
==== Front
Sci Rep
Sci Rep
Scientific Reports
2045-2322
Nature Publishing Group UK London

70524
10.1038/s41598-024-70524-1
Article
Sensory adaptation in the barrel cortex during active sensation in the behaving mouse
Colins Rodriguez Andrea 1
Loft Michaela S. E. 1
Schiessl Ingo 12
Maravall Miguel 3
Petersen Rasmus S. R.Petersen@manchester.ac.uk

1
1 https://ror.org/027m9bs27 grid.5379.8 0000 0001 2166 2407 Division of Neuroscience, School of Biological Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, M13 9PT UK
2 grid.5379.8 0000000121662407 Geoffrey Jefferson Brain Research Centre, Manchester Academic Health Science Centre, Northern Care Alliance NHS Foundation Trust, University of Manchester, Manchester, M6 8HD UK
3 https://ror.org/00ayhx656 grid.12082.39 0000 0004 1936 7590 Sussex Neuroscience, School of Life Sciences, University of Sussex, Brighton, BN1 9RH UK
16 9 2024
16 9 2024
2024
14 2158811 4 2024
19 8 2024
© The Author(s) 2024
2024
https://creativecommons.org/licenses/by/4.0/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Sensory Adaptation (SA) is a prominent aspect of how neurons respond to sensory signals, ubiquitous across species and modalities. However, SA depends on the activation state of the brain and the extent to which SA is expressed in awake, behaving animals during active sensation remains unclear. Here, we addressed this question by training head-fixed mice to detect an object using their whiskers and recording neuronal activity from barrel cortex whilst simultaneously imaging the whiskers in 3D. We found that neuronal responses decreased during the course of whisker-object touch sequences and that this was due to two factors. First, a motor effect, whereby, during a sequence of touches, later touches were mechanically weaker than early ones. Second, a sensory encoding effect, whereby neuronal tuning to touch became progressively less sensitive during the course of a touch sequence. The sensory encoding effect was whisker-specific. These results show that SA does occur during active whisker sensing and suggest that SA is fundamental to sensation during natural behaviour.

Subject terms

Sensorimotor processing
Sensory processing
Barrel cortex
Whisker system
http://dx.doi.org/10.13039/501100002848 Comisión Nacional de Investigación Científica y Tecnológica 72170371 Colins Rodriguez Andrea http://dx.doi.org/10.13039/501100000268 Biotechnology and Biological Sciences Research Council BB/V00817X/1 BB/L007282/1, BB/ P021603/1 Maravall Miguel Petersen Rasmus S. http://dx.doi.org/10.13039/501100000265 Medical Research Council MR/L01064X7/1, MR/ P005659/1 Petersen Rasmus S. Weizmann UK Making Connectionsissue-copyright-statement© Springer Nature Limited 2024
==== Body
pmcIntroduction

Sensory Adaptation (SA), first demonstrated in single neurons by Adrian & Zotterman in 19261, is exemplified by the phenomenon that the response of neurons to a sensory stimulus of fixed strength (e.g., whisker deflection or light flash) progressively decreases during the course of a sequence of such stimuli. In general, SA refers to phenomena where neuronal responses are dynamic – depending not only on the immediate stimulus but also on the temporal context in which they are embedded2–14.

SA has been thoroughly studied in a ‘passive stimulation’ paradigm, where stimuli are applied to an immobilised animal – the animal is the passive recipient of stimulation2,4,5,10,14–20. This paradigm is experimentally convenient for studying SA, since the experimenter has a high degree of control over the sensory stimuli. However, the experimental conditions differ from those of behaving animals in important respects.

Behaving animals acquire sensory information by making exploratory movements of their sense organs (active sensing)21,22. For example, primates explore scenes by eye movements and rats/mice by whisker movements. The value of active sensing is that it allows animals to make actions that sample sensory information most useful to the animal’s current goals and thus to attain those goals more efficiently than would be possible via passive sensing23. Human eye movements can approach the efficiency of an ideal Bayesian planner24 and rats/mice use whisker movements to turn challenging tactile discrimination tasks into more straightforward detection tasks25,26. However, it is technically challenging to study SA during active sensing. In contrast to the passive stimulation paradigm, where SA can be straightforwardly assayed by comparing the response of neurons to a sequence of stimuli of fixed strength, during active sensing, animals continually move their sense organs and thereby change the sensory input. This presents a significant difficulty for the investigation of SA: a change in neuronal response during the course of a sequence of whisker touches might not be due to SA but rather to a change in touch strength. Whether SA is expressed in the actively sensing animal engaged in a goal-driven task is an open question.

Here we address this question by studying the mouse whisker system and taking advantage of recent advances in methods for quantifying whisker-dependent behaviour25,27,28. The whisker system exhibits active sensing in a tractable form: rats/mice explore objects by probing them with back-and-forth movements of their whiskers (“whisking”). There is consistent evidence for SA in the whisker-related zone of primary somatosensory cortex of the anaesthetised rat/mouse4,5,15–19,29,30 and this SA is specific to individual whiskers17,31–36. Using the passive stimulation paradigm, there is also evidence for SA in awake mice10,20,32–37. However, SA has not been studied in awake rats/mice during active touch.

To test whether SA occurs in the behaving, actively sensing mouse, we trained mice to perform a whisker-guided object detection task, and then recorded neural activity from barrel cortex whilst, simultaneously, we imaged their whiskers in 3D. These whisker measurements allowed us to disentangle motor effects on neural activity from sensory ones. We found that cortical responses to whisker-object touch do depend on prior touches. This effect originates not only from a direct effect on the responsiveness of sensory neurons (SA) but also from an indirect effect whereby touch triggers a change in motor action which in turn affects the strength of future touches.

Methods

All experiments were done in accordance with the UK Animals (Scientific Procedures) Act (1986). All experimental protocols described in here were approved by both United Kingdom Home Office national authorities and institutional ethical review, and are in accordance with the ARRIVE guidelines.

Behavioural apparatus

Whiskers were imaged similarly to28 using two high-speed cameras (LTR2, Mikrotron, Unterschleissheim, Germany; 480 × 480 pixels, pixel width 0.047 mm; 1000 frames/s, 0.2 ms exposure time) with telecentric lenses (55–349, Edmunds Optics, Barrington, NJ). In relation to standard anatomical planes for the skull, one camera imaged the horizontal plane; the second camera imaged a plane 25◦ off the sagittal plane and 10◦ off the horizontal plane, chosen to minimise occlusion between the cheek of the mouse and the whiskers. Mice were head-fixed and positioned inside a perspex tube (inner diameter 32 mm). The whiskers were free to move at all times. Whiskers were illuminated from below using an infrared LED array (940 nm, LED 940–66-60, Roithner, Vienna, Austria) via a diffused condenser lens.

Surgical procedure and mice training

Mice (C57; males; N = 11; 6 weeks at the time of implant) were implanted with a titanium head-bar under isoflurane anaesthesia (for details, see38). For all recovery procedures in this study, Buprenorphine (0.1 mg/kg) was injected for postoperative analgesia. After surgery, mice were left to recover for at least 5 days before starting water restriction (1.5 ml water/day). Training began 7–10 days after the start of water restriction. Mice were trained for water reward until they achieved a success rate of 70% on the task defined below. Subsequently, a craniotomy was performed as detailed in the following section and sealed with silicone. Mice were left to recover for 2–3 days until performance had recovered before the first recording session.

Electrophysiological recordings

Mice were trimmed to 3 whiskers in C row (C1, C2 and C3) and intrinsic optical imaging was performed, whilst deflecting whisker C1 or C2 using a piezoelectric actuator to locate C1 or C2 whisker column, under isoflurane anaesthesia39. A craniotomy was then performed at this location and filled with silicone elastomer. At the start of each recording session, the elastomer was removed and a 32-channel linear silicon probe (2 shanks; each shank had a 2 × 8 array of 12 µm recording sites, spanning ~ 200 µm along the long axis of the shank; 250 µm inter-shank spacing, Cambridge NeuroTech, Cambridge, United Kingdom) was inserted into the brain at a rate of approximately 5 μm/s under brief isoflurane anaesthesia so that the depth of the recording sites was 600–800 µm below the pial surface. At the end of a recording session (sampling rate 24.4 kHz, filtering 300–3000 Hz), the microelectrode array was withdrawn, the craniotomy sealed with silicone elastomer, and the mouse returned to its home cage.

