Stereotypical Hippocampal Clustering Predicts Navigational Success in Virtualized Real-World Environments.

Structural differences along the hippocampal long axis are believed to underlie meaningful functional differences. Yet, recent data-driven parcellations of the hippocampus subdivide the hippocampus into a 10-cluster map with anterior-medial, anterior-lateral, and posteroanterior-lateral, middle, and posterior components. We tested whether task and experience could modulate this clustering using a spatial learning experiment where male and female participants were trained to virtually navigate a novel neighborhood in a Google Street View-like environment. Participants were scanned while navigating routes early in training and after a 2 week training period. Using the 10-cluster map as the ideal template, we found that participants who eventually learn the neighborhood well have hippocampal cluster maps consistent with the ideal-even on their second day of learning-and their cluster mappings do not deviate over the 2 week training period. However, participants who eventually learn the neighborhood poorly begin with hippocampal cluster maps inconsistent with the ideal template, though their cluster mappings may become more stereotypical after the 2 week training. Interestingly this improvement seems to be route specific: after some early improvement, when a new route is navigated, participants' hippocampal maps revert back to less stereotypical organization. We conclude that hippocampal clustering is not dependent solely on anatomical structure and instead is driven by a combination of anatomy, task, and, importantly, experience. Nonetheless, while hippocampal clustering can change with experience, efficient navigation depends on functional hippocampal activity clustering in a stereotypical manner, highlighting optimal divisions of processing along the hippocampal anterior-posterior and medial-lateral axes.

The production effect is consistent over material variations: support for the distinctiveness account.

The production effect is the superior memory for items read aloud as opposed to silently at the time of study. The distinctiveness account holds that produced items benefit from the encoding of additional elements associated with the act of production. If so, then that benefit should be consistent regardless of item type. Three experiments, using three different sets of materials and three different methods, tested this hypothesis. Experiment 1, using recognition testing, showed consistent production benefits for high and low frequency words. Experiment 2, using free recall, showed consistent production increments for pictures and words. Experiment 3, using incidental learning, showed consistent production benefits for recognition of nonwords and words. Taken together, these results fit with the distinctiveness account: Production at encoding dependably adds information to the memory record, regardless of item type or method of testing, producing a consistently reliable memory benefit.

Recallable but not recognizable: The influence of semantic priming in recall paradigms.

When people can successfully recall a studied word, they should be able to recognize it as having been studied. In cued-recall paradigms, however, participants sometimes correctly recall words in the presence of strong semantic cues but then fail to recognize those words as actually having been studied. Although the conditions necessary to produce this unusual effect are known, the underlying neural correlates have not been investigated. Across five experiments, involving both behavioral and electrophysiological methods (EEG), we investigated the cognitive and neural processes that underlie recognition failures. Experiments 1 and 2 showed behaviorally that assuming that recalled items can be recognized in cued-recall paradigms is a flawed assumption, because recognition failures occur in the presence of cues, regardless of whether those failures are measured. With event-related potentials (ERPs), Experiments 3 and 4 revealed that successfully recalled words that are recognized are driven by recollection at recall and then by a combination of recollection and familiarity at ensuing recognition. In contrast, recognition failures did not show that memory signature and may instead be driven by semantic priming at recall and followed at recognition stages by negative-going ERP effects consistent with implicit processes, such as repetition fluency. These results demonstrate that recall - long-characterized as predominantly reflecting recollection-based processing in episodic memory - may at times also be served by a confluence of implicit cognitive processes.

Hippocampal and Retrosplenial Goal Distance Coding After Long-term Consolidation of a Real-World Environment.

