Somatosensory, visual, auditory, and gustatory cortex are each reciprocally connected with a specific subset of thalamic nuclei. These nuclei can be subdivided into primary and secondary (often termed ‘higher-order’) nuclei (Guillery and Sherman, 2002; Herkenham, 1980; Phillips et al., 2019). The primary nuclei are the main source of sensory input to the cortex and respond robustly to sensory stimulation with low latency (Chiaia et al., 1991; Constantinople and Bruno, 2013; Sherman and Guillery, 2002; Wimmer et al., 2010). Unlike primary nuclei, the secondary nuclei are interconnected with many cortical and subcortical regions, and their role in sensation and cognition is poorly understood.

In rodents, the facial whisker representation of primary somatosensory cortex (S1) is tightly integrated with two thalamic nuclei: the ventral posteromedial nucleus (VPM) and the posterior medial nucleus (POm) (Deschênes et al., 2016; Petersen, 2007). Compared to the primary nucleus VPM, the secondary nucleus POm has broader receptive fields, longer-latency sensory responses, and poorly encodes fine aspects of whisker touch such as contact timing and stimulus frequency (Diamond et al., 1992; Masri et al., 2008; Moore, 2004; Moore et al., 2015). It receives input from S1, motor cortex, posterior parietal cortex, the zona incerta, and many other subcortical regions in addition to brainstem afferents (Chiaia et al., 1991; Olsen and Witter, 2016; Trageser and Keller, 2004). Whereas VPM innervates cortical layer 4 and the border of layers 5B and 6, POm projects to the apical dendrites of layer 1 as well as layer 5A (Wimmer et al., 2010). POm is a stronger driver of layer 2/3 cells than cortico-cortical synapses and can enhance sensory responses in pyramidal neurons of layers 2/3 and 5 (Mease et al., 2016; Zhang and Bruno, 2019). POm is thus positioned to strongly influence sensory computations in S1 and do so in ways that are highly distinct from VPM. However, what POm activity encodes remains a mystery.

One possibility is that POm activity encodes self-generated movements, through either sensory reafference (stimulation of the sensory receptors by active movement) or motor efference copy (internal copies of motor commands), rather than extrinsic tactile sensations (Yu et al., 2006). If secondary thalamus were a monitor of movements (Sherman and Guillery, 2002), somatosensory cortex could use POm input to differentiate self-generated and externally generated sensory signals. However, recent studies in awake animals have observed that, in comparison to VPM, POm encodes whisker motion and contact less well (Moore et al., 2015; Urbain et al., 2015), which casts doubt on the hypothesis that secondary pathways exist to provide detailed motor information to cortex.

An alternative hypothesis is that secondary thalamic nuclei are key structures for monitoring behavioral state. For instance, several studies have noted that a subset of POm neurons is activated by pain (Frangeul et al., 2014; Masri et al., 2009), a powerful stimulus that can trigger a change in animals’ state. Spatial attention is a more subtle form of behavioral state change and has been implicated repeatedly in studies of primate secondary visual thalamus (lateral pulvinar) (Petersen et al., 1987; Saalmann et al., 2012; Wilke et al., 2010). The rodent homolog to the pulvinar (lateral posterior nucleus, LP) is active during mismatch of movement and visual stimuli (Roth et al., 2016), which might reflect elevations in visual attention or even global arousal. These results raise the possibility that modulation by behavioral state is a general feature of all secondary nuclei.

Here, we investigate how afferent, corticothalamic, and collicular inputs—the three main excitatory pathways to secondary sensory thalamus—influence encoding of movements by POm in the awake mouse. We discovered that removing these circuits enhances rather than reduces modulation of POm activity by movements, suggesting that these pathways may mainly transmit signals of a nature other than movement. We further examine how POm activity relates to that of LP—which has not been directly compared before—to investigate general principles of secondary thalamus function. This comparison reveals that behavioral state, rather than movement itself, prominently dictates the activity of secondary thalamic nuclei.

We characterized the degree to which POm encodes whether or not an animal is whisking versus the fine details of whisking movements. We recorded juxtasomally from single neurons in head-fixed mice while acquiring high-speed video of the contralateral whisker field, from which whisker positions could be algorithmically extracted (Figure 1A and B; Clack et al., 2012). To measure slow aspects of whisking, we calculated whisking amplitude from the median angle across all whiskers. Whisking amplitude is defined as the difference in angle between the minimum and maximum protraction over the whisking cycle (Hill et al., 2011; Moore et al., 2015, see Methods). Whisking amplitude was used to determine periods of quiescence and whisking, as defined by periods of time when whisking amplitude exceeded 20% of the maximum for more than 250 ms (Figure 1B, shaded areas).

Figure 1 with 2 supplements see all

Download asset Open asset

Whisking substantially elevated POm firing rates. We computed the mean firing rate for each cell during periods when the mouse was whisking versus quiescent (Figure 1A ; 22 POm neurons in five mice; recording sites histologically confirmed post hoc, Figure 1—figure supplement 1). The firing rates of POm cells were significantly higher during bouts of whisking, increasing from a mean firing rate of 7.8 Hz to 12.4 Hz (58.5% increase, p<10^–4, paired t-test). To understand which components of whisking might drive POm activity, we calculated the cross-correlation between POm firing rate and three features of whisking activity: the median angle across all whiskers (Figure 1D, gray), the amplitude metric which captures the slow envelope of whisking (green), and the median angle bandpass-filtered from 4 Hz to 30 Hz (black), which reflects fast protractions and retractions of the whiskers. We found that POm neurons had little correlation with the bandpass-filtered angle, but prominent correlations with both whisker angle and whisking amplitude around a time lag of zero. The strongest correlate of POm activity was whisking amplitude, suggesting that POm is coupled to the slow components of whisking, rather than individual whisk cycles.

To investigate the encoding of the fast components of whisking in POm, we analyzed whether individual cells preferred to discharge during a certain phase of the whisking bout. We quantified the phase of whisking by applying the Hilbert transform to the 4–30 Hz bandpass-filtered median whisker. We identified the phase at which each action potential occurred during whisking and plotted distributions of firing rate as a function of phase. For each cell, we fit a sinusoid to characterize the cell’s preferred phases (the phase of the whisk cycle that elicited the highest firing rate) and modulation depth (the degree to which phase impacts firing rate, measured as the peak-trough difference normalized by mean firing rate). Figure 1E shows the phase relationship of two example cells: one with significant coding (top) and the other insignificant (bottom). Most POm cells (18/22) resembled the non-phase coding example, having little or no modulation (right). Taken together, these results indicate that the majority of POm cells do not encode fast whisking dynamics such as whisker angle or the phase of the whisking cycle. Rather, they track overall whisking activity, that is, whisking versus quiescence.

