Neurons exhibit responses that are highly variable (Shadlen and Newsome, 1998), with nearby neurons in the cortex exhibiting correlated variability in their spiking output (Cohen and Kohn, 2011; Smith and Kohn, 2008; Smith and Sommer, 2013; Zohary et al., 1994). It has been estimated on theoretical grounds that even weak correlations substantially reduce the information coding capacity of a population (Zohary et al., 1994). Spatial attention can reduce correlated variability (often referred to as noise correlations) among neurons in macaque visual area V4 (Cohen and Maunsell, 2009; Mitchell et al., 2009), an area that is strongly modulated by spatial attention (Reynolds and Chelazzi, 2004). Mitchell et al., 2009 found that this reduction is restricted to low frequencies below 10 Hz. These studies have estimated, on theoretical grounds, that the reduction in correlated variability accounts for a large fraction (about 80%) of the perceptual benefit due to attention. However, these estimates rely on specific assumptions about the relationship between noise and signal correlations, and thereby, on how population signals are read out in the brain (Abbott and Dayan, 1999; Averbeck et al., 2006; Moreno-Bote et al., 2014; Panzeri et al., 1999). Theoretical studies using heterogeneous tuning curves and optimal readout have concluded that correlated variability does not necessarily limit information (Ecker et al., 2011; Shamir and Sompolinsky, 2006). Consistent with the interpretation that attention-dependent reductions in correlated variability improve perception, a recent study of the effects of naturally occurring fluctuations in neural correlations found improved sensory discrimination when neurons in Area V4 were desynchronized (Beaman et al., 2017). However, other studies have posited that correlations themselves may be induced by fluctuations in attention (Goris et al., 2014), resulting in variation in response gain that is shared across neurons, and other experiments have shown that attention can, under some conditions, also increase correlated neural response variability (Ruff and Cohen, 2014). Taken together, these studies call into question the simple idea that attention reduces correlations so as to improve sensory discrimination. Importantly, all of these studies are correlative in nature. The causal role of correlated variability in perception has not been tested and thus the proposal that low-frequency correlated variability is detrimental for perception has remained purely hypothetical.

Here, we sought to directly test the effects of correlated variability on sensory discrimination by using optogenetic activation to induce correlations in Area V4 as monkeys performed an orientation discrimination task near perceptual threshold. We exploited the fact that attentional modulation of correlated variability is both spatially- and frequency selective: attention-dependent reductions in correlation are restricted to low frequencies (<10 Hz (Mitchell et al., 2009)). We reasoned that the correlations that impair perception may have an inherent time scale, with low- but not high-frequency correlations impairing perception. If so, we would predict that the effects of correlations on perception should be specific to this low-frequency range.

We took advantage of a novel approach to primate optogenetics and electrophysiology (Nassi et al., 2015; Ruiz et al., 2013) in which the native dura mater is replaced by a silicone based artificial dura (Figure 1A, B). This approach provides an optically clear window into the awake-behaving primate brain and allows precise opto-electrophysiology. We injected a lenti-viral construct (lenti-CaMKIIa-C1V1_E162T-ts-EYFP) to preferentially drive expression of the depolarizing opsin C1V1 in excitatory neurons (Yizhar et al., 2011) in a restricted portion (200-300 µm diameter) of dorsal V4 of two macaque monkeys (Figure 1C). Despite some heterogeneity in orientation tuning width at each injection site, overall there was similar tuning among neurons within a site (Figure 2—figure supplement 1).

