Methods Of Assessment Of Learning And Memory

Reprinted from: Brain and Behavior. Raju TR, Kutty BM, Sathyaprabha TN and Shanakranarayana Rao BS (eds.), National Institute of Mental Health and Neuro Sciences, Bangalore, India. 2004:145-151.

METHODS OF ASSESSMENT OF LEARNING AND MEMORY IN RODENTS Srikumar BN, Bindu B, Priya V, Shankaranarayana Rao BS, Raju TR and Bindu M Kutty Among the various functions of the brain, one of the most interesting and puzzling one is the ability to acquire new information and store them for future retrieval. Learning is defined as behavioral modification through experiences or conditioning and when this persists it is termed memory. Memory can be broadly classified into declarative and non-declarative memory (Figure 1) (Milner et al., 1998). Declarative memory answers the question ‘what’, i.e. it is possible to verbally declare this type of memory. Memory of facts and events are examples of declarative memory. Declarative memory can be further classified into episodic and semantic memory. Episodic memory is memory of past and personally encountered events. The knowledge for the meaning of words and how to use them is phrased as semantic memory. Non-declarative (or procedural) memory answers the question ‘how’ and is not stored with respect to time and place. Skills, habits and priming are examples of nondeclarative memory. Priming is change in the ability to detect or identify objects as the result of recent encounters, for example, Mys in Mysore.

Currently several of the mechanisms underlying such a magnificent phenomenon as memory have been understood. However, many of them remain to be completely resolved. Animal models are useful in deciphering this complex machinery at molecular, cellular and systems level. Several of the CNS disorders are often associated with impairment in cognitive functions. Alzheimer’s disease and ageing, for example have a primary impact on learning and memory, and other diseases such as Parkinson’s disease, bipolar depression, schizophrenia, and many other neurological disorders have been demonstrated to have a secondary deficit in learning and memory. There are drugs available for the treatment of these memory disorders, but clinical improvement is far from satisfaction. Presently, the pathophysiological mechanisms underlying these disorders have not been clearly understood and it is imperative that new drugs that would be more useful be developed. Experimental paradigms of learning and memory are very enlightening in both elucidation of the pathophysiology and development of new medications for the treatment of learning and memory disorders.

Figure 1. Different mammalian memory systems (Milner et al., 1998) 145

Paradigms of Learning and Memory Broadly the paradigms can be based on either avoidance conditioning or maze training. Avoidance conditioning can be further classified into active and passive conditioning. The

different conditioning summarized in Table 1.

paradigms

are

Table 1: Summary of the different types of tasks used to assess learning and memory (Brioni et al., 1997; Reddy, 1997)

Passive avoidance

1. 2. 3. 4.

Step-down task Step-through task Two compartment task Up-hill avoidance

Active avoidance

1. 2. 3.

Run way avoidance Shuttle box avoidance Jumping avoidance

Classical conditioning

1. 2. 3.

Classical Pavlovian conditioning Rabbit eye blink response conditioning Operant conditioning

Mazes

1. 2. 3. 4. 5. 6. 7. 8.

Y maze T maze Radial arm maze (RAM) Water maze (WM) Figure 8 maze Complex maze Stone maze Battig maze

The hippocampus has been shown to be an important structure involved in learning and memory. There have been many mazes that have been used to test hippocampal function (Table 1). The radial arm maze (RAM) and water maze (WM) are perhaps the most used among them. A few of the various mazes that are used are depicted in Figure 2.

METHODOLOGY I. Operant Conditioning

Figure 2. Different mazes employed to study learning and memory (Reddy, 1997).

The operant conditioning apparatus consists of a rodent test chamber (Skinner’s box), pellet dispenser, sound attenuating chamber and a control unit. The test chamber contains two stainless steel response levers (5 x 4cm each) at a height of 6cm from the floor, located both on the left and right side (termed left and right lever respectively) on one wall of the chamber. A food hopper (5 x 3cm) is located in between the two levers at the floor level. Each lever can be pressed passively by a weight of about 7-8g. The chamber is connected to a solid food dispenser. Food pellets (approx. 45mg) are delivered to the 146

feeding tray (food hopper) from the pellet dispenser through a plastic tube once lever is pressed. The test chamber and the pellet dispenser are kept in a sound proof box so as to avoid the external disturbances (Bures et al., 1983). Procedure Rats are semi starved for 48hrs to motivate them towards food reward. On subsequent days 5-8 gm of food is provided after the test so as to maintain the body weight at 85% of its initial weight, throughout the experimental period. Acquisition (learning) test The training/shaping period consists of 30min session per day for three days. After three days of training session, the rats are subjected to 15 min test session for seven days. The number of lever press responses per session is recorded. Learning criterion: To assess the acquisition of the operant learning task, the criterion is set in such a way that the rats should reach 75% of the maximum lever press response of normal control rats. Retention (Memory) test Retention test can be carried out, 48h, 72h or 7 days after the acquisition phase of the operant conditioning. Here the rats are subjected to one trial and the number of pedal presses is calculated. Variants of the task 1. Continuous reinforcement schedule (CRF): In CRF, each response is followed by reinforcement after a brief interval (i.e. for each pedal press, there is one food pellet delivered). 2. Fixed ratio schedule (FR): In FR the reinforcement is scheduled after a fixed number of responses or when a fixed minimum time is elapsed. 3. Variable ratio or variable interval schedule (VR or VI): Here the response reward ratio or the inter reward time is randomly varied around some average value.

