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.

  • Article
  • Published:

Action suppression reveals opponent parallel control via striatal circuits

Nature volume 607, pages 521–526 (2022)Cite this article

Subjects

Abstract

The direct and indirect pathways of the basal ganglia are classically thought to promote and suppress action, respectively1. However, the observed co-activation of striatal direct and indirect medium spiny neurons2 (dMSNs and iMSNs, respectively) has challenged this view. Here we study these circuits in mice performing an interval categorization task that requires a series of self-initiated and cued actions and, critically, a sustained period of dynamic action suppression. Although movement produced the co-activation of iMSNs and dMSNs in the sensorimotor, dorsolateral striatum (DLS), fibre photometry and photo-identified electrophysiological recordings revealed signatures of functional opponency between the two pathways during action suppression. Notably, optogenetic inhibition showed that DLS circuits were largely engaged to suppress—and not promote—action. Specifically, iMSNs on a given hemisphere were dynamically engaged to suppress tempting contralateral action. To understand how such regionally specific circuit function arose, we constructed a computational reinforcement learning model that reproduced key features of behaviour, neural activity and optogenetic inhibition. The model predicted that parallel striatal circuits outside the DLS learned the action-promoting functions, generating the temptation to act. Consistent with this, optogenetic inhibition experiments revealed that dMSNs in the associative, dorsomedial striatum, in contrast to those in the DLS, promote contralateral actions. These data highlight how opponent interactions between multiple circuit- and region-specific basal ganglia processes can lead to behavioural control, and establish a critical role for the sensorimotor indirect pathway in the proactive suppression of tempting actions.

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

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

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

Fig. 1: Mice learned to dynamically suppress lateralized actions.
Fig. 2: Opposite modulation of DLS iMSNs and dMSNs during action suppression.
Fig. 3: Optogenetic inhibition of DLS iMSNs, but not dMSNs, disrupted action suppression and selection, whereas inhibition of dMSNs, but not iMSNs, slowed movement.
Fig. 4: A simplified dual agent model reproduces behaviour, neural activity and effects of optogenetic inhibition.

Similar content being viewed by others

Data availability

The data and analysis code that support the findings of this study are available from the corresponding author upon reasonable request. Source data are provided with this paper.

References

  1. Albin, R. L., Young, A. B. & Penney, J. B. The functional anatomy of basal ganglia disorders. Trends Neurosci. 12, 366–375 (1989).

    Article  CAS  PubMed  Google Scholar 

  2. Cui, G. et al. Concurrent activation of striatal direct and indirect pathways during action initiation. Nature 494, 238–242 (2013).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  3. Schultz, W. in Functions of the Cortico-Basal Ganglia Loop (eds Kimura, M. & Graybiel, A. M.) 31–48 (Springer, 1995).

  4. Doya, K. What are the computations of the cerebellum, the basal ganglia and the cerebral cortex? Neural Netw. 12, 961–974 (1999).

    Article  CAS  PubMed  Google Scholar 

  5. Barkley, R. A. Behavioral inhibition, sustained attention, and executive functions: constructing a unifying theory of ADHD. Psychol. Bull. 121, 65–94 (1997).

    Article  PubMed  Google Scholar 

  6. Gerfen, C. R. & Surmeier, D. J. Modulation of striatal projection systems by dopamine. Annu. Rev. Neurosci. 34, 441–466 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Alexander, G. E. & Crutcher, M. D. Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci. 13, 266–271 (1990).

    Article  CAS  PubMed  Google Scholar 

  8. Kravitz, A. V. et al. Regulation of parkinsonian motor behaviours by optogenetic control of basal ganglia circuitry. Nature 466, 622–626 (2010).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  9. Freeze, B. S., Kravitz, A. V., Hammack, N., Berke, J. D. & Kreitzer, A. C. Control of basal ganglia output by direct and indirect pathway projection neurons. J. Neurosci. 33, 18531–18539 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Denny-Brown, D. & Yanagisawa, N. The role of the basal ganglia in the initiation of movement. Res. Publ. Assoc. Res. Nerv. Ment. Dis. 55, 115–149 (1976).

    CAS  PubMed  Google Scholar 

  11. Mink, J. W. The basal ganglia: focused selection and inhibition of competing motor programs. Prog. Neurobiol. 50, 381–425 (1996).

