DispatchR167

Cognitive Neuroscience: NavigatingHuman Verbal Memory

A recent study in humans shows that the same neurons that represent locationduring spatial navigation also code elements of verbal recall. This study thusprovides a critical missing link between two previously unconnected functionsof the hippocampus.

Arne D. Ekstrom1,2,3

Decades of work have established theimportance of neurons called placecells in spatial navigation [1–5]. Aseparate literature, largely based onneuropsychological and brain imagingstudies of human verbal memory,demonstrates the importance of thehippocampus for rememberingspecific events (termed episodicmemory) [6,7]. Yet whether place cells,which are primarily located in thehippocampus, are part of a broadernetwork of cells subserving episodicmemory, is not known. A recentstudy by Miller et al. [8] shows thathippocampal place cells respond in theabsence of spatial cues when spatialcontext is nonetheless retrieved duringverbal free recall.

One of the most striking findingsfrom rodent electrophysiology is theplace cell, first noted by O’Keefeand Dostrovsky [1]. These neurons,located primarily in the hippocampus,increase their firing rate when theanimal is at specific spatial locations.The collection of place cells in anenvironment provides a fairly accurate‘map’ of a rat’s position within theenvironment [2]. The prevalence ofplace cells within the hippocampus,paired with findings that lesions to thehippocampus severely impair theability of a rat to navigate usingexternal-referenced landmarks [9],initially argued for a primary role of thehippocampus in spatial navigation[2,3]. More recently, neurons in themedial entorhinal cortex, which showregularly spaced ‘grid-like’ firingpatterns [10,11], argue for a moregeneral role for the hippocampalcomplex — hippocampus andsurrounding entorhinal andparahippocampal cortices — inspatial navigation.

Patients with damage to medialtemporal lobe, which includes thehippocampus, exhibit severeimpairments in their ability to verballyrecall recently experienced events,

termed ‘episodic memories’ [12].Fueled by this critical advance, muchsubsequent work in humans hasfocused on the role for the medialtemporal lobe in episodic memoryretrieval [13,14]. Episodic memory isthought to involve representation ofobject-related information (forexample, a ‘jar’) in surroundingperirhinal cortex and binding of thisinformation with specific contextualdetails in the hippocampus [6].Benchmark findings in this literaturehave demonstrated greaterhippocampal involvement whenparticipants must remember thelocation or color of a recently learnedword compared to simply indicatingwhether the word was studiedbefore [6,7].

This binding process, wherebyevent-details are combined with objectrepresentations in the hippocampus, isthought to be a fundamental function ofthe hippocampus in humans [6]. Yetgiven the verbal nature of many humanepisodic memory paradigms, exactlyhow and in what manner contextualrepresentations emerge in the firstplace remains unknown. Severalstudies have suggested that contextrepresentation could emerge from theposterior parahippocampal cortex [6],which shows unusual sensitivity whenpeople view scenes compared toother objects [15]. In this way,context-related and object-specificresponses could arrive in thehippocampus through separate yetparallel streams involving medialand lateral entorhinal cortex [16]. Yetthe specific nature of contextualrepresentations in the hippocampusand its relation to object coding,particularly during verbal memoryretrieval, remains unclear.

Capitalizing on a rare situationinvolving epilepsy patients withelectrodes implanted for surgicalmonitoring, Miller et al. [8] recordedsingle neurons directly from the medialtemporal lobes. This allowed theauthors to observe directly how

neurons changed their firing rate as afunction of spatial context during bothnavigation and verbalmemory retrieval.Building on past studies that haveemployed virtual reality with humansand non-human primates to identifyplace responsive neurons [4,5,11], theauthors had patients explore a virtualenvironment by searching for certainstores. Upon locating a specific store,an object appeared, either visually orauditorily, at that store. Thus, if thepatient delivered to the ‘Pickle Store’, ajar might appear when the patientfound that store; other stores involveddifferent object pairings. Analyzingpatient trajectories, Miller et al. [8]demonstrated that significant numbersof medial temporal lobe neuronsincreased their firing rate at specificspatial locations within the virtualenvironment, consistent with pastwork in humans and non-humanprimates [4,5,11].Yet the crucial innovation introduced

by Miller et al. [8] occurred followingexploration of the spatial environmentwhen patients freely verbally recalledobjects from the environment.Critically, when participants recalled anobject such as a jar, place cells withspatial receptive fields near where thisobject was dropped off were moreactive than cells with spatial receptivefields further from that location.Additionally, cellular firing rate patternsduring free recall of objects moreclosely resembled those of place cellswith spatial receptive fields near thatobject compared to place cells withspatial receptive fields further awayfrom the object. Together, thesefindings suggest that when we recallobjects from environments wehave recently visited, place cellsrepresenting areas at or nearby thelocations at which we encounteredthese objects are also active.These findings thus help resolve an

important debate about the functionof the hippocampus. Consistent withtheories positing a role for thehippocampus (across species) inspatial navigation [2,3], Miller et al. [8]show neurons coding specific spatiallocations during virtual exploration.Importantly, the authors demonstratethat these cells are also active duringverbal recall such that they provideinformation about the spatial contextthat participants retrieve. Thesefindings thus demonstrate that theneural machinery of the hippocampus,

Current Biology Vol 24 No 4R168

of which place cells are thought to befundamental, are also active duringverbal tasks that involve retrieval ofthat spatial context. These findingsthus suggest that the spatial functionsof the hippocampus play a moregeneral role in episodic memoryprocesses.

