Memory is coded by patterns of neural activity in distinct circuits. Therefore, it should be possible to reverse engineer a memory by artificially creating these patterns of activity in the absence of a sensory experience. In olfactory conditioning, an odor conditioned stimulus (CS) is paired with an unconditioned stimulus (US; for example, a footshock), and the resulting CS–US association guides future behavior. Here we replaced the odor CS with optogenetic stimulation of a specific olfactory glomerulus and the US with optogenetic stimulation of distinct inputs into the ventral tegmental area that mediate either aversion or reward. In doing so, we created a fully artificial memory in mice. Similarly to a natural memory, this artificial memory depended on CS–US contingency during training, and the conditioned response was specific to the CS and reflected the US valence. Moreover, both real and implanted memories engaged overlapping brain circuits and depended on basolateral amygdala activity for expression.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Data availability

All data supporting the findings of this study are available from the corresponding author upon request.

Code availability

Custom code for behavioral analysis is available from the corresponding author upon request.

Additional information

Journal peer review information: Nature Neuroscience thanks Mark Baxter and other anonymous reviewer(s) for their contribution to the peer review of this work.

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


  1. 1.

    Eichenbaum, H. Still searching for the engram. Learn. Behav. 44, 209–222 (2016).

  2. 2.

    Josselyn, S. A., Kohler, S. & Frankland, P. W. Finding the engram. Nat. Rev. Neurosci. 16, 521–534 (2015).

  3. 3.

    Tonegawa, S., Liu, X., Ramirez, S. & Redondo, R. Memory engram cells have come of age. Neuron 87, 918–931 (2015).

  4. 4.

    Martin, S. J. & Morris, R. G. New life in an old idea: the synaptic plasticity and memory hypothesis revisited. Hippocampus 12, 609–636 (2002).

  5. 5.

    Buck, L. & Axel, R. A novel multigene family may encode odorant receptors: a molecular basis for odor recognition. Cell 65, 175–187 (1991).

  6. 6.

    Buck, L. B. The molecular architecture of odor and pheromone sensing in mammals. Cell 100, 611–618 (2000).

  7. 7.

    Firestein, S. How the olfactory system makes sense of scents. Nature 413, 211–218 (2001).

  8. 8.

    Mombaerts, P. et al. Visualizing an olfactory sensory map. Cell 87, 675–686 (1996).

  9. 9.

    Ressler, K. J., Sullivan, S. L. & Buck, L. B. Information coding in the olfactory system: evidence for a stereotyped and highly organized epitope map in the olfactory bulb. Cell 79, 1245–1255 (1994).

  10. 10.

    Wang, F., Nemes, A., Mendelsohn, M. & Axel, R. Odorant receptors govern the formation of a precise topographic map. Cell 93, 47–60 (1998).

  11. 11.

    Jones, S. V., Choi, D. C., Davis, M. & Ressler, K. J. Learning-dependent structural plasticity in the adult olfactory pathway. J. Neurosci. 28, 13106–13111 (2008).

  12. 12.

    Morrison, F. G., Dias, B. G. & Ressler, K. J. Extinction reverses olfactory fear-conditioned increases in neuron number and glomerular size. Proc. Natl Acad. Sci. USA 112, 12846–12851 (2015).

  13. 13.

    Jiang, Y. et al. Molecular profiling of activated olfactory neurons identifies odorant receptors for odors in vivo. Nat. Neurosci. 18, 1446–1454 (2015).

  14. 14.

    Smear, M., Resulaj, A., Zhang, J., Bozza, T. & Rinberg, D. Multiple perceptible signals from a single olfactory glomerulus. Nat. Neurosci. 16, 1687–1691 (2013).

  15. 15.

    Lammel, S. et al. Input-specific control of reward and aversion in the ventral tegmental area. Nature 491, 212–217 (2012).

  16. 16.

