Learning large-scale spatial relationships in a maze and effects of MK-801 on retrieval in the rhesus monkey

Monkeys have strong abilities to remember the visual properties of potential food sources for survival in the nature. The present study demonstrated the first observations of rhesus monkeys learning to solve complex spatial mazes in which routes were guided by visual cues. Three monkeys were trained in a maze (6 m x 6 m) included of four different mazes. We recorded the cue and cup errors, latencies, and pathway for each trial. The data showed that monkeys learned the target place after three days in the first maze and spent a shorter time in learning the following mazes. The maze was an efficient method to measure the ability and proceeding of spatial memory in monkeys. Moreover, working memory can also be tested by using the maze. MK-801 at 0.02 mg/kg but not at 0.005 mg/kg impaired monkeys' retrieval of spatial memory after they learned all four mazes. The present maze may provide an efficient method to help bridging the gap in cognition between nonhuman primates and humans, and in particular to gain insight into human cognitive function and dysfunction.     

Development of maze navigation by tufted capuchins (Cebus apella)

Theories of spatial navigation hypothesize that animals use vector or topological information to choose routes, often including detours, to move objects or themselves to goals. We assessed adult capuchin monkeys’ (Cebus apella) navigation through 192 virtual two-dimensional mazes that incorporated detour problems. Six monkeys initially were significantly less likely to choose the correct paths when detours were required than when not. Three of the six monkeys repeatedly practiced the 192 mazes to asymptotic mastery; the other three did not practice the mazes again. In a subsequent transfer test, each monkey made correct choices equivalently often on familiar and novel mazes, which suggests that they used general planning skills for maze navigation. Of the three monkeys that practiced the 192 maze-set repeatedly, one efficiently detoured and the other two significantly improved detouring compared to their initial performance. Two monkeys, contrary to their performance when completing the 192 maze-set for the first time, made correct choices at the same rate as chimpanzees. Some evidence suggested that two monkeys used topological information, but utilization of vector information was obvious for all monkeys. Our findings suggest that the boundaries of any individual's navigational abilities are not predicted by species, but depend on experience.     

Dynamics of parietal neural activity during spatial cognitive processing

Dynamic neural processing unrelated to changes in sensory input or motor output is likely to be a hallmark of cognitive operations. Here we show that neural representations of space in parietal cortex are dynamic while monkeys perform a spatial cognitive operation on a static visual stimulus. We recorded neural activity in area 7a during a visual maze task in which monkeys mentally followed a path without moving their eyes. We found that the direction of the followed path could be recovered from neuronal population activity. When the monkeys covertly processed a path that turned, the population representation of path direction shifted in the direction of the turn. This neural population dynamic took place during a period of unchanging visual input and showed characteristics of both serial and parallel processing. The data suggest that the dynamic evolution of parietal neuronal activity is associated with the progression of spatial cognitive operations.     

Places and patterns — a study of context learning in honeybees

To investigate the priming of memories by contextual cues, bees were trained to negotiate two mazes in different places 25?m apart. In the first maze, bees flew leftwards when the inner wall of the maze was covered with 45° stripes or rightwards when the inner wall was coloured yellow. In the second maze, bees flew rightwards on viewing 135° diagonal stripes or leftwards on viewing blue. The trajectories evoked by 45° or 135° stripes were similar in both mazes. However, vertical stripes were treated like 45° stripes in maze 1 and like 135° stripes in maze 2. Contextual cues prime the response to stripes that are oriented in the training condition for that site so influencing responses to stripes in closely neighbouring orientations. What objects in a bee's surroundings determine its sense of place? Bees were trained to different visual patterns at two sites 40?m apart (A+ versus A– at site A, and E+ versus E– at site E). A+ was preferred over A– and E+ was preferred over E– at both training sites. A preference for A+ over E+ exhibited at site A dropped gradually with distance to suggest that spatial context includes both close and distant objects.     

