In-vehicle agent paper

As part of my PhD-studies into user interaction with semi-autonomous systems, we conducted a small survey-based, experimental study comparing participant reactions to different interactions between an in-vehicle agent and a driver. I’ll be presenting the first part of our in-vehicle agent studies at the Workshop on Human Aspects of Ambient Intelligence (HAI) at the Int. Conference on Intelligent Agent Technology in Sydney, early December.

Abstract:

In-vehicle agents can potentially avert dangerous driving situations by adapting to the driver, context and traffic conditions. However, perceptions of system autonomy, the way an agent offers assistance, driving contexts and users’ personality traits can all affect acceptance and trust. This paper reports on a survey-based experiment (N=100) that further investigates how these factors affect attitudes. The 2×2, between-subject, video-based design varied driving context (high, low density traffic) and type of agent (providing information, providing instructions). Both type of agent and traffic context affected attitudes towards the agent, with attitudes being most positive towards the instructive agent in a light traffic context. Participants scoring high on locus of control reported a higher intent to follow-up on the agent’s instructions. Driving-related anxiety and aggression increased perceived urgency of the video scenario.

As soon as the online proceedings are available, I’ll post the link to the full paper.

Update: art recommender article, car agents & robots

My journal article on the effects of transparency on interaction with user-adaptive systems (finally) has been published in User Modeling and User-Adapted Interaction. We used transparent and non-transparent versions of a content-based art recommender to see whether user understanding affects trust and acceptance. Turns out that offering explanations did influence acceptance of systems decisions, but not trust in the system overall.

Cramer H., Evers V., Van Someren, M., Ramlal, S., Rutledge, L., Stash, N., Aroyo, L., Wielinga, B. (2008) ‘The effects of transparency on trust and acceptance in interaction with a content-based art recommender’, User Modeling and User-Adapted Interaction. ‘online first’ pdf

Our more recent research is focusing on concepts such as dealing with autonomy and perceptions of user control, but also on social and affective aspects of interaction, such as empathy.

Creative interfaces at Lowlands’08

Pianos are no match for the remote controlled Giant RoboHand..

..and car doors are meant for play

New survey

I am looking for a lot of respondents for an online survey on car driving and co-drivers. (For those who participated before: this time,  there are no videos involved). The survey is part of a larger study by our lab on assistive technology in cars. We’re mainly interested in people’s experiences with while driving a car (e.g. regarding co-drivers).

You can find the questionnaire here:
http://www.surveymonkey.com/s.aspx?sm=nkcl8hHpuMWjh5GunsFcKA_3d_3d

Filling out the survey will take you about 20 minutes.
The only requirement is having a driver’s license.

Google cars

It’s three little black cars!

Please don’t mind the picture quality, but there they are! Between A’dam Zuid-Oost & Diemen.

Visit to the TU Delft

Friday I visited TU Delft’s industrial design department. First of all, I must say I was really impressed with the building’s hall doubling as a student work space. Big, light, tools everywhere, looked great.

The main reason for my visit were the annual final presentations for a course on ‘Interactive Technology Design’. Students presented a wide variety of prototypes of games, persuasive objects and fitness experiences. Lots of soccer-related designs, which wasn’t all that surprising I was most impressed by a ‘jealous tree’ (can’t remember the project name, unfortunately) . In short, the tree detected when people were interacting with other installations, became jealous and then tried to attract attention from visitors. Conceptually I liked the personality aspects of the tree. Aesthetically, for a first prototype the tree looked clean, but still somewhat mysterious.

The most fun to play with were the “soccer mushrooms”. A strangely addictive game with little mushrooms just begging to be kicked around… The persuasive powers of shower caps and LED’s should never be underestimated, I guess.

I was also very happy to finally see Daniel Saakes’ Skin project. A design tool that supports designers in rapid prototyping and exploring materials, textures and prints for their design prototypes. You choose the objects, pictures, textures, skin projects them on your prototype. Simple idea, very useful results. Daniel has been a well-liked guest speaker for the course I’m teaching in Amsterdam, so it was great to get to play with the tool myself.

