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.

Amygdala inputs to prefrontal cortex guide behavior amid conflicting cues of reward and punishment

Abstract

Orchestrating appropriate behavioral responses in the face of competing signals that predict either rewards or threats in the environment is crucial for survival. The basolateral nucleus of the amygdala (BLA) and prelimbic (PL) medial prefrontal cortex have been implicated in reward-seeking and fear-related responses, but how information flows between these reciprocally connected structures to coordinate behavior is unknown. We recorded neuronal activity from the BLA and PL while rats performed a task wherein competing shock- and sucrose-predictive cues were simultaneously presented. The correlated firing primarily displayed a BLA→PL directionality during the shock-associated cue. Furthermore, BLA neurons optogenetically identified as projecting to PL more accurately predicted behavioral responses during competition than unidentified BLA neurons. Finally photostimulation of the BLA→PL projection increased freezing, whereas both chemogenetic and optogenetic inhibition reduced freezing. Therefore, the BLA→PL circuit is critical in governing the selection of behavioral responses in the face of competing signals.

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: Behavioral tasks to examine the discrimination and competition of reward and fear memories.
Figure 2: Correlated activity between the BLA and PL varied between the retrieval of reward- and fear-related memories.
Figure 3: BLA neurons biased toward encoding CS-Suc or CS-Shock exhibited similar proportions of correlated activity with PL, whereas a greater proportion of fear-biased PL cells exhibited correlated activity with BLA.
Figure 4: Most BLA→PL cells recorded showed selective excitation to the shock-predictive cue.
Figure 5: BLA cells terminating in PL more accurately predicted the animal's behavioral response during competition.
Figure 6: Stimulation of BLA inputs to PL was sufficient to promote fear-related behavior and bias behavioral responses toward fear during competition.
Figure 7: The BLA→PL pathway is necessary for expression of the fear-associated memory, but not for reward-seeking behavior.

Similar content being viewed by others

References

  1. Maren, S. & Quirk, G.J. Neuronal signalling of fear memory. Nat. Rev. Neurosci. 5, 844–852 (2004).

    CAS  PubMed  Google Scholar 

  2. Janak, P.H. & Tye, K.M. From circuits to behaviour in the amygdala. Nature 517, 284–292 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  3. Rogan, M.T., Stäubli, U.V. & LeDoux, J.E. Fear conditioning induces associative long-term potentiation in the amygdala. Nature 390, 604–607 (1997).

    CAS  PubMed  Google Scholar 

  4. Paton, J.J., Belova, M.A., Morrison, S.E. & Salzman, C.D. The primate amygdala represents the positive and negative value of visual stimuli during learning. Nature 439, 865–870 (2006).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Tye, K.M., Stuber, G.D., de Ridder, B., Bonci, A. & Janak, P.H. Rapid strengthening of thalamo-amygdala synapses mediates cue-reward learning. Nature 453, 1253–1257 (2008).

    CAS  PubMed  PubMed Central  Google Scholar 

  6. Shabel, S.J. & Janak, P.H. Substantial similarity in amygdala neuronal activity during conditioned appetitive and aversive emotional arousal. Proc. Natl. Acad. Sci. USA 106, 15031–15036 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  7. Namburi, P. et al. A circuit mechanism for differentiating positive and negative associations. Nature 520, 675–678 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  8. Beyeler, A. et al. Divergent routing of positive and negative information from the amygdala during memory retrieval. Neuron 90, 348–361 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  9. Choi, J.-S. & Kim, J.J. Amygdala regulates risk of predation in rats foraging in a dynamic fear environment. Proc. Natl. Acad. Sci. USA 107, 21773–21777 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  10. Amir, A., Lee, S.-C., Headley, D.B., Herzallah, M.M. & Pare, D. Amygdala signaling during foraging in a hazardous environment. J. Neurosci. 35, 12994–13005 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  11. Burgos-Robles, A., Vidal-Gonzalez, I. & Quirk, G.J. Sustained conditioned responses in prelimbic prefrontal neurons are correlated with fear expression and extinction failure. J. Neurosci. 29, 8474–8482 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  12. Senn, V. et al. Long-range connectivity defines behavioral specificity of amygdala neurons. Neuron 81, 428–437 (2014).

    CAS  PubMed  Google Scholar 

  13. Moorman, D.E. & Aston-Jones, G. Prefrontal neurons encode context-based response execution and inhibition in reward seeking and extinction. Proc. Natl. Acad. Sci. USA 112, 9472–9477 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  14. McGarry, L.M. & Carter, A.G. Inhibitory gating of basolateral amygdala inputs to the prefrontal cortex. J. Neurosci. 36, 9391–9406 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  15. Dilgen, J., Tejeda, H.A. & O'Donnell, P. Amygdala inputs drive feedforward inhibition in the medial prefrontal cortex. J. Neurophysiol. 110, 221–229 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  16. Gabbott, P.L.A., Warner, T.A., Jays, P.R.L., Salway, P. & Busby, S.J. Prefrontal cortex in the rat: projections to subcortical autonomic, motor, and limbic centers. J. Comp. Neurol. 492, 145–177 (2005).

    PubMed  Google Scholar 

  17. Ishikawa, A., Ambroggi, F., Nicola, S.M. & Fields, H.L. Contributions of the amygdala and medial prefrontal cortex to incentive cue responding. Neuroscience 155, 573–584 (2008).