Barnes Maze Testing Strategies with Small and Large Rodent Models

Spatial learning and memory of laboratory rodents is often assessed via navigational ability in mazes, most popular of which are the water and dry-land (Barnes) mazes. Improved performance over sessions or trials is thought to reflect learning and memory of the escape cage/platform location. Considered less stressful than water mazes, the Barnes maze is a relatively simple design of a circular platform top with several holes equally spaced around the perimeter edge. All but one of the holes are false-bottomed or blind-ending, while one leads to an escape cage. Mildly aversive stimuli (e.g. bright overhead lights) provide motivation to locate the escape cage. Latency to locate the escape cage can be measured during the session; however, additional endpoints typically require video recording. From those video recordings, use of automated tracking software can generate a variety of endpoints that are similar to those produced in water mazes (e.g. distance traveled, velocity/speed, time spent in the correct quadrant, time spent moving/resting, and confirmation of latency). Type of search strategy (i.e. random, serial, or direct) can be categorized as well. Barnes maze construction and testing methodologies can differ for small rodents, such as mice, and large rodents, such as rats. For example, while extra-maze cues are effective for rats, smaller wild rodents may require intra-maze cues with a visual barrier around the maze. Appropriate stimuli must be identified which motivate the rodent to locate the escape cage. Both Barnes and water mazes can be time consuming as 4-7 test trials are typically required to detect improved learning and memory performance (e.g. shorter latencies or path lengths to locate the escape platform or cage) and/or differences between experimental groups. Even so, the Barnes maze is a widely employed behavioral assessment measuring spatial navigational abilities and their potential disruption by genetic, neurobehavioral manipulations, or drug/ toxicant exposure.     

Maze Learning and Recall in a Weakly Electric Fish,Mormyrus rume proboscirostris Boulenger (Mormyridae,Teleostei)1

Mormyrus rume proboscirostris, African weakly electric fish, were trained to seek shelter in a meander maze, and following path acquisition released into the empty arena with all maze cues removed, either from the original start box or from a novel site (recall). We demonstrate that fish use their active electrosense, sight, and lateral line synergistically in maze acquisition and recall. In the presence of an electric roadmap consisting of an array of aluminum and Plexiglas objects, fish employed landmark orientation. But fish ignored visual markers and relied on internalized motor routines, which was inconsistent with evidence for cognitive mapping.     

Aging and Sensory Substitution in a Virtual Navigation Task

Virtual environments are becoming ubiquitous, and used in a variety of contexts–from entertainment to training and rehabilitation. Recently, technology for making them more accessible to blind or visually impaired users has been developed, by using sound to represent visual information. The ability of older individuals to interpret these cues has not yet been studied. In this experiment, we studied the effects of age and sensory modality (visual or auditory) on navigation through a virtual maze. We added a layer of complexity by conducting the experiment in a rotating room, in order to test the effect of the spatial bias induced by the rotation on performance. Results from 29 participants showed that with the auditory cues, it took participants a longer time to complete the mazes, they took a longer path length through the maze, they paused more, and had more collisions with the walls, compared to navigation with the visual cues. The older group took a longer time to complete the mazes, they paused more, and had more collisions with the walls, compared to the younger group. There was no effect of room rotation on the performance, nor were there any significant interactions among age, feedback modality and room rotation. We conclude that there is a decline in performance with age, and that while navigation with auditory cues is possible even at an old age, it presents more challenges than visual navigation.     

Can puzzle feeders be used as cognitive screening instruments? Differential performance of young and aged female monkeys on a puzzle feeder task

Conventional cognitive testing of monkeys is time‐consuming and involves single‐caging and food or water deprivation. Here we report a novel test of global cognitive performance that can be completed in a short time period without food/water or social restrictions. Nine mazes of increasing difficulty were developed using a standard puzzle feeder, and the maze‐solving performance of ten young and five aged female cynomolgus monkeys (Macaca fascicularis) was tested. The young monkeys solved maze configurations at higher levels of difficulty and solved the first level of difficulty more quickly than aged monkeys. This task discriminated performance by age in nonhuman primates as do more conventional forms of cognitive testing and indicates that this task may be a quick and easy assessment of global cognitive function. Am. J. Primatol. 49:195–202, 1999. © 1999 Wiley‐Liss, Inc.     