Behavioural task

Mice were trained on the pole detection task detailed in28. Briefly, mice were trained and imaged in a dark, sound-proofed enclosure. A head-fixed mouse was placed inside a perspex tube, from which its head emerged at one end. The stimulus object was a 0.2 mm diameter, vertical carbon fibre pole which could be translated in the horizontal plane by stepper motors. To allow vertical movement of the pole into and out of range of the whiskers, the apparatus was mounted on a pneumatic linear slide, powered by compressed air, controlled from MATLAB via a real-time processor. In this way, the pole moved rapidly (~ 0.15 s) into and out of range of the whiskers. Mouse responses were monitored by a lick port located anterior to the mouth. Licks were detected as described in40. Each lick port consisted of a metal tube connected to a water reservoir via a computer-controlled solenoid valve. Head-fixed mice were trained to detect the presence of the pole using their whiskers, using behavioural procedures similar to40. On each trial, the pole was presented either within reach of the whiskers (‘Go trial’) or out of reach (‘NoGo trial’). At the start of each trial, the computer triggered the pole to move up (travel time ~ 100 ms). The pole stayed up for 1 s, before moving down. On Go trials, the correct response was for the mouse to lick the lick port. Correct responses (hits) were rewarded by a drop of water (~ 10 μl). Incorrect responses on Go trials (not licking, misses) were punished by timeout (3–5 s). On NoGo trials, the correct response was to refrain from licking (correct rejection) and incorrect responses (licking, false alarms) were punished by timeout and tone (frequency 12 kHz).

Whisker tracking

After calibration of the cameras, whiskers were tracked in 3D from the video28. Calibration was performed at the end of each recording session.

The shape of each target whisker was described as a quadratic, 3D Bezier curve bs=xs,ys,zs, defined by 3 control points, where, x, y, z are space coordinates and 0≤s≤1 parameterises location along the curve: bs=0 corresponds to the control point closest to the whisker base and bs=1 to that furthest from the base.

The intrinsic shape of a quadratic curve is fully described by a curvature function κ3Ds 41:κ3Ds=dbsds×d2bsds2dbsds3

Here a denotes the magnitude (2-norm) of vector a. κ3Ds has units of 1/distance and is the reciprocal of the radius of the circle that best fits the curve at point s. For a quadratic curve, in the quasi-static framework, we can apply the standard relationship between bending moment during contacts (Mz) and a change in whisker curvature42,43. Thence it follows that Mz is proportional to:Δκ3Ds=κ3Ds-κ3D,0s

where κ3Ds is the curvature of the whisker during contact and κ3D,0s is the curvature when the whisker is free from contact and in its resting state. All results presented here were evaluated at s=0 and 5 ms after touch onset.

Bezier curves were visually checked and occasional errors were manually corrected. Touches were manually detected at millisecond resolution with the aid of a customised software as in28. This step was the main bottleneck in the analysis. We focused on sessions in which a mouse performed at least 40 Go trials and the simultaneous electrophysiological recordings yielded at least 2 touch-responsive single units. This selection resulted in 6 sessions from 3 mice, containing a total of 443 trials and 1,580,064 frames.

Electrophysiological data analysis

Spike sorting

Single units were isolated from the silicon probe recordings using JRCLUST44. Recordings from each electrode shank were processed independently. A putative single unit was selected for further analysis if (1) its activity could be tracked during the whole recording, (2) its inter-spike interval histogram exhibited a clear absolute refractory period and (3) its waveform was biphasic and consistent across channels38. After the spike sorting process, we selected all units that showed an increase in their mean firing rate shortly after touch onset for further analysis.

ROC analysis

ROC analysis was carried out as in45. Responses were measured as spike counts in 30 ms non-overlapping time windows. The hit rate (correct detection of touch) was calculated using response windows starting at touch onset. For first touch, the false positive rate was calculated using ‘noise’ responses sampled between trial onset and first touch. For later touches, ‘noise’ responses were sampled from inter-touch intervals.

Tuning curves

To measure tuning curves for Δκ3D, we first identified touch onset times and categorised them as belonging to first touch on a trial, second touch, etc. We calculated the mean of Δκ3D in the 5 ms interval following each touch onset. The 5% of touches with greatest absolute value were discarded as outliers. The remaining samples were discretised into 4 equipopulated bins and, for each unit, the mean spike count (calculated in the 30 ms bin starting at touch onset) for each bin was calculated separately for first touch events, second touch events, etc. For each unit, the tuning curve was estimated as the linear correlation between spike count and Δκ3D.

Testing for sensory adaptation

When a mouse touches an object by active touch, it controls the force of whisker-object contact and the strength of the touches vary (Results, Fig. 2). A change in neuronal response to touch during the course of a sequence of touches might therefore reflect SA and/or a change in the strength of touch. To tease these effects apart, we first measured the strength of each touch in the trial (first, second etc.) as |Δκ3D| and measured the tuning curves separately for each touch. The ordinal number of the touch in the trial was defined irrespective of whisker identity. These tuning curves enabled us to estimate (predict) the response of a unit in the state of second touch, for example, to touches of varying |Δκ3D|. To test for SA, we compared the response of units to first touch with that predicted by applying touch strengths measured for first touch through the tuning curves estimated for later touches. This procedure was done using a leave-one-out cross-validation process.

To quantify the contribution of SA to the response attenuation of Fig. 2, we used the following approach. For robustness, this analysis was carried out using the summed activity of each simultaneously recorded population (‘population response’). For each population, we first calculated the median response to first touch (FR1) and the median response to later touch (FRlater). The difference FR1-FRlater is the response attenuation of Fig. 2. Next, we used the tuning curve approach described above to ask how much of this attenuation could be attributed to SA alone. To this end, we calculated the tuning curve of the population response to |Δκ3D| for later touches. We then used this tuning curve to estimate the median population response to first touch (FR1′). In the complete absence of SA, the tuning curve for later touch matches that for first touch, so that FR1′ is equal to FR1 and α = 0 in the following equation:α=FR1-FR1′FR1-FRlater

At the other extreme, if the response attenuation effect is entirely due to SA, α = 1. The index α thus quantifies the degree to which response attenuation in a given population can be explained by SA.

Stimulus-specific adaptation

For this analysis, trials that met all the following criteria were selected: at least the first two touches in a trial (‘adapting’) were produced by the same whisker (C1 or C2); at least one subsequent touch (‘test’) was produced by the other whisker; the time of the test touch did not overlap with that of any other whisker. For comparison between test touch and later (Ntouch≥3) touch responses, we determined tuning curves for later touches, one for each whisker. These tuning curves were obtained as detailed above, using data for all later touches, except (as a cross-validation procedure) the test stimuli. To quantify changes in the responses to the adapting and test stimuli, we used the index defined in Results (Stimulus-Specific Adaptation, Eq. 1).

Results

Variability of whisking behaviour

Our first aim was to establish a behavioural paradigm to investigate whether Sensory Adaptation (SA) occurs during active touch. Head-fixed mice can discriminate the location of a pole using their whiskers and use active touch to do so25,26. As a mouse repeatedly moves its whiskers back and forth against the pole, this results in touch sequences with inter-touch intervals of up to about 100 ms25,26. It is well-established that sequences of touches with such intervals, delivered passively to anaesthetised mice, produce strong SA in barrel cortex15,30. We therefore reasoned that, if SA is expressed in the awake, actively sensing animal, then the process should be engaged by such behaviour. To test this, mice (N = 11) were trained to solve a Go-NoGo pole detection task (Fig. 1A) with a single row of whiskers (C1, C2 and C3). On ‘Go’ trials, the pole was moved within reach of the whiskers: if the mouse licked the lickport (hit), the mouse was rewarded with a water droplet; if not (miss), it was penalised with a time-out. On ‘NoGo’ trials, the pole was placed out of reach of the whiskers: if the mouse withheld licking (correct rejection), the task proceeded immediately to the next trial; if not (false alarm) the mouse was penalised with a time-out and sound. Mice learned to perform the task accurately (81 ± 17%, mean ± SD over mice), performing 135 ± 22 trials per session. Use of this task allowed us to record a substantial number of trials to achieve good statistical power and to test if SA might be decision-dependent.Fig. 1 Variability of whisker-pole touches during the pole detection task. (A) Behavioural task schematic. Mice were trained to report the pole’s presence within whisker reach (Go trials) by licking a port. On NoGo trials, the pole was moved to a position out of whisker reach and mice should withhold licks. False alarms were penalised with a tone, while Misses were penalised with a time-out period. (B) Whisker imaging. Whiskers (C1, C2 and C3) were imaged with two cameras during the behavioural task (top panels). Whiskers were tracked in 3D and 3D curvature (κ3D) was then calculated (bottom panels). (C) Top/Middle: Whisker-pole touches and κ3D/azimuthal whisker angle from an example Go trial. Bottom: Touches for an example session (in this session only C1 (blue) and C2 (green) whiskers touched the pole). (D) Behavioural variability of touches (hit trials green; miss trials red): Strength of touch (Δκ3D), touch-length (inset shows fraction of touches with contact periods > 1000 ms), inter-touch interval and number of touches per trial. Data from all whiskers and behavioural sessions were included.

In order to measure whisking behaviour in 3D, we imaged the whiskers at 1000 frames/s with two video cameras (Fig. 1B, top). Using a 3D whisker tracking algorithm28, both 3D whisker shape and 3D whisker position were extracted from the imaging data (total of 1.8 M frames). From whisker shape, the 3D curvature (κ3D) at the base of each whisker was estimated for each frame (Fig. 1B, bottom). We estimated the strength of whisker-pole touch in each frame as the change in curvature at the whisker base relative to its intrinsic curvature (Δκ3D).

