Tuomo Kujala

The Attentional Demand of Automobile Driving Revisited: Occlusion Distance as a Function of Task- Relevant Event Density in Realistic Driving Scenarios

Objective: We studied the utility of occlusion distance as a function of task-relevant event density in realistic traffic scenarios with self-controlled speed. Background: The visual occlusion technique is an established method for assessing visual demands of driving. However, occlusion time is not a highly informative measure of environmental task-relevant event density in self-paced driving scenarios because it partials out the effects of changes in driving speed. Method: Self-determined occlusion times and distances of 97 drivers with varying backgrounds were analyzed in driving scenarios simulating real Finnish suburban and highway traffic environments with self-determined vehicle speed. Results: Occlusion distances varied systematically with the expected environmental demands of the manipulated driving scenarios whereas the distributions of occlusion times remained more static across the scenarios. Systematic individual differences in the preferred occlusion distances were observed. More experienced drivers achieved better lane-keeping accuracy than inexperienced drivers with similar occlusion distances; however, driving experience was unexpectedly not a major factor for the preferred occlusion distances. Conclusion: Occlusion distance seems to be an informative measure for assessing task-relevant event density in realistic traffic scenarios with self-controlled speed. Occlusion time measures the visual demand of driving as the task-relevant event rate in time intervals, whereas occlusion distance measures the experienced task-relevant event density in distance intervals. Application: The findings can be utilized in context-aware distraction mitigation systems, human–automated vehicle interaction, road speed prediction and design, as well as in the testing of visual in-vehicle tasks for inappropriate in-vehicle glancing behaviors in any dynamic traffic scenario for which appropriate individual occlusion distances can be defined.

Modeling visual sampling on in-car displays

In this article, we study how drivers interact with in-car interfaces, particularly by focusing on understanding driver in-car glance behavior when multitasking while driving. The work focuses on using an in-car touch screen to find a target item from a large number of unordered visual items spread across multiple screens. We first describe a cognitive model that aims to represent a driver?s visual sampling strategy when interacting with an in-car display. The proposed strategy assumes that drivers are aware of the passage of time during the search task; they try to adjust their glances at the display to a time limit, after which they switch back to the driving task; and they adjust their time limits based on their performance in the current driving environment. For visual search, the model assumes a random starting point, inhibition of return, and a search strategy that always seeks the nearest uninspected item. We validate the model?s predictions with empirical data collected in two driving simulator studies with eye tracking. The results of the empirical study suggest that the visual design of in-car displays can have a significant impact on the probability of distraction. In particular, the results suggest that designers should try to minimize total task durations and the durations of all visual encoding steps required for an in-car task, as well as minimize the distance between visual display elements that are encoded one after the other. The cognitive model helps to explain gaze allocation strategies for performing in-car tasks while driving, and thus helps to quantify the effects of task duration and visual item spacing on safety-critical in-car glance durations. We study drivers? in-car glance behavior when multitasking while driving.We describe a cognitive model that represents driver?s visual sampling strategy.Model?s predictions were studied against data from two driving simulator studies.Task length and visual item spacing are critical factors for in-car glance lengths.The model helps explain drivers? gaze allocation strategies while multitasking.

Browsing the information highway while driving: three in-vehicle touch screen scrolling methods and driver distraction

Distraction effects of three alternative touch screen scrolling methods for searching music tracks on a mobile device were studied in a driving simulation experiment with 24 participants. Page-by-page scrolling methods with Buttons or Swipe that better facilitate resumption of visual search following interruptions were expected to lead to more consistent in-vehicle glance durations and thus on less severe distraction effects than Kinetic scrolling. As predicted, Kinetic scrolling induced decreased visual sampling efficiency and increased visual load compared with Swipe, increased experienced workload compared with both Buttons and Swipe, as well as decreased lane-keeping accuracy compared with baseline. However, Buttons did not significantly excel Kinetic with any metric but on subjective ratings. Based on the results, we do not recommend the use of kinetic scrolling with in-vehicle touch screen displays in the manner used in the experiment. Instead, page-by-page swiping seems to suit significantly better for in-vehicle displays due to its systematic nature and low levels of pointing accuracy required for scrolling the pages.

