Behavior is guided not only by sensory input, but also by information held in working memory (WM). In primates, visually guided eye movements are among the most frequently occurring sensorimotor transformations. Saccadic eye movements occur approximately four to five times each second and require the integration of myriad visual features (e.g., motion and shape) into discrete movements that position visual targets onto the fovea. Furthermore, each movement decision reflects not only visual input, but also information held in WM, such as the behavioral relevance of particular objects and features (Bichot et al., 2005; Hollingworth et al., 2013; Hollingworth and Luck, 2009; Wolfe and Horowitz, 2004). How sensory input and WM are integrated in neural circuits to shape behavioral output remains unclear. Studies across multiple species have revealed evidence that WM functions are often associated with networks involved in sensorimotor transformations, including visual-saccadic transformations (Curtis and D'Esposito, 2006; Gnadt and Andersen, 1988; Guo et al., 2014; Knudsen et al., 1995; Knudsen and Knudsen, 1996; Kojima et al., 1996). In these networks, neurons, individually or collectively, often exhibit persistent signaling of information needed to successfully carry out subsequent behaviors. The prevalence of WM-related activity in sensorimotor networks suggests that this may be where WM exerts its influence on sensorimotor transformations. However, the exact mechanism and specific neural circuitry by which WM influences visually guided behaviors are still unknown.

Within primate neocortex, the output of feature-selective neurons in visual cortical areas converges retinotopically onto neurons in the frontal eye field (FEF) (Schall et al., 1995), the prefrontal area mostly directly involved in the control of saccades (Robinson and Fuchs, 1969; Schiller et al., 1979). Neurons within the FEF exhibit functional properties spanning the visual-motor spectrum and also include a substantial portion of neurons with persistent, WM-related activity (Bruce and Goldberg, 1985; Lawrence et al., 2005). These characteristics make the FEF a likely place to observe an influence of WM on incoming visual signals, particularly given that the FEF transmits a predominantly WM signal to visual cortex (Merrikhi et al., 2017). Although much is understood about the role of FEF neurons in the control of visually guided saccades (Schiller et al., 1979; Schlag-Rey et al., 1992; Tehovnik et al., 2000), and in the control of visual spatial attention (Bahmani et al., 2019; Gregoriou et al., 2009; Moore and Fallah, 2001), very little is known about how visual, motor, and memory signals are combined within the FEF. Models of FEF microcircuitry generally predict that visual cortical inputs synapse predominantly onto purely visual FEF neurons (e.g., Heinzle et al., 2007), yet even this is not known. Furthermore, it is also not known how those visual inputs interact with the current content of WM.

We examined the influence of WM on the efficacy of visual cortical input to the FEF in behaving monkeys. First, we identified FEF neurons with direct input from visual cortex using orthodromic stimulation from extrastriate area V4. Despite the common assumption of visual inputs synapsing onto purely visual FEF neurons, our results revealed that visual cortical input to the FEF instead preferentially targets neurons with both visual and motor properties. Next, we measured the effect of spatial WM on orthodromic activation of FEF neurons and found that the efficacy of visual inputs was enhanced by the memory of spatially corresponding locations. Specifically, the activity evoked in the FEF from visual cortex was larger in magnitude, more synchronous, and more rapid when V4 input matched the location held in WM. These results demonstrate how the content of WM can influence visuomotor transformations in the primate brain.

We measured the influence of WM on the efficacy of visual cortical inputs to prefrontal cortex in behaving monkeys. Monkeys performed a spatial WM task classically used to characterize FEF neuronal properties (Figure 1a; Bruce and Goldberg, 1985; Lawrence et al., 2005). Figure 1b shows the activity of an example FEF neuron when a monkey remembered a location either inside or outside of the response field (RF; In and Out conditions, respectively). The neuron responded strongly to a visual cue appearing in the RF and exhibited elevated activity in the delay period when the remembered location coincided with the RF. Prior to saccades to the RF, the neuron also exhibited a burst of motor activity. In primates, direct visual cortical input to prefrontal cortex arrives primarily in the FEF (Markov et al., 2014a; Ungerleider et al., 2008). We orthodromically activated FEF neurons from retinotopically corresponding sites in extrastriate area V4 (Figure 1c) ('Materials and methods'). We targeted overlapping FEF-V4 RFs to maximize the likelihood of finding FEF neurons with input from V4, given the evidence of discrete, retinotopic projections from the latter to the former (Schall et al., 1995; Ungerleider et al., 2008). Of the 311 single FEF neurons recorded, spikes were reliably elicited by V4 stimulation in 115. Latencies of evoked spikes were bimodally distributed (Hartigan’s dip test, p<10⁻⁴⁰), with most neurons having latencies <10 ms (n = 96, latency = 6.53 ± 0.67 ms), consistent with monosynaptic transmission (Figure 1d; Gregoriou et al., 2009; Nowak and Bullier, 1997). We focused our analyses on these visual-recipient neurons. A smaller population of neurons was activated at longer latencies (n = 19, latency = 12.73 ± 1.40 ms). Visual-recipient neurons exhibited a tri-phasic pattern of evoked activity following orthodromic stimulation (Figure 1e, Figure 1—figure supplement 1), similar to previous studies employing orthodromic stimulation in primate neocortex (Matsunami and Hamada, 1984).