The learning of a sequence of visual patterns by the ant Cataglyphis cursor

We used a maze to explore the ability of Cataglyphis cursor to store multiple visual patterns presented in a fixed sequence. Ants were trained individually to negotiate a linear maze that consisted of four boxes connected by tunnels and through which an ant travelled from a sucrose feeder back to its nest. Each box had one entrance and two possible exits. One exit led to a blocked tunnel and the other to an open tunnel leading to the entrance of the next box. The open and closed exits in each box were labelled by different solid, black shapes that were specific to each box. Ants learnt to negotiate the maze using the shapes for guidance rather than a fixed motor strategy. Trained ants could not only discriminate positive from negative shapes, but had also learnt which positive shape belonged to which box. For example, when the positive shape appropriate to box 1 (1+) was pitted against that appropriate to box 3 (3+), ants preferred 1+ to 3+ in box 1, but chose 3+ over 1+ in box 3. We conclude that ants can identify individual positive shapes and expect to encounter them in the correct order independently of extra-maze cues.     

Rearing environment affects spatial learning in juvenile Chinook salmon Oncorhynchus tshawytscha

We tested the prediction that a complex physical rearing environment would enhance short-term spatial memory as assessed by learning ability in a spatial navigation task in juvenile Chinook salmon Oncorhynchus tshawytscha. We reared fish in two low-density treatments, where fish were either in bare fiberglass tanks (bare) or in tanks with physical structure (complex). We also tested conventionally reared high-density hatchery fish to compare with these other experimental treatments. Our reason for including this third hatchery treatment is that the two low-density treatments, aside from the manipulation of structure, followed a rearing programme that is designed to produce fish with more wild-like characteristics. We tested individually marked fish for seven consecutive days and recorded movement and time to exit a testing maze. Stimulus conspecific fish outside the exit of the maze provided positive reinforcement for test fish. Fish from the bare treatment were less likely to exit the start box compared with fish in the complex and hatchery treatments. However, fish in the hatchery treatment were significantly more likely to exit the maze on their own compared with both the bare and complex treatments. Hatchery fish effectively learned the task as shown by a decrease in the number of mistakes over time, but the number of mistakes was significantly greater on the first day of trials. Increasing habitat complexity with structure may not necessarily promote spatial learning ability, but differences between hatchery and experimental treatments in rearing density and motivation to be near conspecifics likely led to observed behavioural differences.     

A comparative study of the performance of individuals with fragile X syndrome and Fmr1 knockout mice on Hebb-Williams mazes

Fragile X syndrome (FXS) is the most prevalent form of heritable mental retardation. It arises from a mutation in the FMR1 gene on the X chromosome that interferes with expression of fragile X mental retardation protein (FMRP) and leads to a wide range of behavioural and cognitive deficits. Previous studies have shown a deficit in basic visual perceptual processing as well as spatial abilities in FXS. How such a deficit may impact spatial navigation remains unknown. The current study extended previous research by evaluating spatial learning and memory using both virtual and physical versions of Hebb-Williams mazes, which allows for testing of humans and animals under comparable conditions. We compared the performance of individuals affected by FXS to typically developing individuals of equivalent mental age as well as the performance of Fmr1 knockout mice to wild-type control mice on the same maze problems. In human participants, performance of the comparison group improved across trials, showing expected significant decreases in both errors and latency. In contrast, the performance of the fragile X group remained at similar levels across trials. Although wild-type control mice made significantly fewer errors than the Fmr1 knockout mice, latencies were not statistically different between the groups. These findings suggest that affected humans and mice show similar spatial learning deficits attributable to the lack of FMRP. The implications of these data are discussed including the notion that Hebb-Williams mazes may represent a useful tool to examine the impact of pharmacological interventions on mitigating or reversing the symptoms associated with FXS.     

Reflexive social attention in monkeys and humans

For humans, social cues often guide the focus of attention. Although many nonhuman primates, like humans, live in large, complex social groups, the extent to which human and nonhuman primates share fundamental mechanisms of social attention remains unexplored. Here, we show that, when viewing a rhesus macaque looking in a particular direction, both rhesus macaques and humans reflexively and covertly orient their attention in the same direction. Specifically, when performing a peripheral visual target detection task, viewing a monkey with either its eyes alone or with both its head and eyes averted to one side facilitated the detection of peripheral targets when they randomly appeared on the same side. Moreover, viewing images of a monkey with averted gaze evoked small but systematic shifts in eye position in the direction of gaze in the image. The similar magnitude and temporal dynamics of response facilitation and eye deviation in monkeys and humans suggest shared neural circuitry mediating social attention.     

