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:

A feedback neural circuit for calibrating aversive memory strength

Nature Neuroscience volume 20, pages 90–97 (2017)Cite this article

Subjects

Abstract

Aversive experiences powerfully regulate memory formation, and memory strength is proportional to the intensity of these experiences. Inhibition of the neural circuits that convey aversive signals when they are predicted by other sensory stimuli is hypothesized to set associative memory strength. However, the neural circuit mechanisms that produce this predictive inhibition to regulate memory formation are unknown. Here we show that predictive sensory cues recruit a descending feedback circuit from the central amygdala that activates a specific population of midbrain periaqueductal gray pain-modulatory neurons to control aversive memory strength. Optogenetic inhibition of this pathway disinhibited predicted aversive responses in lateral amygdala neurons, which store fear memories, resulting in the resetting of fear learning levels. These results reveal a control mechanism for calibrating learning signals to adaptively regulate the strength of behavioral learning. Dysregulation of this circuit could contribute to psychiatric disorders associated with heightened fear responsiveness.

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

Figure 1: Prediction-error signaling in LA neurons re-emerges with optogenetic inhibition of CeA inputs to vlPAG.
Figure 2: CeA–vlPAG pathway controls learning asymptotes by regulating prediction-error coding in LA.
Figure 3: Excitatory auditory-cue processing is reduced in vlPAG neurons with optical inhibition of the CeA–vlPAG pathway.
Figure 4: Auditory predictive cues recruit a CeA–vlPAG–RVM pathway to set behavioral learning asymptotes.

Similar content being viewed by others

References

  1. Knudsen, E.I. Supervised learning in the brain. J. Neurosci. 14, 3985–3997 (1994).

    Article  CAS  Google Scholar 

  2. Schultz, W. Neuronal reward and decision signals: from theories to data. Physiol. Rev. 95, 853–951 (2015).

    Article  CAS  Google Scholar 

  3. McNally, G.P., Johansen, J.P. & Blair, H.T. Placing prediction into the fear circuit. Trends Neurosci. 34, 283–292 (2011).

    Article  CAS  Google Scholar 

  4. Ohmae, S. & Medina, J.F. Climbing fibers encode a temporal-difference prediction error during cerebellar learning in mice. Nat. Neurosci. 18, 1798–1803 (2015).

    Article  CAS  Google Scholar 

  5. Herry, C. & Johansen, J.P. Encoding of fear learning and memory in distributed neuronal circuits. Nat. Neurosci. 17, 1644–1654 (2014).

    Article  CAS  Google Scholar 

  6. Kim, J.J., Krupa, D.J. & Thompson, R.F. Inhibitory cerebello-olivary projections and blocking effect in classical conditioning. Science 279, 570–573 (1998).

    Article  CAS  Google Scholar 

  7. Vallentin, D., Kosche, G., Lipkind, D. & Long, M.A. Neural circuits. Inhibition protects acquired song segments during vocal learning in zebra finches. Science 351, 267–271 (2016).

    Article  CAS  Google Scholar 

  8. Boyden, E.S., Katoh, A. & Raymond, J.L. Cerebellum-dependent learning: the role of multiple plasticity mechanisms. Annu. Rev. Neurosci. 27, 581–609 (2004).

    Article  CAS  Google Scholar 

  9. Rescorla, R.A. & Wagner, A.R. A theory of Pavlovian conditioning: variations in the effectiveness of reinforcement and nonreinforcement. in Classical Conditioning II: Current Research and Theory (eds. Prokasy, W.F. & Black, A.H.) 64–99 (Appleton-Century-Crofts, New York, 1972).

  10. Sutton, R.S. & Barto, A.G. Time-derivative models of Pavlovian reinforcement. in Learning and Computational Neuroscience: Foundations of Adaptive Networks (eds. Gabriel, M. & Moore, J.) 497–537 (MIT Press, Cambridge, MA, 1990).

  11. Fields, H.L. & Margolis, E.B. Understanding opioid reward. Trends Neurosci. 38, 217–225 (2015).

    Article  CAS  Google Scholar 

  12. Fanselow, M.S. Pavlovian conditioning, negative feedback, and blocking: mechanisms that regulate association formation. Neuron 20, 625–627 (1998).

