Triple Loop Learning

The triple loop learning model supports learning and reflection within teams and individuals. It highlights three modes of learning, as shown in the graphic below. Understanding these modes helps teams to reflect in different ways: from the day-to-day activities to the wider context in which they take place.

• Loop 1 helps problem-solving and course-correction, asking: ‘Are we doing things right?’.
• Loop 2 questions the assumptions a set of activities are based on, and asks: ‘Are we doing the right things?’ (1).
• Loop 3 questions the questions, and asks: ‘How do we decide what’s right’?

It digs into the underlying assumptions and values that shape the activities’ approach, inviting ‘continual reflection on the learning process, the contexts within which learning occurs, and the assumptions and values motivating the learning’. (2)

A framework for learning

As shown in the diagram below, this framework actually has four loops – three learning loops and one ‘defensive reasoning’ loop. This first loop is not a learning loop as it is a loop of self-justification that blocks learning. The framework shows what is happening in each loop: its primary concern, the questions it poses, and the learning approach it enables.

Guidance for use

There are different versions of the Triple Loop Learning framework. This is a natural part of collective learning, as different people test and iterate concepts over time. Our version particularly draws on the work of systems practice consultant and academic, Joan O’Donnell, which itself draws on the original Double Loop Learning Framework created by Argyris and Schön. (3)

The following rules of thumb can help guide your use of this framework:

1. It’s non-judgmental: Everyone will find themselves in each of these loops at different times. While we might aspire to be strategic and enquiring at all times, sometimes we are in ‘reactive’ and even ‘defensive’ modes. This framework does not say that Single Loop Learning– the reactive mode– is intrinsically bad: course correction is a necessary part of day-to-day learning and improvement. The important thing is not to just stay in this loop and avoid the broader questions and enquiries in Loops 2 and 3. The aim should be to achieve balance between Loops 1, 2 and 3.
2. It’s non-linear: Learning is never linear. Although there is a logic of moving from the micro-level daily details of Loop 1 out to the bigger picture questions of Loops 2 and 3, in reality we may do some Loop 2 reflections within Loop 1 conversations, or ask some Loop 3 questions within a Loop 2 strategy assessment. The purpose of this framework is to increase our consciousness of these different learning modes and so create spaces for them.
3. Its use can be spontaneous or structured: The framework can be used in either a spontaneous or structured way. Sometimes you might bring the tool into a project review meeting to consider learning at different levels, or you might use it to intentionally structure and plan learning conversations, creating explicit spaces for the different loops at different points in your process.
4. It can be used by individuals or teams: As well as the team uses described above, the framework can also be used as a personal reflection tool. You might print out the framework and think about when you might have found yourself in each of the loops. You might ask yourself some of the questions in Loops 2 and 3 to consider your own approaches to, or assumptions about, a situation.

Triple Loop Learning and System Change

Reflection allows us to consciously direct change. It helps us to understand the existing dynamics of change in a situation and our own role in and relationship to them. By doing this, we are more able to intentionally influence them. Triple loop learning encourages learners to reflect at different levels, and so be able to influence change within them.

The graphic below summarises the kind of change that each learning loop can facilitate.

• Loop 1: Reviewing actions to identify errors and improvements can lead to changing actions. 
• Loop 2: Challenging assumptions about current actions and the course set can lead to changing approaches to situations.
• Loop 3: Challenging assumptions about the approach and the context in which the activities exist can change perspectives and opens the possibility of changing wider norms and structures.

Below we provide further detail on using the framework, along with examples of questions to ask, situations in which you might use these approaches, and potential changes that might happen as a result of using these tools.

Learning Loops Example

Loop: Defensive

Mode: Self-justifying: How do I / we protect my / our own interests?

Description: Although this is a loop, it is one that blocks learning. It avoids feedback and what it perceives as criticism. It believes that if it can convince others that its actions are correct then its position / status will be secure. This mode is often one born of insecurity and fear– the belief that accepting feedback and correction will undermine position, status or advantage.

Questions: How do I / we cover up flaws and ensure ‘business as usual’ continues? How can we present a positive picture at all times?

