Skip to main content
Category: Effects
Type: A metaphor and decision trap: gradual change escapes notice because each step becomes the new baseline
Origin: The frog fable has 19th-century experimental roots; the human pattern is documented in change blindness, normalization of deviance, and shifting baseline research
Also known as: Boiling frog syndrome; 温水煮青蛙, 溫水煮青蛙, and ゆでガエル are the Chinese and Japanese labels. The frog story is a metaphor, not a demonstrated biological effect
Quick Answer — The Boiling Frog Effect is the failure to notice and resist a gradual deterioration because no single step looks like an emergency. The frog story is a metaphor, not a reliable experiment: intact, hydrated frogs try to escape. The useful lesson is structural. Set a fixed baseline and a tripwire before the drift begins, because attention adapts along with the risk.

​
What is the Boiling Frog Effect?

The Boiling Frog Effect is the failure to notice and resist a gradual deterioration because each increment is small enough to be accepted as the new normal.
If you put a frog in boiling water, it won’t jump out. It will die. If you put it in cold water, it will jump before it gets hot—they don’t sit still for you.
That is Douglas Melton, a biologist at Harvard, quoted in a 1995 Fast Company article. His point was blunt: the familiar story is false. A healthy frog with a way out does not sit in a pot while the temperature climbs. It leaves. The story survives because the human part is true enough to be useful. People and organizations often do stay in water that is getting warmer, not because they cannot feel the heat, but because no single reading looks like a reason to jump. The effect is narrower than general optimism and broader than a single bias. Normalcy Bias is the expectation that things will continue as usual, especially when a disaster has not been experienced before. Ostrich Effect is the avoidance of information that might be unpleasant. Status Quo Bias is a preference for the current state over a change. The Boiling Frog Effect is about a sequence: the reference point moves with each step, so the cumulative change is never compared with the starting state. The frog does not decide that hot water is good. It decides that this temperature is normal. The everyday version is a workplace that adds one more weekly report each quarter, a relationship that absorbs one more cutting remark each month, or a city that adds one more lane and one more hour of traffic each year. No single addition is worth a confrontation. The cost is visible only when someone reconstructs the whole sequence. That is why the metaphor is popular in management, public health, and climate writing: it names the moment when a slow trend has already become the background. The metaphor also carries a warning about power. A person can notice the water and still not jump because the pot is now their salary, their mortgage, or their public commitment. The effect is not only a failure of perception. It is a failure of perception plus a rising cost of exit. Escalation of Commitment explains why the person who has already endured three rounds of deterioration is often the least willing to leave before the fourth.

​
Boiling Frog Effect in 3 Depths

  • Beginner: If you cannot remember what this situation looked like a year ago, you do not have a baseline; you have a mood. Write down one number or one sentence that describes the starting state.
  • Practitioner: Pick a threshold before the next review cycle. Define the level at which you will act, who checks it, and what action follows. A threshold chosen in calm conditions is harder to move than a judgment made in the moment.
  • Advanced: Separate the question “has this gotten worse?” from “am I still willing to pay the exit cost?” The first is empirical; the second is strategic. A slow trend can be real and still be rational to accept, but the decision should be explicit rather than absorbed by default.

​
Origin

The fable has a stranger history than its management use suggests. In 1869, the German physiologist Friedrich Goltz reported that a frog with its brain removed would remain in slowly heated water, while an intact frog tried to escape as the water reached about 25°C. In 1872, A. Heinzmann reported that a normal frog would not try to escape if the water was heated slowly enough. In 1875, Carl Fratscher described experiments in which a frog stayed in water heated by about 0.002°C per second, from 20°C to 38°C over roughly two and a quarter hours, and eventually stopped breathing. When Fratscher raised the temperature faster in another experiment, the frog became agitated and tried to escape. In 1888, William Thompson Sedgwick reviewed the conflicting results and proposed the explanation that still fits the evidence: the difference was the heating rate. Heat slowly enough, and the frog’s escape response may never be triggered before it is incapacitated. Heat faster, and it jumps. The story was never a clean demonstration that a healthy animal chooses to be cooked. It was a boundary case about how a stimulus arrives. Modern sources are less ambiguous. In the 1995 Fast Company article, Douglas Melton said a frog in boiling water dies rather than jumping, and a frog in cold water jumps before it gets hot. George R. Zug, then curator of reptiles and amphibians at the Smithsonian’s National Museum of Natural History, said that a frog with a means of getting out would get out. In 2020, Estefany Caroline Guevara-Molina, Fernando Ribeiro Gomes, and Agustín Camacho published a study in the Journal of Thermal Biology showing that hydration level changes the behavioral thermal limits of bullfrogs. Some severely dehydrated individuals did not show a voluntary thermal maximum response at all. That is a useful nuance: the fable is not a description of a healthy frog, but it may describe a compromised one. The human concept has no single scientific coiner. “Boiling frog” entered management and policy writing as a metaphor, popularized in the 1990s by Daniel Quinn’s novel The Story of B and by Al Gore’s climate presentations, including the 2006 film An Inconvenient Truth. The mechanisms it names have separate research traditions. Change blindness was documented in the 1990s. Diane Vaughan’s The Challenger Launch Decision, published in 1996, gave organizational theory the idea of the normalization of deviance. Daniel Pauly’s 1995 paper on fisheries described the shifting baseline syndrome. The metaphor is a useful umbrella, not a diagnosis with a founding paper.