As desired, each trial consisted of a sequence of one or more ‘touches’ (a sequence of frames in which the whisker is in continuous contact with the pole) (Fig. 1C). In contrast to the regular stimulus sequences usually used in sensory physiology, we found that the temporal pattern of touches varied substantially across trials. This variability was manifest in number of touches per trial, interval between touch onsets, duration of touch, strength of touch (Δκ3D) and direction of touch (sign of curvature change) (Fig. 1D). Touches on miss trials were significantly weaker (t-test, p < 0.005), longer (t-test, p < 10–10), sparser (t-test, p < 10–10) and there were considerably fewer touches on miss trials than on hit trials (10.5 ± 5.7 per trial on hit trials, 5 ± 7.7 on miss trials; t-test, p < 10–10. Figure 1D). These observations raise the question of whether, in the behaving animal, SA to repeated touches is a strong enough effect to be significant above the ‘noise’ of behavioural touch variability or whether it might be a phenomenon only observable when touch variability is unusually low, as under passive whisker stimulation.

Response attenuation during a touch sequence

To determine if SA occurs during active touch, we made extracellular recordings of neural activity (single-unit, N = 33 obtained from 6 sessions from 3 mice) from barrel cortex targeted at column C1 or C2 whilst mice performed the pole detection task. Prior to a bout of touches, firing rate was low (red dots prior to touch onset in Fig. 2A, right). During a touch bout, firing rate remained elevated (black dots prior to touch onset in Fig. 2A, right) and each touch elicited a clear phasic response (Fig. 2A, 2B). For each trial, we classified touches according to whether they were first, second, third, etc. on that trial. We compared responses across touches and between hit and miss trials by 2-way ANOVA (main effects of first vs later touch p < 10–6, and hit vs miss trial p = 0.06). We found that the first touch evoked a substantially stronger response than later touches, and that there was a progressive attenuation in response (firing rate) during the course of a series of touches (Fig. 2A-C). We refer to this phenomenon as ‘response attenuation’. Response attenuation was robust across animals: in every mouse, the mean response to second touch was ≤ 50% of that to first touch (range 42–50%). Responses were similar on error (miss) trials: (ratio of firing rate of first touch on miss vs hit trials 1.06; Fig. 2D), indicating that response attenuation is independent of central processing of trial outcomes. Statistically identical results were obtained when we randomly selected later touches to equalise the number of first-touch and later-touch samples (Number of touches = 123, 2-way ANOVA, main effect of first vs later touch p < 10–6; main effect of hit vs miss p = 0.34). The latency of touch responses was greater for later touches (Fig. 2E, t-test, mean p-value across units 3 · 10–4, Number of units = 33), consistent with previous reports of SA during passive stimulation10. We did not observe a significant difference between the width of the peak of the responses to first and later touches (t-test across population responses, p = 0.38). The latency of the touch responses was moderately correlated with the pre-touch firing rate (mean Pearson correlation coefficient across units 0.29 ± 0.06, mean p-value = 10–3). First touch responses were not only stronger than later touch responses but also more discriminable from spontaneous activity (quantified by area under ROC curve; 2-way ANOVA; main effect of first vs later touches, p < 10–6; main effect of hit vs miss, p = 0.53; Fig. 2F). This is in line with the greater pre-touch firing rate for later touches compared to first touch, observed in Fig. 2A (right). Before the first touch in a trial, neurons spike at their baseline level. Pre-touch firing for later touches partly reflects responses to previous touches on a given trial. As detailed below, it is important to distinguish response attenuation from SA, which is just one of several factors that could potentially be responsible.Fig. 2 Response attenuation in barrel cortex during touch sequences. (A) Raster plots and PSTHs of the neural population recorded during one session (Number of units = 7) divided into hit and miss trials. Left panels show data sorted by chronological order of touches, while right panels show the same data sorted by touch order within trials (first touch, second touch etc.). PSTHs were smoothed (5 ms boxcar) for display only. (B) Same as A for miss trials. (C) Normalised responses of all units to touch ordered by their occurrence within trials (grey lines). Firing rate was normalised by responses to the first touch. The black solid line represents mean and SD across neurons in hit trials and the dashed black line represents mean and SD across neurons in miss trials. (D) Comparison of responses to touches (spike count within 30 ms after onset) during hit and miss trials, across the sample of units. Small dots indicate averages within sessions, large dots show the mean across all units and error bars indicate SD across all units. Asterisks show the results of post-hoc testing (Bonferroni). (E) Mean initial latency of the responses of all units to touch ordered by their occurrence within trials (grey lines). Black line shows mean and SD across units. (F) Left: Receiver Operating Characteristic (ROC) curve of touch detection for an example unit. Diagonal line represents chance performance (area under ROC curve is 0.5). Solid lines represent ROC for hit trials; dashed lines for miss trials. Right: Area under ROC curve for first and later touches of hit and miss trials across all units. Grey line represents chance performance. For all panels: * indicates P-value < 0.05, ** < 0.005, *** < 0.0005.

Motor adaptation

Although these response attenuation data are consistent with the hypothesis that SA is elicited by active touch during our task, there is a potential alternative explanation. Since the touches in these touch sequences are far from identical (Fig. 1), the decrease in firing rate from first to later touches might potentially be due to physical differences in stimulus strength. To test this possibility, we used 3D whisker tracking to measure both how far, and how fast, the whiskers were displaced during touches, and also how much the whiskers bent during touches (Fig. 3A and 3B). A significant advantage of 3D whisker imaging is that it captures bending and velocity components in the vertical plane that are not visible from the horizontal view alone (see Fig. 3A and 3C, first touch)28,46.Fig. 3 Strength of whisker touch decreases during a touch sequence. (A) Single-trial example showing whisker tracking solutions (Bezier curves) projected onto both horizontal and vertical imaging planes. Tracking solution at touch onset is white, whilst solutions for the subsequent 20 ms are green. (B) Extraction of whisker shape/displacement parameters from the tracking solutions. Bezier curves describe a whisker at touch onset (t = 0, grey solid curve; circle marks whisker base) and 5 ms later (t = 5, green solid curve). Whisker displacement (Db) is the distance between whisker base at t = 0 and that at t = 5, and is proportional to whisker speed. Curvature of the whisker at its base is derived from the length of the radius (r1; dashed grey line) of the circle (dashed grey curve) whose curvature matches that of the whisker at the base. Dashed green line/curve, with associated radius r2, show corresponding data for t = 5. Touch magnitude (Δκ3D) is the difference in the reciprocals of r1 and r2. X, Y and Z axes correspond to the coordinates defined in Fig. 1B). (C) Δκ3D and Db (mean of their values in the range t = [0,5] ms) for each of the three touches shown in panel A. Shaded areas indicate touches. (D) Left: Δκ3D measured as in panel B as a function of touch number for each session (grey lines), averaged across all trials (black), hit trials (green) and miss trials (red); error bars denote SD across trials. Right: Corresponding data for whisker base speed (Db/dt). E) Comparison of Δκ3D for correct and miss trials. Pink dots indicate averages within a session, red dots and error bars are the average and standard deviation over all trials (from all sessions) respectively. Asterisks show results of post-hoc testing (Bonferroni). For all panels: * indicates P-value < 0.05, ** < 0.005, *** < 0.0005. Dashed brackets indicate no significant difference.

We compared whisker speed Db/Δt and touch strength (Δκ3D) (Fig. 3B), computed at the whisker base, of first touches to later touches. We tested the effect of first vs later touch and of hit vs miss trial on these variables by 2-way ANOVA. For first vs later touch, we found that whisker speed was similar but that there was a substantial decrease in Δκ3D (main effect of first vs later touch on Δκ3D, p < 10–6; on Db/Δt, p = 0.9; Fig. 3D). This indicates that active touch on this task involves pronounced ‘motor adaptation’, whereby animals make progressively weaker pole contact during the course of a trial. The evolution of Δκ3D and Db/Δt was similar on hit and miss trials (main effect of hit vs miss on Δκ3D, p > 0.8; on Db/Δt p = 0.27; Fig. 3D), indicating that, as with the response attenuation reported above, the motor adaptation effect is independent of central processing of trial outcome. Statistically identical results were obtained when we equalised the number of first vs later touch samples (N touches = 123, main effect of first vs later touch on Δκ3D, p < 10–6; on Db/Δt, p = 0.62; main effect of hit vs miss on Δκ3D, p ≥ 0.3 and on Db/Δt, p = 0.49). In these analyses, both protraction and retraction touches were included.

Since whisker bending is the best single predictor of the response of whisker mechanoreceptors to touch38,47–49, these data imply a decrease in strength of afferent drive to barrel cortex during the course of a trial. This suggests that the attenuation of cortical response reported above (Fig. 2) might not be entirely due to SA but potentially could, either entirely or in part, be an indirect effect of motor adaptation.

