The hippocampus and memory: a comparative and ethological perspective Wendy A Suzuki* and Nicola S Clayton

Transcription

1 768 The hippocampus and memory: a comparative and ethological perspective Wendy A Suzuki* and Nicola S Clayton During the past year, considerable progress has been made in our understanding of the wide-ranging functions of the hippocampus. Highlights include the development of new tasks with which to assess spatial/topographic memory in humans and monkeys, novel tests of relational memory in rats, and episodic-like memory tasks in birds. In addition, novel theories of hippocampal function have been developed that are notable for their applicability to both humans and animal models. Addresses *Center for Neural Science, New York University, 4 Washington Place, Room 809, New York, NY 10003, USA; wendy@cns.nyu.edu Department of Experimental Psychology, University of Cambridge, Downing Street, Cambridge CB2 3EB, UK; nsc22@cam.ac.uk Correspondence: Wendy A Suzuki Current Opinion in Neurobiology 2000, 10: /00/$ see front matter 2000 Elsevier Science Ltd. All rights reserved. Abbreviations DNMS delayed-nonmatching-to-sample VPC visual paired comparison Introduction Ever since the seminal description of the well-known amnesic patient HM by Scoville and Milner [1], intense research has focused on defining the role of the hippocampus in declarative memory. Declarative memory refers to the ability to remember personal past experiences in other words, the what, where and when of an event (episodic memory) as well as the ability to acquire knowledge and remember facts about the world that may or may not have been acquired through such personal experiences (semantic memory) [2]. Remembering getting soaked in the London rain last Tuesday is an example of episodic memory, but knowing that it often rains in London is an example of semantic memory, because this knowledge need not be acquired as a result of a personal experience of getting wet. Despite almost 50 years of research following the description of patient HM, a complete understanding of the role of the hippocampus in declarative memory remains elusive. Although it has often been difficult to compare the findings and theories of hippocampal function across species, the recent development of new tasks to probe hippocampal function, as well as the development of new and more inclusive theories of hippocampal function, have served to facilitate cross-species comparisons. In humans and monkeys, for example, new tasks of spatial/topographic memory have been developed using either real-world or virtual environments. Some of the most significant advances in the rat literature have been attributable to the expansion of paradigms beyond the rich inventory of existing spatial memory tasks to include innovative new tasks of olfactory recognition and relational memory that are also sensitive to hippocampal damage. One of the most promising advances has been the development of episodic-like memory tasks in birds; these tasks can also be adopted for use in other species. These recent findings are also of interest from a neuroethological point of view. A common theme of the new developments is that many of these novel tasks take advantage of the animal s natural tendencies and behaviors. Designing tasks that are based on knowledge of the ethology of a given species, or choosing experimental animals with a particularly strong, naturally evolved dependence on certain types of memory, has proven to be remarkably useful for revealing the dependence of these forms of memory on the hippocampus. These new tasks, together with ideas emerging from new theories of hippocampal function, have begun to bridge the gap between human and animal studies. In this review, we adopt a neuroethological perspective to demonstrate how the recent advances in theoretical and empirical research have expanded our understanding of the fundamental mnemonic functions of the hippocampus. We also illustrate how memory tasks of known ethological relevance to an animal can more clearly reveal the fundamental mnemonic functions of the hippocampus. New theories of hippocampal function The structures of the medial temporal lobe that are known to be important for memory include the hippocampus and the surrounding and strongly interconnected entorhinal, perirhinal and parahippocampal cortices. Since the original description of patient HM, there has been a multitude of theories proposed describing the major function of the hippocampus and its relationship with these other brain areas. A common feature of these early theories is that they have tended to focus on either animals or humans, but not on both. An exciting recent development has been the description of two new theories that have attempted to describe a theoretical framework applicable to humans as well as animal model systems. Although these theories have taken very different approaches to defining hippocampal functions, they share the common feature of differentiating the mnemonic functions of the hippocampus on the one hand, and the surrounding entorhinal, perirhinal and parahippocampal cortices on the other. In the first of these new theories, Mishkin and colleagues [3,4] argue that the hippocampus plays a selective role in episodic memory whereas the surrounding cortical areas (i.e. the entorhinal, perirhinal and parahippocampal cortices)

