Parallel processing by cortical inhibition enables context-dependent behavior

Article metrics


Physical features of sensory stimuli are fixed, but sensory perception is context dependent. The precise mechanisms that govern contextual modulation remain unknown. Here, we trained mice to switch between two contexts: passively listening to pure tones and performing a recognition task for the same stimuli. Two-photon imaging showed that many excitatory neurons in auditory cortex were suppressed during behavior, while some cells became more active. Whole-cell recordings showed that excitatory inputs were affected only modestly by context, but inhibition was more sensitive, with PV+, SOM+, and VIP+ interneurons balancing inhibition and disinhibition within the network. Cholinergic modulation was involved in context switching, with cholinergic axons increasing activity during behavior and directly depolarizing inhibitory cells. Network modeling captured these findings, but only when modulation coincidently drove all three interneuron subtypes, ruling out either inhibition or disinhibition alone as sole mechanism for active engagement. Parallel processing of cholinergic modulation by cortical interneurons therefore enables context-dependent behavior.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.


All prices are NET prices.

Figure 1: Task design and auditory cue selection.
Figure 2: Task engagement elicits suppression and facilitation of A1 neurons.
Figure 3: Contextual modulation of inhibitory neurons.
Figure 4: Inhibition acts as a synaptic switch when changing contexts.
Figure 5: Cholinergic activity is necessary and partially sufficient to control behavior via inhibitory networks.
Figure 6: Cholinergic activity impacts inhibition directly.
Figure 7: Network model demonstrates requirement for co-activation of inhibitory and disinhibitory circuit elements.


  1. 1

    Fritz, J., Shamma, S., Elhilali, M. & Klein, D. Rapid task-related plasticity of spectrotemporal receptive fields in primary auditory cortex. Nat. Neurosci. 6, 1216–1223 (2003).

  2. 2

    Otazu, G.H., Tai, L.H., Yang, Y. & Zador, A.M. Engaging in an auditory task suppresses responses in auditory cortex. Nat. Neurosci. 12, 646–654 (2009).

  3. 3

    Atiani, S. et al. Emergent selectivity for task-relevant stimuli in higher-order auditory cortex. Neuron 82, 486–499 (2014).

  4. 4

    Rodgers, C.C. & DeWeese, M.R. Neural correlates of task switching in prefrontal cortex and primary auditory cortex in a novel stimulus selection task for rodents. Neuron 82, 1157–1170 (2014).

  5. 5

    Leonard, M.K. & Chang, E.F. Dynamic speech representations in the human temporal lobe. Trends Cogn. Sci. 18, 472–479 (2014).

  6. 6

    Letzkus, J.J. et al. A disinhibitory microcircuit for associative fear learning in the auditory cortex. Nature 480, 331–335 (2011).

  7. 7

    Fu, Y. et al. A cortical circuit for gain control by behavioral state. Cell 156, 1139–1152 (2014).

  8. 8

    Niell, C.M. & Stryker, M.P. Modulation of visual responses by behavioral state in mouse visual cortex. Neuron 65, 472–479 (2010).

  9. 9

    David, S.V., Fritz, J.B. & Shamma, S.A. Task reward structure shapes rapid receptive field plasticity in auditory cortex. Proc. Natl. Acad. Sci. USA 109, 2144–2149 (2012).

  10. 10

    Tabossi, P. Effects of context on the immediate interpretation of unambiguous nouns. J. Exp. Psychol. Learn. 14, 153–162 (1988).

  11. 11

    Gennari, S.P., MacDonald, M.C., Postle, B.R. & Seidenberg, M.S. Context-dependent interpretation of words: evidence for interactive neural processes. Neuroimage 35, 1278–1286 (2007).

  12. 12

    Williamson, R.S., Hancock, K.E., Shinn-Cunningham, B.G. & Polley, D.B. Locomotion and task demands differentially modulate thalamic audiovisual processing during active search. Curr. Biol. 25, 1885–1891 (2015).

  13. 13

    Jaramillo, S., Borges, K. & Zador, A.M. Auditory thalamus and auditory cortex are equally modulated by context during flexible categorization of sounds. J. Neurosci. 34, 5291–5301 (2014).