To further characterize the relationship between whisking and POm activity, we analyzed the coherence of these signals in the frequency domain. A spectral density analysis of whisking angle indicated that the majority of whisking power is concentrated at low frequencies (<5 Hz) with a prominent peak at higher frequencies (8–13 Hz), corresponding to the dynamics of individual whisk cycles (Figure 1—figure supplement 2A). However, the mean-squared coherence of POm activity and whisking angle was significantly different from chance only in the low-frequency band (Figure 1—figure supplement 2B). For individual cells, mean coherence was significantly greater in the low-frequency band than in the high-frequency band (p=0.0004, paired t-test; Figure 1—figure supplement 2C). This was true even of cells with significant phase modulation depth. Thus, both time- and frequency-domain analyses confirm that POm output predominantly tracks the slow envelope of whisking.

One possible source of whisking-related activity is reafferent sensory input: when the mouse whisks, the self-generated movement could deform the whisker follicle and stimulate mechanoreceptors. To measure the degree to which POm activity is driven by the sensory reafference caused by whisking, we severed the buccal and upper marginal branches of the facial motor nerve on the right side of the face, contralateral to our recordings while taking video of the left (ipsilateral) side of the face (Figure 2A). This manipulation does not damage the sensory neurons and avoids the risk of inducing sensory neuron plasticity (Shetty et al., 2003). Mice were no longer able to move the right whisker pad but whisking on the left side of the face was unaffected (Figure 2B and C). Without whisker movement, there can be no reafferent sensory input from the right whisker pad. As in intact mice, firing rates of POm neurons in nerve cut animals were significantly higher during whisking bouts (Figure 2D, n=12, p=0.0007, paired t-test) and to a similar degree, increasing from quiescent firing rate of 11.6 Hz to a whisking rate of 16.7 Hz (44%). POm firing rates also correlated with ipsilateral whisking amplitude, at a similar magnitude and with a similar lag as in the contralateral whisker field in intact mice (Figure 2E). This demonstrates that the correlation of POm activity and overall whisking is not due to ascending reafferent information.

Figure 2

Download asset Open asset

We also calculated the phase coding of the ipsilateral whisker field (Figure 2F) and compared it to the phase coding of the contralateral whisker field (Figure 1E). While average modulation depth was unchanged (Figure 2E, p=0.12, Wilcoxon rank-sum test), modulation depth by definition is bounded at zero, complicating analysis of medians close to zero. Indeed, there was a noticeable and statistically significant decrease in the range of modulation depths in the transected group (Figure 2G, p=0.0013, two-sample F-test), consistent with eliminating the largest modulation values. These results suggest that reafferent signals do not contribute to changes in POm activity reflecting the slow envelope of whisking (Figure 2E) but are responsible for the small population of POm cells carrying fast phase information (Figure 2F).

POm is reciprocally connected to several cortical areas, potentially making cortical output a strong driver of POm activity (Chiaia et al., 1991; Diamond et al., 2008; Mease et al., 2016). Primary motor cortex (M1) activity encodes the envelope of the whisking in absence of sensory input or motor output (Ahrens and Kleinfeld, 2004; Fee et al., 1997). Output from S1 and M1 could convey sensorimotor information, such as a motor efference copy, that would drive whisking-related activity, independent of ascending sensory input. To test the role of cortical output, we set out to study POm activity during optogenetic suppression of M1 (Figure 3A). We expressed halorhodopsin in all excitatory cortical neurons by crossing Emx1-Cre mice with a conditional halorhodopsin responder line, a technique we previously used to silence S1 (Hong et al., 2018). We first confirmed that the same technique silences M1 by recording it with a 64-channel silicon electrode array. Out of the 131 M1 cells recorded, 50 were fully silenced by the laser and a further 76 were substantially inhibited (Figure 3B, mean laser-on firing rate=17.6% of baseline, median=1.04%). Inhibition was effective across cortical layers, and the laser decreased the firing rates of cells even deeper than 1200 µm from the surface (mean laser-on firing rate of deeper cells=60.1% of baseline, median=28.5%).

Figure 3

Download asset Open asset

M1 suppression reduced the baseline firing rate of POm cells, but POm activity was still elevated during whisking regardless of whether the laser was on or off (Figure 3C). Surprisingly, suppressing M1 increased the correlation between POm firing rate and whisking amplitude (Figure 3D). This suggests that POm encoding of fine whisking kinematics arises from ascending sensory reafference rather than input from motor cortex.

To confirm that these effects were due to inhibition of M1 inputs to POm and not an artifact of optogenetic-induced changes in whisking behavior, we also recorded from cells in VPM. VPM, which does not receive direct projections from M1, was largely unaffected by M1 inhibition. We observed no effect of inhibition on VPM firing rates or cross-correlation between VPM activity and whisking (Figure 3E and F).

In a parallel set of experiments, we silenced S1 using the same cre-dependent halorhodopsin line (Figure 4A). This technique robustly blocks action potentials throughout all cortical layers of S1 in awake behaving mice (Hong et al., 2018). Silencing S1 reduced POm activity whether mice were whisking or quiescent (Figure 4B, n=11 cells, three mice; whisking p=0.0002, laser p=0.024, two-way repeated-measures ANOVA). As in the M1 experiments, the correlation between whisking amplitude and POm activity was, if anything, unchanged or increased by S1 silencing (Figure 4C, laser-off peak correlation=0.036, laser-on peak correlation=0.081). There was a tendency for S1 inhibition to reduce overall activity in VPM, possibly reflecting the known corticothalamic connections between S1 and VPM (Bourassa and Descheˆnes, 1995; Hoogland et al., 1987; Killackey and Sherman, 2003), but this effect did not reach significance (Figure 4D; p=0.1). Suppressing S1 had little impact on the correlation of VPM spiking and whisking (Figure 4E).

Figure 4

Download asset Open asset

Thus, both optogenetic manipulations had qualitatively different effects on VPM and POm activity, consistent with the known anatomical differences in corticothalamic projections onto these two nuclei. Taken together, these results demonstrate that POm does not inherit information about whisking amplitude from M1 or S1. Rather, corticothalamic inputs appear to transmit signals other than whisker movements, and these additional signals normally reduce the correlation of POm activity with whisking amplitude and phase.