    Cousens, G. & Otto, T. Both pre- and posttraining excitotoxic lesions of the basolateral amygdala abolish the expression of olfactory and contextual fear conditioning. Behav. Neurosci. 112, 1092–1103 (1998).

  17. 17.

    Thompson, K. J. et al. DREADD agonist 21 (C21) is an effective agonist for muscarinic based DREADDs in vitro and in vivo. ACS Pharmacol. Transl Sci. 1, 61–72 (2018).

  18. 18.

    Walker, D. L., Paschall, G. Y. & Davis, M. Glutamate receptor antagonist infusions into the basolateral and medial amygdala reveal differential contributions to olfactory vs. context fear conditioning and expression. Learn. Mem. 12, 120–129 (2005).

  19. 19.

    Cowansage, K. K. et al. Direct reactivation of a coherent neocortical memory of context. Neuron 84, 432–441 (2014).

  20. 20.

    Gore, F. et al. Neural representations of unconditioned stimuli in basolateral amygdala mediate innate and learned responses. Cell 162, 134–145 (2015).

  21. 21.

    Liu, X. et al. Optogenetic stimulation of a hippocampal engram activates fear memory recall. Nature 484, 381–385 (2012).

  22. 22.

    Yiu, A. P. et al. Neurons are recruited to a memory trace based on relative neuronal excitability immediately before training. Neuron 83, 722–735 (2014).

  23. 23.

    Denny, C. A. et al. Hippocampal memory traces are differentially modulated by experience, time, and adult neurogenesis. Neuron 83, 189–201 (2014).

  24. 24.

    Tanaka, K. Z. et al. Cortical representations are reinstated by the hippocampus during memory retrieval. Neuron 84, 347–354 (2014).

  25. 25.

    Rashid, A. J. et al. Competition between engrams influences fear memory formation and recall. Science 353, 383–387 (2016).

  26. 26.

    Ramirez, S. et al. Creating a false memory in the hippocampus. Science 341, 387–391 (2013).

  27. 27.

    Redondo, R. L. et al. Bidirectional switch of the valence associated with a hippocampal contextual memory engram. Nature 513, 426–430 (2014).

  28. 28.

    Ohkawa, N. et al. Artificial association of pre-stored information to generate a qualitatively new memory. Cell Rep. 11, 261–269 (2015).

  29. 29.

    Dudai, Y. in Science of Memory: Concepts (eds Roediger, H. L. III, Dudai, Y. & Fitzpatrick, S. M.) 13–16 (Oxford Univ. Press, 2007).

  30. 30.

    Shinkman, P. G., Swain, R. A. & Thompson, R. F. Classical conditioning with electrical stimulation of cerebellum as both conditioned and unconditioned stimulus. Behav. Neurosci. 110, 914–921 (1996).

  31. 31.

    Steinmetz, J. E., Lavond, D. G. & Thompson, R. F. Classical conditioning in rabbits using pontine nucleus stimulation as a conditioned stimulus and inferior olive stimulation as an unconditioned stimulus. Synapse 3, 225–233 (1989).

  32. 32.

    Hegoburu, C., Parrot, S., Ferreira, G. & Mouly, A. M. Differential involvement of amygdala and cortical NMDA receptors activation upon encoding in odor fear memory. Learn. Mem. 21, 651–655 (2014).

  33. 33.

    Sevelinges, Y., Gervais, R., Messaoudi, B., Granjon, L. & Mouly, A. M. Olfactory fear conditioning induces field potential potentiation in rat olfactory cortex and amygdala. Learn. Mem. 11, 761–769 (2004).

  34. 34.

    McGann, J. P. Associative learning and sensory neuroplasticity: how does it happen and what is it good for? Learn. Mem. 22, 567–576 (2015).

  35. 35.

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

  36. 36.

    Ozawa, T. et al. A feedback neural circuit for calibrating aversive memory strength. Nat. Neurosci. 20, 90–97 (2017).

  37. 37.

    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).

  38. 38.

    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).