Operant conditioning has been extensively used in our lab as a learning paradigm, to study neuronal plasticity and as a model to understand the role of different areas in brain’s function. Increased dendritic arborisation and spine density was demonstrated following operant conditioning during the brain growth spurt (Mahajan and Desiraju, 1988). Operant conditioning was found sensitive to detect functional deficits following exposure to several environmental toxins like lead (Kumar and Desiraju, 1992), mercuric chloride (Lakshmana et al., 1993), endosulfan (Lakshmana and Raju, 1994), arsenic (Nagaraja and Desiraju, 1994), and metanil yellow (Nagaraja and Desiraju, 1993). It has been used as a model to understand the effect of subordination stress on brain biochemistry (Dhingra et al., 1996; Dhingra et al., 1997). The role of subiculum (Govindaiah et al., 1997; Nutan and Meti, 2000), fornix (Yoganarasimha and Meti, 1999)and substantia nigra – ventral tegmental area (SN-VTA) (Yoganarasimha et al., 1998) in brain’s function has been examined using operant conditioning paradigm. Furthermore, operant conditioning has been used to demonstrate the capability of intra cranial self-stimulation (ICSS) to facilitate learning (Yoganarasimha et al., 1998) and bring about functional reversal following fornix lesion (Yoganarasimha and Meti, 1999). Thus, operant conditioning is a powerful test to evaluate learning and memory. II. Evaluation of spatial Learning and memory in T-maze The T-maze consists of a start box (12x12cm), stem (35x12cm), a choice area (15x12cm) and two arms (35x12cm); each arm has a goal area (15x12cm) containing a food-well. The sidewalls are 40cm high. The stem and start box are separated by a sliding door, and a cloth curtain separated the arm and goal areas so that the food well from the choice area is not visible to the rat. 16W bulbs illuminated the start box, choice and goal areas. The T-maze is kept in a dimly lit, sound attenuated room (Bures et al., 1983). 147

Procedure Rewarded alternation test This is analogous to non-matching to sample task, where the rat is rewarded only if the current choice does not match the previous one. Behavioural testing is carried out in two phases; a. Orientation and training session and b. Learning performance test (acquisition test) Orientation and training session Rats are semi starved for 48h and allowed to explore the T-maze for 30 min. After the 30 min orientation session, the animals are returned to their home cage. Rats are then trained for alternation food reward. In these sessions rats are subjected to 10 trials / session / day. In each trial, the rat is placed in the start box, then the sliding door is released slowly and the rat is forced to move into one arm that leads to the goal area having the food pellet by blocking the other arm that does not have the food pellet. In the consecutive trials (10 trials), rats are again forced to enter a particular arm for food reward. Acquisition (learning) test The rats are subjected to 10 trials / session (intertrial interval of 30s) until they reach the criteria of 7-8 correct choices out of 10 trials. Acquisition test is conducted for eight days consisting of one session per day. Percentage of correct responses and errors are calculated for each rat. In each trial, the rat is placed in the start box, the sliding door released slowly and the rat is allowed to move into either left or right of the goal area for food reward. In each session of 10 trials, the number of errors, i.e., entry into the non-rewarded arm is recorded. Retention (Memory) test Seven days after the last day of acquisition of the task, the rats are subjected to a retention test. In this, rats are given one session of 30 trials. The performance is evaluated by calculating the number of errors committed during the session. Studies from our lab have documented that Tmaze is a useful tool to evaluate learning and