In addition to afferent inputs from brainstem and efferent inputs from cortex, POm receives excitatory projections from the superior colliculus (SC) (Gharaei et al., 2019), which could also provide a motor efference copy signal similar to known collicular circuits in the visual system (Berman and Wurtz, 2010). SC receives excitatory input from both the trigeminal brainstem and cortex and is whisking modulated (Castro-Alamancos and Bezdudnaya, 2015; Castro-Alamancos and Favero, 2016), making SC a potential source of whisking-related POm activity. To test this possibility, we performed bilateral electrolytic lesions in SC and subsequently recorded POm cells (Figure 5A). Whisking had similar effects on POm activity in both intact and lesioned animals (Figure 5B, n=49 cells from eight animals, 59% increase in mean firing rate, p<10^–9). POm firing rates of lesioned mice were overall higher than those of intact animals, independent of whether animals were whisking or quiescent (Figure 5C, lesion p<10^–3, whisking p<10^–5, two-way ANOVA). There was a slight tendency for SC-lesioned animals to whisk more frequently than intact animals, but this effect was not statistically significant (Figure 5D, p=0.35, Wilcoxon rank-sum test). While small parts of SC remained intact (Figure 5A), these results with such large lesions suggest that SC is not responsible for the whisking-induced elevation of POm activity.

Figure 5

Download asset Open asset

Neither reafference nor the most likely sources of motor efference copy explain the coarse modulation of POm by movement. This raises the question of whether POm encodes movement per se, or another variable that is coupled with whisking and other movements, such as arousal. To investigate this, we measured pupil diameter, which is a known metric of arousal. We acquired videos of the pupil and whiskers while recording from POm (Figure 6A). Pupil diameter was tightly correlated with whisking, with pupil dilation lagging whisking amplitude by 880 ms on average (Figure 6B). Pupil diameter also correlated with POm activity, to a similar degree as whisking and with a lag of 950 ms (Figure 6C, whisking amplitude peak correlation=0.052, pupil diameter peak correlation=0.071, p=0.23, paired t-test).

Figure 6

Download asset Open asset

We reasoned that, if the modulation of POm by whisking was truly due to whisker movement rather than some other correlated variable, non-somatosensory thalamic nuclei would not be expected to track whisking. The secondary visual thalamic nucleus LP is the rodent homolog of the primate lateral pulvinar. LP is primarily coupled with cortical and subcortical visual areas (Nakamura et al., 2015), rather than somatosensory ones. Because of their different connectivity, POm and LP are expected to carry separate sensorimotor signals related to somatosensation and vision, respectively. Therefore, LP would not be expected to encode whisker movement. By contrast, changes in behavioral state, such as overall animal arousal as suggested by our pupil measurements, might modulate all thalamic nuclei, including LP and POm.

We tested this idea by recording juxtasomally from LP neurons (Figure 7A and B; 29 cells from four mice). Surprisingly, we found that LP activity was significantly increased during whisking bouts (Figure 7C, increase from 13.0 Hz to 18.0 Hz, p<10^–4, paired t-test). Like POm, LP activity correlated with both whisking amplitude and median whisker angle with low latency (Figure 7D). Since changes in pupil diameter will cause more light to fall on the retina, the LP correlation with whisking might be an artifact of pupil dilation. To control for this, a subset of cells was recorded in low light. Under these darker conditions, the pupil was maximally dilated and did not change (Figure 7B), rendering input to the retina largely constant. However, these cells still showed an equivalent increase in firing rate during whisking (Figure 7C, orange; n=29 cells, four animals; increase from a mean of 13.0 Hz to 18.0 Hz, or 39%, p<10^–4, paired t-test). Thus, LP activity appears to track whisking independent of changes in visual input, which suggests that the effect in both nuclei is due to the arousal-whisking correlation rather than a direct effect of whisking.

Figure 7 with 1 supplement see all

Download asset Open asset

The activity of individual LP cells was more coherent with the low-frequency components of whisking than high frequencies (Figure 7—figure supplement 1A). Overall LP-whisking coherence was identical to POm-whisking coherence in both the low- and high-frequency bands (Figure 7—figure supplement 1B).

Taken together, our results indicate that the slow component of whisking-related activity in POm is neither a consequence of ascending motion signals from reafferent mechanisms nor corticothalamic or colliculothalamic efferent mechanisms. We conclude instead that behavioral state, such as arousal may strongly dictate the activity of secondary thalamic nuclei, including POm and LP.

Our study tested the idea that the secondary somatosensory thalamus is a monitor of movements or motor commands and manipulated the multiple known pathways to POm that could mediate such signals. Juxtasomal recordings of POm cells revealed that this nucleus mainly tracks slow components of whisking, not detailed kinematics and that mouse POm firing rates are much higher during bouts of whisking than when a mouse is quiescent, consistent with other studies (Moore et al., 2015; Urbain et al., 2015). POm activity mainly correlates with the slow change in whisking amplitude rather than the fast changes of the whisk cycle. Only a minority of our POm cells exhibited any whisking phase information, and phase encoding appeared to depend on sensory reafference. We have demonstrated that, by contrast, the overall elevation of POm activity by whisking is not due to sensory reafference from self-generated movements, as transection of the facial motor nerve did not uncouple POm activity from ipsilateral whisking. We showed that potential motor efference copy via corticothalamic pathways from S1 and M1 cannot account for whisking modulation of POm. Similarly, the phenomenon is independent of SC, the activity of which is linked to movement and orienting.

What appears to be movement-related activity in POm is likely instead a consequence of the encoding of behavioral state. Activity in secondary visual thalamus (LP) exhibits the same correlation with whisker movement that we observed in POm. Though it is possible that POm and LP separately encode correlated sensorimotor information, a more parsimonious explanation is that both POm and LP are modulated by arousal, which is naturally elevated during movement. Modulation of activity by the behavioral state may be a general property of all secondary thalamic nuclei. Future studies are needed to examine if this principle holds in auditory thalamic subnuclei and perhaps even thalamic nuclei connected to motor cortex and other frontal areas. Conceivably, some movement correlations seen even in motor thalamus (Guo et al., 2017) may reflect various states more than specific movements.

