it is inappropriate to consider data from the active treatment arm alone in this way. This may be particularly problematic in the case of psychiatric and chronic pain symptom data, where both within-subject variability and measurement error are likely to be high. Here, we outline some possible solutions to this problem in terms of dataset design and machine learning methodology, and conclude that it is important to carefully consider the kind of inferences that particular training data are able to afford, especially in arenas where the potential clinical benefit is so large.

An improved method for solving Hybrid Influence Diagrams

While decision trees are a popular formal and quantitative method for determining an optimal decision from a finite set of choices, for all but very simple problems they are computationally intractable. For this reason, Influence Diagrams (IDs) have been used as a more compact and efficient alternative. However, most algorithmic solutions assume that all chance variables are discrete, whereas in practice many are continuous. For such ‘Hybrid’ IDs (HIDs) the current-state-of-the-art algorithms suffer from various limitations on the kinds of inference that can be performed. This paper presents a novel method that overcomes a number of these limitations. The method solves a HID by transforming it to a Hybrid Bayesian Network (HBN) and carrying out inference on this HBN using Dynamic Discretization (DD). It generates a simplified decision tree from the propagated HBN to compute and present the optimal decisions under different decision scenarios. To provide satisfactory performance the method uses ‘inconsistent evidence’ to model functional and structural asymmetry. By using the entire marginal probability distribution of the continuous utility and chance nodes, rather than expected values alone, our method also enhances decision analysis by offering the possibility to consider additional statistics other than expected utility, such as measures of risk. We illustrate our method by using the oil wildcatter example and its variations with continuous nodes. We also use a financial score to combine risk and return measures, for illustration.

Scalar and Ignorance Inferences Are Both Computed Immediately upon Encountering the Sentential Connective: The Online Processing of Sentences with Disjunction Using the Visual World Paradigm.

Accounts based on the pragmatic maxim of quantity make different predictions about the computation of scalar versus ignorance inferences. These different predictions are evaluated in two eye-tracking experiments using a visual world paradigm to assess the on-line computation of inferences. The test sentences contained disjunction phrases, which engender both kinds of inferences. The first experiment documented that both inferences are computed immediately upon encountering the disjunctive connective, at nearly identical temporal locations. The second experiment was designed to determine whether or not there exists an intermediate stage at which the truth of the corresponding conjunction phrase is ignored. No such stage was found.

Response heterogeneity: Challenges for personalised medicine and big data approaches in psychiatry and chronic pain

Response rates to available treatments for psychological and chronic pain disorders are poor, and there is a considerable burden of suffering and disability for patients, who often cycle through several rounds of ineffective treatment. As individuals presenting to the clinic with symptoms of these disorders are likely to be heterogeneous, there is considerable interest in the possibility that different constellations of signs could be used to identify subgroups of patients that might preferentially benefit from particular kinds of treatment. To this end, there has been a recent focus on the application of machine learning methods to attempt to identify sets of predictor variables (demographic, genetic, etc.) that could be used to target individuals towards treatments that are more likely to work for them in the first instance. Importantly, the training of such models generally relies on datasets where groups of individual predictor variables are labelled with a binary outcome category - usually 'responder' or 'non-responder' (to a particular treatment). However, as previously highlighted in other areas of medicine, there is a basic statistical problem in classifying individuals as 'responding' to a particular treatment on the basis of data from conventional randomized controlled trials. Specifically, insufficient information on the partition of variance components in individual symptom changes mean that it is inappropriate to consider data from the active treatment arm alone in this way. This may be particularly problematic in the case of psychiatric and chronic pain symptom data, where both within-subject variability and measurement error are likely to be high. Here, we outline some possible solutions to this problem in terms of dataset design and machine learning methodology, and conclude that it is important to carefully consider the kind of inferences that particular training data are able to afford, especially in arenas where the potential clinical benefit is so large.

