Questions about what was actually associated remained unsettled, much because scientists did not yet have the right tools to investigate the neural mechanisms of behavior. Today, more than 50 years later, neuroscience has become a mature discipline, and we know that animals have specialized brain systems for mapping their own location in space, much like Tolman had predicted. The characterization of

map-like neural representations of the external spatial environment began with the discovery of place cells. In 1971, O’Keefe and Dostrovsky described neurons in the rat hippocampus that fire whenever the animal visits certain selleck inhibitor spatial locations but not anywhere else. These neurons were termed “place cells.” Different place cells were shown to fire at different locations (“place fields”). Although there was no apparent topographic arrangement of place cells according to their firing location, the combination Selleck ZD1839 of activity across large ensembles of place cells was unique for every location in the environment, such that as a population, hippocampal cells formed a map-like structure reminiscent of the cognitive map proposed by Tolman in the 1940s (O’Keefe and Nadel, 1978). Already from the earliest days, however, O’Keefe (1976) acknowledged that maps based on place cells would not be sufficient to enable navigation on their own. Navigation has strong metric components that may depend on neural systems measuring distance and direction of the

animal’s movement. O’Keefe and others suggested that the metrics of the spatial map were computed outside the hippocampus (O’Keefe, 1976, Redish, 1999, Redish and Touretzky, 1997, Samsonovich and McNaughton, 1997 and Sharp, 1999), and subsequent studies consequently searched for space-representing of neurons in the entorhinal cortex, from which the hippocampus gets its major cortical inputs. However, evidence for strong spatial signals remained scarce (Barnes et al., 1990, Frank et al., 2000 and Quirk et al., 1992). The search for origins of the place cell signal received new inspiration in 2002, when it was observed that place fields persist in CA1 after disruption of all intrahippocampal input to this

subfield (Brun et al., 2002). This finding raised the possibility that spatial information is transmitted to CA1 through direct connections from the entorhinal cortex, and as a consequence, the search for spatial maps was shifted to this brain region. The first of the new series of studies targeted the dorsal part of the medial entorhinal cortex (MEC), which provides a significant component of the cortical input to the most common recording regions for place cells in the hippocampus. Cells in the dorsal MEC were found to have sharply defined firing fields (Fyhn et al., 2004). These firing fields were similar to the place fields of hippocampal neurons, but the cells invariably had more than one field, and they showed a strikingly regular organization.