This is a response to Bas Dijkstra post on the utility of pair testing.
I fully support the proposition of utilizing pair testing, yet I advocate for the expansion of this collaborative methodology beyond the realms of programming and testing to encompass all forms of serious problem-solving. I also suggest thinking about shifting from pair to mob work.
Individual cognitive processes are inherently limited by biases and mental sets #1. Such mental sets, as conceptualized in cognitive psychology, often restrict our problem-solving capacity to familiar patterns and approaches, seriously reducing the innovative thinking. This idea is also well-described in the work of Charles Sanders Peirce, specifically his discussion of the ‘method of tenacity’ #2 which parallels the concept of mental sets, and the ‘method of authority’ #3 which aligns with the concept of authority bias.
I agree with Peirce’s assertion that the scientific method is the most efficient approach to problem-solving. However, I’d say that this effectiveness is achieved only through collaborative efforts. I contend that a single specialist can effectively overcome abovementioned biases and mental sets, this fight is easier won in teams. This viewpoint is supported by research in the field of operational research #4 and organizational science, which suggests that the diversity of knowledge inherent in teams is not only beneficial but perhaps essential for effective problem analysis and solution synthesis #5.
Our evolution as inherently social beings underscores the importance of teamwork for success.
Thus, I propose that pairing should be a standard practice for any serious work, and mob work should be employed wherever feasible to leverage the collective intelligence and diverse perspectives of the entire team.
References: