Improving Inferences About Null Effects With Bayes Factors and Equivalence Tests

Lakens, Daniël; McLatchie, Neil; Isager, Peder Mortvedt; Scheel, Anne M.; Dienes, Zoltán

doi:10.1093/geronb/gby065

Cited by 251 publications

(205 citation statements)

References 39 publications

Supporting

Mentioning

190

Contrasting

Unclassified

Order By: Relevance

“…In case of significant interactions, follow-up linear mixed models were used to explore their nature. In case of non-significant effects, Bayes factors were calculated as implemented in Version 25 of IBM SPSS Statistics to determine the likelihood of the data under the null hypothesis of no effect and Bayes factors lower than 1/3 were considered as support for the null model 48 .…”

Section: Cs Valence Ratingsmentioning

confidence: 99%

Effects of verbal instructions and physical threat removal prior to extinction training on the return of conditioned fear

Wendt

Hufenbach

König

et al. 2020

Sci Rep

View full text Add to dashboard Cite

Instructions given prior to extinction training facilitate the extinction of conditioned skin conductance (SCRs) and fear-potentiated startle responses (FPSs) and serve as laboratory models for cognitive interventions implemented in exposure-based treatments of pathological anxiety. Here, we investigated how instructions given prior to extinction training, with or without the additional removal of the electrode used to deliver the unconditioned stimulus (US), affect the return of fear assessed 24 hours later. We replicated previous instruction effects on extinction and added that the additional removal of the US electrode slightly enhanced facilitating effects on the extinction of conditioned FPSs. In contrast, extinction instructions hardly affected the return of conditioned fear responses. These findings suggest that instruction effects observed during extinction training do not extent to tests of return of fear 24 hours later which serve as laboratory models of relapse and improvement stability of exposure-based treatments.Exposure-based treatments of pathological anxiety, which involve confronting patients with their feared stimulus or situation until the fear response declines 1 , are considered as highly efficient 2 . However, about one third of patients undergoing exposure-based treatments may not experience a clinically relevant decline in their pathological anxiety 3 and another third may experience a relapse, i.e. a return of their pathological anxiety, within 10 years 4 . These findings highlight the relevance of determining factors that moderate both short-term and long-term outcomes of exposure-based treatments. Pavlovian fear conditioning paradigms allow for the investigation of the development, treatment and maintenance of fear under laboratory conditions and typically comprise several experimental phases 5 : During fear acquisition training, a previously unrelated stimulus (conditioned stimulus, CS) is presented together with an aversive event (unconditioned stimulus, US), acquires signal value for the US and eventually elicits a conditioned fear response (CR). During extinction training, the CS is presented without the US and the CR gradually declines which is why extinction procedures serve as a laboratory analogue for exposure-based treatments 6 . Subsequent retention tests which entail the re-exposure to the CS after experimental manipulations and/or a period of time are used to investigate the maintenance of the extinction memory trace. Thus, retention tests are used to model long-term stability of fear reduction and relapse after exposure-based treatments 7 .Verbal instructions about CS-US associations influence conditioned responding when given prior to acquisition training 8 as well as when given prior to extinction training 9 . Instructions given prior to extinction training serve as a laboratory analogue for cognitive interventions which often accompany exposure treatments, e.g., in form of guided threat reappraisal 10 . In general, instructions informing participants about the subsequen...

show abstract

Section: Cs Valence Ratingsmentioning

confidence: 99%

Effects of verbal instructions and physical threat removal prior to extinction training on the return of conditioned fear

Wendt

Hufenbach

König

et al. 2020

Sci Rep

View full text Add to dashboard Cite

show abstract

“…This may not be the best approach, however, because it is likely that some effect sizes reported in previous research are inflated due to small sample sizes or publication bias (Ioannidis, 2008;Nuijten, van Assen, Veldkamp, & Wicherts, 2015;Szucs & Ioannidis, 2017). We therefore recommend future research to power experiments based on the smallest effect size of interest that is deemed theoretically or practically meaningful (see Lakens, McLatchie, Isager, Scheel, & Dienes, 2018;, and to utilise Bayesian analyses to make accurate statistical inferences (see Dienes, 2014;Lakens, McLatchie, et al, 2018). Moreover, we conducted a Bayesian meta-analysis, pooling the data from Experiment 1 and 2 to provide substantial support for the null hypothesis.…”

