We examined the utility of a PDA-based software system with integrated GPS technology for providing location-aware visual and auditory prompts to enable people with intellectual disability to successfully navigate a downtown bus route. Participants using the system were significantly more successful at completing a bus route than were people in a control group, who used a map and verbal directions. Further, when using the GPS-based system, 73% of participants successfully rang the bell and exited the bus at the right stop compared with only 8% of the control group. This finding was observed for individuals attempting to follow a new bus route for the first time and get off the bus at a previously unknown location.
Community inclusion is a valued goal for most people with intellectual and developmental disabilities and their families. This priority has been stated and codified in many ways, most recently as a part of the United Nations Convention on the Rights and Dignity of Persons With Disabilities, which identified eight general principles: (a) respect for the inherent dignity, autonomy (including the freedom to make one's own decisions), and independence of persons; (b) nondiscrimination; (c) full and effective participation and inclusion in society; (d) respect for difference and acceptance of persons with disabilities as part of human diversity and humanity; (e) equality of opportunity; (f) accessibility; (g) equality between men and women; and (h) respect for the evolving capacities of children with disabilities and for the right of such children to preserve their identities (United Nations, n.d., p. 5).
Verdonschot, de Witte, Reichrath, Buntinx, and Curfs (2009) conducted a systematic review of environmental factors that impact community inclusion and participation of people with intellectual disability. These authors found a number of factors that positively affected community participation, including opportunities to make choices, vocational and social support, family involvement, and assistive technology. Among the variables that negatively affected (e.g., restricted) community participation were the lack of a sense of belonging and the lack of transportation.
The lack of availability of or access to transportation as a barrier to community inclusion is a perpetual problem for many people with intellectual disability. Researchers have identified transportation problems as a obstacle to participation in or access to self-advocacy activities (McNally, 2003), integrated employment (Conley, 2003; Migliore, Grossi, Mank, & Rogan, 2008), religious participation (Minton & Dodder, 2003; Vogel, Polloway, & Smith, 2006), volunteering (Miller, Schleien, & Bedini, 2003), physical activity and exercise (Frey, Buchanan, & Sandt, 2005), integrated leisure activities (Reynolds, 2002), and integrated health care (Hayden, Kim, & DePaepe, 2005; Reichard, Sacco, & Turnbull, 2004). It is evident, then, that transportation limitations affect all aspects of a person's life and constitute a significant barrier to greater community inclusion and, presumably, a better quality of life for many people with intellectual and developmental disability.
Of course, the most frequent solution to transportation difficulties is to own and drive one's own automobile, but for several reasons, including economic pressures and cognitive limitations, most people with intellectual disability do not own or operate a car. Alternatives to driving that have been recommended range from walking (Temple, Anderson, & Walkley, 2000) to building networks of natural supports, such as family members or members of churches and religious communities, who can provide transportation (McNair & Swartz, 1997). If a person receives living or employment supports through a developmental disability service agency, he or she may have access to transportation options provided by that agency, although these options have limitations as well because their availability often is dependent on staff schedules, vehicle or driver accessibility or limitations on staff mileage reimbursement (Huntington, Swanson, & Burmaster, 2003).
Public transit systems (e.g., fixed route public buses) provide the most commonly used or available transportation option and provide, probably, the best option for those living in urban areas for independent, timely, integrated, inexpensive, and relatively unrestricted mobility for people with intellectual and developmental disabilities. However, transit buses also present a unique set of barriers due to route complexity, transfer requirements, unfamiliar destinations, schedule complexity, and other cognitively loaded requirements needed for successful transit system navigation. To access independent public bus travel, individuals must have certain requisite skills, including time management, literacy, problem-solving, attention span requirements, and other cognitive-processing skills. Further, families and support personnel may limit the person's access to public transportation due to fears related to safety.
It is interesting that technology solutions are being explored that reduce barriers to greater independence and community inclusion for people with intellectual and developmental disability (see Wehmeyer, Smith, Palmer, Davies, & Stock, 2004, for a recent overview of the use of technology by people with intellectual disability across several life domains, and Wehmeyer et al., 2006, for an analysis of the efficacy of technology use by people with intellectual disability). The application of technology to the design of transportation supports would seem timely and important, particularly given the rise of mapping and route support devices using Global Positioning System (GPS) that are available for the general public.
