BEHAVIORAL AND BRAIN SCIENCES (1997) 20, 597±655 Printed in the United States of America Long-term potentiation: What's learning got to do with it? Tracey J. Shors Department of Psychology and Program in Neuroscience, Princeton University, Princeton, NJ 08544 Electronic mail: shors࠾princeton.edu Louis D. Matzel Department of Psychology, Program in Biopsychology and Behavioral Neuroscience, Rutgers University, New Brunswick, NJ 08903 Electronic mail: matzel࠾rci.rutgers.edu Abstract: Long-term potentiation (LTP) is operationally defined as a long-lasting increase in synaptic efficacy following high-frequency stimulation of afferent fibers. Since the first full description of the phenomenon in 1973, exploration of the mechanisms underlying LTP induction has been one of the most active areas of research in neuroscience. Of principal interest to those who study LTP, particularly in the mammalian hippocampus, is its presumed role in the establishment of stable memories, a role consistent with ªHebbianº descriptions of memory formation. Other characteristics of LTP, including its rapid induction, persistence, and correlation with natural brain rhythms, provide circumstantial support for this connection to memory storage. Nonetheless, there is little empirical evidence that directly links LTP to the storage of memories. In this target article we review a range of cellular and behavioral characteristics of LTP and evaluate whether they are consistent with the purported role of hippocampal LTP in memory formation. We suggest that much of the present focus on LTP reflects a preconception that LTP is a learning mechanism, although the empirical evidence often suggests that LTP is unsuitable for such a role. As an alternative to serving as a memory storage device, we propose that LTP may serve as a neural equivalent to an arousal or attention device in the brain. Accordingly, LTP may increase in a nonspecific way the effective salience of discrete external stimuli and may thereby facilitate the induction of memories at distant synapses. Other hypotheses regarding the functional utility of this intensely studied mechanism are conceivable; the intent of this target article is not to promote a single hypothesis but rather to stimulate discussion about the neural mechanisms underlying memory storage and to appraise whether LTP can be considered a viable candidate for such a mechanism. Keywords: arousal; attention; calcium; classical conditioning; Hebbian synapses; hippocampus; memory systems; NMDA; spatial learning; synaptic plasticity; theta rhythm 1. Introduction underestimates the research effort, insofar as many articles that address LTP do not use ªlong-term potentiationº in Few topics in neurobiology have attracted as much atten- the title or they refer to the same phenomenon by a tion or resources over the past 20 years as the phenomenon different name (e.g., ªlong-term enhancementº; Mc- of LTP (long-term potentiation), a putative mechanism for Naughton et al. 1986). the induction of stable memories in the mammalian brain. The concerted attention that LTP has attracted over time Long-term potentiation is typically expressed as an in- perhaps carries no surprise for those familiar with the crease in synaptic efficacy lasting from hours to days fol- search for the engram (a neural memory store) and the lowing brief tetanic (high-frequency) stimulation of an associated mechanism that could account for its formation. afferent pathway. [See Vanderwolf & Robinson's ªReticulo- Prior to the observation of LTP, the search had produced Cortical Activity and Behavior. BBS 4(3) 1981.] Thus, fol- virtually no viable candidate mechanisms, at least for the lowing LTP induction, a fixed amount of presynaptic stim- vertebrate nervous system (cf. Kandel & Tauc 1965a; ulation induces a ªpotentiatedº postsynaptic response, for 1965b). In this regard, LTP has been and still may be the example, an increase in EPSPs (excitatory post-synaptic best candidate. In several recent reviews, various authors potentials). The phenomenon of LTP was initially observed have concluded not only that LTP is a viable mechanism for in 1966 by Terje Lomo, then working in the laboratory of the induction and storage of memories but that it is the Per Andersen. In 1973, the first full article described LTP most promising candidate (e.g., Morris et al. 1991). In one in the hippocampus of the rabbit, a collaborative effort article (Martinez & Derrick 1996), the authors review between Lomo and Timothy Bliss (see also Bliss & Gard- recent evidence suggesting that the link between LTP and ner-Medwin 1973). By 1989, the U.S. National Library of memory is in some cases tenuous, and in others even Medicine listed some 312 articles with the term ªlong-term contradictory. Nevertheless, they conclude that ªmost evi- potentiationº in the title, and, in the 1990s alone, over dence firmly supports a role for LTP in learning and 1,000 additional articles have appeared. This search vastly memoryº (see also Eichenbaum & Otto 1993). This conclu- ᮊ 1997 