Memories of Different Durations Form by Different Neurochemical Mechanisms

Memories differ markedly in how long they last, from iconic and short-term memories to long-term and permanent memories, as we saw in Chapter 17. Behavioral evidence suggests that memories of different lengths reflect the operation of different neural processes. For example, individuals who have no trouble forming short-term memories—such as H.M. in Chapter 17—may have an impaired ability to form long-term memories. With habituation and sensitization, the formation of relatively short-term memories in Aplysia involves mainly neurochemical changes at existing synapses, whereas the formation of long-term memories also involves structural changes in existing synapses and changes in the number of synapses.

Drugs have been used extensively to study memory formation; this approach has led to many interesting discoveries and to new concepts. Unlike brain lesions or other permanent interventions, many chemical treatments are advantageous for this research because they are reversible. Drugs can produce relatively brief, accurately timed effects, and subjects can be tested in their normal state both before and after treatment.

Different agents appear to affect different stages of memory formation. This result has given rise to the concept of sequential neurochemical processes in memory formation. Since about 1960, much research on the formation of long-term memory (e.g., H. P. Davis and Squire, 1984; Flood et al., 1977; M. R. Rosenzweig, 1984) has centered on the hypothesis that LTM requires increased protein synthesis during the minutes (perhaps hours) that follow training. More recently, investigators have studied the neurochemical processes involved in earlier stages of memory formation—short-term and intermediate-term memories (Gibbs and Ng, 1977; Mizumori et al., 1985; M. R. Rosenzweig et al., 1992).

Gibbs and Ng (1977) found evidence that short-term, intermediate-term, and long-term types of memory in chicks reflect three sequentially linked neurochemical processes. Their experiments studied chicks trained with a single-trial peck avoidance response: the chicks pecked at a small, shiny bead coated with an aversive liquid; after making a single peck, the chicks usually avoided a similar bead, whether it was presented minutes, hours, or even days later.

The formation of memory for the unpleasant experience could be impaired if an amnestic (memory-impairing) agent were administered close to the time of training. Gibbs and Ng reported that different families of amnestic agents caused memory to fail at different times after training (Figure 1). The agents that caused memory failure by about 5 minutes after training were considered to prevent the formation of STM, those that caused failure about 15 minutes after training were thought to prevent the formation of ITM, and those that caused memory to fail by about 60 minutes after training were thought to be affecting LTM.

Figure 1 Timing of the Effects of Different Amnestic Agents   Because different amnestic agents are thought to impair different stages of memory, they provide a means of observing the duration of each stage. (After Gibbs and Ng, 1977.)

Here are some examples of effects caused by different agents: Potassium chloride (KCl), the NMDA receptor antagonist APV, and lanthanum chloride—all of which inhibit the influx of Ca2+ ions—prevent the formation of STM (and therefore of the subsequent ITM and LTM stages). Ouabain, which inhibits Na+–K+ ATPase, prevents the formation of ITM (and therefore also LTM). Drugs that disrupt the process of protein synthesis in neurons prevent the formation of only LTM. Many agents have been used to disrupt memory at various stages, and although there are some inconsistencies, most of the data confirm at least three stages of memory formation.

The formation of long-term memory requires protein synthesis

Experiments to test the hypothesis that the formation of LTM requires protein synthesis have employed both behavioral intervention (in the form of training) and somatic intervention (in the form of agents that inhibit protein synthesis). Training increases the branching of dendrites and the number of synaptic contacts (Black and Greenough, 1998). The enlarged outgrowths of the neurons are made, in part, of proteins. Furthermore, direct measures of protein in the cerebral cortex of rats show a significant increase with enriched experience (E. L. Bennett et al., 1969).

The other side of the story is that inhibiting the synthesis of proteinxs in the brain at the time of training can prevent the formation of LTM, even though this inhibition does not interfere with acquisition or retrieval during tests of STM or ITM. The antibiotic anisomycin has been used extensively because at proper doses it is very effective at inhibiting protein synthesis without causing toxic side effects. For example, anisomycin treatment prevents LTM storage in mice, without affecting STM (E. L. Bennett et al., 1972; Flood et al., 1973). It may seem strange that the brain can get along with protein synthesis almost completely blocked for hours, but the fact is that cells contain large supplies of proteins; existing enzymes, for example, can continue directing the cell’s metabolism. With repeated administration of anisomycin, Flood et al. (1975, 1977) demonstrated that even relatively strong training could be overcome by inhibition of protein synthesis; the stronger the training was, the longer the inhibition had to be maintained to cause amnesia.

Protein synthesis involved in the formation of LTM appears to occur in two successive waves—the first about 1 hour after training, the second about 5 to 8 hours after training (Matthies, 1989). Administering inhibitors of protein synthesis so that they were effective at either of these periods prevented the formation of LTM, presumably by preventing the structural changes in the neurons that would normally encode the memory trace.


Bennett, E. L., Orme, A. E., and Hebert, M. (1972). Cerebral protein synthesis inhibition and amnesia produced by scopolamine, cycloheximide, streptovitacin A, anisomycin, and emetine in rat. Federation Proceedings, 31, 838.

Bennett, E. L., Rosenzweig, M. R., and Diamond, M. C. (1969). Rat brain: Effects of environmental enrichment on wet and dry weights. Science, 163, 825–826.

Black, J. E., and Greenough, W. T. (1998). Developmental approaches to the memory process. In J. L. Martinez, Jr., and R. P. Kesner (Eds.), Neurobiology of learning and memory (pp. 55–88). San Diego, CA: Academic Press.

Davis, H. P., and Squire, L. R. (1984). Protein synthesis and memory: A review. Psychological Bulletin, 96, 518–559.

Flood, J. F., Bennett, E. L., Orme, A. E., and Rosenzweig, M. R. (1975). Relation of memory formation to controlled amounts of brain protein synthesis. Physiology & Behavior, 15, 97–102.

Flood, J. F., Bennett, E. L., Rosenzweig, M. R., and Orme, A. E. (1973). The influence of duration of protein synthesis inhibition on memory. Physiology & Behavior, 10, 555–562.

Flood, J. F., Jarvik, M. E., Bennett, E. L., Orme, A. E., et al. (1977). The effect of stimulants, depressants and protein synthesis inhibition on retention. Behavioral Biology, 20, 168–183.

Gibbs, M. E., and Ng, K. T. (1977). Psychobiology of memory: Towards a model of memory formation. Biobehavioral Reviews, 1, 113–136.

Matthies, H. (1989). Neurobiological aspects of learning and memory. Annual Review of Psychology, 40, 381–404.

Mizumori, S. J. Y., Rosenzweig, M. R., and Bennett, E. L. (1985). Long-term working memory in the rat: Effects of hippocampally applied anisomycin. Behavioral Neuroscience, 99, 220–232.

Rosenzweig, M. R. (1984). Experience, memory, and the brain. American Psychologist, 39, 365–376.

Rosenzweig, M. R., Bennett, E. L., Martinez, J. L., Colombo, P. J., et al. (1992). Studying stages of memory formation with chicks. In L. R. Squire and N. Butters (Eds.), Neuropsychology of memory (2nd ed., pp. 533–546). New York: Guilford.