Spatial learning and the 'inner GPS' in the mammalian brain Flashcards
How do animals learn spatial locations and routes?
• WM - remember places within a trial (e.g. remember which arms already visited)
• Reference memory - remember places between trials (e.g. remember which arms contain food)
• Rats learn spatial r’ships between arms and external landmarks rather then following a rule or marking visited arms
• Which tests/evidence required for this conclusion?
Rotate the maze - change config of external landmarks
see notes
How do animals learn spatial locations and routes? research
Olton and Samuelson (1976)
Barnes (1988)
Olton and Samuelson (1976)
maze had barriers so couldn’t see rest of room
Tested 24 male albino rats on an 8-arm maze in a paradigm of sampling with replacement from a known set of items until the entire set was sampled. Exps I-III demonstrate that Ss performed efficiently, choosing an average of more than 7 different arms within the 1st 8 choices, and did not utilize intramaze cues or consistent chains of responses in solving the task. Exps IV-VI examined some characteristics of Ss’ memory storage. There was a small but reliable recency effect with the likelihood of a repetition error increasing with the number of choices since the initial instance. This performance decrement was due to interference from choices rather than just to the passage of time. No evidence was found for a primary effect. The data also suggest that there was no tendency to generalize among spatially adjacent arms. Results are discussed in terms of the memory processes involved in this task and human serial learning
Barnes (1988)
There are a number of ways to approach the problem of how nervous systems are modified in response to an organism’s interaction with its environment. One of these has been the study of learning and memory processes in the rat and their underlying physiological mechanisms, an endeavor that has contributed significantly to the overall understanding of the neural basis of behavior. As an example, the neurobiological properties of the rat hippocampal formation in relation to spatial information processing are reviewed, including a variety of behavioral analyses in conjunction with lesion and electrophysiological recording techniques. These approaches have furthered our understanding of cognitive operations that involve the integration of multiple sensory stimuli leading to the production of complex adaptive responses.
Hippocampal lesions impair spatial learning
• Humans
○ Declarative (facts) memories can be lost with hippocampal damage, but not procedural (skills) memories - often old memories are not lost
• Animals
○ Hippocampus imp for acquisition of new info and spatial learning
Control condition in hippocampus. Lesion studies - non-spatial learning (cued learning) - shows can perf other types of learning but doesn’t form spatial memories
see notes
Hippocampal lesions impair spatial learning research
Barnes (1979)
Morris (2008)
O’Keefe and Nadal (1979)
Ruskin et al. (2004)
Barnes (1979)
○ Circular platform maze
- Neurophysiological and behavioral measures were obtained from 32 senescent (28–34 mo) and 32 mature adult (10–26 mo) Long-Evans hooded rats. Extracellularly recorded synaptic responses were obtained from electrodes chronically implanted in the fascia dentata and perforant path. Ss were first tested on a circular platform, which favored the use of spatial cues for its solution; the senescent Ss exhibited poorer memory for the rewarded place. When granule cell synaptic responses were recorded after a single session of very brief high-frequency stimulation, the amount of elevation and time course of decline were equivalent between age groups. After 3 repetitions, however, young Ss maintained the increased synaptic strength for at least 14 days, whereas old Ss declined after the 1st session. The amount of synaptic enhancement was statistically correlated with the ability to perform the circular platform task both within and between groups. Furthermore, the aftereffects of the high-frequency stimulation selectively impaired the old Ss’ spontaneous alternation behavior on a –T maze. Results are discussed in terms of the synaptic theory of memory formation and the aging process.
Morris (2008)
TheMorris water mazeis one of the most widely used tasks inbehavioral neuroscience for studying the psychological processes and neural mechanismsof spatial learning andmemory. The basic task is very simple. Animals, usually rats or mice, are placed in a large circular pool of water and required to escape from water onto a hidden platform whose location can normally be identified only using spatial memory ( Figure1). There are no local cues indicating where the platform is located. Conceptually, the task derives fromplace cellsthat areneuronsin thehippocampuswhich identify or represent points in space in an environment (O’Keefe, 1976).
It was developed by Richard Morris at the University of St Andrews in Scotland and first described in two publications in the early 1980s (Morris, 1981; Morris et al., 1982). Place navigation in the watermaze is now often used as a general assay of cognitive function (Brandeis et al., 1989), for example for testing the impact of various disturbances of thenervous system(e.g. animalmodels of stroke(Nunn et al., 1994), aging (Gallagher and Rapp, 1997), neurodegenerative disease (Hsiao et al., 1996), or the potential impact of novel therapeutic drugs (D’Hooge and De Deyn, 2001). The task has also inspired computational neuroscientists and roboticists interested in navigation (Krichmar et al., 2005).
O’Keefe and Nadal (1979)
○ Theories of spatial cognition are derived from many sources. Psychologists are concerned with determining the features of the mind which, in combination with external inputs, produce our spatialized experience. A review of philosophical and other approaches has convinced us that the brain must come equipped to impose a three-dimensional Euclidean framework on experience – our analysis suggests that object re-identification may require such a framework. We identify this absolute, nonegocentric, spatial framework with a specific neural system centered in the hippocampus.
○ A consideration of the kinds of behaviours in which such a spatial mapping system would be important is followed by an analysis of the anatomy and physiology of this system, with special emphasis on the place-coded neurons recorded in the hippocampus of freely moving rats. A tentative physiological model for the hippocampal cognitive map is proposed. A review of lesion studies, in tasks as diverse as discrimination learning, avoidance, and extinction, shows that the cognitive map notion can adequately explain much of the data.
- The model is extended to humans by the assumption that spatial maps are built in one hemisphere, semantic maps in the other. The latter provide a semantic deep structure within which discourse comprehension and production can be achieved. Evidence from the study of amnesic patients, briefly reviewed, is consistent with this extension
Ruskin et al. (2004)
Prolonged sleep deprivation results in cognitive deficits. In rats, for example, sleep deprivation impairs spatial learning and hippocampal long‐term potentiation. We tested the effects of sleep deprivation on learning in a Pavlovian fear conditioning paradigm, choosing a sleep deprivation paradigm in which REM sleep was completely prevented and non‐REM sleep was strongly decreased. During conditioning, rats were given footshocks, either alone or paired with a tone, and tested 24 h later for freezing responses to the conditioning context, and to the tone in a novel environment. Whereas control animals had robust contextual learning in both background and foreground contextual conditioning paradigms, 72 h of sleep deprivation before conditioning dramatically impaired both types of contextual learning (by more than 50%) without affecting cued learning. Increasing the number of footshocks did not overcome the sleep deprivation‐induced deficit. The results provide behavioural evidence that REM/non‐REM sleep deprivation has neuroanatomically selective actions, differentially interfering with the neural systems underlying contextual learning (i.e. the hippocampus) and cued learning (i.e. the amygdala), and support the involvement of the hippocampus in both foreground and background contextual conditioning
Navigation (Wolf, 2011)
• Getting from one point to another point
• Where am I? - reference to an abstract map, allocentric representation, allows to plan for novel route without learning it
• How do I get from here to a goal location? - following learned routes, novel routes link to learned information, egocentric and allocentric representations
• Multimodal sensory info used
• Cognitive map: using visual allocentric cues in an object-centred reference frame to infer direction and distance - no other animal can do it as well as we can
• View-matching: inferring direction and distance from views that are matched with memorised views in an egocentric frame of reference (e.g. retinotopic maps) - everything in relationship to your body
• Path integration (dead reckoning): updating location and directional orientation by recording idiothetic cues, over long distances (turns, no. of steps, odometry); prone to cumulative error
• Path integration in the desert ant Cataglyphis
○ Has to run v. fast
○ High legs because of heat
○ Has to have well adapted navigation system
○ Have to search and find food - don’t have many landmarks - use path integration - count number of steps - goes in straight line home - knows where it is because of path integration
- Path integration v. inaccurate - larger distance = larger error
see notes
Wolf (2011)
Animals have needed to find their way about almost since a free-living life style evolved. Particularly, if an animal has a home – shelter or nesting site – true navigation becomes necessary to shuttle between this home and areas of other activities, such as feeding. As old as navigation is in the animal kingdom, as diverse are its mechanisms and implementations, depending on an organism’s ecology and its endowment with sensors and actuators. The use of landmarks for piloting or the use of trail pheromones for route following have been examined in great detail and in a variety of animal species. The same is true for senses of direction – the compasses for navigation – and the construction of vectors for navigation from compass and distance cues. The measurement of distance itself – odometry – has received much less attention. The present review addresses some recent progress in the understanding of odometers in invertebrates, after outlining general principles of navigation to put odometry in its proper context. Finally, a number of refinements that increase navigation accuracy and safety are addressed.