Intelligence-Augmented Rat Cyborgs in Maze Solving

Cyborg intelligence is an emerging kind of intelligence paradigm. It aims to deeply integrate machine intelligence with biological intelligence by connecting machines and living beings via neural interfaces, enhancing strength by combining the biological cognition capability with the machine computational capability. Cyborg intelligence is considered to be a new way to augment living beings with machine intelligence. In this paper, we build rat cyborgs to demonstrate how they can expedite the maze escape task with integration of machine intelligence. We compare the performance of maze solving by computer, by individual rats, and by computer-aided rats (i.e. rat cyborgs). They were asked to find their way from a constant entrance to a constant exit in fourteen diverse mazes. Performance of maze solving was measured by steps, coverage rates, and time spent. The experimental results with six rats and their intelligence-augmented rat cyborgs show that rat cyborgs have the best performance in escaping from mazes. These results provide a proof-of-principle demonstration for cyborg intelligence. In addition, our novel cyborg intelligent system (rat cyborg) has great potential in various applications, such as search and rescue in complex terrains.     

Navigation Using Sensory Substitution in Real and Virtual Mazes

Under certain specific conditions people who are blind have a perception of space that is equivalent to that of sighted individuals. However, in most cases their spatial perception is impaired. Is this simply due to their current lack of access to visual information or does the lack of visual information throughout development prevent the proper integration of the neural systems underlying spatial cognition? Sensory Substitution devices (SSDs) can transfer visual information via other senses and provide a unique tool to examine this question. We hypothesize that the use of our SSD (The EyeCane: a device that translates distance information into sounds and vibrations) can enable blind people to attain a similar performance level as the sighted in a spatial navigation task. We gave fifty-six participants training with the EyeCane. They navigated in real life-size mazes using the EyeCane SSD and in virtual renditions of the same mazes using a virtual-EyeCane. The participants were divided into four groups according to visual experience: congenitally blind, low vision & late blind, blindfolded sighted and sighted visual controls. We found that with the EyeCane participants made fewer errors in the maze, had fewer collisions, and completed the maze in less time on the last session compared to the first. By the third session, participants improved to the point where individual trials were no longer significantly different from the initial performance of the sighted visual group in terms of errors, time and collision.     

Eye movements of stumptailed monkeys during discriminative performance: An overview

There are a number of statements that can be made about eye movements of monkeys during the learning of simple and complex discriminative problems that are probably applicable to a wide variety of visual tasks. There are systematic changes in eye movements as a function of practice. Some of these changes occur long after grosser measures of performance, such as frequency of correct choices, have reached an asymptote. Hence, short-term studies of visual information processing may be misleading. Duration of visual fixations and frequency of visual fixations are independent measures, reflecting different cognitive processes. Studies which measure only total looking time confound these two measures and, thus, may miss important information. Eye movements appear to be an important, if not essential, component of the chain of events constituting the cognitive processing underlying performance on visual tasks.     

A weighted constraint satisfaction approach to human goal-directed decision making

When we plan for long-range goals, proximal information cannot be exploited in a blindly myopic way, as relevant future information must also be considered. But when a subgoal must be resolved first, irrelevant future information should not interfere with the processing of more proximal, subgoal-relevant information. We explore the idea that decision making in both situations relies on the flexible modulation of the degree to which different pieces of information under consideration are weighted, rather than explicitly decomposing a problem into smaller parts and solving each part independently. We asked participants to find the shortest goal-reaching paths in mazes and modeled their initial path choices as a noisy, weighted information integration process. In a base task where choosing the optimal initial path required weighting starting-point and goal-proximal factors equally, participants did take both constraints into account, with participants who made more accurate choices tending to exhibit more balanced weighting. The base task was then embedded as an initial subtask in a larger maze, where the same two factors constrained the optimal path to a subgoal, and the final goal position was irrelevant to the initial path choice. In this more complex task, participants’ choices reflected predominant consideration of the subgoal-relevant constraints, but also some influence of the initially-irrelevant final goal. More accurate participants placed much less weight on the optimality-irrelevant goal and again tended to weight the two initially-relevant constraints more equally. These findings suggest that humans may rely on a graded, task-sensitive weighting of multiple constraints to generate approximately optimal decision outcomes in both hierarchical and non-hierarchical goal-directed tasks.     

