Graduate Theses and Dissertations
Iowa State University Capstones, Theses and
Dissertations
2019
Effect of spatial and non-spatial changes on
perceived self-location
Lucia Annaleigh Cherep
Iowa State University
Follow this and additional works at: https://lib.dr.iastate.edu/etd
Part of the Neuroscience and Neurobiology Commons, and the Psychology Commons
This Thesis is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital
Repository. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Iowa State University Digital
Repository. For more information, please contact digirep@iastate.edu.
Recommended Citation
Cherep, Lucia Annaleigh, “Effect of spatial and non-spatial changes on perceived self-location” (2019). Graduate Theses and
Dissertations. 16988.
https://lib.dr.iastate.edu/etd/16988
Effect of spatial and non-spatial changes on perceived self-location
by
Lucia Cherep
A thesis submitted to the graduate faculty
in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
Major: Psychology
Program of Study Committee:
Jonathan Kelly, Major Professor
Eric Cooper
Christian Meissner
The student author, whose presentation of the scholarship herein was approved by the
program of study committee, is solely responsible for the content of this thesis. The
Graduate College will ensure this thesis is globally accessible and will not permit
alterations after a degree is conferred.
Iowa State University
Ames, Iowa
2019
Copyright © Lucia Cherep, 2019. All rights reserved.
ii
TABLE OF CONTENTS
Page
NOMENCLATURE ……………………………………………………………………………………………. iii
ACKNOWLEDGMENTS …………………………………………………………………………………… iiv
ABSTRACT
………………………………………………………………………………………………………… v
CHAPTER 1. INTRODUCTION …………………………………………………………………………… 1
Animal Research on Place Cells
………………………………………………………………………… 2
Human Research on Place Cells ………………………………………………………………………… 6
Current Study. ……………………………………………………………………………………………….. 11
CHAPTER 2. EXPERIMENT 1 …………………………………………………………………………… 17
Method. ………………………………………………………………………………………………………… 18
Analyses
……………………………………………………………………………………………………….. 22
Results …………………………………………………………………………………………………………. 23
Discussion
…………………………………………………………………………………………………….. 26
CHAPTER 3. EXPERIMENT 2 …………………………………………………………………………… 27
Method. ………………………………………………………………………………………………………… 27
Results …………………………………………………………………………………………………………. 28
Discussion
…………………………………………………………………………………………………….. 30
CHAPTER 4. GENERAL DISCUSSION ……………………………………………………………… 35
REFERENCES ………………………………………………………………………………………………….. 39
FIGURES
………………………………………………………………………………………………………….. 43
APPENDIX. IRB APPROVAL ……………………………………………………………………………. 56
iii
NOMENCLATURE
VR
Virtual Reality
VE
Virtual Environment
SAE
Sensorimotor Alignment Effect
JRD(s)
Judgement(s) of Relative Direction
iv
ACKNOWLEDGMENTS
I would like to thank my major professor, Dr. Jonathan Kelly, and my committee
members, Dr. Eric Cooper and Dr. Christian Meissner, for their guidance and support
throughout the course of this research.
In addition, I would also like to thank my friends, colleagues, the department
faculty and staff for making my time at Iowa State University a wonderful experience. I
want to also offer my appreciation to those who were willing to participate in my
experiments, without whom, this thesis would not have been possible.
v
ABSTRACT
Place cell activity is measured through single-cell recording in animals, though place-
responsive cells and related properties have been identified in the human hippocampus.
Human behavioral studies would strengthen these findings, especially given the challenge
of conducting neuroscientific research on human place-responsive cells. The current
study was based on the finding (Lenck-Santini et al., 2005) that rodent place cells
partially remap after spatial environmental changes (rotating objects relative to enclosure)
but are unaffected by non-spatial changes (object substitution). In two completed studies,
human perceived self-location was evaluated in response to spatial and non-spatial
changes in a virtual environment (VE). Participants studied object locations in a learning
VE with three orienting cues: two landmarks and a featural cue (blue stripe on the wall of
the surrounding circular room). Participants then performed judgments of relative
direction (JRD) in which they imagined various perspectives using the learned object
locations. The JRD task was performed while standing in one of four test VEs which
varied in spatial and non-spatial changes relative to the learning VE. Perceived self-
location was inferred from the presence/absence of a sensorimotor alignment effect
(SAE), indicated by facilitation for imagined perspectives aligned with the body at
retrieval. It was expected that the SAE would be present in non-spatial change VEs and
absent in spatial change VEs. As predicted, results indicated that non-spatial changes did
not disrupt perceived self-location (SAE present). Spatial changes did disrupt perceived
self-location (SAE absent), but this effect appeared to depend on participant view at test.
