Do We Become More Aphantasic With Age? A Hungarian Research Team Has Some Surprising Answers

It started, as so much good science does, with a personal question. A graduate student approached Dr. Ilona Kovács at Eötvös Loránd University in Budapest asking whether she would supervise a thesis on aphantasia. What made the request compelling wasn't just academic curiosity — The student is aphantasic herself, and so is most of her family. "Her interest in the topic was much deeper than that of the usual students," Kovács recalls, "which is understandable given that the majority of her family seems to belong to that club as well."

What struck Kovács most, though, wasn't the personal motivation. It was something subtler. "What was very compelling to me was the genuine interest in the topic without her sense of missing out on something." That orientation — curiosity without distress, difference without deficit — would go on to shape the spirit of the research that followed.

Kovács herself is no newcomer to imagery science. A professor of psychology whose career spans both sides of the Atlantic, she began her academic life teaching imagery at the very university she recently rejoined, fell in love with the subject in the 1970s and 80s during the height of the famous Kosslyn-Pylyshyn debate over whether mental images are truly pictorial or merely symbolic, and spent fifteen years in the United States studying visual perception before returning to Hungary to work on sleep, learning, and human development. When this student knocked on her door, Kovács was ready.

The Study: 2,252 Hungarians and One Very Perfect Curve

The team's starting point was a large dataset assembled after the student and her husband gave an interview to a Hungarian online newspaper about aphantasia. A link at the end of the article directed interested readers to the Hungarian version of the Vividness of Visual Imagery Questionnaire (VVIQ), a 16-item self-report measure developed by David Marks in 1973 that remains one of the most widely used tools in imagery research. More than 2,000 people filled it out.

Kovács is clear-eyed about the limitations of this kind of sample. "Obviously it would not be wise to say that we now have the prevalence rate of aphantasia in Hungary," she explains. People who read an article about imagery differences and then choose to fill out a questionnaire are not a random cross-section of the population — they are a self-selected group, likely to include a disproportionate number of people who already suspect something is different about their inner visual experience. Any absolute prevalence numbers from such a dataset should be treated with appropriate skepticism.

But here's what Kovács found when she stopped asking about prevalence and started asking about age: something extraordinary.

When participants were grouped into ten-year age bands and their average VVIQ scores plotted across those groups, a near-flawless declining curve appeared. "A completely and unbelievable perfect declining curve showed up," Kovács says. From the youngest participants to the oldest, imagery vividness dropped with age in a pattern so clean it follows what statisticians call a power trend line — a curve that describes changes which are steepest at younger ages and gradually taper off. "The perfect power curve is actually the dream of an experimentalist," Kovács notes. "It is almost too nice to believe, but it seems to be real."

Who Is Changing? The Four Vividness Groups

To understand what was driving the overall decline, the team divided the VVIQ's 80-point range into four bands. Those scoring 75–80 were classified as hyperphantasic. Scores of 40–74 placed participants in the high intermediate range. Those scoring 20–39 fell into low intermediate imagery vividness. And anyone scoring 16–19 — the questionnaire's floor, corresponding to no mental imagery at all — was classified as aphantasic.

The age-by-group breakdown revealed a clear story. Among the youngest participants, hyperphantasia was relatively common — around 17% of 14-year-olds self-reported it — and aphantasia was essentially absent. "There are no self-reported aphantasics among 14-year-olds," Kovács observes, "while 20% report aphantasia in the oldest group."

Moving up through the age bands, the proportion of hyperphantasic and high-intermediate individuals steadily shrinks, while the proportion of low-intermediate and aphantasic individuals grows. The question this immediately raises: is something being lost over time, or are older people simply more inclined to rate themselves lower?

The Bone Age Finding: Ruling Out a Simple Explanation

This is where Kovács's research takes an unusual turn — one that strengthens the case that the age effect is real rather than a measurement artefact.

