Why Do We Cry Over Fictional Characters?

Sooner or later, when immersed in a compelling book, it happens to everyone: a character dies or leaves the story, and suddenly our eyes well up and a lump forms in our throat; two characters circle each other in a slow-burn romance, and we root for them from the sidelines; a hero achieves their goal, and we celebrate right alongside them.

Yet almost immediately, a thought creeps in, tinged with a hint of embarrassment: "Wait, it's just a book. That character doesn't even exist." That may be true, but what happens beforehand opens the door to a fascinating question that philosophy and psychology have debated for decades: how can we feel genuine grief, joy, or heartache for someone who was never real to begin with? The answer lies in an interplay of several psychological mechanisms: simulation, immersion, identification, and attachment—leading, ultimately, to a deeper moment of introspection.

In 1975, philosopher Colin Radford tackled this very question by formulating what is now known as the paradox of fiction. It rests on three premises, each intuitively true on its own, but contradictory when taken together:

1. We feel genuine emotions toward fictional characters;
2. To feel genuine emotion about something, we must believe it exists or has existed;
3. We know that fictional characters do not exist.

To illustrate this, Radford offered a simple thought experiment: if a stranger in a bar tearfully confides in us about his sister's illness, we are moved to compassion. But if we discover moments later that he is an only child, our emotional reaction changes immediately and radically. When reading a novel, however, knowing full well the characters are entirely made up does nothing to dampen our emotions. Radford concluded from this that feeling moved by fiction is fundamentally irrational.

Other philosophers pushed back against this premise, proposing three alternative schools of thought: some, like Kendall Walton, argue that emotions evoked by reading are "quasi-emotions"—products of a playful, make-believe engagement; others contend that emotional response does not require belief in real existence, only a vivid mental representation; still others suggest that a great novel generates a genuine, temporary illusion of reality as we read.

Psychology and neuroscience have also weighed in. In 2016, a team led by cognitive psychologist Marco Sperduti demonstrated that when people view distressing scenes framed as fictional, their subjective emotional intensity drops slightly compared to scenes presented as real, yet their physiological arousal remains identical. What is more, when a scene resonated with personal experiences or past trauma, the buffering effect of fiction diminished—showing that personal relevance matters far more than whether an event is real or invented. Other research groups have reached similar conclusions: fictional framing creates a quantitative reduction in intensity, not a qualitative difference. Put simply, the emotion may be somewhat muted, but its nature remains unchanged. Consider the smoke detector metaphor: an alarm sounds whenever it detects smoke, unable to distinguish a raging house fire from a lighter held inches from the sensor. Our emotions work much the same way; they respond to what the mind represents, driven by an evolutionary biological system built to react swiftly to stimuli rather than pause to verify their factual authenticity.

In 2008, psychologists Keith Oatley and Raymond Mar introduced the simulation theory of fiction, proposing that stories act as simulators of social life. When we read a novel, we do not merely register information about plot events; we vividly imagine interpersonal scenarios, internal states, and intentions. Fiction essentially functions as a tool for abstraction and simulation of social experience. This theory found empirical support the following year, when a research team led by Nicole K. Speer observed that reading about actions and events involving fictional characters activated brain regions typically associated with visual perception and motor control—the same areas that fire when those sensory and physical experiences occur in real life. These findings revealed that readers do not passively observe a scene from the outside; they mentally simulate it from within. This aligns closely with previous discussions (linked below) on narrative threads and event segmentation theory: the event model constructed by our working memory is anchored in characters, their actions, and their underlying goals. If an event model is a stage or playing field, emotion is what happens when the reader steps onto it.

Of course, not every page we read moves us to tears. Another critical element is required, one familiar to longtime readers of Un tè con Fabulè: narrative transportation, the immersive sensation of being so absorbed "inside the storyworld" that the physical environment fades into the background. Transportation also fosters narrative empathy, as demonstrated in a 2013 experimental study by Matthijs Bal and Martijn Veltkamp, which showed that the more emotionally transported a reader is, the higher their subsequent empathy levels. An important distinction must be made here: identifying with a character does not mean mirroring their exact feelings point for point. Rather, the reader becomes emotionally invested for the character rather than simply with them. When we cry over what befalls a character, we are not directly experiencing their grief; we are experiencing our own grief on their behalf.

This leads to an intriguing realization: our bonds with fictional characters are often stronger and more enduring than those we share with many real-world acquaintances. Why? First, there is the time factor: a substantial novel keeps us intimately connected with a character for days, weeks, or even months, granting us privileged, unmediated access to their inner monologue—an intimacy we rarely achieve with anyone in real life, even our closest loved ones. Second, cognitive event segmentation prioritizes human agents: when building a mental model of any narrative, our minds establish the "who" before anything else.

