Scanning techniques and track spacing (or "Scanners Stink!")

[RiverAux]

Below is the citation and abstract on a recent article on the effectiveness of scanners (spotters) in air search and rescue operations. 

Gaze behavior of spotters during an air-to-ground search.
Croft JL, Pittman DJ, Scialfa CT.
Human Factors. 2007 Aug;49(4):671-8

OBJECTIVE: This study was designed to develop methods for evaluating the gaze behaviors of spotters during air-to-ground search and to compare field-derived measures with previous lab results. Secondary aims were to assess adherence to a prescribed scan path, evaluate search effectiveness, and determine the predictors of task success. BACKGROUND: Crashed aircraft must be located quickly to minimize loss of life, often requiring visual search from the air. METHOD: Eye movements were measured in 10 volunteer spotters while they searched from the air for ground targets. Visual acuity, contrast levels, and performance on a lab-based search task were also measured. RESULTS: Results were similar to those of previous lab-based studies of air-to-ground search. Task success could be predicted best from a combination of gaze and laboratory variables, and as in previous research, experience was not one of them. CONCLUSIONS: In both lab and field research, performance is poor. Improvements in air search and rescue success will depend upon improvements in training, the refinement of scan tactics, changes to the task methods or environment, or modifications to parameters of the search exercise. APPLICATION: Spotters were unable to reliably search their assigned area, which has implications for the current search training program and in-the-air protocol.

The sample size was small, and they were using folks without a lot of experience as scanners, but the upshot was that few scanners followed the scanning techniques we've all been taught and even when they did, they miss a lot. 

The actual article talks about the need to either come up with a scanning technique that works and that people can stick to or to just not worry about "technique" in the first place. 

To some extent it confirms what we know based on search experience -- that you often won't find a target until you fly over it several times.  So, we'll let the eggheads work on improving scanner techniques, but what can we do in the meantime? 

Seems to me that the most effective way to increase our POD is to decrease track spacing.  Typically, for Air SAR we use 1 mile track spacing.  I think this is mostly a legacy left over from the old days where precise navigation wasn't possible and the fact that the charts are gridded in 1 nm increments, so it is just easier to plan that way.  But, we now have good GPS systems which would make it just as easy to fly a 0.5 mile track space as it would 1 mile. 

So, if we use 0.5 mile track spacing rather than 1 mile our POD (from the CAPF-104 chart) is going to increase by 60-300% (depending on terrain and tree cover), assuming 1000'AGL.  Basically, by flying at 0.5 mile track spacing you're really flying over the search area twice since it gives the Scanner and the Observer a shot at looking at the same piece of ground on the same flight. 

One might argue that we're better off trying to cover more ground rather than intensely searching a smaller area.  If you're in a search where your search area encompasses half the state, that might be true.  However, if you've got a relatively small area, a smaller track length makes more sense.  Using the smaller track spacing at high probability sites (near departure and destination sites, known turns in flight path, sites where you've got some some sort of witness report, etc.). 

Dramatically increasing your chances of seeing the target on the first sortie is much more preferable than hoping the 2nd or 3rd or 4th sortie picks it up since those sorties might not take place for a few days, by which time rescue would be unlikely. 

So, am I the only one thinking this?  Are some places using 0.5 mile track spacing on a regular basis? 

[KyCAP]

Interesting.   I would be interested to see our National Conference allow for poster presentations and become a academic outlet for this type of research.

or have abstracts of this type published in the "Volunteer" magazine...

[Eclipse]

You say it yourself, inexperienced spotters and a small sample group - anecdotally interesting, and perhaps fodder for an evening of technique reinforcement, but not exactly scientific.

[RiverAux]

Did you read the article?  They actually did some pretty cool stuff and it goes way beyond "anecdotal". 

And, lets face the fact that most CAP scanners are pretty inexperienced themselves.  They're either pilots who are doing their 2 quick flights before moving on to glory, or somebody that probably only gets 5 hours or less of flight time a year, a fair amount of which is probably taking pictures or flying to and from the grid or mission base rather than trying to spot an airplane. 

[Eclipse]

Sadly, too much of the above is true. 

[Short Field]

Was the study CAP specific - using CAP trained scanners or "off the street" people?