1
CHAPTER 1. INTRODUCTION
The majority of species effectively navigate across vast environments by relying
on strategies that take advantage of available sources of information. External
information, such as visual cues (e.g., proximal and distal landmarks) and the geometric
structure of the environment is one source. Another source originates from self-
movement cues, which include optic flow (visually experienced movement through the
environment), proprioception (sense of body position and effort in movements), and
vestibular cues (Wolbers & Hegarty, 2010). For example, the desert ant (Cataglyphis)
can travel several meters in a curvilinear, outward path and then return in a linear,
homeward path. This feat utilizes by path integration, a process which integrates body-
based self-motion cues over time and polarized light from the sun to sense direction
(Wehner, 2003). These multiple sensory signals input to various cellular networks and
combine to form an internal spatial representation or cognitive map of the environment.
An animal’s self-localization, or understanding of its position in space relative to
the surrounding environment, was based on Tolman’s (1948) theory that mammals use
spatial information as if it was stored in a map-like fashion. This theory, which was
subsumed to be an integration of spatial knowledge and personal experiences, was
elucidated in rodent behavior. A rat was trained to follow a path in a maze to a specified
location where the rat was rewarded with food. After four days, the maze was altered.
The original path was blocked and 12 arms radiated from the central arena. Prevented
from using the original path, the rat explored the environment until it selected a new arm
and traversed the entire length. Nineteen of the 53 rats (36%) chose the arm closest in
distance to the original path (i.e., the selected arm had a location about four inches from
2
the original location). This result suggests that the rats acquired knowledge of the
direction of the original location, and selected a new path with a location spatially close
to the original location (Tolman, Ritchie, & Kalish, 1946). The rodent brain appeared to
form a representation of the rat’s current position while simultaneously integrating the
rat’s previous experience with the original location into a “shortcut” the rat had never
experienced (i.e., the entire length of the new radial arm). This combination of spatial
information and personal experiences encouraged Tolman (1948) to refer to a theory of
how we represent our surrounding environment as a metaphorical cognitive map.
Humans are also quite adept at representing shortcuts in familiar environments. A
conceptually similar paradigm is the triangle completion task, which requires participants
to traverse two path legs then indicate the origin, usually by walking or pointing.
Performance is typically accurate, with an average heading error of about 10˚ when
participants physically walk and turn to complete the task (Klatzky, Loomis, Beall,
Chance, & Golledge, 1998).
Animal Research on Place Cells
The theory of a cognitive map was further specified by work which suggested that
the hippocampus serves as a map-like representation of space (O’Keefe & Dostrovsky,
1971; O’Keefe & Nadel, 1978). The mammalian hippocampus consists of two “C-
shaped” parts, the cornu ammonis (CA) fields, and it is currently suggested that around
11-25% of neurons in the human hippocampus and parahippocampal regions respond
purely to spatial locations (Ekstrom et al., 2003; Miller et al., 2013). To characterize the
role of the hippocampus as a representation of space, O’Keefe, and Dostrovsky (1971)
used electrophysiology, a process which measures electrical activity associated with
activity in the body, to access single pyramidal cells in the dorsal hippocampus (CA1 and
3
4). By inserting microelectrodes into a rat brain, they were able to record action potentials
extracellularly. Wires from a preamplifier were attached to recording equipment and
displayed firing in real-time, while postmortem histology confirmed the location of
recording sites. Out of 76 recorded units, eight cells were of interest due to their
preferential firing in a specific location relative to non-existent firing, or silence, across
other locations. Novel tactile (e.g., placing a hand on the rodent), visual (e.g., rotating the
platform, dimming light sources), and olfactory stimuli were either introduced or
removed in an attempt to alter cell firing, but these unique variations in sensory
information did not produce a differential firing response in those cells. Thus, these cells
appeared to not rely preferentially on any single sensory input but instead weighted them
equally as evidenced by the inability to disrupt firing through single cue alteration. Only
the manipulation of several items in the environment, such as varying the size and shape
of the animal’s environment, elicited altered firing responses of recorded cells. From
these results, O’Keefe and Dostrovsky (1971) proposed that the hippocampus functions
as a spatial map.
