Abstract
Episodic memory requires encoding the temporal structure of experience and relies on brain circuits in the medial temporal lobe, including the medial entorhinal cortex (MEC). Recent studies have identified MEC ‘time cells’, which fire at specific moments during interval timing tasks, collectively tiling the entire timing period. It has been hypothesized that MEC time cells could provide temporal information necessary for episodic memories, yet it remains unknown whether they display learning dynamics required for encoding different temporal contexts. To explore this, we developed a new behavioral paradigm requiring mice to distinguish temporal contexts. Combined with methods for cellular resolution calcium imaging, we found that MEC time cells display context-dependent neural activity that emerges with task learning. Through chemogenetic inactivation we found that MEC activity is necessary for learning of context-dependent interval timing behavior. Finally, we found evidence of a common circuit mechanism that could drive sequential activity of both time cells and spatially selective neurons in MEC. Our work suggests that the clock-like firing of MEC time cells can be modulated by learning, allowing the tracking of various temporal structures that emerge through experience.
This is a preview of subscription content, access via your institution
Access options
Access Nature and 54 other Nature Portfolio journals
Get Nature+, our best-value online-access subscription
$32.99 / 30 days
cancel any time
Subscribe to this journal
Receive 12 print issues and online access
$259.00 per year
only $21.58 per issue
Buy this article
- Purchase on SpringerLink
- Instant access to full article PDF
Prices may be subject to local taxes which are calculated during checkout
Similar content being viewed by others
Data availability
Data presented in the present study are available upon reasonable request to the corresponding author and will be made publicly available at https://github.com/heyslab/Bigus_Lee_NatNeuro_2024 1.5 years after publication.
Code availability
Code is available at https://github.com/heyslab/Bigus_Lee_NatNeuro_2024 or upon reasonable request to the corresponding author.
References
Tulving, E. Précis of elements of episodic memory. Behav. Brain Sci. 7, 223–238 (1984).
Tulving, E. Organization of Memory (Academic Press, 1972).
Kimble, D. P. The effects of bilateral hippocampal lesions in rats. J. Comp. Physiol. Psychol. 56, 273–283 (1963).
Niki, H. Response perseveration following the hippocampal ablation in the rat. Jpn Psychol. Res. 8, 1–9 (1966).
O’Keefe, J. & Conway, D. H. Hippocampal place units in the freely moving rat: why they fire where they fire. Exp. Brain Res. 31, 573–590 (1978).
Olton, D. S., Walker, J. A. & Gage, F. H. Hippocampal connections and spatial discrimination. Brain Res. 139, 295–308 (1978).
O’Keefe, J. & Dostrovsky, J. The hippocampus as a spatial map. Preliminary evidence from unit activity in the freely-moving rat. Brain Res. 34, 171–175 (1971).
Hafting, T., Fyhn, M., Molden, S., Moser, M.-B. & Moser, E. I. Microstructure of a spatial map in the entorhinal cortex. Nature 436, 801–806 (2005).
Muller, R. U. & Kubie, J. L. The effects of changes in the environment on the spatial firing of hippocampal complex-spike cells. J. Neurosci. 7, 1951–1968 (1987).
Fyhn, M., Hafting, T., Treves, A., Moser, M.-B. & Moser, E. I. Hippocampal remapping and grid realignment in entorhinal cortex. Nature 446, 190–194 (2007).
O’Keefe, J. & Nadel, L. The Hippocampus as a Cognitive Map (Oxford University Press, 1978).
Buzsáki, G. & Moser, E. I. Memory, navigation and theta rhythm in the hippocampal-entorhinal system. Nat. Neurosci. 16, 130–138 (2013).
Kacelnik, A., Brunner, D. & Gibbon, J. Behavioural Mechanisms of Food Selection (ed. Hughes, R. N.) 61–82 (Springer, 1990).