    CAS  PubMed  Google Scholar 

  18. Burgos-Robles, A., Bravo-Rivera, H. & Quirk, G.J. Prelimbic and infralimbic neurons signal distinct aspects of appetitive instrumental behavior. PLoS One 8, e57575 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  19. Burgos-Robles, A., Vidal-Gonzalez, I., Santini, E. & Quirk, G.J. Consolidation of fear extinction requires NMDA receptor-dependent bursting in the ventromedial prefrontal cortex. Neuron 53, 871–880 (2007).

    CAS  PubMed  Google Scholar 

  20. Sotres-Bayon, F. & Quirk, G.J. Prefrontal control of fear: more than just extinction. Curr. Opin. Neurobiol. 20, 231–235 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  21. Courtin, J. et al. Prefrontal parvalbumin interneurons shape neuronal activity to drive fear expression. Nature 505, 92–96 (2014).

    PubMed  Google Scholar 

  22. Sierra-Mercado, D., Padilla-Coreano, N. & Quirk, G.J. Dissociable roles of prelimbic and infralimbic cortices, ventral hippocampus, and basolateral amygdala in the expression and extinction of conditioned fear. Neuropsychopharmacology 36, 529–538 (2011).

    PubMed  Google Scholar 

  23. Sangha, S., Robinson, P.D., Greba, Q., Davies, D.A. & Howland, J.G. Alterations in reward, fear and safety cue discrimination after inactivation of the rat prelimbic and infralimbic cortices. Neuropsychopharmacology 39, 2405–2413 (2014).

    PubMed  PubMed Central  Google Scholar 

  24. Laviolette, S.R., Lipski, W.J. & Grace, A.A. A subpopulation of neurons in the medial prefrontal cortex encodes emotional learning with burst and frequency codes through a dopamine D4 receptor-dependent basolateral amygdala input. J. Neurosci. 25, 6066–6075 (2005).

    CAS  PubMed  PubMed Central  Google Scholar 

  25. Sotres-Bayon, F., Sierra-Mercado, D., Pardilla-Delgado, E. & Quirk, G.J. Gating of fear in prelimbic cortex by hippocampal and amygdala inputs. Neuron 76, 804–812 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  26. Nieh, E.H. et al. Decoding neural circuits that control compulsive sucrose seeking. Cell 160, 528–541 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  27. Tye, K.M. et al. Amygdala circuitry mediating reversible and bidirectional control of anxiety. Nature 471, 358–362 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  28. Chow, B.Y. et al. High-performance genetically targetable optical neural silencing by light-driven proton pumps. Nature 463, 98–102 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  29. Mattis, J. et al. Principles for applying optogenetic tools derived from direct comparative analysis of microbial opsins. Nat. Methods 9, 159–172 (2011).

    PubMed  PubMed Central  Google Scholar 

  30. Armbruster, B.N., Li, X., Pausch, M.H., Herlitze, S. & Roth, B.L. Evolving the lock to fit the key to create a family of G protein-coupled receptors potently activated by an inert ligand. Proc. Natl. Acad. Sci. USA 104, 5163–5168 (2007).

    PubMed  PubMed Central  Google Scholar 

  31. Kim, J., Pignatelli, M., Xu, S., Itohara, S. & Tonegawa, S. Antagonistic negative and positive neurons of the basolateral amygdala. Nat. Neurosci. 19, 1636–1646 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  32. Mahn, M., Prigge, M., Ron, S., Levy, R. & Yizhar, O. Biophysical constraints of optogenetic inhibition at presynaptic terminals. Nat. Neurosci. 19, 554–556 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  33. Zhang, W. et al. Functional circuits and anatomical distribution of response properties in the primate amygdala. J. Neurosci. 33, 722–733 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  34. Adhikari, A., Topiwala, M.A. & Gordon, J.A. Single units in the medial prefrontal cortex with anxiety-related firing patterns are preferentially influenced by ventral hippocampal activity. Neuron 71, 898–910 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  35. Jin, J. & Maren, S. Fear renewal preferentially activates ventral hippocampal neurons projecting to both amygdala and prefrontal cortex in rats. Sci. Rep. 5, 8388 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  36. Killcross, S. & Coutureau, E. Coordination of actions and habits in the medial prefrontal cortex of rats. Cereb. Cortex 13, 400–408 (2003).

    PubMed  Google Scholar 

  37. Balleine, B.W. & O'Doherty, J.P. Human and rodent homologies in action control: corticostriatal determinants of goal-directed and habitual action. Neuropsychopharmacology 35, 48–69 (2010).

    PubMed  Google Scholar 

  38. Smith, K.S., Virkud, A., Deisseroth, K. & Graybiel, A.M. Reversible online control of habitual behavior by optogenetic perturbation of medial prefrontal cortex. Proc. Natl. Acad. Sci. USA 109, 18932–18937 (2012).

    CAS  PubMed  PubMed Central  Google Scholar 

  39. Valdés, J.L., Maldonado, P., Recabarren, M., Fuentes, R. & Torrealba, F. The infralimbic cortical area commands the behavioral and vegetative arousal during appetitive behavior in the rat. Eur. J. Neurosci. 23, 1352–1364 (2006).

    PubMed  Google Scholar 

  40. Keistler, C., Barker, J.M. & Taylor, J.R. Infralimbic prefrontal cortex interacts with nucleus accumbens shell to unmask expression of outcome-selective Pavlovian-to-instrumental transfer. Learn. Mem. 22, 509–513 (2015).

    PubMed  PubMed Central  Google Scholar 

  41. Corcoran, K.A. & Quirk, G.J. Activity in prelimbic cortex is necessary for the expression of learned, but not innate, fears. J. Neurosci. 27, 840–844 (2007).

