The Effects of Construal Level on Predictive Heuristics: Disentangling Representativeness From Availability

Abstract

Intuitive predictions about the future can expect upcoming events to conform with a pattern - representativeness heuristic - or to simply repeat the most recent and accessible outcome - availability heuristic. The present work proposes that while representativeness predictions require a relatively abstract comparison process, availability-based predictions reflect a concrete and local use of accessible information. Grounded on the construal level theory, this research pits one heuristic against the other in predictions about random events (rolling dice; coin tosses). Three studies suggest that low construal levels increase predictions consistent with the most accessible and recent outcomes while high levels of construal facilitate representativeness-consistent predictions. The findings highlight how construal level may determine reliance on one heuristic or the other.

Disclosure statement

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

Data availiability statement

The data supporting the findings of this study is available at https://figshare.com/s/cf0f36137f1c1188a9a9. This research was not preregistered.

Author contributions

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the studies 1 to 3. The requirement for approval was waived by the ethics committee since the study posited no risks for participants nor experimental deceiving.

Notes

1 The sample size of this and following studies was determined by the number of participants made available by the university and research center.

2 No differences in the patters of predictions were found between high and low outcome sequences so this variable will not be further explored in the following analysis.

Additional information

Funding

This research was supported by grants from the Portuguese Foundation for Science and Technology (SFRH/BPD/122028/2016) to the first author.