Marshall, A. T. & Kirkpatrick, K. Reinforcement learning models of risky choice and the promotion of risk-taking by losses disguised as wins in rats. J. Exp. Psychol. Anim. Learn. Cogn. 43, 262–279 (2017).
Brunner, D., Fairhurst, S., Stolovitzky, G. & Gibbon, J. Mnemonics for variability: remembering food delay. J. Exp. Psychol. Anim. Behav. Process. 23, 68–83 (1997).
Issa, J. B., Tocker, G., Hasselmo, M. E., Heys, J. G. & Dombeck, D. A. Navigating through time: a spatial navigation perspective on how the brain may encode time. Annu. Rev. Neurosci. 43, 73–93 (2020).
Heys, J. G., Wu, Z., Allegra Mascaro, A. L. & Dombeck, D. A. Inactivation of the medial entorhinal cortex selectively disrupts learning of interval timing. Cell Rep. 32, 108163 (2020).
Vo, A. et al. Medial entorhinal cortex lesions produce delay-dependent disruptions in memory for elapsed time. Neurobiol. Learn. Mem. 185, 107507 (2021).
Dias, M., Ferreira, R. & Remondes, M. Medial entorhinal cortex excitatory neurons are necessary for accurate timing. J. Neurosci. 41, 9932–9943 (2021).
Heys, J. G. & Dombeck, D. A. Evidence for a subcircuit in medial entorhinal cortex representing elapsed time during immobility. Nat. Neurosci. 21, 1574–1582 (2018).
Morris, R. G., Garrud, P., Rawlins, J. N. & O’Keefe, J. Place navigation impaired in rats with hippocampal lesions. Nature 297, 681–683 (1982).
Mishkin, M. Memory in monkeys severely impaired by combined but not by separate removal of amygdala and hippocampus. Nature 273, 297–298 (1978).
Squire, L. R. & Zola-Morgan, S. The medial temporal lobe memory system. Science 253, 1380–1386 (1991).
Otto, T. & Eichenbaum, H. Neuronal activity in the hippocampus during delayed non-match to sample performance in rats: evidence for hippocampal processing in recognition memory. Hippocampus 2, 323–334 (1992).
Verhagen, J. V., Wesson, D. W., Netoff, T. I., White, J. A. & Wachowiak, M. Sniffing controls an adaptive filter of sensory input to the olfactory bulb. Nat. Neurosci. 10, 631–639 (2007).
Heys, J. G., Rangarajan, K. V. & Dombeck, D. A. The functional micro-organization of grid cells revealed by cellular-resolution imaging. Neuron 84, 1079–1090 (2014).
Kraus, B. J. et al. During running in place, grid cells integrate elapsed time and distance run. Neuron 88, 578–589 (2015).
Campbell, M. G., Attinger, A., Ocko, S. A., Ganguli, S. & Giocomo, L. M. Distance-tuned neurons drive specialized path integration calculations in medial entorhinal cortex. Cell Rep. 36, 109669 (2021).
Turi, G. F. et al. Vasoactive intestinal polypeptide-expressing interneurons in the hippocampus support goal-oriented spatial learning. Neuron 101, 1150–1165.e8 (2019).
Hardcastle, K., Maheswaranathan, N., Ganguli, S. & Giocomo, L. M. A multiplexed, heterogeneous, and adaptive code for navigation in medial entorhinal cortex. Neuron 94, 375–387.e7 (2017).
McDonald, R. J. & White, N. M. A triple dissociation of memory systems: hippocampus, amygdala, and dorsal striatum. Behav. Neurosci. 107, 3–22 (1993).
Toda, K. et al. Nigrotectal stimulation stops interval timing in mice. Curr. Biol. 27, 3763–3770.e3 (2017).
Burak, Y. & Fiete, I. R. Accurate path integration in continuous attractor network models of grid cells. PLoS Comput. Biol. 5, e1000291 (2009).
Yoon, K. et al. Specific evidence of low-dimensional continuous attractor dynamics in grid cells. Nat. Neurosci. 16, 1077–1084 (2013).
