Learning plays a fundamental role in an organism’s adaptability to its environment, which, in turn, is crucial for its survival. Learning is an adaptive behavior that is persistent and measurable and is not a result of maturation or altered motivation. An important element of the learning process is the memory, which is required in all the three phases of the process; acquisition, retention, and transfer. Learning can occur under a range of different conditions, and as such, acquisition of a skill or knowledge may or may not be a conscious effort.
Broadly, learning can be classified as associative learning and non-associative learning. When an association between two unrelated stimuli or events often occur with a reinforcement, then that process is termed as associative learning. On the other hand, non-associative learning does not involve reinforcements and is characterized by the change in the strength of response to a single stimulus. The two categorizations are further divided as follows,
- Habituation: This learning is marked by a decrease in the behavioral response to a repeated stimulus that does not result in any consequence. Habituation learning is a more likely reduction in excitatory responses rather than fatigue or sensory adaptation.
- Sensitization: As opposed to habituation, sensitization results in an increased behavioral response to a repeated stimulus that is often aversive in nature.
- Classical Conditioning: This type of learning occurs during the pairing of a conditioned stimulus (usually a neutral stimulus) with an unconditioned stimulus (usually biologically significant). Following repeated conditioning, the organism forms an association between the two unrelated stimuli. The learned association is observed as the unconditioned response to the presentation of the conditioned stimulus alone. The conditioned response is thus a result of the experience.
- Operant Conditioning: Unlike classical conditioning, the associative learning between a conditioned and unconditioned stimulus in operant conditioning is a result of positive or negative reinforcement. Thus, learning in this procedure is voluntary, unlike the unlearned reflexive response in classical conditioning.
- Imprinting: This type of learning is phase-sensitive learning that tends to be rapid and not affected by the consequences of the behaviors. Imprinting involves learning the characteristics of some stimulus.
- Observational Learning (Imitation): This type of learning usually occurs in a social set-up. Organisms learn behaviors of a social model, usually without any reinforcements motivating these behaviors.
Retention of the acquired behavior, response, or skills through learning is dependent on factors such as motivation, the intensity of reinforcement, and repetition. The combination of learning and retention process-based tasks are used to assess a myriad of diseases and disorders to understand the underlying neural mechanisms and develop treatments. In investigatory studies, retention behaviors are often studied within the same context conditions. These studies offer insights into memory capabilities in terms of long-and-short-term retention.
In day to day life, learning is often transferred between similar contexts and/or similar elements. Transfer of learning can easily be observed in educational and training environments, whereas acquired knowledge and skills are utilized or have an impact beyond the context within which they were acquired. Transfer of learning can be positive or negative if it results in improvement of performance or if it results in detrimental performance in another context. Further, the categorization of transfer of learning is based on the type of transfer target. If the transfer of learning takes place in a very similar context, it is categorized as near transfer, while transfer to a context that seems dissimilar is termed as far transfer. However, this categorization is debated due to the possible ambiguity in the judgment of the transfer target.
2. Animal Learning and Memory Assays
Learning and memory assays are applied in the investigation of the cognitive impairments caused by diseases and disorders. These assays can range from spatial learning to social learning, thus allowing observation of different types of learning and memory potentials.
In general, learning and memory assessments include pretraining and test trials; however, depending on the type of investigations, for example, drug testing experiments, pretest trials may also be included to assess baseline performances.
Apparatuses are often equipped with an automated tracking and recording system such as the Noldus EthoVision XT to assist with the observations.
2.1 Non-Associative Learning Tasks
The Open Field test was developed by Hall and Ballachey (1932) and is used in a wide variety of tests involving exploratory behaviors. The subjects are observed over a period of time without any clearly aversive or appetitive stimuli. The apparatus consists of an open arena contained by high walls to prevent the animal from escaping the arena. The apparatus is available in different combinations of clear and opaque walls and the number of arenas in an apparatus to allow observation of multiple animals. Available models include the Rodent Open-Field, the Porcine Open-Field, and the Octopus Round Arena.
Acoustic Startle Response Chamber
The Acoustic Startle Response Chamber comes equipped with an animal holder having a grid floor and transducer and is placed in a sound-attenuated isolation chamber. The stimulus presentation can be done via speakers, visible light bulbs, and shock grid attachment. The chamber can be calibrated using the Conductor Software.
2.2 Choice Mazes
Choice-based assays are a popular variety of learning and memory investigations. These tests are commonly applied in evaluating choice behaviors based on learned associations.
The T-Maze uses a single decision point and is popularly used in spontaneous alternation studies in rodents. The conventional Rodent T-Maze involves a long start arm that ends with two-choice arms extending perpendicularly from it on either side. T-Mazes are available for a range of species, including Cuttlefish T-Maze, Hen T-Maze, Ant T-Maze, and Zebrafish T-Maze. Different variants of the T-Maze, such as Rodent Continuous Angled T-Maze and Zebrafish Bifurcating T-Maze, are also available.
