Even in the absence of external stimuli or overt behavior, the patterns of activity exhibited by the awake cerebral cortex, known as ongoing activity, can display a striking degree of bilateral symmetry. When one region of cortex becomes active, this activity is often accompanied by activity in the symmetrical region in the opposite hemisphere. This bilateral symmetry has been documented both in awake mice (Mohajerani et al., 2010) and in awake humans (Fox et al., 2006; Tyszka et al., 2011). The degree of symmetry varies across cortical areas (Mohajerani et al., 2010), but the rules behind the differences across areas remain unknown.

Bilateral ongoing activity coexists with event-related activity, which is not necessarily bilateral. For instance, when human subjects move their right hand, the movement-related BOLD (Blood-oxygen-level-dependent) response in left motor cortex coexists with bilateral ongoing fluctuations in both left and right motor cortices (Fox et al., 2006).

It is not clear, however, to what extent this bilateral ongoing activity can influence sensory-evoked activity. Similar to the activity associated with movements, the activity evoked by visual stimuli in cortex is not generally bilateral. For instance, a visual stimulus presented to the side of the animal would evoke unilateral responses in the contralateral visual cortex. It is not known whether the bilateral ongoing activity coexists with the visually evoked activity, or whether it is quenched by the onset of visual responses (Churchland et al., 2010; He, 2013).

If bilateral ongoing activity persists during sensory responses, it could contribute to the large trial-by-trial variability seen in these responses. Even in response to an identical visual stimulus, neural responses in primary visual cortex are well-known to vary strongly from trial to trial (Tolhurst et al., 1983). This variability has been observed not only in the spike trains but also in intracellular recordings from single neurons (Anderson et al., 2000; Carandini, 2004) and in voltage imaging of neural populations (Arieli et al., 1996; Carandini et al., 2015). How much of this trial-by-trial variability is explained by bilateral ongoing activity?

If much of trial-by-trial variability is explained by bilateral ongoing activity, then bilateral measurements may become necessary to estimate true sensory responses. How would such a procedure compare to the standard method (Arieli et al., 1996) of measuring activity before a stimulus and removing it from the subsequent activity? Moreover, it is not known how bilateral ongoing activity interacts with visual responses, for example whether the interaction can be described as simply additive or multiplicative (Arieli et al., 1996; Ecker et al., 2014; Goris et al., 2014; He, 2013; Lin et al., 2015; Schölvinck et al., 2015). Likewise, it is not known whether the bilateral fluctuations impair stimulus coding (Averbeck et al., 2006) and therefore impact perception.

To address these questions, we performed widefield optical imaging from both cortical hemispheres of mice performing a visual detection task. Widefield imaging allowed us to monitor bilateral ongoing activity and visually evoked response across hemispheres. To observe these forms of activity both in membrane potential and in spiking activity, we measured both from mice expressing a voltage sensor and from mice expressing a calcium sensor. These measurements revealed that bilateral ongoing activity has a prominent impact on visually evoked responses. An additive model for the interaction of ongoing and evoked activities provided excellent fits to the data, explaining up to 60% of trial-by-trial variance in the observed imaging signal. On the other hand, bilateral ongoing activity did not seem to affect perceptual decisions, suggesting that it does not impact stimulus coding.

We performed widefield imaging in the cerebral cortex of transgenic mice expressing voltage indicator VSFPB1.2 or calcium indicator GCaMP6 in excitatory neurons (Figure 1). The imaging window covered the dorsal part of both cortical hemispheres. We defined the location of sensory and motor areas by aligning the images to the Common Coordinate Framework of the Allen Brain Atlas (Figure 1A). We confirmed the outlines of the visual areas based on functional mapping (visual field sign, Figure 1B).

Figure 1 with 2 supplements see all

Download asset Open asset

We then measured activity while the mice performed a visual two-alternative task (Burgess et al., 2017; Zatka-Haas et al., 2019). In the task, mice reported the location of a grating stimulus presented on the left or right visual field. Before the stimulus application, the mice were required to hold the wheel still for at least 500 ms. We begin by examining the ongoing activity measured in this pre-stimulus interval, focusing on trials when mice held the wheel without making eye movements.

Bilateral ongoing fluctuations differ across areas

The bilateral symmetry of ongoing activity, however, was not equal for all cortical areas, and tended to decrease with distance from the midline (Figure 2). For each cortical area of interest, we computed the temporal correlation of that area (‘seed area’) against all the other pixels. Both in mice-expressing voltage indicators (Figure 2A) and in mice-expressing calcium indicators (Figure 2B), the resulting seed-area maps showed high correlation in the corresponding region in the opposite hemisphere. This bilateral correlation was particularly strong in primary somatosensory cortex (S1) and in medial visual areas such as PM (Figure 2C,D). Bilateral correlation tended to decrease with a region’s distance from the midline, both in mice-expressing voltage indicators (Figure 2C, pairwise correlation ρ = −0.44) and in mice-expressing calcium indicators (Figure 2D, pairwise correlation ρ = −0.42).