Stensola, H. et al. The entorhinal grid map is discretized. Nature 492, 72–78 (2012).
Gardner, R. J. et al. Toroidal topology of population activity in grid cells. Nature 602, 123–128 (2022).
Matell, M. S., Meck, W. H. & Nicolelis, M. A. L. Interval timing and the encoding of signal duration by ensembles of cortical and striatal neurons. Behav. Neurosci. 117, 760–773 (2003).
Hinton, S. C. & Meck, W. H. Frontal-striatal circuitry activated by human peak-interval timing in the supra-seconds range. Brain Res. Cogn. Brain Res. 21, 171–182 (2004).
Meck, W. H. Neuroanatomical localization of an internal clock: a functional link between mesolimbic, nigrostriatal, and mesocortical dopaminergic systems. Brain Res. 1109, 93–107 (2006).
Jin, D. Z., Fujii, N. & Graybiel, A. M. Neural representation of time in cortico-basal ganglia circuits. Proc. Natl Acad. Sci. USA 106, 19156–19161 (2009).
Jazayeri, M. & Shadlen, M. N. A neural mechanism for sensing and reproducing a time interval. Curr. Biol. 25, 2599–2609 (2015).
Mello, G. B. M., Soares, S. & Paton, J. J. A scalable population code for time in the striatum. Curr. Biol. 25, 1113–1122 (2015).
Bakhurin, K. I. et al. Differential encoding of time by prefrontal and striatal network dynamics. J. Neurosci. 37, 854–870 (2017).
Meck, W. H., Church, R. M. & Olton, D. S. Hippocampus, time, and memory. Behav. Neurosci. 98, 3–22 (1984).
Jacobs, N. S., Allen, T. A., Nguyen, N. & Fortin, N. J. Critical role of the hippocampus in memory for elapsed time. J. Neurosci. 33, 13888–13893 (2013).
Fortin, N. J., Agster, K. L. & Eichenbaum, H. B. Critical role of the hippocampus in memory for sequences of events. Nat. Neurosci. 5, 458–462 (2002).
Allen, T. A., Salz, D. M., McKenzie, S. & Fortin, N. J. Nonspatial sequence coding in CA1 neurons. J. Neurosci. 36, 1547–1563 (2016).
MacDonald, C. J., Lepage, K. Q., Eden, U. T. & Eichenbaum, H. Hippocampal ‘time cells’ bridge the gap in memory for discontiguous events. Neuron 71, 737–749 (2011).
Pastalkova, E., Itskov, V., Amarasingham, A. & Buzsáki, G. Internally generated cell assembly sequences in the rat hippocampus. Science 321, 1322–1327 (2008).
Manns, J. R., Howard, M. W. & Eichenbaum, H. Gradual changes in hippocampal activity support remembering the order of events. Neuron 56, 530–540 (2007).
Mankin, E. A. et al. Neuronal code for extended time in the hippocampus. Proc. Natl Acad. Sci. USA 109, 19462–19467 (2012).
Tsao, A. et al. Integrating time from experience in the lateral entorhinal cortex. Nature 561, 57–62 (2018).
Issa, J. B. & Zhang, K. Universal conditions for exact path integration in neural systems. Proc. Natl Acad. Sci. USA 109, 6716–6720 (2012).
Pachitariu, M. et al. Suite2p: beyond 10,000 neurons with standard two-photon microscopy. Preprint at bioRxiv https://doi.org/10.1101/061507 (2017).
Dombeck, D. A., Harvey, C. D., Tian, L., Looger, L. L. & Tank, D. W. Functional imaging of hippocampal place cells at cellular resolution during virtual navigation. Nat. Neurosci. 13, 1433–1440 (2010).
Skaggs, W., McNaughton, B. & Gothard, K. Advances in Neural Information Processing Systems 5 (Morgan-Kaufmann, 1992).
