Humans, and presumably monkeys, effortlessly establish a gist-like memory of their environment whenever they enter a new place. They can then use this memory to guide action even in the absence of vision. The hallmark of this environmental memory is that the objects remembered need not be relevant to the subject’s current behavior. For example, although you may never be asked to point to the door with your eyes closed, you establish a memory of the door’s location automatically. Although normally humans have no trouble pointing to objects in the room with their eyes closed, patients with bilateral parietal lesions cannot do this even though they can easily locate and point to objects in the room with their eyes open (Levine et al., 1985), suggesting an environmental memory impairment.

Visually responsive neurons in the frontal eye field (FEF) exhibit a signal that could represent environmental memory (Umeno and Goldberg, 2001). After monkeys make a number of saccades that bring a task-irrelevant probe stimulus into the receptive field of a visually responsive FEF neuron, many neurons respond on trials when a saccade brings the spatial location of the stimulus into the receptive field, even though the probe stimulus did not appear on the current trial. However, because the investigators always used the same saccade both to establish and evaluate the memory response, it is not clear if the effect is a true spatial memory or merely a memory of receptive field stimulation associated with a saccade.

Here, we asked if neurons in the lateral intraparietal area (LIP), an area with visual, oculomotor, and mnemonic connections that serves as a priority map of the environment (Bisley and Goldberg, 2010), also exhibits an environmental memory signal. Indeed, we found that neurons in LIP did convey an environmental memory signal. Further, we found that the memory signal could be established even when the probe stimulus never appeared in the receptive field of the neuron, and occurred despite the fact that the monkey made a different saccade to bring the spatial location of the vanished probe stimulus into the receptive field. These results suggest that LIP has access to a representation of the visual world in at least supraretinal coordinates. Because saccades and reaching movements are coded in the parietal cortex in retinotopic coordinates (Duhamel et al., 1992; Andersen et al., 1998), these results suggest that one role of LIP is to transform a supraretinal representation of the remembered environment into a retinotopic representation more useful for the generation of action.

In this set of experiments, we demonstrate that LIP neurons exhibit an environmental memory signal, which lasts over a duration of multiple trials, decays slowly, and updates with eye movements. In addition, we demonstrate that a neuron does not require visual stimulation or a particular training saccade to establish memory activity. There is a similar signal in the frontal eye field (Umeno and Goldberg, 2001), in visual and visuomovement cells, but not movement cells. Those preliminary experiments, however, had two confounds. First, the memory was always established by stimulating the receptive field. Second, the saccade that evoked the memory was identical to the saccade used to establish it, so the memory could have been related to the saccadic process. In our experiments, we show that memory can be evoked by a stimulus that never appeared in the neuron’s receptive field, and that the memory response manifested even when the saccade used to evoke it differed from the saccade used to establish it. Furthermore, as in the FEF (Umeno and Goldberg, 2001), memory activity occurred both in neurons which did exhibit saccadic remapping as well as those that did not. In sum, these results suggest that LIP has access to a supraretinal representation of the visual world despite the fact that all of its explicit responses are retinotopic, and that this memory response cannot be simply ascribed to saccadic remapping or saccade planning.

Our results can explain previous clinical observations regarding the role of parietal cortex in spatial memory. Patients with bilateral parietal lesions cannot point to an object in their room with their eyes closed, although they can easily do it with their eyes open (Levine et al., 1985). This suggests that these patients do not have access to a remembered spatial representation of their environment. More importantly, patients with right parietal lesions report their spatial memory in a retinotopic frame. Bisiach and Luzzatti (1978) asked two Milanese patients with right parietal damage to describe their memory of the Piazza del Duomo in Milan. They found that these patients could only describe the recalled environmental landmarks that would fall in their unaffected (right) hemifield, which depended on their (imagined) vantage point. For example, when they described Piazza del Duomo as if they were standing with their back to the cathedral they remembered only the landmarks to the right of the cathedral. However, when they described the Piazza del Duomo as if they were facing the cathedral they only remembered the landmarks on the other side of the square, which were now to the right of their imagined vantage point. This experiment demonstrates two key ideas; (1) The brain stores long-term memory of environments in supraretinal coordinates in an area unaffected by a parietal lesion, since the patients could, under certain circumstances, remember the entire Piazza del Duomo. (2) Retrieving that memory requires the parietal cortex, but the right parietal patients could only report (and therefore mentally represent) objects on the right side of their imagined vantage point. They had a mnemonic left hemianopia for a remembered environment. If their memory were reportable in a supraretinal frame they should have been able to report the entire scene. Because they could not do so, and could only report objects in the non-neglected hemifield, the report of their environmental memory must be processed by the retinotopic map of the parietal cortex. Our results are consistent with this result. The memory of an environment can be stored a supraretinal frame, but must be reported by neurons with retinotopic receptive fields. The neurons that we have reported here fit that description. They have access to a supraretinal representation of the visual field, because environmental memory does not require stimulating their receptive fields, but they only report that memory when the spatial location of the remembered stimuli lies in their receptive field.

