In George Bonanno’s research on loss and trauma, resilience is not bouncing back. It means staying roughly level, with brief disruptions. What most people call bouncing back, a drop followed by a gradual return, is closer to what he calls recovery. Both paths can include distress, and neither is a verdict on character.
The second distinction is about stress. A stress response that switches on to meet a challenge can protect you; one that does not switch off afterwards is one route to wear and tear on the body. The five findings below cover both distinctions, then sleep loss and support. The suggestions under each finding are this page’s operating advice, not interventions the cited studies tested.
| # | Question | What the research found |
|---|---|---|
| 1 | Is resilience rare? | Bonanno’s review describes it as common after loss and trauma |
| 2 | Is resilience the same as bouncing back? | Bouncing back after a sustained drop is closer to recovery; resilience is broadly stable functioning |
| 3 | Is stress itself the problem? | Stress systems can protect or damage the body; failing to switch off is one route to allostatic load |
| 4 | What does sleep loss affect? | Experiments found effects on attention and emotion, with differences across outcomes |
| 5 | Does any support help? | Evidence for buffering came from perceived support that matched the need |
These sources also inform The Psychology of Founder Resilience. None of them studied founders, and none can tell you how you will respond to a particular setback.
1. Resilience is common in the populations studied
In a 2004 review in American Psychologist, George Bonanno argued that psychology had underestimated resilience, because much of what it knew about loss and trauma came from people who sought treatment or showed great distress.
In bereavement studies that reported aggregate data, people with relatively low levels of depression or distress “consistently approached or exceeded 50% of the sample,” while chronic depression and distress tended to occur in 10% to 15% of bereaved people. Bonanno also reviewed a survey by Galea et al. (2002) after the September 11 attacks. As he reports it, over 40% of Manhattan residents did not report a single symptom of PTSD (post-traumatic stress disorder).
For this week. These percentages describe groups, not you, so do not read them as your odds. Read your own week instead: write down what still works, what has become harder, and where you need help. Let that list decide what you protect and what you change.
The limit. Startup setbacks are not bereavement, and they are not violent or life-threatening events. A common outcome is not an obligation, and persistent distress deserves care.
2. Resilience and recovery are different paths
Bonanno separates the two by how functioning changes over time.
Recovery is when “normal functioning temporarily gives way to threshold or subthreshold psychopathology” (symptoms at or just below the level of a diagnosis) “usually for a period of at least several months,” and then gradually returns to pre-event levels. Full recovery may be relatively quick or take as long as one or two years. This is the shape most people mean by bouncing back.
Resilience reflects “the ability to maintain a stable equilibrium.” Resilient people may have short disruptions, such as “several weeks of sporadic preoccupation or restless sleep,” but their functioning stays broadly stable.
Staying level is not the same as feeling nothing. In one prospective study Bonanno describes, 46% of participants had low levels of depression both before losing a spouse and through 18 months afterwards. Even in that group, most reported at least some yearning and emotional pangs.
For this week. Judge the path over several weeks, not by a single night. A few rough nights while you keep working do not tell you which path you are on, and still showing up does not prove you are fine. Once this week, note what has become harder and whether it is easing. A slow climb after a hard drop also fits a path Bonanno describes: recovery usually involves at least several months of reduced functioning.
The limit. These descriptions are not a self-diagnosis or a timetable you must meet, and they are no reason to wait before getting help.
3. A stress response needs to switch off
In a 1998 review in the New England Journal of Medicine, Bruce McEwen of Rockefeller University described the paradox that the body’s stress systems can “protect and restore” but also “damage the body.”
In the normal pattern he describes, a stressor starts the response, the response lasts for an appropriate interval, and then it turns off. The price of adapting to stress, McEwen writes, can be allostatic load: “the wear and tear that results from chronic overactivity or underactivity” of those systems.
He describes four situations associated with allostatic load: frequent stress; failing to adapt to the same stressor when it repeats; a prolonged response that is slow to shut down; and an inadequate response that leaves other systems overcompensating. He also names two factors that largely determine an individual’s response: how the person perceives the situation, and their general physical health.
For this week. Pick an end time for your workday. When it comes, write down the next task and close your email and work chat. Then notice whether you stay on alert anyway; that tells you more than the clock does. The boundary is this page’s advice, not McEwen’s, and not a treatment for allostatic load.
The limit. Not switching off is one of the four situations McEwen describes, not the only one. His review does not show that a hard pitch is harmless or that an evening routine prevents damage.
4. Sleep loss affects attention and emotion
A 2010 meta-analysis by Julian Lim and David Dinges pooled 70 articles and 147 cognitive tests of short-term total sleep deprivation, meaning less than 48 hours without any sleep. Effects ranged from small and nonsignificant for reasoning accuracy to large for lapses in simple attention.