The paralemniscal system has been speculated to be a parallel secondary afferent pathway (Frangeul et al., 2014; Yu et al., 2006). However, in anesthetized rats, POm does not appear to be sensitive to fine aspects of whisker touch, having very large receptive fields and long-latency responses (Diamond et al., 1992; Trageser and Keller, 2004). One might expect that very large synchronized movements of the whiskers, such as during whisking, would elicit a response from POm due to sensory reafference driving coarse receptive fields. However, paralyzing the face did not uncouple POm activity from ipsilateral whisking amplitude (Figure 2). Similarly, mouse barrel cortex is also modulated by whisking and quiescence in absence of sensory input: whisking is associated with a decrease in synchrony between layer 2/3 pyramidal cells in S1 and an increase in discharges by VPM, which is unaltered by bilateral transection of the infraorbital nerve sensory nerve (Poulet and Petersen, 2008). Manipulations of somatosensory thalamus strongly impacted cortical synchronization (Poulet et al., 2012). Further studies are needed to parcel out the extent to which thalamic contributions to cortical synchronization is due to inputs from VPM, POm, or both.

POm receives descending input from many cortical regions including M1 and S1. Conceivably these inputs could modulate ascending sensory input or provide the thalamus with a motor efference copy (Mease et al., 2016; Sherman, 2016). Similarly, LP and LGN axons in V1 exhibit eye movement-related signals (Roth et al., 2016). Previous studies in anesthetized rats have shown that cortical inactivation will silence POm, but not VPM (Diamond et al., 1992). Therefore, cortex might be the primary source of excitatory input to POm. However, we discovered that, in the awake mouse, silencing either M1 or S1 only slightly reduces the firing rate of POm cells and has little to no effect on VPM activity (Figure 3). We conclude that, while S1 and M1 provide significant excitatory inputs to POm, these inputs are not the sole drivers of POm activity during wakefulness.

Moreover, silencing these corticothalamic pathways increased rather than decreased the correlation between POm activity and whisking amplitude. If POm activity were primarily representative of a cortical efference copy, we would expect the opposite effect. While we cannot rule out the possibility that POm receives some efference copy from cortex, such input is not the cause of what at first appears to be whisking modulation. POm might instead be under equal or greater control of subcortical regions such as trigeminal brainstem complex, zona incerta, the thalamic reticular nucleus, and neuromodulatory brainstem centers—all of which receive inputs from broad areas of the nervous system (Pinault and Deschênes, 1998; Trageser and Keller, 2004).

As POm continues to track whisking in absence of both ascending sensory input and descending cortical input, we propose that the activity we observe is not sensorimotor in nature, but rather representative of thalamic coding of internal state. POm axons project to the apical dendrites of pyramidal cells (Meyer et al., 2010; Wimmer et al., 2010), where they might drive state-dependent changes in activity and synchrony. Arousal has dramatic effects on cortical dynamics (Constantinople and Bruno, 2011; Reimer et al., 2014; Vinck et al., 2015). We observed that pupil diameter, which closely tracks arousal, is highly correlated with whisking amplitude. Due to the coupling between pupil and whisking dynamics, they both correlate with POm firing rates (Figure 6). To dissociate the contributions of arousal and whisker movement, we took the novel approach of comparing POm dynamics with those of LP, the rodent homolog of the primate lateral pulvinar. We found a near-identical relationship between LP activity and whisking as we observed in POm (Figure 7), even though there is no known connectivity between LP and the whisker system. As for POm, these shifts in LP activity do not appear to be sensory dependent, as they persist even in low-light conditions where the pupil is maximally dilated and can no longer contribute to changes in retinal activity.

If state-dependent modulation of secondary thalamic nuclei is not derived from sensory reafference or motor efference copy from cortex or SC, the likely remaining candidates would include a large number of neuromodulators. While the neuromodulator acetylcholine is known to mediate the effects of whisking on activity in somatosensory cortex (Eggermann et al., 2014), our silencing experiments rule out any neuromodulation of primary somatosensory cortex as a mechanism. However, acetylcholine and norepinephrine also act directly on thalamic nuclei, such as VPM (Aguilar and Castro-Alamancos, 2005; Hirata and Castro-Alamancos, 2010). Zona incerta terminals within POm are regulated by acetylcholine (Masri et al., 2006) and are likely modulated in the same way within LP. Acetylcholine and norepinephrine both track pupil dynamics (Reimer et al., 2016) and are plausible mechanisms of the POm-whisking correlation. In addition to these two well-studied modulators, many others are known to have direct effects on thalamus (Varela, 2014). Any of these could act directly on POm and LP or indirectly through ZI, TRN, and trigeminal brainstem nuclei, such as principalis and interpolaris. Future studies manipulating these many modulators and at different sites of action are needed.

The arousal effect we have described may be a more general version of modality-specific attentional effects that have been proposed for at least some secondary thalamic nuclei. In primates, pulvinar neurons respond strongest when stimuli are presented in attended regions of visual space (Petersen et al., 1985), and lesion of the pulvinar leads to deficits of selective attention during visual tasks (Ward et al., 2002; Wilke et al., 2010). Human patients with pulvinar damage exhibit spatial neglect, in which a stimulus can be perceived normally in isolation but is missed or distorted in the presence of neighboring stimuli (Karnath et al., 2002; Snow et al., 2009). By analogy, one might hypothesize that POm provides feedback that selects somatosensory stimuli for further cortical processing. Indeed, we and others have already demonstrated that activation of POm sensitizes cortical pyramidal neurons to the occurrence of subsequent tactile stimuli (Mease et al., 2016; Zhang and Bruno, 2019). Thus, POm affords control over the gain of the sensory responsiveness of somatosensory cortex circuitry. Selective enhancement of sensory responses by attention within a modality could be a general principle of all secondary thalamic functions.

Cortex-wide fluctuations in activity are known to correlate with various uninstructed movements (Musall et al., 2019). Cortical activity ceases in the absence of thalamic input (Guo et al., 2017; Reinhold et al., 2015), and secondary thalamic inputs to somatosensory cortex are stronger and longer lasting than corticocortical connections (Zhang and Bruno, 2019). Taking those studies and our study together suggests that secondary thalamus may be the underlying cause of the recently observed patterned fluctuations in activity across cortex. Our study directly tested the multiple known possible sources of afferent and efferent motor signals to secondary thalamus. None of these could explain apparent shifts in thalamic activity. Thus, behavioral state, rather than uninstructed movement, may be a primary driver of thalamic and cortical activity during movement.

