Disfluency attenuates the reception of pseudoprofound and postmodernist bullshit

Abstract

Four studies explore the role of perceptual fluency in attenuating bullshit receptivity, or the tendency for individuals to rate otherwise meaningless statements as “profound”. Across four studies, we presented participants with a sample of pseudoprofound bullshit statements in either a fluent or disfluent font and found that overall, disfluency attenuated bullshit receptivity while also finding little evidence that this effect was moderated by cognitive thinking style. In all studies, we measured participants’ cognitive reflection, need for cognition, faith in intuition, and superstitious beliefs. Superstition strongly predicted bullshit receptivity regardless of fluency. Inconclusive results were found for the remaining scales. Potential links for the role of perceptual disfluency in promoting analytic thinking are discussed.

Keywords:

• Bullshit receptivity
• postmodernism
• processing fluency
• pseudoprofound
• reflexive thinking

Author contributions

All authors conceptualized the study. R. E. Tracy collected and analyzed the data. All authors wrote the manuscript and approved the final version for submission.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Open practices

All data and analysis scripts can be found at https://github.com/ryanetracy/bullshit_fluency.git

Notes

1 Because prior research on bullshit receptivity has been correlational in nature, we conducted simulation analyses to determine the smallest detectable effect size rather than use prior literature as a basis.

2 For all LMM analyses, we report an estimated d value for any comparisons involving the two-level effect of fluency only, in addition to η_p², as an estimate of effect size. For any estimates involving continuous variables or moderations, we only report η_p², owing to the difficulty in interpreting an estimate of d for continuous variables and interactions between continuous and categorical predictors.

3 We included fewer statements than in Study 1, due to these stimuli’s length and increased linguistic complexity to prevent possible fatigue effects in our online sample.

4 We ran additional analyses on participants’ untransformed millisecond RTs using a GLMM that assumed a gamma distribution in accordance with recommendations from Lo and Andrews (2015). While this model produced similar results, we report log-transformed analyses here. The GLMM analyses can be found in the supplementary materials.

5 We also conducted analyses exploring bullshit sensitivity, subtracting participants’ ratings for pseudoprofound bullshit from their ratings for the inspirational quotes (Pennycook et al., 2015). These revealed no significant differences in bullshit sensitivity across the fluency conditions (b = .04, p = .67) when controlling for the individual differences, and thus are not reported in the main text.

6 While not reported here, these effects remained significant when we included the individual difference metrics into the model as predictors and moderators of the fluency and statement type effects.

7 As the three-way interaction subsumed the two-way interactions, we report only the simple slopes analyses for this in the main text. The remaining simple slopes analyses can be found in the Supplementary Material.