Designing browsing for in-car music player: effects of touch screen scrolling techniques, items per page and screen orientation on driver distraction

User interface features of a touch based mobile music player and their comparative impact on driver distraction when searching music albums were investigated. In a driving simulator experiment (N=18) three scrolling methods buttons, swipe and kinetic were compared, whereat the number of music tracks presented in a list-style format varied between three, five and seven items per page. Half of the participants used the music player in a portrait mode and half of them in a landscape mode. It was expected that swipe supports less severe distraction effects than kinetic or button due to systematic page-by-page scrolling and low levels of pointing accuracy required for browsing. Three items should enable more efficient visual sampling efficiency per page, but visual demands are increased compared to five or seven since more scrolling is required. Screen orientation should have no distraction effects. Results indicate that swipe led to less severe distraction effects than buttons or kinetic scrolling methods. Seven items per page was found most distracting, whereas few significant differences were found between three and five. As predicted, screen orientation had no significant effects.

Efficiency of visual time-sharing behavior: the effects of menu structure on POI search tasks while driving

In this paper, the effects of two user interface menu structures on a mobile device display, list and grid, are compared in a driving simulation with the measures of visual time-sharing efficiency, visual load, driving performance and secondary task performance. Eighteen participants conducted a set of eight Point-of-Interest (POI) search tasks with the grid- or list-style menus on navigation software during simulated driving. Between-subject analysis revealed that the list-style menu structure supports more efficient and systematic, and thus, safer interaction while driving than the grid-style menu, in terms of time-sharing and total glance time. However, significant effects of the menu structures were not found in secondary task performance, driving performance measured as lane excursions, or in the measures of average duration of, or total number of glances at the display. The results also suggest that the fewer items in a view, the more efficient and safer the interaction in terms of time-sharing. The sensitivity of the time-sharing metrics for revealing tactical level driver distraction in driving simulation can be argued as being at a higher level than the sensitivity of metrics related to lane maintenance, visual load or secondary task performance.

In-Car Ubiquitous Computing: Driver Tutoring Messages Presented on a Head-Up Display

In this paper, we explore the use of ubiquitous computing in cars from a user psychological point of view. Cars have become an integral part of our modern society and are an attractive opportunity for ubiquitous computing. We chose to study the human dimension of in-car ubiquitous computing with a new driver tutoring system which gives guidance messages for a safer and more economical driving style. The system was tested in a driving simulator and both qualitative and quantitative data was collected. We discuss the safety effects of the system and also the differences in results between novice and experienced male and female drivers. We conclude by presenting the key enhancements revealed in the experiment and by discussing the initial results from the perspective of user psychology. Furthermore, we point out future directions for the work that needs to be done for the system to be actually implemented

Visual-manual in-car tasks decomposed: text entry and kinetic scrolling as the main sources of visual distraction

Distraction effects of in-car tasks with a touch screen based navigation system user interface were studied in a driving simulator experiment with eye tracking. The focus was to examine which particular in-car task components visually distract drivers the most. The results indicate that all of the visual-manual in-car tasks led to increased levels of experienced demands and to lower driving speeds. The most significant finding was that text entry and kinetic scrolling of lists were the main sources of visual distraction whereas simple selection tasks with familiar target locations led to least severe distraction effects.