  39. 39.

    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).

  40. 40.

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

  41. 41.

    Tang, J. et al. Pavlovian fear memory induced by activation in the anterior cingulate cortex. Mol. Pain 1, 6 (2005).

  42. 42.

    Johansen, J. P. & Fields, H. L. Glutamatergic activation of anterior cingulate cortex produces an aversive teaching signal. Nat. Neurosci. 7, 398–403 (2004).

  43. 43.

    Tye, K. M. & Deisseroth, K. Optogenetic investigation of neural circuits underlying brain disease in animal models. Nat. Rev. Neurosci. 13, 251–266 (2012).

  44. 44.

    Carey, R. M. & Wachowiak, M. Effect of sniffing on the temporal structure of mitral/tufted cell output from the olfactory bulb. J. Neurosci. 31, 10615–10626 (2011).

  45. 45.

    Doucette, W. et al. Associative cortex features in the first olfactory brain relay station. Neuron 69, 1176–1187 (2011).

  46. 46.

    Rojas-Libano, D. & Kay, L. M. Interplay between sniffing and odorant sorptive properties in the rat. J. Neurosci. 32, 15577–15589 (2012).

  47. 47.

    Shusterman, R., Smear, M. C., Koulakov, A. A. & Rinberg, D. Precise olfactory responses tile the sniff cycle. Nat. Neurosci. 14, 1039–1044 (2011).

  48. 48.

    Wesson, D. W., Donahou, T. N., Johnson, M. O. & Wachowiak, M. Sniffing behavior of mice during performance in odor-guided tasks. Chem. Senses 33, 581–596 (2008).

  49. 49.

    Kepecs, A., Uchida, N. & Mainen, Z. F. The sniff as a unit of olfactory processing. Chem. Senses 31, 167–179 (2006).

  50. 50.

    Vetere, G. et al. Chemogenetic interrogation of a brain-wide fear memory network in mice. Neuron 94, 363–374 (2017).

  51. 51.

    Paxinos, G. & Franklin, K. B. J. The Mouse Brain in Stereotaxic Coordinates (Academic Press, 2012).

  52. 52.

    Renier, N. et al. Mapping of brain activity by automated volume analysis of immediate early genes. Cell 165, 1789–1802 (2016).

  53. 53.

    Renier, N. et al. iDISCO: a simple, rapid method to immunolabel large tissue samples for volume imaging. Cell 159, 896–910 (2014).

  54. 54.

    Preibisch, S., Saalfeld, S. & Tomancak, P. Globally optimal stitching of tiled 3D microscopic image acquisitions. Bioinformatics 25, 1463–1465 (2009).

Download references


This work was supported by Canadian Institute of Health Research (CIHR) grants to P.W.F. (FDN143227) and S.A.J. (MOP74650) and by National Institute of Mental Health grants to K.J.R. (R01-MH108665) and F.G.M. (F31-MH105237). S.A.J. is a CIHR Canada Research Chair in Memory Function and Dysfunction. P.W.F. is a CIHR Canada Research Chair in Cognitive Neurobiology. The authors thank A. Ramsaran and J. Johansen for comments on an earlier draft of this manuscript.

Author information

Author notes

    • Gisella Vetere

    Present address: Laboratoire Plasticité du Cerveau, ESPCI- École Supérieure de Physique et de Chimie Industrielles, Paris, France


  1. Program in Neurosciences and Mental Health, Hospital for Sick Children, Toronto, Ontario, Canada

    • Gisella Vetere
    • , Lina M. Tran
    • , Sara Moberg
    • , Patrick E. Steadman
    • , Leonardo Restivo
    • , Sheena A. Josselyn
    •  & Paul W. Frankland
  2. Department of Physiology, University of Toronto, Toronto, Ontario, Canada

    • Lina M. Tran
    • , Sheena A. Josselyn
    •  & Paul W. Frankland
  3. Institute of Medical Sciences, University of Toronto, Toronto, Ontario, Canada