Elevated firing rates in secondary thalamus due to arousal or attention could be useful for creating periods of heightened cortical plasticity. Recent studies have shown that repetitive sensory stimuli in anesthetized animals drive POm input to pyramidal neurons, which leads to enhancement of future sensory responses in cortex (Gambino et al., 2014). A potential mechanism of this is that disinhibition of apical dendritic spikes leads to long-term potentiation of local recurrent synapses among cortical pyramidal neurons (Williams and Holtmaat, 2019). Furthermore, an in vivo study found that associative learning can also potentiate long-range POm connections onto pyramidal neurons when subsequently measured in vitro (Audette et al., 2019).

It is conceivable that the arousal modulation of secondary thalamus that we have described is utilized by such processes. Our work opens avenues to examining potential links between arousal, attention, and plasticity.

All experiments complied with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of Columbia University. Thirty C57BL/6 mice were used in these experiments.

Data have been deposited in Dryad.

1. 1. Petty G
   2. Kinnischtzke A
   3. Hong Y
   4. Bruno RM

   (2021) Dryad Digital Repository

   Data from: Effects of arousal and movement on secondary somatosensory and visual thalamus.

   https://doi.org/10.5061/dryad.280gb5mr4

1. 1. Aguilar JR
   2. Castro-Alamancos MA

   (2005) Spatiotemporal gating of sensory inputs in thalamus during quiescent and activated states

   The Journal of Neuroscience Vol. 25:10990–11002.

   https://doi.org/10.1523/JNEUROSCI.3229-05.2005

   • PubMed
   • Google Scholar
2. 1. Ahrens KF
   2. Kleinfeld D

   (2004) Current flow in vibrissa motor cortex can phase-lock with exploratory rhythmic whisking in rat

   Journal of Neurophysiology 92:1700–1707.

   https://doi.org/10.1152/jn.00020.2004

   • PubMed
   • Google Scholar
3. 1. Audette NJ
   2. Bernhard SM
   3. Ray A
   4. Stewart LT
   5. Barth AL

   (2019) Rapid Plasticity of Higher-Order Thalamocortical Inputs during Sensory Learning

   Neuron 103:277–291.

   https://doi.org/10.1016/j.neuron.2019.04.037

   • PubMed
   • Google Scholar
4. 1. Berman RA
   2. Wurtz RH

   (2010) Functional identification of a pulvinar path from superior colliculus to cortical area MT

   The Journal of Neuroscience 30:6342–6354.

   https://doi.org/10.1523/JNEUROSCI.6176-09.2010

   • PubMed
   • Google Scholar
5. 1. Bourassa J
   2. Descheˆnes M

   (1995) Corticothalamic projections from the primary visual cortex in rats: a single fiber study using biocytin as an anterograde tracer

   Neuroscience 66:253–263.

   https://doi.org/10.1016/0306-4522(95)00009-8

   • Google Scholar
6. 1. Castro-Alamancos MA
   2. Bezdudnaya T

   (2015) Modulation of artificial whisking related signals in barrel cortex

   Journal of Neurophysiology 113:1287–1301.

   https://doi.org/10.1152/jn.00809.2014

   • PubMed
   • Google Scholar
7. 1. Castro-Alamancos MA
   2. Favero M

   (2016) Whisker-related afferents in superior colliculus

   Journal of Neurophysiology 115:2265–2279.

   https://doi.org/10.1152/jn.00028.2016

   • PubMed
   • Google Scholar
8. 1. Chiaia NL
   2. Rhoades RW
   3. Bennett-Clarke CA
   4. Fish SE
   5. Killackey HP

   (1991) Thalamic processing of vibrissal information in the rat. I. Afferent input to the medial ventral posterior and posterior nuclei

   The Journal of Comparative Neurology 314:201–216.

   https://doi.org/10.1002/cne.903140202

   • PubMed
   • Google Scholar
9. 1. Clack NG
   2. O’Connor DH
   3. Huber D
   4. Petreanu L
   5. Hires A
   6. Peron S
   7. Svoboda K
   8. Myers EW

   (2012) Automated tracking of whiskers in videos of head fixed rodents

   PLOS Computational Biology Vol. 8:e1002591.

   https://doi.org/10.1371/journal.pcbi.1002591

   • PubMed
   • Google Scholar
10. 1. Constantinople CM
    2. Bruno RM

    (2011) Effects and mechanisms of wakefulness on local cortical networks

    Neuron 69:1061–1068.

    https://doi.org/10.1016/j.neuron.2011.02.040

    • PubMed
    • Google Scholar
11. 1. Constantinople CM
    2. Bruno RM

    (2013) Deep cortical layers are activated directly by thalamus

    Science 340:1591–1594.

    https://doi.org/10.1126/science.1236425

    • PubMed
    • Google Scholar
12. 1. Deschênes M
    2. Takatoh J
    3. Kurnikova A
    4. Moore JD
    5. Demers M
    6. Elbaz M
    7. Furuta T
    8. Wang F
    9. Kleinfeld D

    (2016) Inhibition, Not Excitation, Drives Rhythmic Whisking

    Neuron 90:374–387.

    https://doi.org/10.1016/j.neuron.2016.03.007

    • PubMed
    • Google Scholar
13. 1. Diamond ME
    2. Armstrong-James M
    3. Ebner FF
    4. Armstrong‐James M
    5. Ebner FF

    (1992) Somatic sensory responses in the rostral sector of the posterior group (POm) and in the ventral posterior medial nucleus (VPM) of the rat thalamus

    The Journal of Comparative Neurology 318:462–476.

    https://doi.org/10.1002/cne.903180410

    • PubMed
    • Google Scholar
14. 1. Diamond ME
    2. von Heimendahl M
    3. Knutsen PM
    4. Kleinfeld D
    5. Ahissar E

    (2008) “Where” and “what” in the whisker sensorimotor system

    Nature Reviews. Neuroscience 9:601–612.

    https://doi.org/10.1038/nrn2411

    • PubMed
    • Google Scholar
15. 1. Eggermann E
    2. Kremer Y
    3. Crochet S
    4. Petersen CCH

    (2014) Cholinergic signals in mouse barrel cortex during active whisker sensing

    Cell Reports 9:1654–1660.

    https://doi.org/10.1016/j.celrep.2014.11.005

    • PubMed
    • Google Scholar
16. 1. Fee MS
    2. Mitra PP
    3. Kleinfeld D

    (1997) Central versus peripheral determinants of patterned spike activity in rat vibrissa cortex during whisking

    Journal of Neurophysiology 78:1144–1149.

    https://doi.org/10.1152/jn.1997.78.2.1144

    • PubMed
    • Google Scholar
17. 1. Frangeul L
    2. Porrero C
    3. Garcia-Amado M
    4. Maimone B
    5. Maniglier M
    6. Clascá F
    7. Jabaudon D