​
Key Points

Four findings explain why gradual change is so hard to catch: the comparison is missing, the baseline moves, past survival is misread as safety, and exit gets more expensive.
1

The change arrives between observations

In a 1998 field experiment, Daniel Simons and Daniel Levin had an experimenter ask pedestrians for directions, then swapped him for a different person while two people carried a door between them. Only 7 of 15 pedestrians reported noticing the switch. In a second experiment, when the experimenters dressed as construction workers, only 4 of 12 noticed. The pedestrians were not blind, and the change was not subtle. They had no reason to compare the person before the door with the person after it. Slow trends work the same way: there is no door, but there is also no moment that forces a comparison.
2

The baseline moves with the observer

Daniel Pauly’s 1995 paper described a pattern in fisheries science: each generation accepts the stock size and species composition at the start of its career as the baseline, then evaluates later changes against that. When the next generation arrives, the stock has fallen again, and the lower level becomes the new reference. The result is a gradual accommodation of disappearance. The same mechanism applies to budgets, service levels, noise, workload, and the quality of a relationship. After three rounds of “the new normal,” the original state is no longer the comparison anyone uses.
3

Past survival is read as evidence of safety

Diane Vaughan’s study of the Challenger disaster described the normalization of deviance: an unsafe practice becomes acceptable because it did not immediately produce a catastrophe. The O-ring seals on the shuttle’s solid rocket boosters had shown erosion and blow-by on earlier flights. Each safe return was treated as evidence that the anomaly was tolerable. Richard Feynman, in his appendix to the Rogers Commission report, put the logic plainly: “The argument that the same risk was flown before without failure is often accepted as an argument for the safety of accepting it again.” A track record is not a safety proof. It is a sample, and the sample may be too small.
4

Exit gets more expensive as the water warms

The longer a person stays in a deteriorating situation, the more they have invested in it: time, money, reputation, relationships, or a public promise. That is the terrain of Sunk Cost Fallacy and escalation of commitment. The effect is not only that people fail to see the heat. It is that seeing the heat does not settle the decision. A junior employee can leave a bad role cheaply; a senior employee with a mortgage, a visa, and fifteen years of tenure faces a different calculation. The metaphor is useful only if it includes the exit cost.

​
Applications

These four tactics work because they create a comparison that the situation itself will not provide. Use them where a slow drift has a cost that compounds.

Write the tripwire before the drift

Choose one measurable indicator and one threshold now: error rate, cost per unit, hours of overtime, complaints per month, or the number of exceptions you approve. Write down who checks it and what happens when it is crossed. The value of a tripwire is not that it is clever. It is that it was set before the person doing the checking had an incentive to move it.

Compare with a fixed baseline, not last month

Keep a snapshot of the original state: a photo, a sample, a recording, a metric, or a written description. Review against that snapshot each quarter, not only against the previous period. If the baseline has to change, record why and who approved it. A baseline that is silently renegotiated every period is not a baseline; it is a running average of the drift.

Bring in fresh eyes

Someone who did not watch the change unfold can see the cumulative difference. Show them the first state and the current state side by side, and ask what they would accept if they were starting today. This is the “new employee test,” and it works because the newcomer has no sunk cost in the old normal. The point is not that outsiders are smarter. They have a different reference point.

Log the anomalies you have accepted

Keep a decision journal for deviations, waivers, and workarounds. Record what was accepted, why, and what would have to happen for the acceptance to be reversed. Review the log for the sentence “it has always been like this” and for risks that were tolerated only because they had not yet failed. Then convert the top three into explicit choices: fix, monitor with a threshold, or accept with a named owner.

