How Hunger And Fatigue Shrink The Circle Of Who You Care About

In 2011, Shai Danziger and colleagues published "Extraneous factors in judicial decisions" in PNAS. They analyzed 1,112 parole board hearings before eight experienced Israeli judges over ten months. The dependent variable was simple: did the prisoner get paroled?

The findings: probability of a favorable ruling started at approximately 0.65 right after a break (food break or rest break) and declined steadily, approaching zero just before the next break. After the break, it reset to approximately 0.65 again. The pattern held across three daily sessions.

The researchers controlled for the type of crime, the prisoner's ethnicity, whether the prisoner had legal representation, and the order in which cases were heard. None of these factors explained the pattern as cleanly as session timing.

The default decision when fatigued was denial — requiring no written justification, no risk assessment, no imaginative reconstruction of the prisoner's circumstances. Grant requires all three.

Important caveat: subsequent researchers have questioned whether this specific effect holds across all replications, and some have suggested alternative explanations, including that case difficulty increases over a session (judges front-load easier cases). The debate in the literature is real. But the broader phenomenon that judgment degrades under depletion is robustly supported across hundreds of studies. The parole study is vivid and illustrative; the underlying science is solid even if this single study's mechanism is contested.

---

Roy Baumeister's self-depletion model, developed over two decades at Florida State University, proposes that self-regulation — the ability to override automatic impulses and engage in deliberate, effortful thought — operates like a muscle that fatigues with use.

Early research (Baumeister et al., 1998) showed that participants who resisted eating chocolate cookies before a difficult puzzle-solving task gave up on the puzzle much sooner than those who hadn't had to exert self-control. The glucose hypothesis emerged from follow-up work: acts of self-control reduce blood glucose levels, and restoring glucose (via consuming a sugary drink) partially restores self-regulatory capacity.

This model has faced replication challenges. Some researchers found that simply believing you'd consumed sugar was enough to restore performance, suggesting a motivational rather than purely metabolic mechanism. Others failed to replicate the basic depletion effect.

Where the science has settled, more or less, is this: the original strong glucose model was probably too simple, but resource depletion is real. Whether the mechanism is strictly metabolic or involves motivational and attentional components, the functional result is the same — extended effortful cognition impairs subsequent effortful cognition. The brain is a biological organ. It is not infinitely sustainable.

For our purposes, the mechanism is less important than the demonstrated effect: when people have been thinking hard for a long time, the quality of their moral reasoning, their accuracy of empathy, and their fairness in judgment all decline.

---

Peter Singer's work on the expanding circle of moral concern traces the historical arc of human ethical development: we began caring most about family, then tribe, then nation, then species, and some argue the circle is still expanding to include animals and ecosystems. The direction of moral progress is outward.

Depletion reverses that direction.

This is the crucial point. When you are exhausted, you do not become merely indifferent — you regress. The cognitive effort required to represent a stranger as a full, complex human being with their own interiority, history, and legitimate claims on your attention is genuinely substantial. That effort is among the first things to go when resources are thin.

What remains is a more primitive representational system: fast, categorical, pattern-matching. This person is like people who have given me trouble before. This case looks like cases that ended badly. I don't have the bandwidth to investigate further — default.

That's not a character flaw. That's a brain protecting its remaining resources. But it produces outcomes that a rested version of the same person would recognize as unjust.

The Israeli judges, if you showed them a transcript of their session-end decisions, would likely be troubled by their own behavior. Most of them would not recognize themselves in those choices. That dissociation is itself important: people don't know when they're depleted in ways that compromise their ethics. They feel, in the moment, like they're still thinking clearly.

---

The implications don't stay at the individual level.

Medicine. Hospital systems in the United States routinely schedule residents for 24-to-28-hour shifts. The justification is continuity of care and training rigor. The evidence is that medical errors — including errors with fatal outcomes — increase dramatically with sleep deprivation. A 2004 study in JAMA found that interns made 36% more serious errors during traditional extended shifts than during reduced-hour schedules. A 2009 Harvard study found that fatigued residents made 460% more diagnostic errors in an ICU.

These are not abstract statistics. Each data point is a person who received worse care because the human making decisions about them was running on empty.

The criminal legal system. Beyond the parole study, research on plea bargaining shows that defendants are more likely to accept bad deals when their cases are heard late in the day. Prosecutorial discretion — who gets charged, at what level, whether to negotiate — is exercised by humans who are also subject to depletion. Public defenders, notoriously overworked, make decisions under conditions of chronic fatigue. Every stage of the system is staffed by depleted humans whose default, when tired, is the option requiring least effort — which in an adversarial legal system almost always disadvantages the defendant.

Legislative bodies. Late-night votes are a standard legislative tactic used by majority parties to pass controversial legislation when the opposition is less vigilant, when media attention is lower, and when cognitive resistance is reduced. The people voting on laws that govern millions are doing so in conditions of demonstrated impairment.

Child welfare. A social worker carrying 40-plus active cases, as many do in underfunded systems, cannot give each case the deliberate attention it requires. The caseworker who doesn't have time to fully investigate a home environment, who defaults to removal when uncertain, or defaults to leaving a child in place when uncertain — that person is not failing through negligence in any culpable sense. They're failing because the system is structured to produce depletion-based decisions, then blaming individuals for the outcomes.

