Neurobiology of Learning and Memory 87 (2007) 9–20 www.elsevier.com/locate/ynlme Hippocampal CA1 spiking during encoding and retrieval: Relation to theta phase q Joseph R. Manns, Eric A. Zilli, Kimberly C. Ong, Michael E. Hasselmo, Howard Eichenbaum * Center for Memory and Brain, Boston University, Boston, MA 02215, USA Received 10 March 2006; revised 19 May 2006; accepted 22 May 2006 Available online 12 July 2006 Abstract The hippocampal theta rhythm is a prominent oscillation in the field potential observed throughout the hippocampus as a rat inves- tigates stimuli in the environment. A recent computational model [Hasselmo, M. E., Bodelon, C., & Wyble, B. P. (2002a). A proposed function for hippocampal theta rhythm: separate phases of encoding and retrieval enhance reversal of prior learning. Neural Computa- tion, 14, 793–817. Neuromodulation, theta rhythm and rat spatial navigation. Neural Networks, 15, 689–707] suggested that the theta rhythm allows the hippocampal formation to alternate rapidly between conditions that promote memory encoding (strong synaptic input from entorhinal cortex to areas CA3 and CA1) and conditions that promote memory retrieval (strong synaptic input from CA3 to CA1). That model predicted that the preferred theta phase of CA1 spiking should differ for information being encoded versus information being retrieved. In the present study, the spiking activity of CA1 pyramidal cells was recorded while rats performed either an odor-cued delayed nonmatch-to-sample recognition memory test or an object recognition memory task based on the animal’s spontaneous preference for novelty. In the test period of both tasks, the preferred theta phase exhibited by CA1 pyramidal cells differed between moments when the rat inspected repeated (match) and non-repeated (nonmatch) items. Also in the present study, additional modeling work extended the previous model to address the mean phase of CA1 spiking associated with stimuli inducing varying levels of retrieval relative to encoding, ranging from novel nonmatch stimuli with no retrieval to highly familiar repeated stimuli with extensive retrieval. The modeling results obtained here demonstrated that the experimentally observed phase differences are consistent with different levels of CA3 synaptic input to CA1 during recognition of repeated items. Ó 2006 Elsevier Inc. All rights reserved. Keywords: Hippocampus; Entorhinal; Theta; Model; Memory; Oscillation The hippocampus (defined here as CA fields, dentate and events (Gabrieli, 1998; Manns & Squire, 2002; Schact- gyrus, and subiculum) supports the capacity for declarative er, 1997). In other mammals, declarative memory is typi- memory in collaboration with the adjacent entorhinal, per- cally characterized as being spatial, temporal, associative, irhinal, and postrhinal (parahippocampal in primates) cor- or rapidly acquired (Burgess, Becker, King, & O’Keefe, tices (Brown & Aggleton, 2001; Eichenbaum, 2000; Squire, 2001; Eichenbaum & Cohen, 2001; Morris, 2001; O’Reilly Stark, & Clark, 2004). In humans, declarative memory is & Rudy, 2001). often expressed though the conscious recollection of facts An important area of research has been the attempt to understand the computational principles of how the hippo- campus interacts with adjacent cortical areas to support q Supported by NIMH MH051570, NIMH MH068982, NIMH declarative memory (Hasselmo & McClelland, 1999; Lor- MH60013, NIDA DA16454 (as part of the program for Collaborative incz & Buzsaki, 2000; McClelland, McNaughton, & Research in Computational Neuroscience), and the NSF Science of O’Reilly, 1995; Nadel, Samsonovich, Ryan, & Moscovitch, Learning Center SBE0354378 (CELEST). * Corresponding author. Fax: +1 617 353 1414. 2000; O’Reilly & Rudy, 2001; Teyler & Discenna, 1986). E-mail address: [email protected] (H. Eichenbaum). One particularly compelling puzzle concerns how the 1074-7427/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.nlm.2006.05.007 10 J.R. Manns et al. / Neurobiology of Learning and Memory 87 (2007) 9–20 hippocampus attains a balance between encoding and Encoding Retrieval retrieval processes. Presumably, the hippocampus partici- pates in both encoding and retrieving memory, yet it is Entorhinal Cortex unclear how it does so in a way that optimally compares Entorhinal Cortex new and retrieved information without suffering from potentially incapacitating interference effects. Some attempts to solve this puzzle have focused on the CA3 CA1 CA3CA3 CA1CA1 hippocampal theta rhythm (Hasselmo, 2005). The theta CA3 CA1 rhythm is a 4–12 Hz oscillation that is prominent in the local field potential recorded throughout the hippocampal region and is thought to play an important role in success- ful declarative memory (see Buzsaki, 2002 for a review). One