    Article  CAS  Google Scholar 

  13. LeDoux, J.E. Coming to terms with fear. Proc. Natl. Acad. Sci. USA 111, 2871–2878 (2014).

    Article  CAS  Google Scholar 

  14. LeDoux, J.E. Emotion circuits in the brain. Annu. Rev. Neurosci. 23, 155–184 (2000).

    Article  CAS  Google Scholar 

  15. Duvarci, S. & Pare, D. Amygdala microcircuits controlling learned fear. Neuron 82, 966–980 (2014).

    Article  CAS  Google Scholar 

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

    Article  CAS  Google Scholar 

  17. Tovote, P., Fadok, J.P. & Lüthi, A. Neuronal circuits for fear and anxiety. Nat. Rev. Neurosci. 16, 317–331 (2015).

    Article  CAS  Google Scholar 

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

    Article  CAS  Google Scholar 

  19. Johansen, J.P. et al. Hebbian and neuromodulatory mechanisms interact to trigger associative memory formation. Proc. Nat. Acad. Sci. USA 111, E5584–E5592 (2014).

    Article  CAS  Google Scholar 

  20. Johansen, J.P. et al. Optical activation of lateral amygdala pyramidal cells instructs associative fear learning. Proc. Natl. Acad. Sci. USA 107, 12692–12697 (2010).

    Article  CAS  Google Scholar 

  21. Rosenkranz, J.A. & Grace, A.A. Dopamine-mediated modulation of odour-evoked amygdala potentials during Pavlovian conditioning. Nature 417, 282–287 (2002).

    Article  CAS  Google Scholar 

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

    Article  CAS  Google Scholar 

  23. Annau, Z. & Kamin, L.J. The conditioned emotional response as a function of intensity of the US. J. Comp. Physiol. Psychol. 54, 428–432 (1961).

    Article  CAS  Google Scholar 

  24. Young, S.L. & Fanselow, M.S. Associative regulation of Pavlovian fear conditioning: unconditional stimulus intensity, incentive shifts, and latent inhibition. J. Exp. Psychol. Anim. Behav. Process. 18, 400–413 (1992).

    Article  CAS  Google Scholar 

  25. 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).

    Article  CAS  Google Scholar 

  26. Belova, M.A., Paton, J.J., Morrison, S.E. & Salzman, C.D. Expectation modulates neural responses to pleasant and aversive stimuli in primate amygdala. Neuron 55, 970–984 (2007).

    Article  CAS  Google Scholar 

  27. Ciocchi, S. et al. Encoding of conditioned fear in central amygdala inhibitory circuits. Nature 468, 277–282 (2010).

    Article  CAS  Google Scholar 

  28. Duvarci, S., Popa, D. & Paré, D. Central amygdala activity during fear conditioning. J. Neurosci. 31, 289–294 (2011).

    Article  CAS  Google Scholar 

  29. Ghosh, S. & Chattarji, S. Neuronal encoding of the switch from specific to generalized fear. Nat. Neurosci. 18, 112–120 (2015).

    Article  CAS  Google Scholar 

  30. Fields, H.L. Pain modulation: expectation, opioid analgesia and virtual pain. Prog. Brain Res. 122, 245–253 (2000).

    Article  CAS  Google Scholar 

  31. Bellgowan, P.S. & Helmstetter, F.J. The role of mu and kappa opioid receptors within the periaqueductal gray in the expression of conditional hypoalgesia. Brain Res. 791, 83–89 (1998).

    Article  CAS  Google Scholar 

  32. Gross, C.T. & Canteras, N.S. The many paths to fear. Nat. Rev. Neurosci. 13, 651–658 (2012).

    Article  CAS  Google Scholar 

  33. Carrive, P. The periaqueductal gray and defensive behavior: functional representation and neuronal organization. Behav. Brain Res. 58, 27–47 (1993).