    CAS  PubMed  PubMed Central  Google Scholar 

  42. Klavir, O., Genud-Gabai, R. & Paz, R. Functional connectivity between amygdala and cingulate cortex for adaptive aversive learning. Neuron 80, 1290–1300 (2013).

    CAS  PubMed  Google Scholar 

  43. Likhtik, E., Stujenske, J.M., Topiwala, M.A., Harris, A.Z. & Gordon, J.A. Prefrontal entrainment of amygdala activity signals safety in learned fear and innate anxiety. Nat. Neurosci. 17, 106–113 (2014).

    CAS  PubMed  Google Scholar 

  44. Stujenske, J.M., Likhtik, E., Topiwala, M.A. & Gordon, J.A. Fear and safety engage competing patterns of theta-gamma coupling in the basolateral amygdala. Neuron 83, 919–933 (2014).

    CAS  PubMed  PubMed Central  Google Scholar 

  45. Karalis, N. et al. 4-Hz oscillations synchronize prefrontal-amygdala circuits during fear behavior. Nat. Neurosci. 19, 605–612 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  46. Likhtik, E., Pelletier, J.G., Paz, R. & Paré, D. Prefrontal control of the amygdala. J. Neurosci. 25, 7429–7437 (2005).

    CAS  PubMed  PubMed Central  Google Scholar 

  47. Etkin, A., Egner, T., Peraza, D.M., Kandel, E.R. & Hirsch, J. Resolving emotional conflict: a role for the rostral anterior cingulate cortex in modulating activity in the amygdala. Neuron 51, 871–882 (2006).

    CAS  PubMed  Google Scholar 

  48. Comte, M. et al. Dissociating bottom-up and top-down mechanisms in the cortico-limbic system during emotion processing. Cereb. Cortex 26, 144–155 (2016).

    PubMed  Google Scholar 

  49. Rohr, C.S. et al. The neural networks of subjectively evaluated emotional conflicts. Hum. Brain Mapp. 37, 2234–2246 (2016).

    PubMed  PubMed Central  Google Scholar 

  50. Kanske, P. & Kotz, S.A. Emotion speeds up conflict resolution: a new role for the ventral anterior cingulate cortex? Cereb. Cortex 21, 911–919 (2011).

    PubMed  Google Scholar 

  51. Felix-Ortiz, A.C. et al. BLA to vHPC inputs modulate anxiety-related behaviors. Neuron 79, 658–664 (2013).

    CAS  PubMed  PubMed Central  Google Scholar 

  52. Kremer, E.J., Boutin, S., Chillon, M. & Danos, O. Canine adenovirus vectors: an alternative for adenovirus-mediated gene transfer. J. Virol. 74, 505–512 (2000).

    CAS  PubMed  PubMed Central  Google Scholar 

  53. Blanchard, R.J., Flannelly, K.J. & Blanchard, D.C. Defensive behavior of laboratory and wild Rattus norvegicus. J. Comp. Psychol. 100, 101–107 (1986).

    CAS  PubMed  Google Scholar 

  54. Ferguson, J.E., Boldt, C. & Redish, A.D. Creating low-impedance tetrodes by electroplating with additives. Sens. Actuators A Phys. 156, 388–393 (2009).

    CAS  PubMed  PubMed Central  Google Scholar 

  55. Paxinos, G. & Watson, C. The Rat Brain in Stereotaxic Coordinates (Academic, 1998).

  56. Meyer, R.M., Burgos-Robles, A., Liu, E., Correia, S.S. & Goosens, K.A. A ghrelin–growth hormone axis drives stress-induced vulnerability to enhanced fear. Mol. Psychiatry 19, 1284–1294 (2014).

    CAS  PubMed  Google Scholar 

  57. Do-Monte, F.H., Quiñones-Laracuente, K. & Quirk, G.J. A temporal shift in the circuits mediating retrieval of fear memory. Nature 519, 460–463 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  58. Felix-Ortiz, A.C., Burgos-Robles, A., Bhagat, N.D., Leppla, C.A. & Tye, K.M. Bidirectional modulation of anxiety-related and social behaviors by amygdala projections to the medial prefrontal cortex. Neuroscience 321, 197–209 (2016).

    CAS  PubMed  Google Scholar 

  59. Barthó, P. et al. Characterization of neocortical principal cells and interneurons by network interactions and extracellular features. J. Neurophysiol. 92, 600–608 (2004).

    PubMed  Google Scholar 

  60. Likhtik, E., Pelletier, J.G., Popescu, A.T. & Paré, D. Identification of basolateral amygdala projection cells and interneurons using extracellular recordings. J. Neurophysiol. 96, 3257–3265 (2006).

    PubMed  Google Scholar 

  61. Livneh, U. & Paz, R. Amygdala-prefrontal synchronization underlies resistance to extinction of aversive memories. Neuron 75, 133–142 (2012).

    CAS  PubMed  Google Scholar 

  62. Abeles, M. & Gerstein, G.L. Detecting spatiotemporal firing patterns among simultaneously recorded single neurons. J. Neurophysiol. 60, 909–924 (1988).

    CAS  PubMed  Google Scholar 

  63. Paz, R., Bauer, E.P. & Paré, D. Measuring correlations and interactions among four simultaneously recorded brain regions during learning. J. Neurophysiol. 101, 2507–2515 (2009).

    PubMed  PubMed Central  Google Scholar 

  64. Tye, K.M. & Janak, P.H. Amygdala neurons differentially encode motivation and reinforcement. J. Neurosci. 27, 3937–3945 (2007).

