DG-CA3 circuitry mediates hippocampal representations of latent information

Survival in complex environments necessitates a flexible navigation system that incorporates memory of recent behavior and associations. Yet, how the hippocampal spatial circuit represents latent information independent of sensory inputs and future goals has not been determined. To address this, we imaged the activity of large ensembles in subregion CA1 via wide-field fluorescent microscopy during a novel behavioral paradigm. Our results demonstrated that latent information is represented through reliable firing rate changes during unconstrained navigation. We then hypothesized that the representation of latent information in CA1 is mediated by pattern separation/completion processes instantiated upstream within the dentate gyrus (DG) and CA3 subregions. Indeed, CA3 ensemble recordings revealed an analogous code for latent information. Moreover, selective chemogenetic inactivation of DG-CA3 circuitry completely and reversibly abolished the CA1 representation of latent information. These results reveal a causal and specific role of DG-CA3 circuitry in the maintenance of latent information within the hippocampus.


Abstract (150 words) 26
Survival in complex environments necessitates a flexible navigation system that incorporates 27 memory of recent behavior and associations. Yet, how the hippocampal spatial circuit 28 represents latent information independent of sensory inputs and future goals has not been 29 determined. To address this, we imaged the activity of large ensembles in subregion CA1 via 30 wide-field fluorescent microscopy during a novel behavioral paradigm. Our results demonstrated 31 that latent information is represented through reliable firing rate changes during unconstrained 32 navigation. We then hypothesized that the representation of latent information in CA1 is 33 mediated by pattern separation/completion processes instantiated upstream within the dentate Hippocampal subregions CA1 and CA3 represent space through the coordinated activity 48 of place cellssparsely active cells tuned to different preferred locations that tile the navigable 49 space, collectively forming a 'hippocampal map' specific to each context 1 . Changes to sensory 50 cues or cognitive demands within a context can lead to widespread changes in the firing rates of 51 place cells 2-6 , a type of hippocampal pattern separation known as 'rate remapping' 5 . Previous 52 work has demonstrated that different sources of information and circuit mechanisms are 53 capable of driving rate remapping: sensory-driven remapping 5 is mediated in part by both lateral 54 entorhinal 7 and trisynaptic 8 circuits, while goal-oriented remapping 4 is mediated by a prefrontal-55 thalamic circuit 6 which is distinct from the lateral entorhinal and trisynaptic circuits. 56 However, a flexible navigational system should encode aspects of the current context 57 beyond immediate sensory input and future goals. Indeed, latent information, such as a memory 58 of recent behavior or experiences independent of future goals, can be especially important for 59 discovering and representing relationships which extend beyond the capacity of immediate 60 sensory information. Yet, whether and how the hippocampus represents latent information, as 61 well as the neural circuitry that maintains these representations in the absence of continuous 62 sensory information, has not been determined. 63 PVd] separately at each location in the compartment (Fig. 3b). Reliable differences in map 124 similarity were observed across the compartment in both the CA1 and CA3. The magnitude of 125 remapping was not correlated with differences in median speed across locations (Pearson's 126 correlation, CA1: r = 0.127, p = 0.139; CA3: r = 0.046, p = 0.595). Next, we again computed 127 [PVs -PVd] over the entire compartment, while excluding data from progressively longer times 128 since entering the compartment. This analysis revealed that reliable remapping was observed 129 when including only data recorded at least 5 s since entering the compartment in both CA1 and 130 CA3 (Fig. 3c), well beyond the timescales of both neural activity and calcium reporter dynamics. 131 Together these results indicate that the rate remapping we observed in CA1 and CA3 reflects 132 specific representation of the most recent entryway which persists across both space and time. 133 If CA1 rate remapping in this paradigm is driven by trisynaptic input, then disruption of 134 the trisynaptic circuit should eliminate such remapping. If, on the other hand, remapping of the 135 CA1 and CA3 place codes is driven by common input originating outside of the hippocampus, 136 then inhibition of the trisynaptic circuit should spare this remapping. To causally adjudicate 137 between these possibilities, we repeated our experiment while recording from right CA1 and 138 simultaneously manipulating the trisynaptic circuit in a new cohort of mice. To this end, we 139 employed a Grik4-cre mouse line which expresses cre in both the dentate gyrus and CA3, but 140 not CA1 19 . These regions were transfected bilaterally with the cre-mediated inhibitory designer 141 receptor hm4di 20 (n = 3 mice; Fig. 4a and S3a-c). On interleaved trials, either clozapine-N-oxide 142 (CNO), the designer drug which activates hm4di, or saline, the vehicle control, was 143 administered intraperitonially one hour and fifteen minutes prior to recording (19 Saline 144 sessions, 19 CNO sessions). An additional control cohort not expressing hm4di was also 145 recorded with the same paradigm (n = 3 mice, 13 CNO sessions, 12 saline sessions). 146 Inhibition of the trisynaptic circuit eliminated remapping of the CA1 place code by 147 entryway as assayed by multiple measures (Fig. 4b-d  revealed that the CA1 compartment map was more similar when the mouse entered through the 149 same entryway than when he entered through a different entryway in all conditions except for 150 the CNO with hm4di condition (Fig. 4c). Furthermore, [PVs -PVd] of the CNO with hm4di 151 condition was reliably decreased relative to all other conditions and was not significantly larger 152 than zero. Importantly, PVs remained equally high in all conditions (Wilcoxon rank sum tests: Zs 153 < 1.0920, ps > 0.274), suggesting that the lack of remapping during CNO with hm4di sessions 154 was not driven by a disruption of the CA1 place code more generally (Fig. 4c). Consistent with 155 this, other coding properties such as mean and peak firing rates, spatial information content, 156 and overall split-half reliability did not reliably differ during Saline versus CNO sessions for with 157 hm4di mice ( Fig. S4a-h). At an individual cell level, mean firing rates were reliably more similar 158 when entering through same entryway than different entryways in all conditions except for the 159 CNO with hm4di condition (Fig. 4d). Remapping of the hallway place code persisted across all 160 conditions ( Fig. S4i-o). 161