    Article  CAS  Google Scholar 

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

    Article  CAS  Google Scholar 

  35. Spellman, T. et al. Hippocampal-prefrontal input supports spatial encoding in working memory. Nature 522, 309–314 (2015).

    Article  CAS  Google Scholar 

  36. Di Scala, G., Mana, M.J., Jacobs, W.J. & Phillips, A.G. Evidence of Pavlovian conditioned fear following electrical stimulation of the periaqueductal grey in the rat. Physiol. Behav. 40, 55–63 (1987).

    Article  CAS  Google Scholar 

  37. Kim, E.J. et al. Dorsal periaqueductal gray-amygdala pathway conveys both innate and learned fear responses in rats. Proc. Natl. Acad. Sci. USA 110, 14795–14800 (2013).

    Article  CAS  Google Scholar 

  38. Han, S., Soleiman, M.T., Soden, M.E., Zweifel, L.S. & Palmiter, R.D. Elucidating an affective pain circuit that creates a threat memory. Cell 162, 363–374 (2015).

    Article  CAS  Google Scholar 

  39. Sato, M. et al. The lateral parabrachial nucleus is actively involved in the acquisition of fear memory in mice. Mol. Brain 8, 22 (2015).

    Article  Google Scholar 

  40. Li, H. et al. Experience-dependent modification of a central amygdala fear circuit. Nat. Neurosci. 16, 332–339 (2013).

    Article  CAS  Google Scholar 

  41. Tovote, P. et al. Midbrain circuits for defensive behaviour. Nature 534, 206–212 (2016).

    Article  CAS  Google Scholar 

  42. McNally, G.P. & Cole, S. Opioid receptors in the midbrain periaqueductal gray regulate prediction errors during pavlovian fear conditioning. Behav. Neurosci. 120, 313–323 (2006).

    Article  CAS  Google Scholar 

  43. Hnasko, T.S. et al. Cre recombinase-mediated restoration of nigrostriatal dopamine in dopamine-deficient mice reverses hypophagia and bradykinesia. Proc. Natl. Acad. Sci. USA 103, 8858–8863 (2006).

    Article  CAS  Google Scholar 

  44. Koutsikou, S. et al. The periaqueductal gray orchestrates sensory and motor circuits at multiple levels of the neuraxis. J. Neurosci. 35, 14132–14147 (2015).

    Article  CAS  Google Scholar 

  45. Bakal, C.W., Johnson, R.D. & Rescorla, R.A. The effect of change in US quality on the blocking effect. Pavlov. J. Biol. Sci. 9, 97–103 (1974).

    CAS  PubMed  Google Scholar 

  46. Eshel, N. et al. Arithmetic and local circuitry underlying dopamine prediction errors. Nature 525, 243–246 (2015).

    Article  CAS  Google Scholar 

  47. Tian, J. et al. Distributed and Mixed Information in Monosynaptic Inputs to Dopamine Neurons. Neuron 91, 1374–1389 (2016).

    Article  CAS  Google Scholar 

  48. Silva, B.A. et al. The ventromedial hypothalamus mediates predator fear memory. Eur. J. Neurosci. 43, 1431–1439 (2016).

    Article  Google Scholar 

  49. Reger, M.A., Gahm, G.A., Swanson, R.D. & Duma, S.J. Association between number of deployments to Iraq and mental health screening outcomes in US Army soldiers. J. Clin. Psychiatry 70, 1266–1272 (2009).

    Article  Google Scholar 

  50. Roy, M. et al. Representation of aversive prediction errors in the human periaqueductal gray. Nat. Neurosci. 17, 1607–1612 (2014).

    Article  CAS  Google Scholar 

  51. Callaway, E.M. & Luo, L. Monosynaptic circuit tracing with glycoprotein-deleted rabies viruses. J. Neurosci. 35, 8979–8985 (2015).

    Article  CAS  Google Scholar 

  52. Blair, H.T., Sotres-Bayon, F., Moita, M.A. & Ledoux, J.E. The lateral amygdala processes the value of conditioned and unconditioned aversive stimuli. Neuroscience 133, 561–569 (2005).

