A tutorial on a practical Bayesian alternative to null-hypothesis significance testing

Masson, Michael E. J.

doi:10.3758/s13428-010-0049-5

Cited by 737 publications

(708 citation statements)

References 34 publications

Supporting

Mentioning

695

Contrasting

Unclassified

Order By: Relevance

“…At SOA = -150 ms, the effect of semantic relatedness was significant, F1(1, The central finding concerns the absence of an interaction between semantic and phonological relatedness at the SOA under which both types of effects were present (SOA = 0 ms). Bayesian analysis with the method suggested by Masson (2011) resulted in a Bayes factor of 5.20, with pBIC(H0|D) = .839 and pBIC(H1|D) = .161, which according to the classification suggested by Raftery (1999) constitutes "positive" evidence for the null hypothesis (i.e., an additive pattern between semantic and phonological effects).…”

Section: Resultsmentioning

confidence: 99%

Additivity of semantic and phonological effects: Evidence from speech production in Mandarin

Zhu

Zhang

Damian

2016

Quarterly Journal of Experimental Psychology

View full text Add to dashboard Cite

A number of previous studies using picture-word interference (PWI) tasks conducted with speakers of Western languages have demonstrated non-additive effects of semantic and form overlap between pictures and words, which may indicate underlying non-discrete processing stages in lexical retrieval. The present study used Mandarin speakers and presented Chinese characters as distractors. In two experiments, we crossed semantic relatedness with "pure" phonological (i.e., orthographically unrelated) relatedness, and found statistically additive effects. In a third experiment, semantic relatedness was crossed with orthographic overlap (phonological overlap was avoided), and once again we found an additive pattern. The results are discussed with regard to possible cross-linguistic differences between Western and non-Western languages in terms of phonological encoding, as well as concerning the locus of relatedness effects in PWI tasks.Key words: speech production, semantic interference, phonological facilitation, lexical selection, MandarinAdditivity of semantics and phonology 3 Speaking involves the retrieval of semantic, syntactic, and phonological/phonetic information of the corresponding words. How these forms of information relate to each other remains a controversial issue in speech production theories (Bi, Xu, & Caramazza, 2009;Dell, 1986;Starreveld & La Heij, 1995, 1996a, 1996b. Serial discrete models (e.g., Levelt, Roelofs, & Meyer, 1999) argue that for a given target word, only a single selected lexical-semantic/syntactic node ("lemma") spreads its activation to the phonological level, and semantic processing must be completed before phonological processing can begin. Non-serial models dispute some or all of these assumptions, and assume that phonological encoding can proceed based on partial and continuous input from the semantic level. Cascaded models (e.g., Humphreys, Riddoch, & Quinlan, 1988;Morsella & Miozzo, 2002) propose that multiple lexical-semantic candidates which are co-activated during retrieval of the target word transmit activation to the phonological level. Interactive models (e.g., Dell, 1986) additionally assume that transmission of activation between semantic and phonological encoding is bidirectional. In such models, phonological processing begins on the basis of early partial information provided by semantic processes, and partial phonological processing aids in retrieval at the earlier lexical-semantic level.One way of empirically distinguishing discrete from non-discrete (cascaded or interactive) models is to apply "additive-factors logic" (Sternberg, 1969). Two experimental variables are manipulated and their effect on performance is measured.If the two variables show additive effects, then it can be concluded that they affect different and separate processing stages. By contrast, if the two variables show nonAdditivity of semantics and phonology 4 additive effects, then either they act on a single processing stage, or they affect two processing stages but these two stages are c...

show abstract

Section: Resultsmentioning

confidence: 99%

Additivity of semantic and phonological effects: Evidence from speech production in Mandarin

Zhu

Zhang

Damian

2016

Quarterly Journal of Experimental Psychology

View full text Add to dashboard Cite

show abstract

“…We also report Bayes factors for each of the t tests, because Bayesian analysis provides evidence for how strongly the data favor the alternative versus the null hypothesis (Masson, 2011;Wagenmakers, 2007).…”