2 The hippocampus and memory Suzuki and Clayton 769 play the major role in semantic memory. We will refer to this and similar theories (see below) as the episodic theory of hippocampal memory function. Evidence for this theory comes from the results of testing three patients who sustained hippocampal damage early in life. These patients are described as exhibiting a clear impairment in episodic memory, with a relative sparing of semantic memory [3]. Moreover, using tasks modeled on those used in non-human primates, Vargha-Khadem et al. [3] found that performance on tasks of simple object recognition was spared in their hippocampal patients, suggesting that these tasks may be categorized as semantic-like and that they depend on the surrounding cortical areas. In contrast, their patients were impaired on an object place association task as well as a voice face association task. Because monkeys with hippocampal lesions (that also included the parahippocampal cortex) are also impaired on an object place association task [5], this suggested that certain associative memory tasks used in monkeys may have elements in common with hippocampal-dependant episodic memory in humans. Tulving and Markowitsch [6] propose a similar episodic theory of hippocampal memory function, but in their version, episodic memory is an extension of semantic knowledge: memory about the context of when and where the information was acquired is combined with the semantic knowledge that was gained on that occasion. Thus, according to the Tulving and Markowitsh theory, episodic memory cannot exist without semantic knowledge. According to both versions of the episodic theory of hippocampal memory function, semantic knowledge can be acquired by way of the surrounding cortex even in the face of impaired episodic capabilities. Both theories also state that damage to the hippocampus would result in a selective impairment of episodic memory, whereas damage to the semantic (i.e. cortical) system would impair both semantic and episodic memory. In contrast to the episodic theory of hippocampal memory function, the relational theory of Eichenbaum and colleagues [7 ] suggests that the hippocampus supports the formation of relational representations between all kinds of stimuli. A key feature of hippocampal-dependent relational representations is that they are quickly learned and allow for inferential relationships between representations that can be used in novel situations. Like the episodic theory, this theory suggests that the hippocampus plays an essential role in signaling the unique sets of relationships that make up an episode. Unlike the episodic theory, Eichenbaum and colleagues suggest that the hippocampus is also important for extracting common features across episodes and therefore plays a critical role in semantic memory as well as in simpler forms of recognition and associative memory. In contrast to the relational functions of the hippocampus, the role of the surrounding cortices is described as being important in maintaining the persistence of individual stimuli in memory [8]. This idea is supported by findings in both rats and monkeys showing that damage to these cortical areas results in abnormally rapid forgetting on various recognition memory tasks. Also consistent with this idea is the finding in rats that cortical neurons exhibit sustained stimulus-specific activity for to-be-remembered stimuli during the delay interval of various recognition memory tasks [8]. This relational view of hippocampal function is reminiscent of the theory of Squire and Zola [9], who argue that the individual structures of the medial temporal lobe contribute to many aspects of declarative memory including both semantic and episodic memory. A key question in this review concerns whether these new theories can be extended to a variety of non-human animals. Both the episodic and relational theories emphasize the idea that the hippocampus is important for episodic memory. While this idea is strongly supported by findings in the human literature, recent developments in the bird literature have begun to provide ways to test episodic-like memories in other species. In contrast, only the relational theory of Eichenbaum and colleagues suggests that the hippocampus also plays an important role in other forms of declarative memory, namely semantic memory, which includes some forms of recognition and associative memory. According to the relational theory, spatial tasks may be particularly good markers of hippocampal function because their performance depends on the ability to form relational representations between stimuli, even though they do not necessarily require episodic recall (i.e. recall of what, where and when). As we shall see in the following sections, empirical evidence is most consistent with the relational theory. Moreover, findings suggest that the dependence of a given memory task on hippocampal function may be particularly clear when the task is designed with an awareness of the ethological relevance of the spatial, episodic, or relational memory to the animal in question. Empirical findings Recent experimental findings have provided important insights into three aspects of hippocampal function. The first concerns a long-standing debate over the role of the hippocampus in recognition memory. The second concerns the role of the human and non-human primate hippocampus in spatial representation, and the third concerns the development of episodic-like memory tasks in animals. The hippocampus and recognition memory Recognition memory underlies our ability to detect that a stimulus has been experienced previously. Although it is widely accepted that the perirhinal and entorhinal cortices surrounding the hippcampus are important for recognition memory [10 12], the role of the hippocampus itself in this form of memory has been controversial. The most common task used to test recognition memory in animals is the delayed-nonmatching-to-sample (DNMS) task. In this task, animals are first presented with a sample object. Following a variable delay interval, animals are then given a choice between the sample object and a novel object.