Sensory adaptation

To determine whether SA does contribute to response attenuation, we developed a method to test for potential changes in SA during the course of a touch sequence. The natural variability in touch reported in Fig. 1, allowed us to estimate tuning curves to touch strength (mean of 71 first touches, 560 later touches per session). Consistent with previous studies50–53, we found that barrel cortical units were typically tuned to whisker bending: the greater the bending (Δκ3D), the greater the firing rate (Fig. 4A). If SA is indeed expressed, the prediction is that tuning curves (firing rate as a function of Δκ3D) should shift downwards (units become less sensitive to touch) during the course of a touch sequence. We used linear regression to estimate tuning curves, since we found linearity to be a good approximation within the range of Δκ3D that we were considering (Fig. 4AB).Fig. 4 Sensory Adaptation is prominent during active touch. (A) Left: sorted values of Δκ3D in an example session. Right: Tuning curves for first and later touches for an example unit recorded during session shown in A. (B) Left: Tuning curves for first (red) and later (black) touches for all selected units. Responses in the 30 ms interval before touch are shown for comparison. Right: Prediction of response to first touch using tuning curves estimated for later touch (black circles) and first touch respectively (red circles). (C) Comparison of tuning curves parameters for first and later touches. Left panel shows slope and right panel shows intercept of tuning curves for all selected units. Error bars indicate mean and standard deviation across units. (D) Left: Ratio between firing rate predicted from later touches tuning curve (Ntouch > 4) and mean observed firing rate for each touch. Grey lines represent mean for each unit and black line represents average across units. Right: Intercept of tuning curves for each touch. Grey lines represent mean for each unit and black line represents average across units.

Figure 4A (right) shows results for an example unit. As predicted by the SA hypothesis, there was a downwards shift in tuning curves from first to later touches. This was a typical result (Fig. 4B). To quantify the extent of this adaptation, we used the tuning curves to estimate and compare the responses of units to first vs later touches, keeping touch strength fixed. Specifically, we compared the actual response of each unit to first touch (red dots in Fig. 4B right) to that predicted by feeding the touch strength (Δκ3D) of the first touch through the tuning curve estimated for later touches (black dots in Fig. 4B right). For 85% of units, the responses to first touch predicted in this way were significantly lower than the actual responses (t-test for each unit (α = 0.05), mean p-value across units = 0.0023)—on average, 59% lower. This was also true on an individual animal basis (mean across mice 57% ± 17). Repeating this analysis for tuning curves estimated for second touch, third touch, etc. in turn showed gradual convergence to an adapted state (Fig. 4D). Both the slope and intercept of the tuning function for later touch decreased significantly compared to those for first touch (Fig. 4C), with the strongest effect observed for the intercept: 93% of units had higher intercept for first touch (mean across mice 97% ± 4); 63% had higher slope for first touch (mean across mice 61% ± 15). On average, the ratio between intercepts from later and first touches was 0.32 (range across individual mice 0.24–0.43). The intercept decreased gradually over the course of later touches (Fig. 4D right). These results indicate that a significant part of the response attenuation from first to later touches observed in Fig. 2 can be explained by a change in neuronal sensitivity. Thus, response attenuation is not simply a consequence of motor adaptation – there is also SA.

To estimate how much of the overall response attenuation in Fig. 2 is due to SA, for each simultaneously recorded population of units, we first determined the overall response attenuation from first to later touch due to the combined effects of motor adaptation and SA (calculated from Fig. 2B). Next, we estimated how much of this decrement could be attributed to the effect of tuning curves alone, keeping touch strength constant (Methods). We found that, on average across the sessions, 73% of the overall attenuation in firing rate is attributable to SA (range across individual mice 51–84%).

In sum, these results indicate that both Sensory Adaptation and Motor Adaptation contribute to the attenuation in firing rate observed during a sequence of whisker-pole touches, with Sensory Adaptation playing the major role.

Stimulus-specific adaptation

A hallmark of sensory adaptation in studies of anaesthetised animals is that adaptation can be specific to individual whiskers17,19. That is, stimulation of a given whisker attenuates responses to subsequent stimulation only of that whisker and not to that of other whiskers (whisker-specific adaptation, WSA). The hypothesis that SA is operative in the actively sensing, behaving animal suggests, therefore, that the changes in tuning curve that we reported above should be whisker-specific. To probe for WSA, we selected all trials in a session in which at least the first two touches were produced by the same whisker – the adapting whisker Wad (C1 or C2) – and at least one later touch (3rd or later) was made by a different whisker-the test whisker Wtest. Simultaneous touches were excluded. WSA predicts that the test stimulus should evoke an unadapted response.

To test this, we compared the responses (FRtest) of units to whisker Wtest touches satisfying the definition of ‘test stimuli’ to responses to whisker Wtest in the adapted state (that is, responses to the third or later touch of that whisker within a trial, Fig. 5A). To factor out potential differences in touch strength, we measured tuning curves of units to |Δκ3D| of that whisker in the adapted state, and used these tuning curves to predict responses to touches of identical |Δκ3D| to those measured for the test stimuli (FRpredicted). WSA predicts that FRtest should be greater than FRpredicted and that the following WSA index I should be positive:1 I=FRtest-FRpredictedFRtest+FRpredicted

Fig. 5 Whisker-specific Adaptation. (A) Comparison of response to the test stimulus with that predicted in the adapted state (see main text). Units whose responses were significantly higher than predicted in the adapted state are highlighted in green: * indicates P-value < 0.05. (B) Histogram of WSA index (I) values across the population.

Alternatively, if adaptation spreads across whiskers, then I should be zero. We found that units had significantly higher FRtest compared to FRpredicted (Fig. 5B, t-test p = 8 × 10–3; removing outlier p = 7 × 10–4), that 91% of units had positive I (Fig. 5B) and that, across the population, I was significantly positive (t-test; p < 2 × 10–7). These results suggest that adaptation in our awake, actively sensing paradigm did exhibit WSA.

Discussion

Our aim was to investigate the extent to which Sensory Adaptation (SA) is expressed during behaviour, under conditions of active sensation, where the sensory input to the brain is determined by the animal’s own motor actions. Our main finding was that the response of individual neurons in barrel cortex decreased whilst a mouse repeatedly touched an object, and that this was due to two distinct factors: first, a motor adaptation effect, whereby mice tuned their whisking movements so that later touches were mechanically weaker than early ones; second, a sensory adaptation effect, whereby individual neurons became progressively less sensitive to touch. Thus, under conditions where animals are actively sensing the environment, the activity in cortical circuits reflects a delicate interplay between neuronal and behavioural dynamics.

Most previous research on SA used ‘passive sensation’ paradigms, where the animal–usually anaesthetised—is a passive recipient of stimulation applied by the experimenter. The advantage of passive sensation is that the experimenter can deliver a train of identical stimuli and test for SA by comparing responses to them. Although it has long been recognised as important to test for SA in actively sensing, behaving animals, the challenge has been that, during natural active sensation, animals move their sense organs and this makes it problematic to compare responses across repeated stimuli. Our approach was to train mice to perform a task (pole detection) that naturally elicits a stimulus sequence of whisker-pole touches. By tracking the whiskers, we showed that the resulting whisker-object touches are markedly variable in their temporal pattern, duration and strength. To disentangle the effects of this variability on neuronal activity in barrel cortex from that of SA, we used 3D whisker tracking to detect whisker-pole touches and to estimate their strength28,42,54. This allowed us to measure how neurons are tuned to touch strength (the primary mechanical variable encoded by neurons in the ascending whisker pathway38,47–53 and how this tuning adapts over the course of a touch sequence.

Previous work on SA in the anaesthetised brain or brain slice has found that cortical neurons strongly adapt when stimulated by trains of sensory input with a stimulus rate of a few Hz or more12,15,30. However, of the few studies that have been conducted in awake animals, some report that SA is weak32,34,35 others that it is substantial10,20,55. These differences may reflect cellular/anatomical differences in the recorded neurons, including laminar location and the regions to which the neurons project. The differences may also reflect differences in behaviour such as temporal and mechanical aspects of touch, which is precisely why it is important to study SA under conditions representative of natural behaviour. In particular, since the whisker system is an active sensory system, it is insightful to study SA whilst animals actively control their sensory apparatus to solve behavioural tasks, and to do so in a manner that controls for differences in stimulus strength. Our results indicate that SA is indeed expressed during the course of active whisker exploratory touch sequences. Controlling for differences in mechanical touch strength revealed that neurons were more sensitive to the first touch in a trial than to later touches. Thus, under conditions of active whisker exploration, adaptation significantly affects neuronal responses and tuning.

