Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Episodic sequence memory is supported by a theta–gamma phase code

Nature Neuroscience volume 19, pages 1374–1380 (2016)Cite this article

Subjects

Abstract

The meaning we derive from our experiences is not a simple static extraction of the elements but is largely based on the order in which those elements occur. Models propose that sequence encoding is supported by interactions between high- and low-frequency oscillations, such that elements within an experience are represented by neural cell assemblies firing at higher frequencies (gamma) and sequential order is encoded by the specific timing of firing with respect to a lower frequency oscillation (theta). During episodic sequence memory formation in humans, we provide evidence that items in different sequence positions exhibit greater gamma power along distinct phases of a theta oscillation. Furthermore, this segregation is related to successful temporal order memory. Our results provide compelling evidence that memory for order, a core component of an episodic memory, capitalizes on the ubiquitous physiological mechanism of theta–gamma phase–amplitude coupling.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Get just this article for as long as you need it

$39.95

Learn more

Prices may be subject to local taxes which are calculated during checkout

Figure 1: Theta–gamma model fitting analysis.
Figure 2: Phase analysis of theta–gamma coupling during sequence encoding, plotted by position and subsequent temporal order memory for left posterior cluster of sensors.
Figure 3: Relative biases in gamma power over theta phase by sequence position and subsequent memory.

References

  1. Bliss, T.V. & Lomo, T. Long-lasting potentiation of synaptic transmission in the dentate area of the anaesthetized rabbit following stimulation of the perforant path. J. Physiol. 232, 331–356 (1973).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  2. Nabavi, S. et al. Engineering a memory with LTD and LTP. Nature 511, 348–352 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Lisman, J.E. & Idiart, M.A. Storage of 7 +/− 2 short-term memories in oscillatory subcycles. Science 267, 1512–1515 (1995).

    Article  CAS  PubMed  Google Scholar 

  4. Jensen, O. & Lisman, J.E. Hippocampal CA3 region predicts memory sequences: accounting for the phase precession of place cells. Learn. Mem. 3, 279–287 (1996).

    Article  CAS  PubMed  Google Scholar 

  5. Lisman, J.E. & Jensen, O. The θ-γ neural code. Neuron 77, 1002–1016 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Jensen, O., Idiart, M.A. & Lisman, J.E. Physiologically realistic formation of autoassociative memory in networks with theta/gamma oscillations: role of fast NMDA channels. Learn. Mem. 3, 243–256 (1996).

    Article  CAS  PubMed  Google Scholar 

  7. Koene, R.A. & Hasselmo, M.E. First-in–first-out item replacement in a model of short-term memory based on persistent spiking. Cereb. Cortex 17, 1766–1781 (2007).

    Article  PubMed  Google Scholar 

  8. Buzsáki, G. & Draguhn, A. Neuronal oscillations in cortical networks. Science 304, 1926–1929 (2004).

    Article  CAS  PubMed  Google Scholar 

  9. Axmacher, N. et al. Cross-frequency coupling supports multi-item working memory in the human hippocampus. Proc. Natl. Acad. Sci. USA 107, 3228–3233 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Tort, A.B.L., Komorowski, R.W., Manns, J.R., Kopell, N.J. & Eichenbaum, H. Theta-gamma coupling increases during the learning of item-context associations. Proc. Natl. Acad. Sci. USA 106, 20942–20947 (2009).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  11. Friese, U. et al. Successful memory encoding is associated with increased cross-frequency coupling between frontal theta and posterior gamma oscillations in human scalp-recorded EEG. Neuroimage 66, 642–647 (2013).

    Article  PubMed  Google Scholar 

  12. Fuentemilla, L., Penny, W.D., Cashdollar, N., Bunzeck, N. & Düzel, E. Theta-coupled periodic replay in working memory. Curr. Biol. 20, 606–612 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Lega, B., Burke, J., Jacobs, J. & Kahana, M.J. Slow-theta-to-gamma phase-amplitude coupling in human hippocampus supports the formation of new episodic memories. Cereb. Cortex 26, 268–278 (2014).

    Article  PubMed  PubMed Central  Google Scholar 

  14. Aru, J. et al. Untangling cross-frequency coupling in neuroscience. Curr. Opin. Neurobiol. 31, 51–61 (2015).

    Article  CAS  PubMed  Google Scholar 

  15. Jensen, O. & Lisman, J.E. Hippocampal sequence-encoding driven by a cortical multi-item working memory buffer. Trends Neurosci. 28, 67–72 (2005).

