Part 3:Environmental Deformations Dynamically Shift Human Spatial Memory
Mar 22, 2022
Contact: Audrey Hu Whatsapp/hp: 0086 13880143964 Email: audrey.hu@wecistanche.com
4 | METHODS
4.1 | Participants
Forty-nine participants gave written consent and were paid for participating in Experiment 1, 53 for Experiment 2, and 48 for Experiment 3. One participant from Experiment 1 and four participants from Experiment 2 was excluded for performing worse than chance by the end of the last familiar block. An additional participant was excluded from Experiment 2 as an outlier (shift score > 3 above the mean, in the predicted direction), leaving a final count of48 participants in Experiment 1 (31 female, mean age 23.5, age-range 18–44), 48 in Experiment 2 (30 female, mean age 22.4, age-range 18–33), and 48 in Experiment 3 (38 female, mean age 22.9, age-range 18–44), with 24 participants in each experimental condition. The sample size was chosen prior to conducting all experiments to be double the number of participants in prior experiments studying similar effects (Chen et al., 2015). All participants provided informed consent in accordance with the Institutional Review Board of the University of Pennsylvania.
4.2 | Experimental protocols
4.2.1 | Experiment 1: Desktop virtual environment with full visual information available—We used Source SDK Hammer Editor (http://www.valvesoftware.com, Valve Software, Bellevue, WA) to construct virtual reality environments that were rendered and displayed from the first-person perspective using the commercial game software Portal (http://www.valvesoftware.com, Valve Software, Bellevue, WA). The environment was displayed on a 27-in. LG monitor (resolution: 1920 × 1080) and participants were seated roughly 50 cm from the screen. Participants learned the locations of target objects inside a virtual environment, using the learning procedure described in the main text and illustrated in Figure 2. Participants moved through the environment by using their right hand to operate arrow keys to move forward or backward and turn left or right. During the replace phase, participants navigated to their remembered object location and pressed the “r” key with their left hand to register their response. Virtual heading and location were recorded every 100 ms.
The familiar environment was a square virtual arena, with no ceiling. Each boundary wall was 116 virtual units (vu) in length × 5.6 vu in height relative to a simulated eye-level of4 vu. One virtual unit corresponds to 0.3048 real-world meters (1 ft). The four target objects were a radiator, a lamp, an oil drum, and a cake. At the start of each block, participants collected each target object in pseudo-random order twice without any interspersed replace trials. They then performed 16 replace trials (4 for each object, in pseudo-random order), each of which was immediately followed by a collect trial for the same object to provide feedback. The instructions for each trial (either “Collect” or “Replace,” followed by the target object name) were displayed in the center of the screen for the entirety of the trial. During each “collect” trial, only the to-be-collected object was present in the room. During “replace” trials, no objects were present. The same texture was applied to all walls. Distal cues, in the form of the sun, sky, and a mountain range, surrounded the arena (Figure S1). These distal cues were rendered at infinity, thus providing orientation information but no cues to location.
Participants completed two blocks, a familiar block followed by a deformation block. Only replace trials differed between blocks. The environment used in replacing trials in the deformation block was either stretched 50% along with one-dimension relative to the familiar square environment (width 174 vu × length 116 vu) or compressed 50% (width 58 vu × length 116 vu). To create these deformed environments, the floor, wall, and ceiling textures were not rescaled, but were instead truncated (during compressions) or continued to tile the new space (during stretches). Ten participants noticed a difference between the original and the deformed environment.
improve memory cistanche products
4.2.2 | Experiment 2: Desktop virtual environment with visual information
obscured during replacing trials—The design and procedures were similar to that of Experiment 1, except as described below.
The familiar environment was a square virtual room. Each wall was textured with a unique wallpaper to provide orientational cues. The floor was also repetitively textured to provide optic flow information but the floor texture provided no cues to a location inside the environment. Each boundary wall was 116 virtual units (vu) in length and 19 vu in height relative to a simulated eye-level of4 vu. The environment was completely enclosed by the walls and ceiling (Figure S1).
Participants completed three blocks. In the first block, the environment was square, and visual cues were always visible. In the second block, the environment was also a square, and visual cues during replacing trials (but not collect trials) were masked by a dense fog once the participant traveled at least 3.1 vu away from their starting location. The fog is fully saturated at 12.5 Vu, occluding all visual cues beyond this radius. All objects were located at least 30 vu from all boundaries. In the third block (deformation block), visual cues were also masked by dense fog upon movement from the initial position and the familiar room was replaced by a rectangular room which was either stretched 50% from the original square along one axis (width 174 Vu × length 116 VU) or compressed 50% (width 58 vu × length 116 vu). To create these deformed environments, the floor, wall, and ceiling textures were not rescaled, but were instead truncated (for compressions) or continued to tile the new space (for stretches). Eleven participants noticed a difference between the original and the deformed environment.
