PV interneuron tuning is re-weighted, not re-tuned: what longitudinal imaging does and does not show

Question: How have longitudinal two-photon calcium imaging studies of parvalbumin interneuron sensory tuning in mouse visual and auditory cortex tracked experience-dependent plasticity, and what circuit mechanisms explain shifts in their stimulus selectivity?

Answer

Across the retrieved literature, the direct longitudinal two-photon evidence on PV interneuron sensory tuning comes from somatosensory barrel cortex, not from visual or auditory cortex, and it supports a picture in which experience changes which PV cells respond to a stimulus and how strongly PV inhibition acts on pyramidal cells, rather than systematically re-writing PV feature preference: chronic imaging during whisker trimming reveals substantial recruitment of spared-whisker-responsive PV cells within deprived barrels, reversal of that recruitment on whisker regrowth, and yet persistence of individual PV selectivity changes for weeks [8]. In V1 and A1 the retrieved papers provide causal but cross-sectional evidence that PV cells constrain the gain, tuning width and noise geometry of principal-cell selectivity without themselves constituting an independent feature map [16] [7] [6], and that the instructive signals for PV re-weighting arrive through altered excitatory drive and disinhibitory interneuron crosstalk [15] [14]. The mechanisms that best explain selectivity shifts are therefore circuit-level: cell-type-biased thalamocortical and corticocortical innervation of PV cells, learning-onset suppression of PV→pyramidal synaptic output, and PV–SST–VIP antagonism gating when plasticity is permitted.

Longitudinal two-photon tracking of PV tuning exists only for somatosensory cortex, and it shows recruitment rather than re-tuning

At baseline, PV cells in the barrel field have spatially distributed whisker responses, responding best to the principal whisker of a given barrel and less often to surrounding whiskers in a distance-dependent manner [8]. During chronic whisker trimming, PV cells responsive to the spared whisker are substantially recruited in deprived but not spared barrels, and this recruitment reverses upon whisker regrowth [8]. Critically, the reversal is incomplete at the single-cell level: changes in individual PV cell whisker selectivity can persist for weeks [8]. This dissociation between a reversible population-level redistribution and partially irreversible single-cell states is the strongest constraint the retrieved corpus places on how PV tuning behaves over chronic timescales, and it implies that "return to baseline" at the population level can mask persistent single-cell drift.

PV plasticity is expressed at the inhibitory synapse, is contingency-gated, and can be sex-specific

Evoked PV→pyramidal IPSCs in layer 2/3, but not layer 5, are rapidly suppressed at the onset of a whisker-based sensory-association task, accompanied by a reduced number of PV-associated synapses on somata and dendrites [13]. This reduction occurs only in females, and PV-mediated inhibition remains stable when whisker stimulation is decoupled from reward during pseudotraining, indicating that the change is contingency-dependent rather than stimulus-driven [13]. Experience also interacts with PV maturation retrogradely: whisker trimming restricted to the third postnatal week prevents mTORC1-hyperactivation-induced deficits in PV cell input and output connectivity, in whisker-evoked cortical response precision, and in adult behaviour [11]. Together these results locate learning-related PV plasticity at the inhibitory output synapse, where it can shift the tuning of downstream principal cells without altering PV cells' own preferred stimulus.

In visual cortex PV cells control gain and noise geometry rather than carrying independent orientation preference

In marmoset V1, bidirectional optogenetics shows laminar specificity: in granular layers PV cells implement divisive/multiplicative linear gain control, whereas in extragranular layers they exert tuned nonlinear suppression that sharpens orientation tuning [16]. Modelling attributes gain control to changes in spiking threshold and tuning sharpening to changes in input noise, which regulates the input–output function [16]. Consistently, PV activation in awake marmoset V1 improves population discriminability without expanding the stimulus-related signal, instead compressing shared trial-to-trial variability and rotating it away from the stimulus-coding direction [7]. In mouse V1, population calcium imaging of excitatory and PV neurons shows that suppression in PV neurons is transiently reduced, driven by subcortical input dynamics, and that heterogeneity in suppression strength is linked to population correlations and excitatory–inhibitory balance [20]. Cortical circuits with strong recurrent excitation and stabilizing inhibition that provides normalization, but too weak across columns to account for cross-orientation suppression, are also consistent with interneuron stimulation producing paradoxical decreases in inhibitory firing through withdrawal of recurrent excitation [4]. The implication is that measured "shifts in PV selectivity" in V1 are more likely changes in the tuning of PV cells' normalizing input than changes in a PV feature preference.

