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.     

Theory learning as stochastic search in the language of thought

We present an algorithmic model for the development of children's intuitive theories within a hierarchical Bayesian framework, where theories are described as sets of logical laws generated by a probabilistic context-free grammar. We contrast our approach with connectionist and other emergentist approaches to modeling cognitive development. While their subsymbolic representations provide a smooth error surface that supports efficient gradient-based learning, our symbolic representations are better suited to capturing children's intuitive theories but give rise to a harder learning problem, which can only be solved by exploratory search. Our algorithm attempts to discover the theory that best explains a set of observed data by performing stochastic search at two levels of abstraction: an outer loop in the space of theories and an inner loop in the space of explanations or models generated by each theory given a particular dataset. We show that this stochastic search is capable of learning appropriate theories in several everyday domains and discuss its dynamics in the context of empirical studies of children's learning.     

Comparing Journal of Consumer Psychology and Journal of Consumer Research

The Journal of Consumer Psychology (JCP) was created in 1992 partly in response to what was interpreted by some as a shift in the methodological focus of the Journal of Consumer Research (JCR). JCR began to increase its accommodation of postmodern approaches, compared to more positivistic approaches. JCP's mission was thus perceived by some to be more methodologically homogeneous, with a greater focus on traditional experimental psychology in the domain of consumer behavior. Is JCR really more methodologically diverse than JCP? In what other ways are the two journals similar or different? This article attempts to answer these questions with an in‐depth analysis of the articles published in both journals from 1992 to 1998. The articles are reviewed in terms of overall output, authorship, content, research design, and data analysis. Conclusions and implications are included.     

La réparation psychique

Crime victims are confronted with a three-fold trauma at the physical, psychic and group levels. Indeed, trauma has a disruptive effect on the individual psyche as well as on the feeling of belonging to a larger community. As such, the crime victim must elaborate a personal questioning, while participating to a judicial proceeding aimed at restoring a feeling of belonging. Reparation thus takes place at the individual and collective levels, which implies simultaneously considering psychic reality and judicial reality, which confronts the crime victim with two positions and with two ways of functioning. After dealing with the trial's pacifying and resocializing functions, the three-times assistance support will be considered: formal and psychological preparation before the trial, assistance during the trial, post trial speech group therapy and information.     

Isolating Change And Invariance In Patterns Of Behavior

General approaches which may be taken in the study of change and in- variance in patterns of behavior are presented and compared. These are classified according to whether data are gathered by "ideal" means, in which the same entities are assessed with the same measuring devices on several occasions, or data are gathered by "less-than-ideal" means, in which change of occasion is confounded with change in the sample of entities or is confounded with change in the sample of measuring devices. Several estimation and hypothesis-testing procedures are presented as appropriate for use with data gathered in one or the other of these ways. These include Tucker's multi-mode factoring technique, his inter-battery factoring procedure, Meredith's methods for rotating to achieve factorial invariance and various techniques of use in evaluating the significance of difference of vectors and matrices. Newly developed least-squares techniques, based in part upon discriminant function principles, are presented in some detail. The various study-design and statistical-method possibilities are evaluated for the amount and kind of information they could provide about stable and dynamic patterns of behavior.     

An overview of Bayesian reasoning in the analysis of delay‐discounting data

Statistical inference (including interval estimation and model selection) is increasingly used in the analysis of behavioral data. As with many other fields, statistical approaches for these analyses traditionally use classical (i.e., frequentist) methods. Interpreting classical intervals and p‐values correctly can be burdensome and counterintuitive. By contrast, Bayesian methods treat data, parameters, and hypotheses as random quantities and use rules of conditional probability to produce direct probabilistic statements about models and parameters given observed study data. In this work, we reanalyze two data sets using Bayesian procedures. We precede the analyses with an overview of the Bayesian paradigm. The first study reanalyzes data from a recent study of controls, heavy smokers, and individuals with alcohol and/or cocaine substance use disorder, and focuses on Bayesian hypothesis testing for covariates and interval estimation for discounting rates among various substance use disorder profiles. The second example analyzes hypothetical environmental delay‐discounting data. This example focuses on using historical data to establish prior distributions for parameters while allowing subjective expert opinion to govern the prior distribution on model preference. We review the subjective nature of specifying Bayesian prior distributions but also review established methods to standardize the generation of priors and remove subjective influence while still taking advantage of the interpretive advantages of Bayesian analyses. We present the Bayesian approach as an alternative paradigm for statistical inference and discuss its strengths and weaknesses.     