    • Patrick E. Steadman
    • , Sheena A. Josselyn
    •  & Paul W. Frankland
  4. National Center for PTSD at VA Boston Healthcare System, Boston, MA, USA

    • Filomene G. Morrison
  5. Department of Psychiatry, Boston University School of Medicine, Boston, MA, USA

    • Filomene G. Morrison
  6. McLean Hospital, Harvard Medical School, Belmont, MA, USA

    • Kerry J. Ressler
  7. Department of Psychology, University of Toronto, Toronto, Ontario, Canada

    • Sheena A. Josselyn
    •  & Paul W. Frankland
  8. Brain, Mind and Consciousness Program, Canadian Institute for Advanced Research, Toronto, Ontario, Canada

    • Sheena A. Josselyn
  9. Child and Brain Development Program, Canadian Institute for Advanced Research, Toronto, Ontario, Canada

    • Paul W. Frankland


  1. Search for Gisella Vetere in:

  2. Search for Lina M. Tran in:

  3. Search for Sara Moberg in:

  4. Search for Patrick E. Steadman in:

  5. Search for Leonardo Restivo in:

  6. Search for Filomene G. Morrison in:

  7. Search for Kerry J. Ressler in:

  8. Search for Sheena A. Josselyn in:

  9. Search for Paul W. Frankland in:


G.V. and P.W.F. conceived the study and designed the experiments. G.V., L.M.T., and S.M. conducted the behavioral and immunohistochemical experiments. G.V. and L.R. conducted data and statistical analyses. P.E.S. conducted the iDISCO tissue clearing. P.W.F., G.V., K.J.R., F.G.M., and S.A.J. wrote the paper.

Competing interests

The authors declare no competing interests.

Corresponding authors

Correspondence to Sheena A. Josselyn or Paul W. Frankland.

Integrated supplementary information

  1. Supplementary Figure 1 Testing apparatus.

    a, Odor preference was assessed in a rectangular box. Filter paper containing acetophenone or carvone was placed in the lid of a Petri dish. This was covered by a Petri dish (with 9 drilled holes), and then bedding. b, Photograph and schematic of test apparatus.

  2. Supplementary Figure 2 Distance traveled in test sessions. These graphs depict the distance traveled during the test session in each of the experiments.

    a, Acetophenone paired with shock (ANOVA, F4,46 = 13.10, P < 0.0001). (The data shown here corresponds to groups shown in Fig. 1b, c: CS + US group, n = 15 mice; naïve/home cage, n = 8 mice; US only, n = 15 mice; CS only, n = 8 mice; US│CS group, n = 12 mice). b, M72 photostimulation paired shock (ANOVA, F2,26 = 10.86, P = 0.0004). (The data here corresponds to groups shown in Fig. 1f, g: CS + US group, n = 18 mice; CS only, n = 8 mice; US│CS group, n = 8 mice). c, Acetophenone paired with food (corresponds to data shown in Supplementary Fig. 6, n = 12). d, M72 photostimulation paired with LHb stimulation (2 tailed t-test: t14 = 0.59, P = 0.56). (The data shown here corresponds to groups shown in Fig. 2d, e: CS + US, n = 12 mice; US│CS, n = 8 mice). e, M72 photostimulation paired with LDT stimulation (2 tailed t-test: t18 = 1.29, P = 0.21). (The data shown here corresponds to groups shown in Fig. 2i, j: CS + US, n = 10 mice; US│CS, n = 10 mice). f, Acetophenone paired with shock (2 tailed t-test: t14 = 0.24, P = 0.81) (real memory condition in BLA silencing experiment, corresponding to groups shown in Fig. 4d, e). g. M72 photostimulation paired with LHb photostimulation (2 tailed t-test: t14 = 1.26, P = 0.23) (real memory condition in BLA silencing experiment, corresponding to groups shown in Fig. 4i, j). *P < 0.05, **P < 0.01, ***P < 0.001 ****P < 0.0001 by Newman-Keuls. Error bars = s.e.m.