Recent research indicates the hippocampus may code the distance to the goal during navigation of newly learned environments. It is unclear however, whether this also pertains to highly familiar environments where extensive systems-level consolidation is thought to have transformed mnemonic representations. Here we recorded fMRI while University College London and Imperial College London students navigated virtual simulations of their own familiar campus (>2 years of exposure) and the other campus learned days before scanning. Posterior hippocampal activity tracked the distance to the goal in the newly learned campus, as well as in familiar environments when the future route contained many turns. By contrast retrosplenial cortex only tracked the distance to the goal in the familiar campus. All of these responses were abolished when participants were guided to their goal by external cues. These results open new avenues of research on navigation and consolidation of spatial information and underscore the notion that the hippocampus continues to play a role in navigation when detailed processing of the environment is needed for navigation.

Distinctive encodings and the production effect: failure to retrieve distinctive encodings decreases recollection of silent items.

Studies have shown that when aloud and silent items are studied together, silent items are remembered more poorly than when they are studied independently. We hypothesise that this cost to silent items emerges because, at test, participants search for memories of having said items aloud and when those memory searches fail, participants become uncertain about whether silent items were studied. This effect should be exaggerated if other unique distinctive encoding conditions are also included at study (e.g., mumbling, writing, typing, etc.). To test this prediction, we examined the impact of introducing mumbled, "important" (i.e., words that participants are told are the most important to remember), and mouthed words to a study list of aloud and silent words. Introducing mumbled and "important" words further impaired the recollection of silent items. Introducing mouthed items did not further impair the memorability of silent items because mouthing and speaking aloud are so similar and hence, are not fully unique from each other. The memorability of aloud items was unaffected in all conditions. These results suggest that participants search for distinctive encoding information at test, and only for items that fail those searches (i.e., silent items) do they lose confidence.

Cognitive mapping style relates to posterior-anterior hippocampal volume ratio.

As London taxi drivers acquire "the knowledge" and develop a detailed cognitive map of London, their posterior hippocampi (pHPC) gradually increase in volume, reflecting an increasing pHPC/aHPC volume ratio. In the mnemonic domain, greater pHPC/aHPC volume ratios in young adults have been found to relate to better recollection ability, indicating that the balance between pHPC and aHPC volumes might be reflective of cross-domain individual differences. Here, we examined participants' self-reported use of cognitive map-based navigational strategies in relation to their pHPC/aHPC hippocampal volume ratio. We find that greater reported cognitive map use was related to significantly greater posterior, relative to anterior, hippocampal volume in two separate samples of young adults. Further, greater reported cognitive map usage correlated with better performance on a self-initiated navigation task. Together, these data help to advance our understanding of differences between aHPC and pHPC and the greater role of pHPC in spatial mapping.

The influence of recollection and familiarity in the formation and updating of associative representations.

Prior representations affect future learning. Little is known, however, about the effects of recollective or familiarity-based representations on such learning. We investigate the ability to reuse or reassociate elements from recollection- and familiarity-based associations to form new associations. Past neuropsychological research suggests that hippocampal, and presumably recollective, representations are more flexible than extra-hippocampal, presumably familiarity-based, representations. We therefore hypothesize that the elements of recollective associations, as opposed to familiarity-based representations, may be more easily manipulated and decoupled from each other, and facilitate the formation of new associations. To investigate this hypothesis we used the AB/AC learning paradigm. Across two recall studies we observed an advantage in learning AC word pairs if AB word pairs were initially recollected. Furthermore, AB word pairs were more likely to intrude during a final AC test if those AB word pairs were initially familiarity-based. A third experiment using a recognition version of the AB/AC paradigm ruled out the possibility that our findings were due to memory strength. Our results support the idea that elements in recollective associative traces may be more discretely coded, leading to their flexible use, whereas elements in familiarity-based associative traces are less flexible.

Recollection-dependent memory for event duration in large-scale spatial navigation.