Low-stress Route Learning Using the Lashley III Maze in Mice

Many behavior tests designed to assess learning and memory in rodents, particularly mice, rely on visual cues, food and/or water deprivation, or other aversive stimuli to motivate task acquisition. As animals age, sensory modalities deteriorate. For example, many strains of mice develop hearing deficits or cataracts. Changes in the sensory systems required to guide mice during task acquisition present potential confounds in interpreting learning changes in aging animals. Moreover, the use of aversive stimuli to motivate animals to learn tasks is potentially confounding when comparing mice with differential sensitivities to stress. To minimize these types of confounding effects, we have implemented a modified version of the Lashley III maze. This maze relies on route learning, whereby mice learn to navigate a maze via repeated exposure under low stress conditions, e.g. dark phase, no food/water deprivation, until they navigate a path from the start location to a pseudo-home cage with 0 or 1 error(s) on two consecutive trials. We classify this as a low-stress behavior test because it does not rely on aversive stimuli to encourage exploration of the maze and learning of the task. The apparatus consists of a modular start box, a 4-arm maze body, and a goal box. At the end of the goal box is a pseudo-home cage that contains bedding similar to that found in the animal’s home cage and is specific to each animal for the duration of maze testing. It has been demonstrated previously that this pseudo-home cage provides sufficient reward to motivate mice to learn to navigate the maze¹. Here, we present the visualization of the Lashley III maze procedure in the context of evaluating age-related differences in learning and memory in mice along with a comparison of learning behavior in two different background strains of mice. We hope that other investigators interested in evaluating the effects of aging or stress vulnerability in mice will consider this maze an attractive alternative to behavioral tests that involve more stressful learning tasks and/or visual cues.Download video file.^{(94M, mp4)}     

Reaction times for hand movements made in response to visual versus vibratory cues

Reaction times were determined for monkeys and humans who made wrist flexion and extension movements in response to vibratory and visual cues. Humans initiated movements approximately 50 msec sooner in response to vibratory as compared to visual cues. For monkeys, this difference was approximately 100 msec. Mean daily reaction times for monkeys and humans improved with practice until they reached a steady level of performance. Increased differences between vibratory and visual reaction times were weakly correlated with increased age of humans. The increase in the differences appeared to result from decreased reaction times by older subjects for vibratory-cued movements; reaction times for visually cued movements did not consistently vary across the age range of subjects tested (19-36 years). The results obtained using this novel paradigm suggest that it may be a useful tool for simultaneously testing behavioral performance or neurological function during somatosensorimotor and visuomotor tasks.     

Reaction Times for Hand Movements Made in Response to Visual versus Vibratory Cues

Reaction times were determined for monkeys and humans who made wrist flexion and extension movements in response to vibratory and visual cues. Humans initiated movements approximately 50 msec sooner in response to vibratory as compared to visual cues. For monkeys, this difference was approximately 100 msec. Mean daily reaction times for monkeys and humans improved with practice until they reached a steady level of performance. Increased differences between vibratory and visual reaction times were weakly correlated with increased age of humans. The increase in the differences appeared to result from decreased reaction times by older subjects for vibratory-cued movements; reaction times for visually cued movements did not consistently vary across the age range of subjects tested (19-36 years). The results obtained using this novel paradigm suggest that it may be a useful tool for simultaneously testing behavioral performance or neurological function during somatosensorimotor and visuomotor tasks.     

Integration of local features into global shapes: monkey and human FMRI studies

The integration of local image features into global shapes was investigated in monkeys and humans using fMRI. An adaptation paradigm was used, in which stimulus selectivity was deduced by changes in the course of adaptation of a pattern of randomly oriented elements. Accordingly, we observed stronger activity when orientation changes in the adapting stimulus resulted in a collinear contour than a different random pattern. This selectivity to collinear contours was observed not only in higher visual areas that are implicated in shape processing, but also in early visual areas where selectivity depended on the receptive field size. These findings suggest that unified shape perception in both monkeys and humans involves multiple visual areas that may integrate local elements to global shapes at different spatial scales.