Bayesian inference for psychology,part III: Parameter estimation in nonstandard models

We demonstrate the use of three popular Bayesian software packages that enable researchers to estimate parameters in a broad class of models that are commonly used in psychological research. We focus on WinBUGS, JAGS, and Stan, and show how they can be interfaced from R and MATLAB. We illustrate the use of the packages through two fully worked examples; the examples involve a simple univariate linear regression and fitting a multinomial processing tree model to data from a classic false-memory experiment. We conclude with a comparison of the strengths and weaknesses of the packages. Our example code, data, and this text are available via https://osf.io/ucmaz/.     

Hare Psychopathy Checklist-Revised

The Hare Psychopathy Checklist-Revised (PCL-R) is among the most frequently used instruments for the assessment of psychopathic traits. In the scope of the German adaptation of the PCL-R, normative data were collected from a sample of adult male offenders. Based on these data the factor structure and reliability of the German language version of the PCL-R are discussed in the first section of the article. The assessment of a model with four facets and two superordinate factors showed a very good agreement with the empirical data. In addition, examination of interrater agreement and internal consistency indicated that the PCL-R is a reliable measure. In the second part of the paper recommendations are provided for the correct application of the PCL-R. The focus lies on the correct performance, evaluation and interpretation of PCL-R test results with respect to key indices and the dimensional structure of the psychopathy construct.     

Escalation and Regulation of Emotional Arousal in Couples Predicts Relationship Satisfaction Concurrently and 25 Years Later

Relationship distress and divorce are major risk factors for the development or exacerbation of psychopathology and psychosocial impairments. Given that heightened negative emotions within couples’ interactions may portend negative relationship outcomes, it is critical to understand how emotions unfold across a conversation and how partners may influence each other’s immediate emotional experiences. This study examined whether these regulatory dynamics within one interaction predicted relationship satisfaction concurrently and 25 years later. Vocally‐encoded emotional arousal (f₀) was measured during couples’ (N = 25 couples) conversations about a relationship issue. Across different analytical strategies, results demonstrate that one partner’s f₀ dynamics had immediate and long‐term associations with the other partner’s satisfaction. Partners were less satisfied if the other partner (a) expressed higher f₀ overall and (b) escalated more in f₀ across the conversation. Yet, partners were more satisfied when their f₀ escalated across the conversation. Also, women specifically were more satisfied if their f₀ remained elevated longer before regulating back to their emotional baseline. Thus, higher f₀ was associated with higher satisfaction in the same partner, but associated with less satisfaction in the other partner—particularly when these emotions come from women. It may be that partners have to decide whether to prioritize expressing their emotions fully or limit expression in the service of their partner’s happiness. These findings challenge us to think of ways to address this “win–lose” scenario so that couples can balance both partners’ emotional needs and preserve relationship quality across the life span.     

Responses to exclusion are moderated by its perceived fairness

Experimental exclusion manipulations may induce exclusion in a way that participants perceive as unfair. Groups often use exclusion punitively to correct inappropriate behavior, however, which may lead to perceptions that it is potentially justified or fair. The current studies examined if individuals' perceptions of fairness with respect to an exclusion experience moderated their reactions. Participants wrote about or imagined a time in which they were excluded after they did something wrong (fair exclusion) or excluded even though they did nothing wrong (unfair exclusion) or about a mundane experience unrelated to exclusion (control). Compared with fair exclusion, unfair exclusion resulted in significantly weaker efficacy needs satisfaction (Studies 1, 2, and 4), greater antisocial intent (Study 3), and greater sensitivity to signs of interpersonal acceptance and rejection in a visual search task (Study 4). These results suggest that it is important to consider the role of perceived fairness in shaping responses to exclusion.     

A Bayesian approach to mitigation of publication bias

The reliability of published research findings in psychology has been a topic of rising concern. Publication bias, or treating positive findings differently from negative findings, is a contributing factor to this “crisis of confidence,” in that it likely inflates the number of false-positive effects in the literature. We demonstrate a Bayesian model averaging approach that takes into account the possibility of publication bias and allows for a better estimate of true underlying effect size. Accounting for the possibility of bias leads to a more conservative interpretation of published studies as well as meta-analyses. We provide mathematical details of the method and examples.     

How personal standards perfectionism and evaluative concerns perfectionism affect the error positivity and post-error behavior with varying stimulus visibility

Error detection is required in order to correct or avoid imperfect behavior. Although error detection is beneficial for some people, for others it might be disturbing. We investigated Gaudreau and Thompson’s (Personality and Individual Differences, 48, 532?537, 2010) model, which combines personal standards perfectionism (PSP) and evaluative concerns perfectionism (ECP). In our electrophysiological study, 43 participants performed a combination of a modified Simon task, an error awareness paradigm, and a masking task with a variation of stimulus onset asynchrony (SOA; 33, 67, and 100 ms). Interestingly, relative to low-ECP participants, high-ECP participants showed a better post-error accuracy (despite a worse classification accuracy) in the high-visibility SOA 100 condition than in the two low-visibility conditions (SOA 33 and SOA 67). Regarding the electrophysiological results, first, we found a positive correlation between ECP and the amplitude of the error positivity (Pe) under conditions of low stimulus visibility. Second, under the condition of high stimulus visibility, we observed a higher Pe amplitude for high-ECP–low-PSP participants than for high-ECP–high-PSP participants. These findings are discussed within the framework of the error-processing avoidance hypothesis of perfectionism (Stahl, Acharki, Kresimon, Völler, & Gibbons, International Journal of Psychophysiology, 97, 153–162, 2015).     

Middle ground on liability for costs?

Philosophical Studies - On the strict liability view, excusably ignorant agents must cover all the wrongful costs they have inadvertently brought onto others, although it is undisputed that they...