In fact, GPS technology offers promise for increasing independence in transportation for people with intellectual disability. Maturing GPS technologies coupled with specialized PDA-based software applications present an opportunity for more independent transportation and, consequently, greater community access and inclusion. A number of such devices have been developed for people with disabilities other than intellectual disability. For example, a device called the BrailleNote GPS uses a receiver similar in size to a cell phone that plugs into a BrailleNote device and provides audio cues to blind users for foot traffic navigation (Axistive, 2007). A similar system called the Trekker is based on GPS and digital map technology to support people who are blind or visually impaired to navigate through their environment (Visuaide, 2009). These systems provide examples of the types of technology supports that might enable people with intellectual disability to overcome barriers to transportation use, but these and similar devices are not designed to be cognitively accessible, and all assume the ability to read or use Braille and require the ability to process complex contextual cues. Our purpose in the current study was to evaluate the efficacy of a GPS-enabled device designed specifically to support independent transportation and bus use by people with intellectual disability.
Study participants were adolescents and adults who received support services through either a public school transition program or several community-based developmental disability agencies. A total of 23 individuals participated (14 females and 9 males). Their average age was 31.91 years (SD = 10.44, range = 18 to 49); average IQ was 54.32 (SD = 7.51, range = 40 to 66). Mean age for the experimental group was 30.55 (SD = 8.23), and their mean IQ was 52 (SD = 8.39). Participants in the control group had a mean age of 33.17 years (SD = 12.36) and a mean IQ of 57.50 (SD = 4.96). Consent was obtained from all participants prior to beginning the study in accordance with the approved IRB procedures for this project. It was made clear to the participants that all study data would be kept confidential and that they could discontinue involvement at any point. Subjects were each paid $10 for participation, which took between 30 and 60 min per person to complete.
Prior to the beginning of each session, a brief interview was held with each participant to obtain self-reported information on previous experience using public transit systems. Sixty percent of the participant pool reported having ridden a city bus previously, although only 43.5% of that subgroup reported having ridden a city bus independently (i.e., without the aid of a familiar person).
The travel support device used in this study was a specially designed, cognitively accessible GPS-based software prototype, called WayFinder (currently available commercially in the United Kingdom; availability in the United States is pending). This device was designed to integrate with a Windows mobile-based hand-held computer to facilitate independent transportation while using a public transit system. Multiple travel routes, or GPS-based instruction sets, can be programmed into the device to provide personalized route-by-route travel instructions and, thereby, support independent transportation for people with intellectual disability. For example, different bus routes for different destinations (e.g., going to work, going home after work, visiting the doctor's office, going to the theatre) could be created in the system and launched via an identifiable picture icon and audio description. For this study, we programmed a single bus route into the device for testing along with a sample training route that was developed to teach participants how to use the device. Although any bus or walking route can be programmed by support staff into the unit by riding or walking to the destination and setting waypoints/recording instructions, the following narrative describes operation during the test route used in the study.
With regard to the operation of the travel support device, the user first selects the desired travel route via the full color, on-screen icon (Figure 1, Screen 1), and the initial prompt screen is displayed (Figure 1, Screen 2), which, in this case, shows a picture of the specific bus to take as well as an audio prompt: “This is our starting bus stop. Watch for the green and yellow shuttle bus and press the Start button when you see it coming.” When the bus approaches, the user then presses the Start button and instructions are provided with a picture and audio message to wait for the bus to stop, wait for others to get off the bus, and then to get on and take a seat (Figure 1, Screen 3). The user then presses the OK button to continue.
At this point, the user is sitting on the bus and the display of the travel support device shows a background watermark indicating bus travel and a trip status indicator on the bottom of the screen (Figure 1, Screen 4). This screen is displayed when there is no other travel message that is being displayed at a particular time. The trip status indicator, which is displayed during the entire trip, consists of a line on the bottom of the display with a person icon that moves across the screen from left to right as progress is made on the route, as can be seen on most of the screen shots in Figure 1. This indicator is used to provide feedback to users regarding their progress so that they can have a general idea of how far into the trip they are and how close they are getting to their final destination.
