Interaction between piloting and beacon homing by rats in a swimming pool

The effects of spatial stability and cue type on spatial learning: Implications for theories of parallel memory systems

Some theories of spatial learning predict that associative rules apply under only limited circumstances. For example, learning based on a boundary has been claimed to be immune to cue competition effects because boundary information is the basis for the formation of a cognitive map, whilst landmark learning does not involve cognitive mapping. This is referred to as the cue type hypothesis. However, it has also been claimed that cue stability is a prerequisite for the formation of a cognitive map, meaning that whichever cue type was perceived as stable would enter a cognitive map and thus be immune to cue competition, while unstable cues will be subject to cue competition, regardless of cue type. In experiments 1 and 2 we manipulated the stability of boundary and landmark cues when learning the location of two hidden goals. One goal location was constant with respect to the boundary, and the other constant with respect to the landmark cues. For both cue types, the presence of distal orientation cues provided directional information. For half the participants the landmark cues were unstable relative to the boundary and orientation cues, whereas for the remainder of the participants the boundary was unstable relative to landmarks and orientation cues. In a second stage of training, all cues remained stable so that both goal locations could be learned with respect to both landmark and boundary information. According to the cue type hypothesis, boundary information should block learning about landmarks regardless of cue stability. According to the cue stability hypothesis, however, landmarks should block learning about the boundary when the landmarks appear stable relative to the boundary. Regardless of cue type or stability the results showed reciprocal blocking, contrary to both formulations of incidental cognitive mapping. Experiment 3 established that the results of Experiments 1 and 2 could not be explained in terms of difficulty in learning certain locations with respect to different cue types. In a final experiment, following training in which both landmarks and boundary cues signalled two goal locations, a new goal location was established with respect to the landmark cues, before testing with the boundary, which had never been used to define the new goal location. The results of this novel test of the interaction between boundary and landmark cues indicated that new learning with respect to the landmark had a profound effect on navigation with respect to the boundary, counter to the predictions of incidental cognitive mapping of boundaries.

From objects to landmarks: the function of visual location information in spatial navigation

Landmarks play an important role in guiding navigational behavior. A host of studies in the last 15 years has demonstrated that environmental objects can act as landmarks for navigation in different ways. In this review, we propose a parsimonious four-part taxonomy for conceptualizing object location information during navigation. We begin by outlining object properties that appear to be important for a landmark to attain salience. We then systematically examine the different functions of objects as navigational landmarks based on previous behavioral and neuroanatomical findings in rodents and humans. Evidence is presented showing that single environmental objects can function as navigational beacons, or act as associative or orientation cues. In addition, we argue that extended surfaces or boundaries can act as landmarks by providing a frame of reference for encoding spatial information. The present review provides a concise taxonomy of the use of visual objects as landmarks in navigation and should serve as a useful reference for future research into landmark-based spatial navigation.

Do rufous hummingbirds (Selasphorus rufus) use visual beacons?

Animals are often assumed to use highly conspicuous features of a goal to head directly to that goal ('beaconing'). In the field it is generally assumed that flowers serve as beacons to guide pollinators. Artificial hummingbird feeders are coloured red to serve a similar function. However, anecdotal reports suggest that hummingbirds return to feeder locations in the absence of the feeder (and thus the beacon). Here we test these reports for the first time in the field, using the natural territories of hummingbirds and manipulating flowers on a scale that is ecologically relevant to the birds. We compared the predictions from two distinct hypotheses as to how hummingbirds might use the visual features of rewards: the distant beacon hypothesis and the local cue hypothesis. In two field experiments, we found no evidence that rufous hummingbirds used a distant visual beacon to guide them to a rewarded location. In no case did birds abandon their approach to the goal location from a distance; rather they demonstrated remarkable accuracy of navigation by approaching to within about 70 cm of a rewarded flower's original location. Proximity varied depending on the size of the training flower: birds flew closer to a previously rewarded location if it had been previously signalled with a small beacon. Additionally, when provided with a beacon at a new location, birds did not fly directly to the new beacon. Taken together, we believe these data demonstrate that these hummingbirds depend little on visual characteristics to beacon to rewarded locations, but rather that they encode surrounding landmarks in order to reach the goal and then use the visual features of the goal as confirmation that they have arrived at the correct location.

