Eye Tracking: Evaluating the Impact of Gesturing during Televised Weather Forecasts

Robert Drost Geocognition Research Lab, Department of Geological Sciences, Michigan State University, East Lansing, Michigan

Search for other papers by Robert Drost in
Current site
Google Scholar
PubMed
Close
,
Jay Trobec Chief Meteorologist, KELO-TV. Sioux Falls, South Dakota

Search for other papers by Jay Trobec in
Current site
Google Scholar
PubMed
Close
,
Christy Steffke Geocognition Research Lab, Department of Geological Sciences, Michigan State University, East Lansing, Michigan

Search for other papers by Christy Steffke in
Current site
Google Scholar
PubMed
Close
, and
Julie Libarkin Geocognition Research Lab, Department of Geological Sciences, Michigan State University, East Lansing, Michigan

Search for other papers by Julie Libarkin in
Current site
Google Scholar
PubMed
Close
Full access

We are aware of a technical issue preventing figures and tables from showing in some newly published articles in the full-text HTML view.
While we are resolving the problem, please use the online PDF version of these articles to view figures and tables.

Abstract

Televised media is one of the most frequently accessed sources of weather information. The local weathercaster is the link between weather information and the public, and as such weathercaster characteristics, from vocal cadence to physical appearance, can impact viewer understanding. This study considers the role of weathercaster gesturing on viewer attention during weather forecasts. Two variations of a typical weather forecast were viewed by a total of 36 students during an eye tracking session. The first forecast variation contained physical gestures toward forecast text by the newscaster (Gesture condition) while the second variation contained minimal gesturing (No Gesture condition). Following each eye tracking session, students completed a retention survey related to the forecast. These data were used to identify areas of interest to which students attended during viewing and to ascertain how well the forecast was retained across the gesturing treatments. Study results suggest that the weathercaster’s gesturing during forecasts may have induced confusion among participants, but did not affect retention of the weather information investigated in the study. Gesturing diverted attention from other areas of interest within the forecast by encouraging participants to focus on the weathercaster’s hands. This study indicates that minor modifications to weathercaster behavior can produce significant changes in viewer behavior.

CURRENT AFFILIATION: George Mason University, Center for Climate Change Communication, Fairfax, Virginia

CORRESPONDING AUTHOR: Robert Drost, George Mason University, Center for Climate Change Communication, Fairfax, VA 22030, E-mail: drostrob@msu.edu

Abstract

Televised media is one of the most frequently accessed sources of weather information. The local weathercaster is the link between weather information and the public, and as such weathercaster characteristics, from vocal cadence to physical appearance, can impact viewer understanding. This study considers the role of weathercaster gesturing on viewer attention during weather forecasts. Two variations of a typical weather forecast were viewed by a total of 36 students during an eye tracking session. The first forecast variation contained physical gestures toward forecast text by the newscaster (Gesture condition) while the second variation contained minimal gesturing (No Gesture condition). Following each eye tracking session, students completed a retention survey related to the forecast. These data were used to identify areas of interest to which students attended during viewing and to ascertain how well the forecast was retained across the gesturing treatments. Study results suggest that the weathercaster’s gesturing during forecasts may have induced confusion among participants, but did not affect retention of the weather information investigated in the study. Gesturing diverted attention from other areas of interest within the forecast by encouraging participants to focus on the weathercaster’s hands. This study indicates that minor modifications to weathercaster behavior can produce significant changes in viewer behavior.

CURRENT AFFILIATION: George Mason University, Center for Climate Change Communication, Fairfax, Virginia

CORRESPONDING AUTHOR: Robert Drost, George Mason University, Center for Climate Change Communication, Fairfax, VA 22030, E-mail: drostrob@msu.edu

Communication of weather information through daily forecasts or severe weather warnings is a critical mechanism for both public safety and public knowledge. Agencies such as NOAA and the NWS are constantly striving to produce weather products that will benefit the public by providing information on current and future weather hazards that enable the public to make prudent decisions when faced with potential severe weather. Storm-based warnings and the relatively recent impact-based warning trials are evidence of these initiatives. Delivery of weather information varies greatly among the many sources available to the public. Television, however, remains the most frequently utilized source of weather information. How weather information is communicated on television directly impacts the viewing audience’s attention and retention of the information presented. This article presents research using novel approaches to identify how the public interacts with television-based weather forecasts as well as public preferences for receiving weather information. The use of eye tracking technology, coupled with analyses of prior participant knowledge and weather experience surveys, gives insight into weathercast viewer attention on specific areas of televised forecasts, as well as what information participants find most useful when making decisions based on weather information. Determining participants’ attention by measuring where and for how long they are looking at specific weathercast components, and coupling this information with participant responses on communication preferences, provides insight into methods that may improve traditional weather broadcasts. This research provides the groundwork for future investigations of weather communication, including the evaluation of severe weather warning efficacy.

