We conducted two lexical decision experiments and one replication study to examine the scope of transposed-phoneme effects when the transposition involves non-adjacent phonemes. The critical stimuli were non-words derived from a real word (the base-word) either by transposing two phonemes or by substituting the same phonemes with different phonemes. In Experiment 1, the transposed phonemes belonged either to the same syllable (e.g. /bis.tɔk/ for the French base-word /bis.kɔt/) or to a different syllable (e.g. /ʃo.lo.ka/ for the French base-word /ʃo.ko.la/) and were located either at the beginning of the speech signal (e.g. /sib.kɔt/ for /bis.kɔt/; /ko.ʃo.la/ for /ʃo.ko.la/) or at the end (e.g. /bis.tɔk/ for /bis.kɔt/; /ʃo.lo.ka/ for /ʃo.ko.la/). Experiment 2 compared within-syllable and between-syllable transpositions derived from the same set of bi-syllabic base-words (e.g., /sib.kɔt/, /bik.sɔt/, /bis.tɔk/ for the base-word /biskɔt/). In both experiments, we found clear transposed-phoneme effects with longer “no” decisions for transposed-phoneme non-words compared with the matched substituted-phoneme non-words. The effect was of similar magnitude when the transposed phonemes occurred in the same syllable and across different syllables. Also, for both the within- and between-syllable transpositions, the size of the transposed-phoneme effect did not vary as a function of the position of the transposition. Overall, our results suggest that phonemes can migrate across their respective positions not only within a syllable, but also across syllables. More importantly, they also suggest that position-independent phonemes exert a continuous influence during the entire processing of the auditory stimulus to the extent that there is sufficient time for this influence to manifest itself.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
One possible way to reconcile transposed-phoneme effects with models that code for the precise order of segments is to incorporate the notion of noise in the order encoding process, hence mimicking certain models of orthographic processing (e.g., Gomez et al., 2008). However, such a model has not yet been developed for spoken-word recognition, and thus the TISK model that assumes position-independent phonemes is actually the sole model that can account for transposed-phoneme effects.
We acknowledge that controlling for syllable structure has introduced another confound in terms of the distance between the transposed phonemes, with non-adjacent transpositions in the within-syllable conditions and adjacent transpositions in the between-syllable condition.
Baayen, R. (2008). Analyzing linguistic data: A practical introduction to statistics. Cambridge University Press.
Baayen, R. H., Davidson, D. J., and Bates, D. M. (2008). Mixed-effects modeling with crossed random effects for subjects and items. Journal of Memory and Language, 59, 390–412.
Baayen, R. H., and Milin, P. (2010). Analyzing reaction times. International Journal of Psychological Research, 3, 12–28.
Barr, D. J., Levy, R., Scheepers, C., & Tily, H. J. (2013). Random effects structure for confirmatory hypothesis testing: Keep it maximal. Journal of Memory and Language, 68, 255–278.
Bates, D. M. and Sarkar, D. (2007). lme4: Linear mixed-effects models using S4 classes, R package version 2.6, retrieved 20 November 2017 from http://lme4.r-forge.r-project.org/.
Dufour, S., & Grainger, J. (2021): When you hear /baksɛt/ do you think /baskɛt/? Evidence for transposed-phoneme effects with multi-syllabic words. Journal of Experimental Psychology, Learning, Memory and Cognition, in press.
Dufour, S., & Grainger, J. (2020). The influence of word frequency on the transposed- phoneme priming effect. Attention, Perception, and Pyschophysics, 2060–2069.
Dufour, S., & Grainger, J. (2019). Phoneme-order encoding during spoken-word recognition: A priming investigation. Cognitive Science, 43, 1–16.
Dufour, S., Peereman, R., Pallier, C., & Radeau, M. (2002). VoCoLex: A lexical database on phonological similarity between French words. L’Année Psychologique, 102, 725–746.
Foster, E. D., & Deardorff, A. (2017). Open Science Framework (OSF). Journal of the Medical Library Association, 105(2), 203–206.
Gaskell, M. G., & Marslen-Wilson, W. D. (1997). Integrating form and meaning: a distributed model of speech perception. Language and Cognitive Processes, 12, 613–656.
Goldinger, S. D. (1996). Auditory lexical decision. Language and Cognitive Processes,11, 559–567.
Gomez, P., Ratcliff, R., & Perea, M. (2008). The overlap model: A model of letter position coding. Psychological Review, 115, 577–601.
Grainger, J., & van Heuven, W. J. B. (2004). Modeling letter position coding in printed word perception. In P. Bonin (Ed.), The Mental Lexicon (pp. 1–23). New York: Nova Science.
Gregg, J., Inhoff, A.W. & Connine, C.M. (2019). Re (Re) re-considering the role of temporal order in spoken-word recognition. Quarterly Journal of Experimental Psychology, 72, 2574–2583.
Hannagan, T., Magnuson, J. S., & Grainger, J. (2013). Spoken word recognition without a TRACE. Frontiers in Psychology, 4, 563.