Eti zōsa phlox: Inferring Divine Presence in Euripides’ Bacchae

It has long been recognised by interpreters that Dionysus’ meanings are peculiarly connected with his presence (parousia). Myths of the god typically recount a divine presence undiscerned or misdiagnosed by mortals, with tragic consequences. The vengeful Dionysus of Euripides’ Bacchae comes amongst humans disguised, ironically, to make them learn and recognise his divine identity and that they are in the presence of the truly, if not at first apparently, immortal. The kind of inference that it seems mortals are expected to rely upon in the vicinity of Dionysus in order to secure that all-important knowledge and recognition, is explored in this paper, which is offered as a contribution towards the development of a contemporary anthropology of Greek tragedy and its god. The anthropology of art and agency of Alfred Gell is mobilised in order to explore how different modes of inference are implicitly compared and evaluated in the drama and to suggest what this has to do with the nature of Dionysiac experience as represented by the tragic poet. The inspired inference it requires, envisioned by Euripides at the end of the 5th century, is equated with the logical process of abduction as defined and employed in philosophy, semiotics and the anthropology of art.

Behavioral and Neural Evidence on Pragmatic Processing: A Foreword to XPRAG.it

Behavioral and Neural Evidence on Pragmatic Processing: A Foreword to XPRAG.it

Energy Efficiency Optimization of Secondary Network Considering Primary User Return With Alternating-Phase-Type Traffic

Much work has been done in the area of cognitive radio networks to formulate the inherent tradeoff among different metrics, such as sensing and throughput. Primary user (PU) traffic behavior can affect the mentioned tradeoff and must be considered in analyses, especially when cognitive users coexist with the dynamic-traffic PU. In this case, PU-returns to the spectrum during secondary users' (SUs) transmission are more likely and, hence, such kinds of interference are inevitable. Furthermore, how to model the PU traffic is important. Note that there are two kinds of interference between SU and PU in cognitive radio networks, one due to sensing error and the other due to primary user reoccupancy. In this paper, a new metric called energy per successful transmission time is defined to address both kinds of inference by formulating SUs' successful transmission time as well as energy consumption. Furthermore, collision probability and average packet delay are formulated considering these interference metrics and the primary user traffic is modeled as an alternating phase type renewal process by which many traffic behaviors, such as long range dependency and self-similarity can be modeled. Finally, some numerical examples are given and the corresponding curves for different metrics are discussed.

Two-stage and sequential sampling for estimation and testing with prescribed precision

Statistical data analysis includes several phases. First, there is the phase of data collection. Second, there is the phase of analysis and inference. The two phases are interconnected. There are two types of data analysis. One type is called parametric and the other type is nonparametric. In the present paper, we discuss parametric inference. In parametric inference, we model the results of a given experiment as realization of random variables having a particular distribution, which is specified by its parameters. A random sample is a sequence of independent and identically distributed (i.i.d.) random variables. Statistics are functions of the data in the sample, which do not involve unknown parameters. A statistical inference is based on statistics of a given sample. We discuss two kinds of parametric inference. Estimating the values of parameters, or testing hypotheses concerning the parameters in either kind of inference, we are concerned with the accuracy and precision of the results. In estimation of parameters, the results are precise if, with high probability, they belong to a specified neighborhoods of the parameters. In testing hypotheses, one has to decide which one of two or several hypotheses should be accepted. Hypotheses which are not accepted are rejected. We distinguish between two types of errors. Type I error is the one committed by rejecting a correct hypothesis. Type II is that of accepting a wrong hypothesis. It is desired that both types of errors will occur simultaneously with small probabilities. Both precision in estimation or small error probabilities in testing depend on the statistics used (estimators or test functions) and on the sample size. In this paper, we present sampling procedures that attain the desired objectives. In Sec. 2, we discuss estimation of the parameters of a binomial distribution. In Sec. 3, more general results about estimation of expected values are presented. In Sec. 4, we discuss the Wald Sequential Probability Ratio Test (SPRT), which has optimal properties for testing two simple hypotheses.