Media theorists Donald Horton and Richard Wohl coined the term "parasocial interaction" to describe this one-sided psychological bond formed with individuals who are unaware of our existence, or who do not exist at all. Originally formulated to explain the connection viewers felt with television personalities, the concept was later expanded to literature and fictional protagonists. Building on this, Professor Jonathan Cohen of the University of Haifa conceptualized "parasocial breakups," demonstrating that the conclusion of a series, the final chapter of a book, or the death of a beloved character can trigger genuine feelings of grief and separation distress in readers. Once again, this stems from how the human brain is wired: our emotional system responds to the presence of social cues without verifying whether a living person stands on the other end.

All of this might suggest a tidy conclusion: reading fiction automatically makes us more empathetic. Yet the reality is more nuanced, and empirical studies remain divided. While a clear correlation exists between reading and empathy, the notion that opening a novel instantly boosts one's emotional capacity remains debated; it may simply be that highly empathetic people read more to begin with. What we can say with certainty is that emotional engagement with fictional characters provides a rich testing ground for social and cognitive faculties, even if it does not automatically guarantee higher empathy in daily life.

When browsing for our next read, we often deliberately reach for a book guaranteed to make us weep. Why actively seek out sorrow?

Professors Mary Beth Oliver of Penn State University and Anne Bartsch of the University of Leipzig distinguish between two distinct forms of media enjoyment:

1. Hedonic enjoyment, or lighthearted, immediate pleasure;
2. Eudaimonic appreciation, the rewarding sense of having experienced something deeply meaningful, even when painful. Sad stories rarely provide cheerful amusement, but they leave us with a lasting sense of personal depth and significance.

This model is backed by empirical research. In an academic study led by Kaitlyn Fitzgerald and Matthew Grizzard, readers asked to recall fictional character deaths overwhelmingly remembered meaningful losses rather than gratifying ones (such as the downfall of a villain). The deaths that resonated were those that prompted reflection and sorrow, not celebratory cheers. The researchers suggested that fiction allows us to process disenfranchised grief—unspoken or unacknowledged loss—using fictional characters as safe emotional proxies.

Another study published in 2020 by David Huron and Jonna Vuoskoski, originally conducted in music psychology and later extended to other narrative arts, suggests that the pleasure derived from sadness is rooted in compassion. Readers experience a prosocial, positive emotional state when caring deeply about someone else's fate without facing real-world jeopardy themselves. Crying over a novel gives us the healing, prosocial benefits of tears without the devastating fallout of real loss.

It is often said that experiencing intense feelings through a character's journey provides emotional catharsis. First theorized by Aristotle in his Poetics, catharsis was described as the purification or purgation of pity and fear through tragedy. Centuries later, Josef Breuer and Sigmund Freud integrated the term into clinical psychoanalysis using a hydraulic model, comparing emotions to pressurized fluid trapped in a closed vessel that must be vented to prevent an explosion. This visual metaphor proved so compelling that everyday language still revolves around it: "venting," "letting off steam," "getting it off your chest," "releasing bottled-up feelings." Yet experimentally, when applied to intense emotions such as anger, the hydraulic theory falls apart. Between 1999 and 2002, Professor Brad Bushman of Ohio State University found that participants provoked to anger who were encouraged to vent by punching a bag while focusing on their provoker became more aggressive afterward, not less, compared to those who rested or engaged in a distraction. Venting did not drain the anger; it kept it actively stoked.

These findings on anger do not mean tragic literature serves no purpose, or that reading-induced tears are futile. Sad narratives are cautiously associated with improved emotional processing and long-term positive mood recovery.

Catharsis, then, is best understood not as an explosive release of hydraulic pressure, but as the deliberate processing and integration of emotion. A story does not simply drain our feelings; it provides a structured, contained space to experience them fully, name them, and integrate their meaning.

In the end, we cry over characters who do not exist because our emotional architecture answers to internal representations, not to external validation. Reading is mental simulation, and through simulation, we reconstruct a character's inner world as our own. When narrative transportation works as intended, the boundary between our reality and the narrative world softens; we spend hours in deep, private companionship with characters whose thoughts are laid bare to us. Often, what we seek as readers is not effortless entertainment, but meaning. Within the sanctuary of a book's pages, we are free to feel without consequence.

Tears shed over a novel are the clearest proof that the reading experience succeeded: the narrative took hold, transportation was achieved, an event model was built, and a piece of your heart bonded with an imaginary soul. That familiar post-cry embarrassment—the urge to dismiss it all because "it's just a book"—can safely be put to rest.