Mindfulness and Wellbeing: The Role of Cognitive Biases

We will recruit from the online crowdsourcing platform Prolific, advertising a study that will investigate the effects of mood on decision-making. Half of the participants will be randomized to a mindfullness treatment intervention (14-day, 15 minutes a day of listening to guided meditations (designed to promote attention and presence)); the other half will be randomized to an active control intervention designed to control for the structual effects of the treatment and meditation's relaxing effects (14-day, 15 minutes a day of listening to relaxing music, with the same instructor as the treatment introducing the music). The participants will also take surveys online before and after the interventions to alllow us to measure the key outcome variables and various useful controls.

2020-08-10

2020-09-30

"Attention"-related primary outcomes: inattentive inference; sunk cost effect. "Presence"-related primary outcomes: fantasy (lottery) task; information preference scale.

Inattentive inference: using a hint X+Y for two random variables X,Y, to guess X, where the distributions of X,Y are known (Graeber, T. (2020). Inattentive Inference. (Online Baseline Tasks))
Sunk cost effect: 8-item scale; for each item subjects are presented with a scenario involving a type of sunk cost and have to indicate on a 6-point Likert scale their preference between two hypothetical alternatives (one of which indicates a susceptibility to the sunk cost effect) (Ronayne, D., Sgroi, S., & Tuckwell, A. (2020). Evaluating the Sunk Cost Effect.)
Fantasy (lottery) task: choice of whether to earn £X, or a smaller amount £Y but be allowed to open an envelope which will display the result of a lottery early (rather than waiting until the end of the survey) (Ganguly, A. & Tasoff, J. (2016). Fantasy and Dread)
Information preference scale: 13-item scale; each item subjects are presented with a scenario where they are asked whether they would like to receive some (potentially-negative) information of some kind and have to indicate how willing they are to do so (Ho, E.H., Hagmann, D., & Lowenstein, G. (2020). Measuring Information Preferences).

Trait mindfulness; stress; risk preference; time preference.

Trait mindfulness: FFMQ-15 (Gu, J., Strauss, C., Crane, C., Barnhofer, T., Karl, A., Cavanagh, K., & Kuyken, W. (2016). Examining the factor structure of the 39-item and 15-item versions of the Five-Facet Mindfulness Questionnaire)
Stress: Perceived Stress Scale (PSS-10) (Cohen, S. & Williamson, G.M. (1988): Perceived Stress in a Probabilitiy Sample of the US)
Risk preference: Staircase Risk and the self-reported willingness to take risk question in Falk, A., Becker, A., Dohmen, T. J., Huffman, D., & Sunde, U. (2016). The preference survey module: A validated instrument for measuring risk, time, and social preferences.
Time preference: Unincentivized Time Preference question in Ganguly, A., & Tasoff, J. (2017). Fantasy and dread: The demand for information and the consumption utility of the future. Management Science, 63(12), 4037-4060.
Willingness to give up for future question in Falk, A., Becker, A., Dohmen, T. J., Huffman, D., & Sunde, U. (2016). The preference survey module: A validated instrument for measuring risk, time, and social preferences.

Half of the participants are randomized to a mindfulness treatment intervention (14-day, 15 minutes a day of listening to guided meditations (designed to promote attention and presence)); the other half are randomized to an active control intervention designed to control for the structural effects of the treatment and meditation's relaxing effects (14-day, 15 minutes a day of listening to relaxing music, with the same instructor as the treatment introducing the music). The participants also take surveys online before and after the interventions to allow us to measure the primary and secondary outcomes.

Randomization by computer program

Individual

No

200-260

200-260

100-130

We will have a sample size of 200-260 (100-130 in each treatment). Our main outcome variable is the attention measure, which will be measured once before treatment and once after treatment. Using sampsi package in Stata, assuming ANCOVA as the analysis method, alpha=0.025 to take into account two outcome variables, autocorrelation=0.6, one-sided mean comparison, with a sample size of 200, to reach a power of at least 0.8, we need a minimum effect size (MDE) of 0.32; with a sample size of 260, we need MDE of 0.28; with a sample size of 260 and autocorrelation=0.7, we need MDE of 0.25. Referring to Boettcher et al. (2014), Bennike et al. (2017) and Bostock et al. (2018), which use similar designs of an online mindfulness treatment group compared to an active control group and before-after comparison, and also looking at the treatments in Graeber (2019) where we took the measure for the main outcome variable (a simple intervention like adding enforced deliberation or hint could achieve an effect size of 0.2-0.5), we find a MDE of 0.25-0.32 possible for our study. References: Bennike, I. H., Wieghorst, A., & Kirk, U. (2017). Online-based mindfulness training reduces behavioral markers of mind wandering Boettcher, J., Åström, V., Påhlsson, D., Schenström, O., Andersson, G., & Carlbring, P. (2014). Internet-based mindfulness treatment for anxiety disorders: a randomized controlled trial. Bostock, S., Crosswell, A. D., Prather, A. A., & Steptoe, A. (2019). Mindfulness on-the-go: Effects of a mindfulness meditation app on work stress and well-being. Graeber, T. (2020). Inattentive Inference.