Using the landmark task to test the underlying assumptions of unilateral contraction research

ABSTRACT

The unilateral contractions procedure (i.e., squeezing a ball with one hand) supposedly enhances a wide variety of cognitive functions, from episodic recall to choking under pressure. The practicality and affordability of this procedure makes it highly appealing. But does it work? We addressed this question by testing whether intermittent and sustained unilateral contractions shifted a well-supported hemispheric asymmetry: visuospatial attention. Based on prior research, contracting the left (or right) hand should lead baseline scores on the landmark task—a visuospatial attention measure—to deviate further left (or right). We meta-analysed the results of our six experiments and showed that the unilateral contractions procedure, particularly with intermittent contractions, does not reliably shift landmark task scores measured during (Experiments 4–6) or after (Experiments 1–3 & 6) performing unilateral contractions. Although we question if and how unilateral contractions activate the contralateral hemisphere, Experiment 6 provided some support for the utility of sustained contractions.

KEYWORDS:

• Hemispheric asymmetries
• spatial attention
• hemispheric activation
• unilateral contractions

Acknowledgements

Thanks to Sophie Witcombe, Derryth Lloyd, Jahnavi Ponnen, and Isabella Zannetino for collecting data and to Gina Grimshaw and Laura Kranz for sharing their unpublished results. Ella Moeck was on an Australian Government Research Training Scheme Scholarship when this data was collected.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

Deidentified data for all experiments is publicly available on the Open Science Framework (see http://osf.io/5t39k).

Notes

1 Experiment 6 contributes two effect sizes to the meta-analysis.

2 McCourt, Freeman, Tahmahkera-Stevens, and Chausee (2001) reported a leftward shift of d = 0.37 in landmark responses with left hand responding, suggesting small to medium was an appropriate estimated effect size.

3 To encourage natural behaviour, we did not systematically control squeezing speed (e.g., using a metronome).

4 The palm of the non-squeezing hand faced upwards, which was inconsistent with prior research (e.g., Schiff et al., 1998) where the palm faced downwards. Despite this methodological oversight, if the effects of the unilateral contractions procedure only occur when the palm faces downwards, then the generalizability of this procedure is limited.

5 We report 90% CIs around ${\eta }_p^2$ because ${\eta }_p^2$, unlike d, can only be positive. A 95% CI can include 0, even for significant effects (Läkens, 2014).

6 We ran all Bayes analyses using JASP (2018).