Introduction 

In the last issue of The Human Stream we examined some barriers that get in the way of effective learning (after things go wrong). As The Human Stream is all about connecting theory to practice, we adopted a reasonably ‘zoomed in’ view in that article - focusing on failure in the context of patient safety.

While the intent for this issue was to continue with our practical exploration of how organisations might enhance their capacity to learn from failures, we cannot do so holistically without introducing the concepts of adaptation and adaptive behaviour to the discussion. Therefore, this issue take a slight conceptual detour before coming back to practical implications.

The learning, adaptation nexus

If the rise and falls of human civilisations is of any interest at all, then the work of Jared Diamond should be mandatory reading. In his 2005 best seller ‘Collapse: Why Some Societies Choose to Fail or Survive’, Diamond studies several past and current civilisations. En route, he abstracts a five-point checklist why some societies fall into decline. We need not go into the checklist itself (in brief, it spans human impacts on the environment, climate change, relationships with hostile and friendly societies, political, cultural and economic factors). However, the many examples given in the book serve to drive home a core truth: Societies that adapt endure.

Curiously, even though they had an excellent model to emulate, the Norse settlements persisted with their reliance on cattle farming, European agricultural methods, European living practices and trade dependence with Norway (which became increasingly difficult to sustain due to growing sea ice). This resistance to learn from the Inuit way of life meant that the Norse presence in Greenland vanished within about 200 years.

The Nature of Adaptation

At its most fundamental level, adaptation is about responding to change. In social groups and societies, this process emerges from our collective ability to sense and interpret changes in our environment, understand how our actions influence the context around us, and translate these insights into meaningful action at scale. This capacity for adaptation isn't merely about gathering information—it requires a genuine willingness to enact and sustain change when needed.

The Learning-Adaptation Dynamic

Whilst these principles apply broadly to modern organisations, particularly in healthcare where adaptation and learning are crucial for patient safety, the relationship between learning and adaptation is more nuanced than it might first appear. In the context of the earlier example, it might be tempting to think of adaptation as simply the outcome of learning but this is more the exception than the rule- in fact, the relationship between the two is more complex and bidirectional.

Yet not all adaptation stems from conscious learning. In nature, we see countless examples of adaptive behaviour without any central control or conscious learning process. Bacterial colonies adjust to environmental changes, plants modify their growth patterns in response to light, and animals alter their behaviour based on immediate threats—all without what we would traditionally call "learning."

Understanding the Spectrum

To make sense of this dynamic, we need to distinguish between three interrelated concepts that exist along a spectrum.

At one end, we have adaptive behaviour—immediate responses to new situations that are dynamic, often automatic, and usually short-lived. These responses require no conscious learning yet allow organisms and systems to respond swiftly to immediate challenges.

Further up we have adaptation—a longer-term modification to structures or behaviours that often arise from repeated interactions within an environment, leading to lasting alterations in capabilities or predispositions.

At the other end of the spectrum is true ‘learning’—a conscious process of acquiring knowledge or skills that typically involves intentional practice and reflection. In people, it is this metacognitive element that distinguishes learning from simpler forms of adaptation.

In modern organisations, particularly in healthcare settings, these three processes interact in complex ways. Organisations must be capable of immediate adaptive responses whilst also supporting longer-term adaptation and intentional learning. The challenge lies in creating environments where all three can operate effectively (and synergistically) in the pursuit of safety.

Learning in the context of patient safety

Centrally what the above discussion implies is that we cannot have a complete conversation about learning without also talking about adaptation and adaptive performance. It also follows that to successfully maintain safety in an evolving operational environment, organisations must maintain two somewhat distinct systems for learning.

The first is the more formal analytical system (Fig 1), and relates to the familiar mechanisms we implement to extract generalisable lessons from a small set of noteworthy experiences. Incident investigations, after-action reviews and formal learning cycles after instances of adverse harm are examples of this.

As experienced patient safety practitioners would appreciate, this system of learning is only activated when more significant failures are encountered. While this system of learning from failure allows for more ‘penetrative’ analysis of specific events, it is also more susceptible to the kinds of traps we spoke of in the last issue because of its selectivity (in its bias towards serious harm) and its formal ‘process’ orientation.

In parallel, organisations also need to cultivate a more ‘fluid’ synthesis-oriented learning system, one that can pull together learnings from the many ‘on-the-fly’ (adaptive) adjustments that clinicians make when things aren’t going to plan or when they adapt in small ways in response to previous challenges in service delivery. This example of learning, centred on actions at the adaptive end of the spectrum, can tell us a lot about emerging points of failure and how teams are attempting to manage them.

Structured correctly, this system of learning can operate much more in step with the current state of play and also engage more meaningfully with near-misses and high frequency, low consequence issues before they become major failures.

While this system is less prone to the traps we discussed last week, it lacks the objective weight that is sometimes needed to drive large scale changes.

When we frame learning in this expanded way, it allows us to accommodate a focus on everyday work within a traditional orientation around learning from failure. We can now more easily accept that everyday adaptive work contains useful information on how clinicians perceive, pre-empt and respond adaptively to the many risks present in clinical operations. One system is stronger at learning from looking back at events while the other excels at looking ahead. Consequently both are necessary.

To be fair, there is a lot more to learning from everyday work than the mere avoidance of failure, and we will examine the relevant theory and applications more methodically in the next issue. But for now, it is sufficient to establish some contiguity between serious failures and everyday work so that we can see them as synergistic foci.

Creating the conditions for enhanced learning

With this broadened idea of ‘learning’ in place, we are better equipped to look at some ways in which we could improve organisational capacity for learning after and (around) failure in the context of patient safety.