    CAS  PubMed  PubMed Central  Google Scholar 

  65. Herry, C. et al. Switching on and off fear by distinct neuronal circuits. Nature 454, 600–606 (2008).

    CAS  PubMed  Google Scholar 

  66. Johansen, J.P., Tarpley, J.W., LeDoux, J.E. & Blair, H.T. Neural substrates for expectation-modulated fear learning in the amygdala and periaqueductal gray. Nat. Neurosci. 13, 979–986 (2010).

    CAS  PubMed  PubMed Central  Google Scholar 

  67. Amano, T., Duvarci, S., Popa, D. & Paré, D. The fear circuit revisited: contributions of the basal amygdala nuclei to conditioned fear. J. Neurosci. 31, 15481–15489 (2011).

    CAS  PubMed  PubMed Central  Google Scholar 

  68. Pendyam, S. et al. Fear signaling in the prelimbic-amygdala circuit: a computational modeling and recording study. J. Neurophysiol. 110, 844–861 (2013).

    PubMed  PubMed Central  Google Scholar 

  69. Hastie, T., Tibshirani, R. & Friedman, J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction 2nd edn. (Springer, 2011).

  70. Narayanan, N.S. & Laubach, M. Methods for studying functional interactions among neuronal populations. Methods Mol. Biol. 489, 135–165 (2009).

    PubMed  Google Scholar 

  71. Richmond, B.J. & Optican, L.M. Temporal encoding of two-dimensional patterns by single units in primate inferior temporal cortex. II. Quantification of response waveform. J. Neurophysiol. 57, 147–161 (1987).

    CAS  PubMed  Google Scholar 

  72. Cristianini, N. & Shawe-Taylor, J. An Introduction to Support Vector Machines and Other Kernel-based Learning Methods (Cambridge Univ. Press, 2000).

  73. Schlkopf, B. & Smola, A.J. Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond (MIT Press, 2001).

  74. Good, P.I. Permutation, Parametric, and Bootstrap Tests of Hypotheses (Springer, 2005).

Download references

Acknowledgements

The authors thank S. Sorooshyari for discussion and assistance on Matlab programming, as well as C. Wildes, and the entire Tye laboratory for discussion and support on this project. A.B.-R. was supported by the Brain and Behavior Research Foundation (NARSAD Young Investigator Award) and NIMH (Research Supplement to Promote Diversity in Health-Related Sciences). E.Y.K. was supported by the Collaborative Clinical Neuroscience Fellowship and the American Brain Foundation Clinical Research Training Fellowship. W.A.R.-G. and P.A.P.-R. were supported by the MIT Summer Research Program, which received support from the Center for Brains, Minds and Machines (CBMM), NSF (STC Award CCF-1231216) and NIH (Endure Award 1R25-MH092912-01). E.H.N. was supported by the National Science Foundation Graduate Research Fellowship (NSF GRFP), the Integrative Neuronal Systems Training Fellowship (T32 GM007484) and the Training Program in the Neurobiology of Learning and Memory. A.C.F.-O. was supported by an institutional NRSA training grant (5T32GM007484-38). P.N. was supported by the Singleton, Leventhal and Whitaker fellowships. C.A.L. was supported by an NSF Graduate Research Fellowship, an Integrative Neuronal Systems Fellowship and the James R. Killian Fellowship. M.J.P., K.N.P. and M.A. were supported by the MIT Undergraduate Research Opportunities Program. K.K.A. was supported by the MIT Research Assistantship Program. K.M.T. is a New York Stem Cell Foundation - Robertson Investigator and McKnight Scholar, and this work was supported by funding from the JPB Foundation, PIIF, PNDRF, JFDP, Whitehall Foundation, Klingenstein Foundation, NARSAD Young Investigator Award, Alfred P Sloan Foundation, New York Stem Cell Foundation, McKnight Foundation, NIH R01-MH102441-01 (NIMH), R01-AA023305-01 (NIAAA) and NIH Director's New Innovator Award DP2-DK-102256-01 (NIDDK).

Author information

Authors and Affiliations

Authors

Contributions

A.B.-R. and K.M.T. conceived and designed experiments. A.B.-R. and K.N.P. designed and constructed electrodes and optrodes for neural recordings. A.B.-R. performed surgeries to chronically implant electrodes and performed single-unit recordings. A.B.-R., C.A.L. and K.N.P. sorted extracellular waveforms. A.B.-R. and E.Y.K. analyzed electrophysiology data. E.Y.K. wrote the Matlab scripts for the support vector machine learning algorithms. E.M.I., M.J.P., K.N.P., K.K.A., P.A.P.-R. and M.A. built optical fibers. A.B.-R., E.M.I., M.J.P., W.A.R.-G., K.N.P. and M.A. performed animal training and analyzed behaviors from videos. A.B.-R., E.M.I., M.J.P., W.A.R.-G., A.C.F.-O., K.N.P., K.K.A. and M.A. performed histological assessment. E.H.N. assisted with programing of the neural recording workstation and wrote the Matlab script for quantification of animal movement. P.N. assisted with Med-PC programming for behavioral studies and wrote the Matlab scripts to analyze port entry data and waveform properties. A.B. performed ex vivo whole-cell patch-clamp electrophysiological recordings. A.B. and A.C.F.-O. assisted with figures. A.B.-R., E.Y.K. and K.M.T. made figures and wrote the manuscript. All authors contributed to the editing and revision of the final version of the manuscript.