Transference of information presented during weather forecasts is impacted by a multitude of factors. These may include the type of text, visuals, animations, and colors used during the forecast (Fig. 1), and also the arrangement of forecast elements on the televised weathercast screen. In addition, weathercasters themselves may also influence audience attention and retention of forecast information by their presentation style, tone, terminology, and local knowledge. Combined together, these forecast elements have the potential to impact the viewing audience’s attention to specific weathercast components, and ultimately impact viewer retention of forecast information. Research has shown that visual elements tend to reinforce a viewer’s attention to specific information. Gesturing has also been shown to redirect viewer attention toward specific visual elements. Varying amounts of gesturing as part of a weathercaster’s presentation style is thus expected to influence a viewer’s interaction with and understanding of a severe weather forecast.

Fig. 1.
Fig. 1.

Examples of a typical weather forecast and elements contained within. Both forecasts were identical except for the presence or absence of the weathercaster’s gesturing. (a) No Gesture condition forecast variation. (b) Gesture condition forecast variation. Forecasts provided by KELO-TV, Jay Trobec, chief meteorologist.

Citation: Bulletin of the American Meteorological Society 96, 3; 10.1175/BAMS-D-13-00217.1

To ascertain the impacts of weathercast components and gesturing techniques used by weathercasters, eye tracking technology was used to capture the eye movements of study participants as they viewed a weathercast. Two variations of a typical weather forecast, the Gesture and No Gesture conditions, were viewed by different participants during the eye tracking experiment. The weather forecast presented graphical information indicating the daily high and low temperature as well as future weather conditions (potential for rain, cloud cover) for a seven-day period. The weathercasts contained exactly the same information presented in the same way, except that the first forecast condition contained video of the weathercaster physically gesturing toward the forecast text by turning his body and extending his arms and hands toward the information being displayed (Gesture condition). The second variation contained only minimal gesturing by the weathercaster, as indicated by minimal movement of the weathercaster’s hands and no body movement toward the weather forecast (No Gesture condition). Each participant viewed either the Gesture condition or the No Gesture condition weathercast.

The eye tracking data collected in this study included both the location and time of each participant’s apparent focus while they viewed the weathercast. As the human eyes pursue information from a video, image, or other visual stimulus, eye tracking measures the eyes’ location 60 times per second. These data overlaid with still images from the weathercast videos were used as a basis for investigation in this study. Through these data, participant viewing patterns while viewing weathercasts as a function of the presence or absence of gesturing by the weathercaster were identified and compared.

After viewing the weather forecast videos, participants completed a retention survey containing seven questions directly related to the seven-day forecast. The viewer gaze data collected during the eye tracking experiment were used to determine which screen elements attracted the attention of the participants, and provided a basis for establishing areas of interest (AOI) within the forecast. These AOI (Fig. 2) were the focus of the research and were used to establish the difference in participant attention during both the Gesture and No Gesture conditions. Ultimately, the eye tracking data recorded differences in participants’ overt visual attention to weathercast components as impacted by gesturing of the weathercaster.

Fig. 2.
Fig. 2.

(a) No Gesture condition with associated areas of interest (AOI). (b) Gesture condition with associated AOI. All but one of the AOI are identical; the Hands AOI varies based on main gesturing position in the video. Face–1, Hands–2, Station banner–3, Days of the week–4, Scene–5, Forecast–6, Forecast temperature–7.