As with any important advance,the Miller et al. [8] study also raisesseveral novel and important researchquestions for future consideration.First, given the importance of objectrepresentations within the medialtemporal lobe, how exactly does theactivity of cells representing specificobjects and stores activated duringrecall [17,18] relate to place cellactivity [16,18,19]? Second, given thepotential importance of separatestreams of input for object and spatialcontext into the hippocampus, howexactly is this information combinedfrom perirhinal and parahippocampalcortices [16]? One intriguingpossibility is that the CA1 subfield ofthe hippocampus, which receivesinput from both perirhinal andparahippocampal cortex (via entorhinalcortex), as well as subfield CA3,may play a role in integrating thisinformation [20]. Third, to what extentare place cells and spatial context partof a more general contextual codingapparatus, such as temporal oremotional context [6]? These issuespertaining to object and contextualrepresentation are critical next stepsto better decoding how memory is

represented within the medial temporallobes and further bridging the gapbetween research on rodents andhumans.

References1. O’Keefe, J., and Dostrovsky, J. (1971). The

hippocampus as a spatial map. Preliminaryevidence from unit activity in the freely-movingrat. Brain Res. 34, 171–175.

2. O’Keefe, J., and Nadel, L. (1978). TheHippocampus as a Cognitive Map (Oxford:Clarendon Press).

3. Gallistel, C.R. (1990). The Organization ofLearning (Cambridge, MA: MT Press).

4. Ekstrom, A.D., Kahana, M.J., Caplan, J.B.,Fields, T.A., Isham, E.A., Newman, E.L., andFried, I. (2003). Cellular networks underlyinghuman spatial navigation. Nature 425,184–188.

5. Hori, E., Tabuchi, E., Matsumura, N.,Tamura, R., Eifuku, S., Endo, S., Nishijo, H., andOno, T. (2003). Representation of place bymonkey hippocampal neurons in real andvirtual translocation. Hippocampus 13,190–196.

6. Diana, R.A., Yonelinas, A.P., and Ranganath, C.(2007). Imaging recollection and familiarityin the medial temporal lobe: a three-component model. Trends Cogn. Sci. 11,379–386.

7. Davachi, L., Mitchell, J.P., and Wagner, A.D.(2003). Multiple routes to memory: distinctmedial temporal lobe processes build item andsource memories. Proc. Natl. Acad. Sci. USA100, 2157–2162.

8. Miller, J.F., Neufang, M., Solway, A., Brandt, A.,Trippel, M., Mader, I., Hefft, S., Merkow, M.,Polyn, S.M., Jacobs, J., et al. (2013). Neuralactivity in human hippocampal formationreveals the spatial context of retrievedmemories. Science 342, 1111–1114.

9. Morris, R.G.M., Garrud, P., Rawlins, J.N.P., andO’Keefe, J. (1982). Place navigation impaired inrats with hippocampal lesions. Nature 297,681–683.

10. Fyhn, M., Molden, S., Witter, M.P., Moser, E.I.,and Moser, M.B. (2004). Spatial representationin the entorhinal cortex. Science 305,1258–1264.

11. Jacobs, J., Weidemann, C.T., Miller, J.F.,Solway, A., Burke, J.F., Wei, X.X., Suthana, N.,

Sperling, M.R., Sharan, A.D., Fried, I., et al.(2013). Direct recordings of grid-like neuronalactivity in human spatial navigation. Nat.Neurosci. 16, 1188–1190.

12. Scoville, W.B., and Milner, B. (1957). Loss ofrecent memory after bilateral hippocampallesions. J. Neurol. Neurosurg. Psych. 20,11–21.

13. Squire, L.R., Stark, C.E., and Clark, R.E. (2004).The medial temporal lobe. Annu. Rev. Neurosci.27, 279–306.

14. Tulving, E. (2002). Episodic memory: from mindto brain. Annu. Rev. Psychol. 53, 1–25.

15. Epstein, R., and Kanwisher, N. (1998). A corticalrepresentation of the local visual environment.Nature 392, 598–601.

16. Hargreaves, E.L., Rao, G., Lee, I., andKnierim, J.J. (2005). Major dissociationbetween medial and lateral entorhinal input todorsal hippocampus. Science 308,1792–1794.

17. Ekstrom, A.D., Viskontas, I., Kahana, M.J.,Jacobs, J., Upchurch, K., Bookheimer, S., andFried, I. (2007). Contrasting roles of neuralfiring rate and local field potentials inhuman memory. Hippocampus 17,606–607.

18. Wood, E.R., Dudchenko, P.A., andEichenbaum, H. (1999). The global record ofmemory in hippocampal neuronal activity.Nature 397, 613–616.

19. Eichenbaum, H. (2004). Hippocampus:cognitive processes and neural representationsthat underlie declarative memory. Neuron 44,109–120.

20. Katz, Y., Kath, W.L., Spruston, N., andHasselmo, M.E. (2007). Coincidence detectionof place and temporal context in a networkmodel of spiking hippocampal neurons. PLoSComput. Biol. 3, e234.

1Center for Neuroscience, University ofCalifornia, Davis, 1544 Newton Court, Davis,CA 95618, USA. 2Department of Psychology,University of California, Davis, 1 Shields Ave,Davis, CA 95616, USA. 3NeuroscienceGraduate Group, University of California,Davis, 1 Shields Ave, Davis, CA 95616, USA.E-mail: adekstrom@ucdavis.edu

http://dx.doi.org/10.1016/j.cub.2013.12.043