    Article  CAS  PubMed  Google Scholar 

  12. Redgrave, P., Prescott, T. J. & Gurney, K. The basal ganglia: a vertebrate solution to the selection problem? Neuroscience 89, 1009–1023 (1999).

    Article  CAS  PubMed  Google Scholar 

  13. Gouvêa, T. S. et al. Striatal dynamics explain duration judgments. eLife 4, e11386 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  14. Soares, S., Atallah, B. V. & Paton, J. J. Midbrain dopamine neurons control judgment of time. Science 354, 1273–1277 (2016).

    Article  ADS  CAS  PubMed  Google Scholar 

  15. Matias, S., Lottem, E., Dugué, G. P. & Mainen, Z. F. Activity patterns of serotonin neurons underlying cognitive flexibility. eLife 6, e20552 (2017).

    Article  PubMed  PubMed Central  Google Scholar 

  16. Lima, S. Q., Hromádka, T., Znamenskiy, P. & Zador, A. M. PINP: a new method of tagging neuronal populations for identification during in vivo electrophysiological recording. PLoS One 4, e6099 (2009).

    Article  ADS  PubMed  PubMed Central  CAS  Google Scholar 

  17. Jin, X. & Costa, R. M. Start/stop signals emerge in nigrostriatal circuits during sequence learning. Nature 466, 457–462 (2010).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  18. Klaus, A. et al. The spatiotemporal organization of the striatum encodes action space. Neuron 96, 949 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Markowitz, J. E. et al. The striatum organizes 3D behavior via moment-to-moment action selection. Cell 174, 44–58 (2018).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Han, X. et al. A high-light sensitivity optical neural silencer: development and application to optogenetic control of non-human primate cortex. Front. Syst. Neurosci. 5, 18 (2011).

    Article  PubMed  PubMed Central  Google Scholar 

  21. Nagel, G. et al. Channelrhodopsin-2, a directly light-gated cation-selective membrane channel. Proc. Natl Acad. Sci. USA 100, 13940–13945 (2003).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  22. Turner, R. S. & Desmurget, M. Basal ganglia contributions to motor control: a vigorous tutor. Curr. Opin. Neurobiol. 20, 704–716 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Panigrahi, B. et al. Dopamine Is required for the neural representation and control of movement vigor. Cell 162, 1418–1430 (2015).

    Article  CAS  PubMed  Google Scholar 

  24. Dudman, J. T. & Krakauer, J. W. The basal ganglia: from motor commands to the control of vigor. Curr. Opin. Neurobiol. 37, 158–166 (2016).

    Article  CAS  PubMed  Google Scholar 

  25. Sutton, R. S. & Barto, A. G. Reinforcement Learning: an Introduction (MIT Press, 1998).

  26. Bornstein, A. M. & Daw, N. D. Multiplicity of control in the basal ganglia: computational roles of striatal subregions. Curr. Opin. Neurobiol. 21, 374–380 (2011).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Shen, W., Flajolet, M., Greengard, P. & Surmeier, D. J. Dichotomous dopaminergic control of striatal synaptic plasticity. Science 321, 848–851 (2008).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  28. Collins, A. G. E. & Frank, M. J. Opponent actor learning (OpAL): modeling interactive effects of striatal dopamine on reinforcement learning and choice incentive. Psychol. Rev. 121, 337–366 (2014).

    Article  PubMed  Google Scholar 

  29. Gurney, K. N., Humphries, M. D. & Redgrave, P. A new framework for cortico-striatal plasticity: behavioural theory meets in vitro data at the reinforcement-action interface. PLoS Biol. 13, e1002034 (2015).

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  30. Iino, Y. et al. Dopamine D2 receptors in discrimination learning and spine enlargement. Nature 579, 555–560 (2020).

    Article  ADS  CAS  PubMed  Google Scholar 

  31. Lee, S. J. et al. Cell-type-specific asynchronous modulation of PKA by dopamine in learning. Nature 590, 451–456 (2021).

    Article  ADS  CAS  PubMed  Google Scholar 

  32. Dayan, P. Improving generalization for temporal difference learning: the successor representation. Neural Comput. 5, 613–624 (1993).

    Article  Google Scholar 

  33. Stachenfeld, K. L., Botvinick, M. M. & Gershman, S. J. The hippocampus as a predictive map. Nat. Neurosci. 20, 1643–1653 (2017).