Figure 1

Download asset Open asset

We trained two monkeys to perform an attention-demanding orientation-change detection task (Figure 2A). The monkeys were spatially cued to attend to one of two spatial locations. In the ‘attend in’ condition, the monkeys were instructed to covertly attend to a spatial location within the receptive fields of neurons at the viral injection site, while maintaining fixation at a central fixation point. In the ‘attend-away’ condition attention was directed to a location of equal eccentricity across the vertical meridian. On each trial, a sequence of oriented Gabor stimuli simultaneously flashed on and off at both spatial locations (200 ms on, variable 200–400 ms off). At an unpredictable time (minimum 1 s, maximum 5 s), one of the two stimuli (95% probability at cued location; 5% probability at uncued location, ‘foil trials’) briefly changed in orientation (200 ms) and the monkey was rewarded for making a saccade to the location of orientation change. If no change occurred within 5 s (‘catch trials’, 13% of trials), the monkey was rewarded for holding fixation. We controlled task difficulty by varying the degree of orientation change and thereby obtained behavioral performance curves (psychometric functions) for each recording session (Figure 2B, Figure 2—figure supplement 2A,B). Impaired performance (Figure 2—figure supplement 2A, left panel, square symbol) and slower reaction times (Figure 2—figure supplement 2A, right panel, square symbol) were observed for the foil trials, indicating that the monkey was indeed using the spatial cue in performing the task.

Figure 2 with 5 supplements see all

Download asset Open asset

To test if low-frequency correlations impair discrimination we optically stimulated neurons at the opsin injection site with 4-5Hz sinusoidally modulated low-power laser stimulation, on a randomly chosen subset of trials (‘low-frequency stimulation’ condition). Our goal was to induce correlations without significantly altering the mean firing rates by using low-power stimulation. Significant changes in mean firing rate could have unknown effects such as masking of the stimulus evoked response. Equating firing rates also avoids any indirect effects of mean firing rate changes on spike-count correlations (Cohen and Kohn, 2011). We find that low-frequency optical stimulation modulates the timing of the neural response (Figure 4D) but does not alter the overall magnitude of the population response (Figure 4A, B, C. We replicate previous findings that attention reduces low-frequency spike-count correlations in the baseline (no optical stimulation) condition (Mitchell et al., 2009) (Figure 3A, left panel; gray versus white bars, $p=0.02$, t-test). As predicted, low-frequency optical stimulation increases low-frequency correlations (Figure 3A, left panel; blue versus gray bar, $p=0.045$, t-test). The induced correlations were at a level comparable in strength to that observed when attention was directed away from the RF location in the baseline condition (Figure 3A, left panel; blue versus white bar). Optogenetic activation is accompanied by a period of reduced activity following stimulation. By careful titration of laser intensity (amplitude of sinusoidal modulation) we were able to alter the timing of spiking without altering mean firing rate. This is shown in Figure 4: we see a robust increase in firing rate due to attention in both the low-frequency stimulation and baseline conditions (Figure 4A, B), but there is no significant rate increase due to optical stimulation either during the pre-stimulus blank period (Figure 4C, left panel, $p=0.49$, t-test; Figure 4—figure supplement 1, top-left panel) or during the stimulus presentation period (Figure 4C, right panel, ${\displaystyle p>0.1}$, t-test; Figure 4—figure supplement 1, bottom-left panel). Rather, unit activity shows phase locking to optical stimulation (Figure 4D,E; pre-stimulus period). The distributions of spiking activity with respect to the phase of the optical stimulation show significant deviation from what would be expected from a null distribution (Figure 4D, example units, $p\ll 0.01$; Figure 4E, population, ${\displaystyle p<0.01}$, Rayleigh test). The null distributions were derived from a rate-matched Poisson process. We see a similar phase locking due to optical stimulation during the stimulus presentation period (Figure 4F, ${\displaystyle p<0.01}$, Rayleigh test), although the peak of the phase-lock distribution for the stimulus presentation period occurs earlier (around 120°) compared to that for the pre-stimulus period (around 210°). This would be expected, as the neurons are depolarized by the visual stimulus and hence more easily pushed to spiking threshold by optogenetic depolarization, as compared to when no stimulus is present. Thus, the physiology data shows that we successfully induced correlated activity among neurons at the opsin site without affecting the response rates.

