Aphantasia: the feelings are real, they just don't stick around

Close your eyes and picture an apple.

Most people see something. Maybe it’s dim and flickery, maybe it’s practically photographic, but something shows up. For roughly one in a hundred people, nothing does. They know what an apple looks like. They can describe it, recognise it, draw it badly like the rest of us. But the picture never arrives.

That’s aphantasia. It got its name in 2015, from Adam Zeman at Exeter¹, after a run of people wrote to him saying they’d just discovered that “picture this” was never a figure of speech for anybody else.

I went down this rabbit hole for a work reason. I build Subtext, an app that reads the emotional tone of a message before you send it, and I kept circling a question I couldn’t shake: how much of the feeling in a text actually lives in the words, and how much of it gets built by the reader afterwards? Aphantasia turns out to be a very sharp instrument for answering that. And the answer is stranger than I expected.

The thing nobody warns you about

Ask people with aphantasia what’s hard about it and a lot of them won’t lead with the visual stuff. They’ll tell you that feelings don’t stick around.

Not that they don’t feel things. They do, plenty. But once whatever caused the feeling is gone, the feeling tends to go with it. A friend leaves and the warmth fades within the hour. Someone dies and the grief is real but weirdly abstract, more a fact you know than a thing you’re inside. You try to get excited about a holiday six weeks out and there’s just nothing to grab onto.

There’s now a decent body of research explaining why, and the explanation is elegant enough that it changed how I think about emotion generally.

Mental pictures as a volume knob

The idea goes back to a 2008 paper by Emily Holmes and colleagues with the very useful title “Mental imagery as an emotional amplifier”². Their claim: a thought that arrives as a picture hits you harder than the same thought in words.

Which sounds obvious until you sit with it. “My flight is at 6am” is information. Actually picturing yourself standing in a dark kitchen at 4am with a bag you haven’t packed is a feeling. Same fact, different format, and only one of them makes your stomach drop.

Holmes and Mathews later argued³ that imagery does this through a few routes at once. It pokes the emotional systems directly. It builds a simulation close enough to real perception that your body responds as though the thing were happening. And it hooks into your autobiographical memories and drags their attached feelings up with them.

Here’s the part that matters for the fading problem. Most emotions outlast their trigger only because we keep restaging the scene. You replay the argument on the walk home. You picture the diagnosis again while brushing your teeth. You run the reunion in advance eleven times before it happens. Every replay tops the feeling back up.

Take the picturing away and the loop stops turning. The appraisal is fine. You still know the thing was unfair, or sad, or worth wanting. What’s missing is the machinery that keeps refreshing the bodily half of it.

The study that made this concrete

This was a nice theory for over a decade with no clean way to test it. You can’t ask someone to stop visualising on command.

Then Marcus Wicken, Rebecca Keogh and Joel Pearson at UNSW realised aphantasia was the test. Their 2021 paper in Proceedings of the Royal Society B⁴ is still the single best piece of evidence in this whole area.

They wired people up to measure skin conductance, which is a fancy way of saying they measured how much your palms sweat when you get scared. Then they had participants read frightening stories and imagine them.

Typical imagers: response climbs, as you’d expect.

Aphantasic group: flat. Not lower. Statistically indistinguishable from zero.

Then came the second half, which is what makes the study land. Same kind of participants, but this time they looked at actually frightening photographs. Both groups responded the same way. No gap at all.

So this isn’t emotional numbness. Real fear works perfectly. What’s broken is specifically the pipeline that turns words into felt fear, and that pipeline apparently runs through the mind’s eye.

Worth noting how carefully they did it, because a lot of aphantasia research leans hard on self-report. They screened people two independent ways, a questionnaire plus an objective task measuring whether imagining something biases what your eyes see next. Of 29 people who identified as aphantasic, only 22 passed both and made it into the study. They also checked baseline anxiety across groups and found no difference, which kills the obvious objection that the controls were just jumpier people. Pearson later called it by far the biggest difference⁵ his lab had found between aphantasics and everyone else.

It shows up everywhere you’d expect

Once you know what to look for, the same pattern keeps surfacing.