    Article  CAS  PubMed  Google Scholar 

  34. Tai, L.-H., Lee, A. M., Benavidez, N., Bonci, A. & Wilbrecht, L. Transient stimulation of distinct subpopulations of striatal neurons mimics changes in action value. Nat. Neurosci. 15, 1281–1289 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Majid, D. S. A., Cai, W., Corey-Bloom, J. & Aron, A. R. Proactive selective response suppression is implemented via the basal ganglia. J. Neurosci. 33, 13259–13269 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Watanabe, M. & Munoz, D. P. Presetting basal ganglia for volitional actions. J. Neurosci. 30, 10144–10157 (2010).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Ford, K. A. & Everling, S. Neural activity in primate caudate nucleus associated with pro- and antisaccades. J. Neurophysiol. 102, 2334–2341 (2009).

    Article  PubMed  Google Scholar 

  38. Amita, H. & Hikosaka, O. Indirect pathway from caudate tail mediates rejection of bad objects in periphery. Sci. Adv. 5, eaaw9297 (2019).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  39. Parent, A. & De Bellefeuille, L. Organization of efferent projections from the internal segment of globus pallidus in primate as revealed by flourescence retrograde labeling method. Brain Res. 245, 201–213 (1982).

    Article  CAS  PubMed  Google Scholar 

  40. Lee, J. & Sabatini, B. L. Striatal indirect pathway mediates exploration via collicular competition. Nature 599, 645–649 (2021).

    Article  ADS  CAS  PubMed  Google Scholar 

  41. Tecuapetla, F., Matias, S., Dugue, G. P., Mainen, Z. F. & Costa, R. M. Balanced activity in basal ganglia projection pathways is critical for contraversive movements. Nat. Commun. 5, 4315 (2014).

    Article  ADS  CAS  PubMed  Google Scholar 

  42. Parker, J. G. et al. Diametric neural ensemble dynamics in parkinsonian and dyskinetic states. Nature 557, 177–182 (2018).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  43. Park, J., Coddington, L. T. & Dudman, J. T. Basal ganglia circuits for action specification. Annu. Rev. Neurosci. 43, 485–507 (2020).

    Article  CAS  PubMed  Google Scholar 

  44. Alexander, G. E., DeLong, M. R. & Strick, P. L. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu. Rev. Neurosci. 9, 357–381 (1986).

    Article  CAS  PubMed  Google Scholar 

  45. Prescott, T. J., Montes González, F. M., Gurney, K., Humphries, M. D. & Redgrave, P. A robot model of the basal ganglia: behavior and intrinsic processing. Neural Netw. 19, 31–61 (2006).

    Article  PubMed  MATH  Google Scholar 

  46. Lau, B., Monteiro, T. & Paton, J. J. The many worlds hypothesis of dopamine prediction error: implications of a parallel circuit architecture in the basal ganglia. Curr. Opin. Neurobiol. 46, 241–247 (2017).

    Article  CAS  PubMed  Google Scholar 

  47. Daw, N. D., Niv, Y. & Dayan, P. Uncertainty-based competition between prefrontal and dorsolateral striatal systems for behavioral control. Nat. Neurosci. 8, 1704–1711 (2005).

    Article  CAS  PubMed  Google Scholar 

  48. Dorfman, H. M. & Gershman, S. J. Controllability governs the balance between Pavlovian and instrumental action selection. Nat. Commun. 10, 5826 (2019).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  49. Dayan, P., Niv, Y., Seymour, B. & Daw, N. D. The misbehavior of value and the discipline of the will. Neural Netw. 19, 1153–1160 (2006).

    Article  PubMed  MATH  Google Scholar 

  50. Gerfen, C. R., Paletzki, R. & Heintz, N. GENSAT BAC cre-recombinase driver lines to study the functional organization of cerebral cortical and basal ganglia circuits. Neuron 80, 1368–1383 (2013).

    Article  CAS  PubMed  Google Scholar 

  51. Madisen, L. et al. A toolbox of Cre-dependent optogenetic transgenic mice for light-induced activation and silencing. Nat. Neurosci. 15, 793–802 (2012).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Lopes, G. et al. Bonsai: an event-based framework for processing and controlling data streams. Front. Neuroinform. 9, 7 (2015).