4.2.3 | Experiment 3: Immersive virtual environment with full visual and vestibular information available—The design and procedures for Experiment 3 were similar to those of experiment 1, except as described here. We used Unity game engine version 5.6 (https://unity3d.com, Unity Technologies, San Francisco, CA) to construct and render immersive virtual reality rooms via the stereoscopic HTC Vive head-mounted display and position tracker (resolution of 1,080 × 1,200 per eye; https://www.vive.com/, HTC with technology by the Valve Corporation, New Taipei City, Taiwan). Responses during the replace phase were collected by participants pressing the “trigger” key of a wireless HTC Vive controller with their dominant hand. Participants could freely move their heads and walk around the environment. Their heading and location were recorded every 100 ms. No participants complained of motion sickness during or after the experiment.
The familiar environment was a square virtual room, measuring 2.4 m in length × 2.4 m in width × 2.5 m in height. The positions of2 (north-south) virtual walls matched 2 of the physical tracking room walls, the remaining 2 (east-west) unmatched virtual walls were displaced during deformations. All walls were textured a charcoal grey. The floor and ceiling were textured a lighter grey. A light grey floor-to-ceiling 0.1 m wide × 0.1 m long column was nestled in each corner to deter participants from contacting the tracking equipment (Figure S1).
Participants completed two blocks, a familiar block followed by a deformation block. Only replace trials differed between blocks. The environment used during replacing trials of the deformation block was either stretched along one dimension (east-west) by displacing one or both unmatched walls and their neighboring columns (width 2.8 m × length 2.4 m) or compressed along this dimension (width 2.0 m × length 2.4 m). Between blocks, the display was rendered solid black for 5 s with the instructions “wait for next trial” displayed in the bottom center of the visual field.
Because participants could no longer be teleported between trials, they were instructed to move before each trial to face and nearly touch the center of one of the four walls as indicated by a floating black arrow. To ensure that the participant did not see any walls move during deformation trials, the displaced wall depended on the starting position for that trial. If the trial started from the east wall, then the west wall was displaced by 0.4 m. If the trial started from the west wall, then the east wall was displaced by 0.4 m. If the trial started from either the north or the south walls, then both the east and west walls were displaced by 0.2 m each. From all starting positions, the instantaneous displacement of walls was not visible. No participant noticed the manipulation.
The complete set of target objects was a red sphere, a blue cube, a green cylinder, and a purple capsule. Object locations were all within 0.4 m of the center of the familiar environment. All objects were presented on the same grey 1.5 m tall pedestal in order to raise them to approximately eye level (Figure S1). The target objects for each trial were selected in pseudo-random order. The instructions (either “Collect the” or “Replace the” followed by the target object name in text matching the color of the target object, or “Go to Arrow” to begin the next trial) were displayed in the bottom center of the visual field for the entirety of all trials.
cistanche benefit: enhance memory
4.3 | Analysis
All recorded data were imported into MATLAB (MathWorks) and analyzed using custom-written scripts.
4.3.1 | Object replace location analysis—As described in the main text and Figure
3, to test whether the replaced locations of objects depended on the starting boundary, we first aligned all four objects by subtracting their median replaced locations. Next, for each axis (north-south and east-west) we calculated the displacement along that axis between the median replace locations when starting from one boundary (north or east) minus the opposing boundary (south or west). Lastly, we computed the difference in shift measured along the deformed and undeformed dimensions as the final measure of interest. Medians were chosen as the measure of central tendency to mitigate the effect of outliers in replaced locations.
4.3.2 | Statistics—All statistical tests were two-tailed (unless otherwise noted)
nonparametric tests with the particular test noted accompanying each result. Given the typically long-tailed distribution of the shift data, nonparametric tests were chosen as these tests do not assume a particular shape of the tested distributions. W-statistics were reported for all Wilcoxon signed-rank and rank-sum tests. All box-and-whisker plots indicate the minimum to maximum (whisker), the first to third quartile range (box), and the median (line) of the distribution.
cistanche benefit: enhance memory
Supplementary Material
Refer to the Web version on PubMed Central for supplementary material.
ACKNOWLEDGMENTS
We gratefully acknowledge support from NSF grant PHY-1734030 (VB), NIH grants EY022350 and EY027047 (RAE), and NSF IGERT grant 0966142 (ATK). VB was also partially supported by the Honda Research Institute Curious-Minded Machines program and the Aspen Center for Physics (Aspen, Colorado; NSF grant PHY-1607611) during this time.
Funding information
Honda Research Institute Curious-Minded Machines; National Institutes of Health, Grant/Award Numbers:EY022350, EY027047; National Science Foundation, Grant/Award Numbers: IGERT 0966142, PHY-1607611, PHY-1734030
DATA AVAILABILITY STATEMENT
Data and custom MATLAB scripts implementing all analyses are publicly available at https://github.com/akeinath/HumanMemory_EnvironmentalDeformations.