And after all that I got to talk to Miguel Bruns Alonso about his work on stress and tangible interaction. Very interesting considering I’m setting up a study into stress-adaptive systems. Miguel’s got the most impressive collection of stress balls ever used in academic research

All in all, interesting afternoon and an inspiring way to start the weekend

New survey up!

We have lots of new stuff going on!

We’ve just put up a survey on assistive agents in cars. The survey is part of both my dissertation work on human-computer interaction and of a larger study on assistive in-vehicle technology.

You can find the questionnaire here:
http://staff.science.uva.nl/~hcramer/CarSurvey

Filling out the survey will take you about 10-20 minutes.

Next week’s events

The Phobot team has been asked to present our bot at the ‘Happy Chaos’ event on technology trends on Saturday the 24th of May. Happy Chaos apparently consists of young Vrij Nederland journalists and students who organise events mixing debates on current events with parties. We’ll demo Phobot and I’ll also be talking a bit about current trends in Human-Computer Interaction, such as affective computing, persuasive technologies and social & affective robots. It’s going to be quite interesting introducing a probably very mixed audience to this type of research.

Next week’s Thursday (the 22th of May) I’ll also be visiting ‘The Web and Beyond’, the two-yearly event organised by CHI Netherlands, the Dutch chapter of the ACM’s special interest group on human-computer interaction. Focus at the CHI NL events appears to have shifted a bit since first time I visited one of them. In ‘05 I presented at a dedicated Doctoral Consortium (NB: at a different CHI.NL conference series though), while there’s a lot less attention for academia at the Web and Beyond now. I don’t think this is necessarily a bad thing though. There are quite some events available ‘for us’ and I actually quite like hearing more about commercial&practical developments. We need to leave our academic ivory towers every once in a while, right? Hope to see you there!

Wearables workshop

Last week I went to a wearable computing workshop by Leah Buechley at Mediamatic. I had a great time!

Leah just finished her PhD and is starting at MIT next year. She builds wearable applications using all sorts of materials, ranging from ‘traditional’ sensors, LEDs and circuit boards to vintage 1930’s threads, conductive paint & rubber and color-changing fabric.

During the workshop we worked with Arduino programmable circuit boards. The Arduino can be easily programmed by connecting to a PC/mac via USB. Leah developed the Arduino Lilypad version for easy development of wearable cloth prototypes by novices (hooray! for people like me!). The LilyPad basically is a flatter, washable (!) version of the Arduino which e.g. features holes for conductive thread to facilitate sewing it into clothing. Even middleschool students have successfully used her platform to build their own wearable devices.

It would be great to have our students play with stuff like this. And I want to join in!

A variety of projects and tutorials using the Arduino can be found on YouTube:
-Tabletop visualizing how much you talk
-Mashing physical movement with google earth
-Quick n’ dirty connection of Nintendo Wii Nunchuck w/Arduino to control ‘robot’ movement
-Arduino robots made during a 3 hour University workshop

CHI’08

I’m back from Florence where I found out that:

• Microsoft can succesfully use EEG to read your mind. Alright, not quite, but they did present interesting stuff on distinguishing unconscious brain activity in viewing images: you can detect when people process faces or other images using just 2 sensors. Microsoft had a huge amount of papers at this year’s CHI, I’m not sure how to gauge that presence, but the work they presented was interesting.
• Creative combinations get attention: you can combine knitting with digital memories, just like you can combine eggs with rfid tags
• Homeless people have more trouble keeping in contact with their family and friends than with finding food – I really liked this session, studies that ‘go out there’ and follow less known user groups are incredibly interesting to hear about.
• Urban folks have more distant contacts than rural people, who have less friends and are physically closer to these fewer friends. I really liked this study, but I wish they would have included socio-economic stats though.
• Bill Buxton is a great speaker, his plenary talk was a joy to listen to.
• Florence icecream is great.
• Best goodie award goes to Google, but I still can’t solve Rubik’s cubes

My own presentation at the SWUI’08 workshop went well. Even though I’m not a ’semantic web person’ – I did find the workshop interesting. It did convince me that there is some use after all for Semantic Web techniques, this in contrast to utopic ‘one big Semantic Web’ visions from the past.

Overall CHI’08 was worth it. I met lots of interesting people, had fun and got inspiration for my research.