Figure 1 with 1 supplement see all

Download asset Open asset

Characterizing response properties of visual-recipient neurons

We tested whether the properties of visual-recipient neurons differed from those not activated by orthodromic stimulation (non-activated neurons) (n = 196). We measured the activity of FEF neurons while monkeys performed a spatial WM task that temporally dissociates visual, memory, and motor components of neuronal responses ('Materials and methods'). Figure 2a shows the activity of an example visual-recipient FEF neuron when the monkey remembered a location either inside or outside of the RF. This neuron responded strongly to a visual cue appearing in the RF but exhibited minimal activity in the delay period, and it was not selective for the remembered location. Prior to saccades to the RF, the neuron exhibited a burst of motor activity. Thus, this neuron exhibited visual and motor, not memory-related, activity. We compared the proportions of neurons with significant visual, memory, and motor activity between the visual-recipient and non-activated neuronal populations. Each component of activity was measured as the significant response enhancement in the corresponding behavioral epoch during the In condition (see 'Materials and methods'). We found that visual-recipient FEF neurons exhibited greater proportions of visual (χ² = 9.42, p = 0.002) and motor activity (χ² = 10.71, p = 0.001) than non-activated neurons. However, the proportion of neurons with memory activity did not differ between the two populations (χ² = 0.99, p = 0.318) (Figure 2b, left). Next, we asked whether the greater proportion of neurons with visual and motor activity among visual-recipient neurons corresponded to a larger proportion of visuomotor neurons. To test that, we compared the proportion of visuomotor neurons between the visual-recipient and non-activated neurons with significant selectivity between the In and Out conditions during the visual and/or motor epochs (see 'Materials and methods') (n = 74, non-activated; n = 59, visual-recipient). Overall, the relative proportions of visual, visuomotor, and motor neurons differed between the two populations (χ² = 6.89, p = 0.0319), with a higher prevalence of visuomotor neurons among the visual-recipient population (66 vs 44%, χ² = 11.34, p<10⁻³), and a lower proportion of purely visual (19 vs 31%, χ² = 4.39, p = 0.036) and purely motor neurons (15 vs 26%, χ² = 4.23, p = 0.039) (Figure 2b, right). Thus, the increased prevalence of visual and motor selectivity among the visual-recipient neurons reflected a larger proportion of visuomotor neurons.

Figure 2 with 2 supplements see all

Download asset Open asset

We considered that the larger motor signals among visual-recipient neurons could have resulted from differences in the alignment of the cue stimulus with the centers of FEF visual and movement fields, as they can be significantly misaligned (Bruce and Goldberg, 1985; Schafer and Moore, 2011). Thus, we measured the magnitude of motor activity (selectivity to In vs Out) across varying amounts of visual activity in the two populations of neurons ('Materials and methods') (Figure 2c, left). This comparison revealed that for a given level of visual activity, visual-recipient neurons exhibited a larger component of motor activity than non-activated neurons (ANCOVA, F = 10.15, p = 0.002). In contrast, a corresponding comparison of memory and visual activity in the two populations revealed no differences (ANCOVA, F = 0.23, p = 0.631) (Figure 2c, right). Thus, whereas memory signals among visual-recipient FEF neurons were equal to those of non-activated neurons, motor signals were significantly overrepresented (see Figure 2—figure supplements 1 and 2). It should be noted, however, that in spite of the anatomical evidence of retinotopic projections from V4 to FEF (Schall et al., 1995; Ungerleider et al., 2008), one cannot rule out the possibility that non-activated FEF neurons receive inputs from V4 neurons with non-overlapping RFs. Those inputs could be distributed onto FEF neurons differently from those arriving retinotopically. Nonetheless, our results show that FEF neurons receiving retinotopic inputs from V4 exhibit stronger motor activity than those without such inputs.

WM alters efficacy of V4 input to FEF

In contrast to the disproportionate prevalence of motor signals among neurons receiving input from visual cortex, reciprocal projections of the FEF to visual cortex originate disproportionately from neurons with memory-related activity (Merrikhi et al., 2017). This implicates the FEF as a possible source of the observed memory-dependent modulation in visual cortex (Bahmani et al., 2018; Supèr et al., 2001; van Kerkoerle et al., 2017). It also suggests that memory-related signals may interact with the efficacy of visual inputs arriving in prefrontal cortex. To test this possibility, we compared the efficacy of orthodromic activation of FEF neurons from area V4 between WM for different spatial locations. Previous studies have shown that the efficacy of orthodromic activation of primary visual cortex neurons from the thalamus (Briggs et al., 2013), and of extrastriate cortex from primary visual cortex (Ruff and Cohen, 2016), are both altered during spatial attention. Using a similar approach, we compared the activity evoked from visual cortex by orthodromic stimulation during the delay period of the MGS task when monkeys remembered different cue locations.

We examined the influence of WM across the full population of 96 visual-recipient neurons. On average, the proportion of stimulation-evoked spikes increased by 19% when monkeys remembered locations inside the RF, compared to outside (Spike count_In = 0.22 ± 0.005, Spike count_Out = 0.18 ± 0.004; p<10⁻⁶) (Figure 3—figure supplement 1). For each neuron, we measured the evoked response magnitude to quantify the efficacy of stimulation (see 'Materials and methods') and compared the magnitude between trials with different memory locations. The magnitude of evoked responses was significantly greater when monkeys remembered cue locations inside the RF, compared to outside (Response magnitude_In = 0.77 ± 0.04, Response magnitude_Out = 0.68 ± 0.03; p<10⁻¹⁰) (Figure 3a). The increase in efficacy was independent of the delay period selectivity of FEF neurons (Figure 3—figure supplement 2; r = 0.12, p = 0.235, Pearson correlation). Thus, the efficacy of visual input to the FEF depended on the content of WM. In addition, we found that the latency of evoked spikes was slightly reduced when monkeys remembered locations inside, compared to outside, the RF (Latency_In = 7.88, Latency_Out = 8.04; p<10⁻¹⁰) (Figure 3b). Figure 3c shows the spikes evoked from two example FEF neurons in response to V4 stimulation during the memory period. Evoked spikes from one neuron increased by ~30% during memory of locations inside, compared to outside the RF. The number of evoked spikes from a second, simultaneously recorded, neuron was similar in the two RF conditions, but spike onset appeared more rapid during the In condition, consistent with the latency effects. As a consequence, when combined, the evoked spikes of the two neurons were more synchronous during memory of locations inside of the RF (Figure 3c) ('Materials and methods'). We compared the proportion of joint spikes in all simultaneously recorded pairs of visual-recipient neurons (n = 509) across the different memory locations. Overall, we found that the proportion of joint spiking, when controlled for firing rate (see 'Materials and methods'), increased by nearly 60% during memory of locations inside, compared to outside the RF (Prob_In = 0.103 ± 0.002, Prob_Out = 0.065 ± 0.001;, p<10⁻⁶³) (Figure 3d). Thus, activity evoked in the FEF from visual cortex was larger, more synchronous, and more rapid when monkeys engaged WM at RF locations.

Figure 3 with 2 supplements see all

Download asset Open asset

We found that visual inputs to the FEF disproportionately drove neurons with both visual and motor properties. Rather than exhibiting purely visual properties, as might be expected (Sato and Schall, 2003), visual-recipient FEF neurons also signaled the direction of impending eye movements. Although surprising, this result seems consistent with the pattern of visual cortical connections with the FEF (Barone et al., 2000; Markov et al., 2014b). V4 inputs terminate in all layers of the FEF (Ungerleider et al., 2008), thus potentially distributing inputs across different functional classes of neurons. The bias in those inputs toward visuomotor neurons indicates that rather than being integrated at a subsequent processing stage within the FEF, as proposed by models of FEF microcircuitry (Brown et al., 2004; Heinzle et al., 2007), sensory and motor signals are combined at the interface between visual and prefrontal cortex. Moreover, this result is consistent with the observation of equal visual latencies between visuomotor and visual FEF neurons (Schall, 1991), and the finding that visuomotor neurons exhibit greater visual discrimination than visual neurons during certain saccade tasks (Costello et al., 2013).