Citation: Bulletin of the American Meteorological Society 96, 3; 10.1175/BAMS-D-13-00217.1

The data collected during the eye tracking experiment were analyzed quantitatively. Geographic Information System (GIS) technology was used to aggregate and overlay each participants’ gaze data on still images representative of each weathercast condition. AOIs were identical across both conditions, except for the AOI representing the weathercaster’s hands. This AOI was positioned to span the area within the frame where the hands appeared throughout the weathercast. In the Gesturing condition of the weathercast, the AOI representing the weathercaster’s hands spanned a larger area and was representative of the portion of the video frame in which the weathercaster’s hands gestured. Using GIS, participant gaze within each AOI for each condition of the weathercast was summarized. Focused average gaze points, or fixations, representing where participants focused their attention constituted the eye tracking data and were considered for each AOI. Eye tracking data were analyzed via a Mann-Whitney U test in order to determine which condition had greater fixation and fixation duration for each AOI. Gaze plots and heat maps, two forms of eye tracking data summaries, were generated using ArcGIS in order to graphically illustrate participant gaze patterns.

In addition to eye tracking data, both the demographic and retention survey were analyzed using simple descriptive statistics. The retention survey data were analyzed using an independent samples t test in order to determine differences in retention survey scores between the two conditions; the influence of gender on retention was also evaluated.

Descriptive analysis indicates that data for the study population are normally distributed. Gender also appears to have no impact on the retention scores for both conditions. An independent samples t test indicated that no significant difference existed between the retention survey scores for both conditions. Eye tracking data were used to determine the total number of fixations, average fixation duration, and total fixation duration spent on each AOI among participants.

The total fixation duration for the No Gesture condition was greater than the Gesture condition for all AOIs. Average fixation durations per AOI for the No Gesture condition were also greater than the Gesture condition except for a slightly higher duration in the Weathercaster’s Hand AOI for the Gesture condition. The total number of fixations across most AOIs was also greater for the No Gesture condition compared to the Gesture condition except for the Weathercaster’s Hands and Face AOI. Taken together, these results indicate that participants spent more time fixated on the weathercast when gesturing was absent.

For both conditions, Forecast Temperature, Forecast, and Weathercaster’s Face AOI induced the greatest gaze among participants, while the Station Banner, Days of the Week, Scene Image, and Weathercaster’s Hands AOIs induced the least amount of gaze. The results of statistical comparison of AOI gaze data across the two conditions indicate that gaze per AOI was similar across the two conditions except for the Hands AOI, with more gaze on Hands during the Gesture condition, as expected.

Gaze plots and heat maps were used to visually determine which weathercast components were of interest to the participants and to determine what elements drew the most attention from those viewing the forecast videos. In this study, the gaze plot represents one participant’s gaze path while viewing the weathercast, while the heat map represents a summation of all participants for each condition. A comparison of gaze plots for both conditions (Fig. 3) indicates that the participant’s attention was heavily concentrated on both the weathercaster’s face and the forecast regardless of gesturing. The most apparent difference between the two conditions exists along the participant’s gaze path between the weathercaster’s face and the forecast. During the No Gesture condition, the participant’s gaze path displays a single, uninterrupted movement between the face and the forecast. During the Gesture condition, the participant’s gaze moves repeatedly from the weathercaster’s face to the hands and then to the forecast. This pattern of moving back and forth between elements can indicate viewer confusion or lack of focus. Although it may appear that only minor differences exist between the gaze plots (Figs. 3a, 3b), closer scrutiny reveals that a greater amount and larger clusters (represented as circles) are present in Fig. 3b compared to Fig. 3a. Also apparent are the larger clusters centered on the “days of the week” AOI in Fig. 3b.

Fig. 3.
Fig. 3.

Example gaze plots during No Gesture condition and Gesture condition forecast variations. Each depiction illustrates one participant’s gaze during each forecast. (a) A gaze plot for Participant Elab2011020a during the No Gesture condition forecast variation (b) A gaze plot for Participant Elab2011006b during the Gesture condition forecast variation.

Citation: Bulletin of the American Meteorological Society 96, 3; 10.1175/BAMS-D-13-00217.1

The heat maps developed in GIS summarize all participant gaze data during each weathercast condition (Fig. 4). The heat maps utilize a green-to-red scale to indicate the level of attention by the participants. In this case, green indicates a lower amount of attention, while red areas are indicative of higher amounts of attention. The heat maps indicate an area of higher attention on the weathercaster’s Face and the Forecast for both the Gesture and No Gesture conditions. For the Gesture condition, the area between the weathercaster’s Face and Forecast is also garnering more attention compared to the No Gesture condition, as would be expected for a distracted or confused viewer.

