The Field Guide › Article
Argues that psychological safety and cognitive load are mutually reinforcing: psychological safety frees up capacity by removing the need to run constant mental risk calculations before asking for help, while high cognitive load pushes people toward silence because they lack the spare capacity to run those same calculations properly. Built on Sweller's Cognitive Load Theory (1988) and a three-part model adapted from Fraser et al.'s (2018) work on debriefing simulations: intrinsic cognition (what you already know), extraneous cognition (external information and distractions you must process), and germane cognition (the active problem-solving and learning that actually gets the job done, and through which learning moves from short-term to long-term memory). Good task design, on this account, protects space for germane cognition by minimising extraneous load — through better information access, fewer distractions, easier tools, or simply someone else's help. Speaking up sits squarely inside this: drawing on Amy Edmondson's account of the 'tacit calculus' run before any risky action ('if I do X here, will I be hurt, embarrassed or criticised?'), the piece argues that this constant, largely unconscious risk assessment is itself cognitively expensive, and that under high load there's less spare capacity to run it well, so silence becomes the default. Psychological safety cuts this cost twice over: it removes the need to run the calculus at all when asking a genuine question, and even where it still runs, simply knowing help is available reduces the anxiety that would otherwise consume germane capacity — connecting to Csikszentmihalyi's concept of Flow, which depends on that same freed-up capacity. Extends the idea to the team level via Skelton and Pais's Team Topologies: a team's cognitive load is roughly the sum of its members' load plus the scale of the system or domain it owns, which is why teams tend to grow as their systems grow — but since more people also raises cognitive load, the more durable fix is usually to decompose the system into smaller domains owned by smaller teams, not to keep adding headcount to one large team. Concludes that psychological safety and cognitive load management are jointly necessary for learning and performance, and that reducing colleagues' cognitive load is as much a shared team responsibility as fostering their psychological safety.