memory. Laxmi et al showed ventral subicular lesion results in spatial memory impairment in Tmaze (Laxmi et al., 1999). ICSS-induced facilitation of spatial tasks and stress-induced impairment has been shown using T-maze (Sunanda et al., 2000; Yoganarasimha et al., 1998). III. Evaluation of working and reference memory in Radial arm maze task (RAM) The radial arm maze was introduced in 1976 to study hippocampal dependent learning and memory (Olton and Samuelson, 1976) .The eight arm radial maze is a computer monitored plexiform maze (Columbus Instruments, USA). It consists of eight equally spaced arms radiating from an octagonal central platform. Each arm is 56.2cm long by 7.9cm wide. The entire maze is elevated 80cm above the floor. Acquisition of spatial task The rats are maintained on a restricted feeding at 85%of their free feeding body weight. These rats are allowed to familiarize themselves with the radial maze. Prior to each trial, all the eight arms are baited with food pellets. The rat is placed in the center of the maze and allowed to freely explore the maze. The rats are required to take the food pellets from each arm without making a re-entry into the arm already visited. The trial is terminated when the animal takes the food reward from all the eight arms or after 10 min if all the eight arms are not visited. The average criteria for acquisition are attaining 7.5/8 correct choices. Re-entry into an already visited arm is considered as an error. Each animal is given two trials daily; retention test is carried out ten days following acquisition. The performance of the animal is scored by calculating the percentage of correct responses (a correct entry is when the animal has not previously entered the arm) divided by the total number of entries made by the animal. Variants of Radial arm maze tasks Minor variants 1. Increase or decrease in the number of arms, thereby changing memory load 148

2. Water deprivation with water reward can be used rather than food deprivation with solid reward. 3. Impose increased retention intervals between arm choices. Major variants 1. Partially baited RAM task: In this, four of the eight arms are baited and the rats are trained to choose only the baited arms. This task permits discerning of reference memory and working memory components of spatial memory. An entry into an unbaited arm is regarded as a reference memory error and any re-entry (either to a baited or unbaited arm is considered as a working memory error. 2. Non-spatial working memory: Here, a cue (colour or surface texture) differentiates baited arms from the unbaited arms and allow cue and place memory to be tested. The radial arm maze has been widely used in behavioural neuroscience and behavioural pharmacology. Thus, RAM tests are useful in evaluating the effect of drugs, stress and various other environmental factors on learning and memory (Devi et al., 2003). IV. Assessment of Spatial learning in Water maze task This task is called “Morris water maze task”, named after RGM Morris, who developed it in 1981. The maze used for rats is a circular pool of water, with a diameter of 1.5m, height 57cm, and depth of water 29cm. The pool is a metal cylinder painted white on the inner surface and the escape platform is also made of metal cylinder with flat metallic top having a surface diameter of 10cm and is 1–1.5cm below the water surface during water maze training. The pool is filled with water (23±2.0°C) and made opaque with 1.5 litre of milk in order to obscure the platform and allow efficient tracking of the rats swim paths. The pool is situated in a 3.6 x 3.3 m2 room with black curtains hanging from the ceiling and covering other cues (doors, electrical fixtures,

etc.). An external cue is provided on the wall opposite to the target quadrant. Diffused illumination is provided by indirect lighting from four 200W flood lamp fixtures aimed towards the ceiling. The rats are marked with black colour on their foreheads to facilitate tracking. For the place navigation task, the escape platform is placed in the center of one of the four imaginary quadrants of the pool and kept in this location throughout training (Figure 3). All trials in each experiment are performed at the same time of the day (± 2 hrs) during the animals’ light phase. All rats are given two trials a day with an inter-trial interval of 20 min for 9 days (9 sessions) (Note: the number of trials taken to learn the task varies with the number of trials per day). For each trial, the rat is placed into the pool from one of the two start positions located around the rim of the pool and is then given a maximum of 90s to find the escape platform. If the animal found the platform, it is allowed to rest on it for 15s before being removed from the pool. If the rat does not locate the platform within 90s, the rat is hand guided to the platform. After the experiment, the rats are dried with a towel and placed under warming lamps. On the 10 th training day, the rats are given a probe trial without the platform and rats allowed to swim for 60s. It is generally believed that rats with an intact hippocampus spend more time in the target quadrant. Retention is tested after 10 days. The animals’ movements are tracked with a camera attached to the ceiling. Data are analysed using an automated tracking program (Columbus Instruments). Specialized software provided measures such as latency, path length, swim speed and the amount of time spent in different quadrants of the pool.

Figure 3. Platforms used in different water maze tasks (adapted from (Reddy, 1997). 149

Variants 1. Short-term spatial memory: In this procedure, the hidden escape platform is put in a different location in the pool each day and the focus is upon the savings in escape performance between the first and subsequent trials. The procedure is analogous to matching to sample task. 2. Non-spatial learning: a. The hidden platform is replaced by two visible platforms, one of which is rigid and provides escape from the water and other of which is floating (Figure 3). The floating platform provides insufficient buoyancy to support the rat when it tries to attempt climb and thus the rat should try to avoid it. b. The platform lies above the water level and the rats can easily locate the platform. This is used to demonstrate the effect of the treatment (for example, hippocampus lesion) is on spatial learning and nonspatial learning. 3. Radial arm water maze: This incorporates components of both the radial arm maze and water maze. It is essentially a radial arm maze immersed in a pool of water, with escape platform at the end of one of the arms. The motivation is aversive similar to water maze and can provide spatial information similar to radial arm maze. Ever since Morris developed the task it has been largely used in behavioral neuroscience. Tremendous progress in understanding the molecular mechanism has comes from studies using Morris water maze. Thus different tasks may represent different aspects of spatial memory. Thus this maze is well adapted to study the visuospatial factors in place learning and memory. Conclusion Maze procedures offer a variety of ways in assessing learning and memory in animals and of examining the effects of drugs and brain damage. Mazes tap a variety of processes other than the ability to use visuospatial cues to form cognitive maps of the environment, including associative and