Integrating Covariates into Social Relations Models: A Plausible Values Approach for Handling Measurement Error in Perceiver and Target Effects

The Social Relations Model (SRM) is a conceptual and analytical approach to examining dyadic behaviors and interpersonal perceptions within groups. In an SRM, the perceiver effect describes a person's tendency to perceive other group members in a certain way, whereas the target effect measures the tendency to be perceived by others in certain ways. In SRM research, it is often of interest to relate these individual SRM effects to covariates. However, the estimated individual SRM effects might not provide a very reliable measure of the true, unobserved SRM effects, resulting in distorted estimates of associations with other variables. This article introduces a plausible values approach that allows users to correct for measurement error when assessing the association of individual SRM effects with other individual difference variables. In the plausible values approach, the latent, true individual SRM effects are treated as missing values and are imputed from an imputation model by applying Bayesian estimation techniques. In a simulation study, the statistical properties of the plausible values approach are compared with two approaches that have been used in previous research. A data example from educational psychology is presented to illustrate how the plausible values approach can be implemented with the software WinBUGS.     

Bayesian journey-to-crime modelling of single and multiple crime-type series in Baltimore County,MD

This study tested different Bayesian Journey-to-Crime (JTC) models on a sample group of 850 serial offenders apprehended in Baltimore County, MD from 1993 to 1997. In this research, Bayesian JTC models were being used to predict the home locations of the offenders. The sample group data included 133 assaults, 90 burglaries, 497 larcenies, 81 robberies, and 49 vehicle thefts. The main question this research aimed to answer was whether the addition of crimes of a different type to an existing crime series of a single type would result in more accurate and/or precise Bayesian JTC models. The standard practice by law enforcement has been to consider the same-type crime series only when modelling the anchor point of the offender. Similarly, in research, geographic profiles have been constructed exclusively with the same-type crime series. The results of this study clearly indicated that the inclusion of crimes of a different type into a single crime-type series will result in significantly more accurate and more precise Bayesian JTC models. In contrast, crime series with predominantly assault and burglary showed results that were inconclusive or indicated no significant differences. These results should encourage law enforcement agencies to re-evaluate their standard practice of constructing geographic profiles with only the same-type crime series. Copyright © 2009 John Wiley & Sons, Ltd.     

Tackling Crime by Other Means

If crime is a social problem, then ways of preventing it must be sought. Punishment has been the traditional approach to preventing crime, either as a deterrent, or as a means of reforming the offender. Neither of these approaches is wholly acceptable. Even if deterrence punishment, as grounded in utilitarianism, effectively prevents crime, there may be other methods which produce better results. Reform fails to justify any form of punishment since not only does punishment not reform, but it interferes with the reformative process. Reform is limited in scope to those people who commit crimes and are caught. If crime is to be prevented, then it is necessary to go beyond the crime and the criminal and consider the social contexts in which people act, and the ways in which they learn to react to them. By shaping both the environment and people's responses to it, society can solve the problem of preventing crime, without the need for punishment.     

Unravelling Crime Series Patterns amongst Serial Sex Offenders: Duration,Frequency, and Environmental Consistency

Crime linkage and the investigation of behavioural consistency amongst serial offenders has been a flourishing field of research over the past decade or so, especially with respect to serial sex offenders. The emerging research in this field has often portrayed serial sex offenders as a single, distinct, and homogeneous group. Such an assumption, however, has never been empirically examined. Using a criminal career approach and a sample of 72 serial sex offenders who have committed a total of 361 sexual assaults on stranger victims, the current study aims to examine and describe subgroups of crime series patterns amongst serial sex offenders in terms of duration and frequency of offending. The level of environmental consistency display (i.e. offender's choice of crime location and characteristics of the crime site selected) across subgroups of crime series patterns is also examined. Study findings suggest the presence and heterogeneity of crime series patterns amongst serial sex offenders, advocating for the consideration of subgroups of crime series patterns when studying serial sex offenders. Moreover, the offenders' level of environmental consistency varies across the different crime series patterns identified, allowing for the identification of subgroups of offenders showing a higher or lower level of environmental consistency. Copyright © 2014 John Wiley & Sons, Ltd.     