possible role for the theta rhythm is that it appears Fig. 1. Model of hippocampal network during encoding and retrieval (see to modulate the magnitude of synaptic currents at synapses text for details). During encoding, strong synaptic input from entorhinal in the hippocampus (Brankack, Stewart, & Fox, 1993; Orr, cortex reaches CA1 and CA3 near the trough of the theta rhythm recorded Rao, Houston, McNaughton, & Barnes, 2001; Wyble, Lin- at the hippocampal fissure. During retrieval, strong synaptic input from CA3 reaches CA1 near the peak of fissure theta. ster, & Hasselmo, 2000). This phase specificity of the theta rhythm differs across synapses of the hippocampus. In par- ticular, the synaptic input from entorhinal cortex to CA1 is 2002a, 2002b), the maximum synaptic input to CA1 from strongest at the troughs of the theta rhythm recorded at the entorhinal cortex was described as being 180 degrees out hippocampal fissure and is weakest at the peaks of the fis- of phase from the maximum synaptic input from CA3. sure theta (Brankack et al., 1993). In contrast, the synaptic Yet whether or not the spiking activity of CA1 would transmission from CA3 to CA1 is strongest at the peaks of also differ by 180 degrees between moments of strong the fissure theta and weakest at the troughs of the fissure encoding and strong retrieval was not specified. theta (Brankack et al., 1993). In addition, different phases The prediction that the preferred theta phase of CA1 of theta rhythm are associated with differences in the induc- spiking should differ between strong encoding and strong tion of synaptic modification (Holscher, Anwyl, & Rowan, retrieval was tested in the current study using two recogni- 1997; Hyman, Wyble, Goyal, Rossi, & Hasselmo, 2003; tion memory protocols. The activity of CA1 pyramidal Pavlides, Greenstein, Grudman, & Winson, 1988). For cells was recorded while rats performed either a novel example, in awake, behaving rats, stimulation in stratum object exploration task (Experiment 1) or an odor-cued radiatum of CA1 led to either long-term potentiation delayed nonmatch to sample (DNMS) task (Experiment 2). (LTP) or long-term depression (LTD), depending on We reasoned that inspection of any item would likely whether the stimulation was applied at the peaks or elicit some degree of encoding and some degree of retrieval, troughs of the locally recorded theta rhythm (Hyman regardless of whether the item was entirely novel or was et al., 2003; also see McCartney, Johnson, Weil, & Givens, very familiar. The objective of our analyses was therefore 2004 for a similar result). to compare the preferred theta phase of CA1 spiking Based on these physiological results, a computational between two conditions that differed only in the proportion model was recently proposed to describe how the theta of encoding and retrieval that would be expected to occur. rhythm might reflect the network dynamics that allow Thus, the preferred phase of theta at which cells tended to the hippocampus to oscillate rapidly between conditions spike was compared during the test period of each experi- promoting encoding and conditions promoting retrieval ment between instances when the rat encountered a repeat- (Hasselmo, Bodelon, & Wyble, 2002a; Hasselmo, Hay, ed odor (or object) and instances in which the rat Ilyn, & Gorchetchnikov, 2002b). The model demonstrates encountered a non-repeated odor (or object). The idea that the best retrieval of memories occurs if, over the was that inspection of a repeated item would bias activity course of each theta cycle, the hippocampus transitions in the hippocampus towards retrieval of the memory of from strong input from entorhinal cortex to strong inter- that previously experienced item whereas inspection of a nal retrieval (see Fig. 1). In particular, the model speci- novel item would bias activity towards encoding the new fied that best performance occurs if input to CA1 information. Importantly, during the test period, the rat’s would predominantly come from entorhinal cortex at overt behavior and expectations were constant as it began the trough of fissure theta but would instead mainly sampling of both the novel and familiar items. This consis- come from CA3 at the peak of the fissure theta. The ini- tency in behavior and expectancy would not be available in tial model focused on plasticity and synaptic transmission comparisons between items encountered during the sample and did not derive results for spiking activity. However, period and test period. In both experiments, a difference in a direct extension of the model predicts that the preferred preferred theta phase between conditions was observed for theta phase of CA1 spiking should also differ between CA1 spiking. However, the observed difference in