    Article  CAS  PubMed  Google Scholar 

  16. DuBrow, S. & Davachi, L. Temporal memory is shaped by encoding stability and intervening item reactivation. J. Neurosci. 34, 13998–14005 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  17. Hsieh, L.-T., Gruber, M.J., Jenkins, L.J. & Ranganath, C. Hippocampal activity patterns carry information about objects in temporal context. Neuron 81, 1165–1178 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Jenkins, L.J. & Ranganath, C. Prefrontal and medial temporal lobe activity at encoding predicts temporal context memory. J. Neurosci. 30, 15558–15565 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Tubridy, S. & Davachi, L. Medial temporal lobe contributions to episodic sequence encoding. Cereb. Cortex 21, 272–280 (2011).

    Article  PubMed  Google Scholar 

  20. Davachi, L. & DuBrow, S. How the hippocampus preserves order: the role of prediction and context. Trends Cogn. Sci. 19, 92–99 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  21. Ezzyat, Y. & Davachi, L. Similarity breeds proximity: pattern similarity within and across contexts is related to later mnemonic judgments of temporal proximity. Neuron 81, 1179–1189 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. 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).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Kesner, R.P., Gilbert, P.E. & Barua, L.A. The role of the hippocampus in memory for the temporal order of a sequence of odors. Behav. Neurosci. 116, 286–290 (2002).

    Article  PubMed  Google Scholar 

  24. Dalal, S.S. et al. Spatial localization of cortical time-frequency dynamics. in Conf. Proc. IEEE Eng. Med. Biol. Soc. 2007, 4941–4944 (2007).

    Google Scholar 

  25. Attal, Y. & Schwartz, D. Assessment of subcortical source localization using deep brain activity imaging model with minimum norm operators: a MEG study. PLoS One 8, e59856 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Dalal, S. et al. Simultaneous MEG-intracranial EEG: new insights into the ability of MEG to capture oscillatory modulations in the neocortex and the hippocampus. Epilepsy Behav. 28, 288–292 (2013).

    Article  Google Scholar 

  27. Staudigl, T. & Hanslmayr, S. Theta oscillations at encoding mediate the context-dependent nature of human episodic memory. Curr. Biol. 23, 1101–1106 (2013).

    Article  CAS  PubMed  Google Scholar 

  28. Mills, T., Lalancette, M., Moses, S.N., Taylor, M.J. & Quraan, M.A. Techniques for detection and localization of weak hippocampal and medial frontal sources using beamformers in MEG. Brain Topogr. 25, 248–263 (2012).

    Article  PubMed  Google Scholar 

  29. Quraan, M.A., Moses, S.N., Hung, Y., Mills, T. & Taylor, M.J. Detection and localization of hippocampal activity using beamformers with MEG: a detailed investigation using simulations and empirical data. Hum. Brain Mapp. 32, 812–827 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  30. Ranganath, C. & Hsieh, L.-T. The hippocampus: a special place for time. Ann. NY Acad. Sci. 1369, 93–110 (2016).

    Article  PubMed  Google Scholar 

  31. Harris, K.D. Neural signatures of cell assembly organization. Nat. Rev. Neurosci. 6, 399–407 (2005).

    Article  CAS  PubMed  Google Scholar 

  32. Huxter, J., Burgess, N. & O'Keefe, J. Independent rate and temporal coding in hippocampal pyramidal cells. Nature 425, 828–832 (2003).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Tsodyks, M.V., Skaggs, W.E., Sejnowski, T.J. & McNaughton, B.L. Population dynamics and theta rhythm phase precession of hippocampal place cell firing: a spiking neuron model. Hippocampus 6, 271–280 (1996).

    Article  CAS  PubMed  Google Scholar 

  34. Dalal, S. et al. Simultaneous MEG-intracranial EEG: new insights into the ability of MEG to capture oscillatory modulations in the neocortex and the hippocampus. Epilepsy and Behavior 28, 288–292 (2013).

    Article  Google Scholar 

  35. Zheng, C., Bieri, K.W., Hsiao, Y.-T. & Colgin, L.L. Spatial sequence coding differs during slow and fast gamma rhythms in the hippocampus. Neuron 89, 398–408 (2016).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Summerfield, C. & Mangels, J.A. Coherent theta-band EEG activity predicts item-context binding during encoding. Neuroimage 24, 692–703 (2005).