  14. 14

    McGinley, M.J., David, S.V. & McCormick, D.A. Cortical membrane potential signature of optimal states for sensory signal detection. Neuron 87, 179–192 (2015).

  15. 15

    Miller, J.M. et al. Single cell activity in the auditory cortex of Rhesus monkeys: behavioral dependency. Science 177, 449–451 (1972).

  16. 16

    Hubel, D.H., Henson, C.O., Rupert, A. & Galambos, R. Attention units in the auditory cortex. Science 129, 1279–1280 (1959).

  17. 17

    Zhou, M. et al. Scaling down of balanced excitation and inhibition by active behavioral states in auditory cortex. Nat. Neurosci. 17, 841–850 (2014).

  18. 18

    Kato, H.K., Gillet, S.N. & Isaacson, J.S. Flexible sensory representations in auditory cortex driven by behavioral relevance. Neuron 88, 1027–1039 (2015).

  19. 19

    Tan, A.Y. & Wehr, M. Balanced tone-evoked synaptic excitation and inhibition in mouse auditory cortex. Neuroscience 163, 1302–1315 (2009).

  20. 20

    Wehr, M. & Zador, A.M. Balanced inhibition underlies tuning and sharpens spike timing in auditory cortex. Nature 426, 442–446 (2003).

  21. 21

    Volkov, I.O. & Galazjuk, A.V. Formation of spike response to sound tones in cat auditory cortex neurons: interaction of excitatory and inhibitory effects. Neuroscience 43, 307–321 (1991).

  22. 22

    Anderson, J.S., Carandini, M. & Ferster, D. Orientation tuning of input conductance, excitation, and inhibition in cat primary visual cortex. J. Neurophysiol. 84, 909–926 (2000).

  23. 23

    Ferster, D. Orientation selectivity of synaptic potentials in neurons of cat primary visual cortex. J. Neurosci. 6, 1284–1301 (1986).

  24. 24

    Froemke, R.C. Plasticity of cortical excitatory-inhibitory balance. Annu. Rev. Neurosci. 38, 195–219 (2015).

  25. 25

    Froemke, R.C., Merzenich, M.M. & Schreiner, C.E. A synaptic memory trace for cortical receptive field plasticity. Nature 450, 425–429 (2007).

  26. 26

    Marlin, B.J., Mitre, M., D'amour, J.A., Chao, M.V. & Froemke, R.C. Oxytocin enables maternal behaviour by balancing cortical inhibition. Nature 520, 499–504 (2015).

  27. 27

    Kuhlman, S.J. et al. A disinhibitory microcircuit initiates critical-period plasticity in the visual cortex. Nature 501, 543–546 (2013).

  28. 28

    Bakin, J.S. & Weinberger, N.M. Induction of a physiological memory in the cerebral cortex by stimulation of the nucleus basalis. Proc. Natl. Acad. Sci. USA 93, 11219–11224 (1996).

  29. 29

    Kilgard, M.P. & Merzenich, M.M. Cortical map reorganization enabled by nucleus basalis activity. Science 279, 1714–1718 (1998).

  30. 30

    Hangya, B., Ranade, S.P., Lorenc, M. & Kepecs, A. Central cholinergic neurons are rapidly recruited by reinforcement feedback. Cell 162, 1155–1168 (2015).

  31. 31

    Pi, H.J. et al. Cortical interneurons that specialize in disinhibitory control. Nature 503, 521–524 (2013).

  32. 32

    Chubykin, A.A., Roach, E.B., Bear, M.F. & Shuler, M.G. A cholinergic mechanism for reward timing within primary visual cortex. Neuron 77, 723–735 (2013).

  33. 33

    Lovett-Barron, M. et al. Dendritic inhibition in the hippocampus supports fear learning. Science 343, 857–863 (2014).

  34. 34

    Vogels, T.P. & Abbott, L.F. Gating multiple signals through detailed balance of excitation and inhibition in spiking networks. Nat. Neurosci. 12, 483–491 (2009).

  35. 35

    Rudy, B., Fishell, G., Lee, S. & Hjerling-Leffler, J. Three groups of interneurons account for nearly 100% of neocortical GABAergic neurons. Dev. Neurobiol. 71, 45–61 (2011).