Corresponding author

Correspondence to Kay M Tye.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Progression of port entry and freezing responses across training.

Related to Figure 1. (a-b) Training timeline and behavioral apparatus. Sensory modalities for the sucrose- and shock-predictive cues (conditioned stimuli, CSs) were counterbalanced across animals. During competition trials, the CS-Suc and CS-Shock associations were co-presented to induce conflicting motivational drives and competition between the reward- and fear-related behaviors. (c) Progression of port entry responses during reward conditioning, plotted in blocks of five trials (repeated measures two-way ANOVA with Bonferroni post-hoc tests, epoch x trial-block interaction, F14,448 = 5.36, P < 0.001; t14 > 4.16, ***P < 0.001). (d) Progression of port entry and freezing responses during the discrimination and competition sessions, plotted as session-blocks (trial-type x training-session interactions; port entry during discrimination, F2,60 = 21.4, P < 0.001; freezing during discrimination, F2,60 = 23.4, P < 0.001; port entry during competition, F1,30 = 1.68, P = 0.20; freezing during competition, F1,30 = 7.37, P = 0.011; Bonferroni post-hoc tests, t14 > 3.89, ***P < 0.001). (e-g) Effects of cue modality on behavioral performance. Superimposed dots represent individual subjects. No significant statistical differences were detected between animals that were trained with the light cue for the reward association and the tone cue for the fear association (RLight FTone: n = 8 animals), or vice versa (RTone FLight: n = 8 animals) (Reward Conditioning: two-way repeated measures ANOVA, group x training-session interaction, F2,28 = 0.57, P = 0.57) (Discrimination Session: Bonferroni-corrected unpaired T-Tests, all t14 < 1.71, all P > 0.11) (Competition Session: Bonferroni-corrected unpaired T-Tests, all t14 < 1.61, all P > 0.13). (h) Purity of behavioral responses across trial types during the last competition session. Dots represent individual animals. Values closer to “1” indicate that animals tended to solely perform a behavioral response type (either port entry or freezing) during each trial (see inset heatmap on the left). Values away from “1” indicate that animals tended to perform behavioral transitions (from freezing to port entry, or vice versa) within single trials (see inset heatmap on the right). (i) Effects of previous trial on behavioral output during competition trials. Connecting lines represent individual subjects. No significant differences were detected (repeated measures one-way ANOVA: port entries, F2,15 = 1.81, P = 0.18; freezing, F2,15 = 1.56, P = 0.23). (j) Latency of port entry and freezing responses during competition. Superimposed dots represent latencies for individual subjects. Latencies were capped at 20 s, which was the maximum trial length. Significant differences were detected across trial types for port entry latencies (repeated measures one-way ANOVA with Bonferroni post-hoc tests, F2,47 = 99.7, P < 0.001; all t15 > 5.89, all ***P < 0.001) and freezing latencies (F2,47 = 64.6, P < 0.001; all t15 > 4.29, all ***P < 0.001). In all data panels, error bars represent s.e.m.

Supplementary Figure 2 Histological reconstruction of neural recording sites in the BLA and PL.

Related to Figures 2, 3, 4, and 5. (a-c) Electrodes were chronically implanted in both the BLA and PL for simultaneous single-unit recordings. Error bands for the representative waveforms represent s.d. Circles of matching colors in the coronal drawings correspond to the same animal. (d) Representative confocal images show the recording sites for one animal. (e) Representative waveforms and clusters from well-isolated cells. Waveforms are illustrated in a superimposed manner for each cluster. Clusters are shown in three-dimensional principal component space. (f-g) Distribution of firing frequencies for the BLA and PL cells. Red lines indicate the cutoffs used to exclude low firing rate cells (<0.1 Hz) from the cross-correlation analyses, as cells with such firing rates typically produce unpopulated correlograms with spurious peaks that may result on false positive correlations.

Supplementary Figure 3 Criteria used for the detection of significant correlations between the BLA and PL cells, and side-by-side results with multiple time bin sizes.

