A recent analysis of written accounts from 42 survivors of the 2017 Las Vegas mass shooting suggests that survivors’ memories of the attack were shaped by extreme stress in selective, patterned ways rather than impaired across the board. Survivors documented vivid sensory impressions, spatial awareness focused on escape, and heroic acts by themselves and others, while frequently reporting that time felt slowed down or fractured. The findings, published in Psychological Science, offer a rare look at how people remember a catastrophic event they experienced together.
When individuals face life-threatening danger, their autobiographical memory, which is the mental archive people use to store and recall personal life experiences, responds in complex ways. Psychological theories suggest that acute fear triggers attentional narrowing, meaning that awareness contracts like a spotlight to prioritize information needed for immediate survival while filtering out less relevant surroundings. Encoding, the process by which an experience is stored in memory, may then favor central threat details while background features fade.
Prior evidence supports this view of how severe threat affects recall. For example, a 2017 meta-analysis found that stress around the time of learning tended to impair memory, except when the information was directly tied to the stressful event, in which case memory improved. Similarly, a 2022 study of 21 people recently exposed to trauma indicated that within days of an event, memories of the most distressing moments, often called hotspots, were dominated by physical sensations, motion, and a sense of space.
At the same time, life-threatening experiences often warp the perception of time itself. A 2007 experiment involving frightening free falls suggested that the feeling of time slowing down is an illusion that emerges when looking back on richly encoded memories rather than faster visual perception in the moment. This is still debated, however, with researchers unsure whether the effect arises in perception, memory, or both.
The new study was led by William Ridgway, an operational psychologist at NASA’s Lyndon B. Johnson Space Center and a lecturer in forensic psychology at the University of Nevada, Las Vegas, alongside colleagues at the university and other institutions. All members of the coding team lived in Las Vegas. The researchers set out to examine what survivors themselves emphasized when recalling the shooting, independent of clinical diagnoses.
“Most trauma-narrative research comes from a clinical vantage point, and the memory-focused work that exists tends to pool survivors of very different events, which makes it hard to separate general features of trauma memory from event-specific ones,” Ridgway told PsyPost. “It’s also the case that little qualitative work exists on mass shootings, and on the content of survivors’ own memories.”
“Ultimately, the main catalyst was closer to home,” he explained. “This was our city, and what people wrote in the weeks after October 1, 2017, was a story that needed to be told, and told in survivors’ own words. That’s why we chose an approach that centers what survivors themselves emphasized, and why the paper is built around their excerpts.”
On October 1, 2017, a gunman on the 32nd floor of the Mandalay Bay resort and casino in Las Vegas fired into a crowd of 22,000 concertgoers at the Route 91 Harvest Festival for about 10 minutes. The attack, the deadliest mass shooting in modern American history, killed 60 people and injured more than 800.
Within two weeks of the tragedy, the researchers launched an online survey, recruiting participants through local media, social media, community events, and university announcements. That broader study included both concert attendees and Las Vegas residents who were not at the festival. From it, the authors selected 42 adult survivors who were physically present and submitted written accounts within one month. Of these, 74 percent were women, and their average age was 42.
Participants completed five open-ended prompts at their own pace. The first prompt asked them to recount what happened around them, describe sights, sounds, smells, or other sensations, and explain how they left the festival grounds. The remaining four asked where they were located, who they were with and whether they were separated, what emotions they felt, and what they still wished they understood.
The team analyzed the responses using codebook thematic analysis, a qualitative method (analyzing words and themes rather than numbers) for identifying recurring patterns across texts. Three coders independently reviewed an initial subset of 10 narratives to build a shared codebook before coding the remaining 32 accounts. That process generated 52 initial codes that were consolidated into 39, grouped into six candidate themes, and refined into four final themes: contextual memory, physical injury, heroism, and contemplating mortality.
On December 6, 2023, the day the first coding meeting was originally scheduled, a gunman killed three faculty members on the University of Nevada, Las Vegas campus while some members of the coding team were in an adjacent building. The meeting was postponed until the next day, and the authors acknowledge that the experience may have heightened their sensitivity to fear, spatial orientation, and immediate threat in the survivors’ accounts.
Ridgway said the study’s strength lies in what emerged across the group. “In this case, given the study’s qualitative nature, prevalence is reported in broad terms rather than counts,” he explained. “What carries weight is convergence, not magnitude. Dozens of people independently described the same ten minutes, producing accounts with a recognizable shared structure despite real individual variation.”
Contextual memory was the most pervasive theme and was divided into sensory features, space, and time. Sensory details appeared in all 42 narratives, the only feature present in every account, and frequently began with gunshots being mistaken for fireworks, with survivors registering the sounds distinctly before grasping what they meant. As the attack unfolded, accounts focused on the smell of gunpowder, the sound of automatic gunfire, and blood on sidewalks and doors during the aftermath.
Spatial details appeared in nearly every account, as survivors mapped out food trucks, lawn chairs, dead ends, and landmarks such as nearby casinos, the airport, and fuel tanks while trying to escape. Although spatial context is often treated as background information that degrades under high stress, navigating to safety during an attack makes spatial layout directly relevant to survival, which the authors suggest helps explain why these details were so prominent.
Many survivors also reported distortions in how they experienced time, most commonly feeling as though events unfolded in slow motion or lasted an eternity. Some survivors also described temporal gaps, periods during the escape that were a complete blank in memory, which contrasted with the vivid sensory and spatial details those same individuals reported elsewhere in their narratives.