    (2014) Specific activation of the paralemniscal pathway during nociception

    The European Journal of Neuroscience 39:1455–1464.

    https://doi.org/10.1111/ejn.12524

    • PubMed
    • Google Scholar
18. 1. Gambino F
    2. Pagès S
    3. Kehayas V
    4. Baptista D
    5. Tatti R
    6. Carleton A
    7. Holtmaat A

    (2014) Sensory-evoked LTP driven by dendritic plateau potentials in vivo

    Nature 515:116–119.

    https://doi.org/10.1038/nature13664

    • PubMed
    • Google Scholar
19. Preprint

    1. Gharaei S
    2. Honnuraiah S
    3. Arabzadeh E
    4. Stuart GJ

    (2019) Superior Colliculus Modulates Cortical Coding of Somatosensory Information

    bioRxiv.

    https://doi.org/10.1101/715847

    • Google Scholar
20. 1. Guillery RW
    2. Sherman SM

    (2002) Thalamic relay functions and their role in corticocortical communication: generalizations from the visual system

    Neuron 33:163–175.

    https://doi.org/10.1016/s0896-6273(01)00582-7

    • PubMed
    • Google Scholar
21. 1. Guo ZV
    2. Inagaki HK
    3. Daie K
    4. Druckmann S
    5. Gerfen CR
    6. Svoboda K

    (2017) Maintenance of persistent activity in a frontal thalamocortical loop

    Nature 545:181–186.

    https://doi.org/10.1038/nature22324

    • PubMed
    • Google Scholar
22. 1. Heaton JT
    2. Sheu SH
    3. Hohman MH
    4. Knox CJ
    5. Weinberg JS
    6. Kleiss IJ
    7. Hadlock TA

    (2014) Rat whisker movement after facial nerve lesion: evidence for autonomic contraction of skeletal muscle

    Neuroscience 265:9–20.

    https://doi.org/10.1016/j.neuroscience.2014.01.038

    • PubMed
    • Google Scholar
23. 1. Herkenham M

    (1980) Laminar organization of thalamic projections to the rat neocortex

    Science 207:532–535.

    https://doi.org/10.1126/science.7352263

    • PubMed
    • Google Scholar
24. 1. Hill DN
    2. Curtis JC
    3. Moore JD
    4. Kleinfeld D

    (2011) Primary motor cortex reports efferent control of vibrissa motion on multiple timescales

    Neuron 72:344–356.

    https://doi.org/10.1016/j.neuron.2011.09.020

    • PubMed
    • Google Scholar
25. 1. Hirata A
    2. Castro-Alamancos MA

    (2010) Neocortex network activation and deactivation states controlled by the thalamus

    Journal of Neurophysiology 103:1147–1157.

    https://doi.org/10.1152/jn.00955.2009

    • PubMed
    • Google Scholar
26. 1. Hong YK
    2. Lacefield CO
    3. Rodgers CC
    4. Bruno RM

    (2018) Sensation, movement and learning in the absence of barrel cortex

    Nature 561:542–546.

    https://doi.org/10.1038/s41586-018-0527-y

    • PubMed
    • Google Scholar
27. 1. Hoogland PV
    2. Welker E
    3. Van der Loos H

    (1987) Organization of the projections from barrel cortex to thalamus in mice studied with Phaseolus vulgaris-leucoagglutinin and HRP

    Experimental Brain Research 68:73–87.

    https://doi.org/10.1007/BF00255235

    • Google Scholar
28. 1. Karnath HO
    2. Himmelbach M
    3. Rorden C

    (2002) The subcortical anatomy of human spatial neglect: putamen, caudate nucleus and pulvinar

    Brain 125:350–360.

    https://doi.org/10.1093/brain/awf032

    • PubMed
    • Google Scholar
29. 1. Killackey HP
    2. Sherman SM

    (2003) Corticothalamic projections from the rat primary somatosensory cortex

    The Journal of Neuroscience 23:7381–7384.

    https://doi.org/10.1523/jneurosci.23-19-07381.2003

    • PubMed
    • Google Scholar
30. 1. Masri R
    2. Trageser JC
    3. Bezdudnaya T
    4. Li Y
    5. Keller A

    (2006) Cholinergic regulation of the posterior medial thalamic nucleus

    Journal of Neurophysiology 96:2265–2273.

    https://doi.org/10.1152/jn.00476.2006

    • PubMed
    • Google Scholar
31. 1. Masri R
    2. Bezdudnaya T
    3. Trageser JC
    4. Keller A

    (2008) Encoding of stimulus frequency and sensor motion in the posterior medial thalamic nucleus

    Journal of Neurophysiology 100:681–689.

    https://doi.org/10.1152/jn.01322.2007

    • PubMed
    • Google Scholar
32. 1. Masri R
    2. Quiton RL
    3. Lucas JM
    4. Murray PD
    5. Thompson SM
    6. Keller A

    (2009) Zona incerta: a role in central pain

    Journal of Neurophysiology 102:181–191.

    https://doi.org/10.1152/jn.00152.2009

    • PubMed
    • Google Scholar
33. 1. Mease RA
    2. Metz M
    3. Groh A

    (2016) Cortical Sensory Responses Are Enhanced by the Higher-Order Thalamus

    Cell Reports 14:208–215.

    https://doi.org/10.1016/j.celrep.2015.12.026

    • PubMed
    • Google Scholar
34. 1. Meyer HS
    2. Wimmer VC
    3. Oberlaender M
    4. de Kock CPJ
    5. Sakmann B
    6. Helmstaedter M

    (2010) Number and laminar distribution of neurons in a thalamocortical projection column of rat vibrissal cortex

    Cerebral Cortex 20:2277–2286.

    https://doi.org/10.1093/cercor/bhq067

    • PubMed
    • Google Scholar
35. Book

    1. Mitra P
    2. Bokil H

    (2008) Observed Brain Dynamics

    New York: Oxford University Press.

    https://doi.org/10.1093/acprof:oso/9780195178081.001.0001

    • Google Scholar
36. 1. Moore CI

    (2004) Frequency-dependent processing in the vibrissa sensory system

    Journal of Neurophysiology 91:2390–2399.

    https://doi.org/10.1152/jn.00925.2003

    • PubMed
    • Google Scholar
37. 1. Moore JD
    2. Mercer Lindsay N
    3. Deschênes M
    4. Kleinfeld D

