Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Sep 2.
Published in final edited form as: Neuropsychologia. 2025 Sep 2;219:109252. doi: 10.1016/j.neuropsychologia.2025.109252

Despite dense amnesia, patients with hippocampal lesions show partially intact navigation for recent and remotely learned environments

Joshua D Garren 1, Andrew McAvan 1,2, Melanie Gin 1, Andrew P Yonelinas 3, Matthew D Grilli 1,4, Arne D Ekstrom 1,4
PMCID: PMC12476284  NIHMSID: NIHMS2109700  PMID: 40902821

Abstract

The extent to which medial temporal lobe amnesia affects recently compared to remotely formed memories remains debated. Some studies have investigated this issue by employing either verbal recall of routes or map drawing, suggesting profound loss of recent yet largely intact remote spatial memory. Here, we studied navigation in two patients with amnesia and their matched controls by recreating virtual versions of their recent and remote neighborhoods. Both patients showed largely intact navigation of both their recent and remote neighborhoods, and, in most cases, navigated them better than a completely novel environment. Despite some intact navigation of these environments, patients showed dramatically reduced spontaneously cued episodic memories in these same environments compared to the controls. Our findings suggest that episodic memory and spatial navigation rely on partially distinct neural circuits, and support models of medial temporal lobe function in which extra-hippocampal networks are capable of supporting navigation in the absence of a functioning hippocampus.

Introduction

Lesions to the medial temporal lobes often result in amnesia, with particularly profound effects on episodic memory (Argyropoulos, et al., 2019; Rosenbaum, et al., 2005; Scoville & Milner, 1957; Yonelinas, et al., 2002), the memory for events in our lifetime. Amnesia has both anterograde and retrograde effects, meaning that it affects the ability to encode new information as well as remember information learned prior to the brain lesion (Spiers, Maguire, & Burgess, 2001). In some cases, retrograde amnesia may spare information encoded long ago (remote memories) while impacting more recent memories – revealing a retrograde gradient (Moscovitch, Nadel, Winocur, Gilboa, & Rosenbaum, 2006; Poreh, et al., 2006). Some studies have suggested that all declarative memories (episodic memories + semantic memories, i.e., facts and knowledge about the world) undergo a process of systems consolidation (Alvarez & Squire, 1994; McClelland, McNaughton, & O’Reilly, 1995; Squire & Alvarez, 1995; Squire, Cohen, & Nadel, 1984; Squire, Genzel, Wixted, & Morris, 2015; Squire, Stark, & Clark, 2004; Teng & Squire, 1999), and therefore should show retrograde gradients (see also: Ketonis, McClelland, Parra, & Radvansky, 2024), collectively referred to as the standard consolidation model. Other studies and theoretical work, including multiple trace theory, suggest that only semantic memories should show retrograde gradients in medial temporal amnesia, with episodic memories always dependent on the hippocampus (Herdman, Calarco, Moscovitch, Hirshhorn, & Rosenbaum, 2015; Moscovitch, et al., 2006; Rosenbaum, Gilboa, Levine, Winocur, & Moscovitch, 2009; Rosenbaum, et al., 2008). A third perspective, termed precision and binding, suggests that lesions to the hippocampus will result in reductions in precision of representations involved in a number of different aspects of cognition, including episodic memory, working memory, perception, and navigation (Ekstrom & Yonelinas, 2020; Goodrich, Baer, Quent, & Yonelinas, 2019; Koen, Borders, Petzold, & Yonelinas, 2017; Kolarik, Baer, Shahlaie, Yonelinas, & Ekstrom, 2018; Yonelinas, 2013). Therefore, impairments should be observed in both recent and remote memory, and navigation if the tasks require memory for highly precise representation. Resolving the conditions under which types of memories show retrograde amnesia and anterograde amnesia is important for adjudicating between competing models of medial temporal lobe function and amnesia.

Studying recent and remote memory is complicated by the fact that it can be difficult to verify the content of episodic memories. One approach that some studies have taken is to use memory for places and locations that individuals with amnesia have lived at during different points in their life, with such knowledge about places verifiable with maps. In one study from Teng & Squire (1999) involving a single individual with bilateral medial temporal lobe damage, clear deficits were present in the ability to verbally recall the environment he currently lived in (termed the “recent environment”) compared to 6 age-matched controls. In fact, the amnestic individual remembered no novel or familiar routes of the recent neighborhood he currently resided in. In contrast, he performed nearly as well, or in some cases, better than, age-matched controls for the remote neighborhood where he grew up and left at around 28 years old (Teng & Squire, 1999).These findings were taken to support the idea that declarative memories undergo a process of systems consolidation such that remote memories are no longer dependent on the hippocampus (Squire, et al., 2015). Given that the patient had some word naming difficulties (a low score on the Boston naming task indicating anomia) and marked verbal amnesia, it is also possible that these contributed to some of his learning difficulties in the recent city, which he moved to after his brain injury (Insausti, Annese, Amaral, & Squire, 2013).

A second study from Rosenbaum et al. (2000) studied a single patient with dense amnesia due to a traumatic brain injury affecting the bilateral medial temporal lobes. This patient had lived in the same neighborhood for almost 40 years, along with some of the controls, who were his family members. All participants drew maps of the neighborhood they lived in, which were largely comparable, although those of the patient lacked in some detail compared to the controls. Subsequent studies with more detailed analysis of maps drawn by patients and controls largely replicate this finding of intact schematic details in the maps of amnesia patients with some absence of more detailed information (Herdman, et al., 2015). The patient in the original Rosenbaum study (Rosenbaum, et al., 2000) could not recall or draw details from a library he had worked in for the last two years. One limitation, however, with the Rosenbaum et al. studies is that the amount of time spent in the recent and remote environments were not equated, and thus it is possible that the poorer performance on the recent environment was due to less exposure. Similar to the Teng and Squire study, the patient started working at the library after his brain injury, and thus it is possible that his severe memory deficits for the library stemmed from anterograde, rather than retrograde, effects.

Another important issue to consider with both the Rosenbaum et al. and Teng & Squire studies is that spatial memories acquired primarily through navigation and episodic memory are not necessarily the same (Ekstrom & Hill, 2023). Systems consolidation theory treats navigation-related and episodic memories as the same construct, “declarative memory,” and thus accounts for intact remote spatial memory due to offloading to the neocortex over time such that any declarative memory can exist independent of the hippocampus provided there is sufficient time to “consolidate.” (Squire & Alvarez, 1995). In contrast, multiple trace theory suggests that memories have both a schematic and detailed component (the specific event), with the schematic components accumulating gradually due to multiple traces (Rosenbaum, Winocur, & Moscovitch, 2001). Some versions of multiple trace theory have argued that navigation may involve more schematic forms of knowledge (the cognitive map) and extra-hippocampal cortical areas. While episodic memory always involves the hippocampus (Moscovitch, et al., 2006), the dividing line between episodic memories and those formed as part of navigation are not well defined in these models. Also, these studies involved a strong reliance on explicit and often verbal cues to test individuals, with even the maps requiring verbal labeling. Given that these patients already have significant deficits in verbal memory, it seems important to provide measures that tap into both non-verbal and verbal forms of memory.