These observations align with a 2023 laboratory experiment, which found that acute stress enhanced recognition memory for individual items while impairing memory for the chronological sequence connecting them. Although that laboratory study measured item and order recognition under controlled conditions rather than qualitative narratives from real-world mass shooting survivors, both suggest that stress can heighten recall for isolated moments while disrupting the broader timeline.
Physical injury, whether personally sustained or witnessed, was referenced by nearly all survivors. One survivor who was shot in the leg described the exact entry point in the shin and exit path through the calf, though the authors note that such anatomical precision may reflect information learned later during medical treatment.
Other participants recounted seeing fatal wounds or tripping over someone on the ground without realizing until later that the person had been shot. Several survivors independently described the venue and surrounding streets as looking like a war zone, mirroring language nurses used in a separate study of medical staff who treated victims that night.
Roughly half to a majority of the narratives described acts of heroism, defined as prosocial, or helping, actions carried out at personal risk. Survivors recounted applying pressure to a stranger’s stomach wound while bullets struck the asphalt around them, using belts as makeshift tourniquets, and lying on top of wounded concertgoers to shield them from gunfire.
Other accounts described quiet emotional support, such as wrapping a kimono around a stranger in shock and holding her hand, whereas deliberately self-interested behavior, such as a woman briefly trying to close a hotel room door on others seeking shelter, was rarely reported.
In a separate study, nurses who treated the shooting victims independently described their patients as remarkably selfless, noting that wounded survivors gave up hospital beds and comforted one another. The authors note that this convergence supports the idea that the heroism theme reflects actual conduct rather than merely self-flattering recall, even if social-desirability bias cannot be ruled out.
The prominence of helping actions stood out to the research team, particularly given how memory typically operates in laboratory experiments. “Several things surprised me, but as I write this, what stood out is how prominent acts of heroism were in survivors’ memories,” Ridgway reflected. “Not because I was surprised they happened, but because laboratory research generally finds that people remember positive emotional content less than negative, so I expected these narratives to be organized almost entirely around threat.”
“Instead, many survivors described someone helping under fire as precisely as the gunfire itself, which suggests that, at this level of arousal and personal stakes, positive moments may be encoded much like the threat,” he observed.
Roughly half of the participants also described moments when they confronted the possibility of their own death, most often marked by fear and confusion rather than the calm or transcendence typical of classic near-death accounts, though a few did report a sudden sense of peace and acceptance. Survivors recalled thinking about loved ones, such as children or a fiancé, and feeling an urgent need to call or text family members to tell them they loved them.
One survivor slipped into the present tense while describing preparing to die, as did another when recounting comforting a stranger in shock. Prior research links this linguistic shift to flashback-like reliving, in which an intense moment is recalled as if it were happening now.
Asked what the public should take from the findings, Ridgway pointed to the selectivity of memory. “For the layperson, I’d say these accounts suggest that memory under extreme threat is selective rather than globally impaired,” he stated. “Survivors wrote with notable precision about what they saw, heard, and smelled, and where they ran, while for some the sequence of events fell away, with time both stretched and lost.”
“Our work also captures dimensions of the human experience itself, not just how it was remembered: brutality, compassion, and the confrontation with death,” he added. “Survivors chose to write about more than threat. Many described acts of kindness as they described the shooting, which suggests that, at this level of intensity, what gets remembered is whatever mattered most in that moment, whether that was danger, escape, or an act of heroism.”
There are a few things to keep in mind when interpreting these findings. Because the study did not verify survivors’ accounts against objective records of the event, the findings describe what participants emphasized and how their memories were structured rather than how accurate those recollections were. In addition, participants typed their responses at their own pace online, giving them opportunities to edit and revise their accounts in ways that spontaneous spoken reports do not allow.
Ridgway noted that the phrasing of the survey questions may have shaped the responses. “The caveat I’d highlight is that our first prompt asked directly about sights, sounds, smells, and how people left the grounds, so the prominence of sensory and spatial detail partly reflects what we asked,” he pointed out. “The temporal distortions, the heroism, and the confrontation with death weren’t directly prompted, which is why I find them most compelling.”
The sample was also predominantly female, which may have influenced which details stood out, such as one mother’s intense focus on an abandoned baby stroller during her escape. Festival attendance was also self-reported without independent verification. The authors also note that because fear, urgency, horror, and compassion were woven into every theme rather than analyzed as a separate category, the study may underrepresent the overarching role of emotion in structuring trauma narratives.
Looking ahead, Ridgway outlined several open questions for future work. “For me, this chapter is closed for now, though I may come back to it,” he shared. “The questions I’d want answered are whether these patterns hold under neutral prompts and oral interviews, how the narratives change over years, and whether survivors of other forms of mass violence tell stories with the same shape.”
Ridgway added a final note about who the work belongs to. “What I would add is that the work belongs to the people who shared their memories, to everyone who was there that night, and to the city that lived through it with them, as much as it belongs to us,” he said.
The study, “What Survivors Remembered: Cognitive and Experiential Themes in Trauma Narratives From the Deadliest Mass Shooting in Modern American History,” was authored by William B. Ridgway, Kevin D. Mohawk, Rhiannon N. Soriano Smith, Renato M. Liboro, Charles A. Morgan III, Stephen D. Benning, and Colleen M. Parks.