More importantly, we found that the engagement of WM at RF locations increased the efficacy of visual input to FEF neurons. When monkeys remembered RF locations, activity evoked in the FEF from visual cortex was larger in magnitude, more synchronous, and more rapid. The increased efficacy of visual input to the FEF was independent of delay period activity, and it was observed whether or not neurons exhibited WM-related activity. Given that visual cortical inputs appear to drive neurons across functional classes, this result suggests that those inputs interact constructively with WM-related signals within the FEF. Similarly, previous work has shown that the deployment of attention increases the input efficacy of thalamocortical (Briggs et al., 2013) and corticocortical (Ruff and Cohen, 2016) connections, and, thus, our results indicate that attention and WM exert analogous influences on visual input across multiple levels of the visual hierarchy. In both cases, the cellular-level mechanisms of the observed enhancement remain to be determined, as they could result either from a pre-synaptic or from a post-synaptic facilitative mechanism, or both. Moreover, those mechanisms may differ in important ways between attention and WM.

Evidence from multiple species suggests that top-down attention arises from biases in the selection of sensory input based on the content of WM (Desimone and Duncan, 1995; Knudsen, 2007; Miller and Cohen, 2001; Soto et al., 2010). A recent study indicates that FEF neurons with WM-related activity disproportionately provide input to neurons in area V4 (Merrikhi et al., 2017) and thus underlie the FEF’s contribution to the modulation of visual cortical activity classically observed during spatial attention (Ekstrom et al., 2009; Gregoriou et al., 2014; Moore and Armstrong, 2003). In contrast to visual cortical projections from the FEF, inputs to the FEF did not preferentially target memory-related neurons, but instead neurons with visuomotor properties. This suggests that interactions between the FEF and visual cortex are not strictly recurrent (Knudsen, 2007; Noudoost et al., 2014) and that memory activity within the FEF, rather than reinforcing its own content, may instead facilitate the transformation of visual inputs into motor commands. Combined, our results suggest a basis for the well-documented interdependence of attention, WM, and gaze control (Ikkai and Curtis, 2011; Jonikaitis and Moore, 2019), which at the circuit level remains an important puzzle to solve.

Two adult male rhesus monkeys (Macaca mulatta) were used in this study. All experimental procedures were in accordance with National Institutes of Health Guide for the Care and Use of Laboratory Animals, the Society for Neuroscience Guidelines and Policies, and Stanford University Animal Care and Use Committee.

The source Plexon datafile and the readme file on how to analyze the data are now publicly available via the repository.

1. 1. Noudoost B
   2. Moore T

   (2021) GIN

   FEF MGS with V4 stim.

   https://doi.org/10.12751/g-node.b0lth8

1. 1. Armstrong KM
   2. Moore T

   (2007) Rapid enhancement of visual cortical response discriminability by microstimulation of the frontal eye field

   PNAS 104:9499–9504.

   https://doi.org/10.1073/pnas.0701104104

   • PubMed
   • Google Scholar
2. 1. Bahmani Z
   2. Daliri MR
   3. Merrikhi Y
   4. Clark K
   5. Noudoost B

   (2018) Working memory enhances cortical representations via spatially specific coordination of spike times

   Neuron 97:967–979.

   https://doi.org/10.1016/j.neuron.2018.01.012

   • PubMed
   • Google Scholar
3. 1. Bahmani Z
   2. Clark K
   3. Merrikhi Y
   4. Mueller A
   5. Pettine W
   6. Isabel Vanegas M
   7. Moore T
   8. Noudoost B

   (2019) Prefrontal contributions to attention and working memory

   Current Topics in Behavioral Neurosciences 41:74.

   https://doi.org/10.1007/7854_2018_74

   • Google Scholar
4. 1. Barone P
   2. Batardiere A
   3. Knoblauch K
   4. Kennedy H

   (2000) Laminar distribution of neurons in extrastriate Areas projecting to visual Areas V1 and V4 correlates with the hierarchical rank and indicates the operation of a distance rule

   The Journal of Neuroscience 20:3263–3281.

   https://doi.org/10.1523/JNEUROSCI.20-09-03263.2000

   • PubMed
   • Google Scholar
5. 1. Bichot NP
   2. Rossi AF
   3. Desimone R

   (2005) Parallel and serial neural mechanisms for visual search in macaque area V4

   Science 308:529–534.

   https://doi.org/10.1126/science.1109676

   • PubMed
   • Google Scholar
6. 1. Briggs F
   2. Mangun GR
   3. Usrey WM

   (2013) Attention enhances synaptic efficacy and the signal-to-noise ratio in neural circuits

   Nature 499:476–480.

   https://doi.org/10.1038/nature12276

   • PubMed
   • Google Scholar
7. 1. Britten KH
   2. Shadlen MN
   3. Newsome WT
   4. Movshon JA

   (1992) The analysis of visual motion: a comparison of neuronal and psychophysical performance

   The Journal of Neuroscience 12:4745–4765.

   https://doi.org/10.1523/JNEUROSCI.12-12-04745.1992

   • PubMed
   • Google Scholar
8. 1. Brown JW
   2. Bullock D
   3. Grossberg S

   (2004) How laminar frontal cortex and basal ganglia circuits interact to control planned and reactive saccades

   Neural Networks 17:471–510.

   https://doi.org/10.1016/j.neunet.2003.08.006

   • PubMed
   • Google Scholar
9. 1. Bruce CJ
   2. Goldberg ME

   (1985) Primate frontal eye fields. I. single neurons discharging before saccades

   Journal of Neurophysiology 53:603–635.

   https://doi.org/10.1152/jn.1985.53.3.603

   • PubMed
   • Google Scholar
10. 1. Costello MG
    2. Zhu D
    3. Salinas E
    4. Stanford TR

    (2013) Perceptual modulation of motor--but not visual--responses in the frontal eye field during an urgent-decision task

    Journal of Neuroscience 33:16394–16408.

