'Assessing navigation instructions with emotionally salient narrative' (AsPredicted #98563)
Author(s) Sara Lanini-Maggi (Department of Geography, University of Zurich) - sara.maggi@geo.uzh.ch Christopher Hilton (Department of Geography, University of Zurich; TU Berlin) - c.hilton@tu-berlin.de Sara Irina Fabrikant (Department of Geography, University of Zurich) - sara.fabrikant@geo.uzh.ch
Pre-registered on 2022/05/30 - 11:28 AM (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? Research question: How do navigation instructions containing emotionally engaging narratives affect participants' wayfinding accuracy and the use frequency of mobile maps in a gamified navigation context?
Our research goal is to create navigation aids for pedestrians that make local landmarks more memorable compared to standard instructions, by making them emotionally more salient and by linking them together through a narrative. We contend that emotionally salient instructions might help pedestrians remember the route they traveled more easily; consequently, they perform better, and require the help of a mobile map less frequently, in a wayfinding task. To achieve this aim, participants will navigate a route through a virtual city following instructions that either conform to standard non-emotional navigation instructions or are linked via an emotionally salient narrative (learning phase). In a subsequent route reversal wayfinding task, they will be asked to return to the starting point by walking the same route they traversed previously during the learning phase.
We hypothesize that participants will perform the route reversal wayfinding task significantly more accurately (less wrong turns at decision points) and use the mobile map significantly less frequently for routes navigated with emotionally engaging narrative instructions compared to routes navigated with standard instructions.
In an exploratory analysis, we will further investigate whether and how the emotion type (positive, negative or neutral), nested within the emotionally salient instructions, influences spatial learning and navigation engagement.
3) Describe the key dependent variable(s) specifying how they will be measured. Participants will be asked to perform a navigation task through a virtual urban environment presented online. The following empirical data will be collected from participants during the navigation task to test our main hypothesis:
1) Number of navigation errors (defined as the subject turning into an incorrect path at an intersection) during route reversal. After an error is made, the subject will be notified and prompted back to the previous intersection.
2) Number of mobile map uses during route reversal.
3) Self-reported spatial ability using a standardized questionnaire by Münzer and Hölscher (2011).
Additional DVs for exploratory analyses:
4) A landmark recognition task to assess how accurately participants recognize given landmarks encountered during navigation.
5) A route continuation task (RCT) to assess how accurately they recall movement directions at intersections.
6) A route sequence task to examine how well participants learned the order in which landmarks were encountered along the route.
7) Facial expressions recorded with a participants' webcam using the iMotions software (www.imotions.com) from which we will derive participants' emotions, inclusive engagement, during navigation (as a continuous measure) by using an emotion recognition tool Affectiva (https://www.affectiva.com), available with iMotions.
8) Self-reported felt emotions using the Self-Assessment Manikin (SAM) instrument, i.e., valence and arousal evoked by the narrative given at task-relevant landmarks.
4) How many and which conditions will participants be assigned to? We plan a 2 x 3 nested within-subjects experimental design. The first independent variable (IV) is instruction narrative (IV1) with two levels (standard non-emotional instructions vs. instructions containing emotionally engaging narratives). The second independent variable of emotional cue type (IV2) is nested within the emotionally engaging narrative condition of IV1. Specifically, within the emotionally engaging narratives there will be three conditions: positive emotional narrative, negative narrative, and neutral narrative. Each narrative type will be associated with two of the six total landmarks placed along the route.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. Separate multilevel linear regressions will be used with the predictor variable of instruction narrative (factor, sum contrast coding) for the outcome variables of number of errors during route reversal, and number of map uses. Participant ID will be included as a random factor. In addition, the predictor variable of spatial ability (continuous) will be included to control for this effect in the model.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. Exclusion criteria consist of: (1) Failing to comply with the experimental instructions, and/or withdrawal from the experiment; (2) Technical and data quality issues, including failure to record sensory data online.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. We plan to collect a minimum of 70 participants. We focused on wayfinding accuracy for our sample size calculation, since this is the primary DV of interest for our research question. The minimum sample size (n = 55) was determined a priori using a power analysis with a linear regression for a desired power of at least 0.80 (Faul et al., 2007), and one predictor, using GPower 3.1.9.2. We utilized a medium effect size (f2 = 0.15) to be sure to detect any effects that are large enough to be of interest. As we plan to use an online platform (i.e., Prolific) to recruit participants, we decided to increase the minimum number of participants calculated with G*Power by at least 25%, as recommended by Segen et al. (2021).
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) In the event that a significant difference between navigation instruction types on the route reversal navigation task measures is observed, we plan exploratory analyses to further understand the driving factors behind this effect. Specifically, we will investigate:
1) Spatial knowledge: differences in performance on the landmark recognition task, route continuation task, and the route sequence task between groups in order to identify the specific types of spatial knowledge that may have been affected by the instruction type to drive difference in navigation performance.
2) Task engagement: differences in engagement, derived from facial expression analysis, during navigation between groups to understand whether instruction narrative affected engagement during route learning that may have resulted in better route reversal performance.
3) Cue type: to test differences between positive, negative and neutral emotional cues in the narrative condition on route reversal performance in order to understand whether the type of emotional cue is important in any observed effect of narrative vs standard instructions in the main analysis.