cistanche benefit
REFERENCES
Barry C, Hayman R, Burgess N, & Jeffery KJ (2007). Experience-dependent rescaling of entorhinal grids. Nature Neuroscience, 10, 682–684. 10.1038/nn1905 [PubMed: 17486102]
Bellmund JLS, de Cothi W, Ruiter TA, Nau M, Barry C, & Doeller CF (2020). Deforming the metric of cognitive maps distorts memory. Nature Human Behaviour, 4, 177–188. 10.1038/s41562-019-0767-3
Burak Y, & Fiete IR (2009). Accurate path integration in continuous attractor network models of grid cells. PLoS Computational Biology, 5, e1000291. 10.1371/journal.pcbi.1000291 [PubMed: 19229307]
Bush D, Barry C, Manson D, & Burgess N (2015). Using grid cells for navigation. Neuron, 87, 507–520. 10.1016/j.neuron.2015.07.006 [PubMed: 26247860]
Bush D, & Burgess N (2014). A hybrid oscillatory interference/continuous attractor network model of grid cell firing. The Journal of Neuroscience, 34, 5065–5079. 10.1523/JNEUROSCI.4017-13.2014 [PubMed: 24695724]
Chen G, Lu Y, King JA, Cacucci F, & Burgess N (2019). Differential influences of environment and self-motion on place and grid cell firing. Nature Communications, 10, 630. 10.1038/s41467-019-08550-1
Chen X, He Q, Kelly JW, Fiete IR, & McNamara TP (2015). Bias in human path integration is
predicted by properties of grid cells. Current Biology, 25, 1771–1776. 10.1016/j.cub.2015.05.031 [PubMed: 26073138]
Cheng K, Shettleworth SJ, Huttenlocher J, & Rieser JJ (2007). Bayesian integration of spatial information. Psychological Bulletin, 133, 625–637. 10.1037/0033-2909.133.4.625 [PubMed: 17592958]
Cheung A, Ball D, Milford M, Wyeth G, & Wiles J (2012). Maintaining a cognitive map in darkness: The need to fuse boundary knowledge with path integration. PLoS Computational Biology, 8, e1002651. 10.1371/journal.pcbi.1002651 [PubMed: 22916006]
Deshmukh SS, & Knierim JJ (2011). Representation of non-spatial and spatial information in the lateral entorhinal cortex. Frontiers in Behavioral Neuroscience, 5, 69. 10.3389/fnbeh.2011.00069 [PubMed: 22065409]
Deshmukh SS, & Knierim JJ (2013). Influence of local objects on hippocampal representations: Landmark vectors and memory. Hippocampus, 23, 253–267. 10.1002/hipo.22101 [PubMed: 23447419]
Doeller CF, Barry C, & Burgess N (2010). Evidence for grid cells in a human memory network. Nature, 463, 657–661. 10.1038/nature08704 [PubMed: 20090680]
Doeller CF, & Burgess N (2008). Distinct error-correcting and incidental learning of location relative to landmarks and boundaries. Proceedings of the National Academy of Sciences ofthe United States of America, 105, 5909–5914. 10.1073/pnas.0711433105 [PubMed: 18413609]
Doeller CF, King JA, & Burgess N (2008). Parallel striatal and hippocampal systems for landmarks and boundaries in spatial memory. Proceedings of the National Academy of Sciences ofthe United States of America, 105, 5915–5920. 10.1073/pnas.0801489105 [PubMed: 18408152]
Dordek Y, Soudry D, Meir R, & Derdikman D (2016). Extracting grid cell characteristics from place cell inputs using non-negative principal component analysis. eLife, 5, e10094. 10.7554/ eLife.10094 [PubMed: 26952211]
Ekstrom AD, Harootonian SK, & Huffman DJ (2020). Grid coding, spatial representation, and navigation: Should we assume an isomorphism? Hippocampus, 30, 422–432. 10.1002/hipo.23175 [PubMed: 31742364]
Epstein RA, Patai EZ, Julian JB, & Spiers HJ (2017). The cognitive map in humans: Spatial navigation and beyond. Nature Neuroscience, 20, 1504–1513. 10.1038/nn.4656 [PubMed: 29073650]
Etienne AS, Boulens V, Maurer R, Rowe T, & Siegrist C (2000). A briefview of known landmarks reorientates path integration in hamsters. Naturwissenschaften, 87, 494–498. 10.1007/ s001140050766 [PubMed: 11151669]
Etienne AS, & Jeffery KJ (2004). Path integration in mammals. Hippocampus, 14, 180–192. 10.1002/ hipo.10173 [PubMed: 15098724]
Fiete IR, Burak Y, & Brookings T (2008). What grid cells convey about rat location. The Journal of Neuroscience, 28, 6858–6871. 10.1523/JNEUROSCI.5684-07.2008 [PubMed: 18596161]
Gallistel CR (1990). The organization oflearning. Cambridge, MA: Bradform Books/MIT Press. Hafting T, Fyhn M, Molden S, Moser M-B, & Moser EI (2005). Microstructure of a spatial map in the entorhinal cortex. Nature, 436, 801–806. 10.1038/nature03721 [PubMed: 15965463] Hardcastle K, Ganguli S, & Giocomo LM (2015). Environmental boundaries as an error correction
mechanism for grid cells. Neuron, 86, 827–839. 10.1016/j.neuron.2015.03.039 [PubMed: 25892299] Hartley T, Trinkler I, & Burgess N (2004). Geometric determinants of human spatial memory.
Cognition, 94, 39–75. 10.1016/j.cognition.2003.12.001 [PubMed: 15302327]
