Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 44 of the 44 references it cites were found in a registry or an index; 21 of the 40 sources its claims cite were checked for support.

Long-term memory reorganization of navigational episodes

Open access

Published version of a preprint

Abstract

During navigation, the brain builds representations of self-motion and of environmental information for future action. The classic view suggests that these representations consolidate and eventually stabilize. However, there are no data on their fate at extended memory delays. Here we investigated memory of real-world navigational episodes across memory delays of up to three decades. We show that memory of the spatial aspects of these episodes do not achieve a stable state but rather continue to transform for many years. Our data suggest that at any given point in time, spatial memory of navigational episodes is a changing combination of episode-independent schematic information and several interacting spatial representations directly related to a navigational episode, which may show distinct temporal trajectories. Consistent with recent accounts of memory reorganization, we further show that neither current theories of systems consolidation nor classic models of forgetting fully explain spatial memory performance at extended memory delays.

Overview

CoChat Summary

TL;DR

In a retrospective study of real-world zoo navigation, spatial memory did not stabilize over time but continued to change across delays from 6 days to three decades. Performance reflected a mix of episodic spatial representations and schematic knowledge, and the decay pattern did not fit classic forgetting theories well.

Key findings

  • Zoo visitors outperformed zoo-naive controls on all three tasks: direction-pointing with video, direction-pointing with arrow, and target placement, all with P < 0.001.
  • Spatial memory declined with time since last zoo visit for all tasks, with reported correlations of ρ(102) = 0.25 (video), ρ(54) = 0.370 (arrow), and ρ(102) = 0.350 (target placement).
  • Time since last visit explained only 11–24% of variance in simple models, but generalized additive models explained up to 58.8%, 61.8%, and 64% of variance for the three tasks, respectively.
  • The decline was task-dependent: the target placement task showed the steepest decay, and real performance declined more steeply than a simulated Ebbinghaus forgetting curve across all tasks (time × source: F(1,559) = 50.233, P < 0.001).

Methods

  • Retrospective behavioral study of 134 participants: 104 zoo visitors and 30 zoo-naive controls, aged 14–71; the delay since last zoo visit ranged from 6 days to three decades.
  • Participants completed three spatial-memory tasks in/with the Berlin Tierpark setting: direction-pointing with video (egocentric), direction-pointing with arrow (egocentric-to-allocentric), and target placement on a 2D map (allocentric), plus questionnaires on demographics and spatial ability.
  • Main analyses included z-score standardization against control performance, Mann–Whitney U tests, Kruskal–Wallis tests, Spearman correlations, regression/GAMs, and comparisons with a modified Ebbinghaus forgetting function.

Limitations

  • The paper states that time since last visit explained only up to a quarter of the variance, so many other factors contribute; it also notes that a combination of time since last visit, age at last visit, and spatial orientation skills explained up to two thirds of variance.
  • The authors explicitly caution that the retrospective design brings ecological setting and uncontrolled confounding factors, which may mask temporal differences and make task-specific trajectory differences hard to interpret.
  • Evident from the text: the study uses cross-sectional retrospective recall rather than prospective follow-up, so causal claims about individual memory change over time are limited.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1038/s41562-026-02472-x