Climer, J. R. & Dombeck, D. A. Information theoretic approaches to deciphering the neural code with functional fluorescence imaging. eNeuro 8, ENEURO.0266-21.2021 (2021).
Park, E. H., Keeley, S., Savin, C., Ranck, J. B. & Fenton, A. A. How the internally organized direction sense is used to navigate. Neuron 101, 285–293.e5 (2019).
Acknowledgements
We thank D. Dombeck, M. Long and M. Sheffield for valuable comments on earlier versions of this manuscript. We thank M. Wachowiak for generous support in designing and validating the olfactometers used in the present study. The present study was supported by the US National Institutes of Health (NIH) National Science Foundation (NSF), Whitehall Foundation (J.G.H.), Brain and Behavior Research Foundation (J.G.H.), NIH/National Institute of Mental Health (grant no. 1 DP2 MH129958-01 to J.G.H.), NSF CAREER Award (no. IOS-2145814 to J.G.H.), the Basic Science Research Program through the National Research Foundation of Korea (N.R.F.) funded by the Ministry of Education (grant no. RS-2023-00242639 to H.W.L.) and the University of Utah (J.G.H.). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Author information
Authors and Affiliations
Contributions
E.R.B, H.W.L. and J.G.H. designed experiments and wrote the manuscript. E.R.B, H.W.L. and J.S. collected the data. E.R.B, H.W.L., J.S., J.C.B. and J.G.H. analyzed and interpreted the data. E.R.B. and J.G.H. built equipment used to collect the data.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Neuroscience thanks Mehrdad Jazayeri and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data
Extended Data Fig. 1 tDNMS set-up, controls, and additional behavioral analysis of mice in Fig. 1.
a. Experimental set-up. Odorized air is directed either to the mouse or to a vacuum. A lick spout, connected to a capacitance sensor, delivers water and is used to monitor mouse licking. b. Odor concentration control. Odor concentration was measured using a photoionization detector (PID). Odor can be delivered with high temporal specificity at a constant concentration over 45 minutes, as shown by PID measurements (green) relative to control signal (black). c. Trial structure. Each trial consists of two presentations of the same odor, each for either a 2 or 5 s duration, separated by an interstimulus interval (ISI). Trial start is signified by a visual cue, and trials are separated by a 16-24 s intertrial interval. d. Training protocol. Mice undergo three phases of pretraining (see Methods). e. Odor control session. After completing tDNMS training, mice were tested with no odorant (mineral oil only). Mice failed to solve nonmatch trials in the absence of odor but performed well in a prior session in which odor was used (p = 2.6 × 10−13, two-tailed paired t-test; n = 14 mice). Bars represent mean across mice ± s.e.m. f. Trial length control session. A cohort of mice was trained on a version of the tDNMS task with modified durations. The ISI was then manipulated on a random subset of nonmatch trials (“probe trials”) so that overall trial duration was identical to match trial duration. If mice use total trial duration to solve the task, rather than individual stimulus durations, they should incorrectly withhold licking on probe trials. Instead, there was no significant difference between standard nonmatch and probe trial performance (p = 0.19, two-tailed paired t-test; n = 7 mice). Bars indicate mean across mice ± s.e.m. g. Average performance by trial type during shaping (phase 3) of mice in Fig. 1 (n = 26). Shaping consists of probe trials where mice must correctly trigger reward and automatic trials where reward is automatically delivered. Performance was examined on all probe trials within the first 1/2 session of shaping phase 3, termed “early shaping”, and the last 1/2 session of shaping phase 3, or “late shaping”, for each mouse. Dots represent performance of each mouse, and bars show mean ± s.e.m. across mice. Mice performed better on short-long trials than long-short both early (p = 4.3 × 10−7, two-tailed paired t-test) and late (p = 0.041, two-tailed paired t-test) in shaping. Additionally, performance was higher for both short-long (p = 0.009, two-tailed paired t-test) and long-short (p = 4.7 × 10−10, two-tailed paired t-test) trials in late compared to early shaping. h. Reason for mistakes on long-short trials for mice in Fig. 1 (n = 26). During shaping phase 3 and the tDNMS task, mice can miss nonmatch trials either by withholding licking or by licking prematurely during the first odor and/or interstimulus interval. The percent of incorrect long-short trials in shaping phase 3 and the tDNMS task missed due to licking early is shown. Dots indicate values for each mouse, with red lines showing the median value across mice. i. Average time of first incorrect lick on long-short trials relative to first odor onset for mice in Fig. 1 (n = 26). Black circles represent the average time of first lick across all incorrect long-short trials for a given mouse, and red dots show the median value across mice. j. Average time of first incorrect lick on short-short trials in the tDNMS task relative to first odor onset for mice in Fig. 1 (n = 26). Black circles represent the average time of first lick across all incorrect short-short trials for a given mouse, and red dots show the median value across mice.