Recent theoretical and empirical work suggests that navigation and episodic memory differ both in their neural underpinning and cognitive processes (Ekstrom & Hill, 2023; Ekstrom, Huffman, & Starrett, 2017). Navigation can be thought of as a cognitive-motor skill involving learning a mixture of strategies and general heuristics dependent on areas like the retrosplenial and parietal cortices. In contrast, episodic memory depends primarily on the hippocampus, with newer spatial memories further benefiting from episodic memory and more remote episodic memories less so. Put another way, the cognitive processes underlying navigation and episodic memory are predicted to have only partial overlap, with episodic memory being more critical for new spatial learning and less so for spatial environments learned longer ago (Eichenbaum, 2017; Eichenbaum, 2017; Ekstrom & Hill, 2023; Ekstrom & Yonelinas, 2020). In support for differences between spatial memories formed as part of navigation and episodic memories, lesions to the human medial temporal lobes show somewhat variable effects on spatial navigation. Several studies suggest that patients with medial temporal lobe lesions, despite showing dense amnesia, demonstrate only mild deficits in spatial memory (Bohbot, et al., 1998; Kolarik, et al., 2018; Maguire, Nannery, & Spiers, 2006; McAvan, Wank, Rapcsak, Grilli, & Ekstrom, 2022; although see: Banta-Lavenex, Colombo, Ribordy-Lambert, & Lavenex, 2014; Guderian, et al., 2015 for evidence of more profound navigation deficits with medial temporal lobe lesions). The majority of these studies, however, have involved navigating novel environments for which individuals have little familiarity. Therefore, it is important to test navigation and episodic memory in environments with which participants would have more experience across their lifetime.

Here, we tested two patients with amnesia on three different spatial environments: a town that they had never visited (“novel”), the city or town in which they currently reside (“recent”), and the city or town that they lived in while growing up but had since left and not visited in some time (“remote”). Critically, we tested their spatial navigation and episodic memories using different assays. We tested their navigation by having them find different locations in a virtual recreation of their recent and remote neighborhoods. To avoid possible difficulties related to verbal memory, we removed all street signs and only included a geometrically accurate rendering of the streets in these cities, with buildings providing only coarse shape details. These geometric details nonetheless were expected to serve as strong cues for remembering routes through their neighborhood (Teng & Squire, 1999) and possibly other details. To test their episodic memory in the same paradigm, we included verbal prompts every 4 trials of navigation asking them to verbally elaborate on any memories that may have been triggered from being back in these locations. By including a novel, recent, and remote environment, this also allowed us to compare an environment learned after the brain lesion (novel: subject to anterograde amnesia), an environment learned both before and after the brain lesion (recent: subject to both anterograde and retrograde amnesia), and an environment learned well before the brain lesion (remote: subject only to retrograde amnesia effects).

The three models discussed so far make somewhat competing predictions. The standard model of systems consolidation predicts that patients with hippocampal lesions will show temporally graded retrograde and anterograde amnesia broadly for all declarative memories (i.e., episodic, semantic, and navigation-related memories that need to be explicitly recalled). Therefore, remote neighborhoods should be remembered much better (and with little to no deficit) compared to recent and novel environments. Note that standard consolidation does not distinguish between recent retrograde and anterograde effects as both are assumed to be lost due to hippocampal damage. Multiple trace theory predicts episodic memories should always depend on the hippocampus. Therefore, episodic memories from recent and remote neighborhoods should show severe impairments in patients with lesions to the hippocampus. In contrast, navigation involves some degree of rehearsal, so recent and remotely explored environments should involve “schema” and show some preservation. Instead, the degree of rehearsal should dictate how intact these memories are. The precision and binding theory would suggest that episodic memories, regardless of their age, should show the most profound impairments due to their dependence on both binding and precision. Spatial navigation, particularly when independent from episodic memory, should show mild deficits in precision regardless of their age following lesions to the hippocampus. In other words, both recent and remote navigation-related memories should show equal degrees of precision loss.

Methods

Participants

The study consisted of 4 individuals P1, C1, P2, and C2. Two of the individuals, P1 and P2, were deemed to have hippocampal/medial temporal lobe amnesia based on a combination of their neuroanatomical lesions, etiology of brain injury, and neuropsychological performance. P1 had a history of epilepsy and surgical resection focused on the left medial temporal lobe. The surgery was completed approximately 10 years prior to the current study. Based on an interview done with one of the researchers (MDG), and neuropsychological testing done by our research group (see Table 1), P1 experienced a significant decline in memory and independence after the surgical resection, although he remains partly independent in activities of daily living.

Table 1:

WAIS-IV = Wechlser Adult Intelligence Scale, 4th Edition; FSIQ = Full Scale IQ; VCI = Verbal Comprehension Index; PRI = Perceptual Reasoning Index; WMI = Working Memory Index; PSI = Processing Speed Index; WMS-IV = Wechlser Memory Scale, 4th Edition; AMI = Auditory Memory Index; VMI = Visual Memory Index; IMI = Immediate Memory Index; DMI = Delayed Memory Index; NAART = North American Adult Reading Test; BNT = Boston Naming Test; Qualitative labels were derived from Guilmette et al., 2020.

P1 P2
WAIS-IV WAIS-IV
FSIQ Std= 110 High Average FSIQ Std= 81 Low Average
VCI Std= 98 Average VCI Std= 103 Average
PRI Std= 117 High Average PRI Std= 77 Below Average
WMI Std= 97 Average WMI Std= 80 Low Average
PSI Std= 122 Above Average PSI Std= 79 Below Average
WMS-IV WMS-IV
AMI Std= 67 Exceptionally Low AMI Std= 84 Low Average
VMI Std= 107 Average VMI Std= 84 Low Average
IMI Std= 91 Average IMI Std= 77 Below Average
DMI Std= 80 Low Average DMI Std= 87 Low Average
Supplementary Supplementary
NAART Std=116 High Average NAART Std= 118 High Average
Category Fluency z= −1.1 Low Average Category Fluency z= −2.9 Exceptionally Low
BIMT z= −1.8 Below Average BNT z= 0.5 Average
Open in a new tab

In P2’s case, medical records indicate that approximately nine years prior to the current study, she experienced a hypoxic/ischemic event that may have persisted for one to two days prior to treatment. P2 presented to the emergency department with dense amnesia and disorientation, which was reportedly partially recovered prior to discharge. Neuropsychological testing completed as part of the current study revealed impairment in immediate learning and memory, along with deficits in other cognitive domains (e.g., processing speed, category fluency). Although P2’s scores on standardized memory tests were not severely low (see Table 1), based on interviews with one of the researchers (MDG), it is clear that P2 has difficulty recalling recent, anterograde events (e.g., P2 is usually unaware of current news events and cannot describe meals from earlier in the day or activities from the weekend, among many other probes) and cannot independently manage many activities of daily living due to memory impairment. Each amnestic individual was matched with a control participant (C1, C2) based on age, sex, education, and geographical considerations.