Time and space represent two key aspects of episodic memories, forming the spatiotemporal context of events in a sequence. Little is known, however, about how temporal information, such as the duration and the order of particular events, are encoded into memory, and if it matters whether the memory representation is based on recollection or familiarity. To investigate this issue, we used a real world virtual reality navigation paradigm where periods of navigation were interspersed with pauses of different durations. Crucially, participants were able to reliably distinguish the durations of events that were subjectively "reexperienced" (i.e., recollected), but not of those that were familiar. This effect was not found in temporal order (ordinal) judgments. We also show that the active experience of the passage of time (holding down a key while waiting) moderately enhanced duration memory accuracy. Memory for event duration, therefore, appears to rely on the hippocampally supported ability to recollect or reexperience an event enabling the reinstatement of both its duration and its spatial context, to distinguish it from other events in a sequence. In contrast, ordinal memory appears to rely on familiarity and recollection to a similar extent.

Forgetting Patterns Differentiate Between Two Forms of Memory Representation.

For decades, there has been controversy about whether forgetting is caused by decay over time or by interference from irrelevant information. We suggest that forgetting occurs because of decay or interference, depending on the memory representation. Recollection-based memories, supported by the hippocampus, are represented in orthogonal patterns and are therefore relatively resistant to interference from one another. Decay should be a major source of their forgetting. By contrast, familiarity-based memories, supported by extrahippocampal structures, are not represented in orthogonal patterns and are therefore sensitive to interference. In a study in which we manipulated the postencoding task-interference level and the length of the delay between study and testing, we provide direct evidence in support of our representation theory of forgetting. Recollection and familiarity were measured using the remember/know procedure. We show that the causes of forgetting depend on the nature of the underlying memory representation, which places the century-old puzzle of forgetting in a coherent framework.

The production effect in paired-associate learning: benefits for item and associative information.

In five experiments, we extended the production effect-better memory for items said aloud than for items read silently-to paired-associate learning, the goal being to explore whether production enhances associative information in addition to enhancing item information. In Experiments 1 and 2, we used a semantic-relatedness task in addition to the production manipulation and found no evidence of a production effect, whether the measure was cued recall or item recognition. Experiment 3 showed that the semantic-relatedness task had overshadowed the production effect; the effect was present when the semantic-relatedness task was removed, again whether cued recall or item recognition was the measure. Experiments 4 and 5 provided further evidence that production can enhance recall for word pairs and, using an associate recognition test with intact versus rearranged pairs, indicated that production may also enhance associative information. That production boosts memory for both types of information is considered in terms of distinctive encoding.

Remembered study mode: support for the distinctiveness account of the production effect.

The production effect is the finding that words spoken aloud at study are subsequently remembered better than are words read silently at study. According to the distinctiveness account, aloud words are remembered better because the act of speaking those words aloud is encoded and later recovery of this information can be used to infer that those words were studied. An alternative account (the strength-based account) is that memory strength is simply greater for words read aloud. To discriminate these two accounts, we investigated study mode judgements (i.e., "aloud"/"silent"/"new" ratings): The strength-based account predicts that "aloud" responses should positively correlate with memory strength, whereas the distinctiveness account predicts that accuracy of study mode judgements will be independent of memory strength. Across three experiments, where the strength of some silent words was increased by repetition, study mode was discriminable regardless of strength-even when the strength of aloud and repeated silent items was equivalent. Consistent with the distinctiveness account, we conclude that memory for "aloudness" is independent of memory strength and a likely candidate to explain the production effect.

Hippocampal Signal Complexity and Rate-of-Change Predict Navigational Performance: Evidence from a Two-Week VR Training Program.