Extended Data Fig. 2 Histological verification of in vivo imaging in the MEC, and additional examples of MEC time cells.
a. Sagittal sections of post-mortem histology from all six mice utilized in the in vivo calcium imaging experiments. MEC neurons labelled with GCaMP6s (green). The sections are stained with NeuroTrace 435/455, displaying neuronal morphology in blue. The approximate locations of the two-photon imaging fields of view (FOV) for each mouse are labeled with red Alexa594. This labeling was achieved by inserting a pin coated with Alexa594 at sites corresponding to prominent vascular landmarks visible both in the in vivo two-photon imaging and under a dissecting scope during the ex vivo marking procedure. Confirmation of the imaging sites within the MEC was based on the presence of the lamina dissecans, the relative position of the post-rhinal border to the pin mark, and the characteristic circular shape of the dentate gyrus as observed in the medial-lateral sagittal sections. n = 6 mice. b. For each time cell, mean dF/F displayed for each trial type (top) and dF/F activity on each trial, sorted by trial type (below).
Extended Data Fig. 3 Additional analysis on time cell tuning.
a. Top, sequence of MEC time cells significantly tuned for S-S trials, sorted by S-S trials and displayed for S-S, S-L and L-S trials. Middle, same as above, expect for significant MEC time cells on S-L trials and sorted by S-L trials. Bottom, same as top, except for significant MEC time cells on L-S trials and sorted by L-S trials. b. A generalized linear model was used to assess whether neurons are tuned to one of three variables- time in the trial, distance travelled from trial start, or licking – or a combination of 2 or 3 variables (see Methods). Analysis performed on n = 695 time cells, collected from 10 behavioral sessions, lead to n = 177 significant models. Boxplot showing log-likelihood increase gained by each variable: time (median = 37.15), distance (median = 14.63) and licking (median = 0.82) (two-sided Kruskal-Wallis test, p = 2.6×10-26; followed by two-sided Wilcoxon rank-sum test with Bonferroni-correction: Time vs. Distance: p = 1.7×10-08; Time vs. Licking: p = 3.7×10-22; Distance vs. Licking: p = 8.8×10-13). Log-likelihood was normalized to recording time in minutes. c. Histogram demonstrating the model that best described the calcium activity of each cell and trial type. d. Boxplot showing adjusted variance explained for models that best describe the calcium activity of each cell for the single variable models: time (median = 0.0512), distance (median = 0.0202) and licking (median = 0.0101). Number of models n = 125,396. (One-sided Wilcoxon signed rank test (median greater than 0), p = 0.0001, p = 0, p = 0.0312). For box plots, the line inside of each box is the sample median. The upper quartile corresponds to the 0.75 quantile and the lower quartile corresponds to the 0.25 quantile. The blue dots in b and d represent outliers. Outliers are values that are more than 1.5 × interquartile range (IQR) away from the top or bottom of the box. The whiskers are lines that extend above and below each box. One whisker connects the upper quartile to the nonoutlier maximum (the maximum data value that is not an outlier), and the other connects the lower quartile to the nonoutlier minimum. e. Proportion of MEC time cells that either remained stable, displayed a time shift, displayed rate remapping, or displayed on/off dynamics across trial types. f. Rank order analysis for shuffle distribution (black) and real data (red). The similarity of the sequences of time cells across trial types is examined by comparing their rank orders. Each time cell is assigned three rank orders, corresponding to its sorting by peak timing for each trial type. Subsequently, the mean difference between rank orders within a cell is compared to a shuffle distribution, generated by shuffling rank order of cells 10,000 times. The p-value is computed as the proportion of shuffle values smaller than the actual data. Notably, p-values are zero for all three comparisons. g. Discriminant Index indicates the extent to which the difference in dF/F between trial types deviates from chance level. S-S vs S-L: Day 1 n = 224, Day N n = 221, z = 2.55, p = 0.01; S-S vs L-S: Day 1 n = 225, Day N n = 231, z = 3.80, p = 1.4×10-4, two-sided Wilcoxon rank sum test. Individual data points with median.