P1 was a 57-year-old male at the time of participating and had an education of 18 years. We matched him to a 57-year-old control who also had an education level of 18 years. Our second amnestic individual, P2, was an 80-year-old female at the time of testing with an education level of 14 years. Given the age of P2, it is important to note that medical records, and informant report, indicate that she was cognitively normal prior to the hypoxic/ischemic event, and P2’s Cognitive impairment has been stable since the hypoxic/ischemic event. Her control was 82 with an education level of 16 years. We explain how we performed geographical matching below in the “interview” section.

Neurological profiles

Both individuals with amnesia showed hippocampal volume loss when compared to their healthy controls. The hippocampus proper includes CA1, CA3, anterior CA fields, and the subiculum while the extra hippocampal regions include the entorhinal (ERC), perirhinal (PRC), and posterior parahippocampal cortices (PHC), all of which we were able to analyze and consider as part of our high-resolution scans. P1 underwent a left medial temporal lobe resection to treat his epilepsy, approximately 10 years prior to the current study. Compared to C1, P1 showed a 58% loss in their left hippocampus proper and no appreciable volume loss in the right hippocampus proper. P1 exhibited an 18% volume loss in their left CA3/DG compared to his control. P1 also showed a 95% loss of volume in his left anterior CA fields and a 39% loss of volume in his left subiculum compared to his matched control. P1 also showed volume loss in the left extra hippocampal regions (41% in PHC, 17% in ERC, and 47% in PRC) and no loss in the right extra hippocampal regions. In fact, P1 showed some evidence of an enlarged right hippocampus (31% above control) and extra-hippocampal regions (PHC 102% larger than control, ERC 50% larger than control, and PRC 18% larger than control), possibly due to compensation. P2 showed a 29% loss of volume in the hippocampus proper in their left hemisphere and a 23% loss of volume in the hippocampus proper in their right hemisphere when compared to their healthy control (Figure 1). P2 showed 38% volume loss in CA1 and 22% volume loss in CA3/DG relative to her matched control. P2 also exhibited a 22% volume loss in her anterior CA fields and 29% volume loss in her subiculum compared to her matched control. In the extra hippocampal regions, P2 exhibited a 31% volume loss in her left PHC, a 9% loss in the ERC, and a 23% loss in the PRC. In the right hemisphere, P2 showed a 12% loss of volume in the PHC, a 19% loss in the PRC, and no appreciable volume loss to the ERC. Example images from P1 & P2’s high-resolution T2 hippocampal scans are shown in Figure 1.

Figure 1:

Figure 1:

Open in a new tab

T1 whole MPRAGE brain scan (top row) and T2 high-resolution hippocampal scan (bottom row) from P1 and P2. For P1, clear volume loss is evident on the left side to the resection. For P2, bilateral hippocampal loss is evident due to hypoxia/ischemia.

Interview

All participants were interviewed to ascertain how long they had lived in their hometowns (“remote”) and how long they had lived in their current residence (“recent”). To meet the criteria for inclusion in the study, a participant had to live in their remote neighborhood until roughly 18 years of age and then live in their recent neighborhood for roughly 20–30 years. Controls were interviewed prior to being recruited to ensure they had sufficiently similar exposure to their patient-matched remote and recent environments. Controls also completed a battery of neuropsychological tests to ensure that they were cognitively normal for their age and educational background. Patients and controls all lived in their remote environments during their teen years through the time they left for college. P1 moved frequently but spent the longest stretch of time in their remote environment from the ages of 12–19 years old. C1 was born and raised in his remote environment, leaving for college at the age of 17. P2 was born and raised in her remote environment, leaving for college at the age of 18. C2 lived in her remote environment between the ages of 9–18 but visited on a weekend basis for a period of time after that while attending a nearby college (18–22 years old). P1 lived in his recent environment on a permanent basis since about 2 years prior to the surgical resection, with some time in and out of the environment in the two decades prior. P2 has lived in her recent environment since about 17 years prior to the hypoxic/anoxic event. Remote environments were lived in for an average of 13.75 years (standard deviation of 4 years) and recent environments for an average of 23.5 years (standard deviation of 6.5 years).

Participants were asked to name up to 8 salient landmarks from each of these environments. These landmarks were places that they were highly familiar with and spent a great deal of time at during either their formative years or since moving to their current residence. From this list of 8 familiar landmarks obtained via the interview, we then selected 4 salient landmarks that would constitute their target locations in each environment. The four landmarks were selected to balance the geographical area between the patients and controls such that they were balanced for each remote and recent neighborhood. We did this to try to deal with the potential issue that one environment would be more difficult for one participant due to landmarks being further away from others.

Neuropsychological evaluation

All participants completed neuropsychological testing prior to their inclusion in the study. The results of the testing for P1 and P2 are reported in Table 1. Each amnestic individual had a discrepancy of more than 1.5 standard deviation between a major memory index from the WMS-IV and 1) their Verbal Comprehension Index from the WAIS-IV, which is comprised of tests largely spared by medial temporal lobe amnesia, and 2) their estimated premorbid IQ, based on the North American Adult Reading Test. The healthy controls completed a battery of neuropsychological tests that we use to screen for signs of abnormal cognitive aging (Acevedo-Molina, et al., 2023). Controls performance fell within the normative range of individuals of their age and education level.

Materials and building recent and remote virtual environments

Once the necessary information was gathered during the interview process, we began constructing each participant’s remote and recent environments. We first identified the geographic information for each city using the software CADmapper. This gave us the topographical, road, and building information for each of the participants. We ensured that the portion of each city we used contained all 4 of the target locations and the sizes of the environments for amnestic individuals and matched controls were roughly the same size in km2. After the CAD information was gathered, we then put the blueprint into SketchUp+ to render the environment in 3D. If buildings were not included in the CAD information, we cross-referenced past local map information and Google Street view to manually create prefabs for the buildings in the correct locations. This needed to be done for three of the four remote environments as they were small towns that didn’t have all the necessary building infrastructure available on CAD mapper. All the recent environments had up to date CAD information, so this was not necessary. Once the buildings were created, each aspect of the environment was placed into a group of similar features: major roads, minor roads, walkways, topography, parks, and buildings. Each feature was then colored accordingly before exporting the SketchUp+ file as a .FBX file. The .FBX file for each environment was then imported into Landmarks within the Unity game engine (version 2019.3.7f1).

Upon importing into Unity, the scale was set to keep the features of all environments the same (scale of 100 as opposed to 1). We set up colliders to try to limit exploration outside of the area we had identified from the interviews and to prevent participants from exploring non-goal buildings. Target locations were set up such that they were each a unique color and had their name associated with them. Each target location had a unique starting location associated with it. Invisible to the participant, NavMesh was placed on each environment’s topography to calculate distance error while traveling from the starting locations to the target locations. NavMesh is an AI that can calculate the optimal path from point A to point B based on environmental features, thereby calculating distance error by looking at the excess distance traveled in virtual meters. Within Landmarks, C# code was created to cycle through the various target locations, so participants had different combinations of randomly chosen start and target locations to travel between. The novel environment was a real-world suburban area that all participants were unfamiliar with. The novel environment was created in the same manner via a similar interview with someone who was not a participant in the study. Examples of novel, recent, and remote environments are shown in Figure 2.