Laura Speed’s team at Radboud⁶ had 47 aphantasic and 51 control participants read a short story. The aphantasic readers were less absorbed, less emotionally engaged, less sympathetic toward the characters. But they liked the story just as much. They weren’t having a worse time. They were having a different one.

A 2025 study in Psychophysiology called “Seeing Is Feeling”⁷ sharpened this by changing the format. Twenty-five aphantasic participants, twenty-five controls, emotionally charged stories delivered as audio or video. Comprehension was equal. Attention was equal. Emotional engagement was lower in the aphantasic group, and the heart rate gap appeared during audio but not video.

Read that again, because it’s the whole thesis in one result. When the scene gets handed to you on a screen, everyone’s the same. When you have to build it yourself, the gap opens.

Merlin Monzel’s group in Bonn⁸ found the same split for sympathy. People with aphantasia responded less, physiologically and by self-report, to people in distress, and the gap was concentrated on written descriptions rather than images. In a related task they recognised emotions in faces just as accurately as controls but noticeably slower, which suggests they’re getting there by a longer route.

Sit with that for a second. The place where the difference shows up most is written descriptions, and written descriptions are now the medium most of us run our closest relationships through.

A 2026 review⁹ sums the whole picture up nicely: attenuated embodied emotion, preserved emotional appraisal. The judgment survives. The body’s half of it thins out.

Memory is where this gets sad

The past is the part that seems to hurt most.

Aphantasia overlaps a lot with something called Severely Deficient Autobiographical Memory¹⁰, where people keep the facts of their own lives but can’t re-experience any of it. In Zeman’s original 2015 survey, about two thirds of aphantasic respondents reported autobiographical memory problems, and over 30% rated their memory as bad, against under 10% of controls.

The reason is that remembering isn’t retrieval. It’s reconstruction. A vivid memory is partly you re-simulating the original sensory experience, and the feeling comes back stuck to the pictures. Alexandra Dawes and colleagues showed¹¹ that aphantasic participants generate memories with fewer episodic details, less sensory language, lower richness, and less emotion attached.

Wilma Bainbridge’s drawing studies¹² give the clearest picture of what survives. Ask people to draw a room they’ve seen and aphantasic participants nail the spatial layout almost perfectly. They just put far fewer objects in it, and often write labels instead of drawing things. They also make fewer false-memory errors, which is a small consolation prize. Structure intact, texture gone.

The brain imaging matches. In an fMRI study of 14 congenital aphantasics and 16 controls¹³, autobiographical recall produced less hippocampal activation and more activity in visual-perceptual regions, with altered connectivity between the two. Participants reported less emotion and less confidence in the memories they pulled up.

And a 2025 paper in Scientific Reports¹⁴ adds a layer I find genuinely interesting. Comparing 69 core aphantasics, 266 hypophantasics and 133 typical imagers, it found lower interoceptive awareness in aphantasia, particularly on “emotional awareness,” meaning the ability to link a bodily sensation to a feeling. So it may not only be that the visual replay is missing. The bodily replay might be quieter too.

Which explains something people describe a lot: knowing you loved someone without being able to feel it on demand.

And the future, which is the flip side

Your brain builds the future out of the same parts it uses to rebuild the past. That’s the constructive episodic simulation¹⁵ idea, and it means the gap shows up in anticipation as well.

A 2026 study¹⁶ tested this in a way that stuck with me. Participants wrote out the sentence “I hope [a loved one] is in a car accident” and then simulated it. The aphantasic group (32 people) reported much lower anxiety, guilt, sense of moral violation, and urge to undo the thought than the 48 visualisers. Both groups rated how likely and how bad the event would be identically.

Same judgment about the facts. Completely different emotional weight.

You can feel both edges of that. On the cost side, anticipatory motivation is harder to summon. Excitement about a trip, dread that makes you actually prepare, the pull of an imagined reward. Positive future imagery is used as a lever in depression treatment, though the evidence is shakier than it’s usually reported to be: the largest trial¹⁷ found no overall benefit over a matched control, with an anhedonia improvement showing up only in exploratory analysis, while separate work on repeated positive imagery¹⁸ did find increased behavioural activation.