Working with a sample of adolescents already enrolled in a study of biological development, the team administered both the VVIQ and an assessment of bone age — a measure of biological maturity based on the structure of the wrist bones, assessed using an ultrasound device. Bone age and chronological age don't always match. A 13-year-old with early-onset puberty may have the biological maturity of a 15-year-old; another child the same age may still be developing at a slower rate. Within any single-year age cohort, the spread in biological maturity can be considerable — Kovács notes that a normal range in puberty onset spans roughly six years.

The finding: even within the same chronological age group, more biologically mature adolescents tended to report lower VVIQ scores. "The more mature an individual is within the same age group, the lower the VVIQ score tends to be," Kovács explains. "Therefore, chronological age in itself does not explain the whole story."

This is a meaningful piece of evidence against the simplest dismissal of the age effect — the argument that older people just habitually score themselves lower on self-report measures without any real change in experience. If that were the whole story, biological maturity within a single age cohort shouldn't predict anything. But it does.

The finding also points toward a biological mechanism. Bone age is driven by hormonal changes during puberty, and puberty coincides with extensive structural reorganization of the brain. "It's as if the brain is sort of growing up at the same time as the kid and the skeleton of the kid," Kovács says. If the same maturational processes that reshape the skeleton also reshape the neural architecture underlying mental imagery, then a biologically-mediated decline in vividness begins to make mechanistic sense.

The Mental Sandbox: A New Way of Thinking About Imagery

To explain what might actually be happening in the brain, Kovács introduces a framework she calls the "mental sandbox." Rather than locating imagery primarily in the visual cortex — which has been the dominant approach in neuroimaging research — she proposes that we look instead at the brain's default network: the large-scale system that becomes most active precisely when we are not engaged with the external world.

"Why should we try and look for neural correlates in the visual network when there is a huge network being active just when we need it — in the absence of stimulation?" she asks. The default network, she argues, is where the real work of mental imagery happens. It is less tightly constrained by the architecture of sensory processing, more variable across individuals, and more potentially plastic over a lifetime.

The sandbox metaphor captures something important about how this space might work. "I think we might find that there is a mental sandbox in there that is largely free from the incoming stimuli and the everyday chores of motion control — and can be freely packed with the building blocks of thinking." Those building blocks, Kovács suggests, come in fundamentally different flavors. Some are episodic — rich in color, detail, and sensory texture, resembling actual perceptual experience. Others are more abstract — organized, categorical, stripped of perceptual detail but perhaps more efficient for reasoning and retrieval.

The central developmental hypothesis is this: as we age and accumulate experience, the sandbox gradually fills with more abstract elements at the expense of episodic ones. Young children may operate primarily with vivid, episodic building blocks. As maturation proceeds — particularly through puberty and into adulthood — abstract representations become increasingly dominant. The result, in the aggregate, looks like a decline in imagery vividness.

Kovács uses a memorable analogy to illustrate why this shift might actually be adaptive. "Try to find a picture from six years ago in your Google Drive — it will not be easy unless you've given it a name and put it in a folder, which is already abstraction." As the volume of stored experience grows, pure episodic recall becomes unwieldy. Abstract organization — categories, labels, schemas — makes memory tractable. The price, perhaps, is vividness.

Two Kinds of Aphantasia, One Sandbox

This framework helps make sense of what Kovács sees as a crucial distinction: the difference between developmental and congenital aphantasia.

Developmental aphantasia, in this model, emerges over time as the sandbox shifts toward abstract building blocks. It reflects a process of normal cognitive maturation — the same process that produces the age curve in the VVIQ data. People in this group may have had vivid imagery earlier in life; it has faded as other cognitive strategies have taken over.

Congenital aphantasia is something different. "If you're born with the ability to work with the more abstract ones from the beginning, then this is what you're going to do," Kovács explains. This form of aphantasia isn't a loss — there was nothing to lose. The sandbox was always organized around abstract rather than episodic elements. Those who recognize their aphantasia as lifelong, as something that has simply always been present, likely belong in this category.

Hyperphantasia, meanwhile, represents the other end of the spectrum: a sandbox in which episodic, vivid building blocks persist and remain dominant even as others around them shift toward abstraction. Whether this reflects genetic factors, particular environmental conditions (Kovács wonders aloud whether growing up in a family of photographers might reinforce episodic imagery), or something else remains an open question.