Figure 2:

Figure 2:

Open in a new tab

Examples of recent, remote, and novel environments used for testing. Note that recent and remote environments were specific to each patient and control. The same novel environment was employed for all participants.

Procedure

Participants were seated at a table and navigated each virtual environment using the mouse and keyboard controls. This means of desktop navigation was effective for all participants with the exception of one of the amnestic individuals (P2). Controlling their direction via the mouse, pointing, and using the WASD keys to provide locomotion proved difficult, even after extensive training. Ultimately, this participant controlled the direction of movement (forward, backward, and turns) but the researcher controlled the mouse to keep the field of view as it would naturally be during first person navigation. This simulated the experience of keeping one’s eyes straight ahead to follow head direction such as in naturalistic movement.

All participants went through a brief training period in an environment that they were not later tested on. The purpose of the training was to get participants familiar with the controls and instructions for the task. Once participants were comfortable moving in VR, they began navigating their recent and remote environments. Participant P1 navigated 8 trials in each of their respective recent and remote environments and C1 navigated 16 trials in each of their respective environments. Due to an issue with the randomization code, there were some trials that needed to be discarded due to the program randomly selecting a start location near the target location. These particular trials did not measure navigation accuracy, nor memory, and therefore were not considered in any of our analyses. After excluding these trials, P1 navigated 7 trials in his recent environment, 6 in his remote environment and 7 in the novel environment while C1 navigated 14, 12, and 14 trials in each of his environments used for analysis. The issue was corrected for P2 and C2, who navigated a total of 16 useable trials.

Because we controlled for the overall area of each participant’s environment, we had each pair of participants navigate at the same speed in their matched environments (remote-remote & recent-recent). The speed in each remote environment was akin to an average walking speed, whereas the recent environment’s speed was faster. This increase in speed was due to the rather dramatic increase in total area for the more urban recent environments (on average 10–13 km2 for recent environments) compared to the smaller remote environments that constituted their hometowns (roughly 1–1.8 km2 for remote environments). This ensured that environments with dramatically different surface areas did not involve drastically different times to get from one location to another.

The compass onset was also controlled based on the same parameters as speed. The compass was employed to 1) allow participants to learn in the novel environment, which they otherwise had no exposure to 2) limit frustration due to getting lost. Each compass onset time was matched for each amnestic individual and their control in their respective environments. The compass onset time was constant for the same reason speed was kept constant amongst all participants. The compass functioned in such a way that it would point in the direction of the target location. This gave participants a visual cue to aid them in reaching the prompted target location in cases where they were lost. The onset of this compass corresponded with an amount of time that had passed without trial completion. The onset time was chosen in such a way that participants would be able to make a mistake or two while navigating to their target. This was determined by having a researcher who knew the most direct route complete the route and purposely take a wrong turn at multiple junctions.

Episodic recall

To more directly compare navigation and episodic memory, we included prompts during navigation to determine if any of the cues present provided any episodic memory recall. Every fourth trial, participants were provided with a prompt and asked to state any memories that they may have recalled. They were not specifically prompted to provide accounts of specific memories from their time spent living in any of the environments. We then recorded these spontaneous verbalizations and analyzed them for content containing instances of episodic memory recall using the autobiographical interview method (Levine, Svoboda, Hay, Winocur, & Moscovitch, 2002).

These tallies of total episodic recall using the autobiographical interview method are included in Table 2. We have also included a tally of purely semantic instances of recall (Table 3). This was to ensure that a low score on episodic memory was not due simply to a lack of verbal elaboration. Examples of such semantic information are street names, building names/information, and information about features of the environment. Participants were prompted to recall any information that came to mind. There were next to no instances of recalling episodic information from trials preceding the trials in either patients or controls.

Table 2:

Scoring and categorization of spontaneously retrieved memories using the autobiographical memory interview method

Internal Event (Episodic) External
Spontaneous Recall Event Time Place Perceptual Thought/Emotion Semantic Other Repetition
P1 Remote 1 0 0 1 1 0 0 3 0
Recent 0 0 0 0 0 0 0 0 0
Novel 0 0 0 0 0 0 0 0 0
C1 Remote 4 14 1 5 0 0 3 1 0
Recent 3 21 1 0 0 0 3 5 0
Novel 2 2 0 2 1 0 0 0 0
P2 Remote 0 0 0 0 0 0 0 0 0
Recent 0 0 0 0 0 0 0 0 0
Novel 0 0 0 0 0 0 0 0 0
C2 Remote 9 48 8 19 9 1 40 2 0
Recent 10 25 4 10 7 4 18 3 2
Novel 0 0 0 0 0 0 0 0 0
Open in a new tab

Table 3:

Table tallying the instances of semantic recall that fell outside the bounds of the autobiographical memory interview method

Episodic Semantic
P1 Remote 2 15
Recent 0 3
Novel 0 5
C1 Remote 20 24
Recent 22 25
Novel 5 3
P2 Remote 0 28
Recent 0 12
Novel 0 3
C2 Remote 85 81
Recent 50 75
Novel 0 0
Open in a new tab

Analysis

MATLAB version R2024a Update 2 (24.1.0.2578822) was used for all analyses, with RStudio version 2024.04.2 build 764 used for plotting. We considered three different dependent measures in analyses. Excess path was based on how different the distance of the selected path was from the optimal path. We calculated this by deriving the distance traveled before the compass came on and dividing by the optimal distance. The optimal distance was computed based on the most direct path that could be taken to the target considering any buildings or other obstacles. Values at or close to 1 reflected the optimal distance and values greater than 1 reflected less accurate paths. We also calculated heading direction to determine how accurate their direction was overall before the compass came on. Using the subject’s walked coordinates, we calculated the angle they travelled using MATLAB’s arctangent function, atan. We used the current coordinates and their previous coordinates, then found which unit circle quadrant it fell in based on the coordinates (the previous coordinate acted as the center). The same method was used to find the optimal angle from start to target. With these new heading angles, we found the circular mean of the heading direction using MATLAB’s circ_mean function. To find the heading direction error we simply subtracted the average heading direction we calculated previously with the target direction.

abs(target_angle_rad(trial,col)-avg_rotation(trial,col));

We used Hatchfill2 in a MATLAB package (created by Kesh Ikuma): https://www.mathworks.com/matlabcentral/fileexchange/53593-hatchfill2) to help distinguish the controls from their respective patients by using patterns on the bar plots. Number of compass trials were computed for each participant and then compared between a patient and matched control. This allowed us to determine whether patients were more likely than controls to rely on the compass for any given condition.