In auditory cortex PV and SST inhibition make dissociable contributions to tuning fidelity and scene segregation

PV suppression in auditory cortex broadly alters spatial tuning properties — tuning width, modulation depth, centroid and sparseness — and reduces target-sound discriminability under clean conditions [6]. SST suppression leaves spatial tuning comparatively intact but selectively impairs discrimination during masked trials, weakening spatial release from masking [6]. In an auditory categorization task, two-photon imaging shows that SST interneurons carry both frequency-selective and categorical-choice signals, and their inactivation refines frequency discriminability and decision-boundary selectivity of auditory cortical neurons [5]. Pathological experience-dependent plasticity can be reversed by forcing PV activity: brief 40-Hz optogenetic activation of PV neurons in higher-order auditory cortex durably reverses lateral amygdala sensitization and reinstates discriminative auditory threat memory after noise-induced hearing loss [17]. Thus PV inhibition in A1 is positioned to set the fidelity of frequency and spatial tuning, while SST carries the segregation and categorization functions that learning would most plausibly reshape.

Shifts in PV selectivity are inherited from plasticity of the excitatory and disinhibitory inputs to PV cells

Layer 6 corticothalamic neurons form functional AMPA-receptor-containing synapses preferentially onto PV interneurons rather than promiscuously followed by pruning, establishing a cell-type-biased anatomical substrate through which a defined excitatory source can dominate PV tuning [19]. Plasticity of behaviourally relevant sensory responses is routed through interneuron subtypes: optogenetic climbing-fibre activation suppresses potentiation of whisker responses in layer 2/3 pyramidal cells by modulating SST- and VIP-positive interneurons, with zona incerta PV neurons required as a relay [15]. A model constrained by in vivo imaging of reversal learning assigns context-prediction error to lateral orbitofrontal disinhibitory VIP interneurons driving top-down gain amplification of apical dendritic inputs [14]. Across fear conditioning and extinction, PV and SST interneurons are recruited sequentially, with SST activity facilitating extinction by inhibiting PV cells and disinhibiting principal neurons [9]. Developmental work shows cortical SST and PV circuits act as reciprocal antagonists gating experience-dependent refinement of retinogeniculate connectivity, with PV acting as a brake on pruning once it matures [18]. Experience also realigns feedforward and recurrent drive so that layer 2/3 subthreshold orientation tuning becomes temporally coherent with layer 4 coactivity, a signature that naive cortex lacks [10].

Conflicts and gaps

The central gap is bibliographic: no retrieved paper performs longitudinal two-photon tracking of PV tuning in mouse V1 or A1 during an experience-dependent manipulation, so the visual and auditory mechanisms above are inferred from acute causal perturbations and from somatosensory chronic imaging [8]. The direction of required PV plasticity is contested. Learning-onset suppression of PV→pyramidal output is presented as the enabling event for stimulus–reward association [13], whereas enhancing PV activity is presented as the therapeutic route to restoring discriminative auditory responses [17] and PV suppression is shown to degrade spatial tuning and discriminability [6]; these are reconcilable only if PV tone is optimised rather than maximised, which no retrieved study tests directly. Species and laminar scope diverge: the gain-versus-sharpening dissociation comes from marmoset, and the authors explicitly frame it as reconciling divergent rodent results [16]. The sex specificity of learning-onset PV disinhibition [13] has not been tested for sensory tuning. Finally, cortical excitatory plasticity in V1 is reported as preserving feature preference while downstream parietal cortex reorganises preference [3], which leaves open whether any PV preference shift in V1 would be a cause or a consequence of plasticity elsewhere.

What would change the answer

A chronic, cell-type-identified preparation is the decisive experiment: longitudinal two-photon imaging of PV cells in V1 and A1 across a controlled experience (monocular deprivation, reward-paired orientation or frequency discrimination, or noise trauma), with stable registration of the same cells and interleaved measurement of pyramidal tuning in the same field of view. Pairing this with all-optical interrogation would let PV→pyramidal synaptic efficacy be probed repeatedly in identified cell pairs rather than inferred; existing all-optical work already quantifies how imaging parameters themselves drive opsin desensitisation and inflate trial-to-trial variability, a confound any longitudinal design must control [2]. Denser sampling of PV cells across layers and millimetres of cortex would reduce the selection bias inherent in small fields of view, as demonstrated by volumetric light-beads imaging resolving heterogeneous frequency tuning across cortical areas [1]. Voltage or fast calcium indicators would separate genuine preference shifts from changes in response gain, which the noise-geometry result shows can improve discriminability with no change in signal magnitude [7]. Finally, structural correlates — tracking PV-associated synapse number and perineuronal-net integrity alongside functional tuning — are needed to distinguish recruitment of previously silent PV cells from re-tuning of persistently imaged ones [13] [12].

