The phrase describes a real phenomenon: some people who survive cardiac arrest later report hearing sounds, including insects, during the period when their heart had stopped. Research suggests these experiences arise from residual brain activity in the first minutes after circulation ceases, not from consciousness persisting after death.

Emily Dickinson's 1862 poem gave the phrase its cultural permanence, but the scientific question is narrower and more testable. What, if anything, does a dying brain process? The answer sits at the intersection of resuscitation medicine, electroencephalography, and the psychology of memory under extreme stress.

The dying brain's residual electrical activity

When the heart stops, oxygen delivery to the brain halts within seconds. Consciousness typically fades within 10 to 20 seconds of circulatory arrest. But fading is not the same as switching off. Neurons do not die instantly; they depolarize, release stored neurotransmitters, and generate abnormal electrical patterns for several minutes.

The most studied pattern is a surge of gamma-band activity, roughly 30 to 100 hertz, in the moments around cardiac arrest. A 2022 study of four comatose patients undergoing withdrawal of life support recorded such surges in two of them, in temporal and frontal regions, within seconds before and after the heart stopped (Xu et al., Proceedings of the National Academy of Sciences). Gamma rhythms are associated with conscious processing in healthy brains, which is why the finding attracted attention. The caveat is sample size and the fact that correlation with conscious experience was never established.

A separate 2023 study of cardiac arrest patients at a Michigan hospital found that about 40 percent of survivors who completed interviews reported some awareness during resuscitation, including hearing sounds or feeling sensations (Parnia et al., Resuscitation). That figure comes from a small cohort and should not be treated as a population rate.

Why auditory experience may persist longest

Sensory systems do not shut down in unison. Hearing is plausibly the last sense to go because the auditory pathway is fast and heavily myelinated, and because the cochlea can continue transducing mechanical vibration for a short window after blood flow stops. This is the basis for the common clinical practice of speaking to unconscious patients.

The mechanism matters more than the claim. Sound enters the ear, travels the brainstem, and reaches the auditory cortex, which sits in the temporal lobe. Temporal regions were exactly where the dying-brain gamma surges appeared in the 2022 recordings. A dying brain need not be conscious to generate a fragment of auditory experience; it may produce a brief, unstructured impression that memory later stitches into a narrative.

That stitching is the second half of the story. Memories formed during a medical crisis are reconstructed, not replayed. A patient who heard a monitor beep, a door, or a conversation may encode a fragment. On recall, the brain fills gaps with plausible content. A fly buzzing is a culturally available image, and Dickinson's line is among the most quoted in American poetry.

Quantified context: how big is the phenomenon?

Near-death experiences are reported by roughly 10 to 20 percent of cardiac arrest survivors in most studies, though estimates vary widely with interview timing and definition. The 2023 Michigan cohort's 40 percent awareness figure is higher than older estimates, likely because the study asked specifically about awareness and used structured interviews.

Timing is tighter than most people assume. Effective CPR maintains some cerebral blood flow, which can extend the window for residual activity. Without CPR, brain electrical activity declines substantially within minutes and irreversible injury begins within roughly 4 to 6 minutes at normal body temperature. Hypothermia extends that window, sometimes dramatically; cold-water drowning victims have survived neurologically intact after 30 minutes or more of submersion.

The 2022 gamma-surge recordings occurred within a window of seconds to a few minutes around the cessation of circulation. That is not a long time for a rich experience, which is one reason researchers are cautious about claims of detailed awareness.

What the dying brain can and cannot do

Claim Evidence status What the data support
Hearing persists after consciousness fades Moderate Auditory pathway may transduce sound briefly; awareness unclear
Gamma surges accompany cardiac arrest Emerging Recorded in small samples; meaning for experience unknown
Detailed memories form during arrest Weak Memory formation requires intact hippocampal function, unlikely in arrest
Awareness occurs during CPR Moderate Survivor interviews support fragments of awareness in a minority
Consciousness persists after death Not supported No mechanism; brain activity ceases and does not resume
Visual experiences dominate reports Mixed Many reports are non-visual; auditory and tactile fragments common

Common misconceptions, including one that surprises clinicians

The most widespread misconception is that near-death experiences are proof of an afterlife. The evidence does not support that leap, but it also does not explain the experiences away. The honest position is that residual brain activity is a sufficient mechanism for fragmentary experience, and no study has demonstrated experience without brain activity.

A counterintuitive detail: the vividness of a recalled experience does not correlate well with the amount of brain activity at the time. Memories encoded in fragments and reconstructed later can feel more coherent than the underlying event. This is why clinicians who interview survivors months later may hear detailed narratives that the medical record cannot confirm.

Red flags worth knowing:

Reports of detailed awareness during a period with no documented cerebral blood flow deserve scrutiny, not dismissal. Survivor accounts gathered months or years later are vulnerable to reconstruction and cultural scripting. Media claims that a study "proved" consciousness after death almost always overstate small, preliminary findings.

Practical actions for clinicians, families, and patients

For clinicians, the practical implication is simple and already standard in many intensive care units: speak to unconscious patients as if they can hear, and avoid alarming language at the bedside. This costs nothing and is consistent with the possibility of brief auditory processing.

For families, the same advice applies. Narrate what you are doing. Avoid discussing distressing news as though the patient is absent. The probability of awareness is low but not zero.

Concrete steps with thresholds:

Assume auditory processing for at least the first 10 to 20 seconds after circulatory arrest. During CPR, keep bedside conversation calm and orienting until resuscitation ends. If a survivor reports fragments of awareness, document them without leading questions. Do not promise that a dying person hears you; say it is possible and that speaking kindly is reasonable. For research participation, survivors can be interviewed as early as feasible, ideally within weeks, to reduce reconstruction.

Where the explanation breaks down

The mechanistic account has real limits. Gamma surges have been recorded in only a handful of patients, and recording quality during cardiac arrest is poor. Interview studies rely on self-report, which is subject to memory distortion. No study has simultaneously recorded brain activity and obtained a verified account of experience from the same patient.

There is also a definitional problem. "Awareness" in these studies ranges from hearing a voice to a structured narrative. Lumping them together inflates prevalence estimates. And the cultural availability of the fly image means some reports may be shaped by familiarity with Dickinson's poem, a bias researchers cannot easily control for.

What can be said with confidence is narrower than the headlines: the dying brain shows organized electrical activity for a short period after circulation stops, auditory processing is plausibly among the last functions to fade, and a minority of survivors report fragmentary awareness. Whether that awareness includes a fly is unknowable from the evidence and does not need to be knowable for the bedside lesson to hold.

[!NOTE] This article is general educational information about neuroscience and resuscitation science, not a diagnosis, treatment plan, or medical advice for any individual. If you are caring for a dying person or processing a loss, a palliative care team or a qualified local clinician can provide guidance specific to your situation. For immediate danger or thoughts of self-harm, contact local emergency or crisis services.