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.

  • Brief Communication
  • Published:

An independent, landmark-dominated head-direction signal in dysgranular retrosplenial cortex

Abstract

We investigated how landmarks influence the brain's computation of head direction and found that in a bidirectionally symmetrical environment, some neurons in dysgranular retrosplenial cortex showed bidirectional firing patterns. This indicates dominance of neural activity by local environmental cues even when these conflicted with the global head direction signal. It suggests a mechanism for associating landmarks to or dissociating them from the head direction signal, according to their directional stability and/or utility.

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

Access options

Rent or buy this article

Prices vary by article type

from$1.95

to$39.95

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

Figure 1: Two types of directional encoding by RSC neurons.
Figure 2: BD cells are specific to dysgranular RSC.
Figure 3: WC-BD activity.

Similar content being viewed by others

References

  1. Taube, J.S. Annu. Rev. Neurosci. 30, 181–207 (2007).

    Article  CAS  Google Scholar 

  2. Taube, J.S. & Burton, H.L. J. Neurophysiol. 74, 1953–1971 (1995).

    Article  CAS  Google Scholar 

  3. Knierim, J.J., Kudrimoti, H.S. & McNaughton, B.L. J. Neurosci. 15, 1648–1659 (1995).

    Article  CAS  Google Scholar 

  4. Auger, S.D., Mullally, S.L. & Maguire, E.A. PLoS One 7, e43620 (2012).

    Article  CAS  Google Scholar 

  5. Chen, L.L., Lin, L.H., Green, E.J., Barnes, C.A. & McNaughton, B.L. Exp. Brain Res. 101, 8–23 (1994).

    Article  CAS  Google Scholar 

  6. Chen, L.L., Lin, L.H., Barnes, C.A. & McNaughton, B.L. Exp. Brain Res. 101, 24–34 (1994).

    Article  CAS  Google Scholar 

  7. Cho, J. & Sharp, P.E. Behav. Neurosci. 115, 3–25 (2001).

    Article  CAS  Google Scholar 

  8. Sugar, J., Witter, M.P., van Strien, N.M. & Cappaert, N.L.M. Front. Neuroinform. 5, 7 (2011).

    Article  Google Scholar 

  9. Taube, J.S., Muller, R.U. & Ranck, J.B. Jr. J. Neurosci. 10, 436–447 (1990).

    Article  CAS  Google Scholar 

  10. Yoganarasimha, D., Yu, X. & Knierim, J.J. J. Neurosci. 26, 622–631 (2006).

    Article  CAS  Google Scholar 

  11. Cooper, B.G., Manka, T.F. & Mizumori, S.J. Behav. Neurosci. 115, 1012–1028 (2001).

    Article  CAS  Google Scholar 

  12. Pothuizen, H.H.J., Davies, M., Albasser, M.M., Aggleton, J.P. & Vann, S.D. Eur. J. Neurosci. 30, 877–888 (2009).

    Article  Google Scholar 

  13. Marchette, S.A., Vass, L.K., Ryan, J. & Epstein, R.A. Nat. Neurosci. 17, 1598–1606 (2014).

    Article  CAS  Google Scholar 

  14. Robertson, C.E., Hermann, K.L., Mynick, A., Kravitz, D.J. & Kanwisher, N. Curr. Biol. 26, 2463–2468 (2016).

    Article  CAS  Google Scholar 

  15. Knight, R. et al. Phil. Trans. R. Soc. Lond. B 369, 20120512 (2013).

    Article  Google Scholar 

  16. Kadir, S.N., Goodman, D.F.M. & Harris, K.D. Neural Comput. 26, 2379–2394 (2014).

    Article  Google Scholar 

  17. Bonnevie, T. et al. Nat. Neurosci. 16, 309–317 (2013).

    Article  CAS  Google Scholar 

  18. Giocomo, L.M. et al. Curr. Biol. 24, 252–262 (2014).

    Article  CAS  Google Scholar 

  19. Paxinos, G. & Watson, C. The Rat Brain in Stereotaxic Coordinates, 6th edn. (Elsevier Academic Press, 2007).

Download references

Acknowledgements

This work was supported by grants from the Medical Research Council (G1100669) and Wellcome Trust (103896AIA) to K.J.

Author information

Authors and Affiliations

Authors

Contributions

K.J. conceived the study and obtained funding; P.-Y.J. and K.J. designed the protocol; P.-Y.J. and D.O. performed surgeries and recordings; and P.-Y.J., L.S., G.C. and H.P. analyzed data. All authors interpreted data and discussed results. P.-Y.J. and K.J. wrote the manuscript. All authors commented on and edited the manuscript.

Corresponding author

Correspondence to Kate Jeffery.

Ethics declarations

Competing interests

K.J. is a non-shareholding director of Axona Ltd.