There are multiple mechanisms which contribute to SA, and SA is known to depend on a host of parameters—stimulus frequency, stimulus magnitude and brain state2,9,32,56. This suggests that the mechanisms expressed in any particular experiment are likely to depend on its precise conditions. The first mechanism to be clearly established was synaptic depression in thalamocortical synapses. For SA in the anaesthetised brain, this effect is prominent15,57. On the basis that thalamic activity is higher in activated vs quiescent brain states, it was argued that thalamocortical synapses are chronically depressed in the awake brain and, consequently, that there is little scope for thalamocortical synaptic depression to produce SA in the activated brain32. However, subsequent research has found that mechanisms involving excitatory-inhibitory balance, neuromodulation by acetylcholine, subcortical adaptation and synchrony-sensitivity also contribute to SA2,9,15,58–61. Indeed, a recent study of awake mice, reported that, although SA could not be explained by synaptic depression of thalamocortical synapses, it could be explained by non-linear amplification of thalamic SA, due to the sensitivity of layer 4 cortical neurons to synchrony in thalamic activity10. Taken together, these studies suggest that mechanisms different to synaptic depression gain in relative prominence under awake, behaving conditions. Future work should further probe the mechanisms under these conditions. For this, simultaneous recordings of cortical neurons across layers, including of inhibitory interneurons, and their thalamic inputs will help to disentangle the role of thalamic vs cortical mechanisms, potential laminar differences and the role of intracortical inhibition. Additionally, recordings of local field potential might reveal insights into the interplay between sensory adaptation and the level of arousal of the animal. Given that neuromodulators such as acetylcholine can reduce cortical adaptation61, and the neuromodulatory input increases with the level of arousal and/or attention59, future work should consider its potential impact on SA in behaving animals.

Active sensation is the general principle that animals use motor actions to acquire sensory information and that they select those actions expected to yield the sensory information most useful to their current goal24,62. We found that the strength of whisker-pole contact decreased systematically during the course of a trial (motor adaptation). This is reminiscent of the ‘minimal impingement’ strategy whereby whisking amplitude decreases upon contact with an object63. It is also consistent with extensive evidence for active sensing in animals with mobile whiskers, including rats, mice and aquatic mammals26,40,63–67. Our finding that motor adaptation is similar on correct and incorrect trials suggests that its principal mechanism is insensitive to top-down signals from the brain’s decision-making centres and may involve anatomical loops relatively close to the periphery of the sensorimotor system. Overall, sensory and motor adaptation may be mechanisms to elicit strong neural signals in response to salient, unexpected events.

Acknowledgements

This work was funded by Biotechnology and Biological Sciences Research Council grants (BB/L007282/1, BB/ P021603/1), Medical Research Council grants (MR/L01064X7/1, MR/P005659/1) and a Weizmann UK Making Connections Grant to RSP, a CONICYT Becas Chile contract 72170371 to ACR and BBSRC grant (BB/V00817X/1) to MM. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Author contributions

A.C.R.: Conceptualization, Formal Analysis, Methodology, Software, Writing – Original Draft Preparation, Visualisation, Validation, Data Curation. M.S.E.L.: Investigation, Methodology, Data Curation, Writing – Review and Editing. I.S.: Investigation, Methodology, Writing – Review and Editing. M.M.: Conceptualization, Methodology, Writing – Review and Editing, Funding Acquisition R.S.P.: Conceptualization, Methodology, Formal Analysis, Software, Writing – Original Draft Preparation, Supervision, Project Administration, Funding Acquisition.

Data availability

Data used to produce the figures in this paper (electrophysiological recordings and tracking of behaviour) is available at Figshare (https://doi.org/10.6084/m9.figshare.22788425.v1). All code used to produce the figures in this paper was developed in MATLAB and is available in our GitHub repository https://github.com/AndreaColinsR/Sensory_Adaptation.

Competing interests

The authors declare no competing interests.

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Andrea Colins Rodriguez and Michaela S. E. Loft.
==== Refs
References

