Experiencing Self vs. Remembering Self

Kahneman’s distinction between the self that lives the moments of an experience and the self that retrospectively summarizes them and decides whether to repeat — a distinction that makes experience design a dual-target discipline.

Definition

The experiencing self is the self that exists in the present moment of the experience. It feels what is happening as it happens: the warmth of the lobby, the edge of the seat, the ten seconds of cold while the elevator is called, the music that has been playing at 62 dB for the last forty minutes, the smell of cypress that arrived at 12 lux and is now decaying. Daniel Kahneman’s working operationalization of this self, in Thinking, Fast and Slow and the underlying experimental work, is roughly the answer to “how do you feel right now?” sampled on a fine-grained schedule across the duration of the experience. The integral of those samples across the duration is what economists call instant utility, and Kahneman calls hedonic flow.

The remembering self is the self that exists outside the experience and looks back on it. It does not have access to the fine-grained sample stream. It has access to a compressed summary, typically dominated by the most intense moment and the final moment with the middle largely thrown away, and that summary is what gets reported on a survey, written in a review, told to a friend, and used to decide whether to come back. Kahneman’s working operationalization is the answer to “how was it?” asked some time after the experience has ended.

The cleanest summary of the distinction is Kahneman’s own. The experiencing self lives the present moment; the remembering self keeps the score and writes the story. The two are not redundant. They disagree. And the disagreement is structural, not noise.

The classic demonstration is the cold-pressor study reported in 1993 by Kahneman, Fredrickson, Schreiber, and Redelmeier, which is also the founding study of the peak-end rule. Subjects underwent a 60-second submersion of one hand in painfully cold water (the short trial), and a separate 90-second trial that began with the same 60 seconds at the same temperature and was followed by 30 seconds in which the water was raised to a slightly less painful but still-uncomfortable temperature (the long trial). When asked which trial they would prefer to repeat, a majority chose the longer one: the trial whose experiencing self had endured every second of the short trial plus an additional thirty also-painful seconds. The choice is locally irrational on any preference theory that respects strictly more pain. It makes sense only if the remembering self’s summary, dominated by the milder ending of the long trial, is the self being asked the question.

What the cold-pressor study makes formal is the premise on which experience design rests: an experience can be poorly lived and well remembered, or well lived and poorly remembered. The two selves do not necessarily produce the same answer to “was that good?”, and the operator who designs only for one of them is leaving the other on the table.

Why It Matters

The distinction is not philosophical. It is the framework that resolves several of the most common contradictions a working brief produces, and it is the single most useful piece of cognitive vocabulary the book imports from the lab.

The first contradiction it resolves is the long-pleasant-versus-short-painful paradox. Why does a five-day Caribbean trip with calm weather, decent food, and a small but persistent inconvenience at the close (a delayed flight, a curt last interaction at the front desk) get rated lower in retrospect than a three-day Caribbean trip with one extraordinary peak and a great send-off? On any average-utility model, the longer trip wins; the experiencing self lived through more pleasant minutes. On the two-selves model, the longer trip loses because the remembering self isn’t averaging. It’s pricing the trip on a peak and a close, and the close it has is poor. The contradiction stops looking like a paradox once the operator stops conflating the two selves.

The second contradiction the distinction resolves concerns the in-stay metric trap. Hospitality and themed-entertainment operators measure satisfaction in real time: the in-app pulse after a meal, the front-desk-to-room-key check-in score, the post-ride rider survey at the platform exit. Those metrics sample the experiencing self. Weeks later, the same operators measure remembered satisfaction: net promoter score, the post-stay review, the willingness-to-return survey. Those metrics sample the remembering self. The two metric streams diverge on the same trip routinely, and the divergence is read as noise, methodology error, or population drift. It is none of those things. It is the gap between the two selves talking. An operator who measures only one of the two is missing exactly half the diagnostic the framework provides; an operator who measures both can pinpoint where on the curve the experiencing self was unhappy that the remembering self forgave, and where the experiencing self was content that the remembering self downgraded for a bad close.

The third contradiction the distinction resolves is the design-for-whom question that creeps into briefs once the operator’s revenue model is on the table. A repeat-business model (luxury hotels, theme parks with annual passes, immersive-theatre productions whose run depends on returning audiences) is, financially, a model addressed almost entirely to the remembering self. The remembering self is the one who books again, recommends, writes the LinkedIn post, and signs up for the loyalty tier. A one-shot model (a destination wedding, a museum’s blockbuster touring exhibition, a single-night brand activation) is more nearly a model addressed to the experiencing self, plus the social-residue effect of the remembering self after the fact. The framework lets the brief say which self the budget is buying, rather than smudging the question.