[Pylon]

Was the study CAP specific - using CAP trained scanners or "off the street" people?

From the original abstract (my emphasis added):

Task success could be predicted best from a combination of gaze and laboratory variables, and as in previous research, experience was not one of them

[Flying Pig]

There is nothing scientific about scanning.  The eyeball is very unrealiable and tires easily.  To think a Scanner or Observer is very effective with disciplined scanning for any extended period is silly.  You try to do the best you can, but to scan for 2 1/2 - 3 hours effectively?  Not gonna happen.

[Frenchie]

Was the study CAP specific - using CAP trained scanners or "off the street" people?

Evidently the study took "off the street people".  As far as the study goes, it appears to be more of an exercise in evaluating the optics of the human eye rather than specifically focused on search and rescue.  The complete study is not available for free as far as I can tell, but it's anyone's guess as to what altitude they were using, track spacing, targets, length of time spent scanning, terrain, etc.  Most likely they used $5 per hour college student slugs on their first flight in a small aircraft who were more concerned with looking out the window at all the interesting things you can see from the air and thinking about where their next frat party is going to be.  I would hardly match them up against CAP scanners who know someone's life may be at stake in a real world situation.  As far as something like this negating decades of research by the military on the subject in real world conditions, I'm not really all that optimistic.  Interestingly enough, CAP can and does manage to find downed aircraft and the POD tables developed by the military and perfected by decades of military search and rescue tend to be at least reasonably accurate from what I've seen.

[Smithsonia]

In advertising we study how people scan advertisements.
http://www.uwf.edu/ddawson/graphics/eyetrac.htm
This may be interesting to the discussion. Lateral, horizontal, and 45 Degree angles are self taught, often used, and natural eye scanning that are exercised by all of us everyday. I try to make my scans do the same. Horizontal for awhile, vertical for awhile, and descending 45 degree scans... I change up to relieve natural repetitive exhaustion.

Since I've never been formally tested and never found anyone from the air -- I can't tell you that this is a perfected method... but it is my preferred method. I use an imaginary frame with the tire, bottom of the wing, and wing strut as my guide. I try to limit my focus to a limit area inside this frame and stare at the imaginary space that I'm hitting with my eye. I try to give each space a beat or two, maybe 3-4 10ths of a second.

I drive my eye to the ground, move to another dimension and re-drive my eye to the target base (or ground) again. It's something we practice in rifle target shooting. It can be very fatiguing... which is the reason for varying the pattern every couple of minutes. It gets to be natural within a half hour or so. I try to practice and retrain my brain as I head to the grid.
I think we often fly too fast. If we slow down to 80knots I feel like I get a better look. This conflicts with some pilots I know who like to do the grid and 90 or 95 knots. But that may be because I fly a Taylorcraft for Saturday Burger run-fun and it can't get any faster than 80. So my brain may be slower.
Anyone else?
With regards;
ED OBRIEN

[Short Field]

Correct scanning technique (as covered in AFIDIL 2130A) at least puts a set of eyes over all the ground.  Recogitiion of a ground target relates to your POD and is a separate issue.

Incorrect scanning techniques leave huge holes that NEVER have a set of eyes over them.  Anything in the holes has zero chance of being seen.  Most untrained scanners will not be using correct scanning techniques. 

[RiverAux]

Was the study CAP specific - using CAP trained scanners or "off the street" people?

The study was done with Canadians.  It doesn't really discuss their background other than that went through some sort of basic scanning training before the test -- they may have been affilitated with the Canadian version of CAP, but it didn't say for certain. 

Correct scanning technique (as covered in AFIDIL 2130A) at least puts a set of eyes over all the ground.  Recogitiion of a ground target relates to your POD and is a separate issue.

Incorrect scanning techniques leave huge holes that NEVER have a set of eyes over them.  Anything in the holes has zero chance of being seen.  Most untrained scanners will not be using correct scanning techniques. 

Evidentally these people received the typical training done in Canada, which from what I can tell from the article, is the same technique we use (we probably stole it from them originally). 

[Eclipse]

The study was done with Canadians. 

The test group is shown below:



And here is a group of CT'ers reactions to the study!