Bayesian Group Agents and Two Modes of Aggregation

Suppose we have a group of Bayesian agents, and suppose that theywould like for their group as a whole to be a Bayesian agent as well. Moreover, suppose that thoseagents want the probabilities and utilities attached to this group agent to be aggregated from theindividual probabilities and utilities in reasonable ways. Two ways of aggregating their individual data areavailable to them, viz., ex ante aggregation and ex post aggregation. The former aggregatesexpected utilities directly, whereas the latter aggregates probabilities and utilities separately.A number of recent formal results show that both approaches have problematic implications. This studydiscusses the philosophical issues arising from those results. In this process, I hope to convincethe reader that these results about Bayesian aggregation are highly significant to decision theorists, butalso of immense interest to theorists working in areas such as ethics and political philosophy.     

A generative joint model for spike trains and saccades during perceptual decision-making

Theory development in both psychology and neuroscience can benefit by consideration of both behavioral and neural data sets. However, the development of appropriate methods for linking these data sets is a difficult statistical and conceptual problem. Over the past decades, different linking approaches have been employed in the study of perceptual decision-making, beginning with rudimentary linking of the data sets at a qualitative, structural level, culminating in sophisticated statistical approaches with quantitative links. We outline a new approach, in which a single model is developed that jointly addresses neural and behavioral data. This approach allows for specification and testing of quantitative links between neural and behavioral aspects of the model. Estimating the model in a Bayesian framework allows both data sets to equally inform the estimation of all model parameters. The use of a hierarchical model architecture allows for a model, which accounts for and measures the variability between neurons. We demonstrate the approach by re-analysis of a classic data set containing behavioral recordings of decision-making with accompanying single-cell neural recordings. The joint model is able to capture most aspects of both data sets, and also supports the analysis of interesting questions about prediction, including predicting the times at which responses are made, and the corresponding neural firing rates.     

A Bayesian approach to the evolution of perceptual and cognitive systems

We describe a formal framework for analyzing how statistical properties of natural environments and the process of natural selection interact to determine the design of perceptual and cognitive systems. The framework consists of two parts: a Bayesian ideal observer with a utility function appropriate for natural selection, and a Bayesian formulation of Darwin's theory of natural selection. Simulations of Bayesian natural selection were found to yield new insights, for example, into the co‐evolution of camouflage, color vision, and decision criteria. The Bayesian framework captures and generalizes, in a formal way, many of the important ideas of other approaches to perception and cognition.     

South African Serial Homicide: A Victim‐Focused Behavioural Typology

This current study assessed the value of the Interpersonal Model in differentiating serial homicide offences in South Africa, notably in terms of how the offender may display behavioural patterns that are indicative of them dealing with the victim as a Person, an Object, or a Vehicle. The sample consisted of 302 offences committed by 33 offenders that occurred from 1953 to 2007 in South Africa. Multidimensional scaling analysis was used to test the Interpersonal Model and results indicated that crime themes did not directly correspond to the threefold model. Instead, two crime scene types were present: Victim as Object (where the focus was on tangible gains, interaction with the victim throughout the phases of the crime was limited, and victims included men and live women) and Victim as Vehicle (where the victim might be construed as a conduit through whom the offender could realise their specific psychological needs, the interaction was extensive, and victims tended to be vulnerable). South African serial homicides did not appear to have a specific sub‐theme of Victim as Person, suggesting that the themes engaged in centred more on instrumentally focused actions but in qualitatively different ways. However, victim types were integral to the overall behavioural model. Using the Victim as Object/Vehicle dichotomy, 85.7% of the offences could be seen to engage dominantly in one of the two patterns at the crime scene. This not only supported the results of a twofold model, but having these many offenders showing a dominant crime scene type also shows that the model is an excellent representation of serial homicide offending in South Africa. Copyright © 2014 John Wiley & Sons, Ltd.     