Discussion 162
Characterizing the neural basis of navigation has been a major focus of research since 163 the discovery of hippocampal place cells; yet a mechanistic explanation of how hippocampal 164 spatial circuits represent information beyond sensory inputs and goal-oriented behavior has 165 been limited, likely hindered by difficulties assaying latent information isolated from a behavioral 166 task. Here we used a novel behavioral paradigm aimed to examine how the hippocampus 167 encodes latent information. We first report that latent information, in the form of the most recent 168 entryway in a multientry compartment, is robustly encoded by rate changes in the CA1 place 169 cell representation in the absence of explicit task demands. Building on proposals that CA3 170 recurrent connections are theoretically capable of maintaining information in the absence of 171 continuous input via attractor dynamics, and that the DG is particularly well-suited to 172 disambiguate similar input representations via pattern separation, we hypothesized the DG-CA3 173 circuit mediated the representation of latent information in this paradigm [21][22][23] revealed that this circuit also maintained an analogous code for latent information, indicating 175 that such a role was plausible. Finally, to causally test this hypothesis, we used a transgenic 176 mouse line and chemogenetics to reversibly inactivate the DG-CA3 circuit while recording from 177 CA1 in this paradigm. These experiments revealed that the representation of latent information 178 in CA1 is completely dependent on CA3-DG activity in this paradigm. Importantly, the spatial 179 tuning of the CA1 representation remained intact during DG-CA3 inactivation, offering support to 180 the hypothesis that the key role of the trisynaptic circuit (EC-DG-CA3-CA1) is not spatial coding 181 (similar to the results of 24 ), but rather the encoding of additional information beyond a 182 navigator's immediate spatial location. 183 At a theoretical level, our results contribute to a growing literature demonstrating that 184 hippocampal rate remapping can be driven by multiple sources, potentially mediated by distinct 185 circuit mechanisms. Previous studies have demonstrated that rate remapping can be driven by 186 sensory inputs 5 , and that such representation partially depend on both the lateral entorhinal 187 cortex 7 and NDMA receptors in the DG 8 , though on average these studies report a 20% and 188 50% reduction in rate remapping, respectively. Importantly, such paradigms are not designed to 189 rule out a latent/mnemonic contribution to rate remapping; instead, these paradigms may 190 plausibly assay contributions of both sensory inputs and latent information. In the current study 191 in which we rule out the possibility of sensory-driven remapping, inactivation of DG-CA3 led to 192 an average 86% reduction in remapping, with residual remapping failing to significantly differ 193 from chance. Therefore, it is possible that the lateral entorhinal cortex and the DG-CA3 circuit 194 mediate distinct sensory-and latent-contributions to rate remapping. Similarly, previous studies 195 demonstrating rate remapping on the basis of goal-oriented behavior have not disambiguated a 196 contribution of latent information from a contribution of planned behavior 4 , and have 197 demonstrated that remapping reflecting planned behavior is primarily driven by the Nucleus 198 Reunions, which bypasses the DG-CA3 circuit 6 . Consistent with this distinction, we observed no 199 coding of future exitway, suggesting that DG-CA3-dependent remapping did not reflect planned 200 behavior. We anticipate that a similar experimental approach will help to address several 201 lingering questions, including the relationship between latent, sensory, and goal-oriented rate 202 remapping, the specificity of trisynaptic, lateral entorhinal, and prefrontal-thalamic circuit 203 contributions to these forms of rate remapping, and the contribution of the trisynaptic circuit to 204 other forms of hippocampal remapping 5,25,26 . and velocity to single unit activity in the hippocampus of freely-moving rats. Exp. brain 247 Res. 52, 41-9 (1983 experimental design, analysis of data, as well as drafting and revising the manuscript. 280

