Bayesian Factor Analysis as a Variable-Selection Problem: Alternative Priors and Consequences

Factor analysis is a popular statistical technique for multivariate data analysis. Developments in the structural equation modeling framework have enabled the use of hybrid confirmatory/exploratory approaches in which factor-loading structures can be explored relatively flexibly within a confirmatory factor analysis (CFA) framework. Recently, Muthén & Asparouhov proposed a Bayesian structural equation modeling (BSEM) approach to explore the presence of cross loadings in CFA models. We show that the issue of determining factor-loading patterns may be formulated as a Bayesian variable selection problem in which Muthén and Asparouhov's approach can be regarded as a BSEM approach with ridge regression prior (BSEM-RP). We propose another Bayesian approach, denoted herein as the Bayesian structural equation modeling with spike-and-slab prior (BSEM-SSP), which serves as a one-stage alternative to the BSEM-RP. We review the theoretical advantages and disadvantages of both approaches and compare their empirical performance relative to two modification indices-based approaches and exploratory factor analysis with target rotation. A teacher stress scale data set is used to demonstrate our approach.     

Childhood Interrupted: A Story of Loss,Separation, and Reconciliation

This essay presents a story of personal loss and childhood trauma experienced by the author in 1968. Written in autoethnographic form, the author narrates a particular time in his life when he lost his hearing and subsequently experienced “disrupted attachment”(Becker, 1997 Becker, G. (1997). Disrupted lives: How people create meaning in a chaotic world. Berkeley, CA: University of California Press. [Google Scholar]) caused by forced separation from family on the day he began life at a residential school for deaf children. Forty-six years later, the author weaves together a narrative of loss and trauma followed by his own reflections, showing how he used writing conversation as a source of healing that allowed him reconcile with his past.     

A caveat on the Savage–Dickey density ratio: The case of computing Bayes factors for regression parameters

The Savage–Dickey density ratio is a simple method for computing the Bayes factor for an equality constraint on one or more parameters of a statistical model. In regression analysis, this includes the important scenario of testing whether one or more of the covariates have an effect on the dependent variable. However, the Savage–Dickey ratio only provides the correct Bayes factor if the prior distribution of the nuisance parameters under the nested model is identical to the conditional prior under the full model given the equality constraint. This condition is violated for multiple regression models with a Jeffreys–Zellner–Siow prior, which is often used as a default prior in psychology. Besides linear regression models, the limitation of the Savage–Dickey ratio is especially relevant when analytical solutions for the Bayes factor are not available. This is the case for generalized linear models, non-linear models, or cognitive process models with regression extensions. As a remedy, the correct Bayes factor can be computed using a generalized version of the Savage–Dickey density ratio.     

Sex differences in facial emotion perception ability across the lifespan

Perception of emotion in the face is a key component of human social cognition and is considered vital for many domains of life; however, little is known about how this ability differs across the lifespan for men and women. We addressed this question with a large community sample (N?=?100,257) of persons ranging from younger than 15 to older than 60 years of age. Participants were viewers of the television show “Tout le Monde Joue”, and the task was presented on television, with participants responding via their mobile devices. Applying latent variable modeling, and establishing measurement invariance between males and females and across age, we found that, for both males and females, emotion perception abilities peak between the ages of 15 and 30, with poorer performance by younger adults and declining performance after the age of 30. In addition, we show a consistent advantage by females across the lifespan, which decreases in magnitude with increasing age. This large scale study with a wide range of people and testing environments suggests these effects are largely robust. Implications are discussed.     

潜变量建模的贝叶斯方法

贝叶斯统计是统计学的两大流派之一,近年来贝叶斯统计在社会及行为科学领域日益流行。鉴于国内心理学界对贝叶斯统计应用仍不广泛,本文尝试从非技术性的角度对贝叶斯统计用于潜变量建模的过程进行简要介绍。主要涉及贝叶斯与频率论在统计学基本概念上的对比;贝叶斯统计的基本原理和分析过程。最后以一个验证性因子分析为例,简要介绍贝叶斯统计用于潜变量建模的分析过程。希望本文能为国内心理学者进行潜变量建模提供新的视角。     

THE epistemologies of parsimony: A review of Ockham's razors: A user's manual by Elliott Sober

Sober analyzes two paradigms of parsimony that have been used successfully in science. These are associated with two interpretations of probability: Bayesian and frequentist. Sober applies these paradigms to problems in biology, psychology, and philosophy. In the chapter on psychology, he argues that objective data consisting of environmental input and two or more concurrent responses could be used to refute empirically the radical behaviorist thesis that probability of learned responses can be accounted for solely on the basis of environmental variables. Sober believes that such data are readily available and offers a thought experiment to illustrate his point. Behavior analysts, however, would want actual experimental data, undoubtedly with animals, before accepting any such refutation. Nonetheless, Sober's philosophical point about the type of experiment that would be capable of refuting this thesis is valid. The behavior analytic program, however, does not depend upon the truth of this thesis.     

Blockchain Technology as an Institution of Property

This paper argues that the practical implementation of blockchain technology can be considered an institution of property similar to legal institutions. Invoking Penner's theory of property and Hegel's system of property rights, and using the example of bitcoin, it is possible to demonstrate that blockchain effectively implements all necessary and sufficient criteria for property without reliance on legal means. Blockchains eliminate the need for a third‐party authority to enforce exclusion rights, and provide a system of universal access to knowledge and discoverability about the property rights of all participants and how the system functions. The implications of these findings are that traditional property relations in society could be replaced by or supplemented with blockchain models, and implemented in new domains.     

Parents’ experience of child contact within entrenched conflict families following separation and divorce: a qualitative study

Child contact arrangements with parents following separation and divorce are strongly endorsed for children in both public policy and law where safe, but can be difficult to sustain. Entrenched high-conflict post-separation relationships between parents can cause substantial emotional risks to children as well as impacting severely on parents’ mental health. This paper describes a qualitative study, aimed at examining parents’ experiences of contact arrangements post-separation, undertaken within a mixed methods random allocation study of therapeutic outcomes for parents in entrenched conflict over their children. Two established semi-structured interviews with 22 parents were jointly subjected to thematic analyses. A thematic analysis across interviews revealed three main themes: ‘Dealing with contact evokes extreme states of mind’ for parents; when speaking of contact, the child is ‘everywhere and nowhere’ in the parents’ minds; ‘the hardest thing about contact is dealing with my ex-partner’. These findings indicate the immense strain children and parents are under and shed much light on the desperate states of mind for parents, particularly the anxieties driving relentless child contact disputes. This paper may contribute to the understanding of parents’ experiences of contact arrangements post-separation, potentially providing important information which can inform best practice for professionals working with this population.     

Effect of anxiety on behavioural pattern separation in humans

Behavioural pattern separation (BPS), the ability to distinguish among similar stimuli based on subtle physical differences, has been used to study the mechanism underlying stimulus generalisation. Fear overgeneralisation is often observed in individuals with posttraumatic stress disorder and other anxiety disorders. However, the relationship between anxiety and BPS remains unclear. The purpose of this study was to determine the effect of anxiety (threat of shock) on BPS, which was assessed across separate encoding and retrieval sessions. Images were encoded/retrieved during blocks of threat or safety in a 2?×?2 factorial design. During retrieval, participants indicated whether images were new, old, or altered. Better accuracy was observed for altered images encoded during periods of threat compared to safety, but only if those images were also retrieved during periods of safety. These results suggest that overgeneralisation in anxiety may be due to altered pattern separation.