Figure 3

Download asset Open asset

Figure 4 with 5 supplements see all

Download asset Open asset

Behaviorally, we find that low-frequency stimulation impairs the monkey’s ability to detect fine orientation changes, and does so only at the opsin location, as indicated by impairment in the attend-in condition (Figure 2B, upper panel; Figure 2—figure supplement 2B), not in the attend-away condition (Figure 2B, lower panel), in which the monkey discriminated orientation at the contralateral location. To quantify this behavioral deficit, we estimated the threshold of the monkeys’ psychometric functions and calculated the change in threshold due to optical stimulation as a modulation index (perceptual discrimination modulation index, PDMI; see Materials and methods). We find a significant increase in PDMI due to low-frequency stimulation in the attend-in condition ($p=0.02$, t-test; Figure 2C), indicating impaired detection of fine orientation changes. Unexpectedly, we also found a significant increase in slope ($p=0.007$, t-test; Figure 2—figure supplement 5A), suggesting that the shift from non-detection to detection occurs over a narrower range of orientations in the laser stimulation condition. In a large fraction of individual behavioral sessions (Figure 2—figure supplement 5B), both the changes in threshold and slope were significant (20/42 sessions). 11/42 sessions had significant threshold change only, while 5/42 sessions had significant slope change only.

The impairment due to optical stimulation is location specific: there was no significant change in PDMI on trials when the monkey was cued to detect the target at the unstimulated location (attend-away condition, Figure 2D). Importantly, the impairment is also frequency specific. When we stimulate the neurons with 20Hz sinusoidally modulated low-power laser stimulation (‘high-frequency stimulation’ condition), we observed no significant change in PDMI (Figure 2E), despite a significant increase in high-frequency spike-count correlations (Figure 3B) and phase locking comparable to the low-frequency stimulation condition (Figure 4—figure supplement 2). As in the low-frequency stimulation condition, we find no significant changes in mean firing rates with the high-frequency stimulation condition (Figure 4—figure supplement 1, right panels). The stimuli in each sequence were presented with irregular timing to ensure that any impairment did not stem from stimuli appearing at a particular phase of the laser stimulation, such as the phase at which neural sensitivity was at its nadir. To verify that such phase alignment did not nonetheless occur by chance, we measured the phase of target stimulus onset for both low- and high-frequency stimulation and found no phase preference (Figure 4—figure supplement 3A). Nor was behavioral performance phase dependent (Figure 4—figure supplement 3B). A two-way ANOVA of normalized performance with factors 'laser-phase' (the two different phase bin arrangements are shown in Figure 4—figure supplement 3B) and 'delta-orientation' (the trial-by-trial difference between target and non-target orientation) revealed no significant main effect of laser phase ($F_{\text{laser-phase}}=1.53,p=0.2$, left bin arrangement in Figure 4—figure supplement 3B; $F_{\text{laser-phase}}=0.73,p=0.53$, right bin arrangement in Figure 4—figure supplement 3B), a significant main effect of orientation and no significant interaction between the factors. A second potential concern is that the laser might impair orientation discrimination by distorting or flattening orientation tuning curves. We find that orientation tuning curves were not significantly altered by the laser at either frequency (Figure 4—figure supplement 4), at least in the range of orientations used in the experiment. A two-way ANOVA of normalized firing rates with factors 'laser-condition' (low-frequency laser, no laser) and 'orientation' revealed no significant main effect of laser condition ($F_{\text{laser-condition}}=0.8,p=0.37$), a significant main effect of orientation and no significant interaction between the two factors. Similarly, we found no significant main effect of the high-frequency laser condition ($F_{\text{laser-condition}}=0.05,p=0.82$), a significant main effect of orientation and no significant interaction between the two factors. A third potential concern is that the rhythmic laser stimulation might cause a sort of frequency-dependent adaptation that would cause neurons to be less sensitive to visual stimuli presented at a similar frequency. If so, the low-frequency (4-5Hz) laser stimulation could reduce the responses evoked by 200ms visual stimuli, impairing the monkey’s ability to discriminate the stimuli, while the high-frequency laser stimulation might not cause this effect, explaining the observed impairment. To test this, we measured the firing rates evoked by the first four non-target stimulus flashes, on no-laser, low-frequency laser and high-frequency laser trials, in the 'attend in' condition (Figure 4—figure supplement 5). Though the first stimulus in the sequence evoked a stronger response than the subsequent stimuli (reflecting a form of visual stimulus-driven adaptation), we find no evidence that the addition of the laser at either frequency caused a change in mean firing rate. A two-way ANOVA of normalized firing rate with factors 'flash-position' (1,2,3 or 4) and 'laser condition' (low-frequency laser, high-frequency laser, no laser) revealed a significant main effect of flash position ($F_{\text{flash-position}}=3.33,p=0.02$), no significant main effect of laser condition ($F_{\text{laser-condition}}=0.65,p=0.52$) and no significant interaction between the two factors. For the small number of sessions ($n=15$) over which we could do this analysis, the PDMI trends toward significance for the low-frequency condition ($p=0.07$), but is highly non-significant for the high-frequency condition ($p=0.7$). Additionally, we did not find significant changes in false-alarm rates with either low- or high-frequency stimulation (${\displaystyle p>0.1}$, t-test; Figure 2—figure supplement 2E). This was true for false alarms made during catch trials as well as on non-catch trials. Thus, we find no evidence that laser stimulation caused our subjects to mis-perceive a non-target as a target.

