Navigation is essential for humans in order to adapt to the living environment and get successful mastery of daily life. Memory certainly plays a crucial role on navigation. People spend a lot of time recalling and figuring a route or a shortcut out to reach a place. Of course, well-known paths are remembered easily and require less working memory capacities (Montello, 2009; Nori and Piccardi, 2011). On the contrary, unknown paths require more attention and greater working memory capacities (Montello, 2009). Interestingly, when people recall a pathway from memory a specific spatial memory system is used (e.g., Piccardi et al., 2010). This system is called topographic memory and involves not only visuospatial information, but also vestibular and proprioceptive information relative to the whole-body movements, as well as a continuous change of the person’s point of view that implies an active updating of the mental representation and the person’s position in the environment (e.g., Piccardi et al., 2008, 2013). This system was found to be separated by the standard visuo-spatial memory at neural level in normal subjects (Nemmi et al., 2013), as well as in different neuropsychological disorders, such as brain-damaged patients (Piccardi et al., 2011a) or in patients with Alzheimer’s disease at the early stages (Bianchini et al., 2014).

Topographical memory is widely supported by object location memory, which has a role in maintaining a coherent and meaningful representation of the visual world, as well as in providing a platform from which directional information can be generated (Postma and De Haan, 1996; Gronau et al., 2008; Postma et al., 2008). For example, remembering the position of a landmark implies the capacity to process landmark-identity information (what), landmark-position (where), and the binding of what and where information (Moscovitch et al., 1995). In general, ‘a landmark is a salient environmental cue working as a spatial reference’ (Nico et al., 2008, p. 1898). They refer to any feature of the environment that is recognizable and serve as spatial reference, such as edges or barriers, rivers, squares, lakes, particular buildings, city monuments and so forth. For example, many people use the Coliseum as reference point to get familiarized with Rome and better navigate the city. Therefore, landmarks play a key role on the construction of mental representations of the navigational environment (Raubal and Winter, 2002), given that perceptual representations of the spatial scene in topographical memory encompass landmarks with their features.

In order to be effective for navigation landmarks must be structurally (with a prominent spatial location – Raubal and Winter, 2002), visually (with a particular size, color, and shape) and semantically (depending on cultural, personal and historical influences) relevant (Caduff and Timpf, 2008). In addition, landmarks can also be emotionally loaded (Balaban et al., 2017), enhancing (positively or negatively) the ability to construct a cognitive map. In other words, the emotional salience of the landmark rather than the individual’s emotional state can be relevant as an aid for navigation. According to Gartner (2012) emotions that are linked to specific landmarks may facilitate navigation when the cognitive state is overloaded, as well as enhance the process of cognitive mapping at any moment. In this direction, Kitchin (1996) developed a conceptual model of the environment–behavior interaction. This model includes three basic components: ‘real world’, representing the environment people interact with; ‘working memory’, that encompasses different filters, including the current emotional state filter which works in partnership with the perceptual context; ‘long term memory’, that contains both records of situations within a time framework and an information store. Therefore, in this model emotions also act as filters in the process of building up a cognitive map.

Although emotions were found to influence topographical memory in real environments (e.g., Palmiero et al., 2015, 2016), the literature that specifically addressed the issue of the effectiveness of emotional landmarks in navigation and wayfinding is extremely scarce (Gartner, 2012). To the best of our knowledge, using virtual environments only Balaban et al. (2017) specifically investigated the role of affect-laden landmarks on wayfinding and recognition with respect to neutral landmarks. Results showed that negatively laden landmarks led to better wayfinding and recognition performance than neutral and positively laden landmarks. Furthermore, they found that negative associations are better remembered over time than positive and neutral associations. According to Balaban et al. (2017), the affective valence of the landmarks enhances wayfinding performance.