    Article  PubMed  Google Scholar 

  37. Sederberg, P.B., Kahana, M.J., Howard, M.W., Donner, E.J. & Madsen, J.R. Theta and gamma oscillations during encoding predict subsequent recall. J. Neurosci. 23, 10809–10814 (2003).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Hsieh, L.-T., Ekstrom, A.D. & Ranganath, C. Neural oscillations associated with item and temporal order maintenance in working memory. J. Neurosci. 31, 10803–10810 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Gevins, A., Smith, M.E., McEvoy, L. & Yu, D. High-resolution EEG mapping of cortical activation related to working memory: effects of task difficulty, type of processing, and practice. Cereb. Cortex 7, 374–385 (1997).

    Article  CAS  PubMed  Google Scholar 

  40. Scheeringa, R. et al. Trial-by-trial coupling between EEG and BOLD identifies networks related to alpha and theta EEG power increases during working memory maintenance. Neuroimage 44, 1224–1238 (2009).

    Article  PubMed  Google Scholar 

  41. Raghavachari, S. et al. Gating of human theta oscillations by a working memory task. J. Neurosci. 21, 3175–3183 (2001).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Jensen, O. & Tesche, C.D. Frontal theta activity in humans increases with memory load in a working memory task. Eur. J. Neurosci. 15, 1395–1399 (2002).

    Article  PubMed  Google Scholar 

  43. Blumenfeld, R.S., Parks, C.M., Yonelinas, A.P. & Ranganath, C. Putting the pieces together: the role of dorsolateral prefrontal cortex in relational memory encoding. J. Cogn. Neurosci. 23, 257–265 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  44. Canolty, R.T. et al. High gamma power is phase-locked to theta oscillations in human neocortex. Science 313, 1626–1628 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. O'Keefe, J. A review of the hippocampal place cells. Prog. Neurobiol. 13, 419–439 (1979).

    Article  CAS  PubMed  Google Scholar 

  46. Foster, D.J. & Wilson, M.A. Hippocampal theta sequences. Hippocampus 17, 1093–1099 (2007).

    Article  PubMed  Google Scholar 

  47. Skaggs, W.E. & McNaughton, B.L. Replay of neuronal firing sequences in rat hippocampus during sleep following spatial experience. Science 271, 1870–1873 (1996).

    Article  CAS  PubMed  Google Scholar 

  48. Dragoi, G. & Buzsáki, G. Temporal encoding of place sequences by hippocampal cell assemblies. Neuron 50, 145–157 (2006).

    Article  CAS  PubMed  Google Scholar 

  49. DuBrow, S. & Davachi, L. The influence of context boundaries on memory for the sequential order of events. J. Exp. Psychol. Gen. 142, 1277–1286 (2013).

    Article  PubMed  PubMed Central  Google Scholar 

  50. de Cheveigné, A. & Simon, J.Z. Denoising based on time-shift PCA. J. Neurosci. Methods 165, 297–305 (2007).

    Article  PubMed  PubMed Central  Google Scholar 

  51. Oostenveld, R., Fries, P., Maris, E. & Schoffelen, J.-M. FieldTrip. Open source software for advanced analysis of MEG, EEG, and invasive electrophysiological data. Comp. Intel. and Neurosci. 2011, e156869 (2010).

    Google Scholar 

  52. Tort, A.B.L., Komorowski, R., Eichenbaum, H. & Kopell, N. Measuring phase-amplitude coupling between neuronal oscillations of different frequencies. J. Neurophysiol. 104, 1195–1210 (2010).

    Article  PubMed  PubMed Central  Google Scholar 

  53. Van Veen, B.D., van Drongelen, W., Yuchtman, M. & Suzuki, A. Localization of brain electrical activity via linearly constrained minimum variance spatial filtering. IEEE Trans. Biomed. Eng. 44, 867–880 (1997).

    Article  CAS  PubMed  Google Scholar 

  54. Jenkinson, M., Beckmann, C.F., Behrens, T.E.J., Woolrich, M.W. & Smith, S.M. FSL. Neuroimage 62, 782–790 (2012).

    Article  PubMed  Google Scholar 

  55. Nolte, G. The magnetic lead field theorem in the quasi-static approximation and its use for magnetoencephalography forward calculation in realistic volume conductors. Phys. Med. Biol. 48, 3637–3652 (2003).