Surgeries 377
During all surgeries, mice were anesthetized via inhalation of a combination of oxygen and 5% 378 Isoflurane before being transferred to the stereotaxic frame (David Kopf Instruments), where 379 anesthesia was maintained via inhalation of oxygen and 0.5-2.5% Isoflurane for the duration of 380 the surgery. Body temperature was maintained with a heating pad and eyes were hydrated with 381 gel (Optixcare). Carprofen (10 ml/kg) and saline (0.5 ml) were administered subcutaneously at 382 the beginning of each surgery. Preparation for recordings involved three surgeries per mouse. 383 First, at the age of six to ten weeks, each mouse was transfected with a 400 nl injection 384 of the calcium reporter GCaMP6f according to the specific viral construct and injection 385 coordinates described in Supplemental virus to 6 parts PBS) before surgical microinjection. All injections were administered via glass 396 pipettes connected to a Nanoject II (Drummond Scientific) injector at a flow rate of 23 nl/sec. 397 One week post-injection, either a 1.8mm or 0.5mm diameter gradient refractive index 398 (GRIN) lens (Go!Foton) was implanted above either dorsal CA1 or CA3 as indicated in 399 Supplemental Table 1. Implantation of the 1.8mm diameter GRIN lens required aspiration of 400 intervening cortical tissue, while no aspiration was required for implantation of the 0.5mm 401 diameter GRIN lens. Results observed using 1.8-or 0.5-mm diameter grin lenses were similar. 402 In addition to the GRIN lens, two stainless steel screws were threaded into the skull above the 403 contralateral hippocampus and prefrontal cortex to stabilize the implant. Dental cement (C&B 404 Metabond) was applied to secure the GRIN lens and anchor screws to the skull. A silicone 405 adhesive (Kwik-Sil, World Precision Instruments) was applied to protect the top surface of the 406 GRIN lens until the next surgery. 407 One to three weeks after lens implantation, an aluminum baseplate was affixed via 408 dental cement (C&B Metabond) to the skull of the mouse, which would later secure the 409 miniaturized fluorescent endoscope (miniscope) in place during recording. The 410 miniscope/baseplate was mounted to a stereotaxic arm for lowering above the implanted GRIN 411 lens until the field of view contained visible cell segments and dental cement was applied to affix 412 the baseplate to the skull. A polyoxymethylene cap with a metal nut weighing ~3 g was affixed 413 to the baseplate when the mice were not being recorded, to protect the baseplate and lens, as 414 well as to simulate the weight of the miniscope. 415 After surgery, animals were continuously monitored until they recovered. For the initial three 416 days after surgery mice were provided with a soft diet supplemented with Carprofen for pain 417 management (MediGel CPF, ~5mg/kg/day). Screening and habituation to recording in a 418 rectangular 20 x 40 cm semi-transparent plastic home cage environment began 3 to 7 days 419 following the baseplate surgery and continued for at least 2 days until the quality and reliability 420 of the calcium data were confirmed. 421