The memory response has both a longer latency and a lesser intensity than a simple reafferent visual response. This suggests that it does not arise from a feedforward mechanism, just as task-related pattern selectivity in V4 requires a late, presumably feedback, mechanism (Ipata et al., 2012). There are two known mechanisms by which LIP can create a spatially accurate representation of space despite a constantly moving eye. The first is through remapping, in which a neuron will respond immediately after a saccade brings the spatial location of a recently vanished stimulus into its receptive field (Duhamel et al., 1991; Sun and Goldberg, 2016). This effectively changes the spatial origin of the retinotopic coordinate system from the presaccadic fixation point to the saccade target. The next saccade could just as easily remap the current retinotopic representation into a coordinate system centered on the next saccade target, and so on, through a number of saccades. This is unlikely to be the basis of our results, however, because many cells that show environmental memory do not show pre-saccadic remapping when a saccade brings a stimulus into its receptive field.

The second mechanism creates a spatially accurate retinotopic representation from a supraretinal representation. There is excellent human psychophysical evidence that the saccadic system has access to a supraretinal representation of space that does not involve remapping. Karn et al. (1997) showed that there is little difference in the inaccuracy of memory-guided saccades after two or six intervening visually-guided saccades. Reliance on a remapping mechanism for accuracy would entail the concatenation of errors from saccade to saccade, whereas access to a supraretinal coordinate frame would enable a more stable representation of space across multiple saccades. In a similar experiment, Poletti et al. (2013) asked humans to make different numbers of intervening saccades between the presentation of a target for a memory-guided saccade and executing the actual saccade. For the first few intervening saccades, the variability of the memory-guided saccade increased linearly, as expected if the error of each remapping process would concatenate. However, after the first few saccades, the increase in variability decreased with each subsequent saccade, as if the representation of the saccade target had shifted to a supraretinal frame which would not increase its error with each saccade. To explain this phenomenon, they proposed a model that entails a shift from an early remapping mechanisms to a subsequent supraretinal mechanism. This late supraretinal representation could arise from a gain-field mechanism. Visual responses in LIP are linearly modulated by the position of the eye in the orbit, the gain field. The eye position signal that modulates the gain fields in LIP could come from the representation of eye position in area 3a (Zhang et al., 2008). It is mathematically trivial to calculate target position in supraretinal coordinates from retinotopic gain fields (Zipser and Andersen, 1988; Salinas and Abbott, 1997; Pouget and Sejnowski, 1994). Although gain fields are inaccurate immediately after a saccade (Xu et al., 2012), they could contribute to a late, accurate, supraretinal representation that develops when the eye position signal is accurate, as one would expect if the eye position signal arose from proprioception. The neurons that we studied did not have eye position modulation of their visual responses – such neurons often have a phasic and tonic postsaccadic response that arises from the eye position signal itself (Wang et al.). Indeed, because the memory of the stimulus was established well before the saccade that evoked the memory, the supraretinal representation of the remembered stimulus could easily be implicitly encoded in the gain-field responses of LIP neurons (Zipser and Andersen, 1988).