  36. 36

    Kepecs, A. & Fishell, G. Interneuron cell types are fit to function. Nature 505, 318–326 (2014).

  37. 37

    Xue, M., Atallah, B.V. & Scanziani, M. Equalizing excitation-inhibition ratios across visual cortical neurons. Nature 511, 596–600 (2014).

  38. 38

    Xiang, Z., Huguenard, J.R. & Prince, D.A. Cholinergic switching within neocortical inhibitory networks. Science 281, 985–988 (1998).

  39. 39

    Hasselmo, M.E. & Bower, J.M. Acetylcholine and memory. Trends Neurosci. 16, 218–222 (1993).

  40. 40

    Kruglikov, I. & Rudy, B. Perisomatic GABA release and thalamocortical integration onto neocortical excitatory cells are regulated by neuromodulators. Neuron 58, 911–924 (2008).

  41. 41

    Froemke, R.C. et al. Long-term modification of cortical synapses improves sensory perception. Nat. Neurosci. 16, 79–88 (2013).

  42. 42

    Dorrn, A.L., Yuan, K., Barker, A.J., Schreiner, C.E. & Froemke, R.C. Developmental sensory experience balances cortical excitation and inhibition. Nature 465, 932–936 (2010).

  43. 43

    Metherate, R. & Weinberger, N.M. Cholinergic modulation of responses to single tones produces tone-specific receptive field alterations in cat auditory cortex. Synapse 6, 133–145 (1990).

  44. 44

    Pinto, L. et al. Fast modulation of visual perception by basal forebrain cholinergic neurons. Nat. Neurosci. 16, 1857–1863 (2013).

  45. 45

    Lin, S.C. & Nicolelis, M.A. Neuronal ensemble bursting in the basal forebrain encodes salience irrespective of valence. Neuron 59, 138–149 (2008).

  46. 46

    Schneider, D.M., Nelson, A. & Mooney, R. A synaptic and circuit basis for corollary discharge in the auditory cortex. Nature 513, 189–194 (2014).

  47. 47

    Lee, S., Kruglikov, I., Huang, Z.J., Fishell, G. & Rudy, B. A disinhibitory circuit mediates motor integration in the somatosensory cortex. Nat. Neurosci. 16, 1662–1670 (2013).

  48. 48

    Mayse, J.D., Nelson, G.M., Avila, I., Gallagher, M. & Lin, S.C. Basal forebrain neuronal inhibition enables rapid behavioral stopping. Nat. Neurosci. 18, 1501–1508 (2015).

  49. 49

    Kawai, H., Lazar, R. & Metherate, R. Nicotinic control of axon regulates thalamocortical transmission. Nat. Neurosci. 10, 1168–1175 (2007).

  50. 50

    Goard, M. & Dan, Y. Basal forebrain activation enhances cortical coding of natural scenes. Nat. Neurosci. 12, 1444–1449 (2009).

  51. 51

    Huber, D. et al. Multiple dynamic representations in the motor cortex during sensorimotor learning. Nature 484, 473–478 (2012).

  52. 52

    Chen, J.L., Pfäffli, O.A., Voigt, F.F., Margolis, D.J. & Helmchen, F. Online correction of licking-induced brain motion during two-photon imaging with a tunable lens. J. Physiol. (Lond.) 591, 4689–4698 (2013).

  53. 53

    Vogelstein, J.T. et al. Fast nonnegative deconvolution for spike train inference from population calcium imaging. J. Neurophysiol. 104, 3691–3704 (2010).

  54. 54

    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).

  55. 55

    Miller, K.D. & Troyer, T.W. Neural noise can explain expansive, power-law nonlinearities in neural response functions. J. Neurophysiol. 87, 653–659 (2002).

  56. 56

    Pfeffer, C.K., Xue, M., He, M., Huang, Z.J. & Scanziani, M. Inhibition of inhibition in visual cortex: the logic of connections between molecularly distinct interneurons. Nat. Neurosci. 16, 1068–1076 (2013).

  57. 57

    Hofer, S.B. et al. Differential connectivity and response dynamics of excitatory and inhibitory neurons in visual cortex. Nat. Neurosci. 14, 1045–1052 (2011).