Jaeger, T. F. (2008). Categorical data analysis: away from ANOVAs (transformation or not) and towards logit mixed models. Journal of Memory and Language, 59, 434–446.
Harrison, W., Hepner, C.R.J., & Nozari, N. (2020). Is segmental interference position-dependent. In Proceedings of the 42nd Annual Meeting of the Cognitive Science Society (pp. 681–687).
Marslen-Wilson, W. D., & Welsh, A. (1978). Processing interaction and lexical access during word recognition in continuous speech. Cognitive Psychology, 10, 29–63.
Marslen-Wilson, W. D. (1990). Activation, competition, and frequency in lexical access. In G. T. M. Altmann (Ed.), Cognitive Models of Speech Processing: Psycholinguistic and Computational Perspectives (pp. 148–172). MIT Press.
Marslen-Wilson, W. D., & Warren, P. (1994). Levels of perceptual representation and process in lexical access: Words, phonemes, and features. Psychological Review, 101, 653–675.
McClelland, J. L., & Elman, J. L. (1986). The TRACE model of speech perception. Cognitive Psychology, 18, 1–86.
Norris, D. (1994). SHORTLIST: a connectionist model of continuous speech recognition. Cognition, 52, 189–234.
O'Rourke, T.B., Holcomb, P.J. (2002). Electrophysiological evidence for the efficiency of spoken word processing. Biological Psychology, 60, 121–150.
Perea, M., & Lupker, S. J. (2004). Can CANISO activate CASINO? Transposed-letter similarity effects with nonadjacent letter positions. Journal of Memory and Language, 51, 231–246.
R Development Core Team (2016). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. Available from http://www.R-project.org.
Toscano, J. C., Anderson, N. D., & McMurray, B. (2013). Reconsidering the role of temporal order in spoken-word recognition. Psychonomic Bulletin & Review, 20, 981–987.
This research was supported by ERC grant 742141.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix A Results of the post hoc analysis of Experiment 1
A correlation analysis was conducted to examine whether the magnitude of the /ʃoloka/ effect, that is, the difference between the transposed and substituted non-words in the different syllable final position condition, correlates with the average reaction time (RT) of participants in that condition. For each participant, the size of the transposed-phoneme effect was calculated by subtracting the mean RT (averaged across items) obtained for transposed-phoneme non-words with that obtained for the substituted-phoneme non-words. Pearson correlation coefficients and the associated p-value were calculated using statistica software. As illustrated in Fig. A1, the correlation was positive (r(86) = .32) and significant (p < .01), showing that the size of the transposed-phoneme effect increased as RTs increased.
Appendix B Replication of Experiment 1 with CV.CV.CV type base-words
A total of 185 participants participated in the experiment. Twenty participants were excluded from the analyses. Among them 17 participants had an error rate superior to 50% and the other three had excessively long reaction times (RTs) (greater than 2,000 ms on average and greater than 3 standard deviations (SDs) above the general mean). The mean RT and percentage of correct responses to non-words in each condition are presented in Fig. 4.
RTs to non-words (available at https://osf.io/gwk2j/) were analyzed using linear mixed-effects models following the same procedure as in Experiment 1. The RT analysis was performed on correct responses, thus removing 305 data points out of 8,580 (3.55%). Six RTs < 600 ms and 34 RTs > 5,000 ms were considered as outliers (less than 1%) and were also excluded from the analysis. Additionally, within each condition and for each participant RTs lying more than 3 SDs from the participant’s mean were excluded (29 data points; less than 1%). For the model to meet the assumptions of normally distributed residuals and homogeneity of variance, a log transformation was applied to the RTs (Baayen & Milin, 2010) prior to running the model. The model was run on 8,206 data points and included the variable Non-word Type (transposed, substituted), Position (initial, final), and their interaction as fixed effects. The non-word deviation point was also entered as a factor. The model also included participants and items as random intercepts.
The effect of Position was not significant (b = 0.0051, SE = .0077, t = 0.66). Crucially, the model revealed a significant effect of Non-word Type with RTs being slower for the transposed-phoneme non-words than for the substituted-phoneme non-words (b = 0.0194, SE = 0.0078, t = 2.50). Again, this effect was of similar magnitude when the transposition occurred at the initial or final of the words, as revealed by the lack of interaction between Non-word Type and Position (b = -0.0189, SE = .0155, t = -1.22). As found in the post hoc analysis of Experiment 1 (Appendix A), the correlation illustrated in Fig. 5 was positive (r(163) = .30) and significant (p < .001), showing again that the size of the transposed-phoneme effect increased as RTs increased.
About this article
Cite this article
Dufour, S., Mirault, J. & Grainger, J. Do you want /ʃoloka/ on a /bistɔk/? On the scope of transposed-phoneme effects with non-adjacent phonemes. Psychon Bull Rev 28, 1668–1678 (2021). https://doi.org/10.3758/s13423-021-01926-9
- Spoken word recognition
- Auditory lexical decision