Related to Figure 2. (a) Cross-correlograms were calculated using 25-ms time bins, they were then corrected using two different predictor methods (trial shifting and spike shuffling) to eliminate confounds produced by CS-elicited changes in firing frequency, and they were then z-score transformed to detect significant peaks or troughs. (a1) Correction by shifting trials (repeated 19 times per trial type). (a2) Correction by shuffling spike trains (repeated 100 times per trial type). (a3) Repetition of the analysis using 5-ms time bins to detect and exclude “zero-lag correlations” (i.e., peaks or troughs that are centered at zero, ±2.5 ms), which typically result from common input-induced comodulation. The resulting correlograms from the shift and shuffle predictors (second column) were subtracted from the raw correlograms (first column), and the difference was z-score transformed (third column; zoomed in on the fourth column) to detect significant peaks or troughs within the time window of interest (±100 ms). Peaks or troughs were required to reach a z-value threshold (indicated by dashed horizontal lines) that was calculated based on a P-level of <0.01 (corrected for two-tail comparisons and Bonferroni-corrected for multiple comparisons). Significant peaks in these representative correlograms are indicated with colored dots for each trial type. To classify neural pairs as significantly correlated, peaks or troughs were required to exceed the significance threshold after correcting with both predictor correctors (trial-shifting and spike shuffling), and do not show "zero-lag" latencies in the 5-ms binned correlograms. In addition, neural pairs were classified as either “BLA led” if peaks or troughs occurred within the range of +2.5 to +100 ms, or as “PL led” if peaks or troughs occurred within the range of -100 to -2.5 ms, relative to the BLA reference spikes. (b-c) Side-by-side cross-correlation results using either 25-ms or 10-ms bin widths. Numbers within the heatmaps represent the proportion of significantly correlated BLA/PL neural pairs during the CS-Suc and CS-Shock trials. Inset heatmaps show zoomed in views of the time window of interest (±100 ms). (d-e) Lead and lag results using either 25 ms or 10 ms bins. With both bin widths, excitatory correlations were likelier to be led by the BLA during the CS-Shock trials, whereas inhibitory correlations were likelier to be led by the BLA during the CS-Suc trials. Chi-square tests for the 25-ms binned excitatory CCs: X2 = 29.8 and ***P < 0.001 (BLA vs PL during the CS-Shock), X2 = 16.0 and ***P < 0.001 (CS-Suc vs CS-Shock ratios). Chi-square tests for the 25-ms binned inhibitory CCs: X2 = 5.99 and *P = 0.014 (BLA vs PL during the CS-Suc), X2 = 5.70 and *P = 0.017 (CS-Suc vs CS-Shock ratios). Chi-square tests for the 10-ms binned excitatory CCs: X2 = 21.0 and ***P < 0.001 (BLA vs PL during the CS-Shock), X2 = 14.4 and ***P < 0.001 (CS-Suc vs CS-Shock ratios). Chi-square tests for the 10-ms binned inhibitory CCs: X2 = 0.67 and P = 0.41 (BLA vs PL during the CS-Suc), X2 = 1.18 and P = 0.28 (CS-Suc vs CS-Shock ratios). (f-g) Mean latency of peaks and troughs. Boxes represent the median and the 25th to 75th percentiles, whiskers represent the 10th and 90th percentiles, and the plus signs (+) within the boxes represent the mean latencies per event. The mean latencies for the excitatory correlations with the 25-ms bins were as follow: CS-Suc, 5.9±2.5 ms; CS-Shock, 13.0±2.4 ms (unpaired T-test: t623 = 2.05, *P = 0.041). The mean latencies for the inhibitory correlations with the 25-ms bins were as follow: CS-Suc, 17.1±7.6 ms; CS-Shock, -1.4±5.2 ms (t95 = 1.95, ~P = 0.055). The mean latencies for the excitatory correlations with the 10-ms bins were as follow: CS-Suc, 2.0±2.7 ms; CS-Shock, 8.1±2.5 ms (t623 = 2.05, ~P = 0.064). The mean latencies for the inhibitory correlations with the 10-ms bins were as follow: CS-Suc, 9.5±15.0 ms; CS-Shock, 8.2±8.9 ms (t34 = 0.08, P = 0.94).

Supplementary Figure 4 Representative raw cross-correlations demonstrating potential increased variability with smaller time bin widths.

Related to Figure 2. Cross-correlations for each neural pair (individual columns) are represented in triplicate using three different bin widths: 25 ms (top row), 10 ms (middle row), and 5 ms (bottom row). Significant correlations are displayed as solid lines, and non-significant correlations are displayed as dotted lines. Significant peaks are indicated by colored dots. While these examples illustrate the y-axes as PL spike probability, the significance of each correlation was assessed using our compound criteria based on z-score thresholds after correcting with the trial-shifting and spike-shuffling predictors (as in Supplementary Fig. S3a). (a) Correlograms between representative BLA and PL cells with moderate firing frequencies that showed consistent peaks and lead/lag timings across all three bin widths. (b) Correlograms between representative BLA and PL cells with relatively low firing frequencies that showed peaks and lead/lag timings that were somewhat inconsistent across the distinct bin widths. (c-d) Example correlograms between representative BLA and PL cells with either low or moderate firing frequencies that exhibited spurious significant peaks when using the narrowest 5-ms bin widths, raising concerns for false positives due to the increased variability obtained with the small bins. (e-f) Example correlograms between representative BLA and PL cells both with relatively low firing frequencies that also showed spurious significant peaks with the smaller bin widths. These examples also raised concerns for false positives when using small bin widths due to sparse firing. (g) Example correlograms that represent the potential for false negatives with the 5-ms bins, which may be due to splitting of the peaks among adjacent bins, or due to the more stringent z-score criteria, which required additional corrections for multiple comparison in the case of the 5-ms bins compared to the wider bin widths. In summary, our dataset provided more reliable cross-correlation results when using the relatively wider bin widths (25 ms).

Supplementary Figure 5 BLA and PL exhibited less correlated activity during a habituation session prior to the acquisition of the reward and fear memories, as well as during a neutral cue after acquiring the memories.

Related to Figure 2. (a) On a subset of animals (n = 4), BLA/PL recordings were performed during an initial habituation session in which three cues (light, tone, and white noise) were presented without any outcome. These cues were then paired with either sucrose, shock, or no outcome to become the CS-Suc, CS-Shock, and neutral CS–, respectively. Recordings were then performed during a discrimination session after animals acquired these associations. (b-c) Behavioral responses during the discrimination session. While animals showed selective port entry and freezing responses to the CS-Suc and CS-Shock, respectively, they did not display any of these behaviors during the neutral CS–. Error bands represent s.e.m. (d) Proportion of BLA/PL neural pairs that exhibited significantly correlated activity during either the habituation or discrimination session. Significantly higher proportions of BLA/PL neural pairs exhibited correlated activity during the discrimination session than the habituation session during all epochs, except the CS– (Bonferroni-corrected chi-square tests comparing habituation versus discrimination: ITI, X2 = 21.6, ***P < 0.001; CS-Suc, X2 = 18.6, ***P < 0.001; CS-Shock, X2 = 13.0, **P = 0.0012; CS–, X2 = 2.91, P = 0.36). In addition, during the discrimination session there were fewer correlated neural pairs during the CS– than during the other epochs (Bonferroni-corrected chi-square tests: CS– vs ITI, X2 = 8.40, *P = 0.015; CS– vs CS-Suc, X2 = 3.7, P = 0.22; CS– vs CS-Shock, X2 = 3.11, P = 0.31). These findings support the hypothesis that BLA/PL correlations strengthened with learning.