Pointing tasks (Ekstrom et al., 2014)
• Accuracy can be increased in a JRD task
• Humans may rely on allocentric knowledge for some tasks
• Most ethological situs can be solved with both, and there could be continuum how each contributes
• SOP - visual cues - use egocentric info - primed by presence of objects
• JRD - more accurate as given more reference points - allocentric
Difficult to tell the 2 apart
see notes
Pointing tasks (Ekstrom et al., 2014) research
Epstein and Vass (2014)
Epstein and Vass (2014)
Humans and animals use landmarks during wayfinding to determine where they are in the world and to guide their way to their destination. To implement this strategy, known as landmark-based piloting, a navigator must be able to: (i) identify individual landmarks, (ii) use these landmarks to determine their current position and heading, (iii) access long-term knowledge about the spatial relationships between locations and (iv) use this knowledge to plan a route to their navigational goal. Here, we review neuroimaging, neuropsychological and neurophysiological data that link the first three of these abilities to specific neural systems in the human brain. This evidence suggests that the parahippocampal place area is critical for landmark recognition, the retrosplenial/medial parietal region is centrally involved in localization and orientation, and both medial temporal lobe and retrosplenial/medial parietal lobe regions support long-term spatial knowledge.
What about mice? (Rinaldi et al., 2020)
• Allocentric navigation not only dependent on hippocampus but also distributed neural circuits (dorsomedial striatum, nucleus Accumbens, prelimbic and infralimbic cortex)
- Retrieval of allocentric and egocentric info mediated by distinct neural systems - activate diff neural networks
see notes
What about mice? (Rinaldi et al., 2020) research
Hok et al. (2016)
Hok et al. (2016)
The increasing use of mice models in cognitive tasks that were originally designed for rats raises crucial questions about cross‐species comparison in the study of spatial cognition. The present review focuses on the major neuroethological differences existing between mice and rats, with particular attention given to the neurophysiological basis of space coding. While little difference is found in the basic properties of space representation in these two species, it appears that the stability of this representation changes more drastically over time in mice than in rats. We consider several hypotheses dealing with attentional, perceptual, and genetic aspects and offer some directions for future research that might help in deciphering hippocampal function in learning and memory processes.
Egocentric v allocentric representations? Both exist in humans in some form (Nadel and Hardt, 2004; Burgess, 2006; Galati et al., 2010; Filimon, 2015)
• Widely suggested that humans and mammals have cog maps based allocentric representations in brain
• Many brain areas map spatial location of objects in egocentric reference frame (e.g. relative to eye, head/hand), e.g. in parieto-frontal cortex
• Could allocentric representations be explained via egocentric spatial reference frames?