Statistical comparisons

Because we compared each patient with a matched control, we employed a non-parametric test, the Wilcoxon rank sum in MATLAB. To compare trials in which the compass came on between patients and controls, we employed a Fischer’s exact test using an online calculator (https://www.socscistatistics.com/tests/fisher/default2.aspx). We used RStudio to plot the paths of the subject’s we used geom_point from ggplot2 package along with scale_color_gradient function to represent time (low- green; high- red). Significance was based on an uncorrected p value of p<.05.

Results

Individuals with amnesia show partially intact navigation performance in both recent and remote environments

We considered each amnestic individual and matched control’s trajectories throughout their recent, remote, and novel environments. Note that the novel environment was the same for all participants, but the recent and remote environments were specific to the places they currently lived in and grew up in, respectively. As shown in Figure 3 both patients and controls had trials in which they struggled to find the target (left panels) and others in which they found the target quickly (right panels). The black arrow shows the optimal heading, although they were not able to take a path this direct due to buildings and other obstacles. Based on inspection of the raw trajectories, it was clear that there were instances in which the patients showed intact navigation, including in the novel environment.

Figure 3:

Figure 3:

Open in a new tab

Example patient and control trajectories. “Start” is randomized start location and “target” is location they were cued to find. Left panels for each city (“remote,” “recent,” “novel”) show trials in which the participant did not use an efficient path, right panels show an efficient path. Note that the bird-flight (“optimal” black arrow) was not typically possible due to buildings and other obstacles.

To determine the average accuracy of paths taken by amnestic individuals and controls, we first considered path error, which reflected how far the paths chosen by the participants deviated from the most direct possible path in terms of distance. Both amnestic individuals performed significantly worse in the novel environment than their matched controls (P1 vs. C1 ranksum test W(17)=94, p=0.045; P2 vs. C2 ranksum test W(30)=371, p=0.00006; Figure 4). However, neither patient showed worse excess path from their matched controls in the recent environment (P1 vs. C1, ranksum test W(19)=74, p=0.85; P2 vs. C2 W(29)=237, p=0.92). For the remote environment, P1 did not show any difference in path error from his matched control; however, P2 showed significantly better performance than her matched control at her remote environment (P1 vs. C1, ranksum test W(16)=60, p=0.82; P2 vs. C2 ranksum test W(30)=184, p=0.0027, Figure 4). Both patients performed better on the recent than novel environment (P1 novel vs. P1 recent, ranksum test W(12)=76, p= 0.0012; P2 novel vs. P2 recent ranksum test W(29)=337, p=0.0015). Both patients also performed numerically better on the remote than novel environment in terms of path error, although only P2’s difference reached significance (P1 novel vs. P1 remote ranksum test W(11)=58, p=0.23, P2 novel vs. P2 remote ranksum test W(30)=380, p=0.000013). These findings suggest largely intact spatial navigation abilities in terms of deviation of their paths from ideal on both the recent and remote environments, with evidence of some impairments in both patients for the novel environment.

Figure 4:

Figure 4:

Open in a new tab

Deviation of paths from the most direct possible path (taking into consideration obstacles) for novel, recent, and remote environments.

We next considered heading error, which provided a measure of how far the average angle of the paths that they selected overall deviated from the angle of the bird flight distance to the target before the compass came on the screen. Both P1 and P2 did worse at navigating the novel town than their matched controls (P1 vs. C1, ranksum test, W(17)=99, p=.013; P2 vs. C2, W(30)=324, p=0.025). P1 was not impaired at navigating his recent city compared to C1 (Figure 5, ranksum test, W(19)=65, p=0.39). However, P2 was significantly worse at navigating her recent city compared to C2 (rankusm test, W(29)= 329, p<0.00047). Her heading error in the recent city was not significantly different than that for the novel city (ranksum test of novel vs. recent W(29)=281, p =0.33; this was also true for P1: ranksum test, W(12)=60, p=0.38), although, as can be seen from Figure 5, both patients performed numerically better at navigating their recent environment than the novel one. Neither P1 nor P2 showed any impairment in heading error in their remote city relative to their matched controls (P1 vs. C1 ranksum W(16)=74, p= 0.13; P2 vs. C2 ranksum W(30)=277, p=0.64). They also performed better on the remote environments than their novel environments, with P1 approaching a significant difference and P2 showing a significant difference (P1 novel vs. P1 remote, ranksum W(11)=62, p =0.073; P2 novel vs. P2 remote, ranksum W(30)=367, p=0.00011). These findings suggest that both amnestic individuals had some preserved navigation ability in their recent and remote cities, although the better performance on heading error in the remote city would suggest that the more distant memories were more resistant to retrograde amnesia to a potentially greater extent.

Figure 5:

Figure 5:

Open in a new tab

Heading error of patients and matched controls. Heading error reflects the mean angle taken by the patient compared to the bird flight angle.

We next considered the total number of trials that participants found the target without the aid of the compass compared to their matched controls. Although the previous analyses considered all paths from all trials before the compass came on, it could be instructive to consider which trials were successful in terms of finding the target within the allotted time. Some caution with these analyses is necessary as both patients were somewhat more tentative navigators than the controls, possibly related to awareness of their memory issues (Figure 6).

Figure 6:

Figure 6:

Open in a new tab

Percentage of trials in which the compass came on for patients and controls

In the novel environment, only P1 showed significantly more usage of the compass than his matched control (P1 vs. C1 Fischer’s exact test p=0.017). Although P2 used the compass numerically more than the matched control, there was not a significant difference (75% vs. 56.25%, Fischer’s exact test p=0.46). P1 and P2 also used the compass numerically more than their matched controls in the recent environment, but only P2 needed the compass at a rate that was significantly higher than her control. (P1 vs. C1; 85.71% vs. 57%, Fischer’s exact test p=0.19; P2 vs. C2; 86.67% vs. 31%, Fischer’s exact test 0.0032). This was also true for the remote environment, although neither difference was significant (P1 vs. C1 50% vs. 25%, Fischer’s exact test p=0.34; P2 vs. C2 25% vs. 18.75%, Fischer’s exact test p=1). The compass came on less often in the remote than novel environment, although this was only significant for P2 (P1 novel vs. remote Fischer’s exact test p=0.27; P2 novel vs. remote Fischer’s exact test p=0.012). The compass came on equally often for the patients in the novel compared to recent environment (P1 novel vs. recent, Fischer’s exact test p=1; P2 novel vs. recent Fischer’s exact test p=0.65). Together, these findings suggest that the patients required the compass more often to navigate than the controls, which was more pronounced for the novel and recent environments.

Patients showed dramatic impairments at spontaneous retrieval of episodic memories in their respective environments compared to controls

Every fourth trial, all participants were prompted to retrieve any memories that might come to mind as a result of navigating. Navigation cued many more spontaneous event memories in the controls than in the patients (Figure 7). C1 remembered 14 events from the remote neighborhood and 21 from the recent one. P1 remembered zero episodic details from either one, although he did spontaneously retrieve a single episodic memory from one event containing perceptual information about a parking lot outside one of his target locations in his remote environment. C2 remembered 48 events from the remote environment and 25 from the recent environment while P2 remembered no events from either (Table 2). Aside from the single instance of spontaneous episodic recall from an event coming from P1, both P1 and P2 otherwise recalled purely semantic memories. These were general facts about their experiences in either their respective remote or recent environments. The information that was recalled lacked any episodic richness. We tallied semantic memories regardless of whether or not they were related to events and reported the totals in Table 3. Together, controls recalled far more episodic memories than patients although were roughly matched on memories for semantic details.