Data acquisition 422
In vivo calcium videos were recorded with a miniscope (v3; miniscope.org) containing a 423 monochrome CMOS imaging sensor (MT9V032C12STM, ON Semiconductor) connected to a 424 custom data acquisition (DAQ) box (miniscope.org) with a lightweight, flexible coaxial cable 12 . 425 The DAQ was connected to a PC with a USB 3.0 SuperSpeed cable and controlled with 426 Miniscope custom acquisition software (miniscope.org). The outgoing excitation LED was set to 427 between 3-6%, depending on the mouse to maximize signal quality with the minimum possible 428 excitation light to mitigate the risk of photobleaching. Gain was adjusted to match the dynamic 429 range of the recorded video to the fluctuations of the calcium signal for each recording to avoid 430 saturation. Behavioral video data were recorded by a webcam mounted above the environment. 431 Behavioral video recording parameters were adjusted such that only the red LED on the CMOS 432 of the miniscope was visible. The DAQ simultaneously acquired behavioral and cellular imaging 433 streams at 30 Hz as uncompressed avi files and all recorded frames were timestamped for post-434 hoc alignment. 435 All recording environments were constructed of a grey Lego base and black Lego bricks 436 (Lego, Inc) according to the dimensions specified in the main text and supplemental figures. All 437 external walls had a height of 22 cm; all internal walls had a height of 15 cm. All hallways were 5 438 cm wide; due to the width mice typically ran the length of the hallway rather than turning around 439 in the hallway. During recording, the environment was dimly lit by a nearby computer screen, 440 which served as directional cue. A white-noise generator was placed above the environment to 441 mask uncontrolled ambient sounds. Each recording session lasted 20 minutes, and only one 442 session was recorded per day to avoid photobleaching. The mouse was always placed in the 443 corner of the hallway at the start of the session and was allowed to explore the environment for 444 15 to 30 s prior to data acquisition. Following each recording the environment was cleaned with 445

disinfectant (Prevail). 446
For CA3 inactivation experiments, 5 mg/kg of clozapine-N-oxide (CNO + 0.7% DMSO) 447 was injected Intraperitoneally 1 hr and 15 min prior to recording. This timepoint was chosen 448 based on a separate experiment in which we monitored the time course of hippocampal theta 449 power reduction during hm4di/CNO-mediated inhibition of the medial septum (data not shown). 450 Mice were returned to their home cage between the injection and the start of the recording 451 session. We conducted two separate control experiments to rule out the possibilities that our 452 results could be explained by injection procedure, expression of hm4di-mcherry, or to non-453 specific effects of CNO itself. As our first within-animal control, we injected sterile saline instead 454 of CNO and repeated the recording and analysis procedures (Fig. 2b-d and Figs. S3, S4). The 455 order of Saline and CNO recordings was interleaved within mouse, and whether the first 456 recording session for a mouse followed a Saline or CNO injection was randomized. As a second 457 control, to ensure that any differences between Saline and CNO sessions were attributable to 458 the interaction between hm4di and CNO, and not an effect of CNO or its metabolites alone, 459 CNO injection experiments were repeated in a second across-mouse control group which did 460 not express hm4di (Fig. 2c-d and Figs. S3, S4). were manually inspected to ensure that motion correction was effective and did not introduce 467 additional artifacts. Following this preprocessing pipeline, the spatial footprints of all cells were 468 manually verified to remove lens artifacts. Position data were inferred from this LED offline 469 following recording using a custom written MATLAB (MathWorks) script and were manually 470 corrected if needed. The experimenter manually segmented data recorded in the compartment 471 and hallway, as well as the most recent entryway based on the recorded position data prior to 472 all further analyses. 473

Data analysis 474
All analyses were conducted using the vector of inferred likelihood of spiking events (ILSE), 475 treating this vector as if it were the firing rate of the cell 17 . 476 Rate maps. Rate maps of activity in the compartment were constructed by first binning the 477 position data into pixels corresponding to a 2.5 cm x 2.5 cm grid of locations. To construct a rate 478 map, the mean ILSE was computed for each pixel and then smoothed with a 4 cm standard 479 deviation isometric Gaussian kernel. Rate maps of activity in the hallway were created by first 480 collapsing the position data onto a line drawn through the center of the hallway. Next, the 481 position data on this line were binned into 17 equally spaced pixels. Next, the mean ILSE was 482 computed for each pixel. This map was then smoothed with a 2 pixel standard deviation 483

