Effects of monetary reward and punishment on information checking behaviour: An eye-tracking study. (July 2018)

• Record Type: Journal Article
• Authors: Li, Simon Y.W.; Cox, Anna L.; Or, Calvin; Blandford, Ann
• Is Part Of: Applied ergonomics. Volume 70(2018)
• Page Start: 110
• Page End: 117
• Publication Date: 2018-07
• Subjects: Error -- Reward -- Punishment -- Data-entry -- Eye-tracking; Human engineering -- Periodicals; 620.82
• DOI: 10.1016/j.apergo.2018.02.014
• Languages: English
• ISSNs: 0003-6870

Abstract

Abstract: The aim of the present study was to investigate the effect of error consequence, as reward or punishment, on individuals' checking behaviour following data entry. This study comprised two eye-tracking experiments that replicate and extend the investigation of Li et al. (2016) into the effect of monetary reward and punishment on data-entry performance. The first experiment adopted the same experimental setup as Li et al. (2016) but additionally used an eye tracker. The experiment validated Li et al. (2016) finding that, when compared to no error consequence, both reward and punishment led to improved data-entry performance in terms of reducing errors, and that no performance difference was found between reward and punishment. The second experiment extended the earlier study by associating error consequence to each individual trial by providing immediate performance feedback to participants. It was found that gradual increment (i.e. reward feedback) also led to significantly more accurate performance than no error consequence. It is unclear whether gradual increment is more effective than gradual decrement because of the small sample size tested. However, this study reasserts the effectiveness of reward on data-entry performance. Highlights: Direct comparison of the effect of reward and punishment on data-entry performance. Monetary reward and punishment improve accuracy compared to no incentives. Incremental reward is more effective than punishment. Accurate performance does not necessarily depend on checking frequency or duration.