The Question of Prevalence — and Why It's More Complicated Than It Looks

One of the most practically important points in Kovács's talk concerns something researchers and community members often take for granted: prevalence estimates.

The commonly cited figures for how many people have aphantasia — ranging from roughly 2% to 5% depending on the study — are almost entirely based on samples with significant age biases. Many studies use university students in their twenties. Others draw from online communities where the very act of self-selection skews the sample toward people already curious about, or affected by, imagery differences.

"Can we rely on databases that have an age-biased cohort of only college students, or on studies that do not include teens and kids or old folks?" Kovács asks directly. Her answer is equally direct: no. "At the moment, we do not exactly know what percentage of the population is affected."

If imagery vividness genuinely declines with age — particularly if the proportion of aphantasic individuals grows from near zero in early adolescence to roughly 20% in older age groups — then any prevalence figure drawn from a narrow slice of the lifespan will be systematically misleading. A study of 20-year-olds will undercount aphantasia relative to a representative all-ages sample. The true population prevalence may be considerably higher than current estimates suggest, or it may depend heavily on how you define "the population" in terms of age.

Beyond the Questionnaire: Measuring Imagery Without Asking

Kovács is candid about the limitations of self-report measures — including the VVIQ that anchors most of the work discussed above. "I really don't believe in questionnaires," she says, not dismissively, but with the pragmatism of someone who views them as a first step rather than a final answer. "Questionnaires are excellent in taking the first steps, piloting your thoughts, just trying to see what's out there. But if the tendencies are strong enough, then you start bringing the entire topic into your laboratory and find tools to measure those tendencies more objectively."

That laboratory work is already underway. Kovács's team has developed two paradigms that use binocular rivalry — the phenomenon whereby the brain alternates between two competing visual stimuli presented separately to each eye — to assess imagery vividness without asking participants to report anything at all.

The insight, borrowed and extended from work by Joel Pearson and colleagues, is elegant: if you prime someone with an instruction to imagine a red grating, and then present competing red and green gratings to each eye, neurotypical imagers will subsequently perceive the red grating for longer intervals. Their prior imagery has biased the rivalry. Aphantasic individuals, unable to generate the priming image, show no such modulation.

The clever twist in Kovács's version is that participants don't press buttons or give verbal reports. Instead, the direction of their involuntary eye movements — the slow tracking and fast return that characterizes normal visual following — betrays which grating they are currently perceiving. Brain imaging has confirmed that this "no-report" paradigm activates a very different, and more targeted, neural signature than paradigms that ask for explicit responses, precisely because it avoids engaging the brain areas involved in intentional reporting.

A parallel auditory version of the paradigm has also been developed, using competing sound streams to probe auditory imagery vividness. Early pilot data suggest that visual VVIQ scores predict how auditory imagery priming affects perception — an intriguing hint that imagery vividness may be a domain-general characteristic rather than a purely visual one.

What Comes Next

Kovács lays out an ambitious research agenda. With objective imagery measures in hand, the team hopes to establish the developmental trend on firmer ground — one that cannot be dismissed as a self-report artefact. They plan to correlate VVIQ scores with binocular rivalry results across age groups, bringing participants into the lab based on objectively measured vividness levels to examine the underlying cortical network architecture in detail. Genetic studies on families with high rates of aphantasia or hyperphantasia are also on the horizon.

Throughout all of this, Kovács keeps returning to a broader conceptual point — one she raises at the very beginning and end of her presentation. The term "aphantasia," she suggests, may not be the most accurate frame for what is actually happening. If what distinguishes so-called aphantasic individuals from others is not a missing faculty but a different type of mental building block — abstract rather than episodic, organized rather than pictorial — then calling it an absence may fundamentally mischaracterize the experience.

"Since it is not a lack of something to miss episodic details in our heads, but a difference in the nature of the available building blocks to represent our inner world," she says, "the term aphantasia may not be the best term." What would be better? Kovács smiles and leaves the answer to the audience.