    https://doi.org/10.1523/JNEUROSCI.1899-13.2013

    • PubMed
    • Google Scholar
11. 1. Curtis CE
    2. D'Esposito M

    (2006) Selection and maintenance of saccade goals in the human frontal eye fields

    Journal of Neurophysiology 95:3923–3927.

    https://doi.org/10.1152/jn.01120.2005

    • PubMed
    • Google Scholar
12. 1. Desimone R
    2. Duncan J

    (1995) Neural mechanisms of selective visual attention

    Annual Review of Neuroscience 18:193–222.

    https://doi.org/10.1146/annurev.ne.18.030195.001205

    • PubMed
    • Google Scholar
13. 1. Ekstrom LB
    2. Roelfsema PR
    3. Arsenault JT
    4. Kolster H
    5. Vanduffel W

    (2009) Modulation of the contrast response function by electrical microstimulation of the macaque frontal eye field

    Journal of Neuroscience 29:10683–10694.

    https://doi.org/10.1523/JNEUROSCI.0673-09.2009

    • PubMed
    • Google Scholar
14. 1. Gnadt JW
    2. Andersen RA

    (1988) Memory related motor planning activity in posterior parietal cortex of macaque

    Experimental Brain Research 70:216–220.

    https://doi.org/10.1007/BF00271862

    • PubMed
    • Google Scholar
15. Book

    1. Green DM
    2. Swets JA

    (1966)

    Signal Detection Theory and Psychophysics

    Wiley.

    • Google Scholar
16. 1. Gregoriou GG
    2. Gotts SJ
    3. Zhou H
    4. Desimone R

    (2009) High-frequency, long-range coupling between prefrontal and visual cortex during attention

    Science 324:1207–1210.

    https://doi.org/10.1126/science.1171402

    • PubMed
    • Google Scholar
17. 1. Gregoriou GG
    2. Rossi AF
    3. Ungerleider LG
    4. Desimone R

    (2014) Lesions of prefrontal cortex reduce attentional modulation of neuronal responses and synchrony in V4

    Nature Neuroscience 17:1003–1011.

    https://doi.org/10.1038/nn.3742

    • PubMed
    • Google Scholar
18. 1. Guo ZV
    2. Li N
    3. Huber D
    4. Ophir E
    5. Gutnisky D
    6. Ting JT
    7. Feng G
    8. Svoboda K

    (2014) Flow of cortical activity underlying a tactile decision in mice

    Neuron 81:179–194.

    https://doi.org/10.1016/j.neuron.2013.10.020

    • PubMed
    • Google Scholar
19. 1. Heinzle J
    2. Hepp K
    3. Martin KA

    (2007) A microcircuit model of the frontal eye fields

    Journal of Neuroscience 27:9341–9353.

    https://doi.org/10.1523/JNEUROSCI.0974-07.2007

    • PubMed
    • Google Scholar
20. 1. Hollingworth A
    2. Matsukura M
    3. Luck SJ

    (2013) Visual working memory modulates rapid eye movements to simple onset targets

    Psychological Science 24:790–796.

    https://doi.org/10.1177/0956797612459767

    • PubMed
    • Google Scholar
21. 1. Hollingworth A
    2. Luck SJ

    (2009) The role of visual working memory (VWM) in the control of gaze during visual search

    Attention, Perception, & Psychophysics 71:936–949.

    https://doi.org/10.3758/APP.71.4.936

    • PubMed
    • Google Scholar
22. 1. Ikkai A
    2. Curtis CE

    (2011) Common neural mechanisms supporting spatial working memory, attention and motor intention

    Neuropsychologia 49:1428–1434.

    https://doi.org/10.1016/j.neuropsychologia.2010.12.020

    • PubMed
    • Google Scholar
23. 1. Jonikaitis D
    2. Moore T

    (2019) The interdependence of attention, working memory and gaze control: behavior and neural circuitry

    Current Opinion in Psychology 29:126–134.

    https://doi.org/10.1016/j.copsyc.2019.01.012

    • PubMed
    • Google Scholar
24. 1. Knudsen EI
    2. Cohen YE
    3. Masino T

    (1995) Characterization of a forebrain gaze field in the archistriatum of the barn owl: microstimulation and anatomical connections

    The Journal of Neuroscience 15:5139–5151.

    https://doi.org/10.1523/JNEUROSCI.15-07-05139.1995

    • PubMed
    • Google Scholar
25. 1. Knudsen EI

    (2007) Fundamental components of attention

    Annual Review of Neuroscience 30:57–78.

    https://doi.org/10.1146/annurev.neuro.30.051606.094256

    • PubMed
    • Google Scholar
26. 1. Knudsen EI
    2. Knudsen PF

    (1996) Disruption of auditory spatial working memory by inactivation of the forebrain archistriatum in barn owls

    Nature 383:428–431.

    https://doi.org/10.1038/383428a0

    • PubMed
    • Google Scholar
27. 1. Kojima J
    2. Matsumura M
    3. Togawa M
    4. Hikosaka O

    (1996) Tonic activity during visuo-oculomotor behavior in the monkey superior colliculus

    Neuroscience Research 26:17–28.

    https://doi.org/10.1016/0168-0102(96)01067-X

    • PubMed
    • Google Scholar
28. 1. Lawrence BM
    2. White RL
    3. Snyder LH

    (2005) Delay-period activity in visual, Visuomovement, and movement neurons in the frontal eye field

    Journal of Neurophysiology 94:1498–1508.

    https://doi.org/10.1152/jn.00214.2005

    • PubMed
    • Google Scholar
29. 1. Markov NT
    2. Ercsey-Ravasz MM
    3. Ribeiro Gomes AR
    4. Lamy C
    5. Magrou L
    6. Vezoli J
    7. Misery P
    8. Falchier A
    9. Quilodran R
    10. Gariel MA
    11. Sallet J
    12. Gamanut R
    13. Huissoud C
    14. Clavagnier S
    15. Giroud P
    16. Sappey-Marinier D
    17. Barone P
    18. Dehay C
    19. Toroczkai Z
    20. Knoblauch K
    21. Van Essen DC
    22. Kennedy H

    (2014a) A weighted and directed interareal connectivity matrix for macaque cerebral cortex

    Cerebral Cortex 24:17–36.