    Article  PubMed  Google Scholar 

  56. Berens, P. CircStat: A MATLAB toolbox for circular statistics. J. Stat. Software 31 2009).

  57. Roach, B.J. & Mathalon, D.H. Event-related EEG time-frequency analysis: an overview of measures and an analysis of early gamma band phase locking in schizophrenia. Schizophr. Bull. 34, 907–926 (2008).

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

We thank L. Lohnas and G. Cogan for critical discussions and readings of the manuscript; J. Walker for technical support during MEG recording; S. Haegens for advice and discussion regarding the source localization; and A. Flinker and K. Doelling for critical discussions around methodological considerations and statistical approaches. We'd also like to thank our financial supporters, NIMH grant RO1–MH074692 to L.D.

Author information

Authors and Affiliations

  1. Department of Psychology, New York University, New York, New York, USA

    Andrew C Heusser, David Poeppel, Youssef Ezzyat & Lila Davachi

  2. Center for Neural Science, New York University, New York, New York, USA

    David Poeppel & Lila Davachi

Authors
  1. Andrew C Heusser
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. David Poeppel
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Youssef Ezzyat
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Lila Davachi
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

A.C.H., D.P., Y.E. and L.D. designed the experiment. A.C.H. collected the data. A.C.H. analyzed the data. A.C.H., D.P., Y.E. and L.D. wrote the paper.

Corresponding authors

Correspondence to Andrew C Heusser or Lila Davachi.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Time–frequency power spectrogram during stimulus presentation

A) Average time-frequency spectrogram is plotted across all sensors and all subjects, where 0 represents stimulus onset. Power is calculated relative to a pre-stimulus baseline (-1 to -.5). The line plots on the left side are group-level t-values representing the statistical contrast of the stimulus on period versus the pre-stimulus baseline. The black line represents 0-500ms and the blue line represents 0-2500ms (i.e. the duration of the stimulus on period). B) Time-frequency power is plotted specifically for the two clusters of sensors that showed the pattern of decreasing theta-gamma coupling by sequence position. C) Topographic plots in time bins of 500ms during stimulus presentation. The top row is theta power and the bottom row is gamma power.

Supplementary Figure 2 Theta–gamma coupling topography

Left. Topographic plot of group-level t-statistics representing sensors that showed significant theta-gamma coupling during the sequence encoding task (0-2.5 seconds; thresholded at t(16)>3.96, p<.001). Filled circles on the topographic plot represent significant sensors. Right. Gamma power binned by theta phase for sensors that showed significant theta-gamma coupling. Error bars represent standard error of the group mean. The dotted lines represent the standard error of the mean of the permuted coupling scores (see Methods for details).

Supplementary Figure 3 Power and coupling averaged across significant clusters of sensors

A) Bar graph representing magnitude of theta-gamma coupling (MI or ‘modulation index’) by sequence position averaged across sensors that displayed a significant fit to theta-gamma model. B) Theta power by sequence position averaged across sensors displaying significant fit to theta-gamma model. C) Gamma power by sequence position averaged across sensors displaying significant fit to theta-gamma model. Error bars represent standard error of the mean.

Supplementary Figure 4 Unthresholded theta–gamma model fit data and power control analysis

The left topographic plot is a statistical plot representing the fit of the theta-gamma model prediction to the theta-gamma coupling data prior to applying the cluster threshold (i.e. Figure 1d before cluster correction, See Methods for details of cluster size permutation procedure). The right topographic plot represents the result of an analysis where power effects in the theta and gamma band were first regressed out of the theta-gamma coupling data and then the residuals of this analysis were fit to the predicted pattern from the theta-gamma model.

Supplementary Figure 5 Source localization analysis with various statistical thresholds

The statistical maps represent the group-level fit of the decreasing PAC by sequence position model to the theta-gamma coupling data at various thresholds (p<.001,.01.1, all uncorrected). Coronal slices are on the top row, axial slices in the middle row, and sagittal slices are along the bottom row.

Supplementary Figure 6 Group-averaged gamma power binned by theta phase for significant clusters

On the left, data was extracted from the left posterior cluster of sensors that significantly fit the model of decreasing phase amplitude coupling. On the top is the group-average across all trials (i.e. irrespective of sequence position). In the middle, theta-gamma coupling is plotted as a function of sequence position only for trials where the order was later correct and on the bottom, only for trials where the order was later incorrect. The plots on the right are the same as the left, but for the left lateral cluster of interest.