1. Adrian ED Zotterman Y The impulses produced by sensory nerve endings: Part 3. Impulses set up by touch and pressure J. Physiol. 1926 61 465 83 10.1113/jphysiol.1926.sp002308 16993807
Adrian, E. D. & Zotterman, Y. The impulses produced by sensory nerve endings: Part 3. Impulses set up by touch and pressure. J. Physiol. 61, 465–83. 10.1113/jphysiol.1926.sp002308 (1926).16993807 10.1113/jphysiol.1926.sp002308
2. Wang Q Webber RM Stanley GB Thalamic synchrony and the adaptive gating of information flow to cortex Nat. Neurosci. 2010 13 1534 1543 10.1038/nn.2670 21102447
Wang, Q., Webber, R. M. & Stanley, G. B. Thalamic synchrony and the adaptive gating of information flow to cortex. Nat. Neurosci. 13, 1534–1543. 10.1038/nn.2670 (2010).21102447 10.1038/nn.2670
3. Laughlin SB The role of sensory adaptation in the retina J. Exp. Biol. 1989 146 39 62 10.1242/jeb.146.1.39 2689569
Laughlin, S. B. The role of sensory adaptation in the retina. J. Exp. Biol. 146, 39–62. 10.1242/jeb.146.1.39 (1989).2689569 10.1242/jeb.146.1.39
4. Maravall M Alenda A Bale MR Petersen RS Transformation of adaptation and gain rescaling along the whisker sensory pathway PLoS One. 2013 8 e82418 10.1371/journal.pone.0082418 24349279
Maravall, M., Alenda, A., Bale, M. R. & Petersen, R. S. Transformation of adaptation and gain rescaling along the whisker sensory pathway. PLoS One. 8, e82418. 10.1371/journal.pone.0082418 (2013).24349279 10.1371/journal.pone.0082418
5. Maravall M Petersen RS Fairhall AL Arabzadeh E Diamond ME Shifts in coding properties and maintenance of information transmission during adaptation in barrel cortex PLoS Biol. 2007 5 0323 0334 10.1371/journal.pbio.0050019
Maravall, M., Petersen, R. S., Fairhall, A. L., Arabzadeh, E. & Diamond, M. E. Shifts in coding properties and maintenance of information transmission during adaptation in barrel cortex. PLoS Biol. 5, 0323–0334. 10.1371/journal.pbio.0050019 (2007).10.1371/journal.pbio.0050019
6. Petersen RS Panzeri S Maravall M Neural coding and contextual influences in the whisker system Biol. Cybern. 2009 100 427 446 10.1007/s00422-008-0290-5 19189120
Petersen, R. S., Panzeri, S. & Maravall, M. Neural coding and contextual influences in the whisker system. Biol. Cybern. 100, 427–446. 10.1007/s00422-008-0290-5 (2009).19189120 10.1007/s00422-008-0290-5
7. Díaz-Quesada M Maravall M Intrinsic mechanisms for adaptive gain rescaling in barrel cortex J. Neurosci. 2008 28 696 710 10.1523/JNEUROSCI.4931-07.2008 18199769
Díaz-Quesada, M. & Maravall, M. Intrinsic mechanisms for adaptive gain rescaling in barrel cortex. J. Neurosci. 28, 696–710. 10.1523/JNEUROSCI.4931-07.2008 (2008).18199769 10.1523/JNEUROSCI.4931-07.2008
8. Adibi M Lampl I Sensory adaptation in the Whisker-Mediated tactile system: Physiology, theory, and function Front. Neurosci. 2021 15 1 23 10.3389/fnins.2021.770011
Adibi, M. & Lampl, I. Sensory adaptation in the Whisker-Mediated tactile system: Physiology, theory, and function. Front. Neurosci. 15, 1–23. 10.3389/fnins.2021.770011 (2021).10.3389/fnins.2021.770011
9. Lampl I Katz Y Neuronal adaptation in the somatosensory system of rodents Neuroscience. 2017 343 66 76 10.1016/j.neuroscience.2016.11.043 27923742
Lampl, I. & Katz, Y. Neuronal adaptation in the somatosensory system of rodents. Neuroscience. 343, 66–76. 10.1016/j.neuroscience.2016.11.043 (2017).27923742 10.1016/j.neuroscience.2016.11.043
10. Wright NC Borden PY Liew YJ Bolus MF Stoy WM Forest CR Rapid cortical adaptation and the role of thalamic synchrony during wakefulness J. Neurosci. 2021 41 5421 5439 10.1523/JNEUROSCI.3018-20.2021 33986072
Wright, N. C. et al. Rapid cortical adaptation and the role of thalamic synchrony during wakefulness. J. Neurosci. 41, 5421–5439. 10.1523/JNEUROSCI.3018-20.2021 (2021).33986072 10.1523/JNEUROSCI.3018-20.2021
11. Maravall M Quian Quiroga R Panzeri P Adaptation and sensory coding Principles of neural coding 2013 CRC Press 357 378
Maravall, M. Adaptation and sensory coding. In Principles of neural coding (eds Quian Quiroga, R. & Panzeri, P.) 357–378 (CRC Press, 2013).
12. Ahissar E Sosnik R Haidarliu S Transformation from temporal to rate coding in a somatosensory thalamocortical pathway Nature. 2000 406 302 306 10.1038/35018568 10917531
Ahissar, E., Sosnik, R. & Haidarliu, S. Transformation from temporal to rate coding in a somatosensory thalamocortical pathway. Nature. 406, 302–306. 10.1038/35018568 (2000).10917531 10.1038/35018568
13. Weber AI Krishnamurthy K Fairhall AL Coding principles in adaptation Annu. Rev. Vis. Sci. 2019 5 427 449 10.1146/annurev-vision-091718-014818 31283447
Weber, A. I., Krishnamurthy, K. & Fairhall, A. L. Coding principles in adaptation. Annu. Rev. Vis. Sci. 5, 427–449. 10.1146/annurev-vision-091718-014818 (2019).31283447 10.1146/annurev-vision-091718-014818
14. Ulanovsky N Las L Farkas D Nelken I Multiple time scales of adaptation in auditory cortex neurons J. Neurosci. 2004 24 10440 10453 10.1523/JNEUROSCI.1905-04.2004 15548659
Ulanovsky, N., Las, L., Farkas, D. & Nelken, I. Multiple time scales of adaptation in auditory cortex neurons. J. Neurosci. 24, 10440–10453. 10.1523/JNEUROSCI.1905-04.2004 (2004).15548659 10.1523/JNEUROSCI.1905-04.2004
15. Chung S Li X Nelson SB Short-term depression at thalamocortical synapses contributes to rapid adaptation of cortical sensory responses in vivo Neuron. 2002 34 437 446 10.1016/S0896-6273(02)00659-1 11988174
Chung, S., Li, X. & Nelson, S. B. Short-term depression at thalamocortical synapses contributes to rapid adaptation of cortical sensory responses in vivo. Neuron. 34, 437–446. 10.1016/S0896-6273(02)00659-1 (2002).11988174 10.1016/S0896-6273(02)00659-1
16. Khatri V Hartings JA Simons DJ Adaptation in thalamic barreloid and cortical barrel neurons to periodic whisker deflections varying in frequency and velocity J. Neurophysiol. 2004 92 3244 3254 10.1152/jn.00257.2004 15306632
Khatri, V., Hartings, J. A. & Simons, D. J. Adaptation in thalamic barreloid and cortical barrel neurons to periodic whisker deflections varying in frequency and velocity. J. Neurophysiol. 92, 3244–3254. 10.1152/jn.00257.2004 (2004).15306632 10.1152/jn.00257.2004
17. Katz Y Heiss JE Lampl I Cross-whisker adaptation of neurons in the rat barrel cortex J. Neurosci. 2006 26 13363 13372 10.1523/JNEUROSCI.4056-06.2006 17182787
Katz, Y., Heiss, J. E. & Lampl, I. Cross-whisker adaptation of neurons in the rat barrel cortex. J. Neurosci. 26, 13363–13372. 10.1523/JNEUROSCI.4056-06.2006 (2006).17182787 10.1523/JNEUROSCI.4056-06.2006
18. Ganmor E Katz Y Lampl I Intensity-dependent adaptation of cortical and thalamic neurons is controlled by brainstem circuits of the sensory pathway Neuron. 2010 66 273 286 10.1016/j.neuron.2010.03.032 20435003
Ganmor, E., Katz, Y. & Lampl, I. Intensity-dependent adaptation of cortical and thalamic neurons is controlled by brainstem circuits of the sensory pathway. Neuron. 66, 273–286. 10.1016/j.neuron.2010.03.032 (2010).20435003 10.1016/j.neuron.2010.03.032
19. Musall S Haiss F Weber B von der Behrens W Deviant processing in the primary somatosensory cortex Cereb Cortex. 2017 27 863 876 10.1093/cercor/bhv283 26628563
Musall, S., Haiss, F., Weber, B. & von der Behrens, W. Deviant processing in the primary somatosensory cortex. Cereb Cortex. 27, 863–876. 10.1093/cercor/bhv283 (2017).26628563 10.1093/cercor/bhv283
20. Musall S von der Behrens W Mayrhofer JM Weber B Helmchen F Haiss F Tactile frequency discrimination is enhanced by circumventing neocortical adaptation Nat. Neurosci. 2014 17 1567 1573 10.1038/nn.3821 25242306
Musall, S. et al. Tactile frequency discrimination is enhanced by circumventing neocortical adaptation. Nat. Neurosci. 17, 1567–1573. 10.1038/nn.3821 (2014).25242306 10.1038/nn.3821
21. Yarbus AL Eye movements during perception of complex objects Eye Move. Vis. 1967 10.1007/978-1-4899-5379-7_8
Yarbus, A. L. Eye movements during perception of complex objects. Eye Move. Vis.10.1007/978-1-4899-5379-7_8 (1967).10.1007/978-1-4899-5379-7_8
22. Gibson JJ Observations on active touch Psychol. Rev. 1962 69 477 491 10.1037/h0046962 13947730
Gibson, J. J. Observations on active touch. Psychol. Rev. 69, 477–491. 10.1037/h0046962 (1962).13947730 10.1037/h0046962
23. Prescott TJ Diamond ME Wing AM Active touch sensing Philos. Trans. R. Soc. B Biol. Sci. 2011 366 2989 2995 10.1098/rstb.2011.0167
Prescott, T. J., Diamond, M. E. & Wing, A. M. Active touch sensing. Philos. Trans. R. Soc. B Biol. Sci. 366, 2989–2995. 10.1098/rstb.2011.0167 (2011).10.1098/rstb.2011.0167
24. Yang SCH Lengyel M Wolpert DM Active sensing in the categorization of visual patterns Elife. 2016 5 e12215 10.7554/eLife.12215 26880546
Yang, S. C. H., Lengyel, M. & Wolpert, D. M. Active sensing in the categorization of visual patterns. Elife. 5, e12215. 10.7554/eLife.12215 (2016).26880546 10.7554/eLife.12215
25. Campagner D Evans MH Loft MSE Petersen RS What the whiskers tell the brain Neuroscience. 2018 368 95 108 10.1016/j.neuroscience.2017.08.005 28843998
