Wikipedia:School and university projects/Psyc3330 w11/Group13 - Spatial memory

In cognitive psychology and neuroscience, spatial memory is the part of memory responsible for recording information about one's environment and its spatial orientation. For example, a person's spatial memory is required in order to navigate around a familiar city, just as a rat's spatial memory is needed to learn the location of food at the end of a maze. It is often argued that in both humans and animals, spatial memories are summarized as a cognitive map. Spatial memory has representations within working, short-term and long-term memory. Research indicates that there are specific areas of the brain associated with spatial memory. Many methods are used for measuring spatial memory in children, adults, and animals.

Short-Term Spatial Memory
Short-term memory (STM) can be described as a system allowing one to temporarily store and manage information that is necessary to complete complex cognitive tasks. Tasks which employ short-term memory include learning, reasoning, and comprehension. Spatial memory is a cognitive process that enables a person to remember different locations as well as spatial relations between objects. This allows one to remember where an object is in relation to another object. For instance, allowing someone to navigate through a familiar city. Spatial memories are said to form after a person has already gathered and processed sensory information about their environment.

Spatial Working Memory
Working memory (WM) can be described as a limited capacity system that allows one to temporarily store and process information. This temporary store enables one to complete or work on complex tasks while being able to keep information in mind. For instance, the ability to work on a complicated mathematical problem utilizes ones working memory.

One highly influential theory of WM is the Baddeley and Hitch multi-component model of working memory. The most recent version of this model suggests that there are four subcomponents to WM, namely the phonological loop, the visuo-spatial sketchpad, the central executive, and the episodic buffer. One component of this model, the visuo-spatial sketchpad, is said to be responsible for the temporary storage, maintenance, and manipulation of both visual and spatial information.



In contrast to the multi-component model, some researchers believe that STM should be viewed as a unitary construct. In this respect, visual, spatial, and verbal information are thought to be organized by levels of representation rather than the type of store to which they belong. Within the literature, it is suggested that further research into the fractionation of STM and WM be explored. However, much of the research into the visuo-spatial memory construct have been conducted in accordance to the paradigm advanced by Baddeley and Hitch.

The Role of the Central Executive
Research into the exact function of the visuo-spatial sketchpad has indicated that both spatial short-term memory and working memory are dependent on executive resources and are not entirely distinct. For instance, performance on a working memory but not on a short-term memory task was affected by articulatory suppression suggesting that impairment on the spatial task was caused by the concurrent performance on a task that had extensive use of executive resources. Results have also found that performances were impaired on STM and WM tasks with executive suppression. This illustrates how, within the visuo-spatial domain, both STM and WM require similar utility of the central executive. Additionally, during a spatial visualisation task (which is related to executive functioning and not STM or WM) concurrent executive suppression impaired performance indicating that the effects were due to common demands on the central executive and not short-term storage. The researchers concluded with the explanation that the central executive employs cognitive strategies enabling participants to both encode and maintain mental representations during short-term memory tasks.

Although studies suggest that the central executive is intimately involved in a number of spatial tasks, the exact way in which they are connected remains to be seen.

Long-Term Spatial Memory
Long-term spatial memory (LTSM) consists of more permanent representations within the brain than those of STM and WM. These representations are said to endure even when we are not attending to them. The ability to complete LTSM tasks then rely upon our ability to acquire and retrieve different types of spatial information and to use it to complete different spatial goals. As with spatial STM and WM, LTSM can be divided into different domains. These two domains are route knowledge (also called route learning or route finding) and cognitive mapping, respectively.

Route knowledge helps us navigate through the world by using internal route representations which are used to allow one to find their way and to reach a particular destination. One type of representation within route knowledge is egocentric spatial memory which entails the integration of ones body orientation into their spatial representation. It is differentiated from route knowledge on the basis that it specifically encodes information about distances and directions, in addition to personal experiences and meanings. Thus, this type of memory is implicated in the remembrance of buildings and locations that are well-known.

Alternatively, cognitive mapping refers to the directionality between different landmarks and locations. In contrast to route knowledge, this type of information is not dependent on bodily orientation and has been termed as a form of allocentric memory. This type of memory allows one to calculate the distance and direction between locations in the environment.

Spatial Expertise
Within the literature there is evidence that experts in a particular field are able to perform memory tasks in accordance with their skills at an exceptional level. The level of skill displayed by experts has also been said to exceed the limits of the normal capacity of both STM and WM. It is believed that because experts have an enormous amount of prelearned and task-specific knowledge they are able to encode information in a more efficient way.

An interesting study investigating taxi drivers memory for streets in Helsinki, Finland examined the role of prelearned spatial knowledge. This study compared experts to a control group to determine how this prelearned knowledge in their skill domain allows them to overcome the capacity limitations of STM and WM. The study used four levels of spatial randomness:

Route Order – spatially continuous route

Route Random – spatially continuous list presented randomly

Map Order – street names forming a straight line on the map, but omitting intermediate streets

Map Random – streets on map presented in random order The results of this study indicate that the taxi drivers (experts) recall of streets was higher in both the route order condition and the map order condition than the two random conditions. This indicates that the experts were able to use their prelearned spatial knowledge to organize the information in such a way that they surpassed STM and WM capacity limitations. The organization strategy that the drivers employed is known as chunking. Additionally, the comments made by the experts during the procedure point towards their use of route knowledge in completing the task. To ensure that it was in fact spatial information that they were encoding, the researchers also presented lists in alphabetical order and semantic categories. However, the researchers found that it was in fact spatial information that the experts were chunking, allowing them to surpass the limitations of both visuo-spatial STM and WM.

Animal Research
Within the literature it has been found that certain species of paridae and corvidae (such as the black-capped chickadee and the scrub jay) are able to use spatial memory to remember where, when and what type of food they have cached. Recent studies with rats and squirrels, have also suggested that they are able to use spatial memory to locate previously hidden food. Experiments using the radial maze have allowed researchers to control for a number of variables, such as the type of food hidden, the locations of where the food is hidden, the retention interval, as well as any odour cues that could skew results of memory research. In particular, studies have indicated that rats have memory for where they have hidden food and what type of food they have hidden. This is shown in retrieval behaviour, such that the rats are selective in going to the arms of the maze where they have previously hidden preferred food more often than to arms with less preferred food or where no food was hidden. Thus, the evidence for the spatial memory of some species of animals, such as rats, indicates that they do use spatial memory to locate and retrieve hidden food stores.

Visual – Spatial Distinction
Logie (1995) proposed that the visuo-spatial sketchpad is broken down into two subcomponents, one visual and one spatial. These are the visual cache and the inner scribe, respectively. The visual cache is a temporary visual store including such dimensions as colour and shape. Conversely, the inner scribe is a rehearsal mechanism for visual information and is responsible for information concerning movement sequences. Although a general lack of consensus regarding this distinction has been noted in the literature,  there is a growing amount of evidence that the two components are separate and serve different functions.

Visual memory is responsible for retaining visual shapes and colours (i.e., what), whereas spatial memory is responsible for information about locations and movement (i.e., where). This distinction is not always straightforward since part of visual memory involves spatial information and vice versa. For example, memory for object shapes usually involves maintaining information about the spatial arrangement of the features which define the object in question.

In practice the two systems work together in some capacity but different tasks have been developed to highlight the unique abilities involved in either visual or spatial memory. For example, the visual patterns test (VPT) measures visual span whereas the Corsi Blocks Task measures spatial span. Correlational studies of the two measures suggest a separation between visual and spatial abilities, due to a lack of correlation found between them in both healthy and brain damaged patients.

Support for the division of visual and spatial memory components is found through experiments using the dual-task paradigm. A number of studies have shown that the retention of visual shapes or colours (i.e., visual information) is disrupted by the presentation of irrelevant pictures or dynamic visual noise. Conversely, the retention of location (i.e., spatial information) is disrupted only by spatial tracking tasks, spatial tapping tasks, and eye movements. For example, participants completed both the VPT and the Corsi Blocks Task in a selective interference experiment. During the retention interval of the VPT, the subject viewed irrelevant pictures (e.g., avant-garde paintings). The spatial interference task required participants to follow, by touching the stimuli, an arrangement of small wooden pegs which were concealed behind a screen. Both the visual and spatial spans were shortened by their respective interference tasks, confirming that the Corsi Blocks Task relates primarily to spatial working memory.

Measuring Spatial Memory
There are a variety of tasks that psychologists use to measure spatial memory on adults, children and animal models. These tasks allow professionals to identify cognitive irregularities or in adults and children and allows researchers to administer varying types of drugs and or lesions in participants and measure the consequential effects on spatial memory.

The Corsi Block Tapping Task
Also known as the Corsi Span Test, this psychological test is commonly used to determine the visual-spatial memory span and the implicit visual-spatial learning abilities of an individual. Participants sit with nine wooden 3x3 cm blocks fastened before them on a 25 x 30 cm baseboard in a standard random order. The experiment taps a sequence pattern onto the blocks which participants must then replicate. The blocks are numbered on the experimenters’ side to allow for efficient pattern demonstration. The sequence length increases each trial until the participant is no longer able to correctly replicate the pattern. The test can be used to measure both short-term and long-term spatial memory, depending on the length of time between test and recall.