The cells of interest in O’Keefe and Dostrovsky’s (1971) experiment that fire
preferentially based off of an animal’s occupied location in an environment were first
referred to as “spatial cells” (see Figure 1). This name would later be refined to the
current concept of “place cells.” The discovery of place cells in the hippocampus was
regarded as a prime example for the role of the hippocampus in the formation of a
cognitive map and the beginning of several investigations into elucidating single cell
responses from the hippocampus and surrounding regions (Eichenbaum, 2017; O’Keefe
& Nadel, 1978). Additional evidence for the hippocampus serving as a neural
4
representation of space came from discoveries of a class of cells that respond to the
direction that an animal is facing at a given time, aptly named “head direction cells”
(Muller, Ranck Jr., & Taube, 1996; Taube et al., 1990a, 1990b).
More recently “grid cells” have also been found in regions within the
hippocampal system (e.g., the medial entorhinal cortex (MEC) and in the pre- and
parasubiculum) (Boccara et al., 2010; Moser, Rowland, & Moser, 2015; Hartley, Lever,
Burgess, & O’Keefe, 2014). Grid cells fire in a hexagonal pattern when an animal
navigates a given space and are presumed to support place cell formation through
additive firing (McNaughton, Battaglia, Jensen, Moser, & Moser, 2006; Solstad, Moser,
& Einevoll, 2006). The importance of refining the role of the hippocampus was
recognized in 2014 when the Nobel Prize in Physiology or Medicine was awarded to
John O’Keefe and May-Britt and Edvard Moser for their discoveries of cells that
constitute a “positioning system” in the brain. These findings have led to the view that the
hippocampus and surrounding structures represent an internal system that supports spatial
navigation.
For this paper, the focus will be primarily on experiments that have investigated
properties of place cells. Place cells denote a location in the environment by combining
several sensory inputs (O’Keefe, 1979), and though place cells are typically recorded
from the hippocampus, these cells have also been found in additional regions, such as the
dentate gyrus and MEC (Grieves & Jeffrey, 2017; O’Keefe, 1979; Park, Dvorak, &
Fenton, 2011). One property of place cells is stability over time. For example, Thompson
and Best (1990) recorded a single place cell that fired in the same location during 14
independent sessions over 153 days (about five months). However, if the environment
5
changed (e.g., altered geometric shape) place cells can change the location of their firing,
or cease entirely, and represent the space uniquely (Anderson & Jeffrey, 2003; O’Keefe
& Conway, 1978; Wills, Lever, Cacucci, Burgess, & O’Keefe, 2005) (see Figure 2). This
process of differential firing based on altered environments is referred to as “remapping”
(Muller & Kubie, 1987). This phenomenon of remapping appears to be unpredictable,
since researchers cannot reliably predict where (or if) a particular place cell will fire in a
novel environment.
Remapping in place cells is unique, and this is especially evident when compared
to firing properties from head direction cells. For example, if the geometric configuration
of space is altered (e.g., change from a circular environment to a rectangular
environment), head direction cells will not cease firing or change location (Taube et al.,
1990b). Instead, head-direction cells have been shown to rotate along with the
environment to preserve the cell’s preferred firing direction. For example, the rotation of
a cue card produced almost near-equal rotation in the preferred firing direction of head-
direction cells (Taube et al., 1990b). Conversely, two place cells that represent adjacent
locations in one enclosure may not represent adjacent locations in a different enclosure.
