How Bayes factors change scientific practice

Dienes, Zoltán

doi:10.1016/j.jmp.2015.10.003

Cited by 367 publications

(404 citation statements)

References 47 publications

(23 reference statements)

Supporting

Mentioning

396

Contrasting

Unclassified

Order By: Relevance

“…A B of 0.33 or below indicates substantial evidence for H0 over H1 and a B between 3 and 0.33 indicates data-insensitivity for distinguishing between either hypothesis. Bayesian analysis is more appropriate for our data than null hypothesis significance testing (NHST) as: (i) we require our analysis to enable us to make statements on whether the null hypothesis can be accepted (whether infants treat colors equivalently), whereas NHST provides no measure of credibility in favor of the null and nonsignificant results do not enable a definite conclusion (39); (ii) we needed a statistical approach that guarantees sensitivity with a minimum number of participants (we have a between-subjects design with 16 conditions) and Bayes combined with an optional stopping rule (test until B is sensitive in either direction) enables this, as B retains its exact meaning as the evidence in favor of H1 over H0 with additional data collection (40); and (iii) Bayes factors should not be adjusted for multiple testing (we have 16 tests), as false-alarm rates are dealt with through information in the data with no reference to how many other tests are conducted (41). In addition to Bayes factors, we report associated P values from NHST, although these are for reference only because they are affected by the optional stopping rule and multiple testing.…”

Section: Resultsmentioning

confidence: 99%

Biological origins of color categorization

Skelton

Catchpole

Abbott

et al. 2017

Proc. Natl. Acad. Sci. U.S.A.

106

View full text Add to dashboard Cite

The biological basis of the commonality in color lexicons across languages has been hotly debated for decades. Prior evidence that infants categorize color could provide support for the hypothesis that color categorization systems are not purely constructed by communication and culture. Here, we investigate the relationship between infants' categorization of color and the commonality across color lexicons, and the potential biological origin of infant color categories. We systematically mapped infants' categorical recognition memory for hue onto a stimulus array used previously to document the color lexicons of 110 nonindustrialized languages. Following familiarization to a given hue, infants' response to a novel hue indicated that their recognition memory parses the hue continuum into red, yellow, green, blue, and purple categories. Infants' categorical distinctions aligned with common distinctions in color lexicons and are organized around hues that are commonly central to lexical categories across languages. The boundaries between infants' categorical distinctions also aligned, relative to the adaptation point, with the cardinal axes that describe the early stages of color representation in retinogeniculate pathways, indicating that infant color categorization may be partly organized by biological mechanisms of color vision. The findings suggest that color categorization in language and thought is partially biologically constrained and have implications for broader debate on how biology, culture, and communication interact in human cognition.color lexicons | infant | categorization | color perception | vision

show abstract

Section: Resultsmentioning

confidence: 99%

Biological origins of color categorization

Skelton

Catchpole

Abbott

et al. 2017

Proc. Natl. Acad. Sci. U.S.A.

106

View full text Add to dashboard Cite

show abstract

“…whether p -values cross the level), and inferences are based on unobserved data (Cohen, 1994; Wagenmakers, 2007). Given those limitations, many have suggested alternative approaches such as model selection (Tibshirani, 1996; Yuan & Lin, 2006) or Bayesian statistics (Dienes, 2016; Wagenmakers, Morey, & Lee, 2016; Wagenmakers et al, 2017). Although each approach comes with pros and cons, here we extend on Bayesian hypothesis testing, a Bayesian alternative to NHST, and how it could be particularly useful in analysing conditioning data.…”

Section: P-values and Nhst For Threat Conditioning Datamentioning

confidence: 99%

“…For example, at the end of a threat acquisition phase, one could expect that the odds should be higher for the , indicating differences in CRs, than for the . However, in Bayesian hypothesis testing, those odds are often set to 1, indicating equal odds for both hypotheses (Dienes, 2016; Rouder, Speckman, Sun, Morey, & Iverson, 2009). In our example, those odds were also set to 1.…”

Section: Bayesian Hypothesis Testing For Threat Conditioning Datamentioning

confidence: 99%

Bayesian hypothesis testing for human threat conditioning research: an introduction and the condir R package

Krypotos

Klugkist

Engelhard

2017

European Journal of Psychotraumatology

View full text Add to dashboard Cite

Threat conditioning procedures have allowed the experimental investigation of the pathogenesis of Post-Traumatic Stress Disorder. The ﬁndings of these procedures have also provided stable foundations for the development of relevant intervention programs (e.g. exposure therapy). Statistical inference of threat conditioning procedures is commonly based on p-values and Null Hypothesis Signiﬁcance Testing (NHST). Nowadays, however, there is a growing concern about this statistical approach, as many scientists point to the various limitations of p-values and NHST. As an alternative, the use of Bayes factors and Bayesian hypothesis testing has been suggested. In this article, we apply this statistical approach to threat conditioning data. In order to enable the easy computation of Bayes factors for threat conditioning data we present a new R package named condir, which can be used either via the R console or via a Shiny application. This article provides both a non-technical introduction to Bayesian analysis for researchers using the threat conditioning paradigm, and the necessary tools for computing Bayes factors easily.

show abstract

“…In analyzing similar data, previous studies used null hypothesis significance testing. Although this approach may provide evidence for differences between groups (i.e., individuals with ASD differ from controls), it cannot distinguish between evidence that suggests no group differences and evidence that is inconclusive (i.e., does not provide enough evidence in support of either no group differences or group differences; Dienes, 2016). Therefore, we computed Bayes Factors (BFs) because they can distinguish between these two alternatives (Dienes, 2016;Wagenmakers, Morey, & Lee, 2016;Wiens & Nilsson, 2016).…”

Section: Introductionmentioning

confidence: 99%

Olfactory Functions in Adults With Autism Spectrum Disorders

et al. 2017

View full text Add to dashboard Cite

Autism spectrum disorders (ASD) are often characterized by atypical sensory behavior (hyper-or hyporeactivity) although evidence is scarce regarding olfactory abilities in ASD; 16 adults with highfunctioning ASD (mean age: 38.2, SD: 9.7) and 14 healthy control subjects (mean age: 42.0 years, SD: 12.5) were assessed in odor threshold, free and cued odor identification, and perceived pleasantness, intensity, and edibility of everyday odors. Although results showed no differences between groups, the Bayes Factors (close to 1) suggested that the evidence for no group differences on the threshold and identification tests was inconclusive. In contrast, there was some evidence for no group differences on perceived edibility (BF 01 ¼ 2.69) and perceived intensity (BF 01 ¼ 2.80). These results do not provide conclusive evidence for or against differences between ASD and healthy controls on olfactory abilities. However, they suggest that there are no apparent group differences in subjective ratings of odors.

show abstract

How Bayes factors change scientific practice

Cited by 367 publications

References 47 publications

Biological origins of color categorization

Biological origins of color categorization

Bayesian hypothesis testing for human threat conditioning research: an introduction and the condir R package

Olfactory Functions in Adults With Autism Spectrum Disorders

Contact Info

Product

Resources

About