  3. Supplementary Figure 3 Schematic showing responses of the representative glomerular array (circles) to acetophenone and carvone.

    Acetophenone activates the M72 glomerulus (and other glomeruli). Carvone activates other glomeruli, but not the M72 glomerulus.

  4. Supplementary Figure 4 Training protocol pairing M72 photostimulation with shock.

    Each train included 40 pulses (pulse width 100 ms) delivered at 4 Hz. During the last 1 s of the train, a 0.7 mA shock was delivered. During training, mice received 10 photostimulation-shock pairs.

  5. Supplementary Figure 5 Training protocol pairing M72 photostimulation (artificial CS) with photostimulation of either LHb-VTA or LDT–VTA projections (artificial US).

    (Top) M72 stimulation occurred continuously through training, and consisted of 100 ms pulses, delivered at 4 Hz. (Bottom) Phasic VTA stimulation consisted of trains (0.5 s duration) of 5 ms pulses delivered at 30 Hz. Inter-train interval was 2 s.

  6. Supplementary Figure 6 Formation of an odor-reward memory.

    a, Schematic showing apparatus and training protocol. Mice were food-deprived one day prior to training. During training, acetophenone was paired with food, and preference (acetophenone vs. carvone) was evaluated one day later. b, In the test, mice (n = 12) spent more time on the acetophenone side of the apparatus (2 tailed, one sample t test vs. zero preference: t11 = 2.89, P = 0.015). Shading represents s.e.m. c, Summary data showing preference scores for individual mice. Error bar = s.e.m.

  7. Supplementary Figure 7 VTA photostimulation alone does not act as a US during conditioning.

    During training, acetophenone was paired with VTA photostimulation in mice (n = 12) not expressing ChR2. a, During testing mice did not exhibit conditioned aversion to acetophenone (2 tailed, one sample t test vs. zero preference: t11 = 0.47, P = 0.65). Shading represents s.e.m. b, Summary data showing preference scores for individual mice. Error bar = s.e.m.

  8. Supplementary Figure 8 Recall induced cFos expression.

    During the training phase mice received either CS alone (that is, acetophenone [n = 6] or M72 stimulation [n = 8]) or CS + US (that is, acetophenone paired with shock [n = 6] or M72 stimulation paired with LHb-mVTA stimulation [n = 7]). During testing, the CS only was presented, and c-Fos induction assessed in 18 brain regions (see Fig. 3a, b). The numbers of c-Fos+ nuclei are shown for mice in the ‘real’ (left) and ‘artificial’ (right) memory conditions. Error bars = s.e.m. Groups (CS + US vs. CS only) were compared by unpaired t-tests (2 tailed) in the real and artificial memory conditions. a, APir (real): t10 = 0.52, P = 0.62; APir (artificial): t13 = 0.56, P = 0.59. b, Pir (real): t10 = 0.26, P = 0.80; Pir (artificial): t13 = 0.20, P = 0.84. c, PLCo (real): t10 = 0.17, P = 0.86; PLCo (artificial): t13 = 0.20, P = 0.84. d, LEnt (real): t10 = 0.29, P = 0.77; LEnt (artificial): t13 = 1.17, P = 0.26. e, PMCo (real): t10 = 0.00, P = 0.99; PMCo (artificial): t13 = 0.71, P = 0.49. f, Tu (real): t9 = 1.34, P = 0.21; Tu (artificial): t11 = 1.70, P = 0.12. g, DTT (real): t9 = 0.68, P = 0.52; DTT (artificial): t10 = 0.25, P = 0.80. h, LOT (real): t10 = 0.99, P = 0.35; LOT (artificial): t11 = 3.25, P = 0.0077. i, AO (real): t10 = 0.99, P = 0.34; AO (artificial): t9 = 0.53, P = 0.61. j, V1 (real): t10 = 1.41, P = 0.19; V1 (artificial): t13 = 0.06, P = 0.95. k, VO (real): t9 = 0.86, P = 0.41; VO (artificial): t10 = 0.81, P = 0.43. l, BMA (real): t10 = 0.02, P = 0.99; BMA (artificial): t10 = 0.22, P = 0.83. m, CA1 (real): t10 = 0.36, P = 0.73; CA1 (artificial): t10 = 1.13, P = 0.29. n, RTn (real): t9 = 0.95, P = 0.37; RTn (artificial): t8 = 2.90, P = 0.020. o, Cg (real): t9 = 1.43, P = 0.19; Cg (artificial): t9 = 0.58, P = 0.58. p, Ce (real): t10 = 2.05, P = 0.068; Ce (artificial): t13 = 1.91, P = 0.079. q, La (real): t10 = 1.13, P = 0.29; La (artificial): t12 = 0.92, P = 0.38. Abbreviations: APir = amygdalopiriform transition area; Pir = piriform cortex; PLCo = posterolateral cortical amygdaloid nucleus; LEnt = lateral entorhinal cortex; PMCo = posteromedial cortical amygdaloid nucleus; Tu = olfactory tubercle; DTT = dorsal tenia tecta; LOT = nucleus of the lateral olfactory tract; AO = anterior olfactory nucleus; V1 = primary visual cortex; VO = ventral orbital cortex; BMA = basomedial amygdaloid nucleus; CA1 = field CA1 of the hippocampus; RTn = rsotromedial tegmental nucleus; Cg = cingulate cortex; Ce = central amygdaloid nucleus; La = lateral amygdaloid nucleus.

