Memory is the record of experience represented in the brain. There are multiple forms of memory supported by distinct brain systems. Specific forms of memory are characterized by whether they last a short or long period, by whether they involve unique experiences or accumulated knowledge, and by whether memory is expressed explicitly by conscious remembering or implicitly through changes in the speed or bias of performance in particular tasks. All forms of memory are based on changes in synaptic connections within neural circuits of each memory system. The strength of memory is modulated by emotional arousal and declines in aging.
There is a large variety of forms of memory beginning with elementary and non-associative types of memory, including habituation and sensitization, and reaching the most complex forms of associative memory, including episodic memory and semantic memory. Thus a central tenet in this field is that there are multiple memory systems, each characterized by distinct operating characteristics and brain pathways in which memories are embodied in the plasticity of information processing within the relevant neural circuit (Figure 1; Eichenbaum & Cohen, 2001). Importantly, the topic of memory is closely related to learning, which is the process by which memories are obtained, and learning is not exclusive to but includes various forms of conditioning and reinforcement learning, both of which are also covered as major topical areas in Scholarpedia.
A major breakthrough in understanding memory systems and their underlying brain mechanisms began with the study of a patient known by his initials H.M. (HM Patient; Scoville & Milner, 1957). This case involved an experimental surgical treatment for epilepsy in which the medial temporal lobe was removed. The surgery largely ameliorated the seizures but unexpectedly left H.M. with a severe amnesia which allows him to remember a limited amount of information for a short time (up to a few minutes). Despite his inability to remember new information, H.M. has considerable intact memories from his childhood and information obtained up to a few years before his surgery. From these observations researchers concluded that the parts of the medial temporal lobe that were removed in H.M, including the hippocampus and adjacent parahippocampal region, play a critical role in converting a short-term memory to a long-term, permanent memory store. Furthermore, the fact that H.M. retains memories for events that occurred remotely prior to his surgery indicated that these brain areas are not the site of permanent storage but instead play a role in the organization and permanent storage of memories elsewhere in the brain through the process known as memory consolidation.
Studies using functional brain imaging have confirmed that the hippocampus and parahippocampal region are activated during the encoding and retrieval of memories in humans, and these studies have also identified a large network of areas in the cerebral cortex that work together to support declarative memory, our ability for learning and consciously remembering everyday facts and events (Squire et al., 2004). The cortical areas include the association areas in the prefrontal, parietal, and temporal cortex plus the cingulum and retrosplenial areas, each of which also play distinct roles in complex aspects of perception, movement, emotion, and cognition. Functional imaging studies and studies on experimental animals, in which particular brain areas have been selectively removed and in which one can characterize the information encoded by single neurons, have shown that the hippocampus and the parahippocampal region play specific roles in memory (Eichenbaum et al., 2007; also see, Squire et al., 2007). The cortical-hippocampal system supports both our capacity for a sense of familiarity with previous exposure to specific stimuli and the ability for active recollection of prior experiences. Furthermore, this system is characterized by relational memory processing, in which items to be remembered are bound to associated items, to the context in which the experience occurred, to preceding and succeeding events, and to related memories (Cohen et al., 1997; Eichenbaum, 2004). Notably, false memory typically involves an error of relational processing that retrieves relations that did not actually occur but is embodied in the network of relational memories.
Short term forms of memory
Information from new experiences initially is stored in iconic memory and forms of short term memory that can support brief storage and immediate recall of substantial detail. Some of this information also enters working memory, a form of declarative memory that enables us to transiently retain and manipulate the information “on-line” in consciousness (Dobbins et al., 2002). Working memory depends on the prefrontal cortex as well as a large network of other cerebral cortical areas. Studies on experimental animals have shown that prefrontal neurons maintain relevant information during working memory and can flexibly combine different kinds of sensory information and abstract concepts and rules on which decisions are made (Miller, 2000). In humans, the prefrontal cortex is highly activated during the encoding, retrieval, maintenance, and manipulation of memories. Distinct areas within the prefrontal cortex support different executive functions in cognition, including selection, rehearsal, and monitoring of information being retrieved from long term memory. In performing these functions, the prefrontal cortex interacts with a large network of posterior cortical areas that encode, maintain, and retrieve specific types of perceptual information (Postle, 2006).