What this changes in practice is the shape of the working brief. Once the two selves are part of the operator’s vocabulary, design choices stop reading as a single optimization problem and start reading as a dual-target one. The choreographed beats and the flow-channel pacing target the experiencing self. The peak-end composition, the engineered farewell, and the trophy artefact target the remembering self. The brief allocates budget across the two and accepts that the same budget produces different results depending on which self it is being spent on.

How It Shows Up

Three cases show the dual target in operation, at three time scales and three settings.

A theme-park day at Walt Disney World (Walt Disney Imagineering, 1971–present, with the closing-sequence design recurringly featured in The Imagineering Field Guides and in Cornell Hospitality Quarterly coverage of themed-attraction design). A typical day at Magic Kingdom is engineered as a dual-target sequence rather than a continuous-quality optimization. The experiencing self is held in flow through wayfinding choreography (a clear sightline to the castle, paced reveals down Main Street), through ride pacing that calibrates challenge against guest tolerance, and through ambient-condition tuning that keeps the average sample reasonably high — the parade times, the ambient music level, the queue-line theming that turns a forty-minute wait into a deliberate prologue rather than dead time. The remembering self is paid in large beats: a single signature ride that produces the day’s peak, a fireworks sequence at the end of the night, and the choreographed “kiss goodnight” — the lighting ramp, the John Williams musical cue, the cast-member send-off described in peak-end rule. The day’s working spreadsheet, leaked over years through Imagineering and operations writing, treats the two budgets separately. The middle of the day is held above a floor; the peak and the end are spent into deliberately. A day whose middle-of-day pulses look mediocre to the in-park survey but whose closing-moment NPS is strong is, on the two-selves analysis, a successful day.

A tasting-menu restaurant evening (a representative top-tier example: Eleven Madison Park, New York, opened 1998, three-Michelin-star tasting-menu format documented across The New York Times Magazine, Eater, and Will Guidara’s own Unreasonable Hospitality). A two-and-a-half-hour seven-course tasting menu is one of the cleanest miniature dual-target environments in commerce. The experiencing self is paid second by second: the table choreography, the bread course, the timing of the first wine pour, the moment a server clears the small plate at exactly the right beat. The remembering self is paid in the peak (the dish that lands as the meal’s signature — at EMP, the celery-root dish with fresh black truffle, or whatever the rotating signature is during the season the guest visited) and in the close (the post-dessert kitchen tour, the picnic-basket take-home of granola and a hand-written thank-you note that arrives as the table is being walked to the door). Will Guidara’s Unreasonable Hospitality is explicit about this dual budget under a working frame: the kitchen runs the experiencing self’s clock; the front-of-house team runs the remembering self’s. A meal whose middle is well-paced but whose close is operational rather than ritualized is a meal whose remembering self walks out unsigned-for, even when the experiencing self ate well.

An immersive-theatre run such as Sleep No More at the McKittrick Hotel (Punchdrunk and Emursive, 2011–2024 in the New York run, design analyzed in the Studies in Theatre and Performance environment-behavior paper, 2023, and recurrently in the trade press). Sleep No More offers an unusually pure case where the two selves’ budgets are visibly different design moves. The experiencing self is held in flow and narrative transportation for roughly three hours of free-roam masked exploration: small choreographed rooms, branching scenes, the felt sense of being inside a story rather than watching one. The remembering self is paid in two deliberate moves the production engineered onto the front and the back of that three-hour middle: a briefing ritual at arrival (the mask handoff, the freight-elevator ride, the bartender’s preface) and a strong, often emotional close (the third-act reunion sequence in the ballroom, the final unmasking outside the venue). The reviews and social-media residue, captured across eight years of NY-run press, fixate on the close. The experiencing self lived three hours; the remembering self files the briefing ritual and the ballroom finale. The production’s design budget was visibly allocated to that dual target — the choreography of the middle hours is, by industry standards, austere; the front-and-back beats are not.

The three cases run at three time scales (a 14-hour park day, a 2.5-hour meal, a 3-hour immersive run) and three settings (themed entertainment, hospitality / restaurant service, immersive theatre). Each is the same shape: budget the experiencing self’s curve to stay above a defensible floor, then spend the marginal dollar on the moments the remembering self will price.

Caveats and Open Questions

Four seams matter to working practice, and a serious entry has to name them.

