Self-generated cognitive fluency: consequences on evaluative judgments

ABSTRACT

People can support abstract reasoning by using mental models with spatial simulations. Such models are employed when people represent elements in terms of ordered dimensions (e.g. who is oldest, Tom, Dick, or Harry). We test and find that the process of forming and using such mental models can influence the liking of its elements (e.g. Tom, Dick, or Harry). The presumed internal structure of such models (linear-transitive array of elements), generates variations in processing ease (fluency) when using the model in working memory (see the Symbolic Distance Effect, SDE). Specifically, processing of pairs where elements have larger distances along the order should be easier compared to pairs with smaller distances. Elements from easier pairs should be liked more than elements from difficult pairs (fluency being hedonically positive). Experiment 1 shows that unfamiliar ideographs are liked more when at wider distances and therefore easier to process. Experiment 2 replicates this effect with non-words. Experiment 3 rules out a non-spatial explanation of the effect while Experiments 4 offers a high-powered replication. Experiment 5 shows that the spatial effect spontaneously emerges after learning, even without a task that explicitly focuses on fluency. Experiment 6 employed a shorter array, but yielded no significant results.

KEYWORDS:

• Symbolic distance effect (SDE)
• magnitude processing
• linear orders
• spatial processing
• cognitive fluency

Acknowledgements

We would like to thank Petra Draginiċ, Michael Evans, Dominik Hausin and Emily Wolstenholme for help with data collection and helpful comments, as well as Dr. Justin Savage for programming, Dr. Henrik Singmann for statistical help, and Dr Kent Harber for editing help. We thank four anonymous reviewers and Dr. Klaus Fiedler for helpful comments and suggestions on earlier versions. Zixi Jin’s contribution to this research was supported by the China Scholarship Council.

Disclosure statement

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

Data availability statement

Data and R scripts for analyses can be accessed at https://osf.io/b4p38/?view_only=f1ee575d66d743c3978e565f4d201505.

Notes

1 We do not, in the context of this paper, discuss the question to what extent spatial characteristics are a necessary element of forming such orders, or indeed, to what extent the SDE needs spatial assumptions in order to be explained (but see Leth-Steensen & Marley, 2000 for such a discussion). Instead, we refer the reader to a series of studies in which we provided experimental evidence for spatial characteristics to be genuine and essential for the construction of linear orders such as used here (von Hecker et al., 2016; see also von Hecker & Klauer, 2021).

2 As a reviewer pointed out, another perspective on the obtained results would be to see our target effect as an indirect fluency effect: External experimental constraints make a decision easy/fluent versus difficult/dysfluent (i.e., asking participants to give liking ratings to stimuli of pair type A-F versus pair type C-D). In this perspective, it is not the fluency per se that is self-generated but only the internal model of the stimuli arrangement from which the level of fluency derives, depending on what decision is asked for. The interesting part is that the fluency of the decision is transferred or associated with the stimuli involved in the decision process. In antecipation of future research, one could assume that – if asked – participants also like the decision process A-F better than the decision process C-D. And liking of the decision process would correlate with liking of the stimulus. The interpretation as indirect fluency effect may additionally explain why the more direct blending effect is larger than the more indirect self-generated fluency effect investigated in the present research.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.