A 2024 meta-analysis by Cara Palmer and colleagues pooled 154 studies, 1,338 effect sizes, and 5,717 participants aged 7 to 79. Across the forms of sleep loss it examined (total deprivation, shortened sleep, and fragmented sleep), the pooled results showed less positive mood and more anxiety symptoms. Results for negative mood and depressive symptoms were mixed.
For this week. If a hard-to-reverse decision can wait, such as signing a contract or deciding on a layoff, do not make it on a day after short sleep. After a night without sleep, do not take sound reasoning as proof that your attention is intact: in the Lim and Dinges data, reasoning accuracy held up far better than simple attention. Both are this page’s precautions, not rules either meta-analysis tested.
The limit. Lim and Dinges studied total sleep deprivation, not an ordinary short night. Neither review tells you how well you will judge a particular decision. One bad night is not proof of impaired judgment, and none of this is a sleep prescription.
5. Support needs to fit the pressure
In a 1985 review in Psychological Bulletin, Sheldon Cohen and Thomas Wills found evidence consistent with two different effects of social support.
Evidence for stress buffering came from studies that measured “the perceived availability of interpersonal resources that are responsive to the needs elicited by stressful events.” Evidence for a general benefit, regardless of stress, came from studies that measured “a person’s degree of integration in a large social network.” The authors also noted that correlational results do not, by themselves, allow a causal interpretation.
For this week. Name the help this pressure calls for, then the person who could give it: someone to check the numbers, hear out a hard decision, or take a task off your plate. Then ask them, rather than assuming they are available.
The limit. The review does not prescribe a number of contacts, and it does not show that one person can meet every need.
How does the RESILIENT framework use these findings?
RESILIENT is the authorial framework in The Psychology of Founder Resilience, a book whose sources include these studies. It organizes founder practice into nine capacities. The framework as a whole has not been externally validated and is not a clinical scale, so read its practices as proposals, not proven treatments.
- Reappraisal. Interpreting a setback in a way that does not amplify the threat.
- Emotional granularity. Telling similar feelings apart and responding to each precisely.
- Somatic regulation. Managing the body’s stress signals through sleep, movement, and breath.
- Identity flexibility. Maintaining more than one room of the self, so a setback cannot collapse the whole person.
- Learning orientation. Treating failure as information that feeds the next decision instead of self-doubt.
- Interpersonal support. Building relationships that supply support without demanding performance.
- Executive function. Guarding attention and decision-making clarity from exhaustion.
- Narrative integration. Holding a coherent story of challenge and growth.
- Titrated exposure. Meeting manageable stressors on purpose, in measured doses.
Somatic regulation (findings 3 and 4), executive function (finding 4), and interpersonal support (finding 5) sit closest to the findings above; reappraisal echoes McEwen’s point about perception. That mapping is this page’s reading, not evidence that the framework works.
Take a hypothetical founder who loses her largest customer on a Tuesday. Following this page’s advice, she writes down what has become harder, sets an end time for her workday, and calls the advisor who knows the numbers. On Thursday, after two short nights, she postpones the layoff decision because it can wait. These are operating choices, not outcomes the studies predict or guarantee.
For more, see the RESILIENT glossary entry, What is founder resilience?, and A holiday is not a refill protocol.
What these findings do not say
They do not say adversity makes you stronger. In their 2004 paper, Richard Tedeschi and Lawrence Calhoun, who study growth after trauma, write that the widespread assumption that trauma often leads to disorder should not be replaced with an expectation that growth is inevitable. They report that continuing personal distress and growth often coexist.
They are not founder studies, and this page is not clinical advice. If low mood, broken sleep, or trouble functioning lasts more than a couple of weeks, talk to a doctor or a licensed professional. If you think about harming yourself, contact a crisis line now: in the Netherlands, 113 Zelfmoordpreventie is free and anonymous (call 113 or chat via 113.nl), and in an emergency, call 112.
Try this week (15 minutes)
Use these prompts to choose one action. They are not a resilience score or a diagnostic checklist. Write one line for each:
- What has changed? What is harder in everyday life, and is it improving, stable, or getting worse?
- Does it switch off? When did the pressure last ease, and what will mark the end of today’s work?
- How did I sleep? Note the last three nights, and one consequential decision that can wait if you have slept badly.
- Who fits? Name one person and the specific help you will ask them for.
Put that one action in this week’s calendar. If your answers show the warning signs above lasting more than a couple of weeks, follow that guidance; this exercise does not replace it.
Cite this:What resilience research found: staying level is not bouncing back.Len P. van der Hof. https://lenvanderhof.com/en/blog/resilience-research-findings/ ·