The 36th Sir Frederick Bartlett Lecture: An associative analysis of spatial learning

The ability of animals to find important goals in their environment has been said to require a form of learning that is qualitatively different from that normally studied in the conditioning laboratory. Such spatial learning has been said to depend upon the construction of a global representation of the environment, and the acquisition of knowledge about the position of goals with reference to this representation is said to be unaffected by the presence of other cues or landmarks. To evaluate the first of these claims, experiments are described that investigated the extent to which the effects of training in one environment transfer to another. To evaluate the second claim, experiments are described that investigated whether cue competition effects normally found in conditioning studies can be found in spatial tasks. Overall, the results indicate that most of the phenomena of spatial learning can be explained by the principles of associative learning. The implications of the reported results for an understanding of the neural mechanisms of spatial learning are considered.

Blocking of spatial control by landmarks in rats

We investigated spatial blocking among landmarks in an open-field foraging task in rats. In Phase 1, rats were presented with A+ trials during which landmark (LM) A signaled the location of hidden food. In Phase 2, rats were given AX+ trials in which LM X served as a redundant spatial cue to the location of food. Additionally, BY+ trials were given as a within-subjects overshadowing-control procedure. At test, rats received nonreinforced presentations of LM X and LM Y on separate trials. Rats took longer to find the training goal location in the presence of LM X than of LM Y, thereby demonstrating that spatial control by LM X was blocked by prior learning with LM A. This constitutes the first evidence in rats for spatial blocking of one proximal landmark by another-approximating a conventional blocking design.

Mapping immediate-early gene activity in the rat after place learning in a water-maze: the importance of matched control conditions

The expression of two immediate-early genes (IEGs), Zif268 and c-Fos, was quantified in hippocampal subregions and related structures following spatial learning in the Morris water-maze. A critical feature was the novel control protocol alongside more standard controls, the purpose of which was to test whether hippocampal activity is set automatically when traversing an environment or whether it is dependent on reaching a specific goal using learning that requires the hippocampus (i.e. task dependent). The new control protocol (Procedural Task) made it possible to match swim time, swim distance and learning to escape from water with that of the experimental (Working Memory) group. Unlike the Working Memory group, the Procedural Task animals showed no evidence of learning the absolute platform location during the test session. While the Working Memory rats showed c-Fos increases relative to the Procedural Task controls in the frontal and parahippocampal cortices, hippocampal levels did not differ. Again, for Zif268 there was no evidence of a relative increase of hippocampal activity in the Working Memory group. In fact, hippocampal Zif268 showed evidence of a relative decrease, even though the spatial working memory task is hippocampal dependent. The study not only highlighted the shortcomings of other control procedures used in water-maze studies (free-swimming or home cage control), but also indicated that the expression of these IEGs in the hippocampus is not a direct predictor of explicit spatial location learning. Rather, the activity in combinations of regions, including prefrontal cortex, provides a stronger correlate of water-maze learning.

Distinct error-correcting and incidental learning of location relative to landmarks and boundaries

Associative reinforcement provides a powerful explanation of learned behavior. However, an unproven but long-held conjecture holds that spatial learning can occur incidentally rather than by reinforcement. Using a carefully controlled virtual-reality object-location memory task, we formally demonstrate that locations are concurrently learned relative to both local landmarks and local boundaries but that landmark-learning obeys associative reinforcement (showing "overshadowing" and "blocking" or "learned irrelevance"), whereas boundary-learning is incidental, showing neither overshadowing nor blocking nor learned irrelevance. Crucially, both types of learning occur at similar rates and do not reflect differences in levels of performance, cue salience, or instructions. These distinct types of learning likely reflect the distinct neural systems implicated in processing of landmarks and boundaries: the striatum and hippocampus, respectively [Doeller CF, King JA, Burgess N (2008) Proc Natl Acad Sci USA 105:5915-5920]. In turn, our results suggest the use of fundamentally different learning rules by these two systems, potentially explaining their differential roles in procedural and declarative memory more generally. Our results suggest a privileged role for surface geometry in determining spatial context and support the idea of a "geometric module," albeit for location rather than orientation. Finally, the demonstration that reinforcement learning applies selectively to formally equivalent aspects of task-performance supports broader consideration of two-system models in analyses of learning and decision making.