Landmarks are an optional feature that can be created within the system during route setup that enable users to be alerted to visual landmarks (e.g., businesses, buildings, bridges, parks) along the way in an effort to help the riders learn the travel route. When the bus approaches a designated landmark waypoint, the system automatically displays a picture of the landmark (see Figure 1, Screen 5) and plays a custom audio message such as “Look out the window and you will see the historic Grace Church. Press the OK button when you see it.” The system can also be set to simply display the landmark as an informational prompt and play the associated audio message without requiring any interaction from the user. Then, when the landmark waypoint is passed, the picture goes away and the default screen again appears (Figure 1, Screen 4).
One of the purposes of landmarking, in addition to general trip orientation, is to help keep the user's attention focused on messages provided by the travel support system so as to increase the likelihood that he or she does not become distracted and miss critical messages. Being distracted, or “day-dreaming,” has been informally observed as one of the reasons people with intellectual disability miss their bus stop. For example, Sargent (a person with an intellectual disability) noted that “the biggest challenge with my disability is that I'm easily distracted—I have difficulty staying focused—and as a result I can miss my stop” (Sargent, 2005, p. 1). Landmarking is a strategy that can be used to mitigate this issue.
Although the landmarking capabilities provided by the system are useful, the primary purpose of the device is to use GPS location information and speed detection algorithms to enable users with intellectual disability to know when to get off a bus and, just as importantly, when to stay on the bus when it stops at a different bus stop. There may be dozens of other places a bus stops along any given route to drop off and pick up other passengers as well as at layover points where the bus stops for a period of time simply to maintain a schedule.
The developers of the WayFinder system used GPS speed data to detect when the bus stopped and combined this with GPS location information to identify whether the stop was at a scheduled bus stop location, a layover point, or simply for traffic-related reasons (e.g., red light, stop sign, traffic congestion). Thus, when the bus stops at a known scheduled stop waypoint—but not the designated destination stop for the traveler, the picture shown in Figure 1, Screen 6, is displayed with the verbal message “This is not your stop—stay on the bus.” This message will repeat until the user acknowledges the message by pressing the OK button. Also, the system can detect when the stop may last for several minutes at a layover point and provide additional information to the traveler to stay seated. In this study, there was a layover point where the bus sometimes stopped and waited for up to 5 min. At this stop, the waypoint prompt was created: “This is not your stop—this is a waiting spot for the bus, and we might wait here for a few minutes. Do not get off the bus here.” The route used for the study included a total of five landmarks and four scheduled bus stops.
As the trip nears its destination, a special landmark can be used to tell the traveler that his or her destination stop is coming up next and that it is time to ring the bell (i.e., pull the cord that rings the bell) to tell the bus driver to stop at the next stop. This waypoint is created during route creation in the same manner as other stops and landmarks. Typically, it is only a few blocks prior to the destination bus stop (logically, the prompt for the destination waypoint must be provided after the second to last scheduled stop but before the designated stop), and when the bus reaches this waypoint, a visual prompt of someone pulling the bell cord is displayed (Figure 1, Screen 7; this image is a fairly accurate depiction of the actual pull cord on buses used in the study). In addition, an audio message plays repeatedly that says: “Your bus stop is next—pull the cord now to ring the bell. Press the Next button after you ring the bell.” This message is repeated until the user pulls the bell cord and presses the Next button.
After the participant presses the Next button, additional prompts can be created during route setup with pictures and audio messages that provide further instructions to the traveler according to his or her specific needs. For example, if the person has a habit of forgetting his or her backpack on the bus, soon after the pulling the cord waypoint, another waypoint could be set with a picture of a backpack along with the message “Get your backpack ready and then press the OK button.” For the evaluation study, we set one additional waypoint that provided a prompt showing a picture of a bus seat and stated: “Check around your seat to make sure you have everything before you get off the bus, and then press the OK button” (Figure 1, Screen 8). As the bus comes to a stop at the destination bus stop, the GPS waypoint data as well as the stop detection algorithm confirm that the bus has stopped at the user's destination bus stop, and the final prompt is provided (Figure 1, Screen 9) telling the user “This is your stop. It is time to get off the bus.”