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank A. Benucci, T. McHugh, C. Yokoyama and H. Fields for comments on earlier versions of the manuscript and E. Boyden (Massachusetts Institute of Technology) and K. Deisseroth (Stanford University) for making available the optogenetic constructs used in these studies. This work was supported by funding from the Kao Corporation and Grants-in-Aid for Scientific Research on Innovative Areas (25116531, 15H01301, 16H01291), Wakate A (25710003) and Kiban B (15H04264) to J.P.J., as well as Wakate B (26830023) to T.O.

Author information

Author notes

  1. Takaaki Ozawa

    Present address: Present address: Department of Basic Neurosciences, University of Geneva, Geneva, Switzerland.,

  2. Edgar A Ycu, Ashwani Kumar and Li-Feng Yeh: These authors contributed equally to this work.

Authors and Affiliations

  1. RIKEN Brain Science Institute, Laboratory for Neural Circuitry of Memory, Saitama, Japan

    Takaaki Ozawa, Edgar A Ycu, Ashwani Kumar, Li-Feng Yeh, Jenny Koivumaa & Joshua P Johansen

  2. Electrical and Instrumentation Engineering Department, Sant Longowal Institute of Engineering and Technology (SLIET), Sangrur, India

    Ashwani Kumar

  3. Department of Life Sciences, Graduate School of Arts and Sciences, University of Tokyo, Tokyo, Japan

    Li-Feng Yeh & Joshua P Johansen

  4. National University of Sciences and Technology, Atta-ur-Rahman School of Applied Biosciences, Islamabad, Pakistan

    Touqeer Ahmed

Authors
  1. Takaaki Ozawa
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Edgar A Ycu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ashwani Kumar
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Li-Feng Yeh
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Touqeer Ahmed
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Jenny Koivumaa
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Joshua P Johansen
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

T.O. and J.P.J. designed the experiments and wrote up the manuscript. T.O., E.A.Y., T.A., L.-F.Y. and J.K. carried out the experiments. T.O., A.K. and J.P.J. analyzed the results.

Corresponding author

Correspondence to Joshua P Johansen.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Behavioral-learning asymptote pilot study for electrophysiology studies

Top, Experimental design. Animals were given 12 CS-US pairings with 2mA eyelid shock intensity on Day 1, a single CS presentation on Day 2 (Test 1), 12 more CS-US pairings on Day 3 (Training 2) with some animals receiving the same shock intensity (‘Same’ group, n = 7) and the ‘Stronger’ group (n = 8) receiving a larger shock intensity (5mA). This was followed by another single CS presentation on Day 4 (Test 2). Bar graph (bottom) shows no change in behavioral freezing to the CS from Test 1 to Test 2 when shock intensity was the same during Training 1 and 2 (‘Same’ group), but an increase in CS-evoked freezing from Test 1 to 2 in the ‘Stronger’ group (A two-way repeated-measures ANOVA revealed significant interaction, F(1,13) = 4.804, p = 0.047 and post hoc Bonferroni tests revealed no difference between test 1 and 2, p = 0.929, in the ‘Same’ group, but a significant increase in freezing in the ‘Stronger’ group, p = 0.029).

Supplementary Figure 2 Histological reconstruction of recording or laser illumination sites for electrophysiology and behavioral experiments

(a & b) Locations of recording sites (red dots) in LA or PAG and optical fiber (green dots or bars) in PAG during in which unpredicted and predicted shocks as well as predicted shocks concurrent with laser inactivation of CeA-vlPAG terminals were delivered. (c) Locations of optical fiber tips to deliver laser into dl/lPAG or vlPAG in each behavioral experiment (green dots; inactivation of terminal from CeA, yellow; inactivation of CeA terminal combined with MUS injections into LA, orange; inactivation of RVM projecting cells in vlPAG). (d) Locations of injection cannula tips to deliver drug into LA combined with (orange, in c) or without (purple) optogenetic inactivation of CeA terminal in vlPAG.