  58. 58

    Litwin-Kumar, A., Rosenbaum, R. & Doiron, B. Inhibitory stabilization and visual coding in cortical circuits with multiple interneuron subtypes. J. Neurophysiol. 115, 1399–1409 (2016).

  59. 59

    Ji, X.Y. et al. Thalamocortical innervation pattern in mouse auditory and visual cortex: laminar and cell-type specificity. Cereb. Cortex 26, 2612–2625 (2016).

  60. 60

    Chen, I.W., Helmchen, F. & Lutcke, H. Specific early and late oddball-evoked responses in excitatory and inhibitory neurons of mouse auditory cortex. J. Neurosci. 35, 12560–12573 (2015).

  61. 61

    Mesik, L. et al. Functional response properties of VIP-expressing inhibitory neurons in mouse visual and auditory cortex. Front. Neural Circuits 9, 22 (2015).

Download references


We thank I. Carcea, G. Fishell, B. Kripkee, M. Long, M. Picardo, D. Rinberg, R. Tsien, and D. Vallentin for comments on earlier versions of this manuscript; K. Katlowitz, X. Lin, H. Lu, J. Merel, and L. Paniniski for assistance with statistics and analysis; and G. Kosche, E. Morina, A. Resulaj, and C. Wilson for technical assistance. Artwork in Figures 1, 4, 5 and 6 was made by S.E. Ross. We thank the GENIE Program and the Janelia Farm Research Campus, Howard Hughes Medical Institute for provision of GCaMP6s and ScanImage. Artwork in Figures 1, 4, and 6 was made by S.E. Ross. This work was funded by grants from the NIDCD (DC009635 and DC012557), a Sloan Research Fellowship, and a Klingenstein Fellowship (R.C.F.) and by a grant from the NIDA (T32 DA007254), the Charles H. Revson Senior Fellowship in Biomedical Sciences, and a grant from the NIDCD (DC05014) (K.V.K.).

Author information

K.V.K. and R.C.F. designed experiments and wrote the manuscript. K.V.K., J.V.G., R.E.F., and E.S.P. conducted experiments. K.V.K., T.A.H.S., and J.V.G. performed analysis. G.W.L. and K.D.M. performed network modeling.

Correspondence to Robert C Froemke.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Integrated supplementary information

Supplementary Figure 1 Trial structure for go/no-go behavior

a, Overall timeline of experiments. Day 1: A1 mapping, injection of AAV1-SYN-GCAMP6s, cranial window and headpost implantation. Water restriction typically occurred ~7-10 days before initial training. Day 14-21: baseline imaging session to select the target and foil tones. Days 15-30: animal training. Day 20+: imaging session in trained animals. Variability in timeline in terms of start dates for each phase was largely attributable to GCaMP6s expression timing and behavioral learning trajectory. b, Normalized lick rate for all animals in all sessions in Figure 1, for hit correct trials to the target tone. Thin lines, individual sessions (n=5 mice, 11 sessions); thick line, average performance. c, Normalized lick rate for foil tones (correct reject trials). d, Performance on go/no-go task over sessions for 11 animals (d’ = z(hit rate) – z(false alarm rate). Dashed line indicates d’ of 1.75, criterion for considering animals fully trained on the task.

Supplementary Figure 2 Identification of false positives due to lick-related brain motion

a, Histogram of normalized tone-evoked response for one example mouse (animal Ms22). Lick-related brain motion should produce both positive and negative deflections in fluorescence. We calculated the nominal false positive rate as the percent of negative deflections greater than 5%. Vertical dashed line indicates false positive rate. Each animal used in this study had nominal false positive rates below 5% (i.e., 95% likely that positive deflections are not due to licking). b, Histogram of normalized responses for another mouse (animal Ms31). c, Scatter plot of the nominal false positive rate for all sites in all animals from Figure 2a-d. Orange dots are example sites in a, b. d, Confidence level (positive-negative deflection rate) at each normalized dF value. Even for small changes in fluorescence (<0.05), the probability of lick-related brain motion driving the signals is only <15%. Orange dots are example sites in a, b.