Cites

44 of 44 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Overview map of the zoo—Tierpark Berlin. The locations of the video recordings are marked with a dark blue asterisk. The locations that served as targets are marked with an orange dot. Map data from OpenStreetMap (https://openstreetmap.org/copyright).
Fig. 1 Overview map of the zoo—Tierpark Berlin. The locations of the video recordings are marked with a dark blue asterisk. The locations that served as targets are marked with an orange dot. Map data from OpenStreetMap (https://openstreetmap.org/copyright).CC BY · PubMed Central
Fig. 2 Experimental design. a, The self-localization. Participants are asked to locate themselves in the zoo using the environmental information provided in the 360° video. b, The Berlin Zoo task. The task consists of three different tasks that depend on egocentric and allocentric spatial representations of the zoo. Two tasks are performed from a first-person perspective in a virtual environment: the direction of target location asks participants to point in the direction of a specific location either by using the environmental information provided in the 360° video (egocentric) or by using an arrow pointing in the direction of a third location (egocentric to allocentric). A third task is performed from a bird’s-eye view on a 2D screen. The target placement asks participants to place a specific location in the correct position on an unlabelled map of the zoo (allocentric). Map data from OpenStreetMap (https://openstreetmap.org/copyright).
Fig. 2 Experimental design. a, The self-localization. Participants are asked to locate themselves in the zoo using the environmental information provided in the 360° video. b, The Berlin Zoo task. The task consists of three different tasks that depend on egocentric and allocentric spatial representations of the zoo. Two tasks are performed from a first-person perspective in a virtual environment: the direction of target location asks participants to point in the direction of a specific location either by using the environmental information provided in the 360° video (egocentric) or by using an arrow pointing in the direction of a third location (egocentric to allocentric). A third task is performed from a bird’s-eye view on a 2D screen. The target placement asks participants to place a specific location in the correct position on an unlabelled map of the zoo (allocentric). Map data from OpenStreetMap (https://openstreetmap.org/copyright).CC BY · PubMed Central
Fig. 3 Spatial memory performance of zoo visitors versus zoo-naive controls. a, Zoo visitors showed a lower deviation from target locations (z-score) and higher accuracy (z-score × −1), outperforming zoo-naive controls in all three tasks. Zoo visitors’ performance was better for target placement than for direction-pointing with video or arrow. Box plots are displayed with a centre line as median, Tukey-style whiskers extend 1.5× IQR, dots present individual data points. Zoo visitors: n = 104 (direction-pointing (video) and target placement), n = 56 (direction-pointing (arrow)); zoo-naive controls: n = 30 (direction-pointing (video) and target placement), n = 20 (direction-pointing (arrow)). b, Example maps for performance in all three tasks. Top: direction-pointing with video. Middle: direction-pointing with arrow. Bottom: target placement on an unlabelled zoo map. Coloured direction indicators and dots indicate zoo visitors; grey direction indicators and rhombuses indicate zoo-naive controls. Map data from OpenStreetMap (https://openstreetmap.org/copyright).
Fig. 3 Spatial memory performance of zoo visitors versus zoo-naive controls. a, Zoo visitors showed a lower deviation from target locations (z-score) and higher accuracy (z-score × −1), outperforming zoo-naive controls in all three tasks. Zoo visitors’ performance was better for target placement than for direction-pointing with video or arrow. Box plots are displayed with a centre line as median, Tukey-style whiskers extend 1.5× IQR, dots present individual data points. Zoo visitors: n = 104 (direction-pointing (video) and target placement), n = 56 (direction-pointing (arrow)); zoo-naive controls: n = 30 (direction-pointing (video) and target placement), n = 20 (direction-pointing (arrow)). b, Example maps for performance in all three tasks. Top: direction-pointing with video. Middle: direction-pointing with arrow. Bottom: target placement on an unlabelled zoo map. Coloured direction indicators and dots indicate zoo visitors; grey direction indicators and rhombuses indicate zoo-naive controls. Map data from OpenStreetMap (https://openstreetmap.org/copyright).CC BY · PubMed Central
Fig. 4 Memory of navigational episodes in the zoo over a period of 6 days to three decades since last visit. a, For zoo visitors, data on all three tasks were locally weighted and smoothed for visual inspection. The performance on all tasks was significantly correlated with the time elapsed since the last visit to the zoo. The solid line depicts the regression line; the shaded area indicates the 95% CI. Points represent individual observations. The dashed line indicates mean performance of zoo-naive controls. The dotted lines indicate the first standard deviation above and below the mean performance of zoo-naive controls. b, The power regression for all three tasks revealed a decline in memory performance as a function of the time since the last visit to the zoo. The regression slope differed significantly between direction-pointing (video) and target placement. The solid line depicts the regression line; the shaded area indicates the 95% CI. Points represent individual observations. The dashed line indicates mean performance of zoo-naive controls. The dotted lines indicate the first standard deviation above and below the mean performance of zoo-naive controls. Note: z-scores were shifted into the positive range before applying a log transformation.
Fig. 4 Memory of navigational episodes in the zoo over a period of 6 days to three decades since last visit. a, For zoo visitors, data on all three tasks were locally weighted and smoothed for visual inspection. The performance on all tasks was significantly correlated with the time elapsed since the last visit to the zoo. The solid line depicts the regression line; the shaded area indicates the 95% CI. Points represent individual observations. The dashed line indicates mean performance of zoo-naive controls. The dotted lines indicate the first standard deviation above and below the mean performance of zoo-naive controls. b, The power regression for all three tasks revealed a decline in memory performance as a function of the time since the last visit to the zoo. The regression slope differed significantly between direction-pointing (video) and target placement. The solid line depicts the regression line; the shaded area indicates the 95% CI. Points represent individual observations. The dashed line indicates mean performance of zoo-naive controls. The dotted lines indicate the first standard deviation above and below the mean performance of zoo-naive controls. Note: z-scores were shifted into the positive range before applying a log transformation.CC BY · PubMed Central
Fig. 5 Memory of navigational episodes in the zoo compared with the Ebbinghaus curve and zoo-naive controls. a, The temporal trajectories for all tasks were compared with data simulated using the Ebbinghaus equation (6 days to 30 years). Curves were aligned to a common score at the first minute (outside the observation window), on the basis of their respective power regressions. Each task showed a different slope compared with the simulated data, with the most pronounced deviation for target placement, whose slope differed significantly from both pointing tasks. Solid line, regression; shaded area, 95% CI; points, individual observations; vertical dashed line, 8,460th minute; dashed line, zoo-naive control mean; dotted lines, ±1 s.d. of controls. z scores were shifted into the positive range before log transformation and calculated per individual query. b, The performance of zoo visitors versus controls across two post zoo visit intervals (1–10 years versus 10–30 years). Zoo visitors showed substantial overall retention, with all tasks differing significantly from controls in the first decade. In the later interval, only target placement remained significantly above control levels. A significant main effect of task was found within the early interval, with post hoc tests indicating better performance on direction-pointing (video) versus target placement; this effect was not present in the later interval. Across time intervals, a significant time effect was revealed with post hoc tests indicating a decrease in performance over time for target placement. The dotted line indicates the mean performance of zoo-naive controls. Data presented as box plots, centre line, median; Tukey-style whiskers, 1.5× IQR; dots, individual data points. The 1–10-year interval, n = 44 (pointing (arrow), n = 31); the 10–30-year interval, n = 26 (pointing (arrow): n = 13).
Fig. 5 Memory of navigational episodes in the zoo compared with the Ebbinghaus curve and zoo-naive controls. a, The temporal trajectories for all tasks were compared with data simulated using the Ebbinghaus equation (6 days to 30 years). Curves were aligned to a common score at the first minute (outside the observation window), on the basis of their respective power regressions. Each task showed a different slope compared with the simulated data, with the most pronounced deviation for target placement, whose slope differed significantly from both pointing tasks. Solid line, regression; shaded area, 95% CI; points, individual observations; vertical dashed line, 8,460th minute; dashed line, zoo-naive control mean; dotted lines, ±1 s.d. of controls. z scores were shifted into the positive range before log transformation and calculated per individual query. b, The performance of zoo visitors versus controls across two post zoo visit intervals (1–10 years versus 10–30 years). Zoo visitors showed substantial overall retention, with all tasks differing significantly from controls in the first decade. In the later interval, only target placement remained significantly above control levels. A significant main effect of task was found within the early interval, with post hoc tests indicating better performance on direction-pointing (video) versus target placement; this effect was not present in the later interval. Across time intervals, a significant time effect was revealed with post hoc tests indicating a decrease in performance over time for target placement. The dotted line indicates the mean performance of zoo-naive controls. Data presented as box plots, centre line, median; Tukey-style whiskers, 1.5× IQR; dots, individual data points. The 1–10-year interval, n = 44 (pointing (arrow), n = 31); the 10–30-year interval, n = 26 (pointing (arrow): n = 13).CC BY · PubMed Central