Figure 7:

Figure 7:

Open in a new tab

Episodic and semantic scoring of memories using the autobiographical interview method

We statistically quantified these differences using a χ2 goodness of fit test, which allowed us to determine whether the spontaneously recalled memories were different than an equal distribution between patients and controls. In the remote environment, P1 and C1 (χ2= 11.267, P=.00079,) and P2 and C2 (χ2 =48, P <.00001) significantly differed in episodic recall. P1 and P2 also significantly differed from their controls in spontaneous instances of episodic recall in their recent environments (P1 χ2= 21, p<.00001; P2 χ2= 25, p<.00001). The novel environment served as a control but the only episodic memories that would be possible would be from that specific environment. For the novel environment, only C1 recalled any specific events.

Notably, we also found a highly significant dissociation between patients and controls for episodic and semantic memories across environments. Together, P1 and P2 recalled a total of 1 memory that could be classified as episodic, while 72 of their memories qualified as “semantic”. As indicated in Table 3, P1 had a single instance of a recalled memory, which contained 2 pieces of episodic information and thus received a rating of 2. In contrast, C1 and C2 recalled a total of 182 pieces of information that could be qualified as episodic and 208 pieces of information that could be qualified as “semantic.” An example of how spontaneous free recall was scored can be found in figure 8. While patients recalled numerically fewer semantic memories than controls, the interaction effect was nonetheless highly significant (Fischer’s exact test, p<.001), driven by a dramatic difference in the percent of episodic compared to semantic memories (i.e., patients: 2% vs. controls: 88%). Together, these findings suggest that the patients had dramatically decreased spontaneous episodic memories for both the recent and remote environments compared to the controls, consistent with their amnesia, yet this was largely (although not exclusively) selective to their episodic memory.

Figure 8:

Figure 8:

Open in a new tab

Example scoring of spontaneous recall from a control and patient.

Discussion

To the best of our knowledge, our study is the first to fully recreate the environments in detail that individuals with amnesia experienced at different time points in their life. In our study, we designed virtual renditions of cities and towns that two individuals with amnesia had lived in years ago and currently reside in and compared them with age and education matched controls who had similar geographical exposure to their recent and remote environments. This allowed us to look at how navigating these environments cued episodic recall and at the same time, how well the individuals with amnesia could navigate these environments. With the interests of also maintaining careful experimental control, our renditions focused on capturing the precise spatial layout, important city contents such as buildings, roads, sidewalks, and other environmental features, while omitting signage and other visual details that could inadvertently bias episodic memory and navigation or result in confabulations. Our findings show that while the patients remained densely amnestic for all episodic memories regardless of the age of the memory, they showed at least partially preserved spatial navigation and spatial memory in both the recent and remote environments. These findings suggest that hippocampal lesions have a more profound effect on episodic memory than spatial navigation, regardless of the age of the memory, supporting models that suggest partially independent brain networks for episodic memory and navigation (Ekstrom & Hill, 2023; Fan, Sokolowski, Rosenbaum, & Levine, 2023).

Our findings also have potential implications regarding systems consolidation and models of episodic/semantic memory formation. As outlined in the introduction, the standard model of systems consolidation (Alvarez & Squire, 1994; McClelland, et al., 1995; Squire & Alvarez, 1995; Squire, et al., 1984; Squire, et al., 2015; Squire, et al., 2004; Teng & Squire, 1999) predicts that both memories related to spatial navigation and episodic memory fall into the broad category of declarative memories, and thus should both involve a prolonged period of consolidation. Our findings are overall inconsistent with this model. Episodic memories, even for the remote environment, were nearly absent, which would not be consistent with the idea of preserved remote episodic memories. We also found evidence for at least partially intact recent spatial memories when measuring path error, and, to a lesser extent, heading error. Standard consolidation theory would suggest that declarative memories should still be undergoing consolidation for recent environments, and thus be severely impacted by hippocampal lesions. Thus, our findings are inconsistent with standard consolidation here as well.

Our findings would be more consistent with both multiple trace theory and the precision and binding model. For all dependent measures, patients maintained some schematic knowledge of their recent and remote environments, with the excess path data suggesting that patients performed better than in a novel environment. Heading error and considerations of when the compass came on suggested that the patients also had some preserved knowledge of their recent and remote environments. These findings would appear more consistent, or at least more readily explainable, by the precision and binding model as this model predicts subtle impairments in precision of memory outside of episodic memory regardless of the age (see also: Grilli, Sabharwal-Siddiqi, Thayer, Rapcsak, & Ekstrom, 2023). These findings, though, could also be explained by some versions of multiple trace theory, which would suggest that schematic knowledge can exist for spatial memories, particularly when sufficiently rehearsed, but should lack in detail (Herdman, et al., 2015; Moscovitch, et al., 2006).

What about the novel environment? Several studies with patients with dense amnesia have suggested some intact spatial navigation and spatial memory for novel environments, with some evidence for less precise spatial memory in some studies (Bohbot, et al., 1998; Kolarik, et al., 2018; Maguire, et al., 2006; McAvan, et al., 2022). Individuals with amnesia navigated the novel environments with greater error in excess path and heading, and at least for P1, the compass came on more often during trials when compared to his control. As can be seen from the path data, however, both patients had some trials in which they navigated accurately in the novel environments. In the novel environments, there were about 25% of trials (P2) and 15% of trials (P1) in which the amnestic individuals successfully found the target within the allotted time and on several trials, they took direct routes to the targets (see Figure 3). Chance would involve a random search, in which case they would have needed the compass every trial and taken meandering routes, yet their performance was better than this, suggesting some ability to acquire new trajectories. Multiple trace theory would appear to suggest that the hippocampus is needed for the formation of “first time” memories, so it is not clear how partially intact navigation in a novel environment might be explained. Standard consolidation theory would also predict severe impairments in acquiring new memories following hippocampal damage. Precision and binding, however, would predict some deficits in the novel environment due to the higher demand for binding of new associations but some preserved function as well. Therefore, the findings from the novel environment could be seen as more supportive of precision and binding theory, although notably, both patients had partial hippocampal lesions, and it is possible their residual tissue would have supported some preserved “trace” formation.