    (2015) Vibrissa Self-Motion and Touch Are Reliably Encoded along the Same Somatosensory Pathway from Brainstem through Thalamus

    PLOS Biology 13:e1002253.

    https://doi.org/10.1371/journal.pbio.1002253

    • PubMed
    • Google Scholar
38. 1. Musall S
    2. Kaufman MT
    3. Juavinett AL
    4. Gluf S
    5. Churchland AK

    (2019) Single-trial neural dynamics are dominated by richly varied movements

    Nature Neuroscience 22:1677–1686.

    https://doi.org/10.1038/s41593-019-0502-4

    • PubMed
    • Google Scholar
39. 1. Nakamura H
    2. Hioki H
    3. Furuta T
    4. Kaneko T

    (2015) Different cortical projections from three subdivisions of the rat lateral posterior thalamic nucleus: a single-neuron tracing study with viral vectors

    The European Journal of Neuroscience 41:1294–1310.

    https://doi.org/10.1111/ejn.12882

    • PubMed
    • Google Scholar
40. 1. Olsen GM
    2. Witter MP

    (2016) Posterior parietal cortex of the rat: Architectural delineation and thalamic differentiation

    The Journal of Comparative Neurology 524:3774–3809.

    https://doi.org/10.1002/cne.24032

    • PubMed
    • Google Scholar
41. Preprint

    1. Pachitariu M
    2. Steinmetz N
    3. Kadir S
    4. Carandini M

    (2016) Kilosort: Realtime Spike-Sorting for Extracellular Electrophysiology with Hundreds of Channels

    bioRxiv.

    https://doi.org/10.1101/061481

    • Google Scholar
42. Book

    1. Paxinos G
    2. Franklin KBJ

    (2019)

    Paxinos and Franklin’s The Mouse Brain in Stereotaxic Coordinates (5th ed)

    Elsevier Inc.

    • Google Scholar
43. 1. Petersen SE
    2. Robinson DL
    3. Keys W

    (1985) Pulvinar nuclei of the behaving rhesus monkey: visual responses and their modulation

    Journal of Neurophysiology 54:867–886.

    https://doi.org/10.1152/jn.1985.54.4.867

    • PubMed
    • Google Scholar
44. 1. Petersen SE
    2. Robinson DL
    3. Morris JD

    (1987) Contributions of the pulvinar to visual spatial attention

    Neuropsychologia 25:97–105.

    https://doi.org/10.1016/0028-3932(87)90046-7

    • PubMed
    • Google Scholar
45. 1. Petersen CCH

    (2007) The Functional Organization of the Barrel Cortex

    Neuron 56:339–355.

    https://doi.org/10.1016/j.neuron.2007.09.017

    • Google Scholar
46. 1. Phillips JW
    2. Schulmann A
    3. Hara E
    4. Winnubst J
    5. Liu C
    6. Valakh V
    7. Wang L
    8. Shields BC
    9. Korff W
    10. Chandrashekar J
    11. Lemire AL
    12. Mensh B
    13. Dudman JT
    14. Nelson SB
    15. Hantman AW

    (2019) A repeated molecular architecture across thalamic pathways

    Nature Neuroscience 22:1925–1935.

    https://doi.org/10.1038/s41593-019-0483-3

    • PubMed
    • Google Scholar
47. 1. Pinault D
    2. Deschênes M

    (1998) Projection and innervation patterns of individual thalamic reticular axons in the thalamus of the adult rat: a three-dimensional, graphic, and morphometric analysis

    The Journal of Comparative Neurology 391:180–203.

    https://doi.org/10.1002/(sici)1096-9861(19980209)391:2<180::aid-cne3>3.0.co;2-z

    • PubMed
    • Google Scholar
48. 1. Poulet JFA
    2. Petersen CCH

    (2008) Internal brain state regulates membrane potential synchrony in barrel cortex of behaving mice

    Nature 454:881–885.

    https://doi.org/10.1038/nature07150

    • PubMed
    • Google Scholar
49. 1. Poulet JFA
    2. Fernandez LMJ
    3. Crochet S
    4. Petersen CCH

    (2012) Thalamic control of cortical states

    Nature Neuroscience 15:370–372.

    https://doi.org/10.1038/nn.3035

    • PubMed
    • Google Scholar
50. 1. Reimer J
    2. Froudarakis E
    3. Cadwell CRR
    4. Yatsenko D
    5. Denfield GHH
    6. Tolias ASS

    (2014) Pupil fluctuations track fast switching of cortical states during quiet wakefulness

    Neuron 84:355–362.

    https://doi.org/10.1016/j.neuron.2014.09.033

    • PubMed
    • Google Scholar
51. 1. Reimer J
    2. McGinley MJ
    3. Liu Y
    4. Rodenkirch C
    5. Wang Q
    6. McCormick DA
    7. Tolias AS

    (2016) Pupil fluctuations track rapid changes in adrenergic and cholinergic activity in cortex

    Nature Communications 7:13289.

    https://doi.org/10.1038/ncomms13289

    • PubMed
    • Google Scholar
52. 1. Reinhold K
    2. Lien AD
    3. Scanziani M

    (2015) Distinct recurrent versus afferent dynamics in cortical visual processing

    Nature Neuroscience 18:1789–1797.

    https://doi.org/10.1038/nn.4153

    • PubMed
    • Google Scholar
53. Preprint

    1. Rodgers CC
    2. Nogueira R
    3. Pil BC
    4. Greeman EA
    5. Fusi S
    6. Bruno RM

    (2020) The Sensorimotor Strategies and Neuronal Representations of Tactile Shape Discrimination in Mice

    bioRxiv.

    https://doi.org/10.1101/2020.06.16.126631

    • Google Scholar
54. Software

    1. Rossant C

    (2021) phy: interactive visualization and manual spike sorting of large-scale ephys data, version 9a330b9

    GitHub.

    https://github.com/cortex-lab/phy
55. 1. Roth MM
    2. Dahmen JC
    3. Muir DR
    4. Imhof F
    5. Martini FJ
    6. Hofer SB

    (2016) Thalamic nuclei convey diverse contextual information to layer 1 of visual cortex

    Nature Neuroscience 19:299–307.

    https://doi.org/10.1038/nn.4197

    • PubMed
    • Google Scholar
56. 1. Saalmann YB
    2. Pinsk MA
    3. Wang L
    4. Li X
    5. Kastner S

    (2012) The pulvinar regulates information transmission between cortical areas based on attention demands

    Science 337:753–756.