    https://doi.org/10.1093/cercor/bhs270

    • PubMed
    • Google Scholar
30. 1. Markov NT
    2. Vezoli J
    3. Chameau P
    4. Falchier A
    5. Quilodran R
    6. Huissoud C
    7. Lamy C
    8. Misery P
    9. Giroud P
    10. Ullman S
    11. Barone P
    12. Dehay C
    13. Knoblauch K
    14. Kennedy H

    (2014b) Anatomy of hierarchy: feedforward and feedback pathways in macaque visual cortex

    Journal of Comparative Neurology 522:225–259.

    https://doi.org/10.1002/cne.23458

    • Google Scholar
31. 1. Matsunami K
    2. Hamada I

    (1984) Effects of stimulation of corpus callosum on precentral neuron activity in the awake monkey

    Journal of Neurophysiology 52:676–691.

    https://doi.org/10.1152/jn.1984.52.4.676

    • PubMed
    • Google Scholar
32. 1. Merrikhi Y
    2. Clark K
    3. Albarran E
    4. Parsa M
    5. Zirnsak M
    6. Moore T
    7. Noudoost B

    (2017) Spatial working memory alters the efficacy of input to visual cortex

    Nature Communications 8:15041.

    https://doi.org/10.1038/ncomms15041

    • PubMed
    • Google Scholar
33. 1. Miller EK
    2. Cohen JD

    (2001) An integrative theory of prefrontal cortex function

    Annual Review of Neuroscience 24:167–202.

    https://doi.org/10.1146/annurev.neuro.24.1.167

    • PubMed
    • Google Scholar
34. 1. Moore T
    2. Armstrong KM

    (2003) Selective gating of visual signals by microstimulation of frontal cortex

    Nature 421:370–373.

    https://doi.org/10.1038/nature01341

    • PubMed
    • Google Scholar
35. 1. Moore T
    2. Fallah M

    (2001) Control of eye movements and spatial attention

    PNAS 98:1273–1276.

    https://doi.org/10.1073/pnas.98.3.1273

    • PubMed
    • Google Scholar
36. 1. Noudoost B
    2. Clark KL
    3. Moore T

    (2014) A distinct contribution of the frontal eye field to the visual representation of saccadic targets

    Journal of Neuroscience 34:3687–3698.

    https://doi.org/10.1523/JNEUROSCI.3824-13.2014

    • PubMed
    • Google Scholar
37. 1. Nowak LG
    2. Bullier J

    (1997) The timing of information transfer in the visual system

    Cerebral Cortex 12:205–241.

    https://doi.org/10.1007/978-1-4757-9625-4_5

    • Google Scholar
38. 1. Robinson DA
    2. Fuchs AF

    (1969) Eye movements evoked by stimulation of frontal eye fields

    Journal of Neurophysiology 32:637–648.

    https://doi.org/10.1152/jn.1969.32.5.637

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

    (2016) Attention increases spike count correlations between visual cortical Areas

    Journal of Neuroscience 36:7523–7534.

    https://doi.org/10.1523/JNEUROSCI.0610-16.2016

    • PubMed
    • Google Scholar
40. 1. Sato TR
    2. Schall JD

    (2003) Effects of stimulus-response compatibility on neural selection in frontal eye field

    Neuron 38:637–648.

    https://doi.org/10.1016/S0896-6273(03)00237-X

    • PubMed
    • Google Scholar
41. 1. Schafer RJ
    2. Moore T

    (2011) Selective attention from voluntary control of neurons in prefrontal cortex

    Science 332:1568–1571.

    https://doi.org/10.1126/science.1199892

    • PubMed
    • Google Scholar
42. 1. Schall JD

    (1991) Neuronal activity related to visually guided saccades in the frontal eye fields of rhesus monkeys: comparison with supplementary eye fields

    Journal of Neurophysiology 66:559–579.

    https://doi.org/10.1152/jn.1991.66.2.559

    • PubMed
    • Google Scholar
43. 1. Schall JD
    2. Morel A
    3. King DJ
    4. Bullier J

    (1995) Topography of visual cortex connections with frontal eye field in macaque: convergence and segregation of processing streams

    The Journal of Neuroscience 15:4464–4487.

    https://doi.org/10.1523/JNEUROSCI.15-06-04464.1995

    • PubMed
    • Google Scholar
44. 1. Schiller P
    2. True S
    3. Conway J

    (1979) Effects of frontal eye field and superior colliculus ablations on eye movements

    Science 206:590–592.

    https://doi.org/10.1126/science.115091

    • Google Scholar
45. 1. Schlag-Rey M
    2. Schlag J
    3. Dassonville P

    (1992) How the frontal eye field can impose a saccade goal on superior colliculus neurons

    Journal of Neurophysiology 67:1003–1005.

    https://doi.org/10.1152/jn.1992.67.4.1003

    • Google Scholar
46. 1. Soto D
    2. Wriglesworth A
    3. Bahrami-Balani A
    4. Humphreys GW

    (2010) Working memory enhances visual perception: evidence from signal detection analysis

    Journal of Experimental Psychology: Learning, Memory, and Cognition 36:441–456.

    https://doi.org/10.1037/a0018686

    • Google Scholar
47. 1. Supèr H
    2. Spekreijse H
    3. Lamme VA

    (2001) A neural correlate of working memory in the monkey primary visual cortex

    Science 293:120–124.

    https://doi.org/10.1126/science.1060496

    • PubMed
    • Google Scholar
48. 1. Tehovnik EJ
    2. Sommer MA
    3. Chou IH
    4. Slocum WM
    5. Schiller PH

    (2000) Eye fields in the frontal lobes of primates

    Brain Research Reviews 32:413–448.

    https://doi.org/10.1016/S0165-0173(99)00092-2

    • PubMed
    • Google Scholar
49. 1. Ungerleider LG
    2. Galkin TW
    3. Desimone R
    4. Gattass R

    (2008) Cortical connections of area V4 in the macaque

    Cerebral Cortex 18:477–499.

    https://doi.org/10.1093/cercor/bhm061

    • PubMed
    • Google Scholar
50. 1. van Kerkoerle T
    2. Self MW
    3. Roelfsema PR

    (2017) Layer-specificity in the effects of attention and working memory on activity in primary visual cortex

    Nature Communications 8:13804.

    https://doi.org/10.1038/ncomms13804

    • PubMed
    • Google Scholar
51. 1. Wolfe JM
    2. Horowitz TS

    (2004) What attributes guide the deployment of visual attention and how do they do it?