Similar to the T-Mazes, Y-Mazes also provide a single decision point. However, unlike the unnatural 90 degrees bend, Y-Mazes are designed with a natural 120 degrees angle to connect the arms. Y-Mazes are also available for a range of species, including Rodent Y-Maze, Drosophila Y-Maze, Locust Y-Maze, and Bat Y-Maze. Variants of Y-Mazes, such as Rodent Double Y-Maze and Rodent Visual Cue Y-Maze, are also available.
Radial Arm Maze
Radial Arm Mazes offer more complexity in terms of choices as compared to the bi-choice assays like the T-Maze and Y-Maze. The maze is constructed with a single multi-choice point from which choice arms radiate outwards. The complexity of the Radial Arm Maze task can be varied based on the number of choice arms. While the number of choice arms can be increased or decreased, the 8 arms Radial Arm Maze is a popular choice. Radial Arm Mazes are available for a range of species, including Rodent Radial Arm Maze, Porcine Radial Arm Maze, and Bee Radial Arm Maze.
2.3 Navigational Mazes
Navigational mazes vary in their complexity depending on the number of choice-points they allow and the possible configurations. While commonly used in the assessment of spatial learning and memory, navigational mazes also provide observation of other navigation-based behaviors such as foraging and route learning.
The Hebb-Williams Maze was described by Hebb and Williams in their 1946 paper as a way to evaluate animal intelligence. Popular in spatial working memory studies, the maze offers the opportunity to create different configurations with the help of moveable interior walls. The maze is available as Rodent Hebb-Williams Maze and Pig Hebb-Williams Maze. The Rodent Lattice Maze is similar to the Hebb-Williams Maze and offers greater levels of path complexity.
The maze was designed by Lashley (1929, 1933) in an attempt to localize the brain region involved in learning and storing information regarding maze exit. The apparatus consists of a start box connected to the maze that consists of four interconnected alleyways ending in a finishing box. The maze offers a low-stress environment for the animals, and the task usually uses positive reinforcements such as food rewards or mimicking the animal’s home cage in the finishing box.
The maze is a navigational learning task that mimics human urban set-up. Developed by Ku, Kloosterman, Tanila, and Wilson (2018), the Town maze consists of two potential start areas, landmark holders and winding alleyways designed such that distant view of the paths is obscure.
Unlike other navigational mazes that use structured pathways, the Morris Water Maze utilizes an open pool of water with hidden escape platforms to evaluate spatial navigation and memory. The apparatus was developed by Richard G. Morris in 1981 that relies on the fear of drowning to motivate the subject to find the hidden platform. The apparatus is easy to modify using inserts to create different learning mazes such as the Plus Water Maze and Water Star Maze. A dry version of the maze Dry Morris Water Maze, is also available.
The Ant Binary Tree Maze is used in the investigation of route learning among ant species. The apparatus includes an artificial nest area that is connected to a binary tree by a long arm. The foraging ant is placed on one of the end arms of the tree with a food reward and is observed finding its way to the nest. The experiments in this maze allow insights into how information is communicated between the ants of the colony by the forager.
The Bee Path Regularity Maze is used in the evaluation of learning and memory under different maze configurations. The apparatus includes vertical cylinders having 3 holes, one entrance, and 2 escape holes, that can be arranged in different configurations to create different levels of complexity. The bees are observed for their maze learning abilities under the motivation of food reward placed in the goal cylinder.
2.4 Classical Conditioning Assays
The Fear Conditioning Chamber comes equipped with an animal holder having an electric grid floor, speakers, and lights. Additionally, different context backgrounds are also available. The chamber is placed in a sound-attenuated isolation chamber. The apparatus is a popular choice in classical conditioning assays and allows a range of protocol applications. The chamber can be calibrated using the Conductor Software.
Active-Passive Avoidance Shuttle Box
The Active-Passive Avoidance Shuttle Box is used to assess different forms of fear-based conditioned avoidance learning. The apparatus consists of a dual-chamber separated by an automated guillotine door. Each chamber is equipped with an electrified grid and speakers to assist with fear conditioning. Context plates are also available.
2.5 Operant Assays
Operant assays rely on reinforcements or punishments to encourage learning of the association between the conditioned and the unconditioned stimulus. The reinforcements can be positive by allowing animals access to a reward or negative by the removal of an aversive stimulus. Similarly, subjecting an animal to a negative stimulus (such as shock) serves as positive punishment, while denying the subject a reward serves as a negative punishment.
Five Choice Serial Reaction Time (5CSRT) Task Chamber
The 5CSRT task is an operant assay that is used in the assessment of learning as well as impulsive behaviors. The general task involves the association between the cue and one of the five choices. The task is applied using appetitive motivation. The apparatus is available as Rodent 5CSRT and Zebrafish 5CSRT.