    https://doi.org/10.1126/science.1223082

    • PubMed
    • Google Scholar
57. 1. Sherman SM
    2. Guillery RW

    (2002) The role of the thalamus in the flow of information to the cortex

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 357:1695–1708.

    https://doi.org/10.1098/rstb.2002.1161

    • PubMed
    • Google Scholar
58. 1. Sherman SM

    (2016) Thalamus plays a central role in ongoing cortical functioning

    Nature Neuroscience 19:533–541.

    https://doi.org/10.1038/nn.4269

    • PubMed
    • Google Scholar
59. 1. Shetty P
    2. Shoykhet M
    3. Simons DJ

    (2003) Whisker plucking alters responses of rat trigeminal ganglion neurons

    Somatosensory & Motor Research 20:233–238.

    https://doi.org/10.1080/08990220310001622951

    • PubMed
    • Google Scholar
60. 1. Siegle JH
    2. López AC
    3. Patel YA
    4. Abramov K
    5. Ohayon S
    6. Voigts J

    (2017) Open Ephys: an open-source, plugin-based platform for multichannel electrophysiology

    Journal of Neural Engineering 14:045003.

    https://doi.org/10.1088/1741-2552/aa5eea

    • PubMed
    • Google Scholar
61. 1. Snow JC
    2. Allen HA
    3. Rafal RD
    4. Humphreys GW

    (2009) Impaired attentional selection following lesions to human pulvinar: evidence for homology between human and monkey

    PNAS 106:4054–4059.

    https://doi.org/10.1073/pnas.0810086106

    • PubMed
    • Google Scholar
62. 1. Trageser JC
    2. Keller A

    (2004) Reducing the uncertainty: gating of peripheral inputs by zona incerta

    The Journal of Neuroscience 24:8911–8915.

    https://doi.org/10.1523/JNEUROSCI.3218-04.2004

    • PubMed
    • Google Scholar
63. 1. Urbain N
    2. Salin PA
    3. Libourel PA
    4. Comte JC
    5. Gentet LJ
    6. Petersen CCH

    (2015) Whisking-Related Changes in Neuronal Firing and Membrane Potential Dynamics in the Somatosensory Thalamus of Awake Mice

    Cell Reports 13:647–656.

    https://doi.org/10.1016/j.celrep.2015.09.029

    • PubMed
    • Google Scholar
64. 1. Varela C

    (2014) Thalamic neuromodulation and its implications for executive networks

    Frontiers in Neural Circuits 8:69.

    https://doi.org/10.3389/fncir.2014.00069

    • PubMed
    • Google Scholar
65. 1. Vinck M
    2. Batista-Brito R
    3. Knoblich U
    4. Cardin JA

    (2015) Arousal and locomotion make distinct contributions to cortical activity patterns and visual encoding

    Neuron 86:740–754.

    https://doi.org/10.1016/j.neuron.2015.03.028

    • PubMed
    • Google Scholar
66. 1. Ward R
    2. Danziger S
    3. Owen V
    4. Rafal R

    (2002) Deficits in spatial coding and feature binding following damage to spatiotopic maps in the human pulvinar

    Nature Neuroscience 5:99–100.

    https://doi.org/10.1038/nn794

    • PubMed
    • Google Scholar
67. 1. Wilke M
    2. Turchi J
    3. Smith K
    4. Mishkin M
    5. Leopold DA

    (2010) Pulvinar inactivation disrupts selection of movement plans

    The Journal of Neuroscience 30:8650–8659.

    https://doi.org/10.1523/JNEUROSCI.0953-10.2010

    • PubMed
    • Google Scholar
68. 1. Williams LE
    2. Holtmaat A

    (2019) Higher-Order Thalamocortical Inputs Gate Synaptic Long-Term Potentiation via Disinhibition

    Neuron 101:91–102.

    https://doi.org/10.1016/j.neuron.2018.10.049

    • PubMed
    • Google Scholar
69. 1. Wimmer VC
    2. Bruno RM
    3. de Kock CPJ
    4. Kuner T
    5. Sakmann B

    (2010) Dimensions of a projection column and architecture of VPM and POm axons in rat vibrissal cortex

    Cerebral Cortex 20:2265–2276.

    https://doi.org/10.1093/cercor/bhq068

    • PubMed
    • Google Scholar
70. 1. Yu C
    2. Derdikman D
    3. Haidarliu S
    4. Ahissar E

    (2006) Parallel thalamic pathways for whisking and touch signals in the rat

    PLOS Biology 4:e124.

    https://doi.org/10.1371/journal.pbio.0040124

    • PubMed
    • Google Scholar
71. 1. Zhang W
    2. Bruno RM

    (2019) High-order thalamic inputs to primary somatosensory cortex are stronger and longer lasting than cortical inputs

    eLife 8:e44158.

    https://doi.org/10.7554/eLife.44158

    • PubMed
    • Google Scholar

Author details

1. Gordon H Petty

   1. Department of Neuroscience, Columbia University, New York, United States
   2. Kavli Institute for Brain Science, New York, United States
   3. Zuckerman Mind Brain Behavior Institute, New York, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Writing - original draft, Writing - review and editing

   Contributed equally with

   Amanda K Kinnischtzke

   Competing interests

   No competing interests declared

2. Amanda K Kinnischtzke

   1. Department of Neuroscience, Columbia University, New York, United States
   2. Kavli Institute for Brain Science, New York, United States
   3. Zuckerman Mind Brain Behavior Institute, New York, United States

   Contribution

   Conceptualization, Formal analysis, Investigation, Methodology

   Contributed equally with

   Gordon H Petty

   Competing interests

   No competing interests declared

3. Y Kate Hong

   1. Department of Neuroscience, Columbia University, New York, United States
   2. Kavli Institute for Brain Science, New York, United States
   3. Zuckerman Mind Brain Behavior Institute, New York, United States

   Present address

   Department of Biological Sciences and Carnegie Mellon Neuroscience Institute, Carnegie Mellon University, Pittsburgh, United States

   Contribution

   Methodology, Resources

   Competing interests

   No competing interests declared

4. Randy M Bruno

   1. Department of Neuroscience, Columbia University, New York, United States
   2. Kavli Institute for Brain Science, New York, United States
   3. Zuckerman Mind Brain Behavior Institute, New York, United States

   Contribution

   Conceptualization, Formal analysis, Funding acquisition, Supervision, Writing - original draft, Writing - review and editing

   For correspondence

   randybruno@columbia.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5122-4632

• 2,249

  views

• 296

  downloads

• 25

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.