Distributed Cognition in an Airline Cockpit and Airbus Flight Control Design

Ok, yes, communication between crew members will be altered under stress. In this case from the transcript the P/F was the most dysfunctional. The other two continued to try to make sense of the situation (the key missing info being what the P/F was doing with the controls).

Instead of “how pilot communicated” I should have said "how information is propagated through a system " as in the paper in the OP.

The information that the P/F is pulling back can be “propagated through the system” in the linked / yoke system because, for one, the other pilot can see what the P/F is doing.

This is what “cognition” means:

the mental action or process of acquiring knowledge and understanding through thought, experience, and the senses.

Hutchins however : “takes as its unit of analysis a culturally constituted functional group rather than an individual mind”.

In other words taking a crew as a unit rather then an individual.

What’s being looked at is how a crew processes information: "This theory is concerned with how information is propagated through a system in the form of representational states of mediating structures. "

Trying to figure out the second part; “representable states and mediating data” is how I came across the paper in the OP.

The paper says this:

A cockpit provides an opportunity to study the
interactions of internal and external representational structure and the
distribution of cognitive activity among the members of the crew

Another term for “Internal representable structure” I think, would be a “mental model” and external could be tools.

Here’s another paper on aviation:

How a Cockpit Remembers Its Speeds


Great thread. But many tangents possible.

While riding aboard many cruise ships and watching the whole “cockpit concept” evolve the last several years, there is much to be learned about what was done right and as important, what is wrong. The errors in design by Naval Architects and electronic engineers and designers are too numerous to even begin listing here. So Deck officers do what they can, in spite of bad design. I give credit to those working aboard cruise ships that have to suffer through the necessities forced upon them, attributed entirely to bad design and cockpit layout.

In general cockpit design for ships has been mostly horrific. Have you seen the layout of most Royal Caribbean ships? Putting the Pilot aft of everyone else in the “cockpit” as if his presence was an afterthought. Placing the helmsman in a hole in the deck (literally) so that he sits down while driving, so his head is NOT in the way of the “navigator” and “co-navigator” sitting in the cockpit behind him.

Or my favorite … throttle controls so small the person handling them has to lower his head down within inches, usually squinting, to see what he has set the throttles at?! “Look at the screen!” to see the RPM’s some people always shout. I would, but frequently the glare from the sun makes it hard to see.

Long story short, how is it that those who design control systems are so disconnected from its actual use??

I rarely go aboard ANY type of ship and the Master/Deck officers do NOT complain about some piece of bridge equipment that is NOT user friendly, poorly laid out, hard to use for the most routine tasks.

Do designers even go aboard a ship and see the end result of their work? From what I can see … the answer is a resounding, NO!

After countless accidents and even more “near miss” incidents, how is it that the shipping industry continues to suffer from poor design of the basic instruments and their layouts? Its easy to see why so many accidents are blamed on “human error”. It was probably an error of design or technical or instrument related problem. But in the end, its easy to blame those handling that piece of junk, eh?

How many times do I have to keep walking on a bridge of a ship and see round dials, all the same size and type, with similar colors, and struggle to decipher which one is showing me the rudder angle, RPM, rate of turn, propellor pitch or direction, wind speed, vessel speed, etc … oh, in the dark of night even better!

It defies all common sense that any man/woman going to sea can’t agree on what a “standard” basic information instrumentation package and display on every ship should be. It’s NOT that hard … is it?

As far as physical bridge design goes, you might enjoy reading the guide linked in this thread:

How modern developments influence the bridge team’s mental process is clearly an under explored subject. It seems rather obvious that they are geared towards ease of use with reduced manning. Consider as an example the much maligned tendency to angle increasingly information dense glass panels towards the conning station. I suspect that the driving forces behind this move fail to properly consider its implications for existing navigation culture and bridge team information exchange, as per the original intent of this thread.

IMO not fir for purpose if they are the ones that should be making the guidelines?
They lost the plot on ECDIS and didnt know there would be computers on the bridge.

That’s for sure. Attributing motivations to the crew after an incident without knowing what information they were looking at or how they were interrupting that information is not very productive.

Don’t want to get into on this thread (any posts going down that path will be treated as off-topic) but for example the way the weather information was processed on the El Faro was a hot mess. The incident reports focused more on specific actions of various crew members at different times but never really examined the overall picture from a weather information-processing perspective.


Interesting that Houston pilots come up in this conversation. toughest jobs around.

Airplanes USED to have a standard layout. WWII era and before instruments were randomly scattered all over, but from say the 50s to recently everything from the cheapest 2-seater to a 747 had the “sacred six” arrangement of primary flight instruments.
Then glass came along and now we’re back to potluck depending on the brand and model.

1 Like

After watching traffic go by with the helmsman’s face light up by what looked like a few TV sets one night, I bought an AIS transponder because I realized that anything not on the TV didn’t exist.

1 Like

That is a significant understatement. I genuinely think those involved designing layouts of most bridges and the placement of equipment are not only uninterested in what we do and how we use everything, but generally don’t care. They do what works and is efficient from a construction standpoint. In other words, the least costly thing to do.


Some of the tugs had ALL of the Non Follow Up steering controllers in parallel, meaning any one of them could move the rudder. We had a barge go aground outbound of the Houston Ship Channel while loaded with 200,000 barrels of 6 oil. They only cause that made sense was that the pilot was standing outboard and was leaning on the console right at the NFU Controller. His leaning caused us to go full starboard!

1 Like

Absolutely correct on the NFU. Have had that happen, thank goodness we weren’t on the 6 pm news. My Norwegian trained mate almost bitch slapped the pilot that day, hipping on the NFU steering lever. Good job Gunnar.

BUT you train in a sim with the same layout as you fly with.
Thats a huge safety step forward that shipping

class, class, class

been on a few bridges so badly laid out they were dangerous.