Figure 2 with 1 supplement see all

Download asset Open asset

However, correlation was lower than expected in the monocular region of primary visual cortex, V1m and, when the animal was engaged in the task, in secondary motor cortex (M2, Figure 2C,D). In area M2, this low bilateral correlation may be related to task engagement, because bilateral correlation seemed to be higher when the animal was in a passive condition (Figure 2C,D, white circle). Indeed, the reduction in bilateral fluctuations seen with task engagement was specific to area M2, where it was significant (Figure 2—figure supplement 1). For each cortical area, we compared bilateral correlation in the active task vs. the passive condition using VSFP and GCaMP data. For all areas except M2, there was no significant change in correlations across the two conditions. This included V1m, where correlations were lower than in other areas but not significantly different across the two conditions (p>0.05 after Bonferroni correction, paired t test, n = 6 mice, Figure 2—figure supplement 1D). In M2, instead, there was a significant reduction during the active task (p = 0.0013/0.013 before/after Bonferroni correction, paired t test, n = 6 mice, Figure 2—figure supplement 1H).

Bilateral ongoing fluctuations continue during unilateral visual responses

Bilateral ongoing fluctuations continued following the onset of the visual stimulus and were typically larger than the unilateral visual responses (Figure 3). For example, consider the voltage responses to three presentations of a 50% contrast stimulus in the right monocular visual field (Figure 3A). In all these trials, activity before visual stimulation was largely bilateral and continued to fluctuate after stimulus onset. Only by acquiring multiple trials could these ongoing fluctuations be averaged out to reveal the underlying unilateral response to the visual stimulus (258 trials, Figure 3B). Similar results were obtained with calcium imaging. The bilateral activity seemed to continue unperturbed following presentation of a stimulus in the left monocular visual field (Figure 3C, Figure 3—video 1). Averaging the calcium activity over 104 trials revealed localized visual responses in area V1 and higher visual areas, which grew between 50 and 150 ms after stimulus onset (Figure 3D).

Figure 3 with 1 supplement see all

Download asset Open asset

Confirming these observations, we measured the activity evoked by our stimuli in areas V1 and PM, and observed that in the first 150 ms following stimulus presentation this activity was exclusively unilateral (Figure 4). Task stimuli elicited visual activity solely in contralateral cortex, both in area V1 (Figure 4A, red traces) and in area PM (Figure 4B, red traces). The mean activity rose in the period 50–150 ms after stimulus onset, and increased with contrast as expected (Figure 4E,F). Consistent with known variability of neural activity, this increase in mean was accompanied by an increase in standard deviation across trials (Figure 4C,D). By contrast, in the unstimulated hemisphere, the stimuli affected neither the mean activity (Figure 4A,B, black traces) nor the standard deviation of activity (Figure 4C,D, black traces). These results suggest that bilateral fluctuations are not reduced by the onset of visual stimuli.

Figure 4

Download asset Open asset

Bilateral fluctuations and visually evoked responses add linearly

We exploited the unilateral nature of visual responses to study their interaction with bilateral fluctuations. We illustrate these analyses with calcium imaging, where we have larger numbers of animals.

Inspection of trial-by-trial calcium signals suggested an additive interaction of stimulus-evoked responses and bilateral ongoing activity (Figure 5A,B). Consider, for instance, the trial-by-trial fluorescence measured in left and right area PM following the onset of a 100% contrast stimulus on the left (Figure 5A,B). Activities were highly variable across trials both in the left, unstimulated hemisphere and in the right, stimulated hemisphere (Figure 5A). This variability persisted at the time of the visual response, 50–150 ms after stimulus onset (Figure 5A), and showed strong bilateral correlation (Figure 5B, dots).

Figure 5 with 4 supplements see all

Download asset Open asset

To quantify the relationship between left and right PM in the response period we computed a sliding average of right PM activity as a function of left PM activity, separately for 100% contrast visual stimulus (Figure 5B black curve), and 0% contrast stimulus (Figure 5B gray curve). Consistent with an additive interaction, the effect of the stimulus was to elevate this right-hemisphere activity by a constant amount, irrespective of left-hemisphere activity (Figure 5B, black vs. gray curves).

To test whether the interaction between ongoing activity and visually evoked responses is additive, we examined the effects of variations in each factor’s amplitude: the contrast of the stimulus, and the simultaneous amplitude of the bilateral fluctuation (Figure 5C–H). We divided trials into two groups according to fluorescence at the time of the responses (50–150 ms) in the unstimulated hemisphere (Figure 5B; blue vs. cyan). For all stimulus contrasts, the visual responses obtained when ongoing activity was high (Figure 5C,D, cyan) appeared additively shifted compared to those obtained when ongoing activity was low (Figure 5C,D, blue). To quantify this effect, we fitted the contrast responses with a curve with two free parameters: offset and scale. The offset parameter was significantly higher when ongoing activity was high (p = 0.0009, paired t test, n = 7 mice Figure 5E). The scale parameter was much less affected: it appeared to show a mild increase when bilateral activity was high, but this increase was not significant (p>0.05, paired t test, n = 7 mice Figure 5F). This result held not only for PM but also for all other visual areas: bilateral ongoing activity changed offset but not scale (n = 7 in V1, n = 5 mice in AL, LM and RL, Figure 5G,H). Similar results were seen when we chose different criteria for distinguishing high vs. low ongoing activity.