The effects of exercise on spatial learning and anxiety-like behavior are mediated by an IGF-I-dependent mechanism related to hippocampal neurogenesis

Knowledge about the effects of physical exercise on brain is accumulating although the mechanisms through which exercise exerts these actions remain largely unknown. A possible involvement of adult hippocampal neurogenesis (AHN) in the effects of exercise is debated while the physiological and pathological significance of AHN is under intense scrutiny. Recently, both neurogenesis-dependent and independent mechanisms have been shown to mediate the effects of physical exercise on spatial learning and anxiety-like behaviors. Taking advantage that the stimulating effects of exercise on AHN depend among others, on serum insulin-like growth factor I (IGF-I), we now examined whether the behavioral effects of running exercise are related to variations in hippocampal neurogenesis, by either increasing or decreasing it according to serum IGF-I levels. Mutant mice with low levels of serum IGF-I (LID mice) had reduced AHN together with impaired spatial learning. These deficits were not improved by running. However, administration of exogenous IGF-I ameliorated the cognitive deficit and restored AHN in LID mice. We also examined the effect of exercise in LID mice in the novelty-suppressed feeding test, a measure of anxiety-like behavior in laboratory animals. Normal mice, but not LID mice, showed reduced anxiety after exercise in this test. However, after exercise, LID mice did show improvement in the forced swim test, a measure of behavioral despair. Thus, many, but not all of the beneficial effects of exercise on brain function depend on circulating levels of IGF-I and are associated to increased hippocampal neurogenesis, including improved cognition and reduced anxiety.

Competition between landmarks in spatial learning: The role of proximity to the goal

In two experiments, rats were trained to find a hidden platform in a Morris pool in the presence of two landmarks. Landmark B was present on all training trials, on half the trials accompanied by landmark A, on the remainder by landmark C. For rats in Group Bn, B was near the location of the platform; for those in Group Bf, B was far from the platform. Group Bn performed better than Group Bf on test trials to B alone, but significantly worse on test trials to a new configuration formed by A and C. Thus, the spatial proximity of B to the platform affected not only how well it could be used to locate the platform, but also its ability to prevent learning about other landmarks.

Failure of a landmark to restrict spatial learning based on the shape of the environment

Rats were required to find a submerged platform in the corner of a swimming pool with a distinctive shape. A landmark near the platform did not interfere with the control acquired by the pool's shape over searching for the platform. This outcome was observed with an overshadowing and a blocking design. A comparison of the ease with which the landmark and the pool's shape gained control over searching for the platform indicates that the failure of overshadowing and blocking was not a consequence of the landmark being less salient than the shape of the pool. The results are not readily explained by theories of associative learning, but they are consistent with the claim that learning about the shape of the environment takes place in a dedicated module, which excludes information about the significance of individual landmarks.

Place Versus Response Learning Revisited: Tests of Blocking on the Radial Maze

Neurobiological and behavioral research indicates that place learning and response learning occur simultaneously, in parallel. Such findings seem to conflict with theories of associative learning in which different cues compete for learning. The authors conducted place+response training on a radial maze and then tested place learning and response learning separately by reconfiguring the maze in various ways. Consistent with the effects of manipulating place and response systems in the brain (M. G. Packard & J. L. McGaugh, 1996), well-trained rats showed strong place learning and strong response learning. Three experiments using associative blocking paradigms indicated that prior response learning interferes with place learning. Blocking and related tests can be used to better understand how memory systems interact during learning.

Sequential control of navigation by locale and taxon cues in the Morris water task

The neurobehavioral dissociation between place navigation and cued navigation has been central to contemporary thinking regarding the psychological processes involved in spatial behavior. In cases where locale (place) cues and taxon cues (e.g., beacons) are present it has been suggested that navigation may be controlled by either stimulus type in isolation, or, alternatively, by both simultaneously. In this report we provide evidence that place cues and beacons sequentially control navigation during a single trip to a visible goal. Rats were trained to navigate to a visible escape platform in a circular swimming pool surrounded by numerous visual cues and the kinematics and accuracy of the trajectories to the platform were analyzed. Shortly after initiating a trajectory to the visible platform, animals routinely engaged in stimulus sampling behaviors (e.g., horizontal head scans) which were consistently associated with changes in accuracy (heading error) and swim velocity. Subsequently, animals swam quickly and accurately to the visible platform suggesting that the sampling behaviors correspond to a shift in exteroceptive stimulus control. Consistent with this idea, removal or relocation of the platform disrupted navigation following the stimulus sampling behaviors, whereas the initial trajectory was unaffected. In contrast, changes in the distal cue constellation selectively disrupted the initial trajectory. The results showing that navigation to a visible goal is controlled sequentially by locale and taxon cues are discussed in relation to contemporary theories of navigation.

Absence of an Interaction Between Navigational Strategies Based on Local and Distal Landmarks

In 3 experiments, rats were required to escape from a Morris pool by swimming to a submerged platform that was located at the apex of a notional, equilateral triangle with 2 different landmarks occupying the corners at the base. Training for 1 group was always conducted in view of the landmarks surrounding the pool and with the triangular array in a fixed orientation. Subjects could therefore identify the direction of the platform from a single landmark within the pool by reference to cues outside the pool or to the other landmark within the pool. Both strategies were used, and the results from additional groups revealed that the first of these strategies did not affect the acquisition of the second one.