Data from patients with parietal lesions, such as the previously described results from Bisiach and Luzzatti, suggest that although LIP might calculate target position in retinotopic coordinates, the parietal cortex does not store the results of that calculation. Instead, a supraretinal representation of space could be stored elsewhere, possibly in the medial temporal lobe. O’Keefe discovered that neurons in the rat hippocampus (the ‘place cells’) discharge when rats enter spatial location that they have seen before (O'Keefe and Dostrovsky, 1971). Other studies have duplicated this result in the monkey hippocampus (Ono et al., 1993; Rolls et al., 1989). Grid cells in entorhinal cortex are thought to be the precursors of hippocampal place cells (Moser et al., 2008) and these have been found in the monkey (Buffalo, 2015) as well as in humans (Jacobs et al., 2013). It is known that there is a direct, reciprocal connection between the parahippocampal gyrus and LIP (Suzuki and Amaral, 1994; Baizer et al., 1991). Thus, LIP could send an eye-position modulated visual signal to the medial temporal lobe, which could serve as a building block for a supraretinal memory. The medial temporal lobe could then use this activity to generate the supraretinal component of place cell activity and send this supraretinal signal back to LIP. LIP itself would only be activated when the spatial location of the vanished stimulus can be described by a retinotopic vector from the center of gaze to the receptive field of the neuron. Thus, LIP has access to a supraretinal representation, but expresses only a retinotopic representation. It is important to emphasize that LIP is part of a spatial network that includes the FEF and the superior colliculus. Although neuropsychological studies suggest that parietal damage affects spatial memory, we do not know if damage to other parts of the network has a similar effect.

Although early models posited that the oculomotor system uses a supraretinal representation (Zee et al., 1976), it is clear that the parietal cortex sends a spatially accurate retinotopic signal to the oculomotor (Pare and Wurtz, 1997) and skeletomotor (Batista et al., 1999) systems. Even auditory stimuli, which are initially coded in craniotopic coordinates, are converted to retinotopic coordinates for the oculomotor (Jay and Sparks, 1987b; Jay and Sparks, 1987a) and skeletal motor system (Grunewald et al., 1999; Cohen and Andersen, 2000) in parietal cortex, and the oculomotor system of the superior colliculus (Jay and Sparks, 1984). We suggest that the environmental memory signal in LIP is an example of the conversion of a supraretinal signal to a retinotopic signal for memory and spatial attention as well as for motor control.

The Animal Care and Use Committees at Columbia University and the New York State Psychiatric Institute approved all of the animal protocols in this study as complying with the guidelines established in the United States Public Health Service Guide for the Care and Use of Laboratory Animals (protocol NYSPI-1225-C). We used one female and two male rhesus monkeys (Macaca mulatta) with weights between 6 and 13 kg.

We first trained monkeys to sit in a primate chair using a pole and collar technique. After chair training, we surgically fitted the monkeys with acrylic implants, anchored with titanium or ceramic screws, outfitted with a headpost which stabilized the monkeys head while sitting in a primate chair. During the same surgery, we implanted scleral search coils around both eyes (Judge et al., 1980) and brought the coil wire out to a plug on the acrylic. After basic behavioral training, we performed a craniotomy to expose the intraparietal sulcus (using coordinates determined from a structural MRI), over which we affixed a recording chamber in the acrylic.

We located the actual recording sites in each monkey by performing MRIs with the recording electrode in place (see Figure 1). We tranquilized the monkey in the vivarium with ketamine and atropine for transport to the MRI lab. We then anesthetized the monkey using endotracheal isoflurane, and positioned it in a Kopf MRI-compatible stereotaxic instrument. To create the images, we used a custom-made rigid and flexible RF array coils using GE MR750(3T) and GE Signa(1.5 T) scanners. We used custom Matlab software, Osirix, GE’s MediaViewer, and DicomWorks to analyze the data.

We performed all surgeries using ketamine/isoflurane general anesthesia using aseptic surgical technique. The monkey recovered fully before testing restarted. During testing, monkeys worked for their daily water intake and were supplemented with dried and fresh fruits. We monitored monkeys’ weights and general health on every recording day, and at least once a week.