The additive interaction between bilateral ongoing activity and visual responses held regardless of the degree to which ongoing activity resembled visual responses (Figure 5—figure supplement 1). Ongoing activity sometimes resembles the responses evoked by visual stimulation (Kenet et al., 2003; Mohajerani et al., 2013). To assess whether this similarity affects the interaction between ongoing activity and visually evoked responses, we created a template based on the average response to a visual stimulus, made it bilateral, and divided trials into two groups according to correlation with this template at the time of the stimulus onset. For all stimulus contrasts, the visual responses obtained when ongoing activity was positively or negatively correlated with the template differed by a single additive factor (Offset: p>0.05 in V1 n = 7 mice, p<0.05 in other visual areas n = 5 mice, paired t test; Scale: p>0.05 in all areas, paired t test, n = 7 mice in V1, n = 5 mice in AL, LM, and RL, Figure 5—figure supplement 1F,G).

Bilateral fluctuations are rapid and their effect is independent of arousal. These effects of bilateral ongoing activity were due to fast changes in cortical activation, not to slow changes in cortical state (Figure 5—figure supplement 2). Slow state changes might in principle have been able to explain these results; for instance, there may be more opportunities for bilateral ongoing activity to be ‘high’ when the cortex is in a more synchronized state, and this state might be the fundamental factor that leads to elevated visual responses. To assess slow changes in cortical state, we computed the amplitude of 2–7 Hz fluctuations in bilateral activity before the stimulus, and thus divided trials in four groups, for all combinations of state (more desynchronized vs. more synchronized) and instantaneous activity (low vs. high, Figure 5—figure supplement 2A). The interaction between instantaneous ongoing activity and visually evoked responses was seen in synchronized and desynchronized states (Figure 5—figure supplement 2B,C). The offset associated with high bilateral activity seemed to be larger in the synchronized state (Figure 5—figure supplement 2C). However, this larger offset was explained by the fact that in the synchronized state, the bilateral ongoing activity tended to reach higher levels (Figure 5—figure supplement 2A,D). Similar results were seen across the other mice (Figure 5—figure supplement 2E–G): fast fluctuations, rather than cortical state, determine the size of the additive interaction on visually evoked responses.

Similarly, the effects of bilateral ongoing activity were not related to slow changes in level of arousal (Figure 5—figure supplement 3). We gauged the level of arousal by measuring pupil diameter, and we divided the trials according to pupil diameter (small vs. large, Figure 5—figure supplement 3A), and found that the effect of bilateral activity was present in both cases (Figure 5—figure supplement 3B,C,E,F). Once again, responses could be explained based on fast ongoing fluctuations, with no need to include pupil dilation as a predictor (Figure 5—figure supplement 3D,G).

Because ongoing activity changed rapidly and was largely bilateral, its impact could not be estimated through the typical procedure of measuring unilateral activity before stimulus onset (Figure 5—figure supplement 4). Ongoing activity is often estimated from prestimulus measurements (Arieli et al., 1996), and these estimates are often subtracted from the subsequent measurements. This procedure is likely to be less effective in the face of strong, fast bilateral ongoing activity. Indeed, the cross-correlation between corresponding regions on the two hemispheres typically decreased to 0 in ~100 ms, and in some mice became negative afterwards (Figure 5—figure supplement 4A). As a result, estimates of bilateral ongoing activity obtained from the unstimulated hemisphere were superior to estimates obtained in the stimulated hemisphere before stimulus onset. This can be seen by dividing all trials into four groups, for all combinations of prestimulus activity in the stimulated hemisphere (down vs. up) and poststimulus activity in the unstimulated hemisphere (low vs. high Figure 5—figure supplement 4B). The strong effect of bilateral activity was present in both up and down groups (Figure 5—figure supplement 4C,D,F,G), and responses could be explained based on fast bilateral fluctuations (Figure 5—figure supplement 4E,H).

Bilateral fluctuations do not affect perceptual reports

Do these fast fluctuations in bilateral ongoing activity result in percepts? On the one hand, one might imagine that they do, because they impact cortical areas that are necessary for the performance of the task (Burgess et al., 2017; Zatka-Haas et al., 2019), so one might expect that whatever affects their activity will influence the percept of the animal. On the other hand, as the correlation structure of bilateral spontaneous activity differs from that of unilateral sensory responses, it may have little impact on information coding (Averbeck et al., 2006): interpreted naively, bilateral activations would constitute stimuli with a peculiar mirror symmetry across the vertical midline, which would be unlikely to be confused with actual images.

Bilateral ongoing fluctuations did not affect the behavior of the mouse in the task (Figure 6). We divided trials according to whether bilateral ongoing activity in PM was low or high (as in Figure 5C,D), and computed psychometric curves in each group. The two psychometric curves were essentially identical: there was no effect of bilateral activity on the probability of choosing the correct side (Figure 6A,C), or to choose a no-go response, which would be appropriate when stimulus is zero (Figure 6B,D). We fitted the behavioral responses with a simple probabilistic observer model (Burgess et al., 2017; Zatka-Haas et al., 2019), and measured offset and scale parameters of the perceptual contrast sensitivity function inferred by the model. Neither offset nor scale were significantly affected by bilateral ongoing activity (p>0.05, paired t test, n = 7 mice, Figure 6E,G). Similar results were seen when we used other visual areas to measure bilateral activity (p>0.05, n = 7 mice in V1, n = 5 mice in AL, LM, and RL, Figure 6F,H).