The hippocampus is believed to be an important region for spatial navigation, helping to represent the environment and plan routes. Evidence from rodents has suggested that the hippocampus processes information in a graded manner along its long-axis, with anterior regions encoding coarse information and posterior regions encoding fine-grained information. Brunec et al. (2018) demonstrated similar patterns in humans in a navigation paradigm, showing that the anterior-posterior gradient in representational granularity and the rate of signal change exist in the human hippocampus. However, the stability of these signals and their relationship to navigational performance remain unclear. In this study, we conducted a two-week training program where participants learned to navigate through a novel city environment. We investigated inter-voxel similarity (IVS) and temporal auto-correlation hippocampal signals, measures of representational granularity and signal change, respectively. Specifically, we investigated how these signals were influenced by navigational ability (i.e., stronger vs. weaker spatial learners), training session, and navigational dynamics. Our results revealed that stronger learners exhibited a clear anterior-posterior distinction in IVS in the right hippocampus, while weaker learners showed less pronounced distinctions. Additionally, lower general IVS levels in the hippocampus were linked to better early learning. Successful navigation was characterized by faster signal change, particularly in the anterior hippocampus, whereas failed navigation lacked the anterior-posterior distinction in signal change. These findings suggest that signal complexity and signal change in the hippocampus are important factors for successful navigation, with IVS representing information organization and auto-correlation reflecting moment-to-moment updating. These findings support the idea that efficient organization of scales of representation in an environment may be necessary for efficient navigation itself. Understanding the dynamics of these neural signals provides insights into the mechanisms underlying navigational learning in humans.

Stereotypical hippocampal clustering predicts navigational success in virtualized real-world environments.

Structural differences along the long-axis of the hippocampus have long been believed to underlie meaningful functional differences, such as the granularity of information processing. Recent findings show that data-driven parcellations of the hippocampus sub-divide the hippocampus into a 10-cluster map with anterior-medial, anterior-lateral, and posteroanterior-lateral, middle, and posterior components. We tested whether task and experience could modulate this clustering using a spatial learning experiment where subjects were trained to virtually navigate a novel neighborhood in a Google Street View-like environment over a two-week period. Subjects were scanned while navigating routes early in training and at the end of their two-week training. Using the 10-cluster map as the ideal template, we find that subjects who eventually learn the neighborhood well have hippocampal cluster-maps consistent with the ideal-even on their second day of learning-and their cluster mappings do not change over the two week training period. However, subjects who eventually learn the neighborhood poorly begin with hippocampal cluster-maps inconsistent with the ideal, though their cluster mappings become more stereotypical by the end of the two week training. Interestingly this improvement seems to be route specific as even after some early improvement, when a new route is navigated participants' hippocampal maps revert back to less stereotypical organization. We conclude that hippocampal clustering is not dependent solely on anatomical structure, and instead is driven by a combination of anatomy, task, and importantly, experience. Nonetheless, while hippocampal clustering can change with experience, efficient navigation depends on functional hippocampal activity clustering in a stereotypical manner, highlighting optimal divisions of processing along the hippocampal anterior-posterior and medial-lateral-axes.

Production benefits both recollection and familiarity.

In three experiments, we investigated the roles of recollection and familiarity in the production effect--the finding that words read aloud are remembered better than words read silently. Experiment 1, using the remember/know procedure, and Experiment 2, using the receiver operating characteristic procedure, converged in demonstrating that production enhanced both recollection and familiarity. Experiment 3 supported the role of recollection by demonstrating that specific episodic information--that is, whether a word had been studied aloud or silently--was stronger for items studied aloud. These findings fit with an explanation of the production effect as hinging on two factors: greater recollection of distinctive information from the study episode, and more familiarity due to greater attention allocated to the material studied aloud.

Widening the boundaries of the production effect.

Words that are read aloud are more memorable than words that are read silently. The boundaries of this production effect (MacLeod, Gopie, Hourihan, Neary, & Ozubko, Journal of experimental psychology: learning, memory, and cognition, 36, 671-685, 2010) have been found to extend beyond speech. MacLeod and colleagues demonstrated that mouthing also facilitates memory, leading them to speculate that any distinct, item-specific response should result in a production effect. In experiment 1, we found support for this conjecture: Relative to silent reading, three unique productions-spelling, writing, and typing-all boosted explicit memory. In experiment 2, we tested the sensitivity of the production effect. Although mouthing, writing, and whispering all improved explicit memory when compared to silent reading, these other production modalities were not as beneficial as speech. We argue that the enhanced distinctiveness of speech relative to other productions-and of other productions relative to silent reading-underlies this pattern of results.