3 770 Neurobiology of behaviour They must choose the novel (i.e. non-matching) object to receive a food reward. One convincing study showed that selective ibotenate lesions of the hippocampus and amygdala in monkeys produced no impairment on the DNMS task even when delay intervals were 40 min long [13]. By contrast, a recent convergence of studies using the DNMS task in monkeys [14,15,16], or analogous tasks in rats [17,18] and humans [19], shows that hippocampal lesions produce recognition memory impairments that are most severe at the longest delay intervals tested. How can we reconcile these contradictory results? One theoretical perspective suggests that recognition tasks can be solved in at least two different ways [20] either by recollecting the stimulus or event, thought to be dependent on the hippocampus [21 ], or by detecting stimulus familiarity, thought to be dependent on the surrounding cortex. Thus, performance on DNMS tasks need not involve episodic recognition [22 ]. Other key issues include the specific training history of the animals, differences in the way the lesions have been made across studies, and the sensitivity of the DNMS task to selective hippocampal lesions. For example, Murray and Mishkin [13] gave their animals several hundred training trials on the task before the lesions were made in a two-step procedure. Both pre-operative training and two-stage lesion techniques are known to produce milder memory deficits than post-operative training and single-stage lesions [14 ]. In contrast, in the other animal studies, single-stage, bilateral lesions were performed and all training was done post-operatively. Why is the deficit milder with preoperative training and two-staged lesions? One possibility is that these procedures bias the animals towards a nonhippocampal strategy (i.e. a strategy dependent on the surrounding cortex or a habit-like strategy dependent on the neostriatum), which would remain intact after the lesion and therefore not affect performance. In contrast, with post-operative training and single-stage lesions, animals may use a more hippocampal-based strategy, resulting in impaired performance. Yet another important issue to consider is the sensitivity of the DNMS task itself. Although Murray and Mishkin [13] relied on the DNMS task to test recognition memory, both Zola et al. [14 ] and Pascalis et al. [16] report that animals with hippocampal damage are even more impaired on a visual paired comparison (VPC) task than on the DNMS task. The VPC task is a recognition task that taps the animal s natural tendency to examine novel versus familiar visual stimuli by measuring eye-movement responses. Unlike the DNMS task in which animals must be trained to displace the sample and test objects, the natural eyemovement responses used in the VPC task do not require training. In the VPC task, monkeys are first given two identical novel pictures to look at (i.e. a sample picture). Following a variable delay interval, they can choose to look at either the sample picture or a novel picture. Normal animals spend more time looking at the novel picture even after long delay intervals. In contrast, animals with hippocampal lesions do not show this novelty preference if the delay interval is longer than a few seconds. Similarly, when rats with hippocampal lesions are tested on a task of spatial recognition memory (a task considered by many to be highly ethologically significant in rats), they are severely impaired [18]. More recently, it has been shown that lesioned rats are also severely impaired on a variety of ethologically motivated tasks in which they dig in odorized sand to find food rewards [23] or identify a safe food after a single exposure to the odor of the food on the breath of another (healthy) rat [24]. Like tasks of spatial memory, these latter tasks also tap the exquisite natural olfactory abilities of rats. Taken together, these studies make two important points. First, the findings suggest that the hippocampus plays a key role in many forms of recognition memory. The involvement of the hippocampus in recognition may depend on a complex set of factors including the type of lesion, the training history and the particular strategy the animal is using to solve the task. Second, the findings point to the trend in the literature showing that recognition tasks that are naturally acquired by the animal more clearly reveal the dependence of these tasks on the hippocampus. Below, we will explore in more detail the usefulness of including the ethological relevance of tasks as a factor in studies of hippocampal function. The hippocampus and space The hippocampus has been heavily implicated in solving spatial tasks in rodents and birds [25], yet it is also clear that primates have highly developed spatial abilities. In the laboratory, there have been some exciting new developments in testing spatial/topographic memory in a more realistic way in humans and monkeys. In humans, for example, virtual reality tasks have been developed for use in positron emission tomography (PET) or functional magnetic resonance imaging (fmri) studies to