In fact, place cells might not even respond at all. However, this change in place cell firing
between distinct environments can revert to the original firing pattern if the animal is re-
introduced to the original environment (Muller & Kubie, 1987; O’Keefe & Conway,
1978). In other words, if the animal recognizes a space, then place cell firing will
demonstrate the original firing patterns. These observations suggest that an animal can
either recognize its occupied space, if place field firing is consistent across time, or
regard the space as novel or distinct if place field firing changes. Therefore, place cells
6
are presumed to be a mechanism within the hippocampus that can distinguish locations
both within and between environments.
The majority of research concerned with understanding the neuronal
representations of space have been based on experiments with rats and mice; however,
experiments investigating place-related activity in non-human primates have yielded
results consistent with rodent studies. Hori et al. (2005) recorded cells within the
hippocampal formation in two adult monkeys (Macaca fuscata) as they performed goal-
oriented navigation tasks projected on a screen. Three environments were displayed, with
different arrangements of distal landmarks to distinguish the arenas. The recordings
indicated place-related activity within the hippocampus formation across the virtual
arenas that were consistent with remapping observed in rodent place cell activity across
unique environments. These findings suggest that non-human primates have neuronal
representations of space that respond similarly as the place cells recorded in rodents
across spatially distinct environments. It appears that knowledge from place cell
recordings in rodents can inform predictions for larger, more complex mammals, such as
non-human primates.
Furthermore, virtual reality (VR) has been immensely popular as a tool to
investigate and understand real-world spatial cognition. Several studies have shown that
the use of VR compliments spatial phenomena typically studied in the real world, such as
the sensorimotor alignment effect (Williams, Narasimham, Westerman, Rieser, &
Bodenheimer, 2007) and spatial updating (Ruddle & Lessels, 2006).
Human Research on Place Cells
Although most prior research on place cells used single-cell recordings in
mammals, research with humans also corroborates these findings. The current view of the
7
role of the hippocampus in humans is thought to be involved in more general memory,
including spatial functions essential to navigation. For example, atrophy in the
hippocampus is often associated with neurodegenerative diseases, such as Alzheimer’s
disease (Fox et al., 1996). Patients with a damaged hippocampus often exhibit difficulty
in forming new, long-lasting memories of personally experienced events as well as
deficits in spatial orientation and navigation (Scoville & Milner, 1957; Spiers, Maguire,
& Burgess, 2001; Vargha-Khadem et al., 1997). Based on these observations, the human
hippocampus appears to support navigational systems and warrants further investigation
for the presence and function of neuronal representations of space that complement the
neuronal mechanisms extensively studied in other mammals.
Methods for understanding spatial cognition in the human brain have included
functional neuroimaging, such as functional magnetic resonance imaging (fMRI) which
allows for imaging of the entire brain for the study of structural relations. For example,
participants given active spatial tasks demonstrate activation within the hippocampal
formation (Hartley et al., 2014). Other functional neuroimaging methods, such as
positron emission tomography (PET) which detects areas of high blood flow in the brain,
have also established correlations between activation in the right hippocampus and goal-
oriented navigation (Maguire et al., 1998).
Other studies with humans have used more direct measures, such as intracranial
electrophysiology, a process where microelectrodes are implanted (usually into an
epileptic patient’s brain to identify seizure origins). Once these microelectrodes are
implanted, it is possible to record the activity of individual neurons while participants
perform various tasks. Ekstrom et al. (2003) directly recorded 317 neurons in seven
8
epileptic patients while they performed a navigational “taxi-driver” game in a virtual
town. Targeted brain regions included the hippocampus, parahippocampal region,
amygdala, and areas in the frontal lobes (e.g., anterior cingulate, orbital frontal cortex,
and supplementary motor cortex). To approximate cellular function, spike rates were
compared as a function of the participant’s location in the virtual town (place), the object
viewed (view), and target (goal).
