Background babble in open-plan offices: A natural masker of disruptive speech?

Abstract Numerous studies have shown that task-irrelevant background speech impairs performance of verbal short-term memory. This well-established effect is related to practice in open-plan offices, where employees are potentially disturbed by the speech of their colleagues. One option to reduce the disruptive effect is by masking the speech, for example, using random noise. Based on past research by Jones and Macken (1995), the ISO Standard 3382-3 (2012) assumes that multiple background speakers in open-plan offices may mask each other in a natural way, consequently reducing the disruptive effect of speech. The aim of this study was to check this assumption using a realistic acoustical simulation of an open-plan office situation. A combination of a nearby speaker and a varying number of background speakers was played to 26 participants while they performed on a verbal short-term memory task. Additionally, the intelligibility of the presented speaker sentences, levels of annoyance, and workload were checked. The results show a significant trend towards an improvement of short-term memory performance when the number of babble voices grows from one to six. However, performance levels are far from those reached under silent conditions. Moreover, annoyance and measures of subjective workload did not diminish due to babble masking.