Figure 6 with 1 supplement see all

Download asset Open asset

These results indicate that the bilateral ongoing activity did not affect perceptual reports. We next asked whether this result could be explained by the bilateral activity having very different features from the visually evoked activity. To test this hypothesis, we restricted the analysis to cases when the ongoing activity resembled the visual responses (Mohajerani et al., 2013). Even in these cases, we saw no significant effects on the performance of the animal (p>0.05, n = 7 mice, Figure 6—figure supplement 1).

Bilateral fluctuations dominate trial-by-trial variations in evoked visual responses

To compare the effects of bilateral ongoing fluctuations and evoked visual responses we fit a simple additive model (Figure 7A). In the model, the activity in a predicted visual area in trial i at time t from stimulus onset is the sum of a visual component $V\left(c_i,t\right)$, which depends only on contrast$c_i$, and a contribution from bilateral fluctuation $F_i\left(t\right)$, that varies across trials independently of the stimulus

(1) $f_i\left(t\right)=V\left(c_i,t\right)+F_i\left(t\right).$

To estimate the visual component $V\left(c_i,t\right)$, we used the average across trials of the activity measured at each stimulus contrast. To estimate the amount of bilateral fluctuation $F_i\left(t\right)$ on trial i, we measured the activity $f_i^{\text{'}}\left(t\right)$ contralateral to the predicted area, subtracting any trial-triggered activity $V^{\text{'}}(c_i,t)$ that may be seen there:

(2) $B_i\left(t\right)=h[f_i^{\text{'}}\left(t\right)-V^{\text{'}}(c_i,t\left)\right]$

The weight ${\displaystyle h}$≤1 is a fixed coefficient, and was obtained from trials with no stimulus (0% contrast) by linear regression against the contralateral fluorescence.

The additive model provided good estimates of trial-by-trial activity (Figure 7B). Consider again the activity measured in right area PM in response to 100% contrast stimuli presented in the left visual field (as in Figure 5A). The bilateral ongoing activity $F_i\left(t\right)$ estimated from left area PM showed strong fluctuations across time and trials (Figure 7B₁). These fluctuations were often as large as the visually evoked response $V\left(100,t\right)$, obtained by averaging activity across trials in right area PM (Figure 7B₂). The sum of these two components yields predicted single-trial activities in right area PM (Figure 7B₃). This predicted activity closely resembles the observed data (Figure 7B₄).

The additive model was most successful when bilateral ongoing fluctuations were measured across symmetrical cortical locations (Figure 7C). We asked whether the model would degrade if bilateral activity $F_i\left(t\right)$ was measured not from the contralateral PM but rather from other contralateral areas. Doing so strongly decreased the explained variance: explained variance was highest when the region used to estimate the bilateral activity was contralateral PM or regions near it, and decreased with distance from contralateral PM (Figure 7D). Similar results were obtained in the other visual areas (Figure 7—figure supplement 1), confirming that the bilateral ongoing activity contributing to the observed variability is not global, but rather local to individual areas.

This simple additive model worked well in all visual areas, and revealed that bilateral ongoing fluctuations had a much larger effect on cortical activity than evoked visual responses (Figure 7D). In area PM, the additive model explained 58 ± 3% of the variance (S.E. n = 7 mice). This is a very high performance, and the bilateral component is crucial to obtain it. Indeed, the visually evoked responses $V\left(c_i,t\right)$ were only able to account for 12 ± 4% of the variance. Similar effects were seen in area AM, and to a large extent also in the remaining visual areas (Figure 7D).

Moreover, the model performed well also on the data obtained with voltage imaging (Figure 7E,F). For area PM, for instance, the additive model accounted for up to 34% of trial-by-trial variability in the voltage signals (Figure 7E). This value was significantly higher than the variance explained by the visually evoked responses, which is known to be particularly low for voltage signals (Figure 7F). Much of the large variability observed across trials when performing widefield imaging experiments (Arieli et al., 1996; Carandini et al., 2015), therefore, is not due to imaging noise, but rather to structured ongoing neural activity tightly coordinated across the hemispheres.

Figure 7 with 1 supplement see all

Download asset Open asset

By taking advantage of widefield imaging, we monitored both cortical hemispheres in mice performing a visual detection task, and found that bilateral ongoing activity has a powerful additive impact on sensory processing. As reported previously (Mohajerani et al., 2010; Shimaoka et al., 2017; Vanni et al., 2017; Vanni and Murphy, 2014), voltage and calcium signals exhibited strong bilateral fluctuations in ongoing activity. In both types of signal, the impact of these bilateral fluctuations varied across cortical areas, and was stronger in areas closer to the midline. Two exceptions were the monocular region of area V1, where activity was less bilateral in all conditions, and area M2, where activity was less bilateral during task engagement. Bilateral ongoing fluctuations continued following the onset of the visual stimulus, and typically swamped the resulting unilateral visual responses. Higher bilateral ongoing activity increased visual responses equally at all contrasts, regardless of response amplitude, indicating that the interaction between ongoing activity and visually evoked responses was additive. A simple additive model provided good estimates of trial-by-trial visual response in all visual areas, and revealed that bilateral ongoing fluctuations had a much larger effect on cortical activity than evoked visual responses.