V4 has patchy organization for orientation tuning, so simultaneously recorded neurons tended to prefer similar orientations (see Figure 2—figure supplement 1). Under these conditions, where signal correlations are positive, positive noise correlations should reduce discriminability (Averbeck et al., 2006) by increasing the overlap between neural responses evoked by discriminanda. In the present experiment, where the task was to discriminate target from non-target, this would predict that in sessions where we observed laser-induced perceptual impairment, we should observe laser-induced decreases in discriminability at the level of pairs or populations of neurons, especially on miss trials, when the monkey was unable to discriminate target from non-target. For each recording session, we calculated a neural measure of discriminability between non-target and target stimuli across simultaneously recorded neurons (neural discriminability modulation index, NDMI; Figure 5; see Materials and methods, (Cohen and Maunsell, 2010)). We then examined whether laser-induced increases in perceptual threshold were correlated with laser-induced reductions in neural discriminability on trials in which the target appeared at the opsin location (the 'attend in' condition). Figure 5B shows NDMI for each experimental session (calculated from miss trials where the animal failed to detect a target) against the corresponding PDMI. We examined this in two ways: by measuring the correlation in a N-dimensional space where N is the number of neurons recorded in a given session (left panel) or by measuring the average NDMI across neural pairs (right panel). As predicted, there is a strong and significant negative correlation in both analyses (Figure 5B; $\rho =-0.29\left(p=0.05\right)$, NDMI from all simultaneous neurons, left panel; $\rho =-0.42\left(p=0.01\right)$, average NDMI across all simultaneous pairs, right panel; robust correlation). In both analyses, the sessions in which low-frequency laser-induced correlations caused the strongest perceptual suppression all showed negative NDMIs. In other words, laser-induced reductions in discriminability at the neural level corresponded to increased perceptual thresholds. There was no significant correlation between NDMI and changes in the slope of the psychometric function. NDMI calculated from hit trials had no correlation with threshold or slope changes.

Figure 5

Download asset Open asset

To confirm whether it is possible to induce coherent activity in a neuronal ensemble due to sub-threshold rhythmic stimulation, we examined the consequences of such stimulation on a conductance-based model of excitatory and inhibitory neurons (Figure 6A; see Materials and methods). We calculated the strength of coherent activity in the network (spike-spike coherence, SSC) both with and without sub-threshold stimulation (Figure 6B,C). We quantified the change in coherence due to stimulation as a modulation index (SSC MI; Figure 6D). We find that it is indeed possible to induce coherent activity in the network at a desired frequency (Figure 6D, Figure 6—figure supplement 1D) and that this induction is robust to a wide range of network (Figure 6—figure supplement 1C, Figure 6—figure supplement 2A) and stimulation parameters (Figure 6—figure supplement 2B).