Fig. 4.
Fig. 4.

Example heat maps for the No Gesture condition and Gesture condition forecast variations. Each depiction illustrates the study population’s gaze during each forecast. (a) A heat map for study participants during the No Gesture condition forecast variation. (b) A heat map for study participants during the Gesture condition forecast variation.

Citation: Bulletin of the American Meteorological Society 96, 3; 10.1175/BAMS-D-13-00217.1

Daily weather forecasts are a frequently accessed source of information that the public counts on for timely and accurate weather information. The information conveyed in a forecast allows individuals to make choices regarding daily and future plans. As a trusted source, the weathercaster plays a critical role in the delivery of weather information, and is a key player in the decision-making process. Variations in screen elements, weathercaster gesturing, and presentation style all influence viewer attention and retention of the forecast information being viewed. For these reasons, it is critical that we understand the impacts of gesturing on viewer attention to screen elements to insure that forecast information is being delivered as effectively as possible.

This study indicates that gesturing impacts weathercast viewer attention during weekend weather forecasts. Interestingly, though retention scores for both conditions were similar, differences in attention paid to the weathercaster’s hands and to various screen elements were evident. Given this information, consideration should be given to the overall design of the on-screen elements contained within the forecast and the weathercaster’s use of gesturing during the forecast. Although gesturing has not been shown in this study to influence viewer retention of the forecast information, it has been shown to possess the ability to redirect viewer attention. The ability to redirect viewer attention may impact the ability to reduce viewer confusion and increase understanding of the forecast.

The novel use of eye tracking equipment and GIS software to analyze viewers’ attention to and retention of daily weather forecasts provides a unique method available to researchers to analyze the effectiveness of weather forecasts. Today’s weathercaster is often considered the science expert of the television station and often utilized as the “go to” person to disseminate scientific information to the public. Weather is a daily concern for the public both for everyday decision making and more critical decisions that must be made in the face of climate change or natural disasters. What may be even more critical are local severe-weather events that have the ability to potentially impact viewer lives and property. The fact that viewers may be distracted or confused by gesturing suggests that additional research into the role of gesture during weathercasts is needed. The goal of research focusing on delivering a “better” forecast, whether daily or severe, relies on communicating accurate information that provides ample opportunity for the viewer to make effective decisions. These decisions represent crucial factors that influence immediate and future plans and decisions among viewers. Additional research utilizing novel methods that investigate the impact of televised forecasts on viewer attention and retention is necessary to further improve the usefulness and effectiveness of weather forecasts. In particular, this study focused on retention of information from a seven-day forecast. The impact of gesturing on retention of less complex information, such as the amount of time a viewer may have to seek shelter, should be investigated. The current study suggests that gestures may induce confusion, rather than encouraging attention. Based on these forecast results, we encourage further investigation of the impact of gesturing on retention of severe-weather warnings.

FOR FURTHER READING

  • Drost, R., 2013: Memory and decision-making: Determining action when the sirens sound. Wea. Climate Soc., 5, 4354.

  • Ehmke, C., and S. Wilson, 2007: Identifying web usability problems from eye-tracking data. Proc. HCI 2007, University of Lancaster, UK.

  • ESRI, 2001: ArcGISTM spatial analyst: Advanced GIS spatial analysis using raster and vector data. ESRI White paper J8747, 13 pp.

  • Harris, 2007: Local television news is the place for weather forecasts for a plurality of Americans. The Harris Poll #118, 28 November 2007. [Available online at www.harrisinteractive.com/harris_poll/index.asp?PID=839.]

  • Just, M. A., and P. A. Carpenter, 1976: Eye fixations and cognitive processes. Cognit. Psychol., 8, 441480, doi:10.1016/0010-0285(76)90015-3.

    • Search Google Scholar
    • Export Citation
  • Liversedge, S. P., and J. M. Findlay, 2000: Saccadic eye movements and cognition. Trends Cogn. Sci., 4, 614, doi:10.1016/S1364-6613(99)01418-7.