Using response times to detect aberrant responses in computerized adaptive testing

A lognormal model for response times is used to check response times for aberrances in examinee behavior on computerized adaptive tests. Both classical procedures and Bayesian posterior predictive checks are presented. For a fixed examinee, responses and response times are independent; checks based on response times offer thus information independent of the results of checks on response patterns. Empirical examples of the use of classical and Bayesian checks for detecting two different types of aberrances in response times are presented. The detection rates for the Bayesian checks outperformed those for the classical checks, but at the cost of higher false-alarm rates. A guideline for the choice between the two types of checks is offered.This study received funding from the Law School Admission Council (LSAC). The opinions and conclusions contained in this paper are those of the authors and do not necessarily reflect the policy and position of LSAC. The authors are most indebted to Wim M. M. Tielen for his computational assistance and to the US Defense Manpower Data Center for the permission to use the ASVAB data set in the empirical examples.     

Balancing Justice and Mercy

The concept of forgiveness is analyzed as a moral gesture toward the offender designed to help restore that individual's moral standing. Jewish sources on the conditions under which forgiveness is obligatory are explored and two contrasting positions are presented: one in which the obligation to forgive is conditional on the repentance of the offender and another in which people are required to forgive unconditionally. These two positions are shown to represent different ways of framing the offending behavior that rest, in turn, on different ways of balancing the need for justice and for mercy respectively. In the final analysis, Judaism's two contrasting attitudes toward forgiveness are rooted in different theological assumptions and different ways of construing the very goals of the moral life. The author points out the merits and shortcomings of both positions and concludes with the suggestion that the two complement each other in important ways.     

Why is violent crime so common in the Americas?

Violent crimes, including murders, rapes, and assaults are substantially higher in the Americas than other regions of the world. This study investigated the role of single parenthood ratios in accounting for this regional variation in violent crime of 39 countries using INTERPOL data. It pitted the prediction of parental investment (calling for a delayed relationship between single parenthood and crime) against a mating aggression hypothesis that predicted a contemporaneous effect. Regression analyses found that current single parenthood ratios were strongly and consistently predictive of violent crimes whereas single parenthood ratios 18 years earlier were not and this conclusion remained following controls for income inequality and the population sex ratio. The results indicate that the regional difference in violent crime is best explained in terms of mating competition rather than reduced parental investment. Aggr. Behav. 32:1–9, 2006. © 2006 Wiley‐Liss, Inc.     

Observer Influences on Fathers and Mothers: An Experimental Manipulation of the Structure and Function of Parent – Infant Conversation

From studies over the past 20 years four contrasting hypotheses can be made about the nature of parent–infant communication: (1) mothers and fathers display similar skills to their infants and do not exert a differential influence on their development; (2) fathers are less sympathetic to their infants' level of development and therefore inadvertently stretch the child's development more than mothers; (3) both parents differentially socialize their sons' and daughters' early communicative skills; (4) any apparent differences between parents reflect their expectations about being observed. To examine these hypotheses together, this experiment records the communication of 10 mother–infant and 10 father–infant dyads in two conditions: when an observer was present or absent. The analysis revealed two patterns. Firstly, in keeping with most research on parent–child communication, mothers and fathers both simplified their speech to their infants in similar ways. Secondly, both the structure and function of parental communication showed differences between the two conditions and many of these differences were moderated by interactions between condition and sex of parent or child. The data thus provide more support for the first and fourth hypotheses cited above. It is suggested that analyses of parent–infant interaction should move away from simple assumptions about parental ‘influences’ upon children's development to consider the subtleties of different parental styles in different settings.     

Retributive and restorative justice: Importance of crime severity and shared identity in people's justice responses

Retributive and restorative justice present two different responses to wrongdoing: one that focuses on addressing the moral wrong through punitive sanctions (retribution) and one that focuses on addressing the harm that has been caused through reparative sanctions (restoration). Psychological investigations of what factors influence which justice outcome that people desire (retributive, restorative, or both) have focused on two constructs: crime severity and shared identity. The crime severity approach contends that people can have multiple justice goals, and which justice goals people want to fulfil is dependent on the salient features of the situation (such as offence seriousness). The shared identity approach argues that people's desire for restoration or retribution is dependent on the shared identity and perceptions of value consensus between offender and victim in the judgement context. The findings related to both of these factors are reviewed, and possibilities for future research integrating these two approaches are discussed.