Figure 6 with 2 supplements see all

Download asset Open asset

The location specificity of the impairment also suggests that the impairment is not due to a phosphene effect (Jazayeri et al., 2012). If attention were drawn away from the unstimulated location by a phosphene we would expect impaired performance in the attend-away condition. We also verified that the impairment was not due to a thermal effect by stimulating a location in the chamber a few millimeters from the opsin site (Figure 2—figure supplement 4A) and not due to visual distractions caused by the laser light by stimulating outside the brain (Figure 2—figure supplement 4B). In both cases, we did not observe any changes in behavior.

One of the two animals used in this study was euthanized to verify that, as we have previously found in macaque V1 (Nassi et al., 2015), AAV5/CAMKII leads to selective expression in pyramidal neurons. The dura mater adhered to V4 bilaterally and we were unable to perform histology. The second animal (Monkey A) is currently in use in another study. Thus, we do not have a measure of AAV5/CAMKII selectivity in macaque V4. However, it is reasonable to assume that the opsin was strongly biased toward pyramidal neurons. The viral constructs and injection protocol used in the present study were identical to those used previously in V1 (Nassi et al., 2015). In that study, AAV5/CAMKII expressing C1V1/EYFP led to expression that was highly selective for excitatory neurons (Figure 1B of Nassi et al., 2015): of 119 neurons imaged in five different fields of view, only 2 (1.7%) were double labeled for both EYFP and parvalbumin/calretinin/calbindin, indicating that expression was heavily biased toward excitatory neurons. AAV5/CaMKII has also been used in other macaque brain areas including perirhinal cortex (Tamura et al., 2017), where it also led to opsin expression primarily in excitatory neurons. It is possible that AAV5/CAMKII leads to less selective expression in V4 pyramidal neurons, but whatever the degree of selectivity we were able to activate neuronal responses in a phase-dependent manner with low intensity laser stimulation. as needed to test the effect of correlated variability on perceptual discrimination in the present study.

Our results establish the first causal link between correlated variability and perception. The optogenetic stimulation protocol in our study, using sinusoidal modulation of laser irradiance, induces the kind of correlations in a local population of the cortex that might not be physiologically realistic. It nevertheless establishes the causal relevance of low-frequency correlated variability in perception and supports the hypothesis that attention-dependent reductions in correlated variability enhance perception. Recently, studies have theorized that only certain correlations – those that are indistinguishable from stimulus-induced correlations – are information limiting (Moreno-Bote et al., 2014). We speculate that the correlations induced in our study included such information-limiting correlations, resulting in the observed impairment. The timescale of these low-frequency correlations is consistent with inter-saccadic intervals (200–300 ms), which may be a relevant timeframe for gathering visual information (Yarbus et al., 1967). Decreases in correlated variability at this timescale could therefore be critical for perception. Our study paves the way for investigating the laminar and cell-class specific components of the cortical circuit that determine this critical component of perception.

Data for the main figures are available via Dryad (doi:10.5061/dryad.8v0k1j3).

1. 1. Nandy A
   2. Nassi J
   3. Jadi M
   4. Reynolds J

   (2019) Dryad Digital Repository

   Data from: Optogenetically induced low-frequency correlations impair perception.

   https://doi.org/10.5061/dryad.8v0k1j3

1. 1. Abbott LF
   2. Dayan P

   (1999) The effect of correlated variability on the accuracy of a population code

   Neural Computation 11:91–101.

   https://doi.org/10.1162/089976699300016827

   • PubMed
   • Google Scholar
2. 1. Appelle S

   (1972) Perception and discrimination as a function of stimulus orientation: the "oblique effect" in man and animals

   Psychological Bulletin 78:266–278.

   https://doi.org/10.1037/h0033117

   • PubMed
   • Google Scholar
3. 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
4. 1. Beaman CB
   2. Eagleman SL
   3. Dragoi V

   (2017) Sensory coding accuracy and perceptual performance are improved during the desynchronized cortical state

   Nature Communications 8:1308.

   https://doi.org/10.1038/s41467-017-01030-4

   • PubMed
   • Google Scholar
5. 1. Brunel N
   2. Wang XJ

   (2003) What determines the frequency of fast network oscillations with irregular neural discharges? I. synaptic dynamics and excitation-inhibition balance