Relating accumulator model parameters and neural dynamics

Accumulator models explain decision-making as an accumulation of evidence to a response threshold. Specific model parameters are associated with specific model mechanisms, such as the time when accumulation begins, the average rate of evidence accumulation, and the threshold. These mechanisms determine both the within-trial dynamics of evidence accumulation and the predicted behavior. Cognitive modelers usually infer what mechanisms vary during decision-making by seeing what parameters vary when a model is fitted to observed behavior. The recent identification of neural activity with evidence accumulation suggests that it may be possible to directly infer what mechanisms vary from an analysis of how neural dynamics vary. However, evidence accumulation is often noisy, and noise complicates the relationship between accumulator dynamics and the underlying mechanisms leading to those dynamics. To understand what kinds of inferences can be made about decision-making mechanisms based on measures of neural dynamics, we measured simulated accumulator model dynamics while systematically varying model parameters. In some cases, decision-making mechanisms can be directly inferred from dynamics, allowing us to distinguish between models that make identical behavioral predictions. In other cases, however, different parameterized mechanisms produce surprisingly similar dynamics, limiting the inferences that can be made based on measuring dynamics alone. Analyzing neural dynamics can provide a powerful tool to resolve model mimicry at the behavioral level, but we caution against drawing inferences based solely on neural analyses. Instead, simultaneous modeling of behavior and neural dynamics provides the most powerful approach to understand decision-making and likely other aspects of cognition and perception.

Restoring the integrative value to the notion of executive function. Commentary on: “Advancing understanding of executive function impairments and psychopathology: bridging the gap between clinical and cognitive approaches”

Restoring the integrative value to the notion of executive function. Commentary on: “Advancing understanding of executive function impairments and psychopathology: bridging the gap between clinical and cognitive approaches”

How Visual Displays Affect Cognitive Processing

We regularly consult and construct visual displays that are intended to communicate important information. The power of these displays and the instructional messages we attempt to comprehend when using them emerge from the information included in the display and by their spatial arrangement. In this article, we identify common types of visual displays and the kinds of inferences that each type of display is designed to promote. In particular, we outline different types of semantic and pictorial displays. Then, we describe four main ways in which visual displays can affect cognitive processing including selection, organization, integration, and processing efficiency and how semantic and pictorial displays support these types of processing. We conclude with seven recommendations for designing visual displays and possible directions for future research.

Obligations and Conditionals

The paper considers two kinds of medieval obligational disputations (positio, rei veritas) and the medieval genre of sophismata in relation to the kinds of inferences accepted in them. The main texts discussed are the anonymous Obligationes parisienses from the early 13th century and Richard Kilvington’s Sophismata from the early 14th century. Four different kinds of warranted transition from an antecedent to a consequent become apparent in the medieval discussions: (1) the strong logical validity of basic propositional logic, (2) analytic validity based on conceptual containment, (3) merely semantic impossibility of the antecedent being true without the consequent, and (4) intuitively true counterfactual conditionals. As these different kinds of consequences are spelled out by means of obligational disputations, it appears that the genre of obligations is indeed useful for the “knowledge of consequences,” as the anonymous Obligationes parisienses claims.

Statistical mutual conversion between whole body motion primitives and linguistic sentences for human motions

This paper describes a novel approach to linguistic mutual inference, which enables robots not only to linguistically interpret the motion patterns in the form of sentences but also to generate the motions from the sentences. The inference can be established based on two modules, the motion language model and the natural language model. The motion language model stochastically represents an association structure between symbols of motion patterns and the words in sentences assigned to the motion. This is a statistical model with a three layered structure of motion symbols, latent states and words. The natural language model statistically represents a structure of sentences based on word bigrams. The motion language model and the natural language model correspond to semantics and syntax respectively. An approach to the integration of motion language model with the natural language model allows the linguistic mutual inference for the robots. The two kinds of inference can be made by solving search problems, search for a sequence of words corresponding to a motion and search for a symbol of motion pattern corresponding to a sentence. The proposed approach to interpretation of motion patterns as sentences and generation of motion patterns from the sentences through the integration of motion language model with the natural language model is validated by an experiment on the human behavioral data.

Matrix equations for normalizing factors in local a posteriori inference of truth estimates in algebraic Bayesian networks

A posteriori inference is one of three kinds of inference that underlie the processing of knowledge patterns with probabilistic uncertainty in intelligent decision making systems using algebraic Bayesian networks (ABNs). In this paper, the key terms and formulations of theorems describing local a posteriori inference in algebraic Bayesian networks are given in a matrix-vector language. The main result is that matrix equations are constructed for normalizing factors appearing in the formulas of a posteriori probabilities of proposition-quanta and ideals of conjuncts. The matrix equations of local a priori inference formulated in general not only simplify the preparation of specifications of appropriate inference algorithms and make their implementation more transparent, but also open a possibility for application of classical mathematical techniques to the analysis of the properties of inference results.

