Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.
A Single-Subject Longitudinal Study Using PointAssist
Transcript of A Single-Subject Longitudinal Study Using PointAssist
Her perseverance was challenged by the nature of the length of the study. The major challenges to the stakeholder were the disabilities spanned from her disorder. Implementing a testing method for real-world interactions was a major technical challenge.
Deploying the test remotely also required a secure and reliable implementation. Other Constraints Team challenges were primarily those pertinent to communication with the participant. How you and your team addressed the challenges? Results of your team’s efforts good and bad Technical challenges were addressed via testing.
Stakeholder challenges were addressed by sharing the burden of adaptation with the user.
Visits to the participant's work place helped communication.
Careful observation of the data helped identify and define measurable outcomes. Prior tests using similar remote testing software yielded positive outcomes.
The participant's experience was improved by the interaction.
Communication efforts implemented were effective but time consuming.
Defined measurable outcomes were appropriate to distinguish the effect that PointAssist had in the participant's performance. Summary Limitations and Future Research Challenges Results Methodology A previous study with PointAssist resulted in significant improvements in click accuracy for participants with motor impairments Motivation To test the validity of PointAssist in real-world interactions
To test the long term effects of PointAssist Purpose of the Project Project Summary HCI Best Practice(s) Based Solution Project Narrative HCI Methods used and how Research Design Development Deployment Product Lifecycle Research was based on an existing assistive technology, namely PointAssist, that has been proven to help individuals with motor impairments with pointing tasks. The testing software was developed to collect data from real-world tasks.
The test ran in the background without interfering with regular computer usage. To test long term effects of PointAssist over time and in real-world contexts we used a single-subject longitudinal experiment. Remote testing was implemented. Testing software was deployed to allow the participant to perform the study in personal computer with preferred settings. Testing sofware relies on the real-time analysis of the sub-momevements.
It does not require maintenance. Participants need no training.
Software does not interfere with other installed software. Author: Affiliation: Keywords: Guarionex Salivia Minnesota State University, Mankato motor impairments, sub-movements, pointing tasks, longitudinal studies
CHI’13, April 27 – May 2, 2013, Paris, France.
Copyright 2012 ACM 978-1-XXXX-XXXX-X/XX/XXPurpose of the Project...$10.00.Purpose of the ProjectTo test the validity of PointAssist in real-world interactionsTo test the long term effects of PointAssist
Purpose of the Project Measurement Time Intervals Common Tasks The effect of PointAssist in Alice's sub-movement characteristics improved her fine motor skills by allowing her to control the cursor better with slower and shorter movements that took less time.
We also showed how her path performance measured as the total distance traveled over the total number of clicks was significantly better with assistance. Defining pointing performance in real-world contexts is challenging. Idle time from the user between tasks and the fact that PointAssist is target agnostic requires creativity to define measurable observations from the data. The Null Hypothesis The null hypothesis in these type of experiments is that the assistance will show no effect difference over the measurement times How to measure performance? Observations based on real-world interactions
Participants where encouraged to work or play to promote clicking "A Single-Subject Longitudinal Study Using PointAssist" Difficulty in finding and retaining participants prompted us to use a single-subject longitudinal study design Participant 37-year-old right-handed female with Cerebral Palsy
We call her Alice
Alice works at company A doing Ebay trading
Uses the computer 40 hours/week Alice's Accuracy Measures from Previous Study with PointAssist Innovative Testing Approach Single-subject longitudinal experiments are not common practice in HCI
Experimental design considers PointAssist as a "treatment" over randomly selected "treatment times" Treatment or Assistance? Instead of treatment times we talk of assistance times, since we do not intend to treat any conditions
Assistance times are blocks of time over which a random assignment of assistance occurs Assignment: ABABAABBBBBAAABAAABB 15 minute blocks
total testing time of 5 hours
four blocks per week
total span of 5 weeks. After 5 minutes After 10 minutes After 15 minutes Dots represent clicks Red paths indicate assistance was provided Motivation Single-subject longitudinal experiments have been shown to provide significance statements about the effect of experimental treatments on a particular individual when he/she is the only subject. About PointAssist PointAssist works by slowing the speed of the cursor depending on the real-time analysis of the sub-movements of a task.
PointAssist has been shown to help a diverse population of individuals with target acquisition. Normalization of complete 15 minute session
We defined "tasks" as paths between clicks
Normalization of tasks is achieved by re-locating clicks at center
This allows us to make better qualitative observations
Distance traveled and the total number of clicks measure amount of activity and path efficiency
Define the ratio of the total distance between clicks and the total number of clicks as path performance ratio
Path performance ratio was found to be significantly smaller We looked at sub-movement characteristics of speed, distance and duration
sub-movements were significantly shorter and slower; sub-movements took significantly less time; more precision Sub-movement Characteristics Path Performance Ratio Discussion We tested the value of longitudinal studies in HCI performing an evaluation of PointAssist in a real-world setting
We contributed by defining new ways in which we can measure performance in real-world interactions
We implemented remote testing effectively helping break geographical boundaries that may prevent effective and feasible data collection We tested the value of longitudinal studies in HCI.
Through single-subject longitudinal experimentation we provided evidence that PointAssist works in real-world environments for some users with Motor Impairments.
Real-wold testing is necessary to validate efficiency and effectiveness of assistive technologies.
Remote testing proved beneficial for recruitment and retention efforts.
We defined and studied a variety of dependent variables that were able to describe the performance of the individual in a real-world setting The system was personalized for the participant and as a result we gained insight on the long-term effects of PointAssist and showed that PointAssist works in real-world environments.
However, the system did not account for changes and variability of individuals with motor impairments over time.
To scale the assistance for other individuals with motor impairments we suggest that frequent adjustment intervention is needed.
Further testing may also be explored with individuals with other range of motor disabilities. Snapshots help us understand the tasks, right? PointAssist Off PointAssist On PointAssist Off PointAssist On