Verification

Claims the paper makes

14 Valid · 0 Not valid · 5 Not checked · 76 Own results and interpretations

56 more claims · 5 with their cited sources checked

Create a free account to see all 95 claims and every source verdict.

See all claims — free account
  • It may state two separate claims as one. 22
  • Probably not a substantial claim. 17

Create a free account to see these claims.

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Not confirmedz = 0.59reported p = 1 · recomputed p = .555

    “z = 0.590, p = 1.0”

  • Not confirmedz = 2.451reported p = .043 · recomputed p = .014

    “z = 2.451, p = .043”

  • Consistentχ²(2) = 13.814reported p = .001 · recomputed p = .001

    “χ2(2) = 13.814, p = .001”

  • Consistentz = -3.553reported p = .001 · recomputed p < .001

    “z = −3.553, p = .001”

    The paper prints p at a floor (such as p < .001)

  • Consistentz = -7.383reported p < .001 · recomputed p < .001

    “z = −7.383, p < .001”

5 of 31 · the rest aren’t loaded

26 more recomputed tests

Create a free account to see every test the Stats check recomputed.

See all tests — free account

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Self-check results appear here once an independent check has confirmed them.

Citations

Cites

Show all 33 references

Research works printed but not linked: 1

  1. Kleiner, M. et al. What’s new in Psychtoolbox-3. Perception 36, 1–16 (2007).

Metadata

Authors
Iggena, D., Schmelter, T., Maier, P. M., Reguieg, K., Finke, C., Hildebrand, K., & Ploner, C. J.
Affiliations
  • Humboldt-Universität zu Berlin (DE)
  • Berliner Hochschule für Technik (DE)
  • Charité – Universitätsmedizin Berlin (DE)
Corresponding author
  • Iggena, Deetje
  • Ploner, Christoph J.
Venue
Nature Human Behaviour
Published
2026-05-18
Volume
10
Issue
7
Pages
1327-1339
ISSN
2397-3374 (ISSN-L)
OpenAlex
W4409125915
Cited by
0
References
33
FWCI
0.00
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Human Behaviour (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (submittedVersion, cc-by-nd)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
Funders
  • Berlin Institute of Health at Charité
  • Deutsche Forschungsgemeinschaft (327654276 – SFB 1315)
  • Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie (16SV8171)
Topics
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Memory and Neural Mechanisms
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Memory Processes and Influences
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Spatial Cognition and Navigation
Keywords
memory reorganization, spatial memory, long-term memory, memory delay, environmental representation, systems consolidation, forgetting, memory stability
MeSH terms
Humans, Memory, Long-Term, Spatial Memory, Spatial Navigation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-09, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-09 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Long-term memory reorganization of navigational episodes. https://app.cochat.ai/papers/10.1038/s41562-026-02472-x
The paper
Iggena, D., Schmelter, T., Maier, P. M., Reguieg, K., Finke, C., Hildebrand, K., & Ploner, C. J. (2026). Long-term memory reorganization of navigational episodes. Nature Human Behaviour. https://doi.org/10.1038/s41562-026-02472-x

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.