   Journal of Neurophysiology 90:415–430.

   https://doi.org/10.1152/jn.01095.2002

   • PubMed
   • Google Scholar
6. 1. Campbell FW
   2. Kulikowski JJ
   3. Levinson J

   (1966) The effect of orientation on the visual resolution of gratings

   The Journal of Physiology 187:427–436.

   https://doi.org/10.1113/jphysiol.1966.sp008100

   • PubMed
   • Google Scholar
7. 1. Chapman B
   2. Bonhoeffer T

   (1998) Overrepresentation of horizontal and vertical orientation preferences in developing ferret area 17

   PNAS 95:2609–2614.

   https://doi.org/10.1073/pnas.95.5.2609

   • PubMed
   • Google Scholar
8. 1. Cohen MR
   2. Kohn A

   (2011) Measuring and interpreting neuronal correlations

   Nature Neuroscience 14:811–819.

   https://doi.org/10.1038/nn.2842

   • 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. Cohen MR
    2. Maunsell JH

    (2010) A neuronal population measure of attention predicts behavioral performance on individual trials

    Journal of Neuroscience 30:15241–15253.

    https://doi.org/10.1523/JNEUROSCI.2171-10.2010

    • PubMed
    • Google Scholar
11. 1. De Valois RL
    2. Yund EW
    3. Hepler N

    (1982) The orientation and direction selectivity of cells in macaque visual cortex

    Vision Research 22:531–544.

    https://doi.org/10.1016/0042-6989(82)90112-2

    • PubMed
    • Google Scholar
12. 1. Ecker AS
    2. Berens P
    3. Tolias AS
    4. Bethge M

    (2011) The effect of noise correlations in populations of diversely tuned neurons

    Journal of Neuroscience 31:14272–14283.

    https://doi.org/10.1523/JNEUROSCI.2539-11.2011

    • PubMed
    • Google Scholar
13. 1. Furmanski CS
    2. Engel SA

    (2000) An oblique effect in human primary visual cortex

    Nature Neuroscience 3:535–536.

    https://doi.org/10.1038/75702

    • PubMed
    • Google Scholar
14. 1. Girshick AR
    2. Landy MS
    3. Simoncelli EP

    (2011) Cardinal rules: visual orientation perception reflects knowledge of environmental statistics

    Nature Neuroscience 14:926–932.

    https://doi.org/10.1038/nn.2831

    • 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. Han X
    2. Qian X
    3. Bernstein JG
    4. Zhou HH
    5. Franzesi GT
    6. Stern P
    7. Bronson RT
    8. Graybiel AM
    9. Desimone R
    10. Boyden ES

    (2009) Millisecond-timescale optical control of neural dynamics in the nonhuman primate brain

    Neuron 62:191–198.

    https://doi.org/10.1016/j.neuron.2009.03.011

    • PubMed
    • Google Scholar
17. 1. Heeley DW
    2. Timney B

    (1988) Meridional anisotropies of orientation discrimination for sine wave gratings

    Vision Research 28:337–344.

    https://doi.org/10.1016/0042-6989(88)90162-9

    • PubMed
    • Google Scholar
18. 1. Izhikevich EM

    (2003) Simple model of spiking neurons

    IEEE Transactions on Neural Networks 14:1569–1572.

    https://doi.org/10.1109/TNN.2003.820440

    • PubMed
    • Google Scholar
19. 1. Jazayeri M
    2. Lindbloom-Brown Z
    3. Horwitz GD

    (2012) Saccadic eye movements evoked by optogenetic activation of primate V1

    Nature Neuroscience 15:1368–1370.

    https://doi.org/10.1038/nn.3210

    • PubMed
    • Google Scholar
20. 1. Li B
    2. Peterson MR
    3. Freeman RD

    (2003) Oblique effect: a neural basis in the visual cortex

    Journal of Neurophysiology 90:204–217.

    https://doi.org/10.1152/jn.00954.2002

    • PubMed
    • Google Scholar
21. 1. Mitchell JF
    2. Sundberg KA
    3. Reynolds JH