Magnetic Self-Administration Runway
The Magnetic Self-Administration Runway was developed by Geist and Ettenberg (1990) using a swivel-carriage device for the delivery of drugs to eliminate the interferences of traditional systems on subject behavior. The assay is commonly used in the evaluation of drug-seeking behaviors. See also Self-Administration Runway.
Repeated acquisition and learning chamber (RAPC)
The RAPC task uses positive reinforcement to evaluate route memorization and learning in rodents. The low-stress task uses a chamber that is divided by four panels into 5 equally sized compartments. The panels, each, have three one-way doorways that create a single path between the start and end compartments at the opposite ends of the chamber. The subjects are tasked with learning the correct route in order to reach the reward in the end box. The panels can be rotated to create different pathways.
2.6 Problem Solving Assays
Problem-solving assays are usually goal-directed and offer insights into the cognitive processes and knowledge manipulations involved.
The Puzzle Box utilizes the principles of the Light-Dark Box combined with obstruction puzzles. The apparatus has two chambers with one chamber being brightly lit, serving as the start area, and the other closed safe space. The subject is tasked with overcoming the barrier that separates the two spaces to access the narrow underpass in order to escape the start area. Barriers can include sawdust, plug, or weighted obstacle.
The Fictive Reward Assay is a binary-choice assay where the subject chooses between a real and a fictive reward. The assay is used to evaluate how failed choice experience can influence future decisions. The apparatus consists of an elevated track that has start areas and goal areas at opposite ends. Both areas are compartmentalized by a perforated acrylic wall to create two start and goal boxes, each equipped with a sliding door and photo beam sensor.
The Ziggurat Task is performed in a large, open square arena with 16 ziggurat pyramids placed evenly throughout. The task was developed by Faraji, Lehmann, Metz, and Sutherland (2008). It creates a low-stress environment that mimics the natural environments of the rodent. The task uses positive rewards placed atop the pyramid and is not directly visible to the subject to evaluate working and reference memory-related processes.
The Win Stay Maze is an adaptation of the Elevated Plus Maze that is designed to evaluate learning based on win-stay and win-shift paradigms. The apparatus consists of goal boxes at the ends of each arm and the LED lights to serves as cues during the task. The apparatus is commonly used in the observation of neural activities during learning and choice behaviors.
Delayed Matching to Place (DMP) Barnes Maze
The DMP apparatus is an adaptation of the conventional Barnes Maze and is used in the assessment of working memory. The apparatus is an elevated, open circular arena with three concentric rings of escape holes. The animal is subjected to bright light and an aversive tone to motivate it to find the correct escape hole, the position of which is changed with trials.
2.7 Olfactory Assays
In comparison to their visual systems, rodents have a well-developed olfactory system. The olfactory assays employ different odors to evaluate learning behaviors and other cognitive behaviors.
Attentional Set-Shifting Chamber
The Attentional Set-Shifting Chamber is used in performing intra-dimensional/extra-dimensional (IDED) task to evaluate cognitive flexibility. The chamber consists of two chambers; a testing area and a waiting area separated by a guillotine-like gate. The testing chamber is further divided by an acrylic partition into two, each containing a ceramic bowl. The bowls can be filled with the digging medium and the odorant, which can be selected from the odor and medium kits. Bowtie Mazes are also used in IDED tasks.
The Dig Task is used in the evaluation of cognitive dysfunction using olfactory-based decision-making. The apparatus consists of a chamber with two scent cups placed at one end. The cups can be filled with digging medium and the odorant selected from the odor and medium kits. The subject is tasked with associating a scent with the reward placed within the scent cup.
The Odor Span Test was developed by Dudchenko, Wood, and Eichenbaum (2000) and is used in the assessment of olfactory working memory capacity in rodents. The apparatus is composed of an elevated, large square arena with 24 clear acrylic cups placed along the perimeter. The cups can be filled with digging medium and the odorant selected from the odor and medium kits. The subjects are tasked with remembering the association between scents and rewards.
Drosophila Olfactory Operant Conditioning
The Drosophila Olfactory Operant Conditioning apparatus is a T-Maze styled apparatus used in odor based classical conditioning. The subjects are presented with odor pairings, where one odor results in electric shock. The apparatus is used in the assessment of memory retention and learning.
3. Human Learning and Memory Assays
Learning and memory form an important part of cognitive processing. Impairments in memory and learning capacities are often seen as an observable symptom of neurodegenerative diseases, neuropsychiatric disorders, and brain injuries. While empirical investigations through surveys, observational research, and correlational methods (For digital healthcare tools visit Qolty) are common practices, human studies also involve observation of performances in tasks (For human research apparatuses click here). Additionally, virtual reality tasks are also gaining popularity, especially for their flexibility in environment creation and overall task control (For virtual reality tools visit Simian Labs). The use of virtual reality offers the opportunity to adapt animal assays for human applications, thus allowing easy comparison between the two models. For human virtual reality experiments, click here.