    Nature Reviews Neuroscience 5:495–501.

    https://doi.org/10.1038/nrn1411

    • PubMed
    • Google Scholar

Acceptance summary:

This study documented the properties of neurons in the frontal eye field (FEF), a cortical brain area traditionally thought to receive visual input and transform that input into motor commands. The authors used extracellular recordings and electrical microstimulation in behaving non-human primates (NHPs) to add to this view by showing that visual input to FEF from visual area V4 appears to be gated by working -memory activity in FEF. This kind of circuit-breaking is rare and valuable in NHPs and provides new insights into mechanisms that subserve the flexible, context-dependent flow of information in the primate brain.

Decision letter after peer review:

Thank you for submitting your article "Working Memory Gates Visual Input to Primate Prefrontal Neurons" for consideration by eLife. Your article has been reviewed by 2 peer reviewers, and the evaluation has been overseen by Joshua Gold as the Reviewing and Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Shengtao Yang (Reviewer #1).

The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Essential revisions:

1. Orthodromic activation of FEF neurons via V4 stimulation increases the percentage of FEF events that lead to spikes and decreases their latency during working memory. Such an effect appears expectable if FEF neurons are at a higher level of when a stimulus in their receptive field is held in memory compared to a stimulus out of their receptive field. Are the authors suggesting something special about working memory? Would the same outcome not be expected during fixation or smooth pursuit for FEF neurons that are activated by these states? It was not clear that efficacy of transmission itself improves by working memory, just the likelihood that the spiking threshold would be reached. This was considered a major concern.

2. It would strengthen the authors thesis to discuss the existing functional evidence (in addition to anatomical evidence) that motor FEF neurons receive visual input and can plan movements accordingly. See for example Costello et al. J. Neurosci 2013, 33(41):16394-408.

3. The authors match the receptive location of FEF and V4 neurons to maximize the chances of identifying monosynaptically connected neurons between the two areas. However, a negative finding of no effects of orthodromic activation does not entirely rule out that the FEF neuron under study receive V4 input, from another site. Some discussion is warranted on this point.

4. In classical working memory tasks, the task periods usually consist of fixation, cue, delay and then response period. The neural activity during delay period is considered as working memory related signal. However, in current study, the authors didn't point out whether only delay period activity was included in analysis when they compared synaptic efficacy between stimulation and non-stimulation trials, in Figure 4a. Because the differences of neuronal response during fixation period cannot be viewed as relevant to information held in working memory, it may be better if only neuronal activity in delay period was included in their analysis.

Relatedly, did all of the 96 visual-recipient FEF neurons show working memory-related activity in the memory-guided saccade task? The example neuron in Figure 3a didn't show a significant difference between In and Out trials during delay period. If the visual-recipient neuron didn't show working memory-related activity, what is the basis for the claim that enhanced synaptic efficacy from V4 to FEF was caused by working memory?

5. Did the two example neurons in Figure 4c show adjusted values (subtracting the same measure during non-stimulated trials)? The authors mentioned in Method that Figure 4 showed adjusted values, but it may not be applicable for raster plot in Figure 4c. It may be helpful that using adjusted values show stimulation effects on evoked spike counts during memory In and Out trials.

6. Did the authors find some FEF cells showing elevated firing during delay period in outside-RF trials compared with baseline firing? These elevated firings would not be caused by the RF cue and thus could more readily be interpreted as a working memory signal.

7. The sample size should be indicated in the Figure 3b Venn diagram.

8. It would be useful to indicate the electrical stimulus protocol in Figure 1.

Reviewer #1:

The authors of Working Memory Gates Visual Input to Primate Prefrontal Neurons studied how working memory influence information transmitting from V4 to frontal eye field via extracellular recording and electrical stimulation on behaving primate. They found that V4 neurons target FEF neurons with both visual and motor property, and its synaptic efficacy of V4 to FEF was enhanced by working memory. These findings are interesting and important to our understanding about how our brain acts during daily WM-related activity.

Reviewer #2:

This is a very interesting study, examining the properties of different types of neurons in the primate Frontal Eye Fields. It is commonly assumed that a serial processing of information takes place in the frontal lobe, from visual representation, to working memory maintenance, to motor output. However, some evidence to the contrary has also been reported, creating a debate in the field. The authors have characterized meticulously FEF neurons receiving V4 projections, by means of orthodromic stimulation. They report two main findings: that visual-input recipient neurons in FEF exhibit substantial motor activity and that working memory alters the efficacy of V4 input to FEF. The paper provides an important addition to our understanding of FEF processing. Although the first result is unambiguous, and goes against the traditional view of the FEF, the interpretation of the second is less straightforward and would need to be qualified further.

1. Orthodromic activation of FEF neurons via V4 stimulation increases the percentage of FEF events that lead to spikes and decreases their latency during working memory. Such an effect appears expectable if FEF neurons are at a higher level of when a stimulus in their receptive field is held in memory compared to a stimulus out of their receptive field. Are the authors suggesting something special about working memory? Would the same outcome not be expected during fixation or smooth pursuit for FEF neurons that are activated by these states? It was not clear that efficacy of transmission itself improves by working memory, just the likelihood that the spiking threshold would be reached.

2. It would strengthen the authors thesis to discuss the existing functional evidence (in addition to anatomical evidence) that motor FEF neurons receive visual input and can plan movements accordingly. See for example Costello et al. J. Neurosci 2013, 33(41):16394-408.

3. The authors match the receptive location of FEF and V4 neurons to maximize the chances of identifying monosynaptically connected neurons between the two areas. However, a negative finding of no effects of orthodromic activation does not entirely rule out that the FEF neuron under study receive V4 input, from another site. Some discussion is warranted on this point.

Essential revisions:

1. Orthodromic activation of FEF neurons via V4 stimulation increases the percentage of FEF events that lead to spikes and decreases their latency during working memory. Such an effect appears expectable if FEF neurons are at a higher level of when a stimulus in their receptive field is held in memory compared to a stimulus out of their receptive field. Are the authors suggesting something special about working memory? Would the same outcome not be expected during fixation or smooth pursuit for FEF neurons that are activated by these states? It was not clear that efficacy of transmission itself improves by working memory, just the likelihood that the spiking threshold would be reached. This was considered a major concern.

