SafetyInsights.org

This earlier paper investigated how counterfactual thinking influences subsequent judgements of accidents and the role of people.

Participants were given scenarios and were asked to judge the causality of an event and then develop some remedies to prevent a repeat of the event. The participants were asked to use if-only counterfactuals to determine the remedies.

This was a difficult paper to explain since it’s divided into a pilot study and two experiments, so I’ve left out of a lot of context and findings; all laboratory findings and not workplace-based.

Providing background:

·      Earlier research has highlighted that when managers choose remedies (fixes, improvements etc.) after an accident, they often focus on factors related to people to the exclusion of the worker’s environment

·      An “institutionalised procedure” of analysing incidents includes identifying causal factors via cognitive simulations of the chain of events. In order to reconstruct plausible scenarios, people use counterfactual thinking

·      Counterfactuals can involve “if only” hypotheticals, typically compared to past-events. E.g. “why did this outcome occur on this occasion and not on past occasions?”

·      Lots of different counterfactuals exist. One type, as mentioned, is ‘if only’ types that often occur after negative events. That is, ‘if only’ one causal factor had been changed then this event would have been averted

·      For instance, counterfactuals are used to describe how a change of circumstances could have prevented the incident, “if only they followed the isolation process, this incident would have been prevented”. Counterfactuals come in many forms, but can include things like could of, would of, should of, may have etc. There’s also upwards and downwards counterfactuals

·      They note that, interestingly, counterfactual thinking often results in conjectures about changing one item in the pathway, rather than combinations of multiple factors as are typically present in incidents

·      Some research has also traced how counterfactuals may result in a focus on humans and human action over system level factors

·      One reason why people prefer focusing on human agency/decision making “is that human actions are more “mentally mutable” than other factors; that is, it is easier to imagine a change in human action (a person having done x instead of y) than to imagine a change in an environmental factor (the replacement of one machine with a different machine)” (p580)

Results

Overall, this study found that if-only conjectures after interpreting accidents shape subsequent decisions about how to prevent recurrences.

These findings demonstrated that when participants use if-only conjectures about accident causality, they are more likely to favour factors related to people rather than the environment. Moreover, if-only conjectures also lead to more human-focussed remedies.

In the pilot study without the “if only” conjectures, participants did not favour people-focussed remedies. Instead, they more often preferred fixing the environment/system, and recognised the limitations of trying to focus on people or behaviour. However, this changed when if-only conjectures were introduced in the experiments.

The authors note “our experiment illustrates a domain where “if only” thinking may impede participants from choosing the best decision option” (p584).

The value derived from counterfactuals therefore, and not surprisingly, depend on the context. In some contexts, the use of counterfactuals serves the goals well, but in other cases it may not (like in the current study).

They also talk about the imbalance in counterfactuals towards if-only judgements of accidents, compared to non-accidents, e.g. when things are going to plan. They note that people have far greater repositories if the few cases when things go wrong but relatively few if-only conjectures about successful work.

They discuss the relevance of things like near misses here, which can be seen to fall towards the things going right end of the spectrum (in the case of not propagating all the way to a full accident): “research on aviation pilots’ responses to near accidents observes the frequent occurrence of counterfactual thoughts about how it “could have been worse” as well as thoughts about how it “could have been better,” yet only the latter, upward comparisons enabled adaptive learning” (p584, emphasis added).

Authors: Morris, M. W., Moore, P. C., & Sim, D. L. (1999). Psychonomic Bulletin & Review, 6, 579-585.

Shout me a coffee

Study link: https://link.springer.com/article/10.3758/BF03212966

Link to the LinkedIn post: https://www.linkedin.com/pulse/choosing-remedies-after-accidents-counterfactual-focus-ben-hutchinson