Supplementary Figure 3 Matching of neurons between active and passive context

a,c Average intensity projection of a single field-of-view in the passive (a) and active (c) contexts. b,d, maximum intensity projections for passive (b) and active (d) contexts. Yellow cells showed up in both contexts, while white boxes denote cells that were identified in one context but not the other and were excluded from the analysis.

Supplementary Figure 4 Rapid switching of neuronal activity after changing contexts

a, Heatmap showing trial-by-trial change in calcium response with the transition point (black bar) indicating when the context was switched (~60-90s). Left example shows a “passive-preferring” neuron immediately increasing the tone-evoked calcium response and the right example shows an “active-preferring” neuron with the opposite profile. b, Summary of neurons during contextual switching for the target tone. Data is plotted as the absolute change relative to the final active trial (n=148 cells, ANOVA with Tukey’s post-hoc). *, p<0.05. c, as in b but measuring responses to the foil tone. For the foil tone, the difference was significant by the second passive trial (n=134 cells, ANOVA with Tukey’s post-hoc). *, p<0.05.

Supplementary Figure 5 Task-related suppression was strongest for neurons with best frequencies close to targets or foils

a, Example neuron with best frequency (BF) close to target tone. Note significant suppression in active context. b, Passive frequency tuning curve for neuron in a (500 ms tones, 70 dB SPL, 0.25 octave spacing). Note peak at target frequency. c, Example neuron where target tone was 0.25 octaves from BF. This neuron selectively responded during the active context. d, Passive frequency tuning curve of neuron in c. e, Activation and suppression for targets and foils. Note significantly higher suppression at BF (target: 82% suppressed; foil: 83% suppressed) vs not BF (target: 58% suppressed; foil: 72% suppressed, p<0.01, Fisher’s exact test). f, Context modulation (y-axis, negative values indicate suppression, positive values indicate activation) as function of BF relative to target tone (in octaves). Suppression is greatest when target tone is BF (BF-target=0, p<0.002 versus not BF). g, as in f but measuring responses to foil tone.

Supplementary Figure 6 Motion artifact exclusion for imaging sessions with tdTomato structural marker

a, RGB merge image of a PV+ interneuron (green, GCaMP6s; red, tdTomato). b, Green channel image and associated tone-evoked epoch in the active and passive contexts, showing large tone-evoked response. c, Red channel image of same neuron and associated tone-evoked epoch, showing lack of response. The cross correlation of these two signals for this neuron was low (r: 0.2), and so this neuron was included in the analysis. d, Green channel from another example neuron included in the analysis. e, Red channel of the cell in d (cross-correlation coefficient r: 0.4). f, RGB merge image of a PV+ interneuron that was excluded. g, Green channel image and associated tone-evoked epoch in the active and passive contexts. h, Red channel image and associated tone-evoked epoch. Note the high correlation between the red and green channels (r: 0.9) even after bleedthrough correction with linear spectral unmixing.

Supplementary Figure 7 Spiking model based on synaptic conductances recovers contextual modulation

a, Example neuron (real recording) showing increased inhibition during the active context. b, Results of integrate-and-fire model neuron showing 5 trials of spikes from currents of cell in a for the passive (top) and active (bottom) contexts. c, Summary of simulations for five neurons with statistically significant suppression (n=5 neurons showing suppression, p<0.01, two-sided Wilcoxon rank sum test for active versus passive context). d, A different example neuron showing decreased inhibition during the active context. e, Results of integrate-and-fire model neuron showing 5 trials of spikes from currents of cell in d for the passive (top) and active (bottom) contexts. Please note this is likely an off-responding neuron. f, Only two neurons exhibited significant activation (n=2 neurons showing activation, p<0.01, two-sided Wilcoxon rank sum test). Five neurons had no significant difference between contexts (n=5 cells, 0.36±0.09 spikes, active context, 0.36±0.08 spikes, passive context, average p-value=0.41, gray lines). g, Cell-attached spikes were recorded (top) and then after breaking into the cell, IPSCs and EPSCs were measured (middle). We then put the measured IPSCs and EPSCs through the integrate-and-fire model and predicted spikes and compared that to the experimentally measured spikes. h, Cell-attached spike recordings show that this cell is suppressed (-45%) in the active context (top, passive: 17.0 Hz, 0% failure rate (failure = no tone-evoked spike); active: 9.3 Hz, 0% failure rate; p<0.01, Student’s two-tailed paired t-test). Synaptic measurements demonstrate that IPSCs are higher in the active context, more than the corresponding change in EPSCs (middle, ΔIPSC/ΔEPSC = 2.1). Predicting spikes based on the IPSCs and EPSCs show similar levels of suppression (-62%) compared to the experimentally measured active context suppression (bottom, model passive: 11.5 Hz; model active: 4.4 Hz). i, Cell-attached spike recordings show that this cell shows a 10% increase in activity in the active context (top, passive: 10.0 Hz, 20% failure rate; active: 11.0 Hz, 0% failure rate; p<0.01, Student’s two-tailed paired t-test). Synaptic measurements demonstrate that IPSCs are lower in the active context, more than the corresponding change in EPSCs (middle, ΔIPSC/ΔEPSC = 5.1). Predicting spikes based on the IPSCs and EPSCs showed similar levels of activation (+6%) compared to the experimentally measured results (bottom, model passive: 6.7 Hz, model active: 7.1 Hz).