We controlled all experiments using the REX (downloadable from: https://nei.nih.gov/intramural/software) system (Hays et al., 1982) and recorded single-unit activity with glass-insulated tungsten electrodes introduced through a guide tube positioned plastic grid with 1 mm spacing between possible penetrations (Crist et al., 1988). In general, recording sessions lasted between 5 and 8 hr depending on the stability of the cell and the monkey’s willingness to continue to work. We took care to avoid making penetrations with the electrode in adjacent grid holes from the previous day’s penetration in order to reduce tissue damage. We kept careful logs regarding the location and depth of each penetration, an estimate describing cell types encountered, and the quality of the monkey’s behavior. During every testing and training session, we monitored monkeys using a closed-circuit camera and monitor.

We verified stimulus timing using a photoprobe affixed to the screen, which emitted a pulse for each event. We monitored the monkey’s eye movements using a CNC phase detector (Crist Instruments, Hagerstown, MD) to decode the search coil signal (Judge et al., 1980) and sampled the signal at 1 kHz. During testing, monkeys sat in a primate chair in a sound-attenuated Faraday room. We performed these experiments in two experimental setups. In one set up, the monkeys sat 57 cm away from a CRT monitor (ViewSonic Professional Series P225F) with a refresh rate of 120 Hz. In the second setup, we presented stimuli with a Hitachi CPX275 LCD projector with a refresh rate of 60 Hz, on a screen 72 cm from the monkey.

Once we isolated an LIP neuron and confirmed that it displayed canonical visual, delay, and perisaccadic responses, we probed the boundaries of LIP neuron receptive fields by having monkeys make memory-guided saccades (Hikosaka and Wurtz, 1983) to targets appearing in locations which were inside and outside of the cell’s receptive field. In some cases, we used a fixation task to characterize the cell’s receptive field by having the monkey fixate a point while stimuli briefly flashed (50 ms) pseudo-randomly on the screen, generating a map of locations that evoked visual responses (measured 50–150 ms after the flash). We ensured that saccade target and probe stimuli locations were positioned correctly with respect to future and current receptive field locations to avoid unintentional contamination by a visual response.

We transformed the REX data into a form analyzable by MATLAB (MathWorks, Natick, MA) using the REXTOOLS software (downloadable from https://nei.nih.gov/intramural/software). Although the REX system provides an estimate of activity using the MEX (downloadable from https://nei.nih.gov/intramural/software) spike sorter, for the analyses in this paper we used the MEX system to digitize the neural data and sorted the spikes offline using the MEX offline analysis program. The custom codes, and raw data from every cell analyzed in this paper are provided as. m, and. csv files.

Before performing detailed statistical analysis, we first investigated if the data came from standard normal distributions, using two-sample Kolmogorov-Smirnov (KS) test. When we tested the mean neural activity in the response windows (intervals described below), all of our data sets were found to be distributed non-normally (non-parametric).

To show that the statistically significant cells in both experiments were not just one side of a symmetric distribution we calculated a memory index to quantify the degree to which each cell manifested memory activity, comparing postsaccadic activity in Block 1 (R_pre) with postsaccadic activity in Block 3 (R_mem):

All index distributions were not normally distributed (by KS test), and we showed that the population median was significantly different from zero by using median values for the distribution of the memory indices and a Wilcoxon signed rank test to compare each sample with a test sample identical to the measured sample with the measured mean subtracted from each value. In addition, we calculated medcouple, a non-parametric histogram skewness measure (Brys et al., 2004). This tool is not affected by unsymmetrical tails and outliers, and is also not based on classical skewness represented by the third moment. Since this measure does not depend on mean or standard deviation, it gives an accurate description of a skewed distribution. Its results are bound between values: −1 (left skewed), 0 (symmetric), and 1 (right skewed).

All cells reported here had both statistically significant visual and memory responses (Wilcoxon signed rank p-value for median activity between baseline and response window <0.001) (Whitley and Ball, 2002). We have excluded units with memory activity if median activity in Block 1 (baseline recording block) pre- and post-saccadic were statistically different (indicating eye-position modulation). In other words, even if we could find significant memory activity well beyond what can be described by either eye position or saccadic activity effect, we are not presenting them here.