Example: A team in a youth charity seeking to reduce youth violence has not met the projected engagement targets for a funded project. Participant data is adjusted to meet the targets. Unaware of these issues, the funder accepts the project reports, and similarly presents the positive programme achievements to its stakeholders and funding is renewed. Although this may bring benefit in the short-term, this is unlikely to hold, and opportunities for learning about how engagement can be more effective will be inhibited.

Loop: Single

Mode: Reactive: Am I / are we doing it right?

Description: Single loop learning focuses on correcting actions and behaviours without questioning the fundamental assumptions or beliefs that underpin those actions. The emphasis is on problem-solving and achieving desired outcomes more efficiently through improvements in working practices and operations.

Questions: How do we correct errors or improve processes and practices to meet existing goals? What is contributing to achieving our goals and what is hindering them?

Example: The programme team use data to analyse how many young people they have reached, compared to their targets. They look at what engagement methods have been successful and haven’t, then decide to concentrate their resources on the more successful engagement approaches. In doing so, they are able to improve their engagement data. However, this reactive learning, which focuses mainly on the numbers, may miss opportunities to understand why unsuccessful approaches haven’t worked or to develop new approaches.

Possible changes: Higher numbers of young people are engaged.

Loop: Double

Mode: Reflective: Am I / are we doing the right things?

Description: Double-loop learning encourages learners to not only correct actions but also question the underlying assumptions that guide those actions. Rather than correcting course, it doesn’t assume the course previously set is itself correct. This compels teams to look harder at whether their actions offer the best chance of achieving their goals, and opens the possibility of exploring alternative approaches.

Questions: What assumptions are we making about how to achieve our goals? Are they correct? Do we need to try different approaches? What might they be?

Example: Here, the programme team question whether the activities they are offering are the right ones – they had previously assumed young people would want to come, but perhaps the problem is the activity itself. Perhaps young people aren’t interested. They consider their broader violence prevention goals and they engage young people in exploring what activities they might be more interested in attending, which could deliver the same overall outcomes. They may they do this as part of a strategy review process, for example.

Possible changes: Change in strategy and approach leading to possible innovations in youth engagement. Change in mentality and understanding about the drivers of violence affecting young people.

Loop: Triple

Mode: Subjective: How do we know what’s right?

Description: The third loop goes beyond actions to examine the context– the norms, structures, patterns– in which the actions occur. It questions the relationship between the actions taken and the wider context, how these contribute to each other and how patterns of behaviour are created. Loop 3 moves from a subject-object view of the team/ individual as the protagonist acting upon a situation/ group of people, to one acting with it. This engages those groups/ people in the process, shifting power relationships and democratising learning. In Loop 3 we are zooming out from our normal mental frameworks to question the situation and our agency, role and motives within it. This is deepening our reflective capacity.

Questions: Are our goals the right ones? Why have we set those goals? Are there other views on this problem/ situation or other goals that could be set? What are the wider norms and structures that are creating this problem? Are there ways in which we are contributing to it ourselves (perhaps unwittingly)? What role do we play in this system / situation? Are there different roles we should be playing? What do others think?

Example: In this loop, the programme team decide to establish ongoing reflection sessions. In these sessions, members are invited to reflect on what their experiences are telling them about youth violence in the local area and why it isn’t improving. This may lead them to question the basis for the actions and the approach they take. They bring in external perspectives, and again seek the perspectives of the young people themselves, to help develop a fuller picture of the issue. They consider what role they are playing and how they might use their role in a different way– considering how their activities might contribute more towards breaking cycles of violence.

References

1. Circular organizing and triple loop learning: Romme and Van Witteloostuijn, Journal of Organizational Change Management, 1999.
2. Beyond Agency and Structure: Triple-Loop Learning: Yuthas, K., Dillard J., Rogers R., Journal of Business Ethics, 2004.
3. Argyris, C., & Schon, D. (1978). Organizational learning: A theory of action perspective

Blank Triple Loop Learning templates to download

• NPC Tool 7 Triple Loop Learning Table Blank