Section: Limitationsmentioning

confidence: 99%

“…Unlike NHST, Bayesian analyses do not rely on inferences based on statistical power and can show that a study unable to detect interesting effect sizes due to low statistical power provides evidence for the null hypothesis relative to the alternative hypothesis, or that a high-powered non-significant finding proffers no evidence for the null compared to the alternative hypothesis . We therefore recommend future research to power experiments based on the smallest effect size of interest that is deemed theoretically or practically meaningful (see Lakens, McLatchie, Isager, Scheel, & Dienes, 2018;, and to utilise Bayesian analyses to make accurate statistical inferences (see Dienes, 2014;Lakens, McLatchie, et al, 2018).…”

Section: Limitationsmentioning

confidence: 99%

Stereotype threat may not impact women's inhibitory control or mathematical performance: Providing support for the null hypothesis

Pennington

Litchfield

McLatchie

et al. 2018

Euro J Social Psych

Self Cite

View full text Add to dashboard Cite

Underpinned by the findings of Jamieson and Harkins (2007; Experiment 3), the current study pits the mere effort motivational account of stereotype threat against a working memory interference account. In Experiment 1, females were primed with a negative self‐ or group stereotype pertaining to their visuospatial ability and completed an anti‐saccade eye‐tracking task. In Experiment 2 they were primed with a negative or positive group stereotype and completed an anti‐saccade and mental arithmetic task. Findings indicate that stereotype threat did not significantly impair women's inhibitory control (Experiments 1 and 2) or mathematical performance (Experiment 2), with Bayesian analyses providing support for the null hypothesis. These findings are discussed in relation to potential moderating factors of stereotype threat, such as task difficulty and stereotype endorsement, as well as the possibility that effect sizes reported in the stereotype threat literature are inflated due to publication bias.

show abstract

“…However, future research should examine the strategies of guilty suspects under similar circumstances. Finally, although the Bayes factors observed (BF 01 = 3.90 for quantity of correct details and BF 01 = 7.82 for accuracy rates) suggest that the present findings likely reflect a genuine absence of the predicted effects of the interviewer's belief-led behaviour towards participants on their alibis (see Jeffreys, 1961;Lakens et al, 2018), we acknowledge the possibility of the study being underpowered. Thus, future research should attempt to replicate the present study using a larger sample size.…”

Section: Discussionmentioning

confidence: 62%

“I think you did it!”: Examining the effect of presuming guilt on the verbal output of innocent suspects during brief interviews

Portnoy¹,

Hope²,

Vrij³

et al. 2019

Journal Invest Psychology

View full text Add to dashboard Cite

Innocent suspects interviewed by a guilt‐presumptive versus innocence‐presumptive or neutral interviewer may tend more to display non‐verbal behaviours which neutral judges consider indicative of guilt. We examined the effects of interviewer's presumption of guilt on innocent mock suspects' alibis. Participants (N = 90) provided an alibi to convince an interviewer of their innocence of a theft after she implied that she believed that they were guilty or innocent or that she had no belief about their veracity. On the basis of existing conflicting findings for suspects' verbal behaviour during accusatory interviews, we predicted that alibis in the guilt‐belief condition would contain the highest or lowest number of correct details with overall higher or poorer accuracy rates, respectively. Although participants perceived the interviewer's presumptive approach, the number of correct details provided and accuracy rates of alibis did not differ significantly between conditions. We propose explanations to these findings and future research paths.

show abstract

Improving Inferences About Null Effects With Bayes Factors and Equivalence Tests

Cited by 251 publications

References 39 publications

Effects of verbal instructions and physical threat removal prior to extinction training on the return of conditioned fear

Effects of verbal instructions and physical threat removal prior to extinction training on the return of conditioned fear

Stereotype threat may not impact women's inhibitory control or mathematical performance: Providing support for the null hypothesis

“I think you did it!”: Examining the effect of presuming guilt on the verbal output of innocent suspects during brief interviews

Contact Info

Product

Resources

About