Supplementary Figure 6 Variations on cross-correlation lead/lag across counterbalanced cue conditions, event-biased populations, and putative projection neurons and interneurons.

Related to Figure 2. (a-b) Lead/lag comparisons between animals that received the light cue for the rewarding CS-Suc association and the tone cue for the fearful CS-Shock association (“RewLight FearTone”, n = 6 subjects), or vice versa (“RewTone FearLight”, n = 6 subjects). On both counterbalanced combinations, the BLA was still likelier to lead the excitatory correlations during the CS-Shock (Bonferroni-corrected chi-square tests: RewLight FearTone; BLA vs PL during CS-Shock, X2 = 16.8, ***P < 0.001; CS-Shock vs ITI, X2 = 9.03, **P = 0.008; CS-Shock vs CS-Suc, X2 = 7.52, *P = 0.018; RewTone FearLight; BLA vs PL during CS-Shock, X2 = 13.1, ***P < 0.001; CS-Shock vs ITI, X2 = 2.88, P = 0.27; CS-Shock vs CS-Suc, X2 = 10.2, **P = 0.004). In addition, the BLA was still likelier to lead the inhibitory correlations during the CS-Suc on both cue combinations. However, statistical comparisons for the inhibitory correlations were unreliable due to the overall low numbers. (c-d) Lead/lag on distinct BLA/PL pairwise populations that exhibited correlations during either specific task events (“event-biased populations”) or during various task events (“unbiased populations”). While zero-lag “common-input” correlations were included in this analysis (represented in the bars as “0”), they were not considered for statistical comparisons due to the overall low numbers (Event-Biased BLA vs PL: X2 = 4.66, ~P = 0.087; Unbiased to ITI & CS-Shock BLA vs PL: X2 = 13.2, ***P < 0.001; Unbiased to All Events BLA vs PL: X2 = 13.4, ***P < 0.001; Unbiased to All Events CS-Shock vs ITI: X2 = 7.59, *P = 0.018; Unbiased to All Events CS-Shock vs CS-Suc: X2 = 9.66, **P = 0.006). Representative correlations in the line plots were constructed using 5-ms bins and were smoothed using a Gaussian distribution for illustration purposes. (e-i) Lead/lag comparisons across putative projection cells and interneurons. (e) BLA and PL cells were classified as putative projection cells or interneurons based on three properties: depolarization half-width, hyperpolarization half-width, and mean firing frequency. A hierarchical clustering method was used to separate cells into two populations: wide spike-width (putative projection cells; blue dots) or narrow spike-widths (putative interneurons; red dots). (f) Lead/lag results for the excitatory correlations after excluding putative interneurons. The BLA was still likelier to lead excitatory correlations during the CS-Shock, but not during the CS-Suc (BLA vs PL during CS-Shock: X2 = 27.7, ***P < 0.001; CS-Shock vs CS-Suc: X2 = 13.5, ***P < 0.001). (g) Proportion of BLA-led excitatory correlations across multiple combinations between the putative projection cells and interneurons. (h) Lead/lag results for the inhibitory correlations after excluding putative interneurons. The BLA was still likelier to lead more of the inhibitory correlations during the CS-Suc, but not during the CS-Shock (BLA vs PL during CS-Suc: X2 = 3.75, ~P = 0.053; CS-Shock vs CS-Suc: X2 = 2.92, ~P = 0.087). (i) Proportion of BLA-led inhibitory correlations across multiple combinations between the putative projection cells and interneurons.

Supplementary Figure 7 CS-responsive populations in the BLA and PL, and cross-correlations across responsive and nonresponsive cells during reward–fear discrimination.

Related to Figure 3. (a-b) Representative BLA and PL cells exhibiting significant changes in activity during the reward- or fear-related CSs. Cells were deemed as either “R+”, “R-”, “F+”, or “F-”, respectively, if they exhibited a selective increase or decrease in activity to either of the CSs. Cells that exhibited significant responses to CSs were deemed as “R+F+”, “R-F-”, “R+F-”, or “R-F+”, respectively. (c-d) Mean response for each CS-responsive population in the BLA and PL. Error bands represent s.e.m. (e-f) Response latencies per CS. Numbers within parentheses indicate the overall number of cells that responded to each CS, with either excitation (“Exc”; increased activity) or inhibition (“Inh”; decreased activity). Boxes represent the median and the 25th to 75th percentiles, whiskers represent the 10th and 90th percentiles, and the plus signs (+) within the boxes represent the mean response latencies, using 50-ms time bins. The mean latencies obtained for the BLA cells were as follow: excitation to the CS-Suc, 108±32; excitation to the CS-Shock, 107±21; inhibition to the CS-Suc, 152±18; inhibition to the CS-Shock, 129±24. The mean latencies obtained for the PL cells were as follow: excitation to the CS-Suc, 183±32; excitation to the CS-Shock, 145±23; inhibition to the CS-Suc, 165±31; inhibition to the CS-Shock, 200±42. No significant differences were detected with the Dunn’s post-hoc tests between the response latencies for the CS-Suc and CS-Shock. (g-j) Cross-correlation lead/lag comparisons among populations of BLA and PL cells that exhibited significant responses to the reward-related cue (“R cells”) or the fear-related cue (“F cells”).