    Article  PubMed  PubMed Central  Google Scholar 

  53. Chen, T.-W. et al. Ultrasensitive fluorescent proteins for imaging neuronal activity. Nature 499, 295–300 (2013).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  54. Pisanello, F. et al. Dynamic illumination of spatially restricted or large brain volumes via a single tapered optical fiber. Nat. Neurosci. 20, 1180–1188 (2017).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Siegle, J. H. et al. Open Ephys: an open-source, plugin-based platform for multichannel electrophysiology. J. Neural Eng. 14, 045003 (2017).

    Article  ADS  PubMed  Google Scholar 

  56. Benhamou, L., Kehat, O. & Cohen, D. Firing pattern characteristics of tonically active neurons in rat striatum: context dependent or species divergent? J. Neurosci. 34, 2299–2304 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Yael, D. et al. Haloperidol-induced changes in neuronal activity in the striatum of the freely moving rat. Front. Syst. Neurosci. 7, 110 (2013).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  58. Rennaker, R. L., Miller, J., Tang, H. & Wilson, D. A. Minocycline increases quality and longevity of chronic neural recordings. J. Neural Eng. 4, L1–L5 (2007).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Kvitsiani, D. et al. Distinct behavioural and network correlates of two interneuron types in prefrontal cortex. Nature 498, 363–366 (2013).

    Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 

  60. Chuong, A. S. et al. Noninvasive optical inhibition with a red-shifted microbial rhodopsin. Nat. Neurosci. 17, 1123–1129 (2014).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Mathis, A. et al. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nat. Neurosci. 21, 1281–1289 (2018).

    Article  CAS  PubMed  Google Scholar 

  62. Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting Linear Mixed-Effects Models Using lme4. J. Stat. Softw. 67, https://doi.org/10.18637/jss.v067.i011 (2015).

  63. Searle, S. R., Speed, F. M. & Milliken, G. A. Population marginal means in the linear model: an alternative to least squares means. Am. Stat. 34, 216–221 (1980).

    MathSciNet  MATH  Google Scholar 

  64. Lenth, R. Least-squares means: the R package lsmeans. J. Stat. Softw. 69, https://doi.org/10.18637/jss.v069.i01 (2016).

  65. Gibbon, J. Scalar expectancy theory and Weber’s law in animal timing. Psychol. Rev. 84, 279–325 (1977).

    Article  Google Scholar 

  66. Merel, J., Botvinick, M. & Wayne, G. Hierarchical motor control in mammals and machines. Nat. Commun. 10, 5489 (2019).

    Article  ADS  PubMed  PubMed Central  CAS  Google Scholar 

  67. Motiwala, A., Soares, S., Atallah, B. V., Paton, J. J. & Machens, C. K. Efficient coding of cognitive variables underlies dopamine response and choice behavior. Nat. Neurosci. 25, 738–748 (2022).

  68. Grondman, I., Busoniu, L., Lopes, G. A. D. & Babuska, R. A survey of actor-critic reinforcement learning: standard and natural policy gradients. In IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews) Vol 42, 1291–1307 (IEEE, 2012).

  69. Buşoniu, L., Babuška, R. & De Schutter, B. in Innovations in Multi-Agent Systems and Applications - 1 (eds Srinivasan, D. & Jain, L. C.) 183–221 (Springer, 2010).

  70. Franklin, K. B. J. & Paxinos, G. The Mouse Brain in Stereotaxic Coordinates 3rd edn (Academic Press, 2008).

Download references

Acknowledgements

We thank B. Lau, J. Krakauer, E. DeWitt, B. Atallah, A. Klaus and T. Monteiro for comments on different versions of the manuscript and the entire J.J.P. laboratory for feedback during the course of this project. We also thank the ABBE Facility and the Scientific Hardware, Histopathology and Rodent Champalimaud Research Platforms for technical assistance, and B. Zarov and D. Domingues for helping with the training of some of the mice in this study. This work was developed with the support from the research infrastructure Congento, co-financed by Lisboa Regional Operational Programme (Lisboa2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF) and Fundação para a Ciência e Tecnologia (Portugal) under the project LISBOA-01-0145-FEDER-022170. The work was funded by an HHMI International Research Scholar Award to J.J.P. (55008745); a European Research Council Consolidator grant (DYCOCIRC - REP-772339-1) to J.J.P.; a Bial bursary for scientific research to J.J.P. (193/2016); internal support from the Champalimaud Foundation; and PhD fellowships (PD/BD/105945/2014 and PD/BD/128301/2017) from Fundação para a Ciência e Tecnologia to B.F.C. and G.G. respectively.