On the benefit side, you don’t spiral. Doom-looping about a conversation you haven’t had yet requires being able to stage it, and if you can’t stage it, you’re mostly free of it.

The alexithymia tangle

There’s overlap with alexithymia, the difficulty naming and describing your own emotions. Monzel’s group found higher alexithymic traits in aphantasic participants, especially around describing feelings and what researchers call externally oriented thinking.

But this is where the field has started arguing with itself, and I’d rather show you the mess than pretend it’s settled.

A 2026 study¹⁹ found something properly counterintuitive: within the low-imagery range, people with weak-but-present imagery scored worse on alexithymia and mental health than people with essentially none. Then another team²⁰, using a bigger sample and finer categories, found the externally-oriented-thinking effect holds for the weak-imagery group but not for complete aphantasics, and concluded the shared-cognitive-style story doesn’t survive a closer look.

So the relationship between imagery and emotional wellbeing isn’t a straight line, and having a faint, effortful signal might be harder to live with than having none. If you write about this, that’s a live disagreement, not a finding.

The part that sounds like good news

If imagery amplifies emotion, then not having it should protect you from the disorders built on intrusive imagery. And there’s real evidence for that.

Keogh, Wicken and Pearson ran the trauma film paradigm²¹, a standard lab model of PTSD, showing participants a ten-minute film about the aftermath of a fatal accident. Aphantasic participants reported significantly fewer intrusions immediately afterwards and fewer again in a diary app over the following week. Pearson’s summary was that on essentially everything they measured, the aphantasic group had fewer intrusive thoughts. That one’s still a preprint, so hold it loosely.

Now the correction, which matters more than the finding.

A survey of 2,815 people run with the Aphantasia Network²² found that aphantasia does not protect people from mental illness, PTSD included. Trauma in aphantasia tends to present differently: intrusive verbal thoughts, emotional flooding, bodily sensations, low mood. Because the standard screening tools lean so hard on visual re-experiencing, people get missed.

The Mawtus paper²³ puts it in a line I keep coming back to:

Lack of visual recall may support recovery from some effects of trauma, rather than granting immunity to trauma in the first place.

Monzel, Vetterlein and Reuter’s paper on this is bluntly titled “No general pathological significance of aphantasia”²⁴. Most people, 65.3%, reported little or no distress about having it. A third, 34.7%, reported distress that tracked with lower wellbeing and higher anxiety and depression. Both numbers are true and people quote whichever one suits them.

Why therapy sometimes just bounces off

A surprising amount of front-line psychological treatment assumes a working mind’s eye. Imaginal exposure. Much of EMDR. Imagery rescripting, which has solid meta-analytic support²⁵ across seventeen trials. Positive imagery work. Best-possible-self exercises. Craving and relapse protocols.

For someone with aphantasia, a lot of that machinery simply doesn’t engage. The Mawtus study found people reporting visualisation-heavy CBT as ineffective and describing clinicians who didn’t understand why, with the willingness to adapt mattering most exactly where the stakes were highest, in trauma and complex cases.

But here’s the good bit, and it’s better than I expected.

Monzel’s group ran a registered report in Psychophysiology²⁶ comparing aphantasics, “simulated aphantasics” (controls whose imagery was disrupted with a bright screen) and normal controls on imaginal extinction after fear conditioning. Imaginal extinction still worked in the aphantasic group. Propositional thought, just thinking about the feared thing without seeing it, was enough.

So exposure doesn’t strictly need pictures. It needs the content activated, and there’s more than one door into that room. Some aphantasic patients may even tolerate the work better, since the material is less vivid to sit with.

What people actually do instead

None of this is treatment advice and I’m not qualified to give any. But the workarounds that keep appearing in the research and in what people describe all share one logic: use the channels that still work. Real perception, language, and the body.

Put the cues outside your head. Photos, videos, voice notes, playlists, objects, smells, going back to the actual place. If you can’t restage the scene internally, restage it externally. Bainbridge’s drawing work and the education research both show that aphantasic people naturally externalise what everyone else holds internally, and this generalises from studying to feeling.