  9. Supplementary Figure 9 Regional cFos expression is highly correlated in real vs artificial conditions.

    a, CS only condition. Regional (n = 18 regions) c-Fos expression is highly correlated in mice in the real vs artificial conditioning groups (Pearson’s, r = 0.80, P < 0.0001). b, CS + US condition. Regional (n = 18 regions) c-Fos expression is highly correlated in mice in the real vs artificial conditioning groups (Pearson’s, r = 0.77, P = 0.0002).

  10. Supplementary Figure 10 C21 treatment alone does not affect conditioned odor aversion.

    During training, acetophenone was paired with shock in mice not expressing the DREADD, hM4Di. Before testing, mice were treated with VEH (n = 12) or C21 (n = 12). a. Both VEH-treated (one sample, 2 tailed t test vs. zero preference: t11 = 2.92, P = 0.014) and C21-treated (one sample, 2 tailed t test vs. zero preference: t11 = 2.40, P = 0.033) mice avoided acetophenone. Shaded area depicts s.e.m. b. Summary data showing preference in VEH- and C21-treated mice did not differ (2 tailed t test, t22 = 0.34, P = 0.73). Error bars = s.e.m.

  11. Supplementary Figure 11 C21 treatment reduces recall-induced activation of hM4Di-infected neurons in the BLA.

    a, b. During training, acetophenone was paired with shock. During testing, VEH-treated mice (n = 4) or C21-treated mice (n = 3) were presented with the CS (acetophenone) and c-Fos was quantified in hM4Di+ neurons in the BLA. C21 treatment reduced c-Fos expression (2 tailed t test, t5 = 3.03, P = 0.029). c, d. During training, M72 stimulation was paired with LHb-VTA stimulation. During testing, VEH-treated mice (n = 4) or C21-treated mice (n = 3) were presented with the CS (M72 stimulation) and c-Fos was measured in hM4Di+ neurons in the BLA. C21 treatment reduced c-Fos expression (2 tailed t test, t5 = 3.15, P < 0.05). In all graphs, error bars = s.e.m. Scale bars (in b and d) = 50 µm.

Supplementary information

  1. Supplementary Information

    Supplementary Figures 1–11.

  2. Reporting Summary

About this article

Publication history






Further reading