Integrated supplementary information

Supplementary Figure 1 Experimental setup and recording protocol.

a, Schematic of the two-compartment apparatus, with one compartment scented with lemon and one with vanilla. Note that the global two-compartment reference frame has 180-degree rotational symmetry so that without the odors, a rat randomly placed in one or other compartment would not know which way it was facing. With the odors, symmetry is broken, and consistent firing of a HD cell across repeated trials indicates use of these odors to inform the directional signal.

b, Protocol for recording trials. The entire apparatus was rotated within the curtained-off arena between trials so as to remove the influence of uncontrolled distal cues. In trials 3 and 4, the rat was randomly placed in one or other compartment, and the door closed to confine the rat to that compartment.

Supplementary Figure 2 Directional firing of RSC HD cells.

a, Firing directions in baseline Trial 1 were homogeneous (determined by a Rayleigh test), both for the whole population of RSC HD neurons (left; n = 96; z = 0.63; p = 0.54) or for the subset recorded in random-start conditions (right; n = 59; z = 1.12; p = 0.33). b and c, Test for use of olfactory context cues. Because HD cells tend to rotate coherently, values for co-recorded cells were averaged to provide an ensemble value. Ensembles and trials were, however, treated as independent data points. Data are for trials 2-5 for the 28 random-start ensembles and shaded according to whether start compartment changed between trials. All firing directions are referred to that in the lemon compartment in Trial 1, with firing directions in the remaining trials expressed as a deviation from this. Data are plotted relative to the expected firing direction based on how the cell fired in baseline trial 1 (expected = zero degrees; directions shown by the compass rose). b, polar plots showing the data collapsed across time; statistics indicate deviation from bimodality (= successful context discrimination). c, The same data as in b but linearized and plotted as a function of recording day. There was significant preference for the original firing direction in some trials (data clustered around zero), but also significant ambiguity (data at 180 deg., Supplementary Table 2). Therefore, the HD system can use odor cues to break environmental symmetry, but this effect is weak.

Supplementary Figure 3 Polar plots from six BD cells.

Note that the cells showed bidirectional firing, with two tuning curves at 180 degrees, illustrated as in Fig. 1. Note that when firing is analysed for only one compartment the tuning curves become unidirectional, indicating directional firing that is compartment-specific and reverses between the compartments. The directionally color-coded spatial spike plots (fourth row of each panel) confirm the compartment-specific distribution of directionality.

Supplementary Figure 4 Calculation of the bimodality of directional firing (the ‘flip score’).

a, Computation of the flip score for a BD (top) and a classic HD cell (bottom) across five recording trials. The flip score is the difference of correlation values at 180° and 90° extracted from the circular autocorrelogram. b, Distribution of flip score values of BD and HD cells, showing the bimodality and the index. The black line shows the kernel density estimation of distributions and the red line is set at the natural splitting of flip score (value equal to 0.6). c, Polar distribution of firing direction of all 116 BD cells after using the double-angle procedure (see Methods). Each dot corresponds to a BD cell firing direction and is color-coded according to rat (n = 4). There was no significant clustering of firing direction (Rayleigh test, p = 0.31).

Supplementary Figure 5 Ruling out that bidirectional tuning curves might have been an artefact of recording and/or analysis.

a, BD cells were not two cells because centers of mass (CoM) of clusters from the individual tuning curves were no further apart in cluster space than for HD cells randomly divided in two. The Euclidean distance in μV between the centers of mass (CoM) of BD clusters was compared with the distance generated by a control procedure, in which 96 HD cell clusters were arbitrarily divided in two 400 times. The distances of the resulting sub-clusters were then averaged. Mean (+/- s.d.) CoM distance (uV) between the two sub-clusters of BD cells was 13.79 +/- 10.35 and for the randomized data it was 12.98 +/- 10.25; these did not differ [t(38513) = 0.85, p = 0.40]. b, The angles between tuning curves of randomly selected pairs of HD cells were near 180 degrees for only 4% of the time (red bars) whereas 100% of the 116 double-tuning curve cells in our data had angular separations that fell within this range. (χ2(1) = 185, p < 0.0001). c Peak firing rates for the individual tuning curves of the BD cells were highly correlated (left; R = 0.91) whereas for randomly selected cell-pairs the firing rates showed a low correlation (right; R = 0.00 ± 0.01), which the value for the data (dotted line) far exceeded (z = 12.04, p < 0.00001). And finally, apparent bidirectionality due to confusion of the large vs small headstage lights (used to determine head direction) can be ruled out because HD and BD cells were co-recorded (Fig. 2, Supp. Fig. 6).

Supplementary Figure 6 Co-recording of BD and HD cells reveals dissociation between tuning curve directions in different compartments.

a, Eight examples, from six different days and three different animals, of simultaneous recordings of BD and HD cells. b, Example of one BD cell simultaneously recorded with two HD cells. c, Two examples of two BD cells with two different firing directions simultaneously recorded with one HD cell. In all these examples, note that the BD cell tuning curves reverse between compartments while the HD cell tuning curves do not.