Thus, in the present study, the aim was to explore the extent to which topographical memory can be influenced by emotional landmarks. Two different frame of references were considered: “allocentric” (world-centered) and “egocentric” (body-centered) frames of references (Burgess, 2006, 2008; Arleo and Rondi-Reig, 2007). Indeed, individuals may locate environmental objects by a) referring to their own position, namely Egocentric frame of reference or by b) referring to the spatial and configurational properties of such objects, namely Allocentric frame of reference (Galati et al., 2000). To move in the environment or to provide spatial direction, individuals combine spatial frames of reference with spatial relations (Ruotolo et al., 2011). For example, “the street closer to me/on my left” (egocentric) or “the flower shop closer to the church/on the right of the church” (allocentric). The existence of these two frames of reference is confirmed by fMRI studies (Hartley et al., 2003; Iaria et al., 2007; Latini-Corazzini et al., 2010; Sulpizio et al., 2013, 2016). Specifically, egocentric navigation involves areas including the parahippocampal place area (PPA), precuneus and cuneus, inferior parietal lobe and retrosplenial cortex (RSC) (see also Byrne et al., 2007). Instead, allocentric navigation is mainly supported by areas containing place cells (hippocampus) and grid cells (entorhinal cortex) (Maguire et al., 1998; Byrne et al., 2007).

In addition, egocentric and allocentric frames of reference yield to different types of spatial knowledge, such as Route and Survey knowledge, respectively. Route knowledge is characterized as knowledge of spatial layout from the perspective of a ground-level observer. Survey knowledge is characterized by an external perspective, such as a bird’s-eye view, which allows for direct access to the global spatial layout (Shelton and Gabrieli, 2002). Evidence for a distinction between these types of spatial knowledge comes from behavioral (Siegel and White, 1975; O’Keefe and Nadel, 1978; Montello, 1998; Tversky, 2000) and fMRI studies (e.g., Mellet et al., 2000; Shelton and Gabrieli, 2002; Boccia et al., 2016). In particular, Shelton and Gabrieli (2002) found that survey encoding recruits areas also recruited by route encoding, but with greater activation in the inferior temporal cortex and in the posterior superior parietal cortex. Furthermore, only route encoding recruited the medial temporal lobe structures, anterior superior parietal cortex and postcentral gyrus.

Therefore, with this in mind, topographical memory, as measured by the Walking Corsi Test (WalCT), was explored using positive, negative or neutral emotional landmarks in terms of both egocentric and allocentric references of frames. Three aspects of topographical memory were assessed: learning a path of eight steps (egocentric frame); delayed recall (five minutes later) of the previously acquired path (egocentric frame); reproduction of the path on the outline of the WalCT (allocentric frame). This is the first study that explored such an emotional landmark-based topographical memory taking into account for the two different systems of spatial frame of references (egocentric vs. allocentric coordinates).

In the present study, the role that emotion plays in shaping our topographical memory was investigated. In general, memory may be enhanced for emotional information (e.g., Cahill and McGaugh, 1998; Hamann et al., 1999; Kensinger and Corkin, 2003). Here, the influence of positive and negative emotional landmarks compared to neutral emotional landmark was investigated in topographical memory as measured using the Walking Corsi Test (WalCT). Two different frames of reference were considered: egocentric coordinates during the learning of an eight-square path and the delayed recall of the same path; and allocentric coordinates during the reproduction of the path on the outline of the WalCT. Indeed, learning and delayed recall of the eight-square path requires the participant to process information about the position of the self relative to the WalCT (egocentric), whereas the reproduction of the learned path on the map requires the participant to process the position of the squares relative to each other in the WalCT (allocentric).