We agree that this is an important point worth addressing more clearly. Thus, we clarify in the revised paper in several ways that the observation of enhanced efficacy of orthodromic activation of FEF neurons is not necessarily expected. First, as described in the methods, and clarified further in the revision, evoked spiking responses are always measured as the change from the non-stimulation condition during the same time period. So response magnitude is normalized to this non-stimulation baseline, and thus one could have reasonably expected there to be no change in the evoked activity between the IN and OUT conditions if the combined (stimulation and endogenous) inputs are added linearly. Second, as we clarify in the revised paper, the enhancement effect is observed in neurons whether or not they exhibit delay activity. Indeed, neurons with equal spike rates in the IN and OUT conditions exhibit equally large enhancement effects. To make this point clearer, we have added a supplemental analysis and figure (Figure 3, Supplement 2) that is similar to that used previously in studies demonstrating similar increased efficacy with attention (see Supplementary Figure 1d from Briggs et al., 2013). This analysis shows no relationship between the change in efficacy and delay selectivity (i.e. difference in firing rate between IN and OUT conditions). So, whether or not changes in efficacy are also present during changes in ‘states’ other than WM (it appears so for attention, and could also be so during smooth pursuit), it is clear that the observed change is not merely due to ‘a higher level’ of activity in the FEF neurons. Note also that, on the contrary, we do not argue that WM is special (in the discussion). Instead, we draw parallels with the effects seen during attention and note other evidence of similar underlying mechanisms.

2. It would strengthen the authors thesis to discuss the existing functional evidence (in addition to anatomical evidence) that motor FEF neurons receive visual input and can plan movements accordingly. See for example Costello et al. J. Neurosci 2013, 33(41):16394-408.

We thank the reviewers for the suggestion, and have added the following text to the Discussion section:

“Moreover, this result is consistent with the observation of equal visual latencies between visuomotor and visual FEF neurons (Schall, 1991), and the finding that visuomotor neurons exhibit greater visual discrimination than visual neurons during certain saccade tasks (Costello et al., 2013).”

3. The authors match the receptive location of FEF and V4 neurons to maximize the chances of identifying monosynaptically connected neurons between the two areas. However, a negative finding of no effects of orthodromic activation does not entirely rule out that the FEF neuron under study receive V4 input, from another site. Some discussion is warranted on this point.

We have revised the wording in the results (pages 5 and 8) to clarify that we targeted overlapping FEF-V4 receptive fields based on the fact that projections from V4 to FEF appear to be retinotopic and discrete (Schall et al., 1995; Ungerleider et al., 2008). We now also note that to the extent that there are non-retinotopic projections they could indeed be distributed differently onto the functional subtypes of FEF neurons than the retinotopic projections (page 8).

4. In classical working memory tasks, the task periods usually consist of fixation, cue, delay and then response period. The neural activity during delay period is considered as working memory related signal. However, in current study, the authors didn't point out whether only delay period activity was included in analysis when they compared synaptic efficacy between stimulation and non-stimulation trials, in Figure 4a. Because the differences of neuronal response during fixation period cannot be viewed as relevant to information held in working memory, it may be better if only neuronal activity in delay period was included in their analysis.

This analysis of efficacy was indeed limited to the delay period. We regret that this important aspect of the results was not clear, and we have added language to the results (page 9), methods (page 17), and the figure 3 caption to clarify this point.

Relatedly, did all of the 96 visual-recipient FEF neurons show working memory-related activity in the memory-guided saccade task? The example neuron in Figure 3a didn't show a significant difference between In and Out trials during delay period. If the visual-recipient neuron didn't show working memory-related activity, what is the basis for the claim that enhanced synaptic efficacy from V4 to FEF was caused by working memory?

This is an important point of clarification and we thank the reviewer for bringing it up. Indeed, a majority of the visual-recipient neurons did not exhibit elevated delay period activity. As shown in the bar plots in Figure 2b, ~30% of visual-recipient FEF neurons have significant delay activity (no different from the proportion of the non-activated FEF neuronal population). In addition, as mentioned above, we have added an analysis of the relationship between delay-period selectivity and the change in efficacy with WM (Figure 3 Supplement 2). This figure and analysis demonstrates that differences in firing rates between the IN and OUT conditions do not correlate with the change in efficacy.

As for the basis for the claim that enhanced efficacy is a result of WM, that is due to the fact that the location of WM is the only thing that varies between the two behavioral (experimental) conditions between which efficacy is being compared. That is a conclusion that quite reasonably follows from the observed differences in orthodromic efficacy between the two behavioral conditions; the magnitude, latency and synchrony of evoked activity all depend on the location the monkey must remember. Note that we do not claim that efficacy increased specifically in WM-related neurons. On the contrary, in the discussion, we suggest that the increased input to what are preferentially visuomotor neurons (Figure 2) might “facilitate the transformation of visual inputs into motor commands.” This is an important point, but it hardly detracts from the fact that WM deployment (defined behaviorally) clearly alters the efficacy of visual inputs.

5. Did the two example neurons in Figure 4c show adjusted values (subtracting the same measure during non-stimulated trials)? The authors mentioned in Method that Figure 4 showed adjusted values, but it may not be applicable for raster plot in Figure 4c. It may be helpful that using adjusted values show stimulation effects on evoked spike counts during memory In and Out trials.

Note that the raster plots in 3c (formerly 4c) are necessarily unaltered data, as the adjustment procedure does not operate on individual spikes, and the calculation of joint spiking takes into account the relative timing of those individual spikes. We now clarify this in the methods (page 18).

6. Did the authors find some FEF cells showing elevated firing during delay period in outside-RF trials compared with baseline firing? These elevated firings would not be caused by the RF cue and thus could more readily be interpreted as a working memory signal.

This is a good question. Indeed, it has previously been shown that some FEF neurons exhibit visual, motor and delay properties that are not aligned in space (Bruce and Goldberg, 1985; Lawrence et al., 2005). Consistent with this, we found that some neurons exhibited more delay activity when the cue was presented outside of their visual receptive field (values below 0.5 in the plot shown in new Figure 3 Supplement 2)—although the magnitude of their delay selectivity is lower on average than those with elevated firing rates for the IN condition. Importantly, the enhanced efficacy of orthodromic activation occurs in these neurons as well (Δ response magnitude = 0.08±0.01, p<10^-4, n=26); overall there was no correlation between FEF neurons’ delay selectivity and their change in input efficacy (r = 0.12, p = 0.235, Pearson correlation).