Supplementary Figure 7 Disjunctive rotations of HD and BD cells further reveal dissociated directionality.

a, Simultaneous recordings of HD and BD cells from two recording sessions (top and bottom). b, Example of an HD cell simultaneously recorded with two BD cells. HD cells showed incoherent rotations of their firing directions relative to apparatus rotations, while BD cells follow these apparatus rotations. Comparing baseline with rotated baseline, all 116 BD cells followed rotation of the box, while only 75 of 96 HD cells did so; χ2(1) = 28.165, p= 1.1 x10-7. Comparing trial 1 and trial 5 baselines, all 116 BD cells followed the box, while only 77 of 96 HD cells did so, χ2(1) = 25,218, p= 5.1 x10-7. This demonstrates uncoupling within the unipolar attractor network, and suggests stronger control by visual cues over BD cells than HD cells.

Supplementary Figure 8 Histology.

Coronal slices through the region of electrode implantation from (a) the retrosplenial animals (RSC), (b) the anterodorsal nucleus of the thalamus (ADN), and (c) postsubiculum (PoS). Main images are at low power and insets at high, with scale bars indicating 1 mm. The hollow dots show the deepest location at which directional neurons were found, estimated using the line drawings (right) adapted from Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. 6th Edition. Amsterdam; Boston: Academic Press/Elsevier; 2007.

Supplementary Figure 9 Preservation of firing-direction reversal in the dark but not in an open field.

We turned off the lights and recorded RSC and PoS cells in light followed by two dark trials (n = 12 sessions), the first being a continuation of the last trial, then a second dark trial after removal, mild disorientation and a rotation of the environment. a, Examples of BD and HD cells recorded during two darkness trials, showing preserved directionality. b, Scatter and data density distributions showing a similar distribution between the two populations in darkness as in light (average bidirectionality or ‘flip’ score for BD cells in both ‘darkness’ and ‘rotated darkness’ trials = 1.06 ±0.40 for both). c, Correlation of firing between light-dark trials (baseline vs. dark) and dark-dark trials (dark vs. rotated dark): HD cells: light/dark = 0.91; dark/dark = 0.84; BD cells: light/dark = 0.78; dark/dark = 0.71. These high correlations indicate that the bidirectional firing is maintained without visual information. Both types of cells rotated coherently with the apparatus during the second dark trial, suggesting that directional firing was not due solely to visual cues. d, Examples of three BD cells recorded in an open field. No clear directional activity is apparent. e, Scatter and data density distribution showing that 9 BD cells lost both their directional and bidirectional pattern in an open arena (average flip score = -0.17 ± 0.18).

Supplementary Figure 10 Within-compartment bidirectional activity.

a, Polar plots from six within-compartment bidirectional (WC-BD) cells. Note that for door-closed odor 1 and 2 trials the tuning curves remained bidirectional, indicating directional firing that reversed within the compartments. The directionally color-coded spatial spike plots (fourth row of each panel) show that firing direction is not compartment-specific as it is with the between-compartment bidirectional (BC-BD) cells. b, Distribution of flip score as a function of recording days for BC-BD cells and WC-BD cells, and HD cells, in trial 1 for the four RSC implanted animals. The two types of bidirectional cells were both evident from the first recording day indicating that plasticity, if it occurs, must be rapid.

Supplementary Figure 11 Hypothesis concerning formation of within-compartment bidirectional firing patterns.

a, A RSC HD cell receives an input from the main HD network, and fires when the rat faces in just one direction. b, A BC-BD cell mainly receives inputs from the local landmark array, and fires when the rat faces in a direction defined by the array; hence it reverses firing direction between compartments. c, A WC-BD cell is driven by landmark inputs, and so initially fires “North” in one compartment and “South” in the other. This induces it to be co-active with “North” and South” HD cells, which acquire inputs by Hebbian co-activation. Now, the cell can fire when the rat faces in either direction. However, in one of the directions it also receives inputs from the landmarks, and so fires slightly more strongly.

Supplementary Figure 12 Spatial patterns of firing by BD neurons.

a, Eight BD cells showing spatial activity related to the doorway. b, Examples from a WC-BD cell revealing a spatial reversing pattern of the tuning curves within each compartment. Such examples were extremely rare, but suggest a spatial component in the inputs to RSC directional neurons.

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–12 and Supplementary Tables 1–3 (PDF 2304 kb)

Supplementary Methods Checklist (PDF 445 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Jacob, PY., Casali, G., Spieser, L. et al. An independent, landmark-dominated head-direction signal in dysgranular retrosplenial cortex. Nat Neurosci 20, 173–175 (2017). https://doi.org/10.1038/nn.4465

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nn.4465

This article is cited by

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