Firstly, results showed that positive and negative emotional landmarks facilitated the learning of the eight-square path as compared to neutral emotional landmarks, confirming, and extending Balaban et al.’s (2017) study, that revealed the key role only of negative emotional landmarks on remembering paths in virtual environments. Contrarily to Balaban et al.’s (2017) study, neutral landmarks were selected with low arousal, whereas both positive and negative emotional landmarks were selected with high arousal. Different studies showed that memory performance is affected by the level of arousal of stimuli, that is high arousal items are better remembered than low arousal items (e.g., Bradley et al., 1992); in addition, arousal was found to enhance memory for high priority information (Mather and Sutherland, 2011). Thus, both positive and negative emotional landmarks with high arousal captured individuals’ attention, producing an improvement of topographical memory, producing a binding of emotional high arousal landmarks with the path. Indeed, when an individual views a highly arousing photo he/she attributes to the photo a distinctiveness that causes the memory enhancement (Ochsner, 2000). At biological level specific stress hormones are released under highly arousing conditions interacting with the amygdala and leading to improvements in memory (McGaugh et al., 2000). In this direction, the amygdala plays a key role in providing attentional advantages to emotional stimuli (Vuilleumier and Driver, 2007), and by consequence emotionally arousing objects attract attention that facilitates binding of their constituent features (Mather, 2007). This explanation was also used by Mather and Nesmith (2008) to account for the enhancement of memory for the location of high arousal pictures. According to these authors, arousal may enhance the binding process of location to arousal stimuli in two ways: increasing the selectivity of attention and increasing the activation level of the features associated with the object (Mather and Nesmith, 2008).

Secondly, no effect of emotional landmarks was found on the delayed recall, whereas positive emotional landmarks enhanced the allocentric topographical memory, facilitating the reproduction of the path on the outline of the WalCT, as compared to both negative and neutral emotional landmarks. In other words, positive emotional landmarks promoted the translation of egocentric information in an allocentric representation more effectively than negative and neutral emotional landmarks, giving rise to a stable survey representation. In this direction, positive high-arousal stimuli (landmarks) are generally recalled more often than negative high arousal stimuli (e.g., Schmidt et al., 2011). Interestingly, positive emotions increase the ability to remember general and heuristic aspects of an experience, enhancing activity within neural regions that support feelings of familiarity (e.g., Levine and Bluck, 2004; Mickley and Kensinger, 2008). According to this interpretation, while building up allocentric representations positive landmarks with high arousal were retrieved from memory more often, enhancing the familiarity of positive landmarks. Iachini et al. (2009) revealed that locations of familiar buildings were mentally represented in terms of allocentric frames of reference, whereas egocentric frames of reference were used when the environment was unfamiliar.

Taken together, these results showed that emotional landmarks (positive or negative) can improve topographical memory, acting as a moderating variable between environment, observer and objects (Caduff and Timpf, 2008). However, a limit of these results has to be found in the experimental set-up used. Indeed, even if WalCT permits a high control on variables and gives general indications about how topographical working memory works (Bianchini et al., 2010), it lacks ecological validity. This means that it should be crucial in the next future to analyze the “real” navigation behavior in ecological environments that include more real emotional landmarks, that are relevant for the individual, such as churches, shops, monuments, cemeteries and so forth. Nevertheless, it is noteworthy to highlight that, learning a path on WalCT is not like learning a path in a real environment it allows to get information about predictive relationship between the WalCT subject’s performance and the subject’s spatial orientation behavior in the real world (e.g., Bianchini et al., 2010; Piccardi et al., 2014).

As a further next step, the issue of the emotional landmarks in navigation should be explored considering not only the general valence of landmarks but also the specific emotions, such as fear, disgust, anger, sadness and so forth for landmarks with negative valence, and relax, love, humor, happiness, and so forth for landmarks with positive valence. In this vein, the arousal issue should also be considered. For example, in light of Gilet and Jallais (2011), critical landmarks could be arranged as follows: on the one hand, positive and negative valence with low arousal (e.g., serenity and sadness); on the other hand, positive and negative valence with high arousal (e.g., happiness and anger). Such conditions would better clarify the unique contribution of valence and arousal to navigation. Finally, the extent to which the issue of navigation and wayfinding can be improved by emotional landmarks should also be investigated in clinical populations with topographical disorientation disorders as well as in patients suffering from mood disorders (i.e., anhedonia in patients suffering from trauma brain injury). In other words, the question is if emotional landmarks can be used as an aid for people that show impairments in navigations.

MP planned, arranged the set-up, collected data, and wrote the paper. LP planned and contributed to wrote the paper.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.