Episodic and semantic memory
The medial temporal lobe is critical to the processing and storage of episodic memories, our memories of specific personal experiences that happened at a particular place and time (Baddeley et al., 2002). Studies on human amnesic patients and on animals with experimental brain damage have shown different parts of the parahippocampal region play distinct roles in processing the “what”, “where”, and “when” information about unique events and the hippocampus links these features of an episodic memory (Eichenbaum et al., 2007). These linkages likely do not contain details of memories in the hippocampus, but instead serve to integrate information processed by numerous specific cortical areas. Thus the what-where-when information channeled in parallel into the medial temporal area are linked by the hippocampus, and those linkages connect the representations of each of those kinds of information in the relevant cerebral cortical areas. Additional experiences involving the same or related events, and perhaps memory replay during sleep (Wilson & McNaughton, 1994), generate repeated two-way interactions between widespread areas of the cerebral cortex such that the medial temporal lobe slowly facilitates interconnections between the representations in different cerebral cortical areas (Paller, 1997). Over a prolonged period, these interactions result in a substantial reorganization and consolidation of connections among cortical representations to incorporate the information contained in new episodes and abstracted semantic knowledge into permanent memory (McClelland et al., 1995).
Our permanent storehouse of memories, called semantic memory, involves a large distributed network of cortical areas that are involved in the perception, action, and analysis of the material being learned. Distinct cortical networks are specialized for processing particular types of materials, such as faces, houses, tools, actions, language, and other categories of knowledge. Studies of people with localized cortical damage areas have shown that specific areas and networks are critical to particular categories of semantic knowledge (Damasio et al., 1996). Also studies using functional imaging of normal humans have identified cortical networks that process particular categories of information, including faces, tools, animals, chairs, words, and houses (Martin et al., 1996). These same networks also perform different types of information processing, for example, distinguishing exemplars within a category or imagining the use of an object. Thus, a large and distinct network of cortical areas participates in any particular category of semantic processing. In addition, specific representations in these cortical areas can be shaped by training, allowing for the learning of perceptual skills (Karni, 1996), and are subject to a biasing towards non-conscious retrieval of recently experienced perceptions, called priming (Schacter & Buckner, 1998).
Whereas H.M. and amnesic patients can be severely impaired in declarative memory, they typically retain various forms of implicit memory, which involves the ability to acquire skills, habits, and preferences that can be expressed by improved performance or altered bias without conscious recollection (Schacter, 1987). These various forms of implicit memory are supported by a variety of memory systems and within the cerebral cortex, as in the case of priming introduced above. One implicit memory system involves cortical areas interacting with the striatum in support of procedural memory, the acquisition of skilled behavior and acquired habits. Procedural memories are expressed directly through activation of the brain’s motor coordination system (Packard & Knowlton, 2002). Additionally, memory involving fine precision of timing in motor learning depends on the cerebellum (Krupa et al., 1993). Another distinct memory system involves cortical and subcortical areas interacting with the amygdala to support the attachment of affective significance to otherwise neutral stimuli and events. This system expresses emotional memory by activating the hypothalamus and sympathetic nervous system, which generate emotional reactions and feelings (LeDoux, 1996). Thus, in addition to the declarative memory system, the procedural and emotional memory systems can store and distinctly express different forms of memory even for the same event.
Synaptic plasticity and memory
The cellular basis of memory involves activity dependent plasticity in synaptic connections. An important model in the study of the cellular basis of memory is the phenomenon of long-term potentiation (LTP), a long-lasting increase in the strength of a synaptic response following stimulation (Bliss et al., 2007). LTP is prominent in the hippocampus, as well as in the cerebral cortex and other brain areas that are involved in different forms of memory. LTP is typically induced by the co-occurence of excitatory input and intracellular depolarization at the so-called Hebbian synapse, involving N-methyl-d-aspartate (NMDA) receptors that allow the entry of Ca++ into the synapse, which activates cyclic adenosine monophosphate (cAMP). Subsequently, cAMP activates several kinases, some of which increase the number of synaptic receptors. In addition, cAMP activates cAMP-response element binding protein (CREB), which operates within the nucleus to activate a class of genes called immediate early genes, which, in turn, activate other genes that direct protein synthesis. Among the proteins produced is neurotrophin, which activates growth of the synapse. Thus, a series of molecular reactions plays a vital role in fixating the changes in synaptic function that occur in LTP.