probe spatial memory and navigational ability. The major advantage of these virtual reality tasks is that they capture more accurately the dynamic aspects of navigation compared to traditional table-top tasks of memory, yet maintain some degree of experimental control that cannot be captured when subjects are simply asked to recall the spatial layout of their neighborhood [26] (but see [27 ]). Maguire et al. [26] used PET to assess hippocampal activation while subjects navigated in a familiar yet complex virtual reality town either by relying on their memory for the spatial layout of the environment or by following a trail of arrows. The right hippocampus was more active during navigation than during trail-following, and there was a highly positive correlation between accurate knowledge of where places were located and activation of the right hippocampus. Unlike previous studies using computer touch screens or traditional stationary test boxes, recent developments in the study of spatial representation in the monkey hippocampus

4 The hippocampus and memory Suzuki and Clayton 771 have begun to examine spatial activity in hippocampal neurons as monkeys walk [28,29 ] or drive [30 ] around the environment. One of the two groups reported spatially selective responses in a significantly higher number of neurons during the locomotion version of the task than during an analogous computer-screen version of the task [30 ]. Using a different experimental protocol, however, the other group reported similar numbers of spatially selective neurons during the locomotion task [29 ] as have been reported on computer-screen-based spatial tasks [31]. Although these studies suggest that more realistic spatial tasks can sometimes engage larger numbers of hippocampal neurons, further studies will be needed to confirm this observation. As is becoming increasingly clear, the knowledge of which tasks are particularly ethologically relevant to the animal is extremely useful in studying the role of brain areas. One of the most striking examples of this comes from comparative studies of different species. These studies show that hippocampal volume is increased in mammalian and avian species that depend critically on spatial memory for survival. Examples of such critical memory skills include home-range navigation, migration, brood parasitism, and memory-based cache recovery in birds that hide food [32]. Thus, the more neuroethologically relevant the spatial tasks are to the particular animal, the more the hippocampus can be shown to be prominent and to play an important role in these tasks. Such specialized animals may be particularly useful for revealing basic features of hippocampal function across all animals. Moreover, there may be a direct link between the performance of spatial memory tasks and hippocampal size. In a series of experiments in food-storing birds, Clayton and colleagues have shown that the hippocampus of titmice and chickadees increases in volume in association with the experience of storing and recovering food caches [33,34]. This experience-dependent hippocampal growth occurs at a relatively late stage in development, after the young have fledged from the nest. Furthermore, recent studies show that post-fledging chickadees require experience of both storing and cache recovery not only to initiate normal hippocampal growth but also to maintain hippocampal size [35]. A similar experience-dependent relationship between hippocampal size and navigational ability has also recently been suggested in humans. Maguire et al. [36 ] tested the hypothesis that the ability of taxi cab drivers to navigate the complex streets of London correlates with hippocampal size. London taxi drivers had larger posterior hippocampi and smaller anterior hippocampi than agematched control subjects. It remains to be seen whether navigational learning is directly correlated with the structural change in the hippocampus because control subjects may have differed from the taxi drivers in a number of respects, including lifestyle and the time spent in London. Nonetheless, the results raise the intriguing possibility of hippocampal plasticity in the healthy adult human brain in response to environmental demands. This hypothesis can be tested directly in experimental animals. Testing episodic memory in animals Most animal tasks have not distinguished between familiarity-based recognition and episodic recall. Recently, however, Clayton and Dickinson devised a way to test episodic-like memory using food-caching birds as the model system. Food-caching birds hoard food throughout their winter territories when food is abundant only to retrieve it days, weeks and even months later when food is scarce. A wealth of evidence from both the field and the laboratory shows that spatial memory plays an important role in cache retrieval [37], an ability that depends upon the hippocampus [38]. Some species, including jays, cache perishable items (e.g. worms) in addition to seeds. It therefore may be adaptive for these species to encode and recall information about when a particular food item was cached and what type of