A critical reexamination of doing arithmetic nonconsciously
A recent study claimed to have obtained evidence that participants can solve invisible multistep arithmetic equations (Sklar et al., 2012). The authors used a priming paradigm in which reaction times to targets congruent with the equation’s solution were responded to faster compared with incongruent ones. We critically reanalyzed the data set of Sklar et al. and show that the claims being made in the article are not fully supported by the alternative analyses that we applied. A Bayesian reanalysis of the data accounting for the random variability of the target stimuli in addition to the subjects shows that the evidence for priming effects is less strong than initially claimed. That is, although Bayes factors revealed evidence for the presence of a priming effect, it was generally weak. Second, the claim that unconscious arithmetic occurs for subtraction but not for addition is not supported when the critical interaction is tested. Third, the data do not show well-established features of numerosity priming as derived from V-shaped response time curves for prime-target distances. Fourth, we show that it is impossible to classify reaction times as resulting from congruent or incongruent prime-target relationships, which should be expected if their results imply that participants genuinely solve the equations on each trial. We conclude that the claims being made in the original article are not fully supported by the analyses that we apply. Together with a recent failure to replicate the original results and a critique of the analysis based on regression to the mean, we argue that the current evidence for unconscious arithmetic is inconclusive. We argue that strong claims require strong evidence and stress that cumulative research strategies are needed to provide such evidence.
KeywordsInterocular suppression Unconscious processing Reproducibility
P.M. was supported by the Research Fund Flanders (FWO Vlaanderen) through a doctoral fellowship. GH is supported by the German Research Foundation (grant HE 6244/1-2).
P.M. and G.H. designed research; P.M. analyzed data; P.M. and G.H. wrote the paper.
- Bachmann, T., Breitmeyer, B., & Ogmen, H. (2007). The experimental phenomena of consciousness: A brief dictionary. Oxford: Oxford University Press.Google Scholar
- Dehaene, S., Molko, N., Cohen, L., & Wilson, A. J. (2004). Arithmetic and the brain. Current Opinion in Neurobiology, 14(2), 218–224. doi: 10.1016/j.conb.2004.03.008
- Franz, V. H., & von Luxburg, U. (2014). Unconscious lie detection as an example of a widespread fallacy in the neurosciences. arXiv:1407.4240. Retrieved from http://arxiv.org/abs/1407.4240
- Goodman, S. N., Fanelli, D., & Ioannidis, J. P. A. (2016). What does research reproducibility mean? Science Translational Medicine, 8(341). doi: 10.1126/scitranslmed.aaf5027
- Hesselmann, G., & Knops, A. (2014). No conclusive evidence for numerical priming under interocular suppression. Psychological Science. doi: 10.1177/0956797614548876
- Hesselmann, G., & Moors, P. (2015). Definitely maybe: Can unconscious processes perform the same functions as conscious processes?. Frontiers in Psychology, 6(584). doi: 10.3389/fpsyg.2015.00584
- Leek, J. T., & Jager, L. R. (2016). Is most published research really false? bioRxiv, 50575. doi: 10.1101/050575
- Moors, P., Wagemans, J., & de-Wit, L. (2016). Faces in commonly experienced configurations enter awareness faster due to their curvature relative to fixation. PeerJ, 4. doi: 10.7717/peerj.1565
- Morey, R. D., Rouder, J. N., Love, J., & Marwick, B. (2015). BayesFactor: 0.9.12-2 CRAN [Zenodo]. Retrieved from doi: 10.5281/zenodo.31202
- Naccache, L. (2009). Priming. In T. Bayne, A. Cleeremans, & P. Wilken (Eds.), The Oxford companion to consciousness (pp. 533–536). Oxford: Oxford University Press.Google Scholar
- Phillips, N. (2016). yarrr: A companion to the e-book YaRrr!: The Pirate’s Guide to R. R package version 0.0.5. Retrieved from www.thepiratesguidetor.com
- RStudio Team. (2015). RStudio: Integrated development environment for R (Version 0.99.441). Boston: RStudio, Inc.Google Scholar
- Shanks, D. R. (2016). Regressive research: The pitfalls of post hoc data selection in the study of unconscious mental processes. Psychonomic Bulletin & Review, in press.Google Scholar
- Singmann, H., Bolker, B., Westfall, J., & Aust, F. (2016). afex: Analysis of Factorial Experiments. R package version 0.16-1. Retrieved from http://CRAN.R-project.org/package=afex
- Stein, T., & Sterzer, P. (2014). Unconscious processing under interocular suppression: Getting the right measure. Frontiers in Psychology, 5, 387. doi: 10.3389/fpsyg.2014.00387
- Wolsiefer, K., Westfall, J., & Judd, C. M. (2016). Modeling stimulus variation in three common implicit attitude tasks. Behavior Research Methods Google Scholar