Sleep and Navigation - Testing the effect of sleep on the consolidation of landmark- and boundary-based spatial memory representations

Abstract

Studies in both animals and humans have shown that brain activity recorded during route-learning tasks is re-activated during consecutive sleep (Peigneux et al., 2004; Skaggs & McNaughton, 1996; Wilson & McNaughton, 1994). These findings have prompted the development of theories on the role of sleep for memory consolidation. And indeed, there exists a large body of evidence that sleep may both strengthen and transform memory traces (Diekelmann & Born, 2010; Rasch & Born, 2013). Therefore, propitious effects of sleep on the consolidation of spatial memory must be expected. However, evidence for that prediction is mixed (Ferrara et al., 2008; Nguyen et al., 2013; Noack et al., 2017; Orban et al., 2006; Peigneux et al., 2004; Rauchs et al., 2008; Wamsley et al., 2010). The aim of the present study was to investigate the differential effect of sleep on the consolidation of memory representations within two different spatial representation systems: the striatal landmark- and hippocampal boundary-based spatial memory representations system. To this end, we conducted a study using a virtual environment adopted from Doeller et al. (2008) containing either a single proximal landmark or a circular boundary. Distal cues were always available for orientation. Forty participants, both male and female, were randomly assigned to either landmark or boundary group. They were asked to learn and remember object locations within the virtual environment. After an interval of either two nights of normal sleep or one night of total sleep deprivation and a consecutive recovery night, retrieval knowledge was tested. All participants were tested both in the sleep and wake condition. We expected to find a sleep-related strengthening of spatial knowledge with a greater benefit from post-learning sleep within the boundary group in particular. Contrary to our expectations, though, we found no such results. However, our results might be limited because of various factors: we found a strong influence of gender with women’s performance significantly worse compared to men’s. Furthermore, our cue type, that means landmark and boundary, implied to address egocentric and allocentric spatial orientation strategies respectively, failed to do so. Also, our experimental design, even though carefully counterbalanced across all participants, might have influenced performance. We used two different virtual realities that differed significantly concerning difficulty. We also did not control for video game experience that has been shown to have an impact on performance in similar studies (Tucker & Fishbein, 2008; Wamsley et al., 2010).