Write things down. If the imagery route to emotion is thin, language carries more of the weight, and language responds well to being used on purpose rather than casually. Journaling isn’t a nice-to-have here.

Go through the body. Breath, movement, body scans, mindfulness that skips the “picture a calm beach” step in favour of just noticing what’s actually happening physically. Given the interoception findings this is one of the more promising directions, and also one of the least tested, so treat it as a reasonable bet rather than an established one.

Find out which senses still work. Aphantasia often isn’t visual-only. Most aphantasic people report weaker imagery in other senses too²⁷, and about 26% report a complete absence across all of them. Absent auditory imagery even has its own name, anauralia²⁸. But plenty of people keep one modality intact. Finding it beats trying harder at the one that’s gone.

Tell your therapist early. Both that visualisation instructions don’t work, and that no visual flashbacks doesn’t mean no distress.

What I’d want you to hold loosely

Aphantasia isn’t a disorder. That’s the settled view among the people who study it, including Zeman’s own 2024 review²⁹ of the first decade of research. Everything above describes group averages with a lot of individual variation inside them, and plenty of people with aphantasia have rich emotional lives and deep relationships and would find the framing of this article slightly annoying.

The evidence is also young. Most of it is from 2020 onward, and many key studies run 15 to 50 people per group, which is normal for this kind of work and still a real limit. A lot of it leans on self-report about experiences nobody can verify from outside, though the flagship results are physiological rather than questionnaire-based, which helps.

Causal direction is often murky. Whether weak interoception dampens imagery or the other way round isn’t settled. And the field is currently arguing about whether some form of unconscious representation persists in aphantasia at all, which means the basic question of what’s actually in there is still open.

Why I found this worth writing up

I build a product about emotional tone in writing, so season my enthusiasm accordingly. But this research moved something for me.

The emotional weight of a written message isn’t sitting in the message. It gets manufactured on arrival, by a reader converting your words into something sensory enough that their body responds. The words are instructions. The feeling is whatever the reader assembles from them.

We know this because there’s a group of people whose assembly line runs on different parts, and for them the same words land differently. Not weaker as people. Just differently equipped for a job most of us never noticed we were doing.

Which leaves a practical problem I hadn’t properly seen before. When you reread your own message before hitting send, you’re running it through your simulator. They’ll run it through theirs. You feel the version you meant. They get the version they build. Nearly every text argument I’ve ever had lived in that gap, and the gap is invisible from the inside by design, because the only tool you have for checking is the thing that caused the problem.

That’s the whole reason Subtext exists: a second read from outside your own head, while the message is still yours. The aphantasia work didn’t give me the idea. It gave me a much better explanation of why the problem is so hard to catch on your own.

---

Got aphantasia and think I’ve got something wrong here? I’d like to know. Find me on LinkedIn.

Samet Durgun is the co-founder of Subtext, an app that catches the emotional tone of your messages and rewrites them in your own voice. He’s based in Berlin.

---

Sources

Every link above goes to the primary source where one exists. Full references, in order of appearance:

1. Zeman, A., Dewar, M., & Della Sala, S. (2015). Lives without imagery: Congenital aphantasia. Cortex, 73, 378-380.
2. Holmes, E. A., Geddes, J. R., Colom, F., & Goodwin, G. M. (2008). Mental imagery as an emotional amplifier: Application to bipolar disorder. Behaviour Research and Therapy, 46(12), 1251-1258.
3. Holmes, E. A., & Mathews, A. (2010). Mental imagery in emotion and emotional disorders. Clinical Psychology Review, 30(3), 349-362.
4. Wicken, M., Keogh, R., & Pearson, J. (2021). The critical role of mental imagery in human emotion. Proceedings of the Royal Society B, 288(1946), 20210267.
5. Pearson’s comments via ScienceDaily, March 2021.
6. Speed, L. J., Eekhof, L. S., & Mak, M. (2024). The role of visual imagery in story reading: Evidence from aphantasia. Consciousness and Cognition.
7. Abdelrahman, S., et al. (2025). Seeing Is Feeling: How Aphantasia Alters Emotional Engagement With Stories. Psychophysiology. doi:10.1111/psyp.70100
8. Monzel, M., Karneboge, J., & Reuter, M. (2024). Affective processing in aphantasia and potential overlaps with alexithymia. Biomarkers in Neuropsychiatry, 11, 100106.
9. Emotion Experience in Aphantasia: Insights from Autobiographical Memory, Prospection and Mental Simulation (2026 review).
10. Palombo, D. J., Alain, C., Söderlund, H., Khuu, W., & Levine, B. (2015). Severely deficient autobiographical memory (SDAM) in healthy adults. Neuropsychologia, 72, 105-118.
11. Dawes, A. J., Keogh, R., Robuck, S., & Pearson, J. (2022). Memories with a blind mind. Cognition, 227, 105192.
12. Bainbridge, W. A., Pounder, Z., Eardley, A. F., & Baker, C. I. (2021). Quantifying aphantasia through drawing. Cortex, 135, 159-172.
13. Monzel, M., Leelaarporn, P., Lutz, T., Schultz, J., Brunheim, S., Reuter, M., & McCormick, C. (2024). Hippocampal-occipital connectivity reflects autobiographical memory deficits in aphantasia. eLife, 13, RP94916.
14. Monzel, M., Nagai, Y., & Silvanto, J. (2025). The role of subjective interoception in autobiographical deficits in aphantasia. Scientific Reports.
15. Schacter, D. L., & Addis, D. R. (2007). The cognitive neuroscience of constructive memory. Phil. Trans. R. Soc. B, 362(1481), 773-786.
16. Ji, J. L., Radomsky, A. S., & Andrade, J. (2026). Seeing is believing: mental imagery amplifies moral, emotional, and motivational responding. Neuropsychologia, 231.
17. Blackwell, S. E., et al. (2015). Positive imagery-based cognitive bias modification as a web-based treatment tool for depressed adults. Clinical Psychological Science, 3(1), 91-111.
18. Renner, F., Ji, J. L., Pictet, A., Holmes, E. A., & Blackwell, S. E. (2017). Effects of engaging in repeated mental imagery of future positive events on behavioural activation. Cognitive Therapy and Research, 41(3), 369-380.
19. Kvamme, T. L., Monzel, M., Nagai, Y., & Silvanto, J. (2026). When weak imagery is worse than none.
20. Delem, et al. Complete Aphantasics Process Emotions Differently, But No Less Efficiently. Preprint.
21. Keogh, R., Wicken, M., & Pearson, J. (2023). Fewer intrusive memories in aphantasia. Preprint.
22. Aphantasia Does Not Shield Against PTSD, a survey of 2,815 people with the Aphantasia Network.
23. Mawtus, B., Renwick, F., Thomas, B. R., & Reeder, R. R. (2024). The Impact of Aphantasia on Mental Healthcare Experiences. Collabra: Psychology, 10(1), 127416.
24. Monzel, M., Vetterlein, A., & Reuter, M. (2023). No general pathological significance of aphantasia. Scandinavian Journal of Psychology, 64(3), 314-324.
25. Kip, A., et al. (2023). Efficacy of imagery rescripting in treating mental disorders associated with aversive memories. Journal of Anxiety Disorders, 99, 102772.
26. Monzel, M., Agren, T., Tengler, M., Karneboge, J., & Reuter, M. (2025). Propositional Thought Is Sufficient for Imaginal Extinction. Psychophysiology. Registered Report.
27. Dawes, A. J., Keogh, R., Andrillon, T., & Pearson, J. (2020). A cognitive profile of multi-sensory imagery, memory and dreaming in aphantasia. Scientific Reports, 10, 10022.
28. Hinwar, R. P., & Lambert, A. J. (2021). Anauralia: The silent mind and its association with aphantasia. Frontiers in Psychology, 12, 744213.
29. Zeman, A. (2024). Aphantasia and hyperphantasia: exploring imagery vividness extremes. Trends in Cognitive Sciences, 28(5), 467-480.
30. Greening, S. G., et al. (2022). Mental imagery can generate and regulate acquired differential fear conditioned reactivity. Scientific Reports, 12, 723.