4. Ethical Obligations and Considerations
Animal models provide great assistance when it comes to drug testing and the development of treatments for learning and memory impairments in humans. It is important that efforts be made to perform all investigations as ethically as possible. The following are few guidelines for animal-based experiments,
- Social isolation for some animals can be extremely stressful and thus should be kept to the required minimum for an experiment.
- Animals should be habituated to handling to minimize the effects of handling stress.
- Experiments that involve aggressive interactions should be carefully monitored to prevent injury to the animals involved.
- Animals should not be overworked, and appropriate rest periods should be included to maintain motivation and avoid muscle fatigue.
- Apparatuses should be cleaned as necessary to prevent any lingering stimuli from influencing the subject behavior.
Apart from the above guidelines, efforts should be made to ensure the overall wellbeing of the animals in the laboratory. Animals should not be subjected to unnecessary stress or mishandling at any time.
In human experiments and research, the following are few guidelines that should be followed,
- Explicit consent of the participants should be obtained prior to testing.
- Experiments should be age-appropriate, and all medical factors should be taken into consideration.
- Safety and well-being of the participants should be prioritized above all.
- Experiments that involve potential triggering set-ups should be carefully created so as not to overwhelm or stress the participant.
- Appropriate measures should be taken when using virtual reality for experimentation.
Learning and memory research in animals furthers our understanding of the neural mechanisms and cognitive processing. The use of animals helps with the development of disease and disorder-specific treatments and drugs. While animal models provide great tools for learning and memory research, they do not serve as an exact model of human behaviors or emotions. Further, experimental assays may be focused on a single aspect of learning and memory while overlooking other factors contributing to these cognitive processes.
- Dudchenko, P.A., Wood, E.R., & Eichenbaum, H. (2000). Neurotoxic hippocampal lesions have no effect on odor span and little effect on odor recognition memory but produce significant impairments on spatial span, recognition, and alternation. Journal of Neuroscience 20(8), 2964-77. DOI: 10.1523/JNEUROSCI.20-08-02964.2000
- Faraji, J., Lehmann, H., Metz, G.A., & Sutherland, R.J. (2008). Rats with hippocampal lesion show impaired learning and memory in the ziggurat task: a new task to evaluate spatial behavior. Behavioural Brain Research 189 (2008) 17–31
- Geist, T. D., & Ettenberg, A. (1990). A simple method for studying intravenous drug reinforcement in a runaway. Pharmacology, Biochemistry and Behaviors;36(3):703-6.
- Hall, C. S., & Ballachey, E. L. (1932). “A study of the rat’s behavior in a field: a contribution to method in comparative psychology.” University of California Publications in Psychology, 6: 1–12.
- Hebb, D.O., & Williams, K.A. (1946). A method of rating animal intelligence. The Journal of General Psychology; 34, 59-65
- Kim, K. U., Huh, N., Jang, Y., Lee, D., & Jung, M.W. (2015). Effects of fictive reward on rat’s choice behavior. Scientific Reports, 5, 8040. http://dx.doi.org/10.1038/srep08040
- Ku, S., Kloosterman, F., Tanila, H., & Wilson M.A. (2018). Characteristics of CA1 place fields in a complex maze with multiple choice points. Hippocampus, 28(2), 81-96. doi: 10.1002/hipo.22810
- Lashley, K. S. (1929). Brain mechanisms and intelligence: A quantitative study of injuries to the brain. University of Chicago Press, Chicago.
- Lashley, K. S. (1933). Integrative functions of the cerebral cortex. Physiological Reviews, 13(1), 1–42.doi:10.1152/physrev.19188.8.131.52
- Moore, B. R. (2004). The evolution of learning. Biological Reviews, 79(2), 301–335. doi:10.1017/s1464793103006225
- Morris, R.G.M. (1981). Spatial localization does not require the presence of local cues. Learning and Motivation, 12, 239-260
- Perkins, D. N., & Salomon, G. (1992). Transfer of Learning. In: International Encyclopedia of Education, Second Edition. Oxford, England: Pergamon Press
- Reznikova, J., & Ryabko, B. (n.d.). Using Shannon entropy and Kolmogorov complexity to study the language and intelligence of ants. Proceedings of 1994 IEEE International Symposium on Information Theory. doi: 10.1109/isit.1994.394773
- Woodworth, R. S., & Thorndike, E. L. (1901). The influence of improvement in one mental function upon the efficiency of other functions. (I). Psychological Review, 8(3), 247–261. doi:10.1037/h0074898