Supplementary Figure 7 Gamma power (70–100 Hz) during stimulus presentation

A) Group-averaged time course of gamma power (-.5 to 2.5 seconds) in left posterior cluster of interest for early (dark gray; sequence positions 1&2), middle (blue; 3&4) and later (yellow; 5&6) sequence positions. B) Same as A but for only correctly remembered sequences. C) Same as A but only for incorrect sequences. Error bars represent standard error of the mean.

Supplementary Figure 8 Theta phase locking (3–8 Hz) during stimulus presentation

A) Group-averaged time course of theta phase locking (-.5 to 2.5 seconds) in left posterior cluster of interest for early (dark gray; sequence positions 1&2), middle (blue; 3&4) and later (yellow; 5&6) sequence positions. B) Same as A but for only correctly remembered sequences. C) Same as A but only for incorrect sequences. Error bars represent standard error of the mean.

Supplementary Figure 9 Average theta–gamma model fit statistic with and without first 500 ms

The bar on the left represents the average model fit for the original analysis on the entire stimulus presentation (0-2500ms), averaging over sensors that showed a significant group-level model fit. The bar on the right represents the average model fit for the new analysis where we remove the first 500ms (500-2500ms), eliminating the possible contribution of the evoked response. Error bars represent standard error of the mean. *** p<.001.

Supplementary Figure 10 Theta phase coding after removing first 500 ms

A) Distribution of gamma power over theta phase by sequence position bin for only correctly remembered sequences (Watson William’s Test F(5,96)=9.10, p=1.05e-5). B) Distribution of gamma power over theta phase by sequence position bin for incorrectly remembered sequences (Watson William’s Test F(5,96)=6.42, p=2.5e-2). Sequence by position interaction is significant (Harrison-Kanji Test: F(5,196)=11.03, p=.025). C) Group-level statistics for theta phase coding effect with (solid line) and without first 500ms (dashed line). Error bars represent standard error of the mean. *p<.05

Supplementary Figure 11 Theta-gamma coupling during intertrial interval

(A) Histogram of gamma power over theta phase for the stimulus presentation interval (dark blue) and the ITI (light blue). (B) Decreasing PAC by sequence position model fits during the entire stimulus interval, after removing the first 500ms and during the ITI. (C) Distribution of gamma power over theta by sequence position for correct order sequences during the ITI (Watson William’s Test: F(5,96)=9.10, p=1.07e-5). (D) Distribution of gamma power over theta by sequence position for incorrect sequences during the ITI (Watson-Williams Test: F(5,96)=9.09, p=4.03e-4). Sequence by position interaction trending (Harrison-Kanji Test: F(5,196)=8.02, p=.07). Error bars represent standard error of the mean. ~ p<.1, * p<.05, ** p<.01, *** p<.001.

Supplementary Figure 12 Testing for systematic theta phase shift by sequence position

The top row is simulated data in the theta frequency for each sequence position, where we simulated a systematic linear phase shift over the sequence. The bottom row is the group-averaged data after computing each of these metrics within-subject. The left column is the data filtered in the theta band. The right row is the lag of the peak in the pair-wise cross-correlation. If a systematic phase shift was present in the data, we would expect a monotonically increasing lag as a function of sequence position. Error bars represent standard error of the mean.

Supplementary Figure 13 Theta (3–8 Hz) phase symmetry by sequence position

(A) On the left is a simulated 4 Hz sine wave. Plotted in the middle is the phase time series of the theta wave. On the right, the waveform is plotted in polar coordinates, where the circular angle represents the phase and the distance from the center represents the power. (B) The same plot as listed above, but for a square wave at 4 Hz. (C) Polar plot representing the grand average of the MEG time series during stimulus presentation (0 to 2.5s) filtered in the theta range (3-8 Hz) and averaged over the left posterior cluster of sensors. Each color represents a distinct sequence position.

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Heusser, A., Poeppel, D., Ezzyat, Y. et al. Episodic sequence memory is supported by a theta–gamma phase code. Nat Neurosci 19, 1374–1380 (2016). https://doi.org/10.1038/nn.4374

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nn.4374

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing