Marcial Losada explains his research for our blog readers


Didier Marlier

November 01, 2009

From Disruption to Engagement

We support leaders as they navigate through significant strategic and cultural changes.  We are united by our values of Expertise, Courage and Generosity. Our network operates across the world.

Mass Engagement Process


Cultural Performance Improvement


Innovative Leadership Development


Team & Individual Coaching

Impressed by the quality of some replies of yours regarding the post “Power of Words” and quoting Marcial Losada’s groundbreaking research, I went to him directly and asked to comment and summarize his work for our mutual benefit. Marcial was kind enough to respond and here it is:

For the benefit of the readers who are not familiar with Meta-Learning (ML), here is a very brief introduction. ML has three variables (dimensions) and 3 parameters. The variables are inquiry-advocacy, positivity-negativity (P/N), and other-self (or external-internal orientation). The three parameters are connectivity (the critical, control parameter), negativity bias and resistance to change (system’s viscosity). The model is driven by a set of nonlinear differential equations that have the same mathematical structure as the Lorenz model (the one that generates the famous butterfly-shaped attractor and is used in many branches of science). ML accounts for 92% of variance in a system’s (couple, team, organization) performance. The best linear models only account for 30% of that variance.

A team is most successful when its members are well connected, is able to balance external vs. internal orientation as well as inquiry and advocacy, and keeps a P/N ratio within the Losada Zone (greater than or equal to about 3:1 and not more than about 11:1). We have taken measurements all the way from Finland to the Patagonia (where there is Methanex, one of the world’s largest methanol company) and I have never observed a team which reaches a 6:1 ratio (5.6 is the upper limit so far). In the Losada Zone, a team is able to flourish (and earn a lot of money as a side bonus) and to be “in the flow” (time doesn’t seem to go by and creativity thrives). Out of the zone, a team languishes visiting over and over the same old routines that keep them stuck, without ever getting to know the best of themselves. This is very sad and it is the main reason why I started a consulting business. I wanted to stop the pain (and make money as a bonus).

Marvin and Michael speak about the power of words and specifically about how, when those words are at least in a P/N ratio of 3:1, we find the best of ourselves. The power of language is indeed something that deserves our attention. Wittgenstein saw it as a prison. And he was right, because language can imprison us. There are two types of prison in nonlinear dynamics: fixed-point attractors and limit cycles. We get there when our P/N ratio (and the words that go with it) are outside the Losada Zone. But language can also liberate us (and Wittgenstein also saw this; he was no fool). Meta Learning is a liberating process whereby we dissolve limiting dynamics such as limit cycles and fixed points and evolve complex order dynamics (that I call complexors–complex order). Lorenz’s butterfly is a complexor and my high performing teams all have butterfly-shaped trajectories when I look at their interaction behavior in phase space. When we are in a complexor pattern we accomplish a great feat: we learn who we truly are and what we are capable of doing. Once you taste this fruit, you are able to transform knowledge into wisdom. You are in a kind of paradise from where you will not be expelled, unless you do that to yourself. When my high performing teams reach a complexor they stay there. I have measured them years later and they are still there. Sustainability is one of the greatest benefits of the training provided by the ML methodology.

I must end by saying that the power of words when viewed from the P/N ratio is much more that what we ever imagined. There is a fascinating study by Dr. Grazyna Rajkowska with chronic depressive patients who lose cerebral mass in the prefrontal cortex and hypothalamus. It turns out that the average chronic depressive has a P/N ratio of 0.5 (two negatives for every positive). I discovered a mathematical function that I call “gamma function” which links P/N to gains or losses in a system. The gamma function predicts a 30% loss of cerebral mass in chronic depressives. This prediction was corroborated by Dr. Rajkowska who measures the loss quite precisely using laser interferometry.

I find this astonishing. How can it happen? What is the mechanism by which P/N destroys cells if it is too low? In my view, the explanation goes back to Einstein formula relating energy to mass. To act on the cerebral mass we need energy (or lack of it to feed those cells in the case of depressives). When you look at the P/N pattern over time, you realize it is an oscillatory pattern, a vibrational pattern. Hence an energy pattern; its frequency increases as the P/N ratio does. This being so, we should also be able to predict the opposite effect: a gain when the P/N ratio is above the Losada Line (3:1). Richie Davidson did a famous study published in the Proceedings of the U.S. National Academy of Science on Tibetan monks doing loving-kindness meditation whose P/N ratio is 4:1 as shown by the differential activation of their left and right prefrontal cortex. My gamma function for that ratio predicts a gain of 30%. Richie discovered that gamma synchrony in the monks is increased by about 30%. Gamma waves are the fastest (higher frequency, hence higher energy) and they connect different parts of the brain which makes creativity and intuition available to us. In this case the gain is not so much in cerebral mass as it is on connectivity. I have shown in my paper, The Role of Positivity and Connectivty in Business Teams, that connectivity and the P/N ratio are mathematically equivalent. This might be the explanation at the brain activity level of Barbara Fredrickson’s findings that P/N broadens our thought-action repertoires.

So watch those words! You might be increasing (or decreasing) your cerebral mass. Perhaps we can start our next team meeting with a mission statement: let’s not lose cerebral mass today!


  1. Marvin Faure

    This is fascinating and exciting research, which validates rather dramatically the advice we give our clients. It is encouraging to see we are on the right track. In the interests of going beyond the intellectual argument and increasing our impact with client teams, would anyone like to share practical ideas for experiential workshop activities in this area?