Author information

Author notes

  1. Sofia Soares

    Present address: Department of Neurobiology, Harvard Medical School, Boston, MA, USA

  2. Asma Motiwala

    Present address: Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA, USA

Authors and Affiliations

  1. Champalimaud Neuroscience Programme, Champalimaud Research, Champalimaud Foundation, Lisbon, PT, Portugal

    Bruno F. Cruz, Gonçalo Guiomar, Sofia Soares, Asma Motiwala, Christian K. Machens & Joseph J. Paton

Authors
  1. Bruno F. Cruz
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Gonçalo Guiomar
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sofia Soares
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Asma Motiwala
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Christian K. Machens
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Joseph J. Paton
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

J.J.P. and B.F.C. conceived of the experiments and wrote the manuscript. S.S. assisted in analysis of the fibre photometry data and helped revise the manuscript. B.F.C. performed all experiments and analysed the data. G.G., B.F.C. and J.J.P. conceived of the model with assistance from A.M. G.G. built the model and performed all simulations. C.K.M. provided constructive feedback about the model and helped revise the manuscript. J.J.P. supervised all aspects of the project.

Corresponding author

Correspondence to Joseph J. Paton.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Peer review

Peer review information

Nature thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Histological reconstruction of sites of fibre implantation for photometry, electrophysiology and optogenetic experiments in the DLS and the DMS.

a–d, Photometry (a), electrophysiology (b) and optogenetic experiments in the DLS (c) and the DMS (d). Mice are coloured by their genotype according to the legend. For electrophysiology and optogenetics experiments, the DV coordinate is shown as the deepest position of the shanks or tapered fibre, respectively, was observed in histological slices. All coordinates were projected to the same coronal slice (AP = +0.5 from bregma) adapted from70. Behaviour metrics across genotypes (e-n). A2a-Cre and D1-Cre single mice, included in the photometry experiments, are shown in red and blue, respectively. Single-mouse psychometric curve (e) fits and respective parameters (see Methods for further details, two-tailed t-test, f, p = 0.935, t12 = 0.083, g, P = 0.17, t12 = −1.459, h, P = 0.823, t12 = 0.228, i, P = 0.826, t12 = −0.225). j, Overall probability of breaking fixation (all trials included) (P = 0.665, t12 = 0.445) k, Percentage of trials in which mice attempted to make a choice after breaking fixation (all trials included) (P = 0.703, t12 = 0.391). l, Probability of reporting at the “long choice” port after breaking fixation contingent on whether the mouse aborted before (<−1.5 s) or after (>1.5 s) the decision boundary (P = 0.872, t12 = −0.165). m, Left, Hazard of breaking fixation in time for single mice (thin curves) and overall averages within genotype (thick lines). Right, differences between the hazard of breaking fixation after and before the decision boundary(P = 0.165, t12  = 1.48). n, Mean velocity during the delay period from correct trials of the longest interval (2.4 s, Data from Fig. 1c, P = 0.892, t12 = −0.139). Error bars represent s.e.m. n.s. P > 0.05.

Source data

Extended Data Fig. 2 Photometry activity during the period of active immobility and movement.

a) Average activity across both hemispheres during the immobility period for each hemisphere (data points) relative to baseline, and across mice. Activity for iMSNs and dMSNs is shown in red and blue, respectively (n = 16 hemispheres, from 8 A2a-Cre mice, n = 10 hemispheres from 6 D1-Cre mice, only correct trials were included). b,c, Difference between average activity after the 1.5-s decision boundary, and before the 1.5-s decision boundary (>1.5 s - <1.5 s) for activity recorded in the hemisphere contralateral to the side of the long choice port (CL, contra-long, b, N = 8 A2a-Cre and 5 D1-Cre mice)) and for activity recorded in the hemisphere contralateral to the short choice port (CS, contra-short, c, N = 8 A2a-Cre and 5 D1-Cre mice)). White lines and boxes represent mean and s.e.m., respectively. d-e) Both DLS direct and indirect pathways are more active during contralateral movements. Photometry signal recorded from from A2a-Cre (d) and D1-Cre (e), aligned to leaving the centre port during a near boundary interval (1.74s) where the same stimulus results in different choices (ipsilateral and contralateral to the recorded hemisphere, depicted as cyan and orange, respectively). Same trial selection as Fig. 2e. All boxes or shaded regions represent mean ± s.e.m.