Gaussian kernel. 484
Place cell selection. To identify place cells coding for locations within the compartment in a 485 manner to avoid a bias for or against the observation of remapping, we applied the following 486 procedures. First, we partitioned the data within the compartment according to the most recent 487 entryway, and according to the first and second halves of the recording and created rate maps 488 for each partition. Next, we computed the Pearson correlation between each pair of cross-half 489 rate map comparisons (e.g. Entry A half 1 to Entry A half 2; Entry A half 1 to Entry B half 2; 490 Entry B half 1 to Entry A half 2; Entry B half 1 to Entry B half 2), while subsampling the data to 491 match the spatial sampling distributions (described below). Next, for each cell we created a null across both most recent entryways equally. This procedure identified as place cells an average 499 of 29.3% ± 1.8% (SEM) with split-half reliable compartment rate maps in the initial CA1 500 recordings, 39.2% ± 1.9% in the CA3 recordings, 27.7% ± 1.4% in the recordings of 501 experimental mice with hm4di (26.9% ± 1.9% under Saline; 28.5% ± 2.0% under CNO), and 502 28.5% ± 1.8% in the recordings of control mice without hm4di (24.6% ± 2.3% under Saline; 503 32.1% ± 2.5% under CNO). 504 Matching sampling distributions. To correct for biases in sampled spatial locations, we 505 subsampled our data during all comparisons to match the spatial sampling distributions across 506 all comparisons (Fig. S1). To do so, we computed the minimum number of samples recorded at 507 each pixel location across all comparisons. Next for each comparison we included a random 508 subset of the data recorded at that location to match that minimum number of samples. for each comparison. Next, the Pearson correlation between these two vectors was taken. 515 Similar results were observed for all analyses when cosine distance was instead used to assess 516 population similarity (not shown). 517 Split-half change in firing rate. To quantify changes in the relative firing rates at an individual cell 518 level, we computed the absolute percent change in firing rate within the compartment across 519 both halves of the recording when entering through either the same entryway or different 520 entryways. Firing rates in each half were normalized to the maximum firing rate for that half. For 521 example, the absolute percent change in firing rate between entering from entryway A in the first 522 half and entryway B in the second half ∆F , was computed as: 523 where F 1 denotes the firing rate when entering from entryway A in the first half of the recording, 525 F 2 denotes the firing rate when entering from entryway B in the second half of the recording, 526 etc. 527

Histological confirmation of virus expression and recording targets 528
Once mice had completed all behavioural experiments, animals were perfused to verify GRIN 529 lens placement and virus expression. Mice were deeply anesthetized and intracardially perfused 530 with 4% paraformaldehyde in PBS. Brains were dissected and post-fixed with the same fixative. 531 Coronal sections (40 μm) of the entire hippocampus were cut using a vibratome and sections 532 were mounted directly on glass slides. Sections were split and half of all sections were stained 533 for cresyl violet to localize GRIN lens placement, the other half of sections were stained for 534 DAPI and mounted with Fluromount-G (Southern Biotechnology) to evaluate virus expression. 535 To evaluate the hm4di-mCherry fluorescence for each mouse, pyramidal cells were quantified 536 across eight coronal slices representing the injection sites along the septo-temporal axis of the 537 hippocampus, with three randomly selected images taken across three subregion (CA1, CA3, 538 dentate gyrus) for a total of 72 images analysed per animal. To evaluate the level of expression 539 in each subregion, the number of transfected pyramidal cells was counted using imageJ and 540 virus expression was imaged with an AxioObserver.Z1 microscope (Carl Zeiss). 541

Statistics 542
All statistical tests are noted where the corresponding results are reported throughout the main 543 text and supplement. All tests were uncorrected 2-tailed tests unless otherwise noted. Z-values 544 for nonparametric Wilcoxon tests were not estimated or reported for comparisons with fewer 545 than 15 datapoints. In all figures, the mean served as the measure of central tendency, and 546 error bars reflected ±1 standard error of the mean. 547

Data availability 548
Source data for all experiments are publicly available at [insert Dryad link] or via request to the 549 corresponding author. 550