Supplementary Figure 8 Optical suppression of PV+ and SOM+ interneurons

a, Experimental paradigm. b, Two example cells with whole-cell voltage clamp recording showing loss of context-dependent changes in IPSCs when PV+ interneurons were suppressed. c, Optical suppression of PV+ interneurons reduced context-dependent changes in pyramidal spiking. Summary of cell-attached recordings (n=5 neurons in 3 mice, Spike modulation index (control): 40.3±13.0%, Spike modulation index (opto): 12.8±4.1%, p<0.05, Student’s paired two-tailed t-test). d, Inactivation of SOM+ interneurons also reduced the context dependent spiking of layer 2/3 pyramidal neurons (n=5 neurons, p=0.06, Student’s paired two-tailed t-test).

Supplementary Figure 9 Effect of atropine on behavioral performance and ChAT histology

a, Coronal slice schematic; black square, area of AAV1-Syn-FLEX-GCaMP6s injection in ChAT-Cre mice into basal forebrain (AP: -0.5 mm from bregma, ML: 1.8 mm on right side, DV: 4.5 mm from surface). b, Proportions of GCaMP+/ChAT-, GCaMP+/ChAT+ and GCaMP-/ChAT+ cells (n=3 mice, 15 images, 174 cells). c, GCaMP6s expression in nucleus basalis. d, ChAT immunohistochemistry in same region. e, Overlay showing co-localization of GCaMP6s, ChAT. f, Decoding stimulus based on cholinergic terminal activity is effective only during tone period (stimulus: pre-tone, p=0.95; tone, p=0.04; post-tone, p=0.18; Student’s t-test). g, Decoding action (i.e., licking or no licking) could be predicted by cholinergic terminals even before the tone (pre-tone, p=0.06; tone, p-0.01; post-tone, p=0.04; Student’s t-test). h, Behavioral outcomes with saline applied to auditory cortex. i, Behavior with atropine applied to auditory cortex.

Supplementary Figure 10 Experimental design where blue light tonically illuminates the cortex in a block-based design

These experiments were performed in fully trained mice.

Supplementary Figure 11 Cholinergic control of inhibition during context-switching

a, Example neuron in saline showing that the time lag between excitation and inhibition is smaller in the active context relative to the time lag in the passive context (ΔE-I time lag: ‒3.6 ms, i.e., inhibition and excitation are 3.6 ms closer together in the active context). b, Example neuron in atropine showing an increase in the time lag between excitation and inhibition (ΔE-I time lag: 20.5 ms). c, Summary of the change in time lag of synaptic inputs in the control and atropine conditions (saline: n=12 neurons, ΔE-I time lag ‒4.4±2.2 ms; atropine: n=7 neurons, ΔE-I time lag 17.0±5.7 ms, p<0.05, Student’s paired two-tailed t-test).