procedural learning. Mazes differ in the extent to which they permit the use of diverse source of spatial information. They remain indispensable in behavioral neuroscience and drug discovery research. References 1. Brioni JD, Hock FJ, McGaugh JL (1997) Drug effects on learning and memory. In: Drug Discovery and Evaluation (Vogel HG, Vogel WH, eds), pp 334-349. Berlin: Springer-Verlag. 2. Bures J, Buresova O, Huston JP (1983) Techniques and basic experiments for the study of brain and behaviour.2nd Edition, Amsterdam: Elsevier Science Publishers B.V. 3. Devi L, Diwakar L, Raju TR, Kutty BM (2003) Selective neurodegeneration of hippocampus and entorhinal cortex correlates with spatial learning impairments in rats with bilateral ibotenate lesions of ventral subiculum. Brain Res 960: 9-15. 4. Dhingra NK, Lakshmana MK, Meti BL, Raju TR (1996) Subordination induced decrease in 5-hydroxytryptamine and dopamine levels in the frontal cortex-a study using worker-parasite relationship in rats as a model. Indian J Physiol Pharmacol 40: 213-219. 5. Dhingra NK, Raju TR, Meti BL (1997) Selective reduction of monoamine oxidase A and B in the frontal cortex of subordinate rats. Brain Res 758: 237-240. 6. Govindaiah, Rao BS, Raju TR, Meti BL (1997) Loss of hippocampal CA1 neurons and learning impairment in subicular lesioned rats. Brain Res 745: 121-126. 7. Kumar MV, Desiraju T (1992) EEG spectral power reduction and learning disability in rats exposed to lead through postnatal developing age. Indian J Physiol Pharmacol 36: 15-20. 8. Lakshmana MK, Desiraju T, Raju TR (1993) Mercuric chloride-induced alterations of levels of noradrenaline, dopamine, serotonin and acetylcholine esterase activity in different regions of rat brain during postnatal development. Arch Toxicol 67: 422-427. 9. Lakshmana MK, Raju TR (1994) Endosulfan induces small but significant changes in the levels of noradrenaline, dopamine and serotonin in the developing rat brain and deficits in the operant learning performance. Toxicology 91: 139-150. 10. Laxmi TR, Bindu PN, Raju TR, Meti BL (1999) Spatial memory impairment in ventral subicular lesioned rats. Brain Res 816: 245-248. 150

11. Mahajan DS, Desiraju T (1988) Alterations of dendritic branching and spine densities of hippocampal CA3 pyramidal neurons induced by operant conditioning in the phase of brain growth spurt. Exp Neurol 100: 1-15.

16. Olton DS, Samuelson RJ (1976) Remembrance of places passed: spatial memory in rats. J Exp Psychol 2: 97-115.

12. Milner B, Squire LR, Kandel ER (1998) Cognitive neuroscience and the study of memory. Neuron 20: 445-468.

17. Reddy DS (1997) Assessment of nootropic and amnestic activity of centrally acting agents. Ind J Pharmacol 29: 208-221.

13. Nagaraja TN, Desiraju T (1993) Effects of chronic consumption of metanil yellow by developing and adult rats on brain regional levels of noradrenaline, dopamine and serotonin, on acetylcholine esterase activity and on operant conditioning. Food Chem Toxicol 31: 41-44.

18. Sunanda, Shankaranarayana Rao BS, Raju TR (2000) Chronic restraint stress impairs acquisition and retention of spatial memory task in rats. Curr Sci 79: 1581-1584.

14. Nagaraja TN, Desiraju T (1994) Effects on operant learning and brain acetylcholine esterase activity in rats following chronic inorganic arsenic intake. Hum Exp Toxicol 13: 353-356. 15. Nutan KS, Meti BL (2000) Deficits in operant behavior and alteration of CA1, CA3 hippocampal dendritic

arborization due to subicular lesions. J Neurosci Res 59: 806-812.

19. Yoganarasimha D, Meti BL (1999) Amelioration of fornix lesion induced learning deficits by selfstimulation rewarding experience. Brain Res 845: 246-251. 20. Yoganarasimha D, Shankaranarayana Rao BS, Raju TR, Meti BL (1998) Facilitation of acquisition and performance of operant and spatial learning tasks in self-stimulation experienced rats. Behav Neurosci 112: 725-729.

151