Cambridge University Press 0140-525X/XX $9.00ϩ.10 597 Shors & Matzel: Long-term potentiation sion is based, in part, on a commonly echoed assertion that, (1973), in the CA3 pyramidal cells by stimulation of the although no direct evidence links LTP to memory, no better mossy fibers (see, e.g., Alger & Teyler 1976; Yamamoto & mechanism has been postulated. This assertion is encom- Chujo 1978), and in the CA1 pyramidal cells by stimulation passed by the broader argument that a good theory should of the Schaffer collateral branches of the CA3 neurons not be abandoned until a better one replaces it, an approach (Andersen et al. 1977; Schwartzkroin & Wester 1975). The with obvious merit. On the other hand, explicit confidence initial description of LTP in the hippocampus was probably in the validity of a prevailing theory can interfere with the fortuitous for memory research; had LTP first been identi- development of viable alternatives and new approaches to a fied in a brain region with less of a historical link to memory problem. Einstein once stated that ªit is the theory which formation (see, e.g., Olds 1955; Scoville & Milner 1957), it decides what we can observeº (see also Kuhn 1973). A might not have received such focused attention. Since flawed theory, the explanatory value of which is outweighed 1973, however, LTP has been found to occur in many brain by the inconsistencies that it introduces, can serve only as a regions, including the piriform (Stripling et al. 1988), ento- detriment to empirical progress. To the extent that a theory rhinal (Wilhite et al. 1986), and prefrontal (Laroche et al. is maintained by popular consensus, ªwhat we can observeº 1989) cortices, the septum (Racine et al. 1983), the auto- will necessarily be obscured by the convictions that a nomic (Libet et al. 1975) and superior (Brown & McAfee theory's advocates embrace. 1982) cervical ganglia, and the ventral horn of the spinal Given the vast amount of attention that LTP has gener- cord (Pockett & Figurov 1993). Furthermore, LTP is not ated over the past 20 years, it seems an appropriate time to limited to the mammalian brain but has been described in review the cellular and behavioral characteristics of LTP other vertebrates as well, such as the goldfish (Lewis & that led us to consider it as a memory device in the first Teyler 1986; Yang et al. 1990), bullfrog (Koyano et al. 1985), place. We should evaluate whether these properties remain bird (Scott & Bennett 1993), and lizard (Larson & Lynch viable features of a memory device and, if so, whether LTP 1985) and also in some invertebrates (Glanzman 1995; remains the most viable mechanism to serve that broader Walters & Byrne 1985). Because negative findings are function. Of particular concern here is a distinction that we usually not definitive, it cannot be said with certainty that will draw between LTP and the formation and storage of LTP cannot be induced in a particular brain region, but it is memories versus a link between LTP and the processes that safe to say that phenomena fitting the general description of ªinfluenceº the formation and storage of memories. By LTP occur ubiquitously throughout the nervous system. If ªinfluence,º we mean that LTP may be neither a necessary LTP is a ubiquitous feature of the nervous system, what nor a sufficient condition for the actual storage of memo- might that mean with respect to its potential role in learning ries, but LTP or an endogenous equivalent could act to and memory? Moreover, if LTP is indeed a learning and facilitate and maintain learning indirectly by altering the memory device, what would such a wide distribution tell us organism's responsiveness to, or perception of, environ- about the neural mechanisms of memory formation? mental stimuli. In this target article, we first review a Most researchers would agree that memory formation number of the cellular properties intrinsic to LTP, with a requires, or at least utilizes, wide and distributed brain particular emphasis on hippocampal LTP and the charac- regions, and the hippocampus is clearly not the only ªstor- teristics most commonly presented as evidence for its ageº site for memory; humans and infrahumans do not relationship to memory. It is important to stress that, even if require a hippocampus to acquire many forms of memory, hippocampal LTP was the ªlearning mechanism,º we would and, even in tasks dependent on the hippocampus for not expect individual synapses to express characteristics of acquisition, the structure is typically not required for later learning and memory processes. Nevertheless, we discuss retrieval. If we begin with the premise that many memories them because they are the features commonly cited as are not actually stored in the hippocampus, then what evidence for the role of LTP in learning, and this will allow function might LTP serve there? Before discussing the role us to evaluate the overall consistency of the evidence of LTP in memory or any behavioral processes, however, we supporting LTP as a mechanism of memory storage.