To quantify the magnitude of the memory and visual responses, we did a comparison of the peristimulus time histograms (PSTHs) during baseline, visual, and memory trials, for identical epochs aligned on saccade onset. All PSTHs were normalized to baseline activity (block 1) and were constructed from 2 ms spike bins, smoothed using a 10 ms sliding causal filter. Baseline calculations were made in the −426 to −226 ms (pre-saccade onset) window. To quantify the decay, perisaccadic memory and visual responses, we selected the time point of peak activity in the window from 125 ms before the saccade onset (to include any predictive remapping) to 500 ms after. We then looked at the median of raw spike count over a window extending 100 ms before and after this point to calculate response latencies, as described next.

To obtain a latency value of the memory response, we used a median threshold-crossing and sustained response method. In short, this analysis compared the median activity in the baseline and response windows. We derived a ‘cut off’ value by determining the spike count values which would fall above at least the 75% percentile of the baseline median. By comparing the median spike counts in the response window to this ‘cut off’ value, we could determine when a minimum of five bins (of 2 ms each) contained a greater spike count than the ‘cut off’ value. We took the first of these bins as the response latency, and verified visually by looking at the corresponding time on the PSTH. Since the baseline activity can vary between different trials in the same experiment, we first chose a window within the time from fixation to −226 ms before saccade onset. Within this wide window, we made 200 ms wide sliding windows, in 1 ms steps. By finding the median spike count within these smaller windows, we picked the window closest to the start of the response window (but still had the same median as the population of these several 200 ms wide windows). This median was chosen to look at the response window in which a threshold crossing for at least 20 ms was picked as the response latency.

We could not calculate latencies for a few neurons using this method, because the baseline (pre-saccadic) memory block response was as high as post-saccadic activity. There were also a few other units whose latencies were automatically picked by the aforementioned median method but a more precise time point could be picked by doing yet another median-based analysis. For these units, we implemented Wilcoxon rank sum comparison with the baseline median and a sliding 50 ms window, stepped 1 ms, in the response epoch. The first such 20 ms window, where there was a statistical difference was observed was picked as the latency point.

We observed that the memory response continued to manifest for almost up to 100 trials after the last probe stimulus trial, and so we analyzed the ‘forgetting’ or ‘decay trials’. While individual cell results were noisy and there were differences between cells, we normalized each cell’s decay activity to its peak memory activity and fit the population data to a first order exponential (shown in figures) and quantified it using comparisons of the first and last 10-trials in a 50-trial epoch. We fit the population data similarly.

1. 1. Andersen RA
   2. Snyder LH
   3. Batista AP
   4. Buneo CA
   5. Cohen YE

   (1998)

   Posterior parietal areas specialized for eye movements (LIP) and reach (PRR) using a common coordinate frame

   Novartis Foundation Symposium 218:109–122.

   • PubMed
   • Google Scholar
2. 1. Baizer JS
   2. Ungerleider LG
   3. Desimone R

   (1991)

   Organization of visual inputs to the inferior temporal and posterior parietal cortex in macaques

   Journal of Neuroscience 11:168–190.

   • PubMed
   • Google Scholar
3. 1. Batista AP
   2. Buneo CA
   3. Snyder LH
   4. Andersen RA

   (1999) Reach plans in eye-centered coordinates

   Science 285:257–260.

   https://doi.org/10.1126/science.285.5425.257

   • PubMed
   • Google Scholar
4. 1. Bisiach E
   2. Luzzatti C

   (1978) Unilateral neglect of representational space

   Cortex 14:129–133.

   https://doi.org/10.1016/S0010-9452(78)80016-1

   • PubMed
   • Google Scholar
5. 1. Bisley JW
   2. Goldberg ME

   (2010) Attention, intention, and priority in the parietal lobe

   Annual Review of Neuroscience 33:1–21.

   https://doi.org/10.1146/annurev-neuro-060909-152823

   • PubMed
   • Google Scholar
6. 1. Brys G
   2. Hubert M
   3. Struyf A

   (2004) A robust measure of skewness

   Journal of Computational and Graphical Statistics 13:996–1017.