Several cells responded to place and view, but to qualify as solely place-
responsive, it was imperative that the recorded cells did not respond to view of an object
and location or goal and location. About 11% of the total recorded cells fit the criteria for
place-selectivity (31 out of 279), and these cells were found at a significant frequency in
the hippocampus. Within the hippocampus, around 24% of recorded cells were identified
as place-responsive cells. These results suggest that the human hippocampus has place-
responsive cells that can form a flexible map-like representation of space. Overall, these
studies have established an emerging model of the physiological basis of human
navigation, and a next step is to test established predictions about neural responses in
animals in human behavior. If similar properties are observed between human behavior
and animal studies, then these results will provide converging evidence for characterizing
the mechanisms of human navigation.
One such convergence came from the study of attractor dynamics in the
hippocampus and its role in the formation of distinct spatial contexts. Place cell firing
was measured in rodents exposed to novel square and circular environments that differed
in color and texture (Wills et al., 2005). It was hypothesized that exposure to the square
or circular environment would establish attractor representations and place cells would
9
fire distinctly in each environment. The results showed that place cells did fire distinctly
and could abruptly switch representations when the rodent was placed in either
environment. To investigate whether intermediate room shapes (e.g., an octagon) would
exhibit the distinct representations of either the square or circular environment, place cell
firing in four intermediate room shapes were also recorded. Results showed that place
cells did exhibit a switch from the squarelike to the circlelike pattern across the series of
intermediate room shapes. These results suggest that attractor dynamics can influence
representations of intermediate or ambiguous spaces, which could aid in reducing
interference by creating orthogonal representations of spatially-relevant contexts.
This observable difference in rodent spatial representation fueled investigations
into whether human memory retrieval is also driven by similar mechanisms (e.g.,
attractor dynamics). Participants performed a behavioral task while lying in an fMRI
scanner where objects were learned relative to specific locations within two distinct VEs
(Steemers et al., 2016). During testing, participants were asked to place each object in its
learned location in the two distinct VEs as well as four morphed VEs which resembled
intermediate versions of the two distinct VEs. The results showed that there was an
abrupt shift in object location representation across the morphed VEs, which suggests
that the hippocampus has a remapping-like response to linear changes in spatial contexts.
These experiments demonstrated that properties of rodent place cells (e.g., attractor
dynamics) could be observed in human behavior.
Though research on human spatial cognition has utilized advanced technology
(e.g., fMRI), these methods can be costly. Human behavioral studies would strengthen
these findings, especially given the challenge of conducting neuroscientific research on
10
human place-responsive cells since those studies are rare and typically involve immobile
patients. One example of successfully predicting human behavior from animal
neuroscience came from grid cell properties initially obtained from a study with rodents
(Barry, Hayman, Burgess, & Jeffery, 2007). When an environment expands or contracts,
grid cell firing in rodents will parametrically expand or contract in accordance with the
now “deformed” environment (i.e., if space expands after it has been adequately
explored, grid cells will expand their firing along the axis of the room that has expanded
and vice versa with contracted spaces).
Based on these grid cell properties, Chen, He, Kelly, Fiete, and McNamara (2015)
predicted that humans would exhibit similar biases. After experiencing the original
“primed” room in VR, participants walked an outbound path in a deformed room that was
stretched or compressed along one axis. Participants executed the homeward path to the
origin in the absence of visual cues (i.e., the room was removed from view), and
researchers predicted response biases that would either undershoot or overshoot the
origin depending on whether the room had been stretched or compressed. The results
demonstrated that path responses did show a bias in accordance with the predictions
derived from grid cell properties. For example, if the deformed space was smaller than
the original familiar space and the participant had restricted vision while walking to the
origin, the participant tended to overshoot the goal location since the original grid cell
pattern had presumably been reinstated once the deformed space was removed. The
results of Chen et al. (2015) demonstrate that human behavioral predictions can be made
from animal literature on the neural mechanisms of spatial navigation.
11
Current Study
If human perceived self-location is informed by place-responsive cells that have
similar properties to place cells found in rodents, then modifying a familiar environment
in a way that elicits a change in place cell firing in rodents should also cause humans to
perceive the modified environment as distinct. The current study used environmental
modifications (e.g., spatial changes) shown to cause changes in place cell firing in rats
and evaluated whether those environmental modifications affected perceived self-
location. Currently, there is a dearth of converging evidence from the animal
neuroscience literature and human navigation behavior. The current study aimed to
bridge this knowledge gap and add to the existing literature on human spatial cognition.