However, the suggestion that elevated activity in the OUT condition is more easily interpreted as reflecting working memory is not correct, in our opinion. One could still argue that such elevated activity is cue induced, e.g. resulting from a release of inhibition. The question of whether delay activity in the FEF, which has been studied and described for decades, is a result of the visual cue vs. memory has been addressed in many ways in previous studies, and we don’t attempt to litigate that issue in the current paper. For example, delay activity often occurs in neurons with no visual response to the cue (e.g. see Figure 4A. in Armstrong et al., 2009), and of course the visual responses of FEF neurons do not persist for the 1 second duration of the delay period (e.g. Sommer and Wurtz, 2001). (We have revised the paper to clarify the duration of the delay period which may have been ambiguous in the original version.)

Importantly, as stated above, the interpretation that the increased efficacy during the delay period results from differences in WM is based solely on the fact that we are comparing efficacy between memory locations during the delay period (e.g., the time during which the monkey must remember the cued location), not because individual FEF neurons display delay activity (or not). Again, the difference in efficacy is independent of the amount of delay activity (Figure 3 Supplement 2). Indeed, the interpretation doesn’t even depend on any presumed role of FEF delay activity in the control of WM.

7. The sample size should be indicated in the Figure 3b Venn diagram.

We have added the sample sizes for non-activated and visual-recipient FEF populations to the Venn diagrams in this figure.

8. It would be useful to indicate the electrical stimulus protocol in Figure 1.

We have moved the stimulation technique, previously shown in Figure 2, into Figure 1, and further details of stimulation timing are now illustrated in Figure 1 supplement 1.

Reviewer #2:

This is a very interesting study, examining the properties of different types of neurons in the primate Frontal Eye Fields. It is commonly assumed that a serial processing of information takes place in the frontal lobe, from visual representation, to working memory maintenance, to motor output. However, some evidence to the contrary has also been reported, creating a debate in the field. The authors have characterized meticulously FEF neurons receiving V4 projections, by means of orthodromic stimulation. They report two main findings: that visual-input recipient neurons in FEF exhibit substantial motor activity and that working memory alters the efficacy of V4 input to FEF. The paper provides an important addition to our understanding of FEF processing. Although the first result is unambiguous, and goes against the traditional view of the FEF, the interpretation of the second is less straightforward and would need to be qualified further.

We hope that the revised manuscript provides a clearer picture regarding the significance and interpretation of the second part of the paper. Clarifications and qualifying statements have been added based on reviewer feedback (see responses to points 1-3 below).

1. Orthodromic activation of FEF neurons via V4 stimulation increases the percentage of FEF events that lead to spikes and decreases their latency during working memory. Such an effect appears expectable if FEF neurons are at a higher level of when a stimulus in their receptive field is held in memory compared to a stimulus out of their receptive field. Are the authors suggesting something special about working memory? Would the same outcome not be expected during fixation or smooth pursuit for FEF neurons that are activated by these states? It was not clear that efficacy of transmission itself improves by working memory, just the likelihood that the spiking threshold would be reached.

See response to essential revision #1 above.

2. It would strengthen the authors thesis to discuss the existing functional evidence (in addition to anatomical evidence) that motor FEF neurons receive visual input and can plan movements accordingly. See for example Costello et al. J. Neurosci 2013, 33(41):16394-408.

See response to essential revision #2 above.

3. The authors match the receptive location of FEF and V4 neurons to maximize the chances of identifying monosynaptically connected neurons between the two areas. However, a negative finding of no effects of orthodromic activation does not entirely rule out that the FEF neuron under study receive V4 input, from another site. Some discussion is warranted on this point.

See response to essential revision #3 above.

References:

Armstrong, K. M., Chang, M. H., and Moore, T. (2009). Selection and maintenance of spatial information by frontal eye field neurons. The Journal of Neuroscience, 29(50), 15621–15629. https://doi.org/10.1523/JNEUROSCI.4465-09.2009

Briggs, F., Mangun, G. R., and Usrey, W. M. (2013). Attention enhances synaptic efficacy and the signal-to-noise ratio in neural circuits. Nature, 499(7459), 476–480. https://doi.org/10.1038/nature12276

Bruce, C. J., and Goldberg, M. E. (1985). Primate frontal eye fields. I. Single neurons discharging before saccades. Journal of Neurophysiology, 53(3), 603–635.

Costello, M. G., Zhu, D., Salinas, E., and Stanford, T. R. (2013). Perceptual modulation of motor – But not visual – Responses in the frontal eye field during an urgent-decision task. The Journal of Neuroscience: The Official Journal of the Society for Neuroscience, 33(41), 16394–16408. https://doi.org/10.1523/JNEUROSCI.1899-13.2013

Lawrence, B. M., White, R. L., and Snyder, L. H. (2005). Delay-period activity in visual, visuomovement, and movement neurons in the frontal eye field. Journal of Neurophysiology, 94(2), 1498–1508. https://doi.org/10.1152/jn.00214.2005

Schall, J. D. (1991). Neuronal activity related to visually guided saccades in the frontal eye fields of rhesus monkeys: Comparison with supplementary eye fields. Journal of Neurophysiology, 66(2), 559–579. https://doi.org/10.1152/jn.1991.66.2.559

Schall, J., Morel, A., King, D. J., and Bullier, J. (1995). Topography of visual cortex connections with frontal eye field in macaque: Convergence and segregation of processing streams. The Journal of Neuroscience, 15(6), 4464–4487.

Sommer, M. A., and Wurtz, R. H. (2001). Frontal eye field sends delay activity related to movement, memory, and vision to the superior colliculus. Journal of Neurophysiology, 85(4), 1673–1685.

Ungerleider, L. G., Galkin, T. W., Desimone, R., and Gattass, R. (2008). Cortical connections of area V4 in the macaque. Cerebral Cortex (New York, N.Y.: 1991), 18(3), 477–499. https://doi.org/10.1093/cercor/bhm061

Author details

1. Behrad Noudoost

   Department of Ophthalmology and Visual Sciences, University of Utah, Salt Lake City, United States

   Contribution

   Conceptualization, Data curation, Formal analysis, Validation, Visualization, Methodology, Writing - original draft, Writing - review and editing

   For correspondence

   behrad.noudoost@gmail.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1588-027X

2. Kelsey Lynne Clark

   Department of Ophthalmology and Visual Sciences, University of Utah, Salt Lake City, United States

   Contribution

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

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-1791-3960

3. Tirin Moore

   Department of Neurobiology, and Howard Hughes Medical Institute, Stanford University, Stanford, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Validation, Visualization, Writing - original draft, Project administration, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-3345-2930

• 1,258

  Page views

• 204

  Downloads

• 3

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.