Reconstructing human contributions to accidents: the new view on error and performance

Another earlier (2002) paper from Sid Dekker.

It explores a shift in how the role of people in both (sometimes) failure, but usually success, is seen.

Of course, there’s been a lot more since written on this; but still an interesting read.

Personally, I’m less interested in the new vs old arguments; but the stuff on human performance is solid.

Some points:

·        Some logics of performance emphasised ‘error’ as the cause of most accidents, where systems were basically safe

·        From that logic, progress is made by protecting systems from unreliable people through proceduralisation etc.

·        A contrasting logic sees error as symptomatic of deeper system issues, where safety isn’t inherent but created by people

·        Error then, tends to be more systematically connected to context – the environment, tools, tasks

·        ‘Reconstructing’ the role of people in surprising events allows us a perspective to explain the seemingly ‘befuddling’ actions

·        “In order to understand why people did what they did, it is necessary to go back and triangulate and interpolate, from a wide variety of sources, the kinds of mindsets that they had at the time”

·        Different mechanisms play into hindsight bias. One mechanism is ‘making tangled histories linear by cherry-picking and re-grouping evidence’

·        Another is how investigators can readily trace backwards, since they know the outcome, to the areas where people could have revised their assessment or taken different actions; but this doesn’t explain why they didn’t and is a limitation of counterfactuals

·        Counterfactual reasoning is an important part of human cognition and imagination, useful for many things, but can play into a powerful hindsight bias

·        Counterfactuals “help us impose structure and linearity on tangled prior histories. Counterfactuals can convert a mass of indeterminate actions and events, themselves overlapping and interacting, into a linear series of straightforward bifurcations”…inconsistent with the often complex environment people operate in

·        Counterfactuals often aren’t explanations either; hence, using them also requires an explanation

·        Local rationality is then discussed; people do what makes sense to them at the time, with the knowledge and information that they have

·        It’s argued that to explain failure that there be a move to “convert the search for human failures into a search for human sensemaking”

·        From ‘‘where did people go wrong?’’ to ‘‘why did this assessment or action make sense to them at the time?’’

·        Moving forward, it’s recommended to:

o  lay out the sequence of events in a context-specific language

o  divide the sequence of events into episodes, if necessary

o  find out how the world looked or changed during each episode

o  identify people’s goals, focus of attention and knowledge active at the time

o  step up to a conceptual description

Study link: https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=4421e47042cd20ca49c07d41db18308c34329317

My site with more reviews: https://safety177496371.wordpress.com

LinkedIn post: https://www.linkedin.com/posts/benhutchinson2_another-earlier-2002-paper-from-sid-dekker-activity-7197730018683207681-Ueuv?utm_source=share&utm_medium=member_desktop