Studying human perceived self-location requires an operational definition. The
simplest way to measure human perceived self-location is to ask, “Where do you think
you are?” However, such requests are likely to create demand characteristics. Therefore,
this study used an implicit measure of perceived self-location based on the sensorimotor
alignment effect (SAE), an established effect that reflects the perceived location of the
respondent (Kelly, Avraamides, & Loomis, 2007). The SAE is best illustrated by an
example. Please pay attention to the location of two objects around you, such as the door
to your office and your office phone. Make sure that you are not currently facing the door
to your office. Now, perform the two following imagined perspective-taking trials. First,
close your eyes and point to the location of your door. Next, close your eyes, imagine
rotating your body until you are facing the door and then point to your phone as if you
occupy that new imagined perspective. The first trial should have been easier than the
second trial, and this is an example of the SAE. For our purposes in the current study, the
SAE is applied regarding an advantageous effect of imagined perspective aligned with
12
the physical body (i.e., an advantage of making spatial judgments when the body is
aligned with the imagined perspective during retrieval). Spatial judgments tend to
increase with difficulty as the imagined perspective deviates absolutely from the body, as
you may have experienced in the example illustrated above. These deviations in
perspectives are referred to as imagined perspectives misaligned with the body or body
misaligned perspectives. An example of a spatial judgment typically used in the SAE is
judgments of relative direction (JRDs). This task involves asking participants to imagine
a specific location and orientation and then point to another location/object from that
perspective (e.g., “Imagine standing at Physics hall, facing the Memorial Union, point to
Parks library”).
The current study inferred perceived self-location through the use of the SAE. It
was predicted that similar environments should facilitate the presence of the SAE, and
different environments should eliminate the SAE. This prediction was supported by Kelly
et al. (2007) who asked participants to learn several objects within one VE and then
investigated the SAE when object location retrieval occurred while the participant stood
either in the learning environment (the objects were removed before testing) or a novel
environment. Participants were asked to make judgments about remembered object
locations by imagining facing one object and then indicating with a joystick the direction
of the second object from that imagined perspective. The results showed a presence of the
SAE in the learning room, with an advantage for perspectives aligned with the
participant’s actual facing direction at the time of retrieval, but no SAE in the novel room
(i.e., no advantage for aligned versus misaligned perspectives). These results suggested
that presence or absence of the SAE during memory retrieval indicates whether the
13
participant believes he/she is in a location that resembles the learning environment (the
environment in which the remembered objects were learned) or a novel environment.
A recent study by Riecke and McNamara (2017) found that the SAE can be
instated in participants who experience a real, remote room that differed in both scale and
appearance from the real learning room. Participants first learned object locations in a
rectangular room, were disoriented and moved to a remote, sparse (i.e., contained none of
the learned objects) test room that resembled the learning room. The results showed that
participant JRDs were more accurate for imagined perspectives that were aligned with the
participant’s physical facing direction during testing than imagined perspectives that were
misaligned with physical facing direction. In a follow-up experiment, participants studied
the same objects in the previous learning room, were disoriented, and moved to a remote
test room that was still rectangular but larger in scale and cluttered with random objects
that differed from the learned objects. The results showed that participant JRDs were still
more accurate for imagined perspectives that were aligned with participant physical
facing direction but to a lesser extent. These results suggest that changing aspects of a test
environment (e.g., room scale, adding novel objects) can reduce the magnitude of the
SAE, but preserving other aspects (e.g., room shape) can facilitate SAE presence.
Therefore, the SAE appears to depend on the similarity between the learning and test
environments, with more similar environments yielding a larger SAE. Based off these
results, the SAE will be used as a proxy in the current study for the perception of self-
location, with the prediction that similar environments will yield the SAE and distinct
environments will not yield the SAE.
14
In sum, if the SAE is present in a testing environment that resembles the learning
environment, the presence implies that the participant believes that he or she is in an
environment that is comparable to the learning environment. Hence, the participant
registered these two environments similarly. If the SAE is absent, it suggests that the
participant believes that he or she is in a different space (i.e., an environment that is
distinct from the learning environment). An absence of the SAE would imply that the
participant registered the two environments as distinct or unique.
