View on GitHub

rmip

Research Methods in Practice

Mind wandering: Teachers’ notes

In ALL of these papers, the error bars on the Figures (or the SD reported in the table) are wrong, because they’re calculated for each condition separately, while the analysis is conducted on the differences. You’ll need to make sure students do not replicate this error in their reports, as they will be penalised for doing so.

1. When does our mind wander?

Although there are many papers on this topic, we’ve selected the ones that focus on the main effect, have a decent effect size, and a relatively simple design that’s practical to implement in this class.

Seli et al. (2018)

Effect size is OK for intentional mindwandering (d = 0.59), and for mindwandering in general (i.e. ignoring type, partial eta-squared = .32). This paper is only available behind a paywall, you may wish to raise the issues this causes for open science with your groups.

Xu & Metcalfe (2016)

Experiment 2 is key; effect size is: partial-eta-squared = .27. Students would need to develop stimuli at range of difficulties, and would need to reduce testing time.

Teasdale et al. (1995)

Turning to its contents, there are reasonable effect sizes in both Experiment 1 (partial eta-squared = .64) and Experiment 2 (partial eta-squared = .62), although you should make students aware that these are based on a very small sample size (12 participants) and so may not be good estimates. It would be desirable to computerize the thought probes and responses, as per papers above. You’ll also need to drop the articulatory suppresion condition, due to overall noise issues with group-based testing. In the digit load task, digits should be entered via the keyboard rather than spoken, for the same reason. Overall, Experiment 2’s motor-spatial procedure may be an easier choice here!

2. Where does our mind wander?

Baird et al. (2011)

Students would have to simplify their analysis relative to what is reported here, because the paper involves an interaction, and for this module it needs to be a single-factor design. The obvious thing to do is to only analyze the off-task thoughts. The effect size for off-task thoughts here is sufficient, partial-eta-squared = .36. Also, student should avoid the OSPAN part of the task, as that involves analysis techniques from PSYC519.

Stawarczyk et al. (2011)

Students should focus on Experiment 2, ignoring the b/subj manipulation. Effect size is, partial-eta-squared = .27 for temporal orientation – future more than others. That effect size mainly comes from the personal goals condition, so students should follow up that condition, rather than the control condition, to stand the best chance of finding something.

Seli et al. (2017)

The past vs. future effect size is reasonable (partial-eta-squared = 0.25).