Bilateral fluctuations have been observed in multiple circumstances, but in the absence of overt movements their functional significance is not clear. The cortex exhibits clear bilateral activity during motor behavior (Musall et al., 2018), which is understandable given the symmetry involved in many body movements. It is less clear, however, why there should be bilateral symmetry in sensory responses, and even less clear why there should be symmetry in ongoing activity, both in wakefulness and in anesthesia (Mohajerani et al., 2010; Shimaoka et al., 2017; Vanni and Murphy, 2014).

Regardless of their origin, it is likely that bilateral fluctuations are at least partially mediated by callosal connections. A number of brain structures connect the two cerebral hemispheres, including anterior commissure, cingulum, interhemispheric thalamic connections, and corpus callosum. Among these, the corpus callosum is the largest white matter tract connecting the cerebral hemispheres in placental mammalian brains. Ongoing activity is less correlated across hemispheres in mutant mice that lack the corpus callosum (Mohajerani et al., 2010) and in rats after surgical section of the corpus callosum (Magnuson et al., 2014). However, ongoing bilateral activity is also reported in humans born without a corpus callosum (Tyszka et al., 2011), suggesting that bilateral fluctuations can also be sustained by other brain structures.

Bilateral fluctuation represents shared activity across neurons, which impacts on response variability. Our widefield imaging result conforms to the previous electrophysiological report by Schölvinck et al. (2015), that ongoing activity is shared across large proportions of neurons and interacts additively with visually evoked responses, and that it accounts for a large fraction of trial-by-trial fluctuations in observed visually evoked activities. The present study extends this idea by demonstrating that such global fluctuation is actually shared across hemispheres, yet in an area-specific manner.

Our results indicate that a full understanding of activity in a population of neurons in one hemisphere may require recordings from the corresponding area of the other hemisphere. Because ongoing activity changed rapidly and was largely bilateral, its impact could not be estimated through the typical procedure of measuring unilateral activity before stimulus onset (Arieli et al., 1996). Indeed, compared to bilateral activity, activity in the prestimulus period had a smaller impact on evoked responses than simultaneously measured activity in the contralateral hemisphere. Given the specificity that we have observed in bilateral fluctuations, the ideal placement of a sensor in the contralateral hemisphere would be one that monitored the symmetrical populations to the ones of interest. Experiments with this design are emerging in electrophysiology (Cohen and Maunsell, 2009; Rabinowitz et al., 2015), and a promising strategy would be to pair them with simultaneous widefield imaging.

Even though bilateral fluctuations occurred in regions necessary for this visual task (Burgess et al., 2017), they did not impact choice behavior. Perhaps, a downstream readout mechanism removes the bilateral fluctuations by taking the difference of activity on left and right visual cortices (Zatka-Haas et al., 2019). It might thus be easy for the brain to ignore them during perceptual decisions based on vision, even though the fluctuations continue during visual stimulation and have a powerful additive impact on visual processing. One possible test of this hypothesis might be to perform optogenetic stimulation (e.g. O'Connor et al., 2013; Sreenivasan et al., 2016) both unilaterally and bilaterally, and ask whether the animal finds it easier to ignore bilateral activations during behavior.

Experimental procedures were conducted according to the UK Animals Scientific Procedures Act (1986), under personal and project licenses released by the Home Office following appropriate ethics review.

Data has been deposited to Dryad Digital Repository under the doi:10.5061/dryad.rd00gk7.

1. 1. Daisuke Shimaoka
   2. Nicholas A Steinmetz
   3. Kenneth D Harris
   4. Matteo Carandini

   (2019) Dryad Digital Repository

   Data from: The impact of bilateral ongoing activity on evoked responses in mouse cortex.

   https://doi.org/10.5061/dryad.rd00gk7

1. 1. Akemann W
   2. Mutoh H
   3. Perron A
   4. Park YK
   5. Iwamoto Y
   6. Knöpfel T

   (2012) Imaging neural circuit dynamics with a voltage-sensitive fluorescent protein

   Journal of Neurophysiology 108:2323–2337.

   https://doi.org/10.1152/jn.00452.2012

   • PubMed
   • Google Scholar
2. 1. Anderson JS
   2. Lampl I
   3. Gillespie DC
   4. Ferster D

   (2000) The contribution of noise to contrast invariance of orientation tuning in cat visual cortex

   Science 290:1968–1972.

   https://doi.org/10.1126/science.290.5498.1968

   • PubMed
   • Google Scholar
3. 1. Arieli A
   2. Sterkin A
   3. Grinvald A
   4. Aertsen A

   (1996) Dynamics of ongoing activity: explanation of the large variability in evoked cortical responses

   Science 273:1868–1871.

   https://doi.org/10.1126/science.273.5283.1868

   • PubMed
   • Google Scholar
4. 1. Averbeck BB
   2. Latham PE
   3. Pouget A

   (2006) Neural correlations, population coding and computation

   Nature Reviews Neuroscience 7:358–366.