By investigating spatial navigation based on constructing virtual environments of both recent and remote cities, we were able to assay both spatial navigation and episodic memory in the same study. Including visual and semantic cues such as road signage may have further improved navigation, and possibly episodic memory, in amnesic patients and controls but by removing these features we were better able to understand spatial knowledge elicited by abstract features such as road geometry and building locations. As such, we believe we are tapping into preserved spatial representations that were previously thought to be severely impaired or even absent in past studies involving amnesic individuals. We note that past studies that showed largely preserved remote spatial memory and severely impaired recent memories relied on either verbal descriptions of routes or landmarks (Teng & Squire, 1999) or map drawing (Rosenbaum, et al., 2000). A limitation of these studies, however, is that both required some reliance on verbal memory, particularly the need to explicitly generate spatial knowledge in the absence of spatial cues. Also, the recent environments were learned after the brain lesions in both patients, and thus were subject to anterograde effects. Here, we overcame some of these obstacles by recreating past and present environments, most of which were experienced prior to the brain lesion. We included testing of a novel environment as a control. The major finding of our study is that while verbal episodic memory showed severe impairments regardless of the age of the environment, consistent with amnesia, spatial memory and navigation were at largely intact. Our findings suggest that episodic memory and spatial navigation depend on at least partially dissociable brain networks, with episodic memory likely more dependent on the medial temporal lobes than spatial navigation and spatial memory.

In contrast to the relatively mild effects of medial temporal lobe lesions on navigation reported in this paper, past work has reported striking deficits in navigation in patients who experience stroke or other lesions to more posterior regions of the brain. These typically include areas such as retrosplenial and posterior parietal cortex. For example, Takashi et al. (1997) studied patients (taxi-drivers) who were profoundly disoriented and unable to navigate to their goals following strokes affecting retrosplenial and medial parietal cortex. These patients did not experience landmark agnosia or the presence of other visual agnosias, suggesting a specific effect to “heading direction” (see also Ino et al., 2007). In a case study conducted by Robin et al. (2017) LH, an individual with damage to his posterior retrosplenial cortex, showed significant impairments in tracking layout changes in scenes. Finally, Malhotra et al. (2009) also suggest that damage to the right posterior parietal cortex in humans results in a deficit in attention to spatial but not object details. Together, these findings suggest the necessity of retrosplenial and posterior parietal cortex to spatial navigation. We speculate that the partially intact navigation skills in the two patients studied here relied on their intact retrosplenial and posterior cortices.

Our study has some important limitations. Despite our best efforts, it was impossible to perfectly match controls with amnestic individuals in terms of length of exposure to the remote and recent environment. We also could not match the size of the cities experienced for recent and remote environments, although we tried to compensate for this by restricting the area explored and altering the speed of the avatar. It is possible that some of the differences between patients and controls in the recent environments could have been related to size and time spent living there. Also, neither patient had complete lesions to the hippocampus and one of the patients had unilateral damage consistent with a surgical resection. We note that while left sided-patients often show the most dense memory loss in verbal domain (Lee et al. 2002), P1 also showed a clear deficit in navigating the novel environment compared to his control. While possible, we think it is unlikely this was mediated by a verbal memory deficit, as we note that our environments did not contain signage or other explicitly verbal details and P1 also showed comparatively intact semantic elaborations (Table 3). Nonetheless, both patients had amnesia consistent with medial temporal lobe damage, and in the very least, our findings would provide evidence that episodic memory and spatial navigation are at least partially dissociable at the cognitive level.

Control Free Recall

“When I was gone this summer (External Event) to get out of the heat and when I came home (Event) the next morning (Time) I go out (Event)… The front yard has five trees (Semantic). Mesquites and palo verdes (Semantic). All of which drop pods, I think twice a year (Semantic). But there was nothing there when I left town (Semantic), but when I got back it was a carpet (Perceptual). My neighbor (Event)… my next door neighbor (Repetition), who I don’t know that well (Semantic) – and he was raking up his pods (Event), which were really mine. All of his pods come from my trees (Semantic), he was raking up his stuff (Repetition) and he came over and raked up my yard for me (Event)! And disposed of them (Event)!”

Patient Free Recall

“We go outside. Cross 13th street (Semantic_No Event) and go on the other side of the street. Oh there’s a creek there, Lynn creek (Semantic_ No Event). So there’s a little creek and I cross that (Semantic_No Event). On the other side of that creek is the Maid-Rite (Semantic_ No Event.”