References

[1] Otero-Coronel, S., Hildebrand, D. G. C., Noebauer, T., Demas, J., Chen, B., Weisenburger, S., Tejera, F., Grodem, S., Lensjo, K. K., Vatne, G. H., Fyhn, M., Freiwald, W. A., Vaziri, A.. "Large-scale volumetric two-photon calcium imaging enables cellular-resolution mesoscopic mapping of marmoset cortical areas". bioRxiv. 2026-05-29T00:00:00.000Z. URL: https://doi.org/10.64898/2026.05.28.728595

[2] Russell, B., Lees, R. M., Lak, A., Packer, A. M.. "Desensitization of opsin responses during all-optical interrogation depends on imaging parameters". bioRxiv. 2025-10-03T00:00:00.000Z. URL: https://doi.org/10.1101/2025.10.02.680058

[3] McClure, J. P., Varadi, M., Poort, J.. "Distinct response selectivity changes in primary visual and parietal cortex during visual discrimination learning". bioRxiv. 2025-10-26T00:00:00.000Z. URL: https://doi.org/10.1101/2025.10.24.684194

[4] Wang, S., Palmigiano, A., Miller, K. D., Van Hooser, S. D.. "Targeted cortical stimulation reveals principles of cortical contextual interactions". bioRxiv. 2026-08-31T00:00:00.000Z. URL: https://doi.org/10.1101/2022.06.22.497254

[5] Chai, R., Zhang, Y., Xin, Y., Deng, L., Xu, N.-l.. "Somatostatin interneurons in auditory cortex contribute to perceptual categorization". bioRxiv. 2025-06-09T00:00:00.000Z. URL: https://doi.org/10.1101/2022.07.06.498950

[6] Qu, Z., Nocon, J. C., Sen, K., Gritton, H.. "Parvalbumin and somatostatin interneurons support distinct computations during auditory scene analysis". bioRxiv. 2026-07-10T00:00:00.000Z. URL: https://doi.org/10.64898/2026.07.08.736456

[7] Cole, S., Hildebrand, D. G. C., Nurminen, L.. "Parvalbumin-expressing interneurons improve sensory discrimination by shaping noise geometry in primate V1". bioRxiv. 2026-06-15T00:00:00.000Z. URL: https://doi.org/10.64898/2026.06.11.731577

[8] Campos, B., Yang, X., Marmouset-de la Taille, I., Dipoppa, M., Zeiger, W. A.. "Dynamic recruitment of PV interneurons in the somatosensory cortex induced by experience dependent plasticity". bioRxiv. 2026-10-19T00:00:00.000Z. URL: https://doi.org/10.1101/2025.06.27.662081

[9] Campos Cardoso, R., Franks, H. T., Potter, N. R., Desa, Z. R., Fitzgerald, B. L., Fowler, R. L., Brown, F. A., Landar, V. A., Privette, M., Cummings, K. A.. "Coordinated activity and plasticity of infralimbic cortex GABAergic interneurons are critical for fear extinction encoding". bioRxiv. 2026-02-07T00:00:00.000Z. URL: https://doi.org/10.64898/2026.02.07.704584

[10] Lempel, A. A., Tragenap, S., Tepohl, C., Kaschube, M., Fitzpatrick, D.. "Coherent cortical representations develop after experience via feedforward-recurrent circuit alignment". bioRxiv. 2024-12-24T00:00:00.000Z. URL: https://doi.org/10.1101/2023.07.09.547747

[11] Amegandjin, C. A., Carreno-Munoz, M.-I., Francavilla, R., Damo Kamda, J. L., Roussel, M., Fernandes do Nascimento, A. S., Farley, A., Chehrazi, P., Ciancone Chama, A., Chattopadhyaya, B., Di Cristo, G.. "Sensory experience and mTORC1 interplay orchestrates the maturation of cortical interneuron connectivity and tactile sensitivity". bioRxiv. 2026-07-23T00:00:00.000Z. URL: https://doi.org/10.64898/2026.07.22.740050