   https://doi.org/10.1038/nrn1888

   • PubMed
   • Google Scholar
5. 1. Burgess CP
   2. Lak A
   3. Steinmetz NA
   4. Zatka-Haas P
   5. Bai Reddy C
   6. Jacobs EAK
   7. Linden JF
   8. Paton JJ
   9. Ranson A
   10. Schröder S
   11. Soares S
   12. Wells MJ
   13. Wool LE
   14. Harris KD
   15. Carandini M

   (2017) High-Yield methods for accurate Two-Alternative visual psychophysics in Head-Fixed mice

   Cell Reports 20:2513–2524.

   https://doi.org/10.1016/j.celrep.2017.08.047

   • PubMed
   • Google Scholar
6. 1. Carandini M

   (2004) Amplification of trial-to-trial response variability by neurons in visual cortex

   PLOS Biology 2:E264.

   https://doi.org/10.1371/journal.pbio.0020264

   • PubMed
   • Google Scholar
7. 1. Carandini M
   2. Shimaoka D
   3. Rossi LF
   4. Sato TK
   5. Benucci A
   6. Knöpfel T

   (2015) Imaging the awake visual cortex with a genetically encoded voltage indicator

   Journal of Neuroscience 35:53–63.

   https://doi.org/10.1523/JNEUROSCI.0594-14.2015

   • PubMed
   • Google Scholar
8. 1. Churchland MM
   2. Yu BM
   3. Cunningham JP
   4. Sugrue LP
   5. Cohen MR
   6. Corrado GS
   7. Newsome WT
   8. Clark AM
   9. Hosseini P
   10. Scott BB
   11. Bradley DC
   12. Smith MA
   13. Kohn A
   14. Movshon JA
   15. Armstrong KM
   16. Moore T
   17. Chang SW
   18. Snyder LH
   19. Lisberger SG
   20. Priebe NJ
   21. Finn IM
   22. Ferster D
   23. Ryu SI
   24. Santhanam G
   25. Sahani M
   26. Shenoy KV

   (2010) Stimulus onset quenches neural variability: a widespread cortical phenomenon

   Nature Neuroscience 13:369–378.

   https://doi.org/10.1038/nn.2501

   • PubMed
   • Google Scholar
9. 1. Cohen MR
   2. Maunsell JH

   (2009) Attention improves performance primarily by reducing interneuronal correlations

   Nature Neuroscience 12:1594–1600.

   https://doi.org/10.1038/nn.2439

   • PubMed
   • Google Scholar
10. 1. Drew PJ
    2. Shih AY
    3. Driscoll JD
    4. Knutsen PM
    5. Blinder P
    6. Davalos D
    7. Akassoglou K
    8. Tsai PS
    9. Kleinfeld D

    (2010) Chronic optical access through a polished and reinforced thinned skull

    Nature Methods 7:981–984.

    https://doi.org/10.1038/nmeth.1530

    • PubMed
    • Google Scholar
11. 1. Ecker AS
    2. Berens P
    3. Cotton RJ
    4. Subramaniyan M
    5. Denfield GH
    6. Cadwell CR
    7. Smirnakis SM
    8. Bethge M
    9. Tolias AS

    (2014) State dependence of noise correlations in macaque primary visual cortex

    Neuron 82:235–248.

    https://doi.org/10.1016/j.neuron.2014.02.006

    • PubMed
    • Google Scholar
12. 1. Engbert R
    2. Kliegl R

    (2003) Microsaccades uncover the orientation of covert attention

    Vision Research 43:1035–1045.

    https://doi.org/10.1016/S0042-6989(03)00084-1

    • PubMed
    • Google Scholar
13. 1. Fox MD
    2. Snyder AZ
    3. Zacks JM
    4. Raichle ME

    (2006) Coherent spontaneous activity accounts for trial-to-trial variability in human evoked brain responses

    Nature Neuroscience 9:23–25.

    https://doi.org/10.1038/nn1616

    • PubMed
    • Google Scholar
14. 1. Garrett ME
    2. Nauhaus I
    3. Marshel JH
    4. Callaway EM

    (2014) Topography and areal organization of mouse visual cortex

    Journal of Neuroscience 34:12587–12600.

    https://doi.org/10.1523/JNEUROSCI.1124-14.2014

    • PubMed
    • Google Scholar
15. 1. Goris RL
    2. Movshon JA
    3. Simoncelli EP

    (2014) Partitioning neuronal variability

    Nature Neuroscience 17:858–865.

    https://doi.org/10.1038/nn.3711

    • PubMed
    • Google Scholar
16. 1. He BJ

    (2013) Spontaneous and task-evoked brain activity negatively interact

    Journal of Neuroscience 33:4672–4682.

    https://doi.org/10.1523/JNEUROSCI.2922-12.2013

    • PubMed
    • Google Scholar
17. 1. Kalatsky VA
    2. Stryker MP

    (2003)

    New paradigm for optical imaging: temporally encoded maps of intrinsic signal

    Neuron 38:529–545.