​
Case Study

On 28 January 1986, the Space Shuttle Challenger broke apart 73 seconds after launch, killing all seven crew members. It was the twenty-fifth shuttle flight. The twenty-four that came before it had returned safely, and that record was part of the problem. The immediate cause was a failure in the joint between the two lower segments of the right solid rocket motor. The seals designed to keep hot gas from escaping through the joint had failed. This was not a new part of the system, and the warning was not new either. The Rogers Commission found that O-ring erosion and blow-by had appeared on earlier flights, notably mission 51-C in January 1985 and mission 51-B in April 1985. The O-rings were not designed to erode. Erosion was a sign that the joint was not behaving as expected. But the shuttle had returned safely each time, so each erosion event was absorbed into the definition of normal. The night before the launch, engineers at Morton Thiokol recommended against launching below 53°F. The recommendation was reversed by Thiokol management, and the commission later concluded that the decision to launch was flawed. It also found that the NASA decision-makers were unaware of the recent history of O-ring problems and of the initial written recommendation. The information existed. It had not reached the people who needed it, and the people who had it had already learned to live with the anomaly. Richard Feynman’s appendix to the commission report describes the mechanism without the frog metaphor. Certification criteria, he wrote, “often develop a gradually decreasing strictness.” The fact that the same risk had been flown before without failure was “often accepted as an argument for the safety of accepting it again.” He compared the reasoning to Russian roulette: the first shot getting off safely is little comfort for the next. The commission’s report is careful about the many contributing causes, including schedule pressure and communication failures. The case is not a single moment of stupidity. It is a sequence in which each safe flight made the next warning easier to discount. The lesson generalizes with a boundary. The Challenger case is about an organization normalizing a technical anomaly, not about a literal slow-heating curve, and it does not prove that every accepted risk ends in disaster. What it shows is how a track record can be misread. The question that was missing was not “has this failed before?” It was “what would have to be true for this to be safe, and do we actually know it?” The first question invites the answer “no, it hasn’t failed.” The second one invites evidence.

​
Boundaries and Failure Modes

The Boiling Frog Effect is a metaphor with a real mechanism, not a law that every slow change is a trap. Four boundaries keep it useful. First, gradual change is often benign or desirable. A person who adapts to a louder city, a tighter budget, or a harder job is not necessarily a frog. Adaptation is a skill, and some trends are worth absorbing. The diagnostic is not the speed of the change. It is whether the current level would have been accepted at the start, whether the change is reversible, and whether anyone explicitly chose it. If the answer to all three is yes, the metaphor does not apply. Second, alarm fatigue is the mirror image of the problem. A team that declares every slow trend a crisis trains people to ignore the next warning. The fix is not a louder alarm. It is a small number of thresholds with clear owners and a known response. A tripwire that fires every week is not a tripwire; it is background noise. This is why the applications emphasize one measurable indicator rather than a general instruction to “stay vigilant.” Third, hindsight bias makes the drift look more obvious after the fact than it was at the time. Once a disaster has happened, the earlier warnings look like a clear road map. Before it, the same warnings competed with noisy data, ordinary successes, and other demands. The metaphor should not be used to retroactively blame people for not seeing a trend that was not measurable with the information they had. Ask what decision rule would have caught it, not who should have been smarter. Fourth, the fable’s biology is false, and the metaphor can be used manipulatively. Someone may call a gradual change a “boiling frog” to force urgency when the change is acceptable, reversible, or simply unfamiliar. The counter-move is to ask for the baseline, the threshold, and the decision that would have been different. If no one can name the level at which they would have acted, the metaphor is doing rhetorical work rather than analytical work.

​
Common Misconceptions

Three beliefs about the boiling frog are common, and each one turns a useful metaphor into a false claim or a bad decision.
It is not a reliable one. Goltz’s brain-removed frog stayed in the water, but an intact frog tried to escape. The 19th-century experiments that reported a passive frog depended on very slow heating, and in some cases on a compromised animal. Modern biologists point out that moving to a different temperature is a basic survival behavior for an ectotherm. The story is a metaphor about human attention, not a fact about healthy frogs.
They often notice it. What they lack is a shared baseline, a threshold, or the power to act. The change-blindness experiments show that people can miss a large change when nothing forces a comparison, but they do not show that people are indifferent to a trend they can see. A worker who says “it has been getting worse for years” has noticed. The failure may be that no one with authority treated the observation as a decision point.
Not every slow change is a trap. Some trends are noise, some are reversible, and some are the price of something worth having. The metaphor is most useful when the current level would have been rejected at the start and the cost of leaving rises with every step. Without those conditions, it can turn ordinary adaptation into a false emergency, and it can make people feel foolish for accepting a trade-off they would choose again.
The Boiling Frog Effect sits beside several neighbors that describe why a threat is ignored, but each one has a different mechanism.

Normalcy Bias

The expectation that life will continue as usual, especially when a disaster has not been experienced before. It explains the shock of the first crisis; the boiling frog explains the drift before it.

Gray Rhino Effect

A high-impact, obvious threat that is ignored despite being visible. The gray rhino is already in the room; the boiling frog is a threat that becomes visible only when you reconstruct the sequence.

Ostrich Effect

The avoidance of bad information. The boiling frog can involve avoidance, but it also includes people who see the trend and still do not act because the exit cost is too high.

Status Quo Bias

A preference for the current state over change. It supplies the inertia; the boiling frog supplies the moving baseline that makes the current state look normal.

Broken Windows Theory

The idea that uncorrected small deviations make larger ones easier to accept. It describes the social permission for drift; the boiling frog describes the perceptual and organizational failure to compare.

Sunk Cost Fallacy

The tendency to continue because of what has already been invested. It explains why the frog stays after it notices the heat: leaving would mean writing off the cost of staying.

​
One-Line Takeaway

A frog does not need a bigger alarm. It needs a fixed line in the water: decide now what level you would refuse, and write down who checks it.