    (2009) Spatial attention decorrelates intrinsic activity fluctuations in macaque area V4

    Neuron 63:879–888.

    https://doi.org/10.1016/j.neuron.2009.09.013

    • PubMed
    • Google Scholar
22. 1. Mitra PP
    2. Pesaran B

    (1999) Analysis of dynamic brain imaging data

    Biophysical Journal 76:691–708.

    https://doi.org/10.1016/S0006-3495(99)77236-X

    • PubMed
    • Google Scholar
23. 1. Moreno-Bote R
    2. Beck J
    3. Kanitscheider I
    4. Pitkow X
    5. Latham P
    6. Pouget A

    (2014) Information-limiting correlations

    Nature Neuroscience 17:1410–1417.

    https://doi.org/10.1038/nn.3807

    • PubMed
    • Google Scholar
24. 1. Myme CI
    2. Sugino K
    3. Turrigiano GG
    4. Nelson SB

    (2003) The NMDA-to-AMPA ratio at synapses onto layer 2/3 pyramidal neurons is conserved across prefrontal and visual cortices

    Journal of Neurophysiology 90:771–779.

    https://doi.org/10.1152/jn.00070.2003

    • PubMed
    • Google Scholar
25. 1. Nandy AS
    2. Nassi JJ
    3. Reynolds JH

    (2017) Laminar organization of attentional modulation in macaque visual area V4

    Neuron 93:235–246.

    https://doi.org/10.1016/j.neuron.2016.11.029

    • PubMed
    • Google Scholar
26. 1. Nassi JJ
    2. Avery MC
    3. Cetin AH
    4. Roe AW
    5. Reynolds JH

    (2015) Optogenetic activation of normalization in alert macaque visual cortex

    Neuron 86:1504–1517.

    https://doi.org/10.1016/j.neuron.2015.05.040

    • PubMed
    • Google Scholar
27. 1. Orban GA
    2. Vandenbussche E
    3. Vogels R

    (1984) Human orientation discrimination tested with long stimuli

    Vision Research 24:121–128.

    https://doi.org/10.1016/0042-6989(84)90097-X

    • PubMed
    • Google Scholar
28. 1. Panzeri S
    2. Schultz SR
    3. Treves A
    4. Rolls ET

    (1999) Correlations and the encoding of information in the nervous system

    Proceedings of the Royal Society of London. Series B: Biological Sciences 266:1001–1012.

    https://doi.org/10.1098/rspb.1999.0736

    • Google Scholar
29. 1. Reynolds JH
    2. Chelazzi L

    (2004) Attentional modulation of visual processing

    Annual Review of Neuroscience 27:611–647.

    https://doi.org/10.1146/annurev.neuro.26.041002.131039

    • PubMed
    • Google Scholar
30. 1. Ringach DL
    2. Sapiro G
    3. Shapley R

    (1997) A subspace reverse-correlation technique for the study of visual neurons

    Vision Research 37:2455–2464.

    https://doi.org/10.1016/S0042-6989(96)00247-7

    • PubMed
    • Google Scholar
31. 1. Ruff DA
    2. Cohen MR

    (2014) Attention can either increase or decrease spike count correlations in visual cortex

    Nature Neuroscience 17:1591–1597.

    https://doi.org/10.1038/nn.3835

    • PubMed
    • Google Scholar
32. 1. Ruiz O
    2. Lustig BR
    3. Nassi JJ
    4. Cetin A
    5. Reynolds JH
    6. Albright TD
    7. Callaway EM
    8. Stoner GR
    9. Roe AW

    (2013) Optogenetics through windows on the brain in the nonhuman primate

    Journal of Neurophysiology 110:1455–1467.

    https://doi.org/10.1152/jn.00153.2013

    • PubMed
    • Google Scholar
33. 1. Shadlen MN
    2. Newsome WT

    (1998) The variable discharge of cortical neurons: implications for connectivity, computation, and information coding

    The Journal of Neuroscience 18:3870–3896.

