Write it down on the day

01Why memory drifts

When people look back on an experience, they don’t average it. In one study, the pain people remembered after a medical procedure followed its worst moment and its last few minutes, and long procedures were not remembered as especially bad (Redelmeier and Kahneman, 1996). Over a week, the more a person’s pain went up and down, the higher they remembered it, compared with their own average (Stone et al., 2005).

This matters for a one-person test, because the comparison is you against yourself: the days before a change against the days after it. That is where weekly recall did worst. It ranked people against each other fairly well, but followed each person’s own changes from week to week poorly (Stone et al., 2004).

02A day is short enough

Recall doesn’t fail all at once. In a month-long study, remembered ratings ran consistently higher than the ones people had made at the time. Even so, memory of a single day matched that day’s ratings well, and beyond the last few days people found it harder and harder to remember how they had felt (Broderick et al., 2008).

An end-of-day note isn’t perfect either. In another study, people’s end-of-day scores closely followed their scores from random moments through the day, but ran one to two points higher on a scale of 0 to 10, and end-of-week questionnaires matched less well (Bosman et al., 2025). So compare like with like: record the days before a change the same way, and at the same time of day, as the days after it.

03A note written later is memory again

A diary only works if it is filled in on time. In one study, people kept a paper diary three times a day for three weeks, and the binder secretly recorded each time it was opened. By the times people wrote on their pages, 90% of entries were on time. By the binder’s record, 11% were (Stone et al., 2002).

On almost a third of the days, the binder was never opened at all, yet people’s pages showed 92% of those days’ entries as made on time. In the same study, people given an electronic diary that took entries only at the right times made 94% of them on time.

A late note isn’t useless. It is a different thing, and it should say so: write the time on each note, and mark any day filled in after the fact.

04How often is enough

Prompts at random moments are harder to keep up than a note at the end of the day. Across four studies of people recording gut symptoms on a phone, an end-of-day diary was filled in about nine times in ten, while prompts at random moments were answered about six or seven times in ten (Beckers et al., 2021).

Being asked often didn’t seem to change the symptom itself. When people with long-term pain were asked to rate it at different frequencies for two weeks, there was little sign that rating it more often changed their pain (Stone et al., 2003).

A short daily note won’t catch every moment, but it is close enough to follow the day, and people tend to keep it going.

05Food is easy to misremember too

The same goes for what you eat. When a large study checked people’s food reports against biological markers, energy intake was under-reported by roughly 12–20% when people recalled the previous 24 hours, and by roughly a third on questionnaires about their usual diet (Subar et al., 2003). That study didn’t test notes made while eating, but writing food down as you go at least spares you rebuilding the day from memory.

06What the studies found

• In a 1996 study, pain was rated as it happened during two kinds of medical procedure (154 and 133 people) (Redelmeier and Kahneman, 1996). The pain people later remembered tracked the most painful moment and the last three minutes; longer procedures were not remembered as especially bad.
• In a 2004 study of 68 people with long-term pain, recall of each week was compared with that week’s in-the-moment ratings (Stone et al., 2004). The two agreed fairly well across people, but poorly within one person from week to week.
• In a 2005 study of 68 people with long-term pain, those whose pain varied more during a week recalled it as higher, compared with their own average (Stone et al., 2005).
• In a 2008 study, 83 people from a rheumatology practice rated their pain and tiredness about five times a day for a month, and also recalled periods of 1 to 28 days (Broderick et al., 2008). Recall came out higher than the in-the-moment ratings throughout. A single day’s recall still matched that day well, but memory grew patchy beyond the last several days.
• In a 2002 study of 80 adults with long-term pain, half kept a paper diary whose binder held hidden light sensors (Stone et al., 2002). Their own records put 90% of entries on time, the sensors 11%, and on 32% of days the binder was never opened. The other half used an electronic diary and made 94% of entries on time. Three of the five authors had roles at a company that makes electronic diaries.
• In a 2003 study, 91 people with long-term pain were randomly assigned to rate it more or less often for two weeks (Stone et al., 2003). How often they rated made little apparent difference to their pain or to how they recalled it, and at every level at least 94% of ratings were made.
• In a 2025 study of 230 people with irritable bowel syndrome in three European countries, people rated their symptoms from 0 to 10 at up to ten random moments a day (Bosman et al., 2025). Among the 162 who also kept an end-of-day diary, its scores followed the momentary ones closely but were 0.8 to 1.8 points higher. Questionnaires at the end of the week matched the momentary ratings less well.
• In a 2022 study of 149 young adults meeting research criteria for IBS or functional dyspepsia, recall ran higher than the same week’s phone ratings, most of all in people who said a doctor had diagnosed a gut condition (Jones et al., 2022).
• In a 2021 comparison drawing on four studies, 25 people with IBS and 15 with functional dyspepsia kept up an end-of-day phone diary on about 90–93% of days, but answered only 61–70% of prompts at random moments (Beckers et al., 2021).
• In a 2003 study of 484 adults aged 40 to 69, the energy people reported eating was checked against biological markers: it came out 12–20% short on 24-hour recalls, and 31–38% short on food-frequency questionnaires (Subar et al., 2003).