Section: Ex-gaussian Analysis Of Experiments 3 and 4 Combinedmentioning

confidence: 99%

The semantic Stroop effect: An ex-Gaussian analysis

2016

View full text Add to dashboard Cite

Previous analyses of the standard Stroop effect (which typically uses color words that form part of the response set) have documented effects on mean reaction times in hundreds of experiments in the literature. Less well known is the fact that ex-Gaussian analyses reveal that such effects are seen in (a) the mean of the normal distribution (mu), as well as in (b) the standard deviation of the normal distribution (sigma) and (c) the tail (tau). No ex-Gaussian analysis exists in the literature with respect to the semantically based Stroop effect (which contrasts incongruent color-associated words with, e.g., neutral controls). In the present experiments, we investigated whether the semantically based Stroop effect is also seen in the three ex-Gaussian parameters. Replicating previous reports, color naming was slower when the color was carried by an irrelevant (but incongruent) colorassociated word (e.g., sky, tomato) than when the control items consisted of neutral words (e.g., keg, palace) in each of four experiments. An ex-Gaussian analysis revealed that this semantically based Stroop effect was restricted to the arithmetic mean and mu; no semantic Stroop effect was observed in tau. These data are consistent with the views (1) that there is a clear difference in the source of the semantic Stroop effect, as compared to the standard Stroop effect (evidenced by the presence vs. absence of an effect on tau), and (2) that interference associated with response competition on incongruent trials in tau is absent in the semantic Stroop effect.

show abstract

“…One popular method is to quantify and correct for model complexity solely through the number of free parameters. This procedure provides the basis for statistical indices such as the Akaike information criterion (AIC; Akaike, 1973;Burnham & Anderson, 2002) and the Bayesian information criterion (BIC; Masson, 2011;Myung, 2000;Raftery, 1995;Schwarz, 1978;Wagenmakers, 2007; for a discussion on the differences between the AIC and BIC, see, e.g., Karabatsos, 2006;Vrieze, 2012).…”

Section: Alternative Methods For Comparing Toolbox Modelsmentioning

confidence: 99%

Testing adaptive toolbox models: A Bayesian hierarchical approach.

Scheibehenne

Rieskamp

Wagenmakers

2013

Psychological Review

103

View full text Add to dashboard Cite

Many theories of human cognition postulate that people are equipped with a repertoire of strategies to solve the tasks they face. This theoretical framework of a cognitive toolbox provides a plausible account of intra-and interindividual differences in human behavior. Unfortunately, it is often unclear how to rigorously test the toolbox framework. How can a toolbox model be quantitatively specified? How can the number of toolbox strategies be limited to prevent uncontrolled strategy sprawl? How can a toolbox model be formally tested against alternative theories? The authors show how these challenges can be met by using Bayesian inference techniques. By means of parameter recovery simulations and the analysis of empirical data across a variety of domains (i.e., judgment and decision making, children's cognitive development, function learning, and perceptual categorization), the authors illustrate how Bayesian inference techniques allow toolbox models to be quantitatively specified, strategy sprawl to be contained, and toolbox models to be rigorously tested against competing theories. The authors demonstrate that their approach applies at the individual level but can also be generalized to the group level with hierarchical Bayesian procedures. The suggested Bayesian inference techniques represent a theoretical and methodological advancement for toolbox theories of cognition and behavior.

show abstract

A tutorial on a practical Bayesian alternative to null-hypothesis significance testing

Cited by 737 publications

References 34 publications

Additivity of semantic and phonological effects: Evidence from speech production in Mandarin

Additivity of semantic and phonological effects: Evidence from speech production in Mandarin

The semantic Stroop effect: An ex-Gaussian analysis

Testing adaptive toolbox models: A Bayesian hierarchical approach.

Contact Info

Product

Resources

About