Supplementary Figure 3 Prediction-error coding in LA neurons re-emerges with optogenetic inhibition of CeA inputs to vlPAG

(a & b) Maximum Z-score averaged population response graphs with Unpredicted (navy), Predicted (blue) and Predicted+Laser (orange) bars for prediction error (PE) coding (a) or non-PE-coding (b) cell populations. For PE coding cells (a) one-way repeated measures ANOVA analysis showed a significant effect of group (F(2,72) = 16.85, p = 0.0001). * denotes significant (p<0.05) differences using Bonferroni post hoc tests. No change was observed in the non-PE-coding (b) cell population (F(2,166) = 1.274, p = 0.282). (c) Perievent time histograms (50 ms bins) showing Z-score averaged responses for all LA non-PE-coding cells ‘Unpredicted’ (left panel), ‘Predicted’ (middle panel) and ‘Predicted+laser’ shock periods are shown in gray area (1 sec). (d) Frequency histogram showing number of total cells (y-axis) and baseline firing rates in 1 Hz bins. Blue bar heights show total cell counts and orange bar heights show total PE-coding cells in specific firing rate time bins. Based on firing rate based electrophysiological criteria established in previous work1,2 we found that PE coding occurs in both pyramidal (≤ 1Hz) and fast spiking interneurons (≥ 7Hz) replicating prior results (main ref. 25). (e) Z-score averaged perievent time histogram (bin=50 ms) for all LA neurons recorded (n=180) triggered by laser onset (which occurs 400 ms prior to CS onset in experiments shown in Fig. 3) for ‘Unpredicted’, ‘Predicted’ and ‘Predicted+Laser’ trials. Note, laser was on only in the ‘Predicted+Laser’ trials. No significant changes were detectable at the population level (a one-way repeated measures ANOVA, F(2,166) =1.968, p = 0.1429).

1. Likhtik, E., Pelletier, J.G., Popescu, A.T. & Pare, D. Identification of basolateral amygdala projection cells and interneurons using extracellular recordings. J Neurophysiol 96, 3257-65 (2006).

2. Wolff, S.B. et al. Amygdala interneuron subtypes control fear learning through disinhibition. Nature 509, 453-8 (2014).

Supplementary Figure 4 Prediction-error signaling in dl/lPAG neurons re-emerges with optogenetic inhibition of CeA inputs to vlPAG

(a) Schematic of experimental paradigm showing virus injection in CeA as well as electrophysiological recording and optical stimulation sites in PAG. (b-d) Example cell and population averaged perievent time histograms (PETH, 50 ms bins) of prediction error coding cells in PAG showing disinhibition of shock-evoked responding in dl/lPAG neurons when CeA-vlPAG pathway was optically inhibited. PETHs show firing rate of an example cell (b) or Z-score averaged response (c-d) on y-axis during (gray area, 1 sec) ‘Unpredicted’, ‘Predicted’ and ‘Predicted+light’. For (b), perievent raster plots are above PETHs and insets show average waveforms on two stereotrode channels during the shock period of the different trial types (total time on y-axis is 0.2 mV and on x-axis is 1 ms). PETH in (c) shows population averaged response in dl/lPAG prediction error coding neurons (n=35 out of 108 total shock responsive cells from 3 animals) with significantly (see statistics in e) larger shock-evoked response in the ‘Predicted+laser’ condition compared with cell responses in the ‘Predicted’ condition while no significant differences (see statistics in f) exist in dl/lPAG non-PE-coding neurons (d, n = 73 out of 108 total shock responsive cells.(e & f) Maximum Z-score averaged population response graphs with Unpredicted (purple), Predicted (blue) and Predicted+Laser (orange) bars for prediction error (PE) coding (e) or non-PE-coding (f) dl/lPAG cell populations. For PE coding cells (e) a one-way repeated measures ANOVA analysis showed a significant group differences (F(2,68) = 13.44, p = 0.001) and * denotes significant (p<0.05) differences using Bonferroni post hoc tests. No change was observed in the non-PE-coding (f) cell population (F(2,114) = 2.71, p = 0.118). (g) Circle plots showing proportion of prediction error coding dl/lPAG cells which were shock responsive or non-responsive in ‘Unpredicted’, ‘Predicted’ and ‘Predicted+Laser’ conditions. Chi-squared test with Bonferroni correction (α=0.0166) revealed a significantly higher proportion of shock responsive cells in the unpredicted vs. predicted conditions (χ2 (1) = 7.467, p = 0.006). There was a non-significant trend toward a higher proportion of shock responsive cells in the predicted+laser vs. predicted conditions (χ2 (1) = 2.362, p = 0.124), mainly because a small shock response remained in many cells resulting in a relatively large proportion of cells being classified as significantly shock responsive in the predicted condition despite their being reduced compared with the predicted condition.