Evidence that the permanent fixation of memories depends on this molecular and cellular cascade of events comes from studies showing that memory fixation can be halted by interference with the molecules in this cascade. Many studies have shown that drugs that block NMDA receptors, cAMP, CREB, or other molecules involved in protein synthesis block memory. These treatments are effective when given before or within minutes after learning, and but are not effective if they are delayed, indicating that the molecular cascade leading to protein synthesis is not essential to initial learning or to maintaining short-term memory, but is essential for permanent memory fixation. In addition, studies using genetically modified mice have shown that alterations in specific genes for these molecules can dramatically affect the capacities for LTP and memory fixation.
In addition to LTP, there is also mechanism that diminishes the strength of connections at infrequently used synapses called long term depression (LTD). LTD involves the same molecular substrates as LTP but occurs with different timing rules of activity at synapses. The combination LTP and LTD allow for a sophisticated reorganization of circuits that create neural representations of information. LTP and LTD occur among all brain structures that are known to participate in different kinds of memory. These cellular and molecular events occur on a timescale of seconds and minutes, are essential for the transition from short-term storage to long-term memory, and occur in every brain structure that participates in memory.
In addition, modulation of memory fixation occurs through emotional arousal and stress (McGaugh et al., 1996). Emotional arousal induces the release of glucocorticoids and adrenergic mechanisms via the amygdala, influencing memory fixation in both the declarative and procedural memory systems and in both animals and humans. Thus, emotionally intense experiences can lead to more vivid and lasting declarative memories, sometimes called flashbulb memories, and to deeply seated habits.
Aging and memory
Although aging causes amnesia involving some loss of all types of learning capacity, the most common and most prominent decline is in episodic memory (Wilson et al., 2006). Age related memory loss involves compromises both in the prefrontal cortex, resulting in loss of executive functions, and in the medial temporal lobe, resulting in an impairment in storing new memories. Normal aging is not caused by the loss of neurons, but rather by decreases in the number of synaptic connections and in loss of neuromodulation that normally activates and coordinates processing in these brain areas, particularly involving pathways that use the neurotransmitter acetylcholine. Age-related deterioration results in a slower rate of new learning and a rigidity to modifying existing representations based on new information.
- Baddeley, A., Conway, M., and Aggleton, J. (2002) Episodic Memory. New Directions in Research. Oxford University Press (New York).
- Bliss, T., Collingrindge, G, and Morris, R. (2007) Synaptic plasticity and the hippocampus. In The Hippocampus Book, Andersen, P., Morris, R., Amarakl, D, Bliss, T, and O’Keefe, J., eds. Oxford University Press, New York. pp. 343-474.
- Cohen, N.J., R.A. Poldrack and H. Eichenbaum (1997) Memory for items and memory for relations in the procedural/declarative memory framework. Memory 5:131-178. doi:10.1080/741941149.
- Damasio, H., Grabowski, T. J., Tranel, D., Hichwa, R. D., & Damasio, A. R. (1996). A neural basis for lexical retrieval. Nature, 380, 499–505. doi:10.1038/380499a0.
- Dobbins, I. G., Foley, H., Schacter, D. L., & Wagner, A. D. (2002). Executive control during episodic retrieval: Multiple prefrontal processes subserve source memory. Neuron, 35, 989-996. doi:10.1016/S0896-6273(02)00858-9.
- Eichenbaum, H. (2000). A cortical-hippocampal system for declarative memory. Nature Reviews Neuroscience, 1, 41-50.
- Eichenbaum, H. (2004) Hippocampus: Cognitive processes and neural representations that underlie declarative memory. Neuron 44:109-120.
- Eichenbaum, H., & Cohen, N. J. (2001) From Conditioning to Conscious Recollection: Memory Systems of the Brain. New York: Oxford University Press.
- Eichenbaum, H., Yonelinas A.R., and Ranganath, C. (2007) The medial temporal lobe and recognition memory. Annual Review of Neuroscience 20:123-152. doi:10.1146/annurev.neuro.30.051606.094328.
- Karni, A. (1996). The acquisition of perceptual and motor skills: A memory system in the adult human cortex. Cognitive Brain Research, 5, 39–48. doi:10.1016/S0926-6410(96)00039-0.