07Sources on this page

Each source is listed once, with links to its record. The reference list says what was read for each, and when it was checked.

1. Beckers et al., 2021Beckers AB, Snijkers JTW, Weerts ZZRM, Vork L, Klaassen T, Smeets FGM, et al. 2021. Digital Instruments for Reporting of Gastrointestinal Symptoms in Clinical Trials: Comparison of End-of-Day Diaries Versus the Experience Sampling Method. JMIR Formative Research 5(11):e31678.DOI 10.2196/31678 · PubMed 34821561 · PMC8663435
2. Bosman et al., 2025Bosman M, Vork L, Jonkers D, Snijkers J, Topan R, Aziz Q, et al.; ESM study group. 2025. Results From a Psychometric Validation Study: Patients With Irritable Bowel Syndrome Report Higher Symptom Burden Using End-of-Day Vs Real-Time Assessment. American Journal of Gastroenterology 120(5):1098–1107.DOI 10.14309/ajg.0000000000003091 · PubMed 39311432 · PMC12043265
3. Broderick et al., 2008Broderick JE, Schwartz JE, Vikingstad G, Pribbernow M, Grossman S, Stone AA. 2008. The accuracy of pain and fatigue items across different reporting periods. Pain 139(1):146–157.DOI 10.1016/j.pain.2008.03.024 · PubMed 18455312 · PMC2617790
4. Jones et al., 2022Jones MP, Yun G, Wass F, Rixon H, Shah A, Walker MM, et al. 2022. The role of mood state and emotion regulation in the discrepancy between gastrointestinal symptom burden recorded prospectively and via recall questionnaire. Neurogastroenterology & Motility 34(7):e14304.DOI 10.1111/nmo.14304 · PubMed 34854512
5. Redelmeier and Kahneman, 1996Redelmeier DA, Kahneman D. 1996. Patients' memories of painful medical treatments: real-time and retrospective evaluations of two minimally invasive procedures. Pain 66(1):3–8.DOI 10.1016/0304-3959(96)02994-6 · PubMed 8857625
6. Stone et al., 2002Stone AA, Shiffman S, Schwartz JE, Broderick JE, Hufford MR. 2002. Patient non-compliance with paper diaries. BMJ 324(7347):1193–1194.DOI 10.1136/bmj.324.7347.1193 · PubMed 12016186 · PMC111114
7. Stone et al., 2003Stone AA, Broderick JE, Schwartz JE, Shiffman S, Litcher-Kelly L, Calvanese P. 2003. Intensive momentary reporting of pain with an electronic diary: reactivity, compliance, and patient satisfaction. Pain 104(1–2):343–351.DOI 10.1016/s0304-3959(03)00040-x · PubMed 12855344
8. Stone et al., 2004Stone AA, Broderick JE, Shiffman SS, Schwartz JE. 2004. Understanding recall of weekly pain from a momentary assessment perspective: absolute agreement, between- and within-person consistency, and judged change in weekly pain. Pain 107(1–2):61–69.DOI 10.1016/j.pain.2003.09.020 · PubMed 14715390
9. Stone et al., 2005Stone AA, Schwartz JE, Broderick JE, Shiffman SS. 2005. Variability of momentary pain predicts recall of weekly pain: a consequence of the peak (or salience) memory heuristic. Personality and Social Psychology Bulletin 31(10):1340–1346.DOI 10.1177/0146167205275615 · PubMed 16143666
10. Subar et al., 2003Subar AF, Kipnis V, Troiano RP, Midthune D, Schoeller DA, Bingham S, et al. 2003. Using intake biomarkers to evaluate the extent of dietary misreporting in a large sample of adults: the OPEN study. American Journal of Epidemiology 158(1):1–13.DOI 10.1093/aje/kwg092 · PubMed 12835280