[12] Belz, L., Aouci, R., Allain, B., Henderson, F., Fontaine, A., Desgagne, P., Calzaroni, L., Fabre, V., Pardi, M. B., Levi, G., Narboux-Neme, N.. "Dlx5/6 regulate perineuronal net-synapse coupling and stabilize adult cortical Parvalbumin neurons networks". bioRxiv. 2026-07-24T00:00:00.000Z. URL: https://doi.org/10.64898/2026.07.23.740286

[13] Park, E., Kuljis, D. A., Myal, S. E., Christian, J. A., Barth, A. L.. "Sexually dimorphic plasticity of PV inhibition in sensory neocortex during learning". bioRxiv. 2026-01-12T00:00:00.000Z. URL: https://doi.org/10.64898/2026.01.10.698814

[14] Tsai, M. C., Teutsch, J., Wybo, W. A. M., Helmchen, F., Banerjee, A., Senn, W.. "Hierarchy of prediction errors shapes the learning of context-dependent sensory representations". bioRxiv. 2026-09-15T00:00:00.000Z. URL: https://doi.org/10.1101/2024.09.30.615819

[15] Silbaugh, A., Koster, K. P., Hansel, C.. "Cerebellar climbing fibers impact experience-dependent plasticity in the mouse primary somatosensory cortex". bioRxiv. 2025-11-02T00:00:00.000Z. URL: https://doi.org/10.1101/2025.04.18.649578

[16] Vafaei, A., Clark, A. M., khadir, A., Dai, W., Federer, F., Angelucci, A.. "Laminar-specific control of response gain and orientation-tuning byparvalbumin-expressing inhibitory interneurons in primate visual cortex". bioRxiv. 2025-12-24T00:00:00.000Z. URL: https://doi.org/10.64898/2025.12.23.696300

[17] Awwad, B., Polley, D. B.. "Cortical inhibitory potentiation reverses maladaptive amygdala plasticity after noise-induced hearing loss". bioRxiv. 2026-04-06T00:00:00.000Z. URL: https://doi.org/10.64898/2026.04.02.716147

[18] JIANG, Q., Wu, S. J., Fishell, G., Chen, C.. "Reciprocal interaction between cortical SST and PV interneurons in regulating top-down retinothalamic refinement". bioRxiv. 2025-01-19T00:00:00.000Z. URL: https://doi.org/10.1101/2025.01.18.633032

[19] Gutman-Wei, A. Y., Sudarsanam, S., Cabalinan, A. G., Shahid, N., Shi, A., Guzman Clavel, L. E., Spindler-Krage, S. M., Agarwal, A., Kolodkin, A. L., Brown, S. P.. "Cell-type-selective synaptogenesis during the development of excitatory connectivity in the mammalian neocortex". bioRxiv. 2025-08-09T00:00:00.000Z. URL: https://doi.org/10.1101/2025.08.08.669371

[20] Rockwell, H., MacLean, J. N.. "Mechanistic basis of dynamic and heterogeneous divisive normalization in visual cortex". bioRxiv. 2025-08-24T00:00:00.000Z. URL: https://doi.org/10.1101/2025.08.19.671076

How this survey was made

This survey answers one question: “How have longitudinal two-photon calcium imaging studies of parvalbumin interneuron sensory tuning in mouse visual and auditory cortex tracked experience-dependent plasticity, and what circuit mechanisms explain shifts in their stimulus selectivity?”

We keep our own collection of recent preprints and refresh it every day from bioRxiv, medRxiv, arXiv, PsyArXiv, SocArXiv and ChemRxiv. For each survey, we search that collection for recent work and keep the strongest matches: papers that use the same terms as the question, and papers on the same topic in different words.

The text was written from the titles and abstracts of those papers; the full texts were not read. Statements that draw on a particular paper are marked with a number in brackets. The reference list just above is the 20 papers whose work this text draws on, chosen from the 136 papers those searches returned; the rest of the shortlist is not printed. It is not a complete review of everything published on the question.

Treat this as a starting point and a reading list, not a peer-reviewed review. The searches behind this survey ran on October 22, 2026 and covered work from the previous two years.