    • PubMed
    • Google Scholar
18. 1. Kenet T
    2. Bibitchkov D
    3. Tsodyks M
    4. Grinvald A
    5. Arieli A

    (2003) Spontaneously emerging cortical representations of visual attributes

    Nature 425:954–956.

    https://doi.org/10.1038/nature02078

    • PubMed
    • Google Scholar
19. 1. Lin IC
    2. Okun M
    3. Carandini M
    4. Harris KD

    (2015) The nature of shared cortical variability

    Neuron 87:644–656.

    https://doi.org/10.1016/j.neuron.2015.06.035

    • PubMed
    • Google Scholar
20. 1. Ma Y
    2. Shaik MA
    3. Kim SH
    4. Kozberg MG
    5. Thibodeaux DN
    6. Zhao HT
    7. Yu H
    8. Hillman EMC

    (2016) Wide-field optical mapping of neural activity and brain haemodynamics: considerations and novel approaches

    Philosophical Transactions of the Royal Society B: Biological Sciences 371:.

    https://doi.org/10.1098/rstb.2015.0360

    • Google Scholar
21. 1. Madisen L
    2. Garner AR
    3. Shimaoka D
    4. Chuong AS
    5. Klapoetke NC
    6. Li L
    7. van der Bourg A
    8. Niino Y
    9. Egolf L
    10. Monetti C
    11. Gu H
    12. Mills M
    13. Cheng A
    14. Tasic B
    15. Nguyen TN
    16. Sunkin SM
    17. Benucci A
    18. Nagy A
    19. Miyawaki A
    20. Helmchen F
    21. Empson RM
    22. Knöpfel T
    23. Boyden ES
    24. Reid RC
    25. Carandini M
    26. Zeng H

    (2015) Transgenic mice for intersectional targeting of neural sensors and effectors with high specificity and performance

    Neuron 85:942–958.

    https://doi.org/10.1016/j.neuron.2015.02.022

    • Google Scholar
22. 1. Magnuson ME
    2. Thompson GJ
    3. Pan WJ
    4. Keilholz SD

    (2014) Effects of severing the corpus callosum on electrical and BOLD functional connectivity and spontaneous dynamic activity in the rat brain

    Brain Connectivity 4:15–29.

    https://doi.org/10.1089/brain.2013.0167

    • PubMed
    • Google Scholar
23. 1. Markowitz JE
    2. Gillis WF
    3. Beron CC
    4. Neufeld SQ
    5. Robertson K
    6. Bhagat ND
    7. Peterson RE
    8. Peterson E
    9. Hyun M
    10. Linderman SW
    11. Sabatini BL
    12. Datta SR

    (2018) The striatum organizes 3D behavior via Moment-to-Moment action selection

    Cell 174:44–58.

    https://doi.org/10.1016/j.cell.2018.04.019

    • PubMed
    • Google Scholar
24. 1. Mohajerani MH
    2. McVea DA
    3. Fingas M
    4. Murphy TH

    (2010) Mirrored bilateral slow-wave cortical activity within local circuits revealed by fast bihemispheric voltage-sensitive dye imaging in anesthetized and awake mice

    Journal of Neuroscience 30:3745–3751.

    https://doi.org/10.1523/JNEUROSCI.6437-09.2010

    • PubMed
    • Google Scholar
25. 1. Mohajerani MH
    2. Chan AW
    3. Mohsenvand M
    4. LeDue J
    5. Liu R
    6. McVea DA
    7. Boyd JD
    8. Wang YT
    9. Reimers M
    10. Murphy TH

    (2013) Spontaneous cortical activity alternates between motifs defined by regional axonal projections

    Nature Neuroscience 16:1426–1435.

    https://doi.org/10.1038/nn.3499

    • PubMed
    • Google Scholar
26. Preprint

    1. Musall S
    2. Kaufman MT
    3. Gluf S
    4. Churchland AK

    (2018) Movement-related activity dominates cortex during sensory-guided decision making

    bioRxiv.

    https://doi.org/10.1101/308288

    • Google Scholar
27. 1. O'Connor DH
    2. Hires SA
    3. Guo ZV
    4. Li N
    5. Yu J
    6. Sun QQ
    7. Huber D
    8. Svoboda K

    (2013) Neural coding during active somatosensation revealed using illusory touch

    Nature Neuroscience 16:958–965.

    https://doi.org/10.1038/nn.3419

    • PubMed
    • Google Scholar
28. 1. Rabinowitz NC
    2. Goris RL
    3. Cohen M
    4. Simoncelli EP

    (2015) Attention stabilizes the shared gain of V4 populations

    eLife 4:e08998.

    https://doi.org/10.7554/eLife.08998

    • PubMed
    • Google Scholar
29. 1. Ratzlaff EH
    2. Grinvald A

    (1991) A tandem-lens epifluorescence macroscope: hundred-fold brightness advantage for wide-field imaging

    Journal of Neuroscience Methods 36:127–137.

    https://doi.org/10.1016/0165-0270(91)90038-2

    • PubMed
    • Google Scholar
30. 1. Schölvinck ML
    2. Saleem AB
    3. Benucci A
    4. Harris KD
    5. Carandini M

    (2015) Cortical state determines global variability and correlations in visual cortex

    Journal of Neuroscience 35:170–178.

    https://doi.org/10.1523/JNEUROSCI.4994-13.2015

    • PubMed
    • Google Scholar
31. 1. Sereno MI
    2. McDonald CT
    3. Allman JM