Source data

Extended Data Fig. 3 The photo-identified iMSN population is enriched in cells with a higher firing rate during the delay period.

a, Activity profile of photo-identified indirect-pathway MSNs (photo-Ided iMSNs) and non-photo-identified putative MSNs (see methods for details). Each row represents a unit’s z-scored activity aligned to trial initiation that results from averaging the activity for all intervals cropped at second tone. Units are ordered by the angular position formed by the first two principal component projections. PCs were computed using a period of −2 to 2.4 s from trial initiation. b, Cumulative distribution of changes in firing rate during the delay period of photo-Ided iMSN (red) and all other putative MSNs (blue). Average ΔFR is significantly larger for iMSNs when compared to the distribution of non-identified cells (two-tailed t-test, P = 0.0196, t286 = 2.35). c, Proportion of up, down, and not modulated cells during the delay period (see methods for details). Proportions are significantly different between the two groups (Chi-squared test, P = 0.0115, χ22 = 8.939). d, Example of putative MSNs (pMSNs) classified as photo-identified. Top) Raster plot with single spikes aligned to laser pulse onset (2-ms duration). Bottom-Left) Distribution of latencies of the first spike observed in a 1–10ms after laser pulse onset. Bottom-right) mean waveform (black) and mean laser-triggered waveform (red). e,h, Distribution of: e, probability of observing a spike between 1 and 10ms after laser onset pulse, f, differences in firing rate between the baseline and 1–10ms post-pulse window, g, correlation coefficient (ρ) between mean waveform and mean laser-triggered waveform, h, median latency of the first spike in a window 1–50 ms after pulse onset (trials wherein no spike was observed in this window were not included).

Source data

Extended Data Fig. 4 Changes of activity in the indirect pathway preceding broken fixations.

a, Differences on the rate (derivative) of activity change (d(ΔF/FBrokenFixation)/dt- d(ΔF/FValid)/dt) aligned on broken fixations calculated from data recorded from A2a-Cre mice (16 hemispheres from 8 mice). b, Difference of mean activity (FRBrokenFixation- FRValid, Hz) of all photo-identified units not modulated (left, N = 46 units), positively modulated (centre, N = 12 units) and negatively modulated (right, N = 31 units) during the delay period. Blue lines depict periods during which there are significant differences between the average activity average activity on broken fixations and valid trials, across cells (two-tailed, paired t-test, p \(\leqslant \) 0.05). c, Schematic depicting the analysis performed in a) and b) in order to compared activity aligned to broken fixation trials. In brief, we took valid (black), or broken fixation (green), trials aligned to trial initiation (first-tone onset) cropped them at second tone or at the broken fixation event, respectively. We used the valid trials to compute a reference “valid trial” that reflected the average activity of all valid trials, cropped at second tone (mid, orange). Averaging available data (i.e. up until second tone) guarantees that only data from the fixation period is used, without incurring into contamination due to movement onset after the cue is delivered. We subsequently align each broken fixation trial to its occurrence (right) and take, from the reference valid trace, a time-matched fragment which we align to the same time since first tone. To compare traces aligned on broken fixation events to valid trials, we then average all broken fixation trials and the corresponding time-matched valid reference traces (see methods). d, Same analysis as in Fig. 3l but for an epoch [−0.5:−0.4]s relative to broken fixation events (Linear regression, slope = −0.042, P = 0.132, t87 = −1.522). Error bars represent s.e.m. n.s. P > 0.05.

