It would depend on what you are trying to do with your data? 

Numerous groups, check PubMed (http://www.ncbi.nlm.nih.gov/pubmed/?term=visual+scanpath). Methods have improved, e.g. see papers of Filipe Cristino, Ian Gilchrist, Kenneth Holmqvist...

There is a lot of similar research to gaze behaviour in traffic context.

See for example :  

http://www.tandfonline.com/doi/abs/10.1080/0014013031000090116#.VRLSyPyG-So

http://www.ncbi.nlm.nih.gov/pubmed/16802889

http://www.tandfonline.com/doi/abs/10.1080/00140139.2014.990524#.VRLXKfyG-So

http://www.ncbi.nlm.nih.gov/pubmed/18484857

http://onlinelibrary.wiley.com/doi/10.1111/j.1475-1313.2005.00290.x/full

The book 'Eye tracking - a comprehensive guide to methods and measures' of K. Holmqvist might also provide some interesting ideas for data analysis. 

Hi Andreas,

As Folami said, it depends on what is your question. There are so many different aspects you can explore with eye-tracking data.

For data-driven analysis of fixation patterns, you could have a look to the iMap toolbox: http://perso.unifr.ch/roberto.caldara/index.php?page=3

Thank you guys! The references given by you all helped quite a lot - now it's more than I can read... We did a literature review finding over 80 published articles based on the above mentioned things...

Another thing is, who is recently doing such studies? It's always good to know people doing the same things and facing the same issues...

If you attend ECEM in Vienna (http://ecem2015.univie.ac.at/) this year you will meet most of us :-)

 Hi Andreas,

Some recent research has used visual scanning entropy as a measure of visual search in aviation. See these:- 

https://www.researchgate.net/publication/268282126_Individual_reactions_to_stress_predict_performance_during_a_critical_aviation_incident

https://www.researchgate.net/publication/262017842_Flying_Under_Pressure_Effects_of_Anxiety_on_Attention_and_Gaze_Behavior_in_Aviation

I think its a really nice measure but whether its useful to you will depend on your question. I hope this helps,

Good luck with your research,

Greg

Article Individual reactions to stress predict performance during a ...

Article Flying Under Pressure: Effects of Anxiety on Attention and G...

I attended the ECEM last week, but it was not that helpful as I initially thought. One half of all activities was about reading. The other half was very fundamental research oriented with only very few applied studies. It was an interesting insight into this very theoretical field, however, a more practical focused conference would be also of interest. Maybe these contributions can be found at monstrous conferences like AHFE.

I’d like to give you all an update on this question/discussion. I recently published a paper together with my excellent former master student and coauthor Bo Zhang:

Haslbeck, A., & Zhang, B. (2017). I spy with my little eye: Analysis of airline pilots’ gaze patterns in a manual instrument flight scenario. Applied Ergonomics, 63(September 2017), 62–71. doi:10.1016/j.apergo.2017.03.015 

We used a saccade-based gaze pattern analysis developed by our former student (Hang Yin) to analyze pilots’ visual behavior. We were able to show differences between well practiced and poorly practiced pilots as we hypothesized before.

Thank you all for your support, ideas and literature recommendations.  

I'm interested to know more about how you apply HMM as a practical aspect.  I am looking at driver behaviour and how to predict their next fixation that will make them either accelerate or decelerate based on queue or traffic aspect colour.  Any advice or examples?

Hi guys, you may want to talk to Dr. Michael Raschke from Blickshift, he is an expert in visual analysis of eye tracking data.

https://www.blickshift.com/?lang=en

Chapter Visual Analysis of Eye Tracking Data

Hi Andreas,

Our group analyzes car driver gaze behavior in curve driving. 

Just out is a paper where we apply a qualitative approach to produce a taxonomy of fixation classes, which can be applied to annotate video gaze data on a fixation by fixation basis:

http://journal.frontiersin.org/article/10.3389/fpsyg.2017.00620/abstract

A more rigorous approach is to derive quantitative predictions from theoretical models that can be directly tested experimentally:

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0135505

I have also written a paper on the methodological peculiarities of studying real-work gaze behavior, including some problems to applying the AOI approach which might not be immediately obvious:

https://bop.unibe.ch/index.php/JEMR/article/view/2410

Hope these are useful.

Saw your recent paper on Researchgate and your work seems interesting! 

I think there are many parallels in driving/aviation as a task (in terms of the basic brain processes involved).

Best,

0tt0

Wow, I didn’t expect to revive this two years old discussion. But it’s inspiring to discuss this topic.

Thanks for your responses and attached papers.

I did several eye-tracking studies in the last ten years. I performed experiments in real driving cars, in usability labs and finally in flight simulators. In my former university (TUM) we had a lot of expertise on “simple” analysis methods with gaze-based metrics (ISO 15007) such as number of glances, percent time on AOI, etc. And we developed & used our own eye-tracker (today provided by the spin off “Ergoneers”). However, I never was completely satisfied with these simple metrics. These are the analyses funders or companies pay for and everything on top is quite effortful and time consuming.

Years ago, I had a former colleague at TUM who also investigated drivers’ scanning behavior in curve driving. Unfortunately, he gave up this topic without publishing anything. I have seen a very similar research presented at the AHFE 2012 in San Francisco in one of the road driving sessions.

As a non-mathematician I never completely understood the HMM approach and, therefore, I haven’t used this method. It also took me some time to find a useful differentiation between “scanpath” and “gaze pattern”. As my experimental tasks aren’t bottom-up visual scanning tasks, I also didn’t care for scanpath analysis. As scanpath analysis I consider the geometrical analysis of the scan vectors. However, I was interested in the sequence-based analysis of gaze pattern. The works of the Underwood & Chapman Group did some fascinating analyses on drivers’ behavior and this made me believe to find some interesting insights in pilots’ gaze data. However, I assume that the more skilled a psychomotor process is the easier it could be to identify behavioral patterns in corresponding eye-data. Consequently, I think that it would be easier to find recurring gaze patterns in instrument flight compared to road traffic. Nevertheless, the Underwood & Chapman group found recurring gaze patterns in car drivers.   

I think we found a very useful approach to consider the dominant saccades in the abovementioned paper. Of course this is not the only useful approach.  

Hi all

just got reviewers comment on my PhD thesis. He wants a little more info on analysis of radiologists scan path over a chest image when looking for a nodule. What method would i use to analyse the search pattern etc.

Thanks

PS I have 3 months so re-submit and I'm done!

This year I released an R package 'GrpString' for scanpath (or string in general) analysis: https://CRAN.R-project.org/package=GrpString

The package emphasizes quantifying the comparison of groups of scanpaths. It can export patterns, return transition matrices, compute entropy, calculate statistical difference, etc. More functions such as HMM will be added in the near future.