The first is the recursive-return-visit case. Repeat customers are not naïve subjects. The second visit’s experiencing self has memories of the first visit, which means the first visit’s remembering self is sitting inside the second visit’s experiencing self, shaping expectations, anchoring comparisons, and mediating the in-the-moment hedonic stream. In the lab, the cold-pressor study isolated the two selves with a gap and a survey. In the field, the gap is filled by reviews, by photographs, by anticipation, and by social conversation, all of which reshape the remembering self’s summary continuously between visits. The implication for design is that the “remembering self” the operator is paying is not static; it is being re-edited every time the guest tells a story about the place. Some patterns (the trophy artefact, the shareable moment) are explicitly designed to take advantage of this re-editing.

The second is the aggregate vs. individual seam. The two-selves framework was developed and validated at the level of individual decision-making. Group experiences (a family at a theme park, a four-top at a restaurant, a couple on a Punchdrunk weekend) have a remembering self that is socially mediated. The summary that gets stored isn’t the average of the four individual summaries; it’s a co-constructed account stitched together over the post-trip dinner conversation. The dominant moments of that group account aren’t always the dominant moments of any individual member’s curve. The honest reading is that the two-selves model predicts well at the individual level and predicts the shape of the group-level summary without quite resolving how the group’s voices are weighted. Operators serving groups (most hospitality and themed-entertainment operators) should design for moments that read as peaks for at least one member of the group rather than for an averaged curve no group will actually compute.

The third seam is the cultural variation one. The lab work has been replicated in North American, Western European, and Japanese samples with the basic shape holding. Less is known about whether the same compression operates the same way in cultures where retrospective summarization is socially mediated to a degree the WEIRD-sample literature does not capture, or where the retrospective-evaluation moment is anchored to a culturally specific ritual (a wedding peak, a religious procession, a celebratory toast) rather than to the close of the experience itself. Practitioners working across cultures should treat the framework as a strong predictor where it has been tested and as a working hypothesis elsewhere.

The fourth and most important seam is the ethical guardrail. Once the operator knows that the remembering self is what gets billed, a tempting reading of the framework is that the experiencing self’s distress can be traded for the remembering self’s satisfaction. Kahneman has been explicit, in the colonoscopy case in particular, that this trade is permissible only when the experiencing self isn’t worse off in net terms. Extending a procedure with a milder ending was acceptable because the original procedure’s distress was already determined; making the experiencing self’s life worse (extending genuine suffering) to give the remembering self a better story is a manipulation, not a design move. The book’s antipatterns shelf names the corresponding traps under Synthetic Scarcity and the broader Manufactured Authenticity family. Operators who internalize the two-selves model without internalizing this guardrail end up with experiences that are remembered fondly and lived badly, and, eventually, with reviewers who notice.

A separate caveat about the trade-press misuse of the framework. “Experiencing self / remembering self” appears in customer-experience copy with a regularity that often skips the experimental basis and the caveats. The book’s job, per policies/style.md, is to be the reference that surfaces the limits the surrounding literature isn’t surfacing. Where this entry is cited, it should be cited with the caveats above attached.

Related Patterns

Sources

• Daniel Kahneman, Thinking, Fast and Slow (Farrar, Straus and Giroux, 2011), Part V, chapters 35–38. The canonical book-length treatment of the two-selves distinction in Kahneman’s own voice, with the colonoscopy-extension thought experiment, the vacation-memory thought experiment, and the formal contrast between experienced utility and decision utility.
• Daniel Kahneman, Barbara L. Fredrickson, Charles A. Schreiber, and Donald A. Redelmeier, “When More Pain Is Preferred to Less: Adding a Better End,” Psychological Science (November 1993), Vol. 4, No. 6, pp. 401–405. The cold-pressor study and the founding experimental statement of the two-selves distinction; cited together with the peak-end rule entry that draws on the same paper.
• Daniel Kahneman, “The Riddle of Experience vs. Memory,” TED talk, February 2010. The shortest and most-cited public explanation of the two-selves distinction in Kahneman’s own voice; the version most readers will have encountered first.
• Donald A. Redelmeier and Daniel Kahneman, “Patients’ Memories of Painful Medical Treatments: Real-Time and Retrospective Evaluations of Two Minimally Invasive Procedures,” Pain (July 1996), Vol. 66, Nos. 1–2, pp. 3–8. The clinical replication of the cold-pressor finding and the source of the colonoscopy thought experiment that Thinking, Fast and Slow later popularized.
• Daniel Kahneman, Peter P. Wakker, and Rakesh Sarin, “Back to Bentham? Explorations of Experienced Utility,” The Quarterly Journal of Economics (1997), Vol. 112, No. 2, pp. 375–406. The formal economic statement of the experienced-utility framework that distinguishes the experiencing self’s instant utility from the remembering self’s decision utility; the substrate paper most subsequent two-selves work cites.