, 2011, Kim et al., 2011, Squire and Wixted, 2011, Squire and GABA pathway Zola-Morgan, 1991 and Suzuki, 2009), in spite of recent reports suggesting that perception may also be compromised (Barense et al., 2007, Barense et al., 2010b, Barense et al., 2011b, Bartko et al., 2007, Baxter, 2009, Buckley et al., 2001, Lee et al., 2005a, Lee et al.,

2005b and Lee and Rudebeck, 2010). A recent representational-hierarchical account unites these findings, suggesting that apparently distinct mnemonic and perceptual functions may arise from common representations and computational mechanisms. The representational-hierarchical account proposes that the perirhinal cortex (PRC) can be considered an extension of the representational hierarchy within the ventral visual stream (VVS) (Barense et al., 2005, Bussey and Saksida, 2002, Bussey et al., 2002, Desimone and Ungerleider,

1989, Graham et al., 2010 and Riesenhuber and Poggio, 1999). It is well-established that as information flows from posterior to anterior regions of the VVS, representations of visual stimulus features are organized hierarchically in increasingly complex conjunctions (Figure 1; Desimone and Ungerleider, 1989, Riesenhuber and Poggio, 1999 and Tanaka, 1996). When an object is viewed, multiple representations of this object are activated throughout the entire VVS, with different representations occurring at different stages of the pathway. The object’s low-level features are represented in early posterior regions, whereas conjunctions of features Linifanib (ABT-869) are represented in more anterior regions, Dasatinib with the most complex feature conjunctions—perhaps at the level of the whole object—being represented in regions such as the PRC. The traditional memory systems view argues that MTL structures such as PRC support exclusively mnemonic functions (Clark et al., 2011, Kim et al., 2011, Squire and Wixted, 2011, Squire and Zola-Morgan, 1991 and Suzuki, 2009). In contrast, the representational-hierarchical view proposes that stimulus representations throughout the VVS and MTL are useful for any cognitive

function that requires them (Bussey and Saksida, 2002, Cowell et al., 2006 and Cowell et al., 2010a). This account seeks to explore whether damage to the high-level representations maintained in MTL regions can account for a variety of deficits observed in amnesia. Under this model, one need not postulate separate memory and perceptual systems. One important prediction of this view—yet to be tested in humans—is that if the complex, object-level representations within the PRC are damaged, interference from incidental, irrelevant features can become catastrophic (Cowell et al., 2006 and McTighe et al., 2010). A stream of visual input (such as that encountered over a delay) can create interference at the level of individual features, simply because different objects tend to share lower-level features (e.g.