Two heads are (not) better than one
From: Dr. Ruth Lassalle
Two heads are better than one – this is not only a common phrase but a design principle in management and organisations. Cockpit, plant and ship crews work in accordance with the four-eyes principle, in production two individuals independently check the correct installation of critical components, and important documents are double checked before they are finalised.
All these examples incorporate the same assumption: The assignment of at least one other person doing the same task in parallel or serial contributes to minimizing error with the effect of enhanced reliability and quality. So, analogous to technical redundancy (e.g. if one generator fails another generator starts running) human redundancy is applied.
But there is one important difference between technical and human redundancy: Technical redundant components are indeed independent where as humans are not. An individual working in accordance to the four-eyes principal knows that there is another individual assigned with the same task and responsibility. This knowledge influences human behaviour (Sagan, 1993).
Studies revealed that teams of two redundantly working individuals do not achieve higher performance than individuals working alone (Skitka, Mosier, Burdick & Rosenblatt, 2000). More over research showed, that the individual monitoring performance is higher when people monitor a technical system on their own compared to people accomplishing the monitoring task in teams of two (Domeinski, Wagner, Schöbel & Manzey, 2007).
An explanation is provided by the phenomenon of social loafing: People exert less effort to achieve a goal when they work in a group and when their individual contributions can’t be related to them than when they work alone. These motivational decrements leading to performance decrements are not necessarily volitional and appear with regard to physical and mental tasks (Williams & Karau, 1991).
In practice, this does not mean that the four-eyes principal does not work. But one’s attention should consider the underlying conditions if improvement in quality needs to be achieved. Simply assigning a task to more than one person in terms of redundancy is hardly sufficient. That is to say, two heads are not unconditionally better than one.
Domeinski, J., Wagner, R., Schöbel, M. & Manzey, D. (2007). Human redundancy in automation monitoring: Effects of social loafing and social compensation. Proceedings of the HFES 51st Annual Meeting, 587-591. Santa Monica: Human Factors and Ergonomics Society
Sagan, S. D. (1993). The limits of safety: Organizations, accidents, and nuclear weapons. Princeton, N.J [u.a.]: Princeton University Press.
Skitka, L. J., Mosier, K. L., Burdick, M. & Rosenblatt, B. (2000). Automation bias and errors: Are crews better than individuals? International Journal of Aviation Psychology, 10(1), 85‐97.
Williams & Karau. (1991). Social loafing and social compensation ‐ the effects of expectations of coworker performance. Journal of Personality and Social Psychology, 61(4), 570‐581.