Supplementary Figure 8 Histological assessment for the BLA→PL stimulation experiments and behavioral effects during a discrimination session with a neutral cue.

Related to Figure 6. (a) Histological reconstruction of viral infusions and location of optical fibers (eYFP: n = 10 animals; ChR2: n = 8 animals). Circles in the BLA drawings represent the center of viral infusions. Dashes in the PL drawings represent the optical fiber tips. (b) Histology for the pharmacology experiment (n = 10 animals). Dashes in the PL drawings represent the position of cannulas and optical fibers. (c) Photostimulation during the discrimination of a CS-Suc, CS-Shock, and a neutral cue (CS–). Animals were tested over two days shortly after infusion of either ACSF or NBQX+AP5. Difference scores are plotted relative to laser-OFF trials. A significant interaction between the drug and laser treatments was detected for freezing responses during CS-Shock trials (repeated measures two-way ANOVA with Bonferroni post-hoc tests: F1,18 = 7.57, P = 0.013; t18 = 3.89, ***P < 0.001). No significant interaction effects were detected for port entries during CS-Suc trials (F1,18 = 0.01, P = 0.97). No significant interaction effects were detected for either behavior during the CS– (freezing, F1,18 = 0.50, P = 0.49; port entry, F1,18 = 2.84, P = 0.11). Error bars represent s.e.m.

Supplementary Figure 9 Validation of the ArchT viral construct and effects of optogenetically mediated inhibition of BLA inputs on the spontaneous firing of PL neurons.

Related to Figure 7. (a) Single-unit activity was monitored in the BLA at several time points after viral infusion to determine whether the ArchT construct produced reliable silencing. In the coronal drawings, four recording sites are illustrated per animal (n = 4 subjects), as microlesions were performed on four representative channels along the circumference of the optical fiber. (b-c) BLA units exhibiting ArchT-induced inhibition or excitation. (d) Quantification of the BLA cells that exhibited significant inhibition (“-”), excitation (“+”), or no change (“No Δ”). ArchT-induced inhibition predominated at all recording time points (chi-square tests comparing inhibited and excited populations: 10 days, X2 = 9.07, **P = 0.003; 20 days, X2 = 19.4, ***P < 0.001; 30 days, X2 = 20.0, ***P < 0.001). (e-f) Population histograms for the inhibited and excited BLA cells. (g) Assessment of spontaneous activity in PL upon local ArchT-induced inhibition of BLA inputs. Coronal drawings show the recording sites in PL. (h) Quantification of the PL cells that exhibited significant changes in activity during inhibition of BLA inputs. Inhibition of BLA inputs produced sparse effects on the spontaneous activity of PL neurons (Fisher exact probability tests: 10 days, P = 1.00; 20 days, P = 0.50; 30 days, P = 0.25). Error bands represent s.e.m.

Supplementary Figure 10 Histological assessment for the BLA→PL inhibition experiments and behavioral effects during discrimination sessions with a neutral cue.

Related to Figure 7. (a) Histology for the optogenetic inhibition experiments (GFP: n = 6 animals; ArchT: n = 6 animals). (b) ArchT-mediated photoinhibition during the discrimination of a CS-Suc, CS-Shock, and a neutral cue (CS–). Difference scores are plotted relative to laser-OFF trials. A significant group x laser treatment interaction was detected for freezing responses during CS-Shock trials (repeated measures two-way ANOVA with Bonferroni post-hoc tests: F1,10 = 6.17, P = 0.032; t10 = 3.51, **P = 0.006). Significant interaction effects were also detected for port entries during CS-Suc trials (F1,10 = 6.86, P = 0.026; t10 = 3.70, **P = 0.004). No significant effects were detected for either behavior during the CS– (freezing, F1,10 = 0.65, P = 0.44; port entry, F1,10 = 0.06, P = 0.82). (c) Histology for the chemogenetic inhibition experiments (mCherry: n = 7 animals; M4D(Gi): n = 7 animals). CAV2-Cre infusions in PL were determined from needle tracks, and two infusion sites are illustrated per animal as the CAV2-Cre virus was infused into two dorsal-ventral PL locations to maximize tissue coverage. (d) Chemogenetic inhibition during the discrimination of a CS-Suc, CS-Shock, and CS–. Animals were tested over three days after systemic injections of either CNO or vehicle. Difference scores are plotted relative to the first vehicle session. A significant group x drug treatment interaction was detected for freezing during the CS-Shock (F2,24 = 6.22, P = 0.007; t12 = 4.32, ***P = 0.001). No significant interaction effects were detected for port entries during the CS-Suc (F2,24 = 2.25, P = 0.13). While a significant interaction effect was detected for freezing during the CS– (F2,24 = 4.95, P = 0.016; t12 = 3.00, *P = 0.011), this effect was not related to the CNO treatment (group difference during CNO, P = 0.58). Port entry responses during the CS– were also unaffected (F2,24 = 0.42, P = 0.66). Error bars represent s.e.m.

Supplementary information

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Burgos-Robles, A., Kimchi, E., Izadmehr, E. et al. Amygdala inputs to prefrontal cortex guide behavior amid conflicting cues of reward and punishment. Nat Neurosci 20, 824–835 (2017). https://doi.org/10.1038/nn.4553

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nn.4553

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