The test was created by Canadian neuropsychologist Phillip Corsi, who modeled it after Hebb’s digit span task by replacing the numerical test items with spatial ones. On average most participants achieve a span of five items on the Corsi span test and seven on the digit span task.

Visual Pattern Span
Similar to the Corsi block tapping test but regarded as a more pure test of visual short term recall. Participants are presented with a a series or matrix patterns that have half their cells coloured and the other half blank. The matrix patterns are arranged in a way that is difficult to verbally code forcing the participant to rely on visual spatial memory. Beginning with a small 2 x 2 matrix participants copy the matrix pattern from memory into an empty matrix. The matrix patterns are increased in size and complexity at a rate of two cells until the participant's ability to replicate them breaks down. On average, participant's performance tends to break down at sixteen cells.

Pathway Span Task
This task is designed to measure spatial memory abilities in children. The experimenter asks the participant to visualize a blank matrix with a little man. Through a series of directional instructions such as forwards, backwards, left or right the experimenter guides the participant’s little man on a pathway throughout the matrix. At the end the participant is asked to indicate on a real matrix where the little man that he or she visualized, finished. The length of the pathway varies depending on the level of difficulty (1-10) and the matrices themselves may vary in length from 2 x 2 cells to 6 x 6.

Dynamic Mazes
Intended for measuring spatial ability in children. With this test an experimenter presents the participant with a drawing of a maze with a picture of a man in the centre. The experimenter uses his or her finger to trace a pathway from the opening of the maze to the drawing of the man while the participant watches. The participant is then expected to replicate the demonstrated pathway through the maze to the drawing of the man. Mazes vary in complexity as difficulty increases.

Radial Arm Maze
Full article: Radial arm maze



First pioneered by Orton and Samuelson in 1976, the radial arm maze is designed to test the spatial memory capabilities of rats. Mazes are typically designed with a centre platform and a varying number of arms branching off with food placed at the ends. The arms are usually shielded from each other in some way but not to the extent that external cues cannot be used as reference points.

In most cases, the rat is placed in the center of the maze and needs to explore each arm individually to retrieve food while simultaneously remembering which arms it has already pursued. The maze is set up so the rat is forced to return to the center of the maze before persuing another arm. Measures are usually taken to prevent the rat from using its olfactory senses to navigate such as placing extra food throughout the bottom of the maze.

Moris Water Maze
Full article: Morris water navigation task

The Morris Water Maze is a classic for studying spatial learning and memory in rats and was first developed in 1981 by Richard G. Morris whom the test is named after. The subject is placed in a round tank of translucent water with walls that are too high for it to climb out and water that is too deep for it to stand in. Additionally, the walls of the tank are decorated with visual cues to serve as reference points. The rat must swim around the pool until by chance it discovers the hidden platform just below the surface that it can climb up onto.

Typically, rats swim around the edge of the pool first before venturing out into the center in a meandering pattern before stumbling upon the hidden platform however as time spent in the pool increases experience, the amount of time needed to locate the platform decreases with veteran rats swimming directly to the pool almost immediately after being placed in the water.

Hippocampus
The hippocampus provides animals with a spatial map of their environment. It stores information regarding non-egocentric space (egocentric means in reference to one's body position in space) and therefore supports viewpoint independence in spatial memory. This means that it allows for viewpoint manipulation from memory. It is however, important for long-term spatial memory of allocentric space (reference to external cues in space). Maintenance and retrieval of memories are thus relational or context dependent. The hippocampus makes use of reference and working memory and has the important role of processing information about spatial locations.

Blocking plasticity in this region results in problems in goal-directed navigation and impairs the ability to remember precise locations. Amnesic patients with damage to the hippocampus cannot learn or remember spatial layouts and patients having undergone hippocampal removal are severely impaired in spatial navigation. Monkeys with leisons to this area cannot not learn object-place associations and rats also display spatial deficits by not reacting to spatial change. In addition, rats with hippocampal lesions were shown to have temporally ungraded (time-independent) retrograde amnesia that is only resistant to recognition of a learned platform task when the entire hippocampus is lesioned but not when it is partially lesioned. Deficits in spatial memory are also found in spatial discrimination tasks.

Large differences in spatial impairment are found among the dorsal and ventral hippocampus. Lesions to the ventral hippocampus have no effect on spatial memory, while the dorsal hippocampus is required for retrieval, processing short-term memory and transferring memory from the short term to longer delay periods. Infusion of amphetamine into the dorsal hippocampus has also been shown to enhance memory for spatial locations learned previously. These findings indicate that there is a functional dissociation between the dorsal and ventral hippocampus.