T9+HUD: Physical Keypad and HUD can Improve Driving Performance while Typing and Driving

We introduce T9+HUD, a text entry method designed to decrease visual distraction while driving and typing. T9+HUD combines a physical 3x4 keypad on the steering wheel with a head-up-display (HUD) for projecting output on the windshield. Previous work suggests this may be a visually less demanding way to type while driving than the popular case which requires shifts of visual attention away from the road. We present a prototype design and report first results from a controlled evaluation in a driving simulator. While driving, the T9+HUD text entry rate was equal compared to a dashboard-mounted touchscreen device, but it reduced lane deviations by 70%. Furthermore, there was no significant difference between T9+HUD and baseline driving in lane-keeping performance. T9+HUD decreased glance time off road by 64% in comparison to the touchscreen QWERTY. We conclude that the data are favorable and warrant more research on attention-reducing text input methods for driving.

Critical Analysis on the NHTSA Acceptance Criteria for In-Vehicle Electronic Devices

We tested a commercial in-car navigation system prototype against the NHTSA criteria for acceptance testing of in-vehicle electronic devices, in order to see what types of in-car tasks fail the acceptance test and why. In addition, we studied the visual demands of the driving scenario recommended by NHTSA for task acceptance testing. In the light of the results, NHTSA guidelines and acceptance criteria need to be further developed. In particular visual demands of the driving scenario and for different simulators need to be standardized in order to enable fair testing and comparable test results. We suggest the visual occlusion method for finding a driving scenario that corresponds better with real-life driving in visual demands as well as for standardizing the visual demands of the scenario when applied to different driving simulators. Furthermore, the acceptance criteria need to be re-evaluated. Especially the TEORT limits applicability to a variety of test tasks needs to be validated and exceptions for certain task types considered. The utility of the average glance duration criterion should be reconsidered.

HCI community in Brazil---sweet 16!

across the country and around the world. The main challenge we now face is supporting collaboration among professors to improve teaching and curricula, possibly by curating existing teaching material and sharing it among educators, as well as defining recommended syllabuses for basic and advanced courses. As for research, IHC is where most young and senior researchers meet. Brazilian HCI research has also been published in international venues, allowing the HCI community worldwide to learn about our work. As a tribute to Brazilian scientific contributions in HCI, Clarisse de Souza was recently nominated to the CHI Academy. Several Brazilian HCI researchers are active internationally, participating in many program committees for international HCI events, journal and book editorial boards, and other community initiatives. Unfortunately, language is still a barrier for many researchers; this Portuguese-speaking comtance rate. In 2002 the Brazilian Computer Society created a Special Commission on HCI, and its Executive Council (CEIHC) is responsible for the quality of IHC and for promoting the HCI field in Brazil. CEIHC has assigned members of the community to lead efforts in HCI education, research, and academia-industry relations in Brazil. Regarding HCI education, a survey was conducted in 2009, distributed through HCI and computer science discussion lists. Answers came from 91 professors teaching HCI courses in 63 universities across the country, located in 18 different states (out of 27), covering all five geographic regions of Brazil. In total, 141 HCI courses were described, 57 percent of which were offered only to undergraduate programs, 23 percent only to graduate courses, and 18 percent to both. In 2012, the survey conducted by the SIGCHI EC Education Committee [1] was translated into Portuguese and distributed in Brazil to help us all understand the differBack in 1996, a small group of Brazilian Ph.D. students attended CHI as student volunteers and were encouraged to find out who else was working with HCI in Brazil. In 1997 the first efforts to organize the Brazilian HCI community came about, with significant contributions from Richard Anderson, then SIGCHI’s chair of Local Chapters. In addition to identifying other people working in HCI, Brazilian HCI researchers created a webpage with their news and contact information and set up both a mailing list and a SIGCHI Prospective Local Chapter: BR-CHI. In 2000 these efforts culminated with the chartering of BR-CHI and the organization of the first national research workshop in HCI. Sixteen years later, the community has consolidated and blossomed. The Brazilian Symposium on Human Factors in Computing Systems (IHC) is an annual event. At recent symposiums we counted some 200 participants and about 100 full paper submissions, with a 30 percent accepHCI Community in Brazil— Sweet 16!