Emotional response failing to scale with the size of a problem — feeling similarly about 100 victims and 100,000.
Imagine you see a sad puppy and feel really bad. Now imagine someone tells you there are 10 sad puppies — you feel a little worse. But what about 10,000 sad puppies? You don't feel 10,000 times worse. Your heart just can't 'count' that high. So whether it's 10 puppies or 10,000, your sadness is about the same, even though the problem is way, way bigger.
Scope insensitivity describes the systematic failure of human emotional and evaluative responses to increase proportionally with the size or scale of a problem. When people are told that 2,000 or 200,000 birds are at risk, their willingness to help barely changes because they respond to a mental prototype — a single suffering creature — rather than to the actual numbers. This bias is particularly dangerous in domains involving large-scale humanitarian crises, environmental destruction, and existential risks, where the stakes differ by orders of magnitude but the emotional urgency remains flat. The effect is driven by the affect heuristic: feelings are generated by vivid mental images, and those images do not multiply with quantity, so enormous problems receive roughly the same emotional weight as small ones.
The same glitch looks different depending on the terrain. Finance, medicine, a relationship, a team — same mechanism, different costume.
Investors and donors tend to allocate similar amounts to address financial risks or charitable causes regardless of whether the affected population or dollar amount at stake differs by orders of magnitude, leading to systematic underfunding of large-scale problems and overfunding of smaller but more emotionally vivid ones.
Public health officials and patients often treat disease risks with similar urgency regardless of whether the condition affects hundreds or hundreds of thousands, leading to disproportionate funding for rare diseases with compelling individual narratives versus common conditions with far greater total burden.
Educators and administrators may invest equal effort and resources addressing learning gaps affecting a handful of students versus those affecting an entire district, treating both as 'an education problem' without proportionally scaling their response to the magnitude of the issue.
People may react with similar emotional intensity when a friend describes one personal setback versus a cascade of serious problems, offering the same consolation and support regardless of how much worse the situation actually is.
Product teams may treat a bug affecting 100 users with the same urgency as one affecting 1 million users, particularly when both are represented by a single angry support ticket or prototype user story rather than raw metrics.
Organizations often allocate similar budgets and attention to safety initiatives regardless of whether they protect 50 or 50,000 employees, especially when the initiative is framed around a single compelling accident narrative.
Media coverage and public outrage tend to be disproportionately similar for crises affecting thousands versus millions of people, with coverage volume driven more by the availability of compelling individual stories than by the actual scale of suffering.
Daniel Kahneman and Jack Knetsch formalized scope insensitivity in their 1992 paper 'Valuing Public Goods: The Purchase of Moral Satisfaction.' The concept was further developed by William Desvousges and colleagues in their 1992 bird oil-pond willingness-to-pay study, and extensively elaborated by Paul Slovic through his work on psychic numbing beginning in the late 1990s and 2000s.
Human cognition evolved in small-group environments where threats and social obligations involved dozens to hundreds of individuals, never millions. Emotional systems were calibrated for face-to-face encounters with identifiable kin and allies, where linear sensitivity to individual suffering was sufficient and computationally efficient. There was no ancestral selection pressure to develop proportional emotional responses to abstract quantities in the thousands or millions.
Machine learning models trained on human preference data can inherit scope insensitivity, treating harms of vastly different magnitudes as similarly important. Reward models and RLHF systems may learn to value helping one user and helping millions of users similarly if human raters do not distinguish scale in their feedback. Recommendation algorithms may also fail to weight the severity or breadth of misinformation proportionally, treating a post misleading 100 people with similar flags as one misleading millions.
50 cards are free to preview. Buyers unlock the rest of the deck plus the interactive training — Spot-the-Bias Quiz unlimited, Swipe Deck with spaced repetition, My Blindspots, Decision Pre-Flight, the Printable Deck + Cheat Sheets, and the Field Guide e-book. $29.50$59.