Extended Data Fig. 4 Trial type decoding analysis for each mouse, and comparison of time cell activity on correct and error trials.
a. For each panel: Left, LDA plots for each mouse on each trial type (S-S – orange, S-L – Blue, L-S – Magenta). K-means clustering is then applied to the LDA plots to categorize the dots into three clusters. The background colors indicate the clustering result. The accuracy of clustering analysis is determined by the proportion of dots correctly classified into their respective trial type. Right, clustering accuracy is compared between the bootstrapped shuffle distribution with randomly assigned trial labels (grey) and the actual data (red). The p-value is computed as the proportion of shuffle values larger than the actual data (one-sided). For mouse 1 through 6: p = 0, 0.04, 0.08, 0, 0, 0.001, respectively. No adjustment was made for multiple comparisons. b. Representative examples of MEC time cells on day 1 (left) and day N (right), depicting activity for both correct and error trials across all three types of trials. c. Cumulative distribution functions of correlation coefficients calculated for each MEC time cell, comparing activity across different trial conditions on day 1 and day N. session type main effect: p = 2.3×10-5, F(1, 288) = 18.5; trial type main effect: p = 0.12, F(1.93, 555.15) = 2.1; interaction: p = 0.33, F(1.93, 555.15) = 1.1, two-way mixed ANOVA with trial type and session factors. d. Sequence of activity of MEC time cells recorded on day 1, arranged according to their activity during error trials on Short-Short (S-S) trials. This sequence is then applied to display cell activity for all three trial conditions during correct trials, maintaining the order from the error trials. e. Same as in d, but for recordings on day N.
Extended Data Fig. 5 MEC time cells on ISI probe trials.
a. Schematic for probe trials. b. Comparative analysis of mean velocity and licking behaviors under three different conditions: Short-Short, Short-Long, and Long-Short, during standard (black) and probe (red) trials. c. Four example MEC time cells during control (black) and probe (red) trials. d. Aggregated data showing the timing of peak responses across the MEC time cell population under Short-Short, Short-Long, and Long-Short conditions, compared between standard (black) and probe (red) trials. Short-Short trial types: n = 120, p = 9.0×10-5, z = 3.9; Short-Long trial types: n = 67, p = 7.4×10-6, z = 4.5; Long-Short trial types: n = 67, p = 3.2×10-5, z = 4.2, two-sided Wilcoxon signed-rank.