Code availability 551
All custom code written for reported analyses are publicly available at [insert Github link] or via 552 request to the corresponding author. 553 Figure S1. Additional analysis of initial CA1 and CA3 recordings. a) Biases in the sampling 555 of spatial locations within the compartment may be correlated with the most recent entryway. To 556 control for these possible biases, we subsampled the data to match the sampling distributions 557 across all conditions prior to all analyses. b) CA1 place field locations within the compartment 558 when the mouse entered from entryway A versus entryway B were highly correlated. c) 559 Population vector correlations of the CA1 code within the compartment when the mouse entered 560 from entryway A versus entryway B, when the relative orientation of the entryway B map is 561 rotated. In all cases, no rotation (0°) yielded the maximum correlation between maps. d-e) as in 562 (b-c) except for all CA3 recordings. f) Schematic of the hallway bounds and linearization. g) 563 Example linearized rate maps for all place cells coding for locations in the hallway partitioned by 564 most recent entryway from the example CA1 recording depicted in Fig. 1d. Maps are normalized 565 to the maximum across both maps separately for each cell. h) As in (g) except for the example 566 CA3 recording depicted in Fig. 2a. i) Split-half correlations of population activity within the 567 hallway when the mouse entered from the same versus the different entryway for initial CA1 568 recordings. Correlations were significantly higher when the mouse entered from the same 569 entryway (Wilcoxon signed rank test: Z = 4.5301, p = 5.90e-6). j) As in (i) except for CA3 570 recordings. Correlations were significantly higher when the mouse entered from the same 571 entryway (Wilcoxon signed rank test: Z = 4.9365, p = 7.95e-7). k) Relationship between 572 population vector measures of remapping within the compartment versus within the hallway for 573 initial CA1 recordings. l) As in (k) except for CA3 recordings. 574 Figure S2. The CA1 rate code depends on entryway in a larger environment. We tested 576 whether the remapping of the CA1 rate code by entryway could also be observed in a larger 577 environment (60 x 36 cm). To this end, we recorded 2601 place cells among 9315 cells during 578 55 sessions from a new cohort of 4 mice (AKCA102, AKCA110, AKCA115, and AKCA119). a) 579 Schematic of the larger recording environment. b) Twenty-nine example rate maps from 580 simultaneously recorded place cells in this larger environment for one session from one mouse 581 when the data are divided by entryway and session half. Rate maps are normalized to the peak 582 across all four maps. c) Split-half correlations of population activity within the compartment 583 when the mouse entered from the same versus the different entryway. Correlations were 584 significantly higher when the mouse entered from the same entryway (Wilcoxon signed rank 585 test: Z = 6.125, p = 9.08e-10). d) Same minus different entryway split-half correlations of 586 population activity within the compartment as a function of recording session (left) and grouped 587 (right). Correlations were significantly higher when the mouse entered from the same entryway 588 in all groups (Wilcoxon signed rank test: p < 0.0086), with no reliable differences between 589 groups (Wilcoxon rank sum test: Zs < 1.1565, ps > 0.247). e) Cumulative distribution of split-half 590 changes of mean firing rates within the compartment when the mouse entered from the same 591 versus the different entryway. Mean firing rates were significantly more similar when the mouse 592 entered from the same entryway (Wilcoxon rank sum test: Z = 9.609, p < 1e-12). **p < 0.01, ***p 593 < 0.001 594 when only including data from progressively longer times since entering the compartment for 601 Saline and CNO sessions for mice with and without hm4di. sum tests, Saline 1 to 4 versus CNO 1 to 4: p = 0.0488; Saline 1 to 4 versus CNO 5 to 9: p = 608 0.0259; Saline 5 to 9 versus CNO 1 to 4: p = 0.0620; Saline 5 to 9 versus CNO 5 to 9: p = 609 0.0148; All other comparisons: p > 0.3949). *p < 0.05 **p < 0.01 610 Figure S4. Other representational characteristics are preserved during DG-CA3 612 inhibition. All statistical results denote the outcome of uncorrected Wilcoxon rank sum tests 613 between conditions. a) Cumulative distribution of mean firing rates of the whole-session 614 unpartitioned data during Saline and CNO sessions for mice with hm4di. No reliable difference 615 was observed (Z = 1.474, p = 0.140). b) Cumulative distribution of peak firing rates during 616 Saline and CNO sessions for mice with hm4di. Peak firing rate defined as the maximum rate 617 across the whole-session unpartitioned rate map. coordinates indicate mm from brain surface. 655