Campagner, D., Evans, M. H., Loft, M. S. E. & Petersen, R. S. What the whiskers tell the brain. Neuroscience. 368, 95–108. 10.1016/j.neuroscience.2017.08.005 (2018).28843998 10.1016/j.neuroscience.2017.08.005
26. O’Connor DH Clack NG Huber D Komiyama T Myers EW Svoboda K Vibrissa-based object localization in head-fixed mice J. Neurosci. 2010 30 1947 1967 10.1523/JNEUROSCI.3762-09.2010 20130203
O’Connor, D. H. et al. Vibrissa-based object localization in head-fixed mice. J. Neurosci. 30, 1947–1967. 10.1523/JNEUROSCI.3762-09.2010 (2010).20130203 10.1523/JNEUROSCI.3762-09.2010
27. Clack NG O’Connor DH Huber D Petreanu L Hires A Peron S Automated tracking of whiskers in videos of head fixed rodents PLoS Comput. Biol. 2012 8 e1002591 10.1371/journal.pcbi.1002591 22792058
Clack, N. G. et al. Automated tracking of whiskers in videos of head fixed rodents. PLoS Comput. Biol. 8, e1002591. 10.1371/journal.pcbi.1002591 (2012).22792058 10.1371/journal.pcbi.1002591
28. Petersen RS Colins Rodriguez A Evans MH Campagner D Loft MSE A system for tracking whisker kinematics and whisker shape in three dimensions PLoS Comput. Biol. 2020 16 e1007402 10.1371/journal.pcbi.1007402 31978043
Petersen, R. S., Colins Rodriguez, A., Evans, M. H., Campagner, D. & Loft, M. S. E. A system for tracking whisker kinematics and whisker shape in three dimensions. PLoS Comput. Biol. 16, e1007402. 10.1371/journal.pcbi.1007402 (2020).31978043 10.1371/journal.pcbi.1007402
29. Khatri V Bruno RM Simons DJ Stimulus-specific and stimulus-nonspecific firing synchrony and its modulation by sensory adaptation in the whisker-to-barrel pathway J. Neurophysiol. 2009 101 2328 2338 10.1152/jn.91151.2008 19279146
Khatri, V., Bruno, R. M. & Simons, D. J. Stimulus-specific and stimulus-nonspecific firing synchrony and its modulation by sensory adaptation in the whisker-to-barrel pathway. J. Neurophysiol. 101, 2328–2338. 10.1152/jn.91151.2008 (2009).19279146 10.1152/jn.91151.2008
30. Simons DJ Temporal and spatial integration in the rat SI vibrissa cortex J. Neurophysiol. 1985 54 615 635 10.1152/jn.1985.54.3.615 4045540
Simons, D. J. Temporal and spatial integration in the rat SI vibrissa cortex. J. Neurophysiol. 54, 615–635. 10.1152/jn.1985.54.3.615 (1985).4045540 10.1152/jn.1985.54.3.615
31. Erchova IA Lebedev MA Diamond ME Somatosensory cortical neuronal population activity across states of anaesthesia Eur. J. Neurosci. 2002 15 744 752 10.1046/j.0953-816x.2002.01898.x 11886439
Erchova, I. A., Lebedev, M. A. & Diamond, M. E. Somatosensory cortical neuronal population activity across states of anaesthesia. Eur. J. Neurosci. 15, 744–752. 10.1046/j.0953-816x.2002.01898.x (2002).11886439 10.1046/j.0953-816x.2002.01898.x
32. Castro-Alamancos MA Absence of rapid sensory adaptation in neocortex during information processing states Neuron. 2004 41 455 464 10.1016/S0896-6273(03)00853-5 14766183
Castro-Alamancos, M. A. Absence of rapid sensory adaptation in neocortex during information processing states. Neuron. 41, 455–464. 10.1016/S0896-6273(03)00853-5 (2004).14766183 10.1016/S0896-6273(03)00853-5
33. Fanselow EE Nicolelis MAL Behavioral modulation of tactile responses in the rat somatosensory system J. Neurosci. 1999 19 7603 7616 10.1523/jneurosci.19-17-07603.1999 10460266
Fanselow, E. E. & Nicolelis, M. A. L. Behavioral modulation of tactile responses in the rat somatosensory system. J. Neurosci. 19, 7603–7616. 10.1523/jneurosci.19-17-07603.1999 (1999).10460266 10.1523/jneurosci.19-17-07603.1999
34. Crochet S Petersen CCH Correlating whisker behavior with membrane potential in barrel cortex of awake mice Nat. Neurosci. 2006 9 608 610 10.1038/nn1690 16617340
Crochet, S. & Petersen, C. C. H. Correlating whisker behavior with membrane potential in barrel cortex of awake mice. Nat. Neurosci. 9, 608–610. 10.1038/nn1690 (2006).16617340 10.1038/nn1690
35. Crochet S Poulet JFA Kremer Y Petersen CCH Synaptic mechanisms underlying sparse coding of active touch Neuron. 2011 69 1160 1175 10.1016/j.neuron.2011.02.022 21435560
Crochet, S., Poulet, J. F. A., Kremer, Y. & Petersen, C. C. H. Synaptic mechanisms underlying sparse coding of active touch. Neuron. 69, 1160–1175. 10.1016/j.neuron.2011.02.022 (2011).21435560 10.1016/j.neuron.2011.02.022
36. Yamashita T Pala A Pedrido L Kremer Y Welker E Petersen CCH Membrane potential dynamics of neocortical projection neurons driving target-specific signals Neuron. 2013 80 1477 1490 10.1016/j.neuron.2013.10.059 24360548
Yamashita, T. et al. Membrane potential dynamics of neocortical projection neurons driving target-specific signals. Neuron. 80, 1477–1490. 10.1016/j.neuron.2013.10.059 (2013).24360548 10.1016/j.neuron.2013.10.059
37. Hentschke H Haiss F Schwarz C Central signals rapidly switch tactile processing in rat barrel cortex during whisker movements Cerebral Cortex. 2006 16 1142 1156 10.1093/cercor/bhj056 16221924
Hentschke, H., Haiss, F. & Schwarz, C. Central signals rapidly switch tactile processing in rat barrel cortex during whisker movements. Cerebral Cortex. 16, 1142–1156. 10.1093/cercor/bhj056 (2006).16221924 10.1093/cercor/bhj056
38. Campagner D Evans MH Bale MR Erskine A Petersen RS Prediction of primary somatosensory neuron activity during active tactile exploration Elife. 2016 5 e10696 10.7554/eLife.10696 26880559
Campagner, D., Evans, M. H., Bale, M. R., Erskine, A. & Petersen, R. S. Prediction of primary somatosensory neuron activity during active tactile exploration. Elife. 5, e10696. 10.7554/eLife.10696 (2016).26880559 10.7554/eLife.10696
39. Bray N Burrows FE Jones M Berwick J Allan SM Schiessl I Decreased haemodynamic response and decoupling of cortical gamma-band activity and tissue oxygen perfusion after striatal interleukin-1 injection J. Neuroinflammation. 2016 13 1 13 10.1186/s12974-016-0664-x 26728181
Bray, N. et al. Decreased haemodynamic response and decoupling of cortical gamma-band activity and tissue oxygen perfusion after striatal interleukin-1 injection. J. Neuroinflammation. 13, 1–13. 10.1186/s12974-016-0664-x (2016).26728181 10.1186/s12974-016-0664-x
40. Campagner D Evans MH Chlebikova K Colins-Rodriguez A Loft MSE Fox S Prediction of choice from competing mechanosensory and choice-memory cues during active tactile decision making J. Neurosci. 2019 39 3921 3933 10.1523/jneurosci.2217-18.2019 30850514
Campagner, D. et al. Prediction of choice from competing mechanosensory and choice-memory cues during active tactile decision making. J. Neurosci. 39, 3921–3933. 10.1523/jneurosci.2217-18.2019 (2019).30850514 10.1523/jneurosci.2217-18.2019
41. Marsh D Applied Geometry for Computer Graphics and CAD 2006 2 Springer
Marsh, D. Applied Geometry for Computer Graphics and CAD 2nd edn. (Springer, 2006).
42. Birdwell JA Solomon J Thajchayapong M Taylor M Cheely M Towal RB Biomechanical models for radial distance determination by the rat vibrissal system J. Neurophysiol. 2007 98 2439 2455 10.1152/jn.00707.2006 17553946
Birdwell, J. A. et al. Biomechanical models for radial distance determination by the rat vibrissal system. J. Neurophysiol. 98, 2439–2455. 10.1152/jn.00707.2006 (2007).17553946 10.1152/jn.00707.2006
43. Quist BW Hartmann MJZ Mechanical signals at the base of a rat vibrissa: The effect of intrinsic vibrissa curvature and implications for tactile exploration J. Neurophysiol. 2012 107 2298 2312 10.1152/jn.00372.2011 22298834
Quist, B. W. & Hartmann, M. J. Z. Mechanical signals at the base of a rat vibrissa: The effect of intrinsic vibrissa curvature and implications for tactile exploration. J. Neurophysiol. 107, 2298–2312. 10.1152/jn.00372.2011 (2012).22298834 10.1152/jn.00372.2011
44. Jun JJ Mitelut C Lai C Gratiy SL Anastassiou CA Harris TD Real-time spike sorting platform for high-density extracellular probes with ground-truth validation and drift correction bioRxiv. 2017 2 17
Jun, J. J. et al. Real-time spike sorting platform for high-density extracellular probes with ground-truth validation and drift correction. bioRxiv. 2, 17 (2017).
45. Safaai H von Heimendahl M Sorando JM Diamond ME Maravall M Coordinated population activity underlying texture discrimination in rat barrel cortex J. Neurosci. 2013 33 5843 5855 10.1523/JNEUROSCI.3486-12.2013 23536096
Safaai, H., von Heimendahl, M., Sorando, J. M., Diamond, M. E. & Maravall, M. Coordinated population activity underlying texture discrimination in rat barrel cortex. J. Neurosci. 33, 5843–5855. 10.1523/JNEUROSCI.3486-12.2013 (2013).23536096 10.1523/JNEUROSCI.3486-12.2013
46. Knutsen PM Biess A Ahissar E Vibrissal kinematics in 3D: Tight coupling of azimuth, elevation, and torsion across different whisking modes Neuron. 2008 59 35 42 10.1016/j.neuron.2008.05.013 18614027
Knutsen, P. M., Biess, A. & Ahissar, E. Vibrissal kinematics in 3D: Tight coupling of azimuth, elevation, and torsion across different whisking modes. Neuron. 59, 35–42. 10.1016/j.neuron.2008.05.013 (2008).18614027 10.1016/j.neuron.2008.05.013