Source data

Extended Data Fig. 5 Electrophysiological identification of putative MSNs and summary effect of opsin activation.

a, Identification of putative MSNs based on firing statistics and waveform duration (see methods). Green data points indicate units ultimately identified as putative MSNs (pMSNs). b, Changes in firing rate during light delivery and baseline period versus the baseline firing rate of all recorded isolated units (Including non- putative MSN units) from A2a-Cre (red) and D1-Cre (blue) mice infected with ArchT. Significantly negatively or positively modulated cells are shown as closed and open circles, respectively. Maximum theoretical inhibition is plotted as a grey dashed line (−ΔFR = Baseline FR). c, Distribution of changes in firing rate (Hz) during the period of light delivery, versus baseline, for putatively labelled MSN units recorded from A2a (Red) and D1-Cre (Blue) mice expressing ArchT, outside the context of the task. Grey depicts non-significantly modulated cells, closed and open shapes depict significantly down- and upmodulated cells, respectively. d, Overall average peristimulus time histogram of all negatively light-modulated cells, putatively labelled as MSNs, recorded from A2a-Cre and D1-Cre mice during the ArchT acute experiment. All units were z-scored (see methods). Shaded coloured area depicts the time of laser illumination. e–h, Same as b–d but for mice expressing ChR2 in MSNs. i, Summary of overall modulation effects of ArchT versus ChR2 activation in pMSNs.

Source data

Extended Data Fig. 6 Manipulation-induced changes in vigour and action selection depend on MSN type and striatum sub-location.

a, Cartoon depicts the three different manipulated trial types:Choice Time (Delay), laser was ramped off as the second tone is played. BrokenFix Choice Time (Delay), laser was ramped off as the mouse leaves the centre port causing a broken fixation. MovementTime (Decision) laser was turned on as the second tone is played until the mouse either performs its choice or 400ms elapse, whichever occurs first. b, Differences in single mouse’s median choice time between inhibited and non-inhibited trials (ΔChoiceTime = ChoiceTimeManipulation - ChoiceTimeControl). For each mouse, we concatenated all sessions and split trials in manipulated versus non-manipulated. From top to bottom: two-tailed t-test, P = 0.198, t5 = 1.482; P = 0.198, t5 = 1.482; P = 0.005, t5 = 4.834; P = 0.128, t2 = 2.523; P = 0.718, t4 = 0.388; P = 0.46, t4 = −0.817; P = 0.617, t5 = −0.533; P = 0.774, t5 = −0.304; P = 0.032, t5 = 2.959; P = 0.005, t5 = 4.876; P = 0.247, t5 = 1.309; P = 0.306, t5 = 1.141. c, Differences in probability of reporting a contralateral choice, relative to inhibition side (ΔP = P_ContraManipulation - P_ContraControl). For each mouse, we concatenated sessions from sessions with unilateral perturbation and normalized choices to the side contralateral to inhibition site. From top to bottom:P = 0.597, t11 = −0.545; P = 0.427, t11 = −0.824; P = 0.259, t5 = −1.274; P = 0.149, t7 = 1.623; P << 0.001, t7 = 6.605; P = 0.365, t11 = 1.623; P = 0.028, t11 = −2.528; P = 0.94, t11 = 0.077; P = 0.933, t11 = −0.086; P = 0.985, t11 = −0.02; P = 0.143, t11 = 1.577. All boxes represent mean ± s.e.m. *P ≤ 0.05, **P ≤ 0.01, **P ≤ 0.001.

Source data

Extended Data Fig. 7 Further quantification of the effects of selective unilateral inhibition of DLS MSNs on broken fixation trials (see Fig. 3).

a,b, Same as Fig. 3m,q, respectively, but expressed as the hazard rate of breaking fixation (see Methods). c, Quantification of the effect shown in Fig. 3m. We calculated the hazard of breaking fixation during the period where the choice contralateral to the site of inhibition would be correct or incorrect (before/after 1.5 s and after/before 1.5 s for CS and CL, respectively). Data shown are the differences between session matched controls and manipulations. Each pair of points depicts data from the same hemisphere and the colour the site of manipulation. (contralateral_correct: two-tailed t-test, P = 0.005, t7 = 4.055, contralateral_incorrect: P = 0.215, t7 = −1.363, N = 8 Hemispheres). d, Same as a) but referent to Fig. 3q. (contralateral_correct: two-tailed t-test, P = 0.711, t11 = 0.38, contralateral_incorrect: P = 0.211, t11 = 1.329, N = 12 Hemispheres) e, Bias to report a contralateral choice after inhibition of iMSNs is not explained by the tendency of mice to make particular choices after breaking fixation early or late in the delay. Each panel, one for manipulations performed in each hemisphere, depicts the data shown in Fig. 3l further split by whether fixation was broken before or after the 1.5-s decision boundary. All Error bars or boxes represent mean ± s.e.m. n.s. P > 0.05, **P ≤ 0.01.