Ontology-Based Interpretation of Natural Language Philipp Cimiano, Christina Unger, and John McCrae (University of Arminia Bielefeld, Germany) Morgan & Claypool, Synthesis Lectures on Human Language Technologies, March 2014, 178 pages, (doi:10.2200/S00561ED1V01Y201401HLT024) , $45.00

<b>Ontology-Based Interpretation of Natural Language</b> <b>Philipp Cimiano, Christina Unger, and John McCrae</b> (University of Arminia Bielefeld, Germany) Morgan &amp; Claypool, Synthesis Lectures on Human Language Technologies, March 2014, 178 pages, (doi:10.2200/S00561ED1V01Y201401HLT024) , $45.00

English and Film: Connecting Children to the World

This paper explores the processes behind drawing different kinds of inference from a single short film. It examines the range knowledge that groups of viewers are able to derive from listening to film sound, and from summarizing the work of a whole film, using the concepts of ‘aesthetic’ and ‘efferent’ reading set out by Louise Rosenblatt. The activities described come from work with groups of PGCE English students, compared with similar activities carried out with primary-aged children.

Children's Knowledge of Free Choice Inferences and Scalar Implicatures

This article presents experimental results showing that 4- and 5-year-old children are capable of drawing free choice inferences from disjunctive statements and from statements containing free choice indefinites, despite not being able to compute inferences of exclusivity for disjunctive statements, or other scalar implicatures. The findings appear to challenge accounts that attempt to unify the two kinds of inferences (Kratzer & Shimoyama 2002; Alonso Ovalle 2005; Fox 2007; Klinedinst 2007; Chemla 2010; van Rooij 2010; Franke 2011; Chierchia 2013). We discuss, however, the compatibility of the child data with a recent approach in the experimental literature, which attributes children's failures to compute scalar implicatures to a difficulty with alternatives (Chierchia et al. 2001; Gualmini et al. 2001; Reinhart 2006; Barner et al. 2011; Singh et al. 2013). Based on the results of two experiments, we propose an explanation for children's selective success on scalar inferences, according to which scalar inferences are generally unproblematic for children, unless they necessitate lexical retrieval of the required alternatives.

Moral inferentialism and the Frege-Geach problem

Despite its many advantages as a metaethical theory, moral expressivism faces difficulties as a semantic theory of the meaning of moral claims, an issue underscored by the notorious Frege-Geach problem. I consider a distinct metaethical view, inferentialism, which like expressivism rejects a representational account of meaning, but unlike expressivism explains meaning in terms of inferential role instead of expressive function. Drawing on Michael Williams’ recent work on inferential theories of meaning, I argue that an appropriate understanding of the pragmatic role of moral discourse—the facilitation of coordinated social behavior—suggests the kind of inferences we should expect terms with this function to license. I offer a sketch of the inferential roles the moral ‘ought’ plays, and argue that if we accept that the relevant inferential roles are meaning-constitutive, we will be in a position to solve the Frege-Geach problem. Such an inferentialist solution has advantages over those forwarded by expressivists such as Blackburn and Gibbard. First, it offers a more straightforward explanation of the meaning of moral terms. It also gives simple answers to at least two semantic worries that have vexed contemporary expressivists—the “problem of permissions” and the commitment to “mentalism”, both of which I argue are problems that don’t get traction with an inferentialist approach. I conclude by considering ways in which this approach can be expanded into a more robust semantic account.

Abduction: A categorical characterization

Scientific knowledge is gained by the informed (on the basis of theoretic ideas and criteria) examination of data. This can be easily seen in the context of quantitative data, handled with statistical methods. Here we are interested in other forms of data analysis, although with the same goal of extracting meaningful information. The idea is that data should guide the construction of suitable models, which later may lead to the development of new theories. This kind of inference is called abduction and constitutes a central procedure called Peircean qualitative induction. In this paper we will present a category-theoretic representation of abduction based on the notion of adjunction, which highlights the fundamental fact that an abduction is the most efficient way of capturing the information obtained from a large body of evidence.