    (1994) Analysis of retinotopic maps in extrastriate cortex

    Cerebral Cortex 4:601–620.

    https://doi.org/10.1093/cercor/4.6.601

    • PubMed
    • Google Scholar
32. 1. Shimaoka D
    2. Song C
    3. Knöpfel T

    (2017) State-Dependent modulation of slow wave motifs towards awakening

    Frontiers in Cellular Neuroscience 11:108.

    https://doi.org/10.3389/fncel.2017.00108

    • PubMed
    • Google Scholar
33. 1. Sreenivasan V
    2. Esmaeili V
    3. Kiritani T
    4. Galan K
    5. Crochet S
    6. Petersen CCH

    (2016) Movement initiation signals in mouse whisker motor cortex

    Neuron 92:1368–1382.

    https://doi.org/10.1016/j.neuron.2016.12.001

    • PubMed
    • Google Scholar
34. 1. Steinmetz NA
    2. Buetfering C
    3. Lecoq J
    4. Lee CR
    5. Peters AJ
    6. Jacobs EAK
    7. Coen P
    8. Ollerenshaw DR
    9. Valley MT
    10. de Vries SEJ
    11. Garrett M
    12. Zhuang J
    13. Groblewski PA
    14. Manavi S
    15. Miles J
    16. White C
    17. Lee E
    18. Griffin F
    19. Larkin JD
    20. Roll K
    21. Cross S
    22. Nguyen TV
    23. Larsen R
    24. Pendergraft J
    25. Daigle T
    26. Tasic B
    27. Thompson CL
    28. Waters J
    29. Olsen S
    30. Margolis DJ
    31. Zeng H
    32. Hausser M
    33. Carandini M
    34. Harris KD

    (2017) Aberrant cortical activity in multiple GCaMP6-Expressing transgenic mouse lines

    eNeuro 4:ENEURO.0207-17.2017.

    https://doi.org/10.1523/ENEURO.0207-17.2017

    • PubMed
    • Google Scholar
35. 1. Tolhurst DJ
    2. Movshon JA
    3. Dean AF

    (1983) The statistical reliability of signals in single neurons in cat and monkey visual cortex

    Vision Research 23:775–785.

    https://doi.org/10.1016/0042-6989(83)90200-6

    • PubMed
    • Google Scholar
36. 1. Tyszka JM
    2. Kennedy DP
    3. Adolphs R
    4. Paul LK

    (2011) Intact bilateral resting-state networks in the absence of the corpus callosum

    Journal of Neuroscience 31:15154–15162.

    https://doi.org/10.1523/JNEUROSCI.1453-11.2011

    • PubMed
    • Google Scholar
37. 1. Vanni MP
    2. Chan AW
    3. Balbi M
    4. Silasi G
    5. Murphy TH

    (2017) Mesoscale mapping of mouse cortex reveals Frequency-Dependent cycling between distinct macroscale functional modules

    Journal of Neuroscience 37:7513–7533.

    https://doi.org/10.1523/JNEUROSCI.3560-16.2017

    • PubMed
    • Google Scholar
38. 1. Vanni MP
    2. Murphy TH

    (2014) Mesoscale transcranial spontaneous activity mapping in GCaMP3 transgenic mice reveals extensive reciprocal connections between areas of somatomotor cortex

    Journal of Neuroscience 34:15931–15946.

    https://doi.org/10.1523/JNEUROSCI.1818-14.2014

    • PubMed
    • Google Scholar
39. 1. Wekselblatt JB
    2. Flister ED
    3. Piscopo DM
    4. Niell CM

    (2016) Large-scale imaging of cortical dynamics during sensory perception and behavior

    Journal of Neurophysiology 115:2852–2866.

    https://doi.org/10.1152/jn.01056.2015

    • PubMed
    • Google Scholar
40. Preprint

    1. Zatka-Haas P
    2. Steinmetz NA
    3. Carandini M
    4. Harris KD

    (2019) Distinct contributions of mouse cortical areas to visual discrimination

    bioRxiv.

    https://doi.org/10.1101/501627

    • Google Scholar

Author details

1. Daisuke Shimaoka

   UCL Institute of Ophthalmology, University College London, London, United Kingdom

   Present address

   Biomedicine Discovery Institute & Department of Physiology, Monash University, Clayton, Australia

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Funding acquisition, Investigation, Visualization, Methodology, Writing—original draft, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9435-5448

2. Nicholas A Steinmetz

   1. UCL Institute of Ophthalmology, University College London, London, United Kingdom
   2. UCL Institute of Neurology, University College London, London, United Kingdom

   Present address

   Department of Biological Structure, University of Washington, Seattle, United States

   Contribution

   Resources, Software, Investigation, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-7029-2908

3. Kenneth D Harris

   UCL Institute of Neurology, University College London, London, United Kingdom

   Contribution

   Conceptualization, Supervision, Funding acquisition, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5930-6456

4. Matteo Carandini

   UCL Institute of Ophthalmology, University College London, London, United Kingdom

   Contribution

   Conceptualization, Supervision, Funding acquisition, Investigation, Project administration, Writing—review and editing

   For correspondence

   m.carandini@ucl.ac.uk

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0003-4880-7682

• 2,927

  views

• 542

  downloads

• 37

  citations

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