Supplementary Figure 5 Establishment of a behavioral-learning asymptote paradigm to examine effects of CeA–vlPAG manipulations on behavior

(a) Top, Experimental design. Animals are given 8 CS-US pairings with 0.3 mA foot shock intensity on Day 1, a single CS presentation on Day 2 (Test 1), 8 more CS-US pairings on Day 3 (Training 2) with some animals receiving the same shock intensity (‘Same’ group, n=8) and another receiving a larger shock intensity (‘Stronger’, 0.8 mA, n=8) followed by another single CS presentation on Day 4 (Test 2) (‘Stronger’ group). Bar graph (bottom) shows no change in behavioral freezing to the CS from Test 1 to Test 2 when shock intensity was the same during Training 1 and 2 (‘Same’ group) but a significant increase in freezing when the shock intensity was increase during Training 2 (‘Stronger’ group)(A two-way repeated-measures ANOVA revealed significant interaction, F(1,14) = 4.711, p = 0.048, Bonferroni post hoc test revealed no difference between test 1 and 2, p = 0.499 in the ‘Same’ group, but a significant increase in freezing in the Stronger group, p = 0.032). (b) Control experiment to test whether a single CS presentation on Test 1 had an effect on learning asymptotes. Same as a, but in one group (‘Single test’) Test 1 was omitted and shock intensity was held constant on Training 2 (n = 8 and 7, respectively). No differences were seen between Test 2 freezing (black bar) and the Single CS test 2 (grey bar) (A two-tailed unpaired t-test, t(13) = 0.3090, p = 0.762) revealed that learning asymptote had been achieved after Training day 1 and the CS presentation on Test 1 had no effect on that.

Supplementary Figure 6 Offset CeA–dl/lPAG terminal inactivation, CeA–vlPAG terminal inactivation effects on single-trial learning and muscimol effects on learning asymptotes

(a) Schematic of experimental paradigm showing optogenetic manipulation of CeA-dl/lPAG pathway specifically during overtraining/learning asymptote. (b) Inhibition of CeA-dl/lPAG pathway during auditory CS and shock US periods of overtraining (n = 10) does not change freezing levels (y-axis) from Test 1 compared with Test 2 (Two-tailed paired t-test, t(9) = 0.041, p=0.9682) (c) Top, Schematic of experimental paradigm showing optogenetic inhibition of CeA-vlPAG pathway during auditory CS and strong shock US (0.8mA) periods of single trial (1 CS-US pairing) fear conditioning. Bottom, inhibition of CeA-vlPAG pathway has no effect on single trial learning as measured 24 hours following learning and optogenetic manipulation comparing ‘Offset” to ‘Overlap’ group (n = 7 and n = 7 respectively) (Two-tailed paired t-test, t( 12) = 0.5571, p = 0.588). (d) Muscimol alone in LA has no effect on learning asymptotes. Top, Experimental design. Animals are given 8 CS-US pairings with 0.3 mA foot shock intensity on Day 1, a single CS presentation on Day 2, 8 more CS-US pairings on Day 3 (Test 1) at the same shock intensity and 1 more CS presentation on Day 4 (Test 2). Prior to Training 2 animals received microinjections of phosphate buffered saline (PBS) as a control or the GABAA agonist Muscimol to inactivate LA neurons. Bar graph (bottom) shows no change (Two-way repeated measures ANOVA, F(1,14) = 0.0567, p = 0.815) in behavioral freezing to the CS from Test 1 (white bar) to Test 2 (black bar) in PBS and Muscimol treated animals.