47. Bush NE Schroeder CL Hobbs JA Yang AE Huet LA Solla SA Decoupling kinematics and mechanics reveals coding properties of trigeminal ganglion neurons in the rat vibrissal system Elife. 2016 5 e13969 10.7554/eLife.13969 27348221
Bush, N. E. et al. Decoupling kinematics and mechanics reveals coding properties of trigeminal ganglion neurons in the rat vibrissal system. Elife. 5, e13969. 10.7554/eLife.13969 (2016).27348221 10.7554/eLife.13969
48. Severson KS Xu D Van de Loo M Bai L Ginty DD O’Connor DH Active touch and self-motion encoding by merkel cell-associated afferents Neuron. 2017 94 666 676.e9 10.1016/j.neuron.2017.03.045 28434802
Severson, K. S. et al. Active touch and self-motion encoding by merkel cell-associated afferents. Neuron. 94, 666-676.e9. 10.1016/j.neuron.2017.03.045 (2017).28434802 10.1016/j.neuron.2017.03.045
49. Tonomura S Ebara S Bagdasarian K Uta D Ahissar E Meir I Structure-function correlations of rat trigeminal primary neurons: Emphasis on club-like endings, a vibrissal mechanoreceptor Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 2015 91 560 576 10.2183/pjab.91.560 26666306
Tonomura, S. et al. Structure-function correlations of rat trigeminal primary neurons: Emphasis on club-like endings, a vibrissal mechanoreceptor. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 91, 560–576. 10.2183/pjab.91.560 (2015).26666306 10.2183/pjab.91.560
50. Petreanu L Gutnisky DA Huber D Xu NL Oconnor DH Tian L Activity in motor-sensory projections reveals distributed coding in somatosensation Nature. 2012 489 299 303 10.1038/nature11321 22922646
Petreanu, L. et al. Activity in motor-sensory projections reveals distributed coding in somatosensation. Nature. 489, 299–303. 10.1038/nature11321 (2012).22922646 10.1038/nature11321
51. Peron SP Freeman J Iyer V Guo C Svoboda K A cellular resolution map of barrel cortex activity during tactile behavior Neuron. 2015 86 783 799 10.1016/j.neuron.2015.03.027 25913859
Peron, S. P., Freeman, J., Iyer, V., Guo, C. & Svoboda, K. A cellular resolution map of barrel cortex activity during tactile behavior. Neuron. 86, 783–799. 10.1016/j.neuron.2015.03.027 (2015).25913859 10.1016/j.neuron.2015.03.027
52. Hires SA Gutnisky DA Yu J O’Connor DH Svoboda K Low-noise encoding of active touch by layer 4 in the somatosensory cortex Elife. 2015 4 e06619 10.7554/eLife.06619 26245232
Hires, S. A., Gutnisky, D. A., Yu, J., O’Connor, D. H. & Svoboda, K. Low-noise encoding of active touch by layer 4 in the somatosensory cortex. Elife. 4, e06619. 10.7554/eLife.06619 (2015).26245232 10.7554/eLife.06619
53. Sofroniew N Vlasov YA Hires SA Freeman J Svoboda K Neural coding in barrel cortex during whisker-guided locomotion Elife. 2015 4 1 19 10.7554/eLife.12559
Sofroniew, N., Vlasov, Y. A., Hires, S. A., Freeman, J. & Svoboda, K. Neural coding in barrel cortex during whisker-guided locomotion. Elife. 4, e12559. 10.7554/eLife.12559 (2015).10.7554/eLife.12559
54. Pammer L O’Connor DH Hires SA Clack NG Huber D Myers EW The mechanical variables underlying object localization along the axis of the whisker J. Neurosci. 2013 33 6726 6741 10.1523/JNEUROSCI.4316-12.2013 23595731
Pammer, L. et al. The mechanical variables underlying object localization along the axis of the whisker. J. Neurosci. 33, 6726–6741. 10.1523/JNEUROSCI.4316-12.2013 (2013).23595731 10.1523/JNEUROSCI.4316-12.2013
55. Ollerenshaw DR Zheng HJV Millard DC Wang Q Stanley GB The adaptive trade-off between detection and discrimination in cortical representations and behavior Neuron. 2014 81 1152 1164 10.1016/j.neuron.2014.01.025 24607233
Ollerenshaw, D. R., Zheng, H. J. V., Millard, D. C., Wang, Q. & Stanley, G. B. The adaptive trade-off between detection and discrimination in cortical representations and behavior. Neuron. 81, 1152–1164. 10.1016/j.neuron.2014.01.025 (2014).24607233 10.1016/j.neuron.2014.01.025
56. Katz Y Okun M Lampl I Trial-to-trial correlation between thalamic sensory response and global EEG activity Eur. J. Neurosci. 2012 35 826 837 10.1111/j.1460-9568.2012.08006.x 22384999
Katz, Y., Okun, M. & Lampl, I. Trial-to-trial correlation between thalamic sensory response and global EEG activity. Eur. J. Neurosci. 35, 826–837. 10.1111/j.1460-9568.2012.08006.x (2012).22384999 10.1111/j.1460-9568.2012.08006.x
57. Castro-Alamancos MA Oldford E Cortical sensory suppression during arousal is due to the activity-dependent depression of thalamocortical synapses J. Physiol. 2002 541 319 331 10.1113/jphysiol.2002.016857 12015438
Castro-Alamancos, M. A. & Oldford, E. Cortical sensory suppression during arousal is due to the activity-dependent depression of thalamocortical synapses. J. Physiol. 541, 319–331. 10.1113/jphysiol.2002.016857 (2002).12015438 10.1113/jphysiol.2002.016857
58. Gabernet L Jadhav SP Feldman DE Carandini M Scanziani M Somatosensory integration controlled by dynamic thalamocortical feed-forward inhibition Neuron. 2005 48 315 327 10.1016/j.neuron.2005.09.022 16242411
Gabernet, L., Jadhav, S. P., Feldman, D. E., Carandini, M. & Scanziani, M. Somatosensory integration controlled by dynamic thalamocortical feed-forward inhibition. Neuron. 48, 315–327. 10.1016/j.neuron.2005.09.022 (2005).16242411 10.1016/j.neuron.2005.09.022
59. Himmelheber AM Sarter M Bruno JP Increases in cortical acetylcholine release during sustained attention performance in rats Cogn. Brain Res. 2000 9 313 325 10.1016/S0926-6410(00)00012-4
Himmelheber, A. M., Sarter, M. & Bruno, J. P. Increases in cortical acetylcholine release during sustained attention performance in rats. Cogn. Brain Res. 9, 313–325. 10.1016/S0926-6410(00)00012-4 (2000).10.1016/S0926-6410(00)00012-4
60. Nuñez A Domínguez S Buño W Fernández de SD. Cholinergic-mediated response enhancement in barrel cortex layer V pyramidal neurons J. Neurophysiol. 2012 108 1656 1668 10.1152/jn.00156.2012 22723675
Nuñez, A., Domínguez, S., Buño, W. & Fernández de SD.,. Cholinergic-mediated response enhancement in barrel cortex layer V pyramidal neurons. J. Neurophysiol. 108, 1656–1668. 10.1152/jn.00156.2012 (2012).22723675 10.1152/jn.00156.2012
61. Martin-Cortecero J Nuñez A Tactile response adaptation to whisker stimulation in the lemniscal somatosensory pathway of rats Brain Res. 2014 1591 27 37 10.1016/j.brainres.2014.10.002 25307139
Martin-Cortecero, J. & Nuñez, A. Tactile response adaptation to whisker stimulation in the lemniscal somatosensory pathway of rats. Brain Res. 1591, 27–37. 10.1016/j.brainres.2014.10.002 (2014).25307139 10.1016/j.brainres.2014.10.002
62. Ahissar E Assa E Perception as a closed-loop convergence process Elife. 2016 10.7554/eLife.12830.001 27159238
Ahissar, E. & Assa, E. Perception as a closed-loop convergence process. Elife. 5, e12830 10.7554/eLife.12830.001 (2016).27159238 10.7554/eLife.12830.001
63. Mitchinson B Martin CJ Grant RA Prescott TJ Feedback control in active sensing: Rat exploratory whisking is modulated by environmental contact Proc. R. Soc. B Biol. Sci. 2007 274 1035 1041 10.1098/rspb.2006.0347
Mitchinson, B., Martin, C. J., Grant, R. A. & Prescott, T. J. Feedback control in active sensing: Rat exploratory whisking is modulated by environmental contact. Proc. R. Soc. B Biol. Sci. 274, 1035–1041. 10.1098/rspb.2006.0347 (2007).10.1098/rspb.2006.0347
64. Towal RB Hartmann MJ Right-left asymmetries in the whisking behavior of rats anticipate head movements J. Neurosci. 2006 26 8838 8846 10.1523/JNEUROSCI.0581-06.2006 16928873
Towal, R. B. & Hartmann, M. J. Right-left asymmetries in the whisking behavior of rats anticipate head movements. J. Neurosci. 26, 8838–8846. 10.1523/JNEUROSCI.0581-06.2006 (2006).16928873 10.1523/JNEUROSCI.0581-06.2006
65. Sofroniew N Cohen JD Lee AK Svoboda K Natural whisker-guided behavior by head-fixed mice in tactile virtual reality J. Neurosci. 2014 34 9537 9550 10.1523/JNEUROSCI.0712-14.2014 25031397
Sofroniew, N., Cohen, J. D., Lee, A. K. & Svoboda, K. Natural whisker-guided behavior by head-fixed mice in tactile virtual reality. J. Neurosci. 34, 9537–9550. 10.1523/JNEUROSCI.0712-14.2014 (2014).25031397 10.1523/JNEUROSCI.0712-14.2014
66. Arkley K Grant RA Mitchinson B Prescott TJ Strategy change in vibrissal active sensing during rat locomotion Curr. Biol. 2014 24 1507 1512 10.1016/j.cub.2014.05.036 24954047
Arkley, K., Grant, R. A., Mitchinson, B. & Prescott, T. J. Strategy change in vibrissal active sensing during rat locomotion. Curr. Biol. 24, 1507–1512. 10.1016/j.cub.2014.05.036 (2014).24954047 10.1016/j.cub.2014.05.036
67. Grant R Wieskotten S Wengst N Prescott T Dehnhardt G Vibrissal touch sensing in the harbor seal (Phoca vitulina): How do seals judge size? J. Comp. Physiol. A Neuroethol. Sens. Neural Behav. Physiol. 2013 199 521 533 10.1007/s00359-013-0797-7 23397461
Grant, R., Wieskotten, S., Wengst, N., Prescott, T. & Dehnhardt, G. Vibrissal touch sensing in the harbor seal (Phoca vitulina): How do seals judge size?. J. Comp. Physiol. A Neuroethol. Sens. Neural Behav. Physiol. 199, 521–533. 10.1007/s00359-013-0797-7 (2013).23397461 10.1007/s00359-013-0797-7