The current study explored the effect of environmental manipulations on
perceived self-location in VR. The selected environmental manipulations are based on
rodent research evaluating the effects of similar manipulations on place cell activity
(Lenck-Santini, Rivard, Muller, & Poucet, 2005). If human perception of self-location is
influenced by neurons similar to the rodent place cell system, then manipulations that
disrupt rodent place cells might also affect human perception of self-location, and the
contribution of this work will further characterize the neural mechanisms underlying
human navigation through means of behavioral predictions informed from animal
neuroscience.
One of the properties of place cells is the ability to fire differentially across
unique environments, referred to as remapping. The process of remapping can be best
characterized by recording from place cells during repeated visits to one environment,
and recording the same place cells in a novel environment and correlating the firing
fields. Place cell firing should be highly correlated across repeated visits to the same
environment, but uncorrelated across two unique environments. Lenck-Santini et al.
(2005) hypothesized that the hippocampus might be more sensitive to detecting
15
differences in spatial arrangements than to object substitution. When introduced to a
novel environment, rats will freely explore the space until, over time, exploratory
behaviors are reduced. It is then presumed that the animal has habituated to the
environment (i.e., encoded and stored critical properties of the space). The existence of a
stored representation after initial exploration is evidenced by potential reexploration,
which occurs after spatial changes (e.g., shift object locations) and non-spatial changes
(e.g., object substitution) to the habituated environment (Poucet, Chapus, Durup, &
Thinus-Blanc, 1986; Thinus-Blanc et al., 1987). If the hippocampus is damaged,
reexploration of the environment is reduced or eliminated for rats after spatial changes in
a learned environment, but reexploration occurs after non-spatial changes (Save, Poucet,
Foreman, & Buhot, 1992). Therefore, Lenck-Santini et al. (2005) predicted that a spatial
change would disrupt locational place cell firing in the hippocampus, while a non-spatial
change would not affect locational place cell firing.
Lenck-Santini et al. (2005) recorded place cells in the CA1 of the hippocampus in
rats across three environments. Rats were first familiarized in a cylindrical arena that
contained two distinct objects and a featural cue card attached to one wall. After exposure
to the first environment, rats were removed and placed into another environment where
both objects rotated 90˚, disrupting the original spatial relationship between the objects
and the card. Rats were also exposed to a third environment where one of the original,
familiar objects was replaced with a novel object, but preserved the spatial relationship
between the objects and the featural cue card (see Figure 3).
After excluding cells that were either lost too early during the experiment or fired
too scarcely, the resulting cells were analyzed across sessions. Object substitution had
16
little to no discernable effect on place cell activity regardless of the firing field’s
proximity to the substituted objects. For object rotation, some place cells were unaffected
while other place cells showed partial remapping (see Figure 4). Unaffected place cells
were mostly located near the border of the environment or far away from the rotated
objects. Place cells that were affected by object rotation were typically close to the
objects or located in between them. The results of the experiment suggest that changes in
place cell firing occurred near and between the two rotated objects, and little to no
changes in place cell firing occurred during object substitution regardless of proximity to
the substituted objects. Lenck-Santini et al. (2005) proposed that the featural cue card
provided a stable reference frame that kept the far firing field intact compared to the near-
firing fields that were modified by the object rotation. These results suggest that spatial
changes (object rotation) cause partial remapping in place cells while non-spatial changes
(object substitution) leave place cells unaffected.
17
CHAPTER 2. EXPERIMENT 1
Motivated by the results of Lenck-Santini et al. (2005) that object rotation, but not
object substitution, caused partial remapping, the current study investigated the effects of
object rotation and substitution on human perception of self-location. Human perceived
self-location is presumed to be informed by a combination of several external and
internal cues, such as place-responsive cells (Ekstrom et al., 2003). Therefore,
investigations connecting properties from animal neuroscience on place cell activity to
behavioral measures of perceived self-location warrant further investigation.
