1066 nature neuroscience • volume 3 no 11 • november 2000 news and views ticity. The exchange of AMPA-R subunits is suggestive of a form of synaptic tagging or synaptic memory 7 , in which informa- tion is encoded at individual synapses, perhaps via a change in the composition of synaptic proteins, such that enhanced transmission is conserved despite protein turnover. The authors have proposed a related model 3 . The findings of Zhu et al. provide insights into the mechanisms involved in the maturation of functional excitatory synapses. However, before completely accepting the idea that excitatory synaps- es routinely begin their lives in relative silence with only NMDA-Rs and become functional components of excitatory neural circuitry by adding AMPA-Rs in an activity- and NMDA-R-dependent manner, a few conflicting observations should be mentioned. First, in acutely dissociated neurons in culture, function- al AMPA-R-containing synapses clearly can form without activation of any glu- tamate receptor, including NMDA-Rs 8–10 , although extended blockade of NMDA- Rs may decrease the overall AMPA-R content of synapses 10,11 . Second, a recent study, in which individual synapses were directly visualized in culture soon after they formed, found that relatively few nascent synapses express only NMDA-Rs and that AMPA-Rs can be found at synapses at the very earliest stage of synaptogenesis 12 . Third, prolonged block- 1. Zhu, J. J., Esteban, J. A., Hayashi, Y. & Malinow, R. Nat. Neurosci. 3, 1098–1106 (2000). 2. Malenka, R. C. & Nicoll, R. A. Science 285, 1870–1874 (1999). 3. Malinow, R., Mainen, Z. F. & Hayashi, Y. Curr. Opin. Neurobiol. 10, 352–357 (2000). 4. Feldman, D. E. & Knudsen, E. I. Neuron 20, 1067–1071 (1998). 5. Hayashi, Y. et al. Science 287, 2262–2267 (2000). 6. Dingledine, R., Borges, K., Bowie, D. & Traynelis, S. F. Pharmacol. Rev. 51, 7–61 (1999). 7. Frey, U. & Morris, R. G. Trends Neurosci. 21, 181–188 (1998). 8. Rao, A. & Craig, A. M. Neuron 19, 801–812 (1997). 9. O’Brien, R. J. et al. J. Neurosci. 17, 7339–7350 (1997). 10. Gomperts, S. N., Carroll, R., Malenka, R. C. & Nicoll, R. A. J. Neurosci. 20, 2229–2237 (2000). 11. Liao, D., Zhang, X., O’Brien, R., Ehlers, M. D. & Huganir, R. L. Nat. Neurosci. 2, 37–43 (1999). 12. Friedman, H. V., Bresler, T., Garner, C. C. & Ziv, N. E. Neuron 27, 57–69 (2000). 13. Turrigiano, G. G., Leslie, K. R., Desai, N. S., Rutherford, L. C. & Nelson, S. B. Nature 391, 892–896 (1998). 14. O’Brien, R. J. et al. Neuron 21, 1067–1078 (1998). 15. Carroll, R. C., Lissin, D. V., von Zastrow, M., Nicoll, R. A. & Malenka, R. C. Nat. Neurosci. 2, 454–460 (1999). ade of activity in cultured neurons, albeit after many synapses have already formed, causes an increase (not decrease) in quan- tal size 13,14 likely because of an increased complement of synaptic AMPA-Rs 14 . These observations indicate that the role of activity in the control of AMPA-R expression at synapses is certainly going to be more complicated than the results of Zhu et al. would suggest. Fourth, even the solid observation that a large propor- tion of excitatory synapses early in devel- opment contain NMDA-Rs but not AMPA-Rs 2–4 is subject to the alternative interpretation that these synapses once expressed AMPA-Rs that were subse- quently removed, as has been suggested to occur during LTD 15 . The results of Zhu et al. nevertheless raise many new questions about the role of glutamate receptor trafficking in the formation of excitatory neural circuitry and the activity-dependent regulation of synaptic strength. The idea that over the course of development distinct AMPA-R subunits and signaling mechanisms con- trol the synaptic delivery of AMPA-Rs has implications for our understanding of how activity and experience modify neur- al circuitry throughout the lifespan. The story certainly is not going to be simple but, as this paper indicates, new approaches are available that will facilitate the efforts of investigators for many years to come. ception first reaches the cerebral cortex. The understanding that this region con- tributes to visual analysis by detecting the presence of ‘low-level’ stimulus features (such as the orientation of edges) at spe- cific locations on the retina has co- evolved with our knowledge that these attributes are mapped smoothly across the cortical surface 1 . At further stages of cortical processing, however, the func- tional architecture (and hence the func- tion) is not as clear. For example, neurons In at least one respect, our brains are like tables or chairs— structure and function are intimately related. Take, for example, the so-called primary visual area, where the visual information needed for per- in the temporal lobe that are responsible for ‘higher-order’ analysis of shape are insensitive to low-level features and instead are selectively activated by multi- faceted, ‘real world’ objects such as faces 2 . Although we know that they tend to colo- calize or cluster into functional groups or columns with similar stimulus prefer- ences 2,3 , the fundamental principles of this organization and, consequently, the basic analyses conducted by these neu- rons are not well understood. Erickson and colleagues, in this issue 4 , present the results of an experiment that offers an intriguing possible basis for this organi- zation. They propose that a temporal lobe cortical area important for visual mem- ory, the perirhinal cortex, has a dynamic functional architecture, and that this architecture is shaped by and reflects a monkey’s experience. The perirhinal cortex is a major con- duit by which visual information from the cerebral cortex reaches the hip- pocampus, a brain structure important The author is at the Center for Learning and Memory, RIKEN-MIT Neuroscience Research Center, and Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA. e-mail: ekm@ai.mit.edu Organization through experience Earl K. Miller Erickson and colleagues suggest that nearby neurons in the perirhinal cortex share similar object preferences, and that these groups may develop based on visual experience. © 2000 Nature America Inc. • http://neurosci.nature.com © 2000 Nature America Inc. • http://neurosci.nature.com