Comparing eye fixation and mouse cursor response modes in number line estimation

ABSTRACT

Recently, responding with eye fixations in the number line estimation (NLE) task was introduced. To determine if this new mode of responding replicates typical mouse cursor NLE results, 33 adults completed a 0 –1000 NLE task and responded with their eyes (eye condition) or the mouse (mouse condition), while their eyes were tracked. In the mouse condition, participants had lower error rates, higher response times, and looked at the three quartiles of the number line and back to the presented target number for a longer time compared to the eye condition. Despite these differences, both conditions elicited an “M-shaped pattern” in response times and error rates, thus replicating the typical NLE results in both response modes. Overall, a similar (strategic) process occurs for both response modes, suggesting that both are appropriate to use in the NLE task. Researchers should, however, proceed with caution, as there are mode-related constraints on response times and error rates.

KEYWORDS:

• Number line estimation
• magnitude
• mouse cursor
• adults
• eye tracking

Acknowledgements

The authors would like to thank all those who were able to participate in this study as well as numerous colleagues for the discussion regarding these findings. In addition, we would like to thank Tom Faulkenberry and another anonymous reviewer for their helpful and constructive comments on our manuscript.

Disclosure statement

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

Data availability

The data that support the findings of this study are openly available in OSF at https://doi.org/10.17605/OSF.IO/SCD6B.

Notes

1 Other authors refer to benchmarks by other names such as landmarks (Siegler & Thompson, 2014) or reference points (Link, Huber, et al., 2014). External (i.e., Visible, explicit) benchmarks on the number line are also possible and have been investigated (Peeters, Sekeris, et al., 2017). However, hereafter in the current paper, the term benchmarks refers to internal benchmarks, unless otherwise stated.

2 Before the analysis, we checked for potential practice effects, effects of presentation side of the target number and the order of condition on the PAE and RT_tot. No significant effects of any of these variables were found either on PAE or RT_tot, thus data were collapsed across these variables.

3 When measuring eye movement behaviour, the eye-tracker must record the location of the eye. Although the eye-tracker used in the current study has exceptional accuracy (∼0.25°-0.5°), there is nevertheless always some error attributed to the eye-tracker’s measurement error for every sample recorded.

4 The descriptive statistics are the mean of the medians across all participants collapsed across every target number for the target AOI. The frequency of 0 ms was very high as some participants did not look back to this AOI for every target number, thus explaining the very low means and standard deviations.

Additional information

Funding

This work was supported by the KU Leuven under Grant KA/16/009.