CORDIS - Forschungsergebnisse der EU

The Role of the Vertical Dimension in Memory and Navigation

Final Report Summary - VERTICAL DIMENSION (The Role of the Vertical Dimension in Memory and Navigation)

All of us have experienced the uncomfortable feeling of being spatially disoriented, for example, coming out from the subway, or waking up in a dark, unfamiliar room. However, as soon as we recognize a key aspect of the environment (e.g. a road sign or a distinctive building), we are usually able to realize where we are and which direction we are facing. This process, called reorientation, is the first step required for navigation, and it is an essential ability for any mobile animal challenged with finding a goal. There are many kinds of spatial cues that can be used for reorientation – landmarks, celestial cues, magnetic fields, odors, sounds, etc. Most research attention has focused on the role of visual cues, in particular of the geometric shape of the environment and of landmarks, provided on a flat, horizontal surface (for a review see Cheng & Newcombe, 2005). However, we live and move in a three-dimensional world, and surfaces extended in the vertical dimension are common in natural (hills, mountains) and man-made (stairs, ramps) environments. Therefore, it is necessary to examine how navigation and spatial memory operate on vertically-extended surfaces, and to what degree these environments are represented differently from horizontal environments.

Project objectives
This project focuses on the simplest case of vertically-extended surface: a tilted floor. Prior studies have shown that terrain slope can be used as a compass to reorient and to encode the location of a target; e.g. think of parking your car on a hill and remembering that “the car is parked uphill (or downhill)”. It has also been suggested that the slope may be a salient spatial cue and may be relied upon more than other competing cues, probably because of two reasons: (1) walking on a slope involves effort, and the energetic expenditure increases the awareness of the slant; (2) unlike other studied spatial cues (e.g. landmarks), a slope is perceived through vision but it is also felt through kinesthesia (the angles of the joints and the weight of the body). Theoretically, each of these sensory modalities are sufficient for accurate perception of underfoot tilt, and the multimodal nature of slope information renders this cue perceptually unique. Despite these distinctive properties, there is only scant research on how vertically-extended surfaces differ from purely horizontal surfaces with respect to memory and navigation.
We take this issue in exam with two different, but connected, approaches which can be articulated into two main objectives. The first objective is to examine at a basic level how slope is perceived and represented in memory. E.g. when walking on a hill, do we perceive the slope more through vision or kinesthesia? Also, do we use the information regarding the slope more or less than other competing cues for reorienting and representing a target location? To address this objective, we use a simplified and controlled lab environment where we can directly manipulate the variables. The second main objective is to study slope in a complex and realistic environment, and to test if the presence of the slope improves navigation abilities compared to a horizontal (flat) environment. For this purpose, we use a natural, outdoor environment.

Work performed
First objective. In order to address this, we have generated a sloped, Virtual Environment (VE) and presented 2D images of it on a computer monitor equipped with eye-tracking device. In this condition, the participant can experience the sloped environment only using vision. The goal was to investigate the role of visual cues for slope perception. Participants viewed an image of the sloped environment, and had to point to the uphill direction as quickly as possible. The eye-tracking device enabled us to understand which cues of the image the participant was focusing on in order to solve the task, and, thus, to infer the visual-processing strategy used. We completed two experiments: (1) to test if visual cues are sufficient for accurate slope perception, and (2) to assess which visual cues are more attended to (e.g. the global layout of the environment, like floor and walls, or objects landmarks in the environment). The first objective can be considered completed.
Second objective. We have only started piloting the study, but have not been able to collect enough participants to draw inferences or to publish the results. This delay has been caused by the adjustment due to using a different experimental set-up for objective 1 (VE with eye-tracking device) compared to the planned one (real-world environment). Furthermore, the time involved with testing participants in a large-scale, outdoor experiment has been underestimated.

Main results achieved
Our main findings are that people are able to accurately and quickly identify uphill in a sloped environment using only visual information. This suggests that vision plays an important role in slope perception, that it is sufficient for slope processing, and that it might be relied upon just as much as kinesthesia – if not more. Furthermore, data indicated that, when walking on a hill, identifying the uphill direction is more difficult when directly facing the top or the bottom of the hill; instead, it is easier when facing the side of the hill, probably because the horizon appears tilted. Finally, we found longer reaction time and more errors for women compared to men. Previous studies have shown a large female disadvantage in performance during a reorientation task on slope, when both visual and kinesthetic cues could be used (Nardi, Newcombe, & Shipley, 2011). Our finding confirms this, and suggests that the female difficulty is present within the visual modality. However, interestingly, our results also suggest that there are no sex differences in visual processing strategies; i.e. men and women tend to rely on the same visual cues for solving the task.

Expected final results, with impact and use
Our questions are important because most of the research on navigation and spatial memory has been carried only considering landmarks, and the role of a multimodal cue such as terrain topography is uncertain. We showed that slope is considerably salient and could improve navigation ability, which has important implications for urban planning. Think of how useful it would be if direction signs included more indications related with the vertical dimension; for example, a sign that says “for the castle, take the path uphill” might be clearer and more practical than giving a complex set of directions like “turn left, then turn right.”