Source data

Extended Data Fig. 8 Unilateral DLS indirect-pathway inhibition did not systematically affect movement trajectories across subject mice.

. a, Example trajectories for a single mouse aligned to centre-out for choices to the “Long port” (red) or “Short pot” (blue) for completed trials (Left) and Broken fixation trials (Right). Trials are further broken down by whether the indirect pathway was inhibited (bottom) or not (top) in mice implanted in the DLS. b, same as a) but for trajectories measured during the delay period (up until second tone or broken fixation events). c) Quantitative differences between trajectory distributions among different conditions. In brief, we computed a mean reference trajectory from “Valid & Non-inhibited” condition and computed, for each trial from each condition, the average Euclidean distance to this reference trajectory. Values shown in the heat maps correspond to the means of these distributions. Significance was accessed by computing a two-sample Kolmogorov–Smirnov test between the reference and testing condition (P ⩽ 0.05 is reported as a red dots).

Source data

Extended Data Fig. 9 Inhibition of DMS dMSNs has a mild effect on the probability of reporting a contralateral action, in the absence of changes in broken fixation rates.

a, Overall probability of breaking fixation during dMSN inhibition experiments. Coloured and black dots represent data from laser-on and laser-off trials, respectively. “Bilateral” condition represents data from sessions wherein light was delivered bilaterally to the DMS (two-tailed t-test, P = 0.893, t5 = −0.141, N = 6 pairs of hemispheres) whereas “unilateral” represents data from sessions wherein light was delivered to a single hemisphere. (two-tailed t-test, P = 0.09, t11 = −1.861,N = 12 hemispheres. Green/Purple code for CS/CL manipulation sessions, respectively). b, Change in probability of registering a choice at the port contralateral to the hemisphere manipulated, after breaking fixation (ΔP = PManipulation - PControl, two-tailed t-test, P = 0.028, t11 = −2.528, N = 12 hemispheres). c, Quantification of the effect shown in d) (contralateral_correct: two-tailed t-test, P = 0.076, t11 = −1.961; contralateral_incorrect: P = 0.383, t11 = −0.909). We calculated the hazard of breaking fixation during the period where the choice contralateral to the site of inhibition would be correct or incorrect (before/after 1.5 s and after/before 1.5 s for CS and CL, respectively). d, Distribution of broken fixation times, expressed as the histogram of trial counts normalized over all included trials of a given condition. e, Change in hazard (ΔH = HManipulation - HControl, N = 12 Hemispheres) due to inhibition of the CS (green) or CL (pruple) relative to session matched control trials. f–j, Same as a-e) but for A2a-Cre mice implanted in the DMS. (f) Bilateral: two-tailed t-test, P = 0.463, t5 = 0.795, N = 6 pairs of hemispheres; Unilateral: two-tailed t-test, P = 0.233, t11 = 1.262, 12 hemispheres. g) two-tailed t-test, P = 0.985, t11 = −0.02, N = 12 hemispheres. h, contralateral_correct: two-tailed t-test, P = 0.149, t11 = 1.551; contralateral_incorrect: P = 0.207, t11 = 1.339, N = 12 hemispheres. All error bars or boxes represent mean ± s.e.m. n.s. P > 0.05, *P ≤ 0.05.

Source data

Extended Data Fig. 10 The model’s pattern of broken fixations when inhibiting the DLS indirect pathway.

Same data as Fig. 4m but expressed as hazard rate (see Methods for details). a, Hazard of breaking fixation for the control (black outline) and inhibition conditions resulting from the selective reduction of action preference values for the Indirect pathway (AL,I(stL,at)) for the SHORT (green) or LONG (purple) actions. b, Change in hazard rate (ΔHR = HRManipulation- HRControl).

Supplementary information

Supplementary Tables

This file contains Supplementary Table 1: Summary of statistical tests performed in the study and additional relevant information; and Supplementary Table 2: Summary of the training procedure for the Opponent Multi-Agent Actor-Critic algorithm.

Reporting Summary

Peer Review File

Source data

Rights and permissions

Springer Nature or its licensor 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.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Cruz, B.F., Guiomar, G., Soares, S. et al. Action suppression reveals opponent parallel control via striatal circuits. Nature 607, 521–526 (2022). https://doi.org/10.1038/s41586-022-04894-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/s41586-022-04894-9

This article is cited by

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

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