    • Search Google Scholar
    • Export Citation
  • Morrow, B. H., J. K. Lazo, and J. L. Demuth, 2008: Communicating weather forecast uncertainty: An exploratory study with broadcast meteorologists. Final Report of Focus Groups, 36th AMS Conf. on Broadcast Meteorology, Denver, CO.

    • Search Google Scholar
    • Export Citation
  • Morss, R. E., J. L. Demuth, and J. K. Lazo, 2008: Communicating uncertainty in weather forecasts: A survey of the U.S. public. Wea. Forecasting, 23, 974991.

    • Search Google Scholar
    • Export Citation
  • NRC, 2003: Communicating Uncertainties in Weather and Climate Information: A Workshop Summary. National Academies Press, 68 pp.

  • Socci, A. D., 2007: Are TV stations undermining the standards and credibility of weather forecasting and meteorology? Bull. Amer. Meteor. Soc., 88, 578580.

    • Search Google Scholar
    • Export Citation
Save
  • Drost, R., 2013: Memory and decision-making: Determining action when the sirens sound. Wea. Climate Soc., 5, 4354.

  • Ehmke, C., and S. Wilson, 2007: Identifying web usability problems from eye-tracking data. Proc. HCI 2007, University of Lancaster, UK.

  • ESRI, 2001: ArcGISTM spatial analyst: Advanced GIS spatial analysis using raster and vector data. ESRI White paper J8747, 13 pp.

  • Harris, 2007: Local television news is the place for weather forecasts for a plurality of Americans. The Harris Poll #118, 28 November 2007. [Available online at www.harrisinteractive.com/harris_poll/index.asp?PID=839.]

  • Just, M. A., and P. A. Carpenter, 1976: Eye fixations and cognitive processes. Cognit. Psychol., 8, 441480, doi:10.1016/0010-0285(76)90015-3.

    • Search Google Scholar
    • Export Citation
  • Liversedge, S. P., and J. M. Findlay, 2000: Saccadic eye movements and cognition. Trends Cogn. Sci., 4, 614, doi:10.1016/S1364-6613(99)01418-7.

    • Search Google Scholar
    • Export Citation
  • Morrow, B. H., J. K. Lazo, and J. L. Demuth, 2008: Communicating weather forecast uncertainty: An exploratory study with broadcast meteorologists. Final Report of Focus Groups, 36th AMS Conf. on Broadcast Meteorology, Denver, CO.

    • Search Google Scholar
    • Export Citation
  • Morss, R. E., J. L. Demuth, and J. K. Lazo, 2008: Communicating uncertainty in weather forecasts: A survey of the U.S. public. Wea. Forecasting, 23, 974991.

    • Search Google Scholar
    • Export Citation
  • NRC, 2003: Communicating Uncertainties in Weather and Climate Information: A Workshop Summary. National Academies Press, 68 pp.

  • Socci, A. D., 2007: Are TV stations undermining the standards and credibility of weather forecasting and meteorology? Bull. Amer. Meteor. Soc., 88, 578580.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Examples of a typical weather forecast and elements contained within. Both forecasts were identical except for the presence or absence of the weathercaster’s gesturing. (a) No Gesture condition forecast variation. (b) Gesture condition forecast variation. Forecasts provided by KELO-TV, Jay Trobec, chief meteorologist.

  • Fig. 2.

    (a) No Gesture condition with associated areas of interest (AOI). (b) Gesture condition with associated AOI. All but one of the AOI are identical; the Hands AOI varies based on main gesturing position in the video. Face–1, Hands–2, Station banner–3, Days of the week–4, Scene–5, Forecast–6, Forecast temperature–7.

  • Fig. 3.

    Example gaze plots during No Gesture condition and Gesture condition forecast variations. Each depiction illustrates one participant’s gaze during each forecast. (a) A gaze plot for Participant Elab2011020a during the No Gesture condition forecast variation (b) A gaze plot for Participant Elab2011006b during the Gesture condition forecast variation.

  • Fig. 4.

    Example heat maps for the No Gesture condition and Gesture condition forecast variations. Each depiction illustrates the study population’s gaze during each forecast. (a) A heat map for study participants during the No Gesture condition forecast variation. (b) A heat map for study participants during the Gesture condition forecast variation.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1201 819 427
PDF Downloads 360 115 11