Extended Data Fig. 6 MEC DREADD inactivation.
a. Ability to inhibit MEC was confirmed using in-vivo 2-photon imaging combined with hM4D(Gi) inactivation. Histology showing co-expression of GCaMP6s and hM4D(Gi)-mCherry in MEC, one section from one mouse is shown. b. Activation of inhibitory DREADDs by 1 mg/kg I.P. injection of DCZ reduces average number of Ca2+ transients in MEC neurons by 80% at 30 minutes post injection compared to before DCZ injection. Top, example neuron before and after DCZ administration. Bottom, population response. In both the control (blue) and DCZ (red) conditions, GCaMP activity was monitored over 5-minute periods, and the change in activity was measured for each cell (n = 205 neurons in DCZ condition and n = 364 neurons in control condition, measured across two sessions in each condition; p < 0.01, two-tailed Kolmogorov-Smirnov test). c. Average time of first lick relative to first odor onset on session 1 of the tDNMS task for mice in Fig. 4. Dots show average time of first lick for each mouse, with bars showing mean ± s.e.m. across mice. There is no difference in average time of first lick for DREADD (n = 15) and Control (n = 16) mice in any trial type (Short-Short: p = 0.56, Short-Long: p = 0.14, Long-Short: p = 0.52, two-tailed unpaired t-tests). d. Average performance by trial type during shaping for mice in Fig. 4 (n = 31 mice). Shaping consists of probe trials where mice must correctly trigger reward and automatic trials where reward is automatically given. Performance was examined on all probe trials within the first 1/2 session of shaping phase 3, termed “early shaping”, and the last 1/2 session of shaping phase 3, or “late shaping”, for each mouse. Mice performed better on short-long trials than long-short both early (p = 2.9 × 10−8, two-tailed paired t-test) and late (8.5 × 10−4, two-tailed paired t-test) in shaping. Additionally, performance was higher for both short-long (p = 1.3 × 10−8, two-tailed paired t-test) and long-short (p = 3.1 × 10−15, two-tailed paired t-test) trials in late compared to early shaping. Dots represent performance of each mouse, with blue dots for Control mice (n = 16) and red for DREADD mice (n = 15). Bars show mean ± s.e.m. across all mice. e. Reason for mistakes on long-short trials for mice in Fig. 4. During shaping phase 3 and the tDNMS task, mice can miss nonmatch trials either by withholding licking or by licking prematurely during the first odor and/or interstimulus interval. The percent of incorrect long-short trials missed due to licking early is shown. Dots indicate values for each mouse, with DREADD mice shown in red (n = 15) and Control in blue (n = 16), and black lines show the median value across all mice. f. Average time of first incorrect lick on long-short trials relative to first odor onset for mice in Fig. 4. Blue (Control, n = 16) and red (DREADD, n = 15) circles represent the average time of first lick on all incorrect long-short trials for a given mouse, and black dots show the median value across all mice.
Extended Data Fig. 7 MEC is not required for all interval timing behavior.
a. Schematic for inhibiting MEC after learning in the tDNMS task. After experiments testing the role of MEC during learning (Fig. 4), a subset of mice (n = 7 DREADD, n = 10 Control) underwent extended training to determine whether MEC is necessary for ongoing task performance. b. Though MEC inhibition impaired learning in the tDNMS task (Sessions 1-8), DREADD mice learned the task in the absence of MEC inhibition (Sessions 9-14). Following learning, subsequent administration of DCZ to inactivate MEC did not affect performance in Sessions 15-16. Bars indicate mean performance ± s.e.m calculated across mice. c. Fixed interval task schematic. MEC DREADD (n = 9) and Control (n = 10) mice were trained on a fixed interval (FI) task (Toda et al. 2017). A droplet of water (4-6ul) was delivered every 10 s to head-fixed mice. Licking was measured; time-locked predictive licking indicates learning the timing of water delivery. The DREADD agonist DCZ (1 mg/kg) was delivered 5 min prior to each session. d. Licking behavior of DREADD (n = 9) and Control (n = 10) mice on sessions 1 and 5 of the FI task. Licking was normalized to the maximum lick frequency with each session for each mouse. All trials for all mice are shown; water delivery occurs at 0 s, indicated by a yellow line. Average lick response for each session is shown in white. e. Fixed interval learning. Predictive licking is defined as an increase in lick rate, measured over 5 seconds preceding the upcoming reward delivery. Both DREADD and Control mice learn the temporal structure of the task, as demonstrated though more frequent engagement in predictive licking from sessions 1-5. Data represent mean ± s.e.m. averaged across mice. f. Average time of peak predictive licking activity in FI task relative to upcoming water delivery. From session 1 to 5, peak predictive licking activity moves closer to reward delivery (0 s) for both Control (p = 2.2 × 10−4, two-tailed paired t-test) and DREADD mice (p = 0.0013 two-tailed paired t-test). Data points represent average time of peaking licking on Session 1 and 5 for each mouse; bars indicate mean ± s.e.m. across mice.