food was cached, as well as where the food was hidden. In a series of laboratory experiments, Clayton and Dickinson [39,40 ] showed that jays remember a series of facts about an object (the food item), a place (where they stored it), a time (how long it was since they stored the item) and an action (food-storing versus cache recovery) based on a single caching experience. When these facts are integrated, the jay has sufficient information to recall the episode of caching [22 ]. The ability to remember the what, where, and when of an event on the basis of a single past experience is termed episodic-like memory [22,39] because the results fulfill the behavioral criteria for episodic memory [2]. An exciting prospect for the future is the development of other animal models of episodic-like memory tasks based on the food-caching paradigm in birds. This could be achieved either by testing natural food-caching memory in other animals such as rats [37], or by having an experimenter hide food and testing the subjects ability to remember where that food was hidden. The latter model could be used to investigate the development of episodic memory in a variety of different animal species and also in humans. The development of an effective test of episodic memory in language-impaired patients opens up the possibility of distinguishing between neurological deficits that cause language impairments and those resulting in a disruption of episodic memory. The neuroethology of the hippocampus Although the hippocampus is involved in a wide range of mnemonic functions, three defining features can be identified. These features include the ability to form new, complex memories in a single trial, to flexibly update those memories, and to extract common information between experiences. Across many species, there are ethologically motivated tasks that share these defining features, including the VPC task in monkeys [14 ], social learning of food odors in rats [24], food caching in jays [39,40 ] and episodic memories in humans [3]. Hippocampal-dependent

5 772 Neurobiology of behaviour memories also include the ability to flexibly update memory on the basis of current experience. This also applies to navigation, which has long been known to be highly sensitive to hippocampal damage in animals [41]. Recent functional imaging studies have confirmed the important role of the hippocampus for navigation in humans [26]. For the same reasons, food caching tasks [38] in which animals must update their memory in a flexible way are also highly sensitive to hippocampal lesions. Finally, the ability to extract common information across trials may be critically dependent on hippocampal function. This idea is supported by the fact that selective hippocampal lesions in adulthood are often associated with semantic memory impairments [42]. Conclusions In conclusion, this brief survey of the current literature suggests that the growing body of more-ethologically based memory tasks in animals tap some of the same key features emphasized in current theories of hippocampal function. For example, the episodic theory of hippocampal memory function suggests that the hippocampus is essential for memories of the what, where and when of common every day events. Tasks designed with the ethological significance of the animal in mind may create a more naturalistic event for the subjects under study to remember and in this way may be particularly well suited to engage hippocampal memory function. Similarly, consistent with the relational theory of hippocampal function, several of the more ethologically motivated memory tasks include the requirement of forming fast and flexible relational representations between stimuli (i.e. spatial/topographic memory). For these reasons, we believe that an ethological perspective is critical for understanding the role of the hippocampus in declarative memory [43]. The use of highly specialized tasks of known ethological relevance may help us to understand how the hippocampus is involved in memory and what distinguishes this type of memory from non-hippocampal memory. Acknowledgments The authors wish to thank Larry Squire, Anthony Dickinson, Nathan Emery and the editors for their very helpful comments on this manuscript. References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest of outstanding interest 1. Scoville WB, Milner B: Loss of recent memory after bilateral hippocampal lesions. J Neurol Neurosurg Psych 1957, 20: Tulving E: Episodic and sematic memory. In Organization of Memory. Edited by Tulving E, Donaldson W. New York: Academic Press; 1972: Vargha-Khadem F, Gadian DG, Watkins KE, Connelly A, Van Paesschen W, Mishkin M: Differential effects of early hippocampal pathology on episodic and semantic memory. Science 1997, 277: Mishkin M, Suzuki WA, Gadian DG, Vargha-Khadem F: Hierarchical organization of cognitive memory. Philos Trans R Soc Lond B Biol Sci 1997, 352: Parkinson JK, Murray EA, Mishkin M: A selective mnemonic role for the hippocampus in monkeys: memory for the location of objects. J Neurosci 1988, 8: Tulving E, Markowitsch HJ: Episodic and declarative memory: role of the hippocampus. Hippocampus 1998, 8: Eichenbaum H, Dudchenko P, Wood E, Shapiro M, Tanila H: The hippocampus, memory, and place cells: is it spatial memory or a memory space? Neuron 1999, 23: An excellent review of research on hippocampal place cells, which describes the firing properties of these neurons and the role they may play in memory. In this paper, the authors call into question the theory that the hippocampus is involved in forming complex spatial maps (i.e. the cognitive mapping theory) and argue that the hippocampus is involved in a broader class of information processing that supports the formation and organization of relational representations between all kinds of stimuli, including between episodes. They refer to the set of linked stimuli or episodes as a memory space. 8. Eichenbaum H: Cortical-hippocampal networks for declarative memory. Nat Neurosci Rev 2000, in press. 9. Squire LR, Zola SM: Episodic memory, semantic memory and amnesia. Hippocampus 1998, 8: Meunier M, Bachevalier J, Mishkin M, Murray EA: Effects on visual recognition of combined and separate ablations of the entorhinal and perirhinal cortex in rhesus monkeys. J Neurosci 1993, 13: Leonard BW, Amaral DG, Squire LR, Zola-Morgan S: Transient memory impairment in monkeys with bilateral lesions of the entorhinal cortex. J Neurosci 1995, 15: Suzuki WA, Zola-Morgan S, Squire LR, Amaral DG: Lesions of the perirhinal and parahippocampal cortices in the monkey produce long-lasting memory impairment in the visual and tactual modalities. J Neurosci 1993, 13: Murray EA, Mishkin M: Object recognition and location memory in monkeys with excitotoxic lesions of the amygdala and hippocampus. J Neurosci 1998, 18: Zola SM, Squire LR, Teng E, Stefanacci L, Buffalo EA, Clark RE: Impaired recognition memory in monkeys after damage limited to the hippocampal region. J Neurosci 2000, 20: This paper examines the performance of 22 animals with hippocampal lesions on two recognition memory tasks the DNMS task with trial-unique objects, and the VPC task. Although animals with hippocampal lesions were impaired on both tasks, their impairment was even more striking on the VPC task which takes advantage of the animal s natural recognition capabilities. 15. Beason-Held L, Rosene DL, Killainy RJ, Moss MB: Hippocampal formation lesions produce memory impairment in the rhesus monkey. Hippocampus 1999, 9: This paper reports that even very small bilateral hippocampal lesions can significantly affect performance on the DNMS task as well as on the spatial, color and object versions of the delayed recognition span task. These findings therefore support the idea that hippocampal lesions can cause significant recognition memory impairments. 16. Pascalis O, Bachevalier J: Neonatal aspiration lesions of the hippocampal formation impair visual recognition memory when assessed by paired-comparison task but not by delayed nonmatching-to-sample task. Hippocampus 1999, 9: Dudchenko PA, Wood ER, Eichenbaum H: Neurotoxic hippocampal lesions have no effect on odor span and little effect on odor recognition memory but produce significant impairments on spatial span, recognition and alternation. J Neurosci 2000, 20: Hampson RE, Jarrard LE, Deadwyler SA: Effects of ibotenate hippocampal and extrahippocampal destruction on delayedmatch and -nonmatch-to-sample behavior in rats. J Neurosci 1999, 19: Reed JM, Squire LR: Impaired recognition memory in patients with lesions limited to the hippocampal formation. Behav Neurosci 1997, 111: Mandler G: Recognising: the judgement of previous experience. Psychol Rev 1987, 87: Aggleton JP, Brown MW: Episodic memory, amnesia and the hippocampal anterior thalamic axis. Behav Brain Sci 1999, 22: An interesting if controversial paper that describes a theory for how the medial temporal lobe and medial diencephalic structures contribute to memory. The authors argue that while the hippocampal system is critical for

6 The hippocampus and memory Suzuki and Clayton 773 episodic memory recall, it is not important for familiarity-based recognition memory. In contrast, they argue that the perirhinal cortex and medial dorsal nucleus of the thalamus are critical for familiarity-based recognition, but not episodic memory recall. 22. Griffiths DP, Dickinson A, Clayton NS: Declarative and episodic memory: what can animals remember about their past? Trends Cogn Sci 1999, 3: In this article, the authors describe the critical features of episodic memory in humans and its relationship with declarative memory; they also address the question of whether episodic memory is restricted to humans. They argue that recent experiments on the memory capabilities of food-caching jays suggest that these birds show some of the critical features of episodic recall (the what, where and when of an event), which they term episodic-like memory. 