   https://doi.org/10.1198/106186004X12632

   • Google Scholar
7. 1. Buffalo EA

   (2015) Bridging the gap between spatial and mnemonic views of the hippocampal formation

   Hippocampus 25:713–718.

   https://doi.org/10.1002/hipo.22444

   • PubMed
   • Google Scholar
8. 1. Cohen YE
   2. Andersen RA

   (2000) Reaches to sounds encoded in an eye-centered reference frame

   Neuron 27:647–652.

   https://doi.org/10.1016/S0896-6273(00)00073-8

   • PubMed
   • Google Scholar
9. 1. Crist CF
   2. Yamasaki DS
   3. Komatsu H
   4. Wurtz RH

   (1988) A grid system and a microsyringe for single cell recording

   Journal of Neuroscience Methods 26:117–122.

   https://doi.org/10.1016/0165-0270(88)90160-4

   • PubMed
   • Google Scholar
10. 1. Duhamel J-R
    2. Colby CL
    3. Goldberg ME

    (1991)

    Neurons in the lateral intraparietal area of the monkey remap visual space in conjunction with saccadic eye movements. I. Presaccadic events

    Soc. Neurosci. Abs 17:1282.

    • Google Scholar
11. 1. Duhamel JR
    2. Colby CL
    3. Goldberg ME

    (1992) The updating of the representation of visual space in parietal cortex by intended eye movements

    Science 255:90–92.

    https://doi.org/10.1126/science.1553535

    • PubMed
    • Google Scholar
12. 1. Grunewald A
    2. Linden JF
    3. Andersen RA

    (1999) Responses to auditory stimuli in macaque lateral intraparietal area. I. Effects of training

    Journal of Neurophysiology 82:330–342.

    https://doi.org/10.1152/jn.1999.82.1.330

    • PubMed
    • Google Scholar
13. Conference

    1. Hays AV
    2. Richmond BJ
    3. Optican LM

    (1982)

    A UNIX- based multiple process system for real-time data acquisition and control

    Conf Proc.

    • Google Scholar
14. 1. Hikosaka O
    2. Wurtz RH

    (1983) Visual and oculomotor functions of monkey substantia nigra pars reticulata. III. Memory-contingent visual and saccade responses

    Journal of Neurophysiology 49:1268–1284.

    https://doi.org/10.1152/jn.1983.49.5.1268

    • PubMed
    • Google Scholar
15. 1. Ipata AE
    2. Gee AL
    3. Goldberg ME

    (2012) Feature attention evokes task-specific pattern selectivity in V4 neurons

    Proceedings of the National Academy of Sciences 109:16778–16785.

    https://doi.org/10.1073/pnas.1215402109

    • PubMed
    • Google Scholar
16. 1. Jacobs J
    2. Weidemann CT
    3. Miller JF
    4. Solway A
    5. Burke JF
    6. Wei XX
    7. Suthana N
    8. Sperling MR
    9. Sharan AD
    10. Fried I
    11. Kahana MJ

    (2013) Direct recordings of grid-like neuronal activity in human spatial navigation

    Nature Neuroscience 16:1188–1190.

    https://doi.org/10.1038/nn.3466

    • PubMed
    • Google Scholar
17. 1. Jay MF
    2. Sparks DL

    (1984) Auditory receptive fields in primate superior colliculus shift with changes in eye position

    Nature 309:345–347.

    https://doi.org/10.1038/309345a0

    • PubMed
    • Google Scholar
18. 1. Jay MF
    2. Sparks DL

    (1987a) Sensorimotor integration in the primate superior colliculus. I. Motor convergence

    Journal of Neurophysiology 57:22–34.

    https://doi.org/10.1152/jn.1987.57.1.22

    • PubMed
    • Google Scholar
19. 1. Jay MF
    2. Sparks DL

    (1987b) Sensorimotor integration in the primate superior colliculus. II. Coordinates of auditory signals

    Journal of Neurophysiology 57:35–55.

    https://doi.org/10.1152/jn.1987.57.1.35

    • PubMed
    • Google Scholar
20. 1. Judge SJ
    2. Richmond BJ
    3. Chu FC

    (1980) Implantation of magnetic search coils for measurement of eye position: an improved method