Production benefits learning: the production effect endures and improves memory for text.

The production effect is the superior retention of material read aloud relative to material read silently during an encoding episode. Thus far it has been explored using isolated words tested almost immediately. The goal of this study was to assess the efficacy of production as a study strategy, addressing: (a) whether the production benefit endures beyond a short session, (b) whether production can boost memory for more complex material, and (c) whether production transfers to educationally relevant tests. In Experiment 1 a 1-week retention interval was included, and a production effect was observed. In Experiment 2 a production effect was observed for both word pairs and sentence stimuli. In Experiment 3 educationally relevant essays were read and tested with a fill-in-the-blanks test: Memory was superior for questions that probed information that had been read aloud relative to information that had been read silently. We conclude that the production benefit is enduring and generalises to text and different test formats, indicating that production constitutes a worthwhile study strategy.

Turns during navigation act as boundaries that enhance spatial memory and expand time estimation.

Ongoing experience unfolds over time. To segment continuous experience into component events, humans rely on physical and conceptual boundaries. Here we explored the subjective representation of turns along travelled routes as boundaries. Across two experiments, turns selectively enhanced participants' subjective recollection of locations immediately preceding them, compared to their recollection of locations in the middle of a route straightaway or immediately following turns. In Experiment 2, we also observed a subjective expansion of the time spent at pre-turn, relative to post-turn, locations. These results highlight the influence of turns on memory for travelled routes and provide further evidence for a link between subjective episodic re-experiencing and temporal memory. Taken together, this evidence suggests that turns during navigation act much as boundaries do for events, enhancing memory and processing of pre-boundary locations.

Directed forgetting of visual symbols: evidence for nonverbal selective rehearsal.

Is selective rehearsal possible for nonverbal information? Two experiments addressed this question using the item method directed forgetting paradigm, where the advantage of remember items over forget items is ascribed to selective rehearsal favoring the remember items. In both experiments, difficult-to-name abstract symbols were presented for study, followed by a recognition test. Directed forgetting effects were evident for these symbols, regardless of whether they were or were not spontaneously named. Critically, a directed forgetting effect was observed for unnamed symbols even when the symbols were studied under verbal suppression to prevent verbal rehearsal. This pattern indicates that a form of nonverbal rehearsal can be used strategically (i.e., selectively) to enhance memory, even when verbal rehearsal is not possible.

Multiple Scales of Representation along the Hippocampal Anteroposterior Axis in Humans.

The ability to represent the world accurately relies on simultaneous coarse and fine-grained neural information coding, capturing both gist and detail of an experience. The longitudinal axis of the hippocampus may provide a gradient of representational granularity in spatial and episodic memory in rodents and humans [1-8]. Rodent place cells in the ventral hippocampus exhibit significantly larger place fields and greater autocorrelation than those in the dorsal hippocampus [1, 9-11], which may underlie a coarser and slower changing representation of space [10, 12]. Recent evidence suggests that properties of cellular dynamics in rodents can be captured with fMRI in humans during spatial navigation [13] and conceptual learning [14]. Similarly, mechanisms supporting granularity along the long axis may also be extrapolated to the scale of fMRI signal. Here, we provide the first evidence for separable scales of representation along the human hippocampal anteroposterior axis during navigation and rest by showing (1) greater similarity among voxel time courses and (2) higher temporal autocorrelation in anterior hippocampus (aHPC), relative to posterior hippocampus (pHPC), the human homologs of ventral and dorsal rodent hippocampus. aHPC voxels exhibited more similar activity at each time point and slower signal change over time than voxels in pHPC, consistent with place field organization in rodents. Importantly, similarity between voxels was related to navigational strategy and episodic memory. These findings provide evidence that the human hippocampus supports an anterior-to-posterior gradient of coarse-to-fine spatiotemporal representations, suggesting the existence of a cross-species mechanism, whereby lower neural similarity supports more complex coding of experience.