Supplementary Figure 7 CeA–vlPAG trans-synaptic rabies virus tracing, effect of CeA–vlPAG inactivation on auditory responding in dl/lPAG cells and baseline firing rates in PAG neurons, and effect of inactivation of RVM-projecting vlPAG neurons on single-trial fear learning

(a) Schematic showing injection of rabies virus (SADΔG-mCherry) along with 3 different cre-dependent (FLEX) AAVs expressing TVA, glycoprotein (RG), eYFP and a ubiquitously expressed AAV-cre-recombinase. Bottom right inset shows starter cells in vlPAG coexpressing mCherry (red) and eYFP (green) and the overlay. Scale bar=10μm. Top right inset showing retrogradely labeled, rabies virus infected mCherry expressing cells in CeA. Scale bar=200μm (b) Z-score averaged perievent time histograms (bin size=50 ms) for vlPAG neurons (n=147) triggered by laser onset (which occurs 400 ms prior to CS onset in experiments shown in Fig. 1 & 2) for ‘Unpredicted’, ‘Predicted’ and ‘Predicted+Laser’ trials. Note, laser was on only in the ‘Predicted+Laser’ trials. No significant changes were detected (One-way repeated measures ANOVA, F(2,292) = 0.557, p = 0.574). (c) PETH showing population averaged firing rates (y-axis) of all CS-responsive cells in dl/lPAG (n=50 out of 113 total cells) during the auditory CS period (20 sec starting at ‘0’, x-axis) without (blue trace) or with (orange trace) light induced inhibition of CeA-vlPAG pathway. Auditory CS (blue bar) and laser on (orange bar) periods are denoted above PETH. A two-tailed paired t-test revealed no effect of terminal inhibition on CS-evoked excitatory response in LA (t( 64) = 0.0508, p = 0.96). (d) Circle plots showing proportion of dl/lPAG cells which were significantly auditory CS-responsive (blue) or non-responsive (grey) with and without laser inactivation of the CeA-vlPAG pathway. Chi-squared test revealed no significant difference in the proportion of CS-responsive cells between with and without laser conditions (χ2 (1) = 2.918, p = 0.0876) (e) Top, Schematic of experimental paradigm showing optogenetic inhibition of RVM projecting vlPAG neurons during auditory CS and strong shock US (0.8 mA) periods of single trial (1 CS-US pairing) fear conditioning. Bottom, inhibition of RVM projecting vlPAG neurons had no effect on single trial learning as measured 24 hours following learning and optogenetic manipulation comparing ‘Offset” to ‘Overlap’ group (n = 8 and n = 8 respectively)((Two-tailed paired t-test, t(14) = 0.3728, p = 0.715).

Supplementary Figure 8 RVM-projecting vlPAG neurons are mostly glutamatergic while a smaller subpopulation is GABAergic

Retrograde tracer (CTb Alexa 647) injections were made into the RVM and tissues were immunostained for vesicular glutamate transporter-2 (vGlut2) or GABA. (a) Top images (‘CTb+/vGlut2+’) show examples of vGlut2+ cells colabeled by CTb. Note vGlut2 fiber label labeling along with clear soma labeling. Bottom ‘CTb+’ images show CTb labeled cell without vGlut2 cell body labeling. Pie-chart shows proportion of double labeled cells (vGlut2+/CTb+) in all CTb+ cells counted (n = 528, from 3 animals). (b) Top images (‘CTb+/GABA+’) show examples of GABA+ colabeled by CTb+. Bottom ‘CTb’ images show CTb labeled cells without GABA labeling. Pie-chart shows proportion of double labeled cells (GABA+/CTb+) in all CTb+ cells counted (n = 1575, from 3 animals). Scale bar = 10μm.

Supplementary information

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ozawa, T., Ycu, E., Kumar, A. et al. A feedback neural circuit for calibrating aversive memory strength. Nat Neurosci 20, 90–97 (2017). https://doi.org/10.1038/nn.4439

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nn.4439

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

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

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