Supplementary Figure 12 Optogenetic activation of cholinergic fibers in mouse auditory cortex

a, Overlay of LFP traces from 10 trials showing effect of optical stimulation. Blue bar: 5 sec period of laser stimulation. b, Amplitude spectra during 2 s before (gray) and during (black) stimulation, averaged from 10 optical stimulation trials. c, LFP spectrogram averaged from 10 trials, normalized by average power during pre-stimulation period. Blue bar: laser stimulation. d, Low-frequency power (1-10 Hz) was significantly decreased by optical stimulation of cholinergic fibers in auditory cortex (Pre-stim vs. During-stim, P < 0.01, Wilcoxon sign-rank test). Blue bar: laser stimulation. Interruptions in axes in d,e are periods of light-induced artifact on tungsten electrode.

Supplementary Figure 13 Additional model results and tests for robustness

a, Changes in firing rates in the 4-unit model (active context minus passive context), when ACh is modeled as input to PV and SOM cells alone (left) or all three inhibitory subtypes (right). PV & SOM alone can recapitulate the changes in evoked rates but causes VIP to decrease during baseline, pre-tone firing. Activating all three subtypes causes correct changes in evoked rates and correctly increases VIP in baseline rates. Excitatory baseline firing is slightly reduced. b, Robustness of firing rate changes to uneven strengths of ACh across inhibitory cell types. Each panel shows the effects of changing IA to the given subtype while holding it constant at 5 for the other 2 subtypes. The directional changes in firing rates remain correct for all 4 cell types over wide ranges. c, Multi-unit model results are similar to 4-unit model. Upper-left shows average baseline (black) and evoked (white) firing rates in the passive context. Upper-right: average changes in firing rates (active-passive). Lower-left: Diversity of firing rate changes in each cell type population represented as the percentage of active-preferring cells (data versus model). Color scheme as in b. Errorbars represent +/- 1 std from 20 network realizations. Lower-left: Tests of optical suppression experiments. Inactivation of PV cells in the model (cells shown here from a single example network) shifts total inhibition (sum of inhibitory inputs in Passive and Active conditions) downward on average but some cells have increased inhibition similar to observed results (Supplementary Fig. 8b). The direction of inhibition changes does not depend on whether the cell was active (magenta) or passive (blue) -preferring before inactivation.

Supplementary Figure 14 Histology

a, We confirmed that our optical suppression of PV+ neurons was selective (PV-cre mice injected with AAV1.EF1a.DIO.eNpHR3.0) by co-staining neurons using an anti-PV antibody (red) and one for GFP to recognize the eNpHR3.0-YFP (green). DAPI nuclear staining is in blue. b, Similar co-staining for somatostatin in SOM-cre mice injected with AAV1.EF1a.DIO.eNpHR3.0. c, CaMK2-GCaMP6f is largely excluded from PV+ interneurons, the largest inhibitory subtype in the cortex (93.3±3.3% of GCaMP6+ neurons do not exhibit detectable staining for PV, n=70 GCaMP6+ neurons in 2 mice).

Supplementary information

Supplementary Text and Figures

Supplementary Figures 1–14 (PDF 2673 kb)

Supplementary Methods Checklist (PDF 470 kb)

Context switching behavior in head-fixed mouse.

Video of trained mice in the context switching behavior. In the first part of the video, the animal is executing a learned sensorimotor response to the target tone (licking) and withholding from licking to the foil tone. Midway through the video the licktube is removed and the animal immediately stops licking for the target tone. (MP4 4450 kb)

Motion correction for passive context data.

Video of neurons expressing GCaMP6s sampled at 3.91 Hz. Left panel is not aligned and right panel is aligned based on correcting for x-y motion. (MP4 3783 kb)

Motion correction for active context data.

Video of neurons expressing GCaMP6s sampled at 3.91 Hz. Left panel is not aligned and right panel is aligned based on correcting for x-y motion. Please note that motion is limited to the x-y plane and there is little to no out-of-plane shifting. (MP4 3748 kb)

Cholinergic axon imaging in active and passive contexts.

Video of cholinergic axons in the auditory cortex that are projecting from the nucleus basalis and expressing GCaMP6s. Data is sampled at 3.91 Hz. Left panel is the active context and right panel is the passive context of the same field imaged immediately after removing the licktube. (MP4 2888 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kuchibhotla, K., Gill, J., Lindsay, G. et al. Parallel processing by cortical inhibition enables context-dependent behavior. Nat Neurosci 20, 62–71 (2017) doi:10.1038/nn.4436

Download citation

Further reading