Procedural Learning: Animals
Neurobehavioral research conducted primarily between 1980 and 2000 employed dissociation methodology to provide compelling evidence for the existence of multiple memory systems in the mammalian brain. Although evidence of neuroanatomical dissociations of the role of various brain structures in different memory tasks is a critical step in advancing the multiple-memory systems hypothesis, a full understanding of memory organization involves elucidation of the psychological operating principles that distinguish different types of memory. One putative form of memory present in the mammalian brain is declarative memory and involves the acquisition, consolidation, and retrieval of cognitive information for facts and events (Cohen and Squire, 1980; Eichenbaum and Cohen, 2001). Declarative memory appears to rely on a neuoranatomical system composed of various structures in the medial temporal lobe, most notably the hippocampus. In contrast, procedural memory involves the acquisition, consolidation, and retrieval of information acquired in tasks involving various types of Pavlovian classical conditioning and instrumental/operant learning paradigms. This article describes evidence indicating that procedural memory involves several, largely separable neuroanatomical systems, including the cerebellum, basal ganglia, and amygdala.
Cerebellum and the Classically Conditioned Eyeblink Response
In the early 1980s Richard Thompson and his colleagues (McCormick et al., 1981) demonstrated a critical role for the cerebellum in the classically conditioned eyeblink response in the rabbit. In this paradigm animals receive pairings of an auditory conditioned stimulus (CS—a tone), and an unconditioned stimulus (UCS—an air puff delivered to the eye). The naturally occurring rabbit eyeblink response is a defensive mechanism initially emitted in response to unconditioned stimulus. However, following several presentations of the tone-air puff contingency, the animal learns to produce the eye blink (i.e., a conditioned response) in the absence of the air puff. A series of elegant studies employing irreversible and reversible lesions, neuroanatomical tracing, and electrophysiological techniques have mapped out the neural circuitry underlying the processing of unconditioned and conditioned stimuli in eyeblink conditioning (Kim and Thompson, 1997). This circuitry includes an unconditioned stimulus pathway in which the dorsal accessory olive projects through the inferior cerebellar penduncle and a conditioned stimulus pathway involving mossy-fiber projections to the cerebellum via the pontine nuclei. The interpositus nucleus of the cerebellum receives converging CS-US information, and this deep cerebellar nuclei plays a critical role in both the acquisition and expression of the classically conditioned eyeblink response. The lack of impairment of performance of unconditioned eyeblink behavior following lesions of the interpositus nucleus provides critical evidence of a selective role for this nucleus in learned or conditioned eyeblink behavior. Study of the role of cerebellar circuitry in the acquisition and expression of conditioned eye-blink behavior represents perhaps the best example of a model approach to understanding the neurobiological bases of a form of procedural learning in the mammalian brain.
The Basal Ganglia and Stimulus-Response Habit Learning
In a multiple-memory systems approach to memory organization, there is evidence that components of the basal ganglia (specifically the caudate-putamen or dorsal striatum) mediate the acquisition of tasks that involve the formation of stimulus-response habits (Mishkin and Petri, 1984; Packard, Hirsh, and White, 1989). In several studies, the role of the rodent basal ganglia in procedural/habit learning has been dissociated from the role of the hippocampal system in cognitive/declarative memory. An early study examining the selective role of the basal dorsal striatum in S-R habit learning involved an experiment using two food-rewarded, eight-arm radial maze tasks; a cognitive/declarative win-shift task that required rats to remember the arms they had visited earlier in a daily training session; and a procedural/S-R habit win-stay task that required rats to acquire a simultaneous visual (light-dark) discrimination. Lesions of the dorsal striatum impair acquisition of the win-stay task and do not affect acquisition of the win-shift task, whereas lesions of the hippocampal system produce the opposite dissociation (Packard, Hirsh, and White, 1989).
An additional study used two water-maze tasks to investigate the selective role of the basal ganglia in S-R memory (Packard and McGaugh, 1992). In these tasks two rubber balls protruding above the water surface served as cues. One ball (correct) was on top of a platform that could be used to escape from the water, and the other ball (incorrect) was on top of a thin rod and thus did not provide escape. The two balls also differed in visual appearance (i.e., vertical versus horizontal black/white stripes). In a cognitive/declarative version of the task, the correct platform was in the same spatial location on every trial, but the appearance of the ball varied. Therefore, this version of the task requires rats to learn to approach the correct ball on the basis of spatial location, not visual pattern. In a procedural/S-R habit version of the task, the rats located the correct platform in different spatial locations across trials, but the visual pattern was consistent. Therefore, this task could be acquired by learning an approach response to the visual cue. Lesions of the dorsal striatum impair acquisition of the S-R habit task without affecting acquisition of the spatial task (Packard and McGaugh, 1992). Other findings indicating that lesions of the rodent dorsal striatum impair two-way active avoidance behavior, simultaneous tactile discriminations, egocentric response learning, and conditional visual and auditory conditioning also support a role for the dorsal striatum in S-R habit learning and memory (Kirkby and Kimble, 1968; Colombo, Davis, and Volpe, 1989; Reading, Dunnett, and Robbins, 1991; Adams, Kesner, and Raggozino, 2001; Kesner, Bolland, and Dakis, 1993).