    Vision Research 20:535–538.

    https://doi.org/10.1016/0042-6989(80)90128-5

    • PubMed
    • Google Scholar
21. 1. Karn KS
    2. Møller P
    3. Hayhoe MM

    (1997) Reference frames in saccadic targeting

    Experimental Brain Research 115:267–282.

    https://doi.org/10.1007/PL00005696

    • PubMed
    • Google Scholar
22. 1. Levine DN
    2. Warach J
    3. Farah M

    (1985) Two visual systems in mental imagery: dissociation of "what" and "where" in imagery disorders due to bilateral posterior cerebral lesions

    Neurology 35:1010.

    https://doi.org/10.1212/WNL.35.7.1010

    • PubMed
    • Google Scholar
23. 1. Moser EI
    2. Kropff E
    3. Moser MB

    (2008) Place cells, grid cells, and the brain's spatial representation system

    Annual Review of Neuroscience 31:69–89.

    https://doi.org/10.1146/annurev.neuro.31.061307.090723

    • PubMed
    • Google Scholar
24. 1. O'Keefe J
    2. Dostrovsky J

    (1971) The hippocampus as a spatial map. Preliminary evidence from unit activity in the freely-moving rat

    Brain Research 34:171–175.

    https://doi.org/10.1016/0006-8993(71)90358-1

    • PubMed
    • Google Scholar
25. 1. Ono T
    2. Nakamura K
    3. Nishijo H
    4. Eifuku S

    (1993) Monkey hippocampal neurons related to spatial and nonspatial functions

    Journal of Neurophysiology 70:1516–1529.

    https://doi.org/10.1152/jn.1993.70.4.1516

    • PubMed
    • Google Scholar
26. 1. Pare M
    2. Wurtz RH

    (1997) Monkey posterior parietal cortex neurons antidromically activated from superior colliculus Modulation of neuronal activity by target uncertainty Shared motor error for multiple eye movements

    Journal of Neurophysiology 78:3493–3497.

    https://doi.org/10.1152/jn.1997.78.6.3493

    • Google Scholar
27. Book

    1. Paxinos G
    2. Huang X-F
    3. Toga AW

    (1999)

    The Rhesus Monkey Brain in Stereotaxic Coordinates

    New York: Academic Press.

    • Google Scholar
28. 1. Poletti M
    2. Burr DC
    3. Rucci M

    (2013) Optimal multimodal integration in spatial localization

    Journal of Neuroscience 33:14259–14268.

    https://doi.org/10.1523/JNEUROSCI.0523-13.2013

    • PubMed
    • Google Scholar
29. 1. Pouget A
    2. Sejnowski TJ

    (1994) A neural model of the cortical representation of egocentric distance

    Cerebral Cortex 4:314–329.

    https://doi.org/10.1093/cercor/4.3.314

    • PubMed
    • Google Scholar
30. 1. Rolls ET
    2. Miyashita Y
    3. Cahusac PM
    4. Kesner RP
    5. Niki H
    6. Feigenbaum JD
    7. Bach L

    (1989)

    Hippocampal neurons in the monkey with activity related to the place in which a stimulus is shown

    Journal of Neuroscience 9:1835–1845.

    • PubMed
    • Google Scholar
31. Book

    1. Saleem KS
    2. Logothetis NK

    (2012)

    A Combined MRI and Histology Atlas of the Rhesus Monkey Brain in Stereotaxic Coordinates (2nd Edition)

    San Diego: Academic Press.