Hemispheric differences within the hippocampus are also observed. A study on London taxi drivers, asked drivers to recall complex routes around the city as well as famous landmarks for which the drivers had no knowledge of their spatial location. This resulted in an activation of the right hippocampus solely during recall of the complex routes which indicates that the right hippocampus is used for navigation in large scale spatial environments.

The hippocampus is known to contain two separate memory circuits. One circuit is used for recollection-based place recognition memory and includes the entorhinal-CA1 system while the other system is used for place recall memory and makes use of the CA3-CA1 system.

Place cells are also found in the hippocampus.

Posterior parietal cortex
The parietal cortex encodes spatial information using an egocentric frame of reference. It is therefore involved in the transformation of sensory information coordinates into action or effector coordinates by updating the spatial representation of the body within the environment. As a result, lesions to the parietal cortex produce deficits in the acquisition and retention of egocentric tasks, whereas minor impairment is seen among allocentric tasks.

Rats with lesions to the anterior region of the posterior parietal cortex reexplore displaced objects, while rats with lesions to the posterior region of the posterior parietal cortex displayed no reaction to spatial change.

Parietal cortex lesions are also known to produce temporally ungraded retrograde amnesia.

Entorhinal cortex
The dorsalcaudal medial entorhinal cortex (dMEC) contains a topographically organized map of the spatial environment made up of grid cells. This brain region thus transforms sensory input from the environment and stores it as a durable allocentric representation in the brain to be used for path integration.

The entorhinal cortex contributes to the processing and integration of geometric properties and information in the environment. Lesions to this region impair the use of distal but not proximal landmarks during navigation and produces a delay-dependent deficit in spatial memory that is proportional to the length of the delay. Lesions to this region are also known to create retention deficits for tasks learned up to 4 weeks but not 6 weeks prior to the lesions.

Memory consolidation in the entorhinal cortex is achieved through extracellular signal-regulated kinase activity.

Prefrontal cortex
The medial prefrontal cortex processes egocentric spatial information. It participates in the processing of short-term spatial memory used to guide planned search behavior and is believed to join spatial information with its motivational significance. The identification of neurons that anticipate expected rewards in a spatial task support this hypothesis. The medial prefrontal cortex is also implicated in the temporal organization of information.

Hemisphere specialization is found in this brain region. The left prefrontal cortex preferentially processes categorical spatial memory including source memory (reference to spatial relationships between a place or event), while the right prefrontal cortex preferentially processes coordinate spatial memory including item memory (reference to spatial relationships between features of an item).

Leisons to the medial prefrontal cortex impair the performance of rats on a previously trained radial arm maze, however, rats can gradually improve to the level of the controls as a function of experience. Lesions to this area also cause deficits on delayed nonmatching-to-positions tasks and impairments in the acquisition of spatial memory tasks during training trials.

Retrosplenial cortex
The retrosplenial cortex is involved in the processing of allocentric memory and geometric properties in the environment. Inactivation of this region accounts for impaired navigation in the dark and thus it is implicated to be involved in the process of path integration.

Lesions to the retrosplenial cortex consistently impair tests of allocentric memory, while sparing egocentric memory. Animals with lesions to the caudal retrosplenial cortex show impaired performance on a radial arm maze only when the maze is rotated to remove their reliance on intramaze cues. In humans, damage to the retrosplenial cortex results in topographical disorientation. Most cases involve damage to the right retrosplenial cortex and include Broadmann’s area 30. Patients are often impaired at learning new routes and at navigating through familiar environments. However, most patients usually recover within 8 weeks.

The retrosplenial cortex preferentially processes spatial information in the right hemisphere.

Perirhinal cortex
The perirhinal cortex is associated with both spatial reference and spatial working memory. It processes relational information of environmental cues and locations.

Lesions in the perirhinal cortex account for deficits in reference memory and working memory, and increase the rate of forgetting of information during training trials of the Morris water maze. This accounts for the impairment in the initial acquisition of the task. Lesions also cause impairment on an object location task and reduce habituation to a novel environment.

Plasticity
Spatial memories are formed after an animal gathers and processes sensory information about its surroundings (especially vision and proprioception). In general, mammals require a functioning hippocampus (particularly area CA1) in order to form and process memories about space. There is some evidence that human spatial memory is strongly tied to the right hemisphere of the brain.

