Strategies in unbounded number line estimation? - Evidence from eye-tracking

Background: The unbounded number line estimation task (with only the start point and a unit but no endpoint given) has been claimed to provide a purer measure of numerical estimation than the standard bounded version of the task (with given start and endpoint). For the latter, recent studies found influences of solution strategies on performance, for instance, proportional judgments- as documented amongst others by eye-tracking data.
Aims and Method: In the current study, we evaluated these different solution strategies by directly comparing participants' estimation performance as well as their corresponding eye-fixation behavior in bounded and unbounded number line estimation in the range of 0 to 50.
Results: In line with previous studies, we found evidence suggesting bounded number line estimation to be solved by proportion-based strategies: participants' eye-fixation behavior indicated the prominent use of reference points with increased numbers of fixations at and around these (i.e., start-, mid- and endpoint). This was different for unbounded number line estimation. Here, the fixation pattern did not indicate the use of reference points but showed a continuous decline of the number of fixations along the number line with increasing number size. Additionally, we observed that in unbounded number line estimation participants' first fixation on the number line was a valid predictor of the location of the target number.
Conclusion: In sum, these data corroborate the idea that unbounded number line estimation is influenced less by propositional strategies not directly related to spatial-numerical estimations as compared to its bounded counterpart.