- Knowlton, B. J., Mangels, J. A., & Squire, L. R. (1996). A neostriatal habit learning system in humans. Science, 273, 1399–1401. doi:10.1126/science.273.5280.1399.
- Krupa, D. J., Thompson, J. K., & Thompson, R. F. (1993). Localization of a memory trace in the mammalian brain. Nature, 260, 989–991. doi:10.1126/science.8493536.
- LeDoux, J. (1996). The emotional brain. New York: Simon and Schuster.
- Martin, A., Wiggs, C. L., Ungerleider, L. G., & Haxby, J. V. (1996). Neural correlates of category-specific knowledge. Nature, 379, 649–652. doi:10.1038/379649a0.
- McClelland, J. L., McNaughton, B. L., & O'Reilly, R. C. (1995). Why are there complementary learning systems in the hippocampus and neocortex: Insights from the successes and failures of connectionist models of learning and memory. Psychological Review, 102, 419-457. doi:10.1037/0033-295X.102.3.419.
- McGaugh, J. L., Cahill, L., & Roozendaal, B. (1996). Involvement of the amygdala in memory storage: Interactions with other brain systems. Proceedings of the National Academy of Sciences, 93, 13508–13514. doi:10.1073/pnas.93.24.13508.
- Miller, E. K. (2000). The prefrontal cortex and cognitive control. Nature Reviews Neuroscience, 1, 59–65.
- Packard, M. G., & Knowlton, B. J. (2002). Learning and memory functions of the basal ganglia. Annual Review of Neuroscience, 25, 563–593.
- Paller, K. A. (1997). Consolidating dispersed neocortical memories: The missing link in amnesia. Memory, 5, 73–88. doi:10.1080/741941150.
- Postle, B. R. (2006). Working memory as an emergent property of the mind and brain. Neuroscience, 139, 23–38. doi:10.1016/j.neuroscience.2005.06.005.
- Schacter, D.L. (1987) Implicit memory: History and current status. J. Experimental Psychol. Learning, Memory, Cognition 13:501-518. doi:10.1037/0278-7318.104.22.1681.
- Schacter, D. L., & Buckner, R. L. (1998). Priming and the brain. Neuron, 20, 185–195. doi:10.1016/S0896-6273(00)80448-1.
- Scoville, W.B., Milner, B. 1957. Loss of recent memory after bilateral hippocampal lesions. J. Neurol. Neurosurg. Psychiat. 20: 11-12. doi:10.1136/jnnp.20.1.11.
- Squire LR, Stark CE, Clark RE. 2004. The medial temporal lobe. Annu Rev Neurosci 27: 279-306. doi:10.1146/annurev.neuro.27.070203.144130.
- Squire LR, Wixted JT, Clark RE. 2007. Recognition memory and the medial temporal lobe: a new perspective. Nat Rev Neurosci 8: 872 - 883. doi:10.1038/nrn2154.
- Wilson, I.A., Gallagher, M., Eichenbaum, H., and Tanila, H. (2006) Neurocognitive aging: Prior memories hinder new hippocampal encoding. Trends in Neurosciences 29:662-670. doi:10.1016/j.tins.2006.10.002.
- Wilson, M.A. and McNaughton, B.L. (1994) Reactivation of hippocampal ensemble memories during sleep. Science 265: 676-679. doi:10.1126/science.8036517.
- Nestor A. Schmajuk (2008) Classical conditioning. Scholarpedia, 3(3):2316. doi:10.4249/scholarpedia.2316.
- William D. Penny and Karl J. Friston (2007) Functional imaging. Scholarpedia, 2(5):1478. doi:10.4249/scholarpedia.1478.
- Norman M. White (2007) Multiple memory systems. Scholarpedia, 2(7):2663. doi:10.4249/scholarpedia.2663.
- Florentin Woergoetter and Bernd Porr (2008) Reinforcement learning. Scholarpedia, 3(3):1448. doi:10.4249/scholarpedia.1448.
- Eichenbaum, H. (2008) Learning and Memory. New York: W.W. Norton & Co.
- Squire, L. R., & Kandel, E. R.. (1999). Memory: From mind to molecules. New York: Freeman & Co.
Associative memory, Conditioning, Cortical memory, Emotional memory, Episodic memory, False memory, HM patient, Iconic memory, Implicit learning, Implicit memory, Memory modulation, Multiple memory systems, Relational memory, Serial learning, Sleep and learning,