Seeing Is Believing: Blind Putting Drills Confer No Advantage to the Novice Golfer

ABSTRACT

Purpose: Golf coaches may recommend “blind” putting drills in which golfers close their eyes to improve their feel. Research on specificity of learning suggests, however, that adding or removing a source of sensory information after practicing under differing circumstances can cause performance decrements. Specificity of learning is also dependent upon specific task requirements. The purpose of this study was to examine whether golf putting, requiring body positioning and aiming an implement, would benefit from blind training. Method: Novice golfers (n = 24) in Vision Training (VT) & No-Vision Training (NVT) groups completed 108 trials of a 10-ft putt. After a 24-hr delay, both groups completed sighted- and blind-putting tests. Results: Acquisition results revealed Group × Block interactions in RE (p = .025) and y-VE (p = .032). Post hoc procedures revealed significant differences between the groups on Block 2 (p = .017), with the NVT group producing longer mean RE. During testing, RE and x-VE results revealed Group × Test interactions (p = .027 & .041), such that performance of the VT group suffered when transferred to blind putting, while performance of the NVT group did not differ when transferred. Conclusion: NVT did not confer any advantage for subsequent performance with vision. Moreover, results were not consistent with the specificity of learning hypothesis and suggest that putting does not rely on complete sensory integration to support subsequent performance. Presumably, task requirements related to body positioning provided adequate sensory cues for successful performance.

KEYWORDS:

• Golf putting
• specificity of learning
• specificity of practice
• sensory integration

Notes

¹ The df differed for the two acquisition analyses because the data for the first half generated eight error-count categories while the data for the second half generated seven categories.

² The sighted test generated three error-count categories, yielding df = 2, while the Blind Test generated six categories, yielding df = 5.