    • Google Scholar
32. 1. Salinas E
    2. Abbott LF

    (1997) Invariant visual responses from attentional gain fields

    Journal of Neurophysiology 77:3267–3272.

    https://doi.org/10.1152/jn.1997.77.6.3267

    • PubMed
    • Google Scholar
33. 1. Sun LD
    2. Goldberg ME

    (2016) Corollary discharge and oculomotor proprioception: cortical mechanisms for spatially accurate vision

    Annual Review of Vision Science 2:61–84.

    https://doi.org/10.1146/annurev-vision-082114-035407

    • PubMed
    • Google Scholar
34. 1. Suzuki WA
    2. Amaral DG

    (1994) Perirhinal and parahippocampal cortices of the macaque monkey: cortical afferents

    The Journal of Comparative Neurology 350:497–533.

    https://doi.org/10.1002/cne.903500402

    • PubMed
    • Google Scholar
35. 1. Umeno MM
    2. Goldberg ME

    (2001) Spatial processing in the monkey frontal eye field. II. Memory responses

    Journal of Neurophysiology 86:2344–2352.

    https://doi.org/10.1152/jn.2001.86.5.2344

    • PubMed
    • Google Scholar
36. 1. Whitley E
    2. Ball J

    (2002) Statistics review 6: Nonparametric methods

    Critical Care 6:509–513.

    https://doi.org/10.1186/cc1820

    • PubMed
    • Google Scholar
37. 1. Xu BY
    2. Karachi C
    3. Goldberg ME

    (2012) The postsaccadic unreliability of gain fields renders it unlikely that the motor system can use them to calculate target position in space

    Neuron 76:1201–1209.

    https://doi.org/10.1016/j.neuron.2012.10.034

    • PubMed
    • Google Scholar
38. 1. Zee DS
    2. Optican LM
    3. Cook JD
    4. Robinson DA
    5. Engel WK

    (1976) Slow saccades in spinocerebellar degeneration

    Archives of Neurology 33:243–251.

    https://doi.org/10.1001/archneur.1976.00500040027004

    • PubMed
    • Google Scholar
39. 1. Zhang M
    2. Wang X
    3. Goldberg ME

    (2008) Monkey primary somatosensory cortex has a proprioceptive representation of eye position

    Progress in Brain Research 171:37–45.

    https://doi.org/10.1016/S0079-6123(08)00606-7

    • PubMed
    • Google Scholar
40. 1. Zipser D
    2. Andersen RA

    (1988) A back-propagation programmed network that simulates response properties of a subset of posterior parietal neurons

    Nature 331:679–684.

    https://doi.org/10.1038/331679a0

    • PubMed
    • Google Scholar

Author details

1. Mulugeta Semework

   1. Mahoney-Keck Center for Brain and Behavior Research, Department of Neuroscience, Columbia University College of Physicians and Surgeons, New York, United States
   2. Department of Neuroscience, Columbia University College of Physicians and Surgeons, New York, United States
   3. Division of Neurobiology and Behavior, New York State Psychiatric Institute, New York, United States

   Contribution

   Conceptualization, Resources, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   For correspondence

   mulugetas@gmail.com

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-6070-0119

2. Sara C Steenrod

   1. Mahoney-Keck Center for Brain and Behavior Research, Department of Neuroscience, Columbia University College of Physicians and Surgeons, New York, United States
   2. Department of Neuroscience, Columbia University College of Physicians and Surgeons, New York, United States
   3. Division of Neurobiology and Behavior, New York State Psychiatric Institute, New York, United States

   Present address

   Institute of Clinical and Translational Sciences, Washington University School of Medicine, St Louis, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, 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:" 0000-0002-7932-7385

3. Michael E Goldberg

   1. Mahoney-Keck Center for Brain and Behavior Research, Department of Neuroscience, Columbia University College of Physicians and Surgeons, New York, United States
   2. Department of Neuroscience, Columbia University College of Physicians and Surgeons, New York, United States
   3. Division of Neurobiology and Behavior, New York State Psychiatric Institute, New York, United States
   4. Department of Neurology, Columbia University College of Physicians and Surgeons, New York, United States
   5. Department of Psychiatry, Columbia University College of Physicians and Surgeons, New York, United States
   6. Department of Ophthalmology, Columbia University College of Physicians and Surgeons, New York, United States
   7. Kavli Institute for Neuroscience, Columbia University, New York, United States

   Contribution

   Conceptualization, Resources, Data curation, Software, Formal analysis, Supervision, Funding acquisition, Validation, Investigation, Visualization, Methodology, Writing—original draft, Project administration, Writing—review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-0728-2464

• 1,604

  views

• 214

  downloads

• 7

  citations

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