A DCM analysis showed that the HC influenced activity in PHC. When considered alongside the results of the adaptation

analyses, where PHC, RSC and VC responded to the subjective perception of scenes, this indicates that these brain areas play a more active role in the second, BE error, phase of BE. This accords with the PHC and RSC findings of Park et al. (2007), where they specifically focussed on the BE error, and not the initial BE effect. Overall, therefore, our results serve to underscore the two-stage nature of BE whilst also characterising the underlying neuroanatomy associated with each phase. We next consider in more detail Selleckchem Proteasome inhibitor the role of the HC in the BE effect, and how this might provide insights into the nature of hippocampal processing. The HC is known to be involved in spatial navigation, recalling past experiences, and imagining fictitious and future scenes and events (Buckner and Carroll, 2007; Hassabis and Maguire, 2007; Addis and Schacter, 2011; Spreng et al., 2009). Hassabis et al. (2007)

found that patients with selective hippocampal damage Metformin and amnesia were unable to construct and visualise fictitious and future scenes and events in their imagination (see also Klein et al., 2002; Hassabis et al., 2007; Rosenbaum et al., 2009; Andelman et al., 2010; Race et al., 2011). This led to the proposal that the HC supports scene construction, defined as the process of mentally generating and maintaining a complex and spatially coherent scene or event (Hassabis and Maguire, 2007, 2009). It was further argued that key functions such as episodic

memory and spatial navigation may critically depend on scene construction (Hassabis and Maguire, 2007). In line with previous reports, the patients in Mullally et al.’s (2012) study with selective bilateral hippocampal damage and amnesia were also unable to explicitly construct and visualise scenes in the imagination. BE, which depends on the ability to construct coherent representations of pheromone scenes beyond the view, was also attenuated in these patients. This demonstrated the automatic and implicit role of the HC in scene construction. Our fMRI data corroborate and extend the results of Mullally et al. (2012) by now pinpointing that the precise contribution of the HC to BE is the initial, rapid extrapolation of scenes. That the intact PHC and RSC of Mullally et al.’s (2012) patients were unable to compensate for their damaged hippocampi and could not rescue BE, resonates with our finding of the HC being the driving force behind scene construction and BE, and subsequently influencing other areas such as PHC.