23. Bunsey M, Eichenbaum H: Conservation of hippocampal memory functions in rats and humans. Nature 1996, 379: Bunsey M, Eichenbaum H: Selective damage to the hippocampal region blocks long term retention of a natural and nonspatial stimulus stimulus association. Hippocampus 1995, 5: Nadel L: The hippocampus and space revisited. Hippocampus 1991, 1: Maguire EA, Burgess N, Donnett JG, Frackowiak RS, Frith CD, O Keefe J: Knowing where and getting there: a human navigation network. Science 1998, 280: Teng E, Squire LR: Memory for places learned long ago is intact after hippocampal damage. Nature 2000, 400: This paper describes the spatial memory capabilities of a deeply amnesic patient (EP) who has extensive damage to the hippocampus and adjacent medial temporal lobe structures. Although EP has no knowledge of his current neighborhood, his memory for places learned long ago, prior to the onset of amnesia, is intact. The authors conclude that the medial temporal lobe is essential for the formation, but not the retrieval, of remote memories. 28. Robertson RG, Rolls ET, Georges-Francois P: Spatial view cells in the primate hippocampus: effects of removal of view details. J Neurophysiol 1998, 79: George-Francois P, Rolls ET, Robertson RG: Spatial view cells in the primate hippocampus: allocentric view not head direction or eye position. Cereb Cortex 1999, 9: Previous studies have shown that some hippocampal neurons respond selectively whenever the animal views a particular part of the environment irrespective of the monkey s absolute position in the environment. This study further shows that neither head direction nor eye position can explain the selectivity of these neurons. Instead, the authors argue that these neurons represent an allocentric view of a particular part of the world. Thus, these spatial view cells may represent the memory of that particular part of the world. 30. Matsumura N, Nishijo H, Tamura R, Eifuku S, Endo S, Ono T: Spatial- and task-dependent neuronal responses during real and virtual translocation in the monkey hippocampal formation. J Neurosci 1999, 19: In this study, monkeys drove around in a motorized cab as single-unit activity in the hippocampus was recorded. Place-cell-like activity was observed in the monkey hippocampus in that some neurons increased their firing rate when the animal passed through a particular location in the environment irrespective of the direction of motion. 31. Rolls ET, Miyashita Y, Cahusac PM, Kesner RP, Niki H, Feigenbaum JD, Bach L: Hippocampal neurons in the monkey with activity related to the place in which a stimulus is shown. J Neurosci 1989, 9: Lee DW, Miyasato L, Clayton NS: Neurobiological basis of spatial learning in the natural environment: neurogenesis and growth in the avian and mammalian hippocampus. Neuroreport 1998, 9: Clayton NS, Krebs JR: Hippocampal growth and attrition in birds affected by experience. Proc Natl Acad Sci USA 1994, 91: Clayton NS, Soha J: Memory in avian food-storing and song learning: a general mechanism or different processes? Adv Study Behav 1999, 28: Banta Lavenex PA, Clayton NS: Caching experience is necessary for maintaining hippocampal size as well as triggering hippocampal growth in food-caching mountain chickadees. Soc Neurosci Abstr 1999, 25: Maguire EA, Gadian DG, Johnsrude IS, Good CD, Ashburner J, Frackowiak RS, Frith CD: Naviation-related structural changes in the hippocampus of taxi drivers. Proc Natl Acad Sci USA 2000, 98: This is an intriguing study that compares structural MRI brain scans of experienced, licensed taxi drivers and control subjects who did not drive taxis. The authors found that taxi drivers had significantly larger posterior hippocampi and significantly smaller anterior hippocampi than non-taxi drivers. The amount of time spent as a taxi driver correlated positively with the size of the posterior hippocampus and negatively with the size of the anterior hippocampus. These results suggest that an enlargement of the posterior hippocampus may be particularly important in people that depend heavily on using their navigational skills, and raise the intriguing possibility of experience-dependent structural plasticity in the adult human brain. 37. Vander Wall SB: Food Hoarding in Animals. Chicago: University of Chicago Press; Sherry DF, Vaccarino AL: Hippocampus and memory for food caches in black-capped chickadees. Behav Neurosci 1989, 103: Clayton NS, Dickinson A: What where and when: evidence for episodic-like memory during cache recovery by scrub jays. Nature 1998, 395: Clayton NS, Dickinson A: Memory for the content of caches by scrub jays. J Exp Psychol Anim Behav Process 1999, 25: This study shows that scrub jays can not only remember the contents and location of their caches, but also can update their memory on the basis of what they have recovered from a cache site. Furthermore, the results of this experiment suggest that jays remember the specific foods that they cache and recover in a way that cannot be explained in terms of simple associations between the food and their cache locations. 41. O Keefe J, Nadel L: The Hippocampus as a Cognitive Map. New York: Oxford University Press; Hamann SB, Squire LR: On the acquisition of new declarative knowledge in amnesia. Behav Neurosci 1995, 109: Gerlai R, Clayton NS: Analysing hippocampal function in transgenic mice: an ethological perspective. Trends Neurosci 1999, 22:47-51.