• Potential allocentric task effects:
○ Mental shift of object to centre it frontally (egocentric left-right decisions)
○ Mental rotation
View-dependent object or scene recognition
see notes
Nadal and Hardt (2004)
Themes emerging from the collection of articles in the Special Section on Long-Term Spatial Memory include the notion of multiple spatial systems, the relation between spatial representations and episodic memory, the role of context, and the neural systems involved in space. The authors conclude that distinguishing between egocentric and allocentric spatial systems makes sense of both behavioral and neurobiological data. The special role of the hippocampal system in allocentric space, and as a consequence, in context, suggests how a spatial system might end up central to the ability to remember episodes
Burgess (2006)
Recent experiments indicate the need for revision of a model of spatial memory consisting of viewpoint-specific representations, egocentric spatial updating and a geometric module for reorientation. Instead, it appears that both egocentric and allocentric representations exist in parallel, and combine to support behavior according to the task. Current research indicates complementary roles for these representations, with increasing dependence on allocentric representations with the amount of movement between presentation and retrieval, the number of objects remembered, and the size, familiarity and intrinsic structure of the environment. Identifying the neuronal mechanisms and functional roles of each type of representation, and of their interactions, promises to provide a framework for investigation of the organization of human memory more generally.
Galati et al. (2010)
We review human functional neuroimaging studies that have explicitly investigated the reference frames used in diVerent cortical regions for representing spatial locations of objects. Beyond the general distinction between “egocentric” and “allocentric” reference frames, we provide evidence for the selective involvement of the posterior parietal cortex and associated frontal regions in the speciWc process of egocentric localization of visual and somatosensory stimuli with respect to relevant body parts (“body referencing”). Similarly, parahippocampal and retrosplenial regions, together with speciWc parietal subregions such as the precuneus, are selectively involved in a speciWc form of allocentric representation in which object locations are encoded relative to enduring spatial features of a familiar environment (“environmental referencing”). We also present a novel functional magnetic resonance imaging study showing that these regions are selectively activated, whenever a purely perceptual spatial task involves an object which maintains a stable location in space during the whole experiment, irrespective of its perceptual features and its orienting value as a landmark. This eVect can be dissociated from the consequences of an explicit memory recall of landmark locations, a process that further engages the retrosplenial cortex.
Filimon (2015)
The use and neural representation of egocentric spatial reference frames is well-documented. In contrast, whether the brain represents spatial relationships between objects inallocentric, object-centered, orworld-centeredcoordinates is debated. Here, I review behavioral, neuropsychological, neurophysiological (neuronal recording), and neuroimaging evidence for and against allocentric, object-centered, or world-centered spatial reference frames. Based on theoretical considerations, simulations, and empirical findings from spatial navigation, spatial judgments, and goal-directed movements, I suggest that all spatial representations may in fact be dependent on egocentric reference frames.
Hippocampus lesions prior to training do not specifically impair WM or reference memory, but spatial task (Morris et al., 1982)
• All rats showed same escape latency in second phase of exp - cue-based navigation - reversal to hidden platform in 3rd phase - rats with hippocampal lesions perf poorly again
• Lesions after training have less strong effects - hippocampus is not the site for permanent memory storage but important in consolidation
Still perform well with cortical lesions - spatial task
see notes
Hippocampus lesions prior to training do not specifically impair WM or reference memory, but spatial task (Morris et al., 1982) research
Clark et al. (2013)
Morris et al. (1982)
Electrophysiological studies have shown that single cells in the hippocampus respond during spatial learning and exploration1–4, some firing only when animals enter specific and restricted areas of a familiar environment. Deficits in spatial learning and memory are found after lesions of the hippocampus and its extrinsic fibre connections5,6following damage to the medial septal nucleus which successfully disrupts the hippocampal theta rhythm7, and in senescent rats which also show a correlated reduction in synaptic enhancement on the perforant path input to the hippocampus8. We now report, using a novel behavioural procedure requiring search for a hidden goal, that, in addition to a spatial discrimination impairment, total hippocampal lesions also cause a profound and lasting placenavigational impairment that can be dissociated from correlated motor, motivational and reinforcement aspects of the procedure
Clark et al. (2013)
