Modified Navigation Instructions for Spatial Navigation Assistance Systems Lead to Incidental Spatial Learning

Abstract

Spatial cognitive skills deteriorate with the increasing use of automated GPS navigation and a general decrease in the ability to orient in space might have further impact on independence, autonomy, and quality of life. In the present study we investigate whether modified navigation instructions support incidental spatial knowledge acquisition. A virtual driving environment was used to examine the impact of modified navigation instructions on spatial learning while using a GPS navigation assistance system. Participants navigated through a simulated urban and suburban environment, using navigation support to reach their destination. Driving performance as well as spatial learning was thereby assessed. Three navigation instruction conditions were tested: (i) a control group that was provided with classical navigation instructions at decision points, and two other groups that received navigation instructions at decision points including either (ii) additional irrelevant information about landmarks or (iii) additional personally relevant information (i.e., individual preferences regarding food, hobbies, etc.), associated with landmarks. Driving performance revealed no differences between navigation instructions. Significant improvements were observed in both modified navigation instruction conditions on three different measures of spatial learning and memory: subsequent navigation of the initial route without navigation assistance, landmark recognition, and sketch map drawing. Future navigation assistance systems could incorporate modified instructions to promote incidental spatial learning and to foster more general spatial cognitive abilities. Such systems might extend mobility across the lifespan.

Keywords: automation; cognitive decline; navigation assistance; spatial memory.

FIGURE 1

(A) Schematic of the experimental setup using a projector behind the experimenter (E) located behind the participant (P). Participants were seated in a mock-up of a VW Touran with the steering wheel connected to a game controller. To their right, a whiteboard was located to allow questionnaire answers and map drawing. (B) Full rendering of the environment used with suburban areas and the city center as displayed in (C) as a real map. (D) Example of a left-turn instruction at a landmark (gas station).

FIGURE 2

NASA Task Load Index (NASA-TLX) scores for the mental demand subscale as a function of navigation instruction for the first (with instruction, light gray bars) and second (without instruction, black bars) navigation phase. Error bars indicate one standard error.

FIGURE 3

Mean percent correct recognized landmarks as a function of navigation instruction dependent on landmark category (relevant, irrelevant, and novel). The rate of correctly recognized landmarks was defined as the ratio of correctly recognized landmarks and the number of all landmarks in the respective landmark category (relevant, irrelevant, and novel) with a value of one indicating 100% correct responses. Error bars indicate one standard error.

FIGURE 4

Mean percent correct reproduced features in the sketch map task as a function of navigation instruction. Error bars indicate one standard error.