---

Here is where this becomes personal — and where it gets uncomfortable.

If you are in any position of authority, responsibility, or decision-making power over other people's lives, your physical state is not a private matter. It is an ethical one.

The manager who prides herself on her work ethic — in early, out late, no lunch break, always available — is not the hero of the story if her cognitive state means her employees receive depleted, pattern-matching versions of her judgment. The person who gets fired, or passed over, or misunderstood in a critical performance review because the manager was running on four hours of sleep — that person deserves better.

This is not about blaming yourself. It's about restructuring.

Practical reframings:

Sleep is not laziness. Matthew Walker's research (Why We Sleep, 2017) demonstrates that sleep deprivation at the level of seven or fewer hours per night — which the majority of working adults experience chronically — produces measurable deficits in emotional regulation, empathy accuracy, and moral reasoning. A sleep-deprived brain is an empathy-impaired brain. Full stop.

Breaks before decisions. Institutions that take this seriously should build in mandatory breaks before high-stakes decisions. Judges reviewing cases in blocks. Doctors limited in consecutive decision-making hours. Hiring panels required to take a break before final candidate deliberation.

Depletion disclosure. Some medical and legal contexts require disclosure of conflicts of interest. The science suggests that severe fatigue, in contexts of high-stakes decisions, might warrant something similar — not legally mandated disclosure, but a cultural norm: "I'm running on low, I want to revisit this when I'm sharper" as an acceptable professional statement rather than a sign of weakness.

Recognizing the pattern in yourself. The cruel irony is that you cannot reliably detect your own depletion in the moment. The subjective sense of "I'm still thinking clearly" persists even as performance degrades. The best proxy is behavioral: What time is it? How long have I been at this without a break? When did I last eat? These are better indicators than your internal confidence level.

---

There is a structural injustice embedded in depletion that goes beyond individual decision-makers.

People with less power — in prisons, in hospitals, in immigration proceedings, in child welfare systems, in low-wage workplaces — are disproportionately on the receiving end of depleted decisions. They are the ones whose cases pile up at the end of sessions. They are the ones whose attorneys are overworked. They are the ones whose caseworkers are understaffed. They are the ones who can't afford private care that routes them to less depleted providers.

The people who most need someone to see them as fully human are systematically matched with the people who are least biologically capable of doing so at that moment.

This is not random. It is structural. Systems are not organized around the humanity of the least powerful; they are organized around the throughput of the most powerful. The institution doesn't care if the judge is hungry. The institution cares if the docket is cleared.

Changing this requires understanding that human biological limits are not personal failures — they are design constraints. Building humane institutions means building around those constraints, not demanding that people transcend them.

---

Carl Jung's concept of the shadow — the repository of everything the ego refuses to acknowledge about itself — maps interestingly onto the depletion literature. The depleted self is closer to the shadow. It's less curated, less controlled, more reactive, more categorical, more prone to projection.

This doesn't make it truer in some romantic sense. The rested self is not false. But the gap between our curated values and our depletion-state behavior is diagnostic. What do you do to other people when you are tired? Who do you fail to see? What snap judgments do you make that your rested self would interrogate?

That gap is an honest answer to the question: who are you when the performance drops?

The work of integration — bringing the depleted self into conscious awareness, understanding your particular regression patterns, building structures that protect yourself and others from your worst moments — is part of what it means to take your humanity seriously. Not just asserting good values, but accounting for the biological reality that those values are implemented by a body with limits.

---

1. Depletion audit. For one week, track your most important decisions and interactions alongside three data points: hours of sleep the previous night, time since last meal, hours since last real break. At the end of the week, look for patterns. Do your worst moments cluster around specific states?

2. Pre-decision check-in. Before any consequential decision — a performance conversation, a hiring choice, a significant commitment — spend 30 seconds on three questions: Am I rested? Am I fed? Have I been at this long enough that I should pause? Make it a habit, not a luxury.

3. The institution question. Identify one context where you have decision-making power over others. What are the structural conditions of that context? Are people being asked to make high-stakes calls in conditions that guarantee depletion? What would need to change?

4. Empathy accuracy practice. When you notice yourself making a quick categorical judgment about someone — in traffic, in a meeting, in a difficult interaction — ask: Am I depleted right now? Would I think this if I were rested? The question doesn't have to produce a different answer. But it builds the habit of treating your own state as relevant data.

---

If every person in a position of power over others — every judge, doctor, manager, teacher, caseworker, legislator, officer — genuinely accepted that their biological state affects the quality of justice they dispense, and built their practices accordingly, the downstream effects would be enormous.

Fairer parole decisions. More accurate diagnoses. Better employment decisions. More nuanced child welfare outcomes. More careful legislation.

The people who would benefit most are the people who are already most vulnerable to the whims of other people's decisions. That's not a coincidence. The moral weight of this is not hypothetical.

Taking care of yourself is not separate from caring about others. At the level of biology, it is exactly the same thing.