Navigation depends on a network of neural systems that accurately monitor an animal’s spatial orientation in an environment. Within this navigation system are head direction (HD) cells which discharge as a function of an animal’s directional heading, providing an animal with a neural compass to guide ongoing spatial behavior. Experiments were designed to test this hypothesis by damaging the dorsal tegmental nucleus (DTN), a midbrain structure that plays a critical role in the generation of the rodent HD cell signal, and evaluating landmark based navigation using variants of the Morris water task. In Experiments 1 and 2, shams and DTN-lesioned rats were trained to navigate toward a cued platform in the presence of a constellation of distal landmarks located outside the pool. After reaching a training criteria, rats were tested in three probe trials in which (a) the cued platform was completely removed from the pool, (b) the pool was repositioned and the cued platform remained in the same absolute location with respect to distal landmarks, or (c) the pool was repositioned and the cued platform remained in the same relative location in the pool. In general, DTN-lesioned rats required more training trials to reach performance criterion, were less accurate to navigate to the platform position when it was removed, and navigated directly to the cued platform regardless of its position in the pool, indicating a general absence of control over navigation by distal landmarks. In Experiment 3, DTN and control rats were trained in directional and place navigation variants of the water task where the pool was repositioned for each training trial and a hidden platform was placed either in the same relative location (direction) in the pool or in the same absolute location (place) in the distal room reference frame. DTN-lesioned rats were initially impaired in the direction task, but ultimately performed as well as controls. In the place task, DTN-lesioned rats were severely impaired and displayed little evidence of improvement over the course of training. Together, these results support the conclusion that the DTN is required for accurate landmark navigation
Place cells in the hippocampus: encoding of an observer-independent spatial location (O’Keefe and Dostrovsky, 1971)
• Popns of neurons (extracellular recordings in freely moving rats) with diff spatial prefs in hippocampus - collectively cells form spatial maps
- Defined by spatial layout
see notes
Place cells in the hippocampus: encoding of an observer-independent spatial location (O’Keefe and Dostrovsky, 1971) research
Calton et al. (2003)
O’Keefe and Dostrovsky (1971)
Responses of cells in the dorsal hippocampus of rats to restraining tactile stimuli as a function of spatial orientation suggest that the hippocampus provides the rest of the brain with a spatial reference map. The activity of cells in such a map would specify the direction in which the s was pointing relative to environmental landmarks and the occurrence of particular tactile, visual, etc., stimuli while facing in that orientation. It is hypothesized that activation of those cells specifying a particular orientation together with a signal indicating movement or intention to move in space would tend to activate cells specifying adjacent or subsequent spatial orientations. In this way, the map would “anticipate” the sensory stimuli consequent to a particular movement.
Calton et al. (2003)
The occurrence of cells that encode spatial location (place cells) or head direction (HD cells) in the rat limbic system suggests that these cell types are important for spatial navigation. We sought to determine whether place fields of hippocampal CA1 place cells would be altered in animals receiving lesions of brain areas containing HD cells. Rats received bilateral lesions of anterodorsal thalamic nuclei (ADN), postsubiculum (PoS), or sham lesions, before place cell recording. Although place cells from lesioned animals did not differ from controls on many place-field characteristics, such as place-field size and infield firing rate, the signal was significantly degraded with respect to measures of outfield firing rate, spatial coherence, and information content. Surprisingly, place cells from lesioned animals were more likely modulated by the directional heading of the animal. Rotation of the landmark cue showed that place fields from PoS-lesioned animals were not controlled by the cue and shifted unpredictably between sessions. Although fields from ADN-lesioned animals tended to have less landmark control than fields from control animals, this impairment was mild compared with cells recorded from PoS-lesioned animals. Removal of the prominent visual cue also led to instability of place-field representations in PoS-lesioned, but not ADN-lesioned, animals. Together, these findings suggest that an intact HD system is not necessary for the maintenance of place fields, but lesions of brain areas that convey the HD signal can degrade this signal, and lesions of the PoS might lead to perceptual or mnemonic deficits, leading to place-field instability between sessions
Place field maps: signal place recognition (Muller et al., 1987)
• Diff cells encode diff locations
• Shape and size of firing fields vary
• Not all complex spike cells act like place cells (D)
• In novel env pref can change
- Single cell can encode for diff spatial locations in diff contexts firing with varied patterns - may have diff patterns in same env (e.g. lights switched on/off)
see notes
Place field maps: signal place recognition (Muller et al., 1987) research
Kinsky et al. (2018)