Internal (Episodic) = Event Time Place Perceptual Thought/Emotion

External = Semantic Event Repetition Other

Semantic information not related to an event = Semantic_No Event

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Acevedo-Molina MC, Thayer SC, Horn K, Nkulu H, Ryan L, Andrews-Hanna JR, & Grilli MD (2023). Past and future episodic detail retrieval is reduced among clinically normal older adults at higher genetic risk for late-onset Alzheimer’s disease. Neuropsychology, 37, 194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Alvarez P, & Squire LR (1994). Memory consolidation and the medial temporal lobe: a simple network model. Proc Natl Acad Sci U S A, 91, 7041–7045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Argyropoulos GP, Loane C, Roca-Fernandez A, Lage-Martinez C, Gurau O, Irani SR, & Butler CR (2019). Network-wide abnormalities explain memory variability in hippocampal amnesia. Elife, 8, 1–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Banta-Lavenex P, Colombo F, Ribordy-Lambert F, & Lavenex P (2014). The human hippocampus beyond the cognitive map: Evidence from a densely amnesic patient. Frontiers in Human Neuroscience, 8, 1–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bohbot VD, Kalina M, Stepankova K, Spackova N, Petrides M, & Nadel L (1998). Spatial memory deficits in patients with lesions to the right hippocampus and to the right parahippocampal cortex. Neuropsychologia, 36, 1217–1238. [DOI] [PubMed] [Google Scholar]
  6. Eichenbaum H (2017). On the Integration of Space, Time, and Memory. Neuron, 95, 1007–1018. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Eichenbaum H (2017). The role of the hippocampus in navigation is memory. Journal of Neurophysiology, 117, 1785–1796. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Ekstrom AD, & Hill PF (2023). Spatial navigation and memory: A review of the similarities and differences relevant to brain models and age. Neuron, 111, 1037–1049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ekstrom AD, Huffman DJ, & Starrett M (2017). Interacting networks of brain regions underlie human spatial navigation: A review and novel synthesis of the literature. Journal of Neurophysiology, 118, 3328–3344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ekstrom AD, & Yonelinas AP (2020). Precision, binding, and the hippocampus: Precisely what are we talking about? Neuropsychologia, 107341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fan CL, Sokolowski HM, Rosenbaum RS, & Levine B (2023). What about “space” is important for episodic memory? Wiley Interdisciplinary Reviews: Cognitive Science, e1645. [DOI] [PubMed] [Google Scholar]
  12. Goodrich RI, Baer TL, Quent JA, & Yonelinas AP (2019). Visual working memory impairments for single items following medial temporal lobe damage. Neuropsychologia, 134, 107227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Grilli MD, Sabharwal-Siddiqi S, Thayer SC, Rapcsak SZ, & Ekstrom AD (2023). Evidence of impaired remote experience-near semantic memory in medial temporal lobe amnesia. Journal of Cognitive Neuroscience, 35, 2002–2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Guderian S, Dzieciol AM, Gadian DG, Jentschke S, Doeller CF, Burgess N, Mishkin M, & Vargha-Khadem F (2015). Hippocampal volume reduction in humans predicts impaired allocentric spatial memory in virtual-reality navigation. Journal of Neuroscience, 35, 14123–14131. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Herdman KA, Calarco N, Moscovitch M, Hirshhorn M, & Rosenbaum RS (2015). Impoverished descriptions of familiar routes in three cases of hippocampal/medial temporal lobe amnesia. Cortex, 71, 248–263. [DOI] [PubMed] [Google Scholar]
  16. Ino T, et al. , Directional disorientation following left retrosplenial hemorrhage: a case report with fMRI studies. Cortex; a journal devoted to the study of the nervous system and behavior, 2007. 43(2): p. 248–54. [DOI] [PubMed] [Google Scholar]
  17. Insausti R, Annese J, Amaral DG, & Squire LR (2013). Human amnesia and the medial temporal lobe illuminated by neuropsychological and neurohistological findings for patient E.P. Proc Natl Acad Sci U S A, 110, E1953–1962. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ketonis PP, McClelland TQ, Parra D, & Radvansky GA (2024). Human retrograde amnesia and memory consolidation. Psychonomic Bulletin & Review, 1–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Koen JD, Borders AA, Petzold MT, & Yonelinas AP (2017). Visual short-term memory for high resolution associations is impaired in patients with medial temporal lobe damage. Hippocampus, 27, 184–193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kolarik BS, Baer T, Shahlaie K, Yonelinas AP, & Ekstrom AD (2018). Close but no cigar: Spatial precision deficits following medial temporal lobe lesions provide novel insight into theoretical models of navigation and memory. Hippocampus, 28, 31–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Lee TM, Yip JT, and Jones-Gotman M, Memory deficits after resection from left or right anterior temporal lobe in humans: a meta-analytic review. Epilepsia, 2002. 43(3): p. 283–91 [DOI] [PubMed] [Google Scholar]
  22. Levine B, Svoboda E, Hay JF, Winocur G, & Moscovitch M (2002). Aging and autobiographical memory: Dissociating episodic from semantic retrieval. Psychology and Aging, 17, 677–689. [PubMed] [Google Scholar]
  23. Maguire EA, Nannery R, & Spiers HJ (2006). Navigation around London by a taxi driver with bilateral hippocampal lesions. Brain, 129, 2894–2907. [DOI] [PubMed] [Google Scholar]
  24. Malhotra P, Coulthard EJ, and Husain M, Role of right posterior parietal cortex in maintaining attention to spatial locations over time. Brain, 2009. 132(3): p. 645–660. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. McAvan AS, Wank AA, Rapcsak SZ, Grilli MD, & Ekstrom AD (2022). Largely intact memory for spatial locations during navigation in an individual with dense amnesia. Neuropsychologia, 170, 108225. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. McClelland JL, McNaughton BL, & O’Reilly RC (1995). Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. Psychol Rev, 102, 419–457. [DOI] [PubMed] [Google Scholar]
  27. Moscovitch M, Nadel L, Winocur G, Gilboa A, & Rosenbaum RS (2006). The cognitive neuroscience of remote episodic, semantic and spatial memory. Current opinion in neurobiology, 16, 179–190. [DOI] [PubMed] [Google Scholar]
  28. Poreh A, Winocur G, Moscovitch M, Backon M, Goshen E, Ram Z, & Feldman Z (2006). Anterograde and retrograde amnesia in a person with bilateral fornix lesions following removal of a colloid cyst. Neuropsychologia, 44, 2241–2248. [DOI] [PubMed] [Google Scholar]
  29. Robin J, et al. , Selective scene perception deficits in a case of topographical disorientation. Cortex, 2017. 92: p. 70–80. [DOI] [PubMed] [Google Scholar]
  30. Rosenbaum RS, Gilboa A, Levine B, Winocur G, & Moscovitch M (2009). Amnesia as an impairment of detail generation and binding: evidence from personal, fictional, and semantic narratives in K.C. Neuropsychologia, 47, 2181–2187. [DOI] [PubMed] [Google Scholar]
  31. Rosenbaum RS, Kohler S, Schacter DL, Moscovitch M, Westmacott R, Black SE, Gao F, & Tulving E (2005). The case of K.C.: contributions of a memory-impaired person to memory theory. Neuropsychologia, 43, 989–1021. [DOI] [PubMed] [Google Scholar]
  32. Rosenbaum RS, Moscovitch M, Foster JK, Schnyer DM, Gao F, Kovacevic N, Verfaellie M, Black SE, & Levine B (2008). Patterns of autobiographical memory loss in medial-temporal lobe amnesic patients. J Cogn Neurosci, 20, 1490–1506. [DOI] [PubMed] [Google Scholar]
  33. Rosenbaum RS, Priselac S, Kohler S, Black SE, Gao F, Nadel L, & Moscovitch M (2000). Remote spatial memory in an amnesic person with extensive bilateral hippocampal lesions. Nature neuroscience, 3, 1044–1048. [DOI] [PubMed] [Google Scholar]
  34. Rosenbaum RS, Winocur G, & Moscovitch M (2001). New views on old memories: re-evaluating the role of the hippocampal complex. Behavioural brain research, 127, 183–197. [DOI] [PubMed] [Google Scholar]
  35. Scoville WB, & Milner B (1957). Loss of recent memory after bilateral hippocampal lesions. J Neurol Neurosurg Psychiatry, 20, 11–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Spiers HJ, Maguire EA, & Burgess N (2001). Hippocampal amnesia. Neurocase, 7, 357–382. [DOI] [PubMed] [Google Scholar]
  37. Squire LR, & Alvarez P (1995). Retrograde amnesia and memory consolidation: a neurobiological perspective. Curr Opin Neurobiol, 5, 169–177. [DOI] [PubMed] [Google Scholar]
  38. Squire LR, Cohen NJ, & Nadel L (1984). The medial temporal region and memory consolidation: A new hypothesis. Memory consolidation: Psychobiology of cognition, 185–210. [Google Scholar]
  39. Squire LR, Genzel L, Wixted JT, & Morris RG (2015). Memory consolidation. Cold Spring Harbor perspectives in biology, 7, a021766. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Squire LR, Stark CE, & Clark RE (2004). The medial temporal lobe. Annu Rev Neurosci, 27, 279–306. [DOI] [PubMed] [Google Scholar]
  41. Takahashi N, et al. , Pure topographic disorientation due to right retrosplenial lesion. Neurology, 1997. 49(2): p. 464–9. [DOI] [PubMed] [Google Scholar]
  42. Teng E, & Squire LR (1999). Memory for places learned long ago is intact after hippocampal damage. Nature, 400, 675–677. [DOI] [PubMed] [Google Scholar]
  43. Yonelinas AP (2013). The hippocampus supports high-resolution binding in the service of perception, working memory and long-term memory. Behav Brain Res, 254, 34–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Yonelinas AP, Kroll NE, Quamme JR, Lazzara MM, Sauve MJ, Widaman KF, & Knight RT (2002). Effects of extensive temporal lobe damage or mild hypoxia on recollection and familiarity. Nat Neurosci, 5, 1236–1241. [DOI] [PubMed] [Google Scholar]

RESOURCES