The Amygdala and Stimulus-Affect Conditioning
The mammalian amygdala has long been implicated in the neurobiology of emotional behavior; this function of the amygdala seems to extend to emotional learning. Bruce S. Kapp and his colleagues provided some of the early evidence of a role for the amygdala in conditioned fear learning (Kapp, Frysinger, Gallagher, and Haselton, 1979). Joseph E. LeDoux's team has extensively investigated the role of the amygdala in a fear-conditioning paradigm in rats in which brief electrical shock is paired with exposure to either a discrete auditory cue or a specific environmental context (LeDoux, 1992; Fendt and Fanselow, 1999). In a series of anatomical, lesion, and pharmacological studies these investigators have mapped out a potential neural circuit in which conditioned stimulus (e.g., tone) and unconditioned stimulus (e.g., shock) information converge in the basolateral amygdala. Projections from the basolateral nucleus to the central nucleus of the amygdala and subsequent downstream projections to various brain-stem regions allow for the expression of various autonomic and behavioral responses (e.g., freezing behavior, heart-rate and blood-pressure changes) in response to discrete and contextual conditioned stimuli. The auditory cortex is not necessary for the acquisition of discrete auditory cue fear conditioning. Rather, auditory information projected to the amygdala from the medial geniculate nucleus of the thalamus is apparently sufficient to support this form of fear conditioning. These findings suggest a rapid-response procedural learning system that can bypass neocortical involvement in cognitive assessment of threatening stimuli.
Further evidence of a role for the amygdala in learned fear responses in the rat emerged from an investigation of the fear-potentiated startle response by Michael Davis and his colleagues (e.g. Davis, 1992). In these experiments, rats initially receive pairings of a light or tone with foot shock. During subsequent training, the normal startle response that results from exposure to a loud noise is potentiated by concurrent presentation of the light or tone conditioned stimulus. Findings from lesion, pharmacological, and anatomical studies converge to indicate a role for the amygdala in the acquisition and expression of fear-potentiated startle.
Aside from its involvement in aversively motivated learning, evidence indicates a role for the amygdala in the acquisition and expression of appetitively motivated stimulus-affect learning tasks. For example, pretraining and posttraining basolateral amygdala lesions in rats impair acquisition and expression of conditioned place preference behavior for both natural rewards such as food and sex (Cador, Robbins, and Everitt, 1989) and addictive drugs such as amphetamines (Hiroi and White, 1991; Hsu, Schroeder, and Packard, 2002). In this task, rats are confined in contrasting environments that alternate daily: on one day in an environmental context paired with natural or drug rewards and on the next day in a second context that is not paired with the rewarding treatment. On a reward-free test session given following training, the amount of time spent in the two environments is measured, and rats demonstrate a reliable conditioned place preference for—that is, spend longer amounts of time in—the environment previously paired with the rewarding stimulus. Investigators interested in the neurobiological bases of addiction have traditionally used this task. However, animals display approach behavior to specific environmental stimuli in a drug-free state on the test day, and therefore expression of a conditioned place preference requires memory for an association between previously neutral stimuli and the rewarding affective consequences of the treatment.
It should be noted that not all investigators agree that the mnemonic function of the amygdala involves a long-term role in memory storage. James L. McGaugh and his colleagues argue that amygdala damage impairs unconditioned fear responses (specifically freezing behavior), and therefore suggest caution in proposing a role for the amygdala role in learned fear responses (Cahill, Weinberger, Roozendaal, and McGaugh, 1999; Fanselow and LeDoux, 1999). In addition, evidence indicates that the amygdala (specifically the basolateral nucleus) modulates memory processes occurring in other brain structures, including declarative memory processes mediated by the hippocampus and stimulus-response habit learning mediated by the basal ganglia (Packard, Cahill, and McGaugh, 1994). However, a modulatory role for the amygdala in some types of learning does not itself rule out a possible long-term role in memory storage in stimulus-affect conditioning. There is a need for further task-specific analysis of the potential short-term versus long-term roles of the amygdala in memory and for consideration of the role of separate amygdala nuclei in different types of learning and memory.
There has been significant progress in identifying neuroanatomical components of separable procedural memory systems in the mammalian brain, including the cerebellum, basal ganglia, and amygdala. Although this article has focused on studies of lower animals, there is similar evidence for the role of these various structures in some of these forms of procedural memory in nonhuman primates (Zola-Morgan, Squire, and Mishkin, 1982; Fernandez-Ruiz, Wang, Aigner, and Mishkin, 2001;) and humans (Knowlton, Mangels, and Squire, 1996; Cohen and Squire, 1980; Johnsrude et al., 2000).
Adams, S., Kesner, R. P., and Ragozzino, M. E. (2001). Role of the medial and lateral caudate-putamen in mediating an auditory conditional response association. Neurobiology of Learning and Memory 76, 106-116.