Participants learned the locations of small objects placed on the floor of a VE,
referred to as the learning environment. The learning environment (see Figure 5)
consisted of a circular room with three distinct cues: a featural cue (blue stripe on one
wall) and two landmarks (plant and cone). In this way, the layout was conceptually
similar to the initial environment used by Lenck-Santini et al. (2005). After learning,
participants were disoriented and randomly placed into one of four test VEs (see Figure
6). The four test VEs included an unchanged condition (no change condition) that was
visually identical to the learning VE (see Figure 6, top left panel), an object rotation
condition (see Figure 6, top right panel), a stripe rotation condition (see Figure 6, lower
left panel), and an object substitution condition (see Figure 6, lower right panel).
The hypotheses for this study follow the results of Lenck-Santini et al. (2005). A
participant tested in the no change condition should demonstrate facilitated JRD
performance with the presence of the SAE (i.e., response errors should be the lowest
when the participant’s body is aligned with the imagined perspective). Similar results
would be expected for participants tested in the object substitution condition since the
18
spatial arrangement was preserved from the learning VE. Participants tested in the object
rotation condition should not exhibit the SAE since the spatial arrangement from the
learning VE was disrupted, leading participants to perceive the surrounding VE as
distinct. The stripe rotation condition was not initially included in the study by Lenck-
Santini et al. (2005), and a prediction for this condition follows the prediction for the
object rotation condition since the spatial arrangement was also disrupted from the
learning VE.
Hypothesis 1A: If the spatial arrangement is not disrupted from the learning VE
(e.g., no change and object substitution conditions), the SAE will be present, indicating
participants regard the test VE as similar to the learning VE.
Hypothesis 1B: If the spatial arrangement is disrupted from the learning VE (e.g.,
object and stripe rotation conditions), the SAE will be absent, indicating participants
regard the test VE as distinct from the learning VE.
Method
Participants
Seventy-one undergraduate students from Iowa State University (F = 42)
participated in exchange for course credit. The first 64 participants were randomly
assigned to one of the four test conditions. Data from seven participants (F = 4) were
removed after outlier analysis (see Results). After outlier removal, additional participants
were assigned to fill out the conditions that lost participant data. The final sample was 64
(F = 38), and the size of each condition was as follows: no change condition (n = 16, F =
10), object rotation condition (n = 16, F = 10), stripe rotation condition (n = 17, F = 9), or
object substitution condition (n = 15, F = 9). Gender was approximately balanced across
19
conditions. This project was approved by the Iowa State University’s Institutional
Review Board (see Appendix).
Stimuli and Design
The VEs were displayed on an HTC Vive HMD. Graphics displayed in the Vive
were generated on a Windows 10 computer with an Intel 6700K processor and Nvidia
GeForce GTX 1070 graphics card. Vizard (Santa Barbara, CA) software displayed
stereoscopic images at 1080 x 1200 resolution with 100˚ horizontal x 110˚ vertical.
Images refreshed at a rate of 90 Hz and reproduced head movement and orientation of
participants as they navigated the VE.
Each VE had the same dimensions (11.4m diameter x 7.5m height) and texture on
the floor, wall, and ceiling. The five VEs included a learning VE, a no change condition
(see Figure 7, top left), an object rotation condition (see Figure 7, top right), a stripe
rotation condition (see Figure 7, lower left), and an object substitution condition (see
Figure 7, lower right). In the learning VE, the two landmarks (plant and cone) were
placed 2.5m from the center of the room, with a total of 5m of distance between them.
The no change condition was the same as the learning VE where participants learned the
small object locations. In the object rotation condition, both the plant and the cone rotated
90˚, with the cone in front of the blue stripe on the wall. In the stripe rotation condition,
the blue stripe was rotated 90˚ to be beside the cone, and the two landmarks remained in
the original position. Lastly, in the object substitution condition, the spatial arrangement
was preserved from the learning VE, but the cone was replaced with a fire hydrant.
Participants learned locations of seven small objects (e.g., tape, stapler, penguin,
ball, CD, book, and mug) organized into a pattern on the floor (see Figure 8) in the
