The systematic tendency to solve problems by adding new elements rather than removing existing ones, even when subtraction would produce a simpler or more effective solution.
Imagine you're building a Lego bridge and one tower is taller than the other. Most people reach for more blocks to make the short tower taller. But a kid might just pull one block off the tall tower—that's quicker and easier! Our brains are like that: whenever something is wrong, we think 'what can I add to fix it?' instead of 'what can I take away?' It's like always putting more toppings on a pizza that already has too much instead of removing the one topping that makes it taste bad.
Additive bias describes the deeply ingrained human default of searching for solutions that involve adding components—features, rules, steps, materials—rather than considering whether removing components would be more effective. When people attempt to improve objects, ideas, or situations, additive transformations come to mind more quickly and with less cognitive effort, causing subtractive alternatives to be systematically overlooked. This bias intensifies under cognitive load, time pressure, and when individuals have limited opportunities to reconsider their initial approach. The result is that people routinely miss simpler, more elegant solutions that involve taking away rather than piling on, contributing to bloated systems, overburdened schedules, and unnecessary complexity across virtually every domain of life.
The same glitch looks different depending on the terrain. Finance, medicine, a relationship, a team — same mechanism, different costume.
Investors and financial planners tend to address portfolio weaknesses by adding new asset classes, instruments, or hedging strategies rather than removing underperforming holdings or simplifying overly complex allocation structures, leading to portfolio bloat and higher management fees.
When treatment plans aren't working optimally, clinicians tend to add new medications, therapies, or diagnostic tests rather than considering deprescribing unnecessary drugs or eliminating redundant procedures—a pattern that contributes to polypharmacy and iatrogenic harm, particularly in elderly patients.
Gabrielle S. Adams, Benjamin A. Converse, Andrew H. Hales, and Leidy E. Klotz, University of Virginia, 2021. Formally published as 'People systematically overlook subtractive changes' in Nature, volume 592, April 2021.
In ancestral environments characterized by scarcity, accumulation was nearly always adaptive. Gathering more food, more tools, more shelter materials, and more allies increased survival probability. Subtraction—voluntarily giving things up—was rarely advantageous when resources were scarce and unpredictable. The brain therefore developed a rapid default toward accumulation and addition as the primary problem-solving strategy, as the cost of adding unnecessarily was low compared to the cost of subtracting something vital.
Machine learning systems exhibit additive bias when models are made more complex (adding layers, features, or parameters) to improve performance rather than pruning unnecessary components. In recommendation systems and content algorithms, the default is to add new signals and data sources rather than simplifying models by removing noisy or low-value inputs. LLMs tend to generate verbose, additive responses and struggle to recommend removal or simplification unless explicitly prompted, reflecting training data where human solutions disproportionately favor addition.
This page taught you the name. The deck turns the name into reflex. 1,100+ swipeable scenarios, 1,100+ defenses, 650+ detection prompts — spaced-repetition Swipe Deck, unlimited Spot-the-Bias Quiz, Defense Playbook, Pre-Flight, My Blindspots, Cheat Sheets, Field Guide e-book. $39.53$59.