Cador, M., Robbins, T. W., and Everitt, B. J. (1989). Involvement of the amygdala in stimulus-reward associations: Interaction with the ventral striatum. Neuroscience 30, 77-86.
Cahill, L., Weinberger, N. M., Roozendaal, B., and McGaugh, J. L.(1999). Is the amygdala a locus of "conditioned fear"? Some questions and caveats. Neuron 23, 227-228.
Cohen, N. J., and Squire, L. R. (1980). Preserved learning and retention of pattern analyzing skill in amnesics: Dissociation of knowing how and knowing that. Science 210, 207-210.
Colombo, P. J., Davis, H. P., and Volpe, B. T. (1989). Allocentric spatial and tactile memory impairments in rats with dorsal caudate lesions are affected by preoperative training. Behavioral Neuroscience 103, 242-250.
Davis, M. (1992). The role of the amygdala in conditioned fear. InJ. P. Aggleton, ed., The amygdala: Neurobiological aspects of emotion, memory, and mental dysfunction. New York: Wiley-Liss.
Fanselow, M. S., and LeDoux, J. E. (1999). Why we think plasticity underlying Pavlovian fear conditioning occurs in the basolateral amygdala. Neuron 23, 229-232.
Fendt, M., and Fanselow, M. S. (1999). The neuroanatomical and neurochemical basis of conditioned fear. Neuroscience and Biobehavioral Reviews 23, 743-760.
Fernandez-Ruiz, J., Wang, J., Aigner, T. G., and Mishkin, M.(2001). Visual habit formation in monkeys with neurotoxic lesions of the ventrocaudal neostriatum. Proceedings of the National Academy of Sciences of the United States of America 984,4,196-4,201.
Hiroi, N., and White, N. M. (1991). The lateral nucleus of the amygdala mediates expression of the amphetamine conditioned place preference. Journal of Neuroscience 11, 2,107-2,176.
Johnsrude, I. S., Owen, A. M., White, N. M., Zhao, W. V., and Bohbot, V. (2000). Impaired preference conditioning after anterior temporal lobe resection in humans. Journal of Neuroscience 20, 2,649-2,656.
Kapp, B. S., Frysinger, R. C., Gallagher, M., and Haselton, J. R.(1979). Amygdala central nucleus lesions: Effects on heart rate conditioning in the rabbit. Physiology and Behavior 23, 1,109-1,117.
Kesner, R. P., Bolland, B. L., and Dakis, M. (1993). Memory for spatial locations, motor responses, and objects: Triple dissociation among the hippocampus, caudate nucleus, and extrastriate visual cortex. Experimental Brain Research 93, 462-470.
Kim, J. J., and Thompson, R. F. (1997). Cerebellar circuits and synaptic mechanisms involved in classical eyeblink conditioning. Trends in Neuroscience 20, 177-181.
Kirkby, R. J., and Kimble, D. P. (1968). Avoidance and escape behavior following striatal lesions in the rat. Experimental Neurology 20, 215-227.
Knowlton, B. J., Mangels, J. A., and Squire, L. R. (1996). A neostriatal habit learning system in humans. Science 273, 1,399-1,402.
LeDoux, J. E. (1992). Emotion and the amygdala. In J. P. Aggleton, ed., The amygdala: Neurobiological aspects of emotion, memory, and mental dysfunction. New York: Wiley-Liss.
McCormick, D. A., Lavond, D. G., Clark, G., Kettner, R. E., Rising, C. E., and Thompson, R. F. (1981). The engram found? Role of the cerebellum in classical conditioning of nictitating membrane and eylid responses. Bulletin of the Psychonomic Society 18, 103-105.
Mishkin, M., and Petri, H. L. (1984). Memories and habits: Some implications for the analysis of learning and retention. In L. R. Squire and N. Butters, eds., Neuropsychology of memory. New York: Guilford.
Packard, M. G., Cahill, L., and McGaugh, J. L. (1994). Amygdala modulation of hippocampal-dependent and caudate nucleus-dependent memory processes. Proceedings of the National Academy of Sciences of the United States of America 91, 8,477-8,481.
Packard, M. G., Hirsh, R, and White, N. M. (1989). Differential effects of fornix and caudate nucleus lesions on two radial maze tasks: Evidence for multiple memory systems. Journal of Neuroscience 9, 1,465-1,472.
Packard, M. G., and McGaugh, J. L. (1992). Double dissociation of fornix and caudate nucleus lesions on acquisition of two water maze tasks: Further evidence for multiple memory systems. Behavioral Neuroscience 106, 439-446.
Reading, P. J., Dunnett, S. B., and Robbins, T. W. (1991). Dissociable roles of the ventral, medial, and lateral striatum on the acquisition and performance of a complex visual stimulus response habit. Behavioral Brain Research 45, 147-161.
Zola-Morgan, S., Squire, L., and Mishkin, M. (1982). The neuroanatomy of amnesia: Amygdala-hippocampus versus temporal stem. Science 218, 1,337-1,339.