Search

Sex differences in mathematical performance have frequently been examined over the last decades indicating an advantage for males especially when numerical problems cannot be solved by (classroom-)learnt strategies and/or estimation. Even in basic numerical tasks such as number line estimation, males were found to outperform females - with sex differences argued to emerge from different solution strategies applied by males and females. We evaluated the latter using two versions of the number line estimation task: a bounded and an unbounded task version. Assuming that women tend more strongly to apply known procedures, we expected them to be at a particular disadvantage in the unbounded number line estimation task, which is less prone to be solved by specific strategies such as proportion judgement but requires numerical estimation. Results confirmed more pronounced sex differences for unbounded number line estimation with males performing significantly more accurately in this task version. This further adds to recent evidence suggesting that estimation performance in the bounded task version may reflect solution strategies rather than numerical estimation. Additionally, it indicates that sex differences regarding the spatial representation of number magnitude may not be universal, but associated with spatial-numerical estimations in particular., (© 2016 The British Psychological Society.)

For bounded number line estimation, recent studies indicated influences of proportion-based strategies as documented by eye-tracking data. In the current study, we investigated solution strategies in bounded and unbounded number line estimation by directly comparing participants' estimation performance as well as their corresponding eye-fixation behaviour. For bounded number line estimation, increased numbers of fixations at and around reference points (i.e. start, middle and endpoint) confirmed the prominent use of proportion-based strategies. In contrast, in unbounded number line estimation, the number of fixations on the number line decreased continuously with increasing magnitude of the target number. Additionally, we observed that in bounded and unbounded number line estimation participants' first fixation on the number line was a valid predictor of the location of the target number. In sum, these data corroborate the idea that unbounded number line estimation is less influenced by proportion-based estimation strategies not directly related to numerical estimations.