Competition among spatial cues in a naturalistic food-carrying task

Rats collected nuts from a container in a large arena in four experiments testing how learning about a beacon or cue at a goal interacts with learning about other spatial cues (place learning). Place learning was quick, with little evidence of competition from the beacon (Experiments 1 and 2). Rats trained to approach a beacon regardless of its location were subsequently impaired when the well-learned beacon was removed and other spatial cues identified the location of the goal (Experiment 3). The competition between beacon and place cues reflected learned irrelevance for place cues (Experiment 4). The findings differ from those of some studies of associative interactions between cue and place learning in other paradigms.

Absence of overshadowing and blocking between landmarks and the geometric cues provided by the shape of a test arena

In three experiments rats were required to escape from a pool of water by swimming to a submerged platform. The position of the platform was determined by the shape of the pool, which was either rectangular or triangular. A landmark that was located on the surface of the pool near the platform failed to overshadow (Experiment 1) or block (Experiment 2) learning about the position of the platform with reference to the shape of the pool. Experiment 3 revealed a similar outcome with cues outside the pool, which could be used, in addition to the shape of the pool, to identify the location of the platform. These findings imply that theories of learning that assume that stimuli must compete with each other for the control that they acquire may not apply to spatial learning based on the shape of the environment.

Acquisition of knowledge about spatial location: assessing the generality of the mechanism of learning

A selection of studies in the last 20 years is reviewed. These studies show basic Pavlovian phenomena in the spatial domain (like blocking, overshadowing, latent inhibition, and perceptual learning) with nonhuman subjects, specifically with rats, both in the radial maze and in the circular pool. The generality of these phenomena with respect to other species and to other spatial preparations is also discussed. The conclusion is that the mechanism responsible for the acquisition of knowledge about spatial location seems to be associative in nature.

Influence of a beacon on spatial learning based on the shape of the test environment

In 5 experiments rats were required to escape from a triangular shaped pool by swimming to a submerged platform. The principal group of interest in each experiment received training with a beacon attached to the platform. The purpose of the experiments was to assess if the beacon overshadowed (Experiments 1-4) or blocked (Experiment 5) learning about the position of the platform with reference to the shape of the pool. The platform was located in the center of the pool for the first 2 experiments and in a corner for the remaining experiments. Although there was an overshadowing effect in Experiment 1, the remaining experiments failed to reveal any disruptive influence of the beacon on learning based on the shape of the pool. Moreover, in Experiments 3-5 there was an indication that the beacon facilitated such learning. The results suggest that spatial learning based on the shape of a test environment may not take place in the same way as that based on more discrete landmarks.

Dissociation of place and cue learning by telencephalic ablation in goldfish

This study examined the spatial strategies used by goldfish (Carassius auratus) to find a goal in a 4-arm maze and the involvement of the telencephalon in this spatial learning. Intact and telencephalon-ablated goldfish were trained to find food in an arm placed in a constant room location and signaled by a local visual cue (mixed place-cue procedure). Both groups learned the task, but they used different learning strategies. Telencephalon-ablated goldfish learned the task more quickly and made fewer errors to criterion than controls. Probe trials revealed that intact goldfish could use either a place or a cue strategy, whereas telencephalon-ablated goldfish learned only a cue strategy. The results offer additional evidence that place and cue learning in fish are subserved by different neural substrates and that the telencephalon of the teleost fish, or some unspecified structure within it, is important for spatial learning and memory in a manner similar to the hippocampus of mammals and birds.

Effects of varying the amount of preexposure to spatial cues on a subsequent navigation task

In each of two experiments rats were preexposed to four compound landmarks (AX, BX, CX, and DX) one at a time; they were then trained to find a submerged platform located in a fixed position in a swimming pool using these same landmarks. When the preexposure was SHORT (4 sessions) it facilitated subsequent learning (a perceptual learning effect), whereas when rats were given a LONG preexposure phase (8 sessions) this facilitatory effect disappeared. EXTRA-LONG preexposure (16 sessions) reversed the facilitatory effect--that is to say, we observed retarded learning. The results show that rats' ability to navigate towards an invisible goal is affected by the length of their preexposure to the spatial cues that signal the location of the goal. These data are consistent with an associative analysis of the swimming pool navigation task.