We utilized a complete between-subjects design (Campbell & Stanley, 1963) with two treatment conditions: use of the technology-based system for the experimental group), and use of traditional support methods (i.e., a written bus schedule and printed map that is available to all riders of the Downtown Shuttle) for the control group. The duration of each evaluation session for each participant ranged from approximately 30 to 60 min, depending upon how long the wait was for the next bus at the bus stop. Because buses came every 10 min, the wait was fairly short. The bus trip itself took in approximately 10 to 15 minutes, which constituted the testing period.
Because none of the participants had ever ridden the Downtown Shuttle bus, they were not familiar with the specific route used for testing. Each participant was accompanied to the starting bus stop, which was one block away from the research location, by the experimenter(s). The selected bus route was circular in that the bus was boarded at the starting bus stop and ridden north for several blocks, west for a block, then south for a block, west for a block, and back south for several blocks. This enabled the destination stop to be the Monument stop, which was one block north of the research office. If the participant failed to ring the bell to get off at the target bus stop, the experimenter prompted the participant to ring the bell to get off at the following stop, which was one block south of the research office. This provided the opportunity for participants to successfully get off at the destination bus stop, but if it was missed, he or she was able to exit the bus at the next stop along with the experimenter(s). In either case, the walk back to the research facility was one block. The specific steps of the bus travel session are listed in Table 1. Travel tasks were divided into two types: travel steps and landmarking steps. Travel steps were steps required to get on and off the bus at the correct location. Landmarking steps were activities associated with correct identification of landmarks during the trip.
Data-collection forms were used to record errors and prompts during each experimental session. There were two dependent measures, each with two parts categorized as travel errors, landmarking errors, travel prompts, and landmarking prompts. Table 2 is a summary of the criteria used for scoring each of the dependent measures. Each participant was accompanied on the bus by two researchers, who closely monitored the performance of each person, with verbal prompts and assistance provided as soon as errors were made. In the case of the target bus stop, if the user missed the stop, he or she was prompted to ring the bell for the next bus stop, which was the same walking distance away from the research facility.
Two researchers who rode along during test sessions each scored subjects to allow for interrater reliability measures to be obtained. Interrater reliability of ratings for recorded errors was .92, p = .003, and for recorded prompts, .85, p = .016, both significant correlations. In addition to the quantitative data collected, there was room provided on the data collection forms to record additional observations as well as statements made by subjects during the test sessions. These observations were helpful for identifying areas for more rigorous assessment during future research.
We used SPSS to analyze the data and independent samples t tests to evaluate mean differences in travel errors, landmarking errors, travel prompts, and landmarking prompts between the experimental group using the WayFinder system and the control group. A unidirectional test was employed to analyze mean differences, with a significance level of .05.
Table 3 provides means, SDs, and standard error of the mean for each dependent measure by experimental group. Analysis of the data showed statistically significant differences between experimental and control groups in travel errors, landmarking errors, and landmarking prompts when tested for significance with a unidirectional test for mean differences. The difference between the two groups on travel prompts approached significance, p = .088, but did not achieve the prescribed criteria. Table 4 is a summary of results for each mean difference.
The maximum number of errors that a person could make in completing the experimental task varied, depending upon how many bus stops were made prior to the destination stop. In general, most bus trips provided two to three opportunities for error. The observed mean difference for travel errors was statistically significant, p = .001, with subjects making significantly fewer errors when using the device. This difference was evidenced by the important finding that in the control group (no GPS system), only 1 out of 12 subjects (8%) successfully got off the bus at the correct bus stop, whereas in the experimental group, 8 of 11 subjects (73%) successfully got off the bus at the correct stop.
The maximum number of landmarking errors possible was four per experimental group because there were a total of four buildings that needed to be identified during the bus route. The observed mean difference for landmarking errors was also statistically significant, p = .003, with subjects making significantly fewer errors identifying the target buildings when traveling with the device.
The maximum number of prompts that an individual could receive in the experimental task was three prompts per step. Overall, the average number of prompts individuals received was less in the experimental group, but this difference was not statistically significant; however, a number of the travel prompts in the experimental group related directly to operation of the device. For example, several times subjects had to be reminded to press the OK button to stop a message such as “this is not your stop, please do not get off the bus” because this message would repeat over and over again when at a bus stop until the user pressed OK.