Extended Data Fig. 8 Histology showing expression of hM4D(Gi)-mCherry in MEC.
Injections were performed bilaterally; sections spanning one hemisphere are shown for each DREADD mouse from Fig. 4 (n = 15, mouse identity and hemisphere noted). Five sagittal sections are shown per mouse, ranging from lateral (left) to medial (right). The middle three sections include MEC.
Extended Data Fig. 9 Additional details on Poisson Regression analysis of behavior.
a. Example of modeling behavior during tDMNS task. Top. Baseline cue-based model including only odor offset, long cue, and second cue features. Bottom. Strategy-based model. Observed anticipatory licks are in black above each model’s predicted value. b. Comparison of cue-based and strategy-based models for an individual animal. Mean lick rate for each trial type in black, model predictions in color. c. Fitting a Gaussian Mixture Model with 2 components to the best fitting LLHi over the cue-based model reveals two populations: one cue-based where additional features do not improve model performance (n = 10 mice) and one strategy-based with improved model predictions (n = 20 mice). d. A strategy-based model fits better on hit/correct reject trials compared to miss/false alarm (based on Strategy 3) (n = 30 mice, two-factor ANOVA significant effect for trial type F(2,155) = 7.64, p = 6.88×10-4 and result F(1,155) = 35.6, p = 1.58×10-8, but not type x result F(5,155) = 2.26, p = 0.11, post-hoc tests: * - p < 0.05, ** - p < 0.01, *** - p < 0.001). e. Example cue and strategy-based models for an animal where both fit similarly. f. Relationship between average mouse performance on the task and how well models fit the data (linear regression, r = 0.369, p = 0.045). g. No significant relationship between average percent correct and evidence for animals using a strategy-based solution, based on best fitting strategy (linear regression, r = 0.025, p = 0.895). h. No significant relationship between ability to decode trial type from the neural data and the ability to fit a model to the behavior, based on best fitting strategy (strategy and cue-based models, ρ = 0.600, p = 0.208, two-sided Spearman’s rank test for correlation and t-statistic). Colors in panels f-h indicate imaged mice, gray indicates not imaged. Center line on box plots depicts the median, the first and third quartiles are indicated by extent of the box and whiskers indicate the outlier cutoff (1.5x inter-quantile range).
Extended Data Fig. 10 Activity correlation between time cells maintained across conditions.
a. Pairwise correlation between all time cells during correct trials and error trials in the tDNMS task shown for real (blue) and shuffled data (black). n = 15,727, z = 69.6, p = 0, two-sided Chi-squared test. b. Same as in a, but shown for correct trials on each trial type (S-S vs S-L, right; S-S vs L-S, middle; L-S vs S-L, right). S-S vs S-L: n = 15,727, z = 72.4, p = 0; S-S vs L-S: n = 15,727, z = 66.3, p = 0; L-S vs S-L: n = 15,727, z = 65.8, p = 0, two-sided Chi-squared test.
Supplementary information
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Bigus, E.R., Lee, HW., Bowler, J.C. et al. Medial entorhinal cortex mediates learning of context-dependent interval timing behavior. Nat Neurosci 27, 1587–1598 (2024). https://doi.org/10.1038/s41593-024-01683-7
Received:
Accepted:
Published:
Issue date:
DOI: https://doi.org/10.1038/s41593-024-01683-7