    https://doi.org/10.1523/JNEUROSCI.18-10-03870.1998

    • PubMed
    • Google Scholar
34. 1. Shamir M
    2. Sompolinsky H

    (2006) Implications of neuronal diversity on population coding

    Neural Computation 18:1951–1986.

    https://doi.org/10.1162/neco.2006.18.8.1951

    • PubMed
    • Google Scholar
35. 1. Smith MA
    2. Kohn A

    (2008) Spatial and temporal scales of neuronal correlation in primary visual cortex

    Journal of Neuroscience 28:12591–12603.

    https://doi.org/10.1523/JNEUROSCI.2929-08.2008

    • PubMed
    • Google Scholar
36. 1. Smith MA
    2. Sommer MA

    (2013) Spatial and temporal scales of neuronal correlation in visual area V4

    Journal of Neuroscience 33:5422–5432.

    https://doi.org/10.1523/JNEUROSCI.4782-12.2013

    • PubMed
    • Google Scholar
37. 1. Tamura K
    2. Takeda M
    3. Setsuie R
    4. Tsubota T
    5. Hirabayashi T
    6. Miyamoto K
    7. Miyashita Y

    (2017) Conversion of object identity to object-general semantic value in the primate temporal cortex

    Science 357:687–692.

    https://doi.org/10.1126/science.aan4800

    • PubMed
    • Google Scholar
38. 1. Wang G
    2. Ding S
    3. Yunokuchi K

    (2003) Representation of cardinal contour overlaps less with representation of nearby angles in cat visual cortex

    Journal of Neurophysiology 90:3912–3920.

    https://doi.org/10.1152/jn.00219.2003

    • PubMed
    • Google Scholar
39. Book

    1. Yarbus AL
    2. Haigh B
    3. Rigss LA

    (1967)

    Eye Movements and Vision

    New York: Plenum press.

    • Google Scholar
40. 1. Yizhar O
    2. Fenno LE
    3. Prigge M
    4. Schneider F
    5. Davidson TJ
    6. O'Shea DJ
    7. Sohal VS
    8. Goshen I
    9. Finkelstein J
    10. Paz JT
    11. Stehfest K
    12. Fudim R
    13. Ramakrishnan C
    14. Huguenard JR
    15. Hegemann P
    16. Deisseroth K

    (2011) Neocortical excitation/inhibition balance in information processing and social dysfunction

    Nature 477:171–178.

    https://doi.org/10.1038/nature10360

    • PubMed
    • Google Scholar
41. 1. Zohary E
    2. Shadlen MN
    3. Newsome WT

    (1994) Correlated neuronal discharge rate and its implications for psychophysical performance

    Nature 370:140–143.

    https://doi.org/10.1038/370140a0

    • PubMed
    • Google Scholar

Author details

1. Anirvan Nandy

   1. Systems Neurobiology Laboratories, The Salk Institute for Biological Studies, La Jolla, United States
   2. Department of Neuroscience, Yale University, New Haven, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Funding acquisition, Investigation, Methodology, Writing—original draft, Writing—review and editing, Designed the experiments, Collected and analyzed the data, Developed the computational model, Ran the simulations

   For correspondence

   anirvan.nandy@yale.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-4225-5349

2. Jonathan J Nassi

   Systems Neurobiology Laboratories, The Salk Institute for Biological Studies, La Jolla, United States

   Present address

   Inscopix Inc, Palo Alto, United States

   Contribution

   Conceptualization, Investigation, Designed the experiments

   Competing interests

   No competing interests declared

3. Monika P Jadi

   Department of Psychiatry, Yale University, New Haven, United States

   Contribution

   Formal analysis, Visualization, Writing—review and editing

   For correspondence

   monika.jadi@yale.edu

   Competing interests

   No competing interests declared

4. John Reynolds

   Systems Neurobiology Laboratories, The Salk Institute for Biological Studies, La Jolla, United States

   Contribution

   Conceptualization, Supervision, Writing—original draft, Writing—review and editing, Designed the experiments, Wrote the manuscript

   Competing interests

   No competing interests declared

• 3,778

  views

• 546

  downloads

• 28

  citations

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