The maximum number of prompts that an individual in the experimental group could receive was three prompts per route step. The average number of prompts individuals received was less in the experimental group, and this difference was statistically significant, p = .03. Thus, device users were able to identify landmarks along the route significantly more independently than were subjects who were simply referring to a printout with pictures of each landmark.
The findings from this study demonstrate that a GPS-enabled portable device with specialized prompting software designed with features that promote cognitive access has promise for supporting independent bus travel for people with intellectual disability. Use of the device provides greater accuracy and increased independence as compared to traditional bus travel methods. In fact, when using the device, 73% of study participants with intellectual disability were able to identify the correct destination bus stop, ring the bell at the proper time, and then exit the bus at the right location as compared with only 8% of the control group participants. This finding is particularly striking given that it was observed for people attempting to follow a bus route for the first time and get off the bus at a previously unknown location combined with challenges of using a new technology device for the first time. In addition, 3 of the 8 participants who successfully rang the bell and exited the bus at the correct bus stop reported that this was the first time they had ever ridden any city bus.
A final note on the findings relates to situations where the GPS-based travel prompting system may or may not be appropriate. The most appropriate applications may be for more experienced bus users when learning a new route; the system does not teach bus-riding skills in and of themselves; therefore, training on safety and social issues would still need to be provided. It may also be inappropriate for the system to be used by itself the very first time a user rides a new route. Further, users should never be instructed or trained to use the system for emergency purposes.
There are a number of limitations that must be considered regarding generalization. First, although participants were randomly assigned to groups that differed according to the transportation method used, the size of each group was relatively small and replications are needed with larger samples. Second, participants were accompanied by research staff on the bus to enable observation and data collection. The presence of researchers certainly had some effect on the behavior of the participants. For example, in a couple of cases it appeared that even though the participants properly understood the prompt that it was time to pull the bus cord to signal the bus driver to stop the bus, they still looked to the researcher for affirmation that it was OK to go ahead and do it. The next phase of study for the device to promote independence in bus travel will require use of the device without the presence of a companion/researcher. Many people with intellectual disability defer to others in authority to prompt them to action or provide affirmation that it is time to act, even when they may know what to do and are fully capable of performing in an independent fashion when they are alone. A third limitation may be the 10-min intervals between buses, which could have negated typical waiting issues or a need for reading bus schedules or other alternative strategies beyond the maps and picture cues provided for use during transit to the control group.
Notably, participants both reported and demonstrated their enjoyment with their success in using the system that enabled them to take the bus. Participant comments referred to ease of use of the prompting system (“It's pretty easy doing this cause it told me to pull the bell”; “You just touch it”; “It said ring the bell and I ringed it”). Even more important was their belief that such a system could help alleviate fears of getting lost (“This is neat. That way you won't get lost” or “That way, I'd never get lost when I ride the bus”).
In summary, we believe that the application of GPS technology to support people with intellectual and developmental disabilities to successfully navigate public transit, fixed-route bus systems provides a promising direction in addressing the significant barrier that transportation difficulties pose for greater community inclusion.
This article is based on work supported by the U.S. Department of Education, National Institute on Disability and Rehabilitation Research, under Project H133S050015. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views or policies of the Department of Education. AbleLink researchers co-authoring this manuscript intend to use these and other research findings to create a commercial product. The opportunity to interview support professionals was critical to the success of this project. The researchers thank the staff at the Transition Program of Colorado Springs School District 11, Carmel Community Living, Goodwill Industries of Colorado Springs, Cheyenne Village, Inc., and MOSAIC, Inc., who provided opinions, feedback, and otherwise facilitated various tasks in the project. Finally, we thank those individuals who volunteered to participate in the WayFinder study and other testing activities conducted as part of this project. Without their help, the project could not have been accomplished.
Editor-in-Charge: Steven J. Taylor
Daniel K. Davies, MA, President (E-mail: firstname.lastname@example.org), Steven E. Stock, MA, Vice President, and Shane Holloway, BS, Software Engineer, AbleLink Technologies, Colorado Springs, CO 80903. Michael L. Wehmeyer, PhD, Professsor/Director, University of Kansas, Department of Special Education, Lawrence, KS 66045.