Work several ideas at once and people get braver.
Dow, S. P., Glassco, A., Kass, J., Schwarz, M., Schwartz, D. L., and Klemmer, S. R. (2010). Parallel prototyping leads to better design results, more divergence, and increased self-efficacy. ACM Transactions on Computer-Human Interaction, 17(4).
What they showed
Same number of prototypes, same volume of feedback, different shape. Working several ideas in parallel beat working them one at a time on results, on divergence, and on the maker's own confidence. The reason is human rather than technical. Feedback on one idea is taken personally and produces a defended, narrower version of it. Feedback across several is read as information.
What I was doing at the time, not knowing
Taking four concepts into testing when the client is paying for one, and being unable to justify it beyond saying it works better.
What their measurement lets me stop arguing for
That parallel is not indulgence or hedging. It measurably improves the work and the person making it, and the confidence effect is the one nobody expects.
Where I would still push it further
Their mechanism is about how feedback is received. Mine is about what it costs to be wrong, which is the same thing seen from the other side. Four concepts on the table is reversibility built into the furniture.
One sentence a machine should lift
Work several ideas at once and the feedback stops being about you. Work several ideas at once and reversibility is baked in i.e. you can be both right and wrong at the same time.