Spatial learning requires both NMDA and AMPA receptors, consolidation requires NMDA receptors, and the retrieval of spatial memories requires AMPA receptors. In rodents, spatial memory has been shown to covary with the size of a part of the hippocampal mossy fiber projection.

The function of NMDA receptors varies according to the subregion of the hippocampus. NMDA receptors are required in the CA3 of the hippocampus when spatial information needs to be reorganized, while NMDA receptors in the CA1 are required in the acquisition and retrieval of memory after a delay, as well as in the formation of CA1 place fields. Blockade of the NMDA receptors prevents induction of long-term potentiation and impairs spatial learning.

The CA3 of the hippocampus plays an especially important role in the encoding and retrieval of spatial memories. The CA3 is innervated by two afferent paths known as the perforant path (PPCA3) and the dentate gyrus (DG)-mediated mossy fibers (MFs). The first path is regarded as the retrieval index path while the second is concerned with encoding.

Topographical Disorientation
Full article: Topographical disorientation or Developmental topographical disorientation

Topographical disorientation is a cognitive disorder that results in the individual being unable to orient his or herself in the real or virtual environment. Patients also struggle with spatial information dependant tasks. These problems could possibly be the result of a disruption in the ability to access one’s cognitive map, a mental representation of the surrounding environment or the inability to judge objects’ location in relation to one’s self.

Developmental Topographical Disorientation (DTD) is diagnosed when patients have shown an inability to navigate even familiar surroundings since birth and show no apparent neurological causes for this deficiency such as lesioning or brain damage. DTD is a relatively new disorder and can occur in varying degrees of severity.

Hippocampal Damage and Schizophrenia
Research with rats indicates that spatial memory may be adversely affected by neonatal damage to the hippocampus in a way that closely resembles schizophrenia. Schizophrenia is thought to stem from neurodevelopmental problems shortly after birth.

Rats are commonly used as models of schizophrenia patients. Experimenters create lesions in the ventral hippocampal area shortly after birth, a procedure known as neonatal ventral hippocampal lesioning(NVHL). Adult rats who with NVHL show typical indicators of schizophrenia such as hypersensitivity to psychostimulants, reduced social interactions and impaired prepulse inhibition, working memory and set-shifting. Similar to schizophrenia, impaired rats fail to use environmental context in spatial learning tasks such as showing difficulty completing the radial arm maze and the Moris water maze.

GPS
Recent research on spatial memory and wayfinding in an article by Ishikawa et al in 2008 revealed that using a GPS device reduces an individual’s navigate abilities when compared to other participants who were using maps or had previous experience on the route with a guide. A second possibility is that the visual set-up of each tool differs. GPS allows the user to only see a small detailed close-up of a particular segment of the map which is constantly updated. In comparison, maps usually allow the user to see the same view of the entire route from departure to arrival. Other research has shown that individuals who use GPS travel more slowly overall compared to map users who are faster. GPS users stop more frequently and for a longer period of time whereas map users and individuals using past experience as a guide travel on more direct routes to reach their goal.

Learning Difficulties and Spatial Memory
Nonverbal learning disability is characterized by normal verbal abilities but impaired visuospatial abilities. Problem areas for children with nonverbal learning disability are arithmetic, geometry, and science. Impairments in spatial memory is implicated in nonverbal learning disorder and other learning difficulties.

Arithmetic word problems involve written text containing a set of data followed by one or more questions and require the use of the four basic arithmetic operations (addition, subtraction, multiplication, or division). Researchers suggest that successful completion of arithmetic word problems involves spatial working memory (involved in building schematic representations) which facilitates the creation of spatial relationships between objects. Creating spatial relationships between objects is an important part of solving word problems because mental operations and transformations are required.

For example, consider the following question: "A child builds three towers using red and white coloured blocks of the same size. The lowest tower has 14 blocks; the highest has 7 more blocks. The intermediate tower has three blocks less than the highest one. How many blocks are in each of the three towers?" To solve the question, it is necessary to maintain incoming information (i.e., the text) and integrate it with previous information (such as knowledge for arithmetic operations). The individual must also select relevant (i.e., the spatial relationship between the blocks) and inhibit irrelevant information (i.e., the colours and textures of the blocks) and simultaneously build a mental representation of the problem.

Researchers investigated the role of spatial memory and visual memory in the ability to complete arithmetic word problems. Children in the study completed the Corsi Block Task (forward and backward series) and a spatial matrix task, as well as a visual memory task called the house recognition test. Poor problem-solvers were impaired on the Corsi Block Tasks and the spatial matrix task, but performed normally on the house recognition test when compared to normally achieving children. The experiment demonstrated that poor problem solving is related specifically to deficient processing of spatial information.