    One approach that occurs to me is to video a team meeting and play it back later, highlighting occurrences of each variable (inquiry/advocacy, P/N, internal/external). This would have the advantage of making participants’ behaviour very explicit to them and thus preventing denial, but the disadvantage of the need to manage the video system.

    Another approach could be simply to call time-out every 5-10 minutes to debrief the meeting process. Each person could have a score-sheet to note the variables as they occur. This would have the advantage of encouraging participants to focus on the process of their meeting as much as on the content and could be an effective approach to learning. It would also be easier for teams to sustain on their own once the consultant has left.

    Has anybody tried these methods, or similar? What seems to work best?

  2. Didier Marlier

    Dear Marvin,
    Let me be the first one to try and share further with you:
    – Indeed we sometimes used the video. Disadvantage, as you say is the filming ( to do well you need a pro behind the camera rather than leave it fix), the editing (sometimes one can do without when keeping a strict timing on your own sheet of paper, tracking what happens) and the time the video takes. It works and personally I am not a big fan (costly, time consuming and cumbersome)
    – We also did try the sheets where we record the Value Building and Value Destroying Behaviours. This is our standard routine and quite helpfull to review/provide feedback.
    – We also use the time-out, specially when we practice “Team Coaching” observing a Board or Executives team. We tend to have them not at regular time but when needed in order to interfere the least possible with their content but only when a solid blocage (Marcial Losada would call it a “fixed point” emerges and the situation is stalled)
    – For the participants to take ownership in the observation and feedback, we have started to apply to our own meetings a somewhat “gamier” method which we found good: We all receive a collection of green and red casino “jetons” and hand them to the person who just displayed (in our view) a Value Building (green jeton) or Value Destroying (Red jeton) behaviour. Feels a bit artificial for 5 minutes but rapidly becomes natural. Works well and without interruption the feedback mechanism is immediate.
    Thanks for your own comment. Hope this is useful. Have a good week Didier

  3. Marcial Losada

    For me, the most interesting and challenging endeavor regarding Meta Learning was to translate all the knowledge gained through ten years of research in two state-of-the art labs in Ann Arbor and Cambridge, into training programs to develop high performing teams. People tend to think that the mathematics is the hard part. The hardest and most beautiful challente for me was to translate the mathematics into programs to develop high performing teams.

    From the start, I realized that the success we had in finding what where the critical variables and parameters to decode complex human interaction processes was due mostly to the fact that we were able to develop a nonlinear model to account for the hundred of thousands of data points we generated at the labs observing hundreds of teams. Human interaction cannot be properly understood with linear models because whenever you have interaction among the components of a system linearity is useless. The reason for this is that linarity assumes independence of the system’s components. Nothing could be further from the truth when we deal with humans interacting, there is no independence whatsoever, we affect each other for better or for worse, not only by waht we say, but by waht we don’t say. This realization was the key to develop our training programs. We concluded they had to be nonlinear also. All the training we had seen in many corporations was linear, and hence the changes introduced were rarely if ever sustainable. As you all know, organizations spend enormous amounts of money in training and development and they deserve a good return in their investment –which basically translates into sustainable change.

    At the start, we realized that each program had to be custom made for each team, because they had to be developed following the interaction patterns we find in the team when we observe them for the first time. How do we go about this? After about an hour of observation of a typical team meeting we can tell what is the P/N ratio, the level of disconnection in the team, how able they are to expand (or contract) their emotional fields and, most importantly what are the gains or the losses in the interaction process. In other words, whether, as a team, they subtract, add or multiply each other. Few teams are able to multiply each other, to create synergy, which is our objective in the training process.

    To accomplish all this, I work with a world expert (Geralda Paulista) on reading nonverbal behavior and modifying this behavior according to the ML model. While Geralda observes the nonverbal behavior, I observe the verbal behavior. Then we compare notes and produce the diagnosis of the team. We translate all the information into compelling graphics, so that the team can observe, for example, their interaction patterns in real time. This type of feedback has an anormous impact on the team and we have seen they start changing just as a result of the diagnostic process. We recently observed this in Brazil with the big electricity companies in Rio and Fortaleza and did observe it previously in the BHP Billiton mining operations in Chile. Both of these are tough customers who work in tough environments, so they provide good evidence of how even the diagnostic process can be compelling.

  4. Jonathan Wilson

    Dear Didier,

    Thank you for your very interesting article and for prompting the valuable responses from Marvin and Marcial.

    I thought you might be interested that my Finnish colleagues Matti and Tapio created a company, Dialogue Life, to assess teams using Marcial’s principles after listening to his speech at the famous Team Academy for entrepreneuring in Jyvaskyla.

    I have the privilege of helping them to promote Dialogue Life in the UK, having seen the use of the tool very powerfully with the Humap team in Finland (

    Marvin asks a question to which we think we have developed a really good answer; a very practical tool that helps teams to develop awareness of how they really work together, without distracting them from their work purpose, which many self-assessment or role preference teams seem to do. I do not mean to criticise those tools, which have helped thousands of teams in the last century. I think that part of the value of Marcial Losada’s principles are that they approach team performance from a radically different direction that is complementary to traditional approaches, and freshly innovative.

    The Dialogue Life process is to leave) a digital video recorder with a 360 lens in the middle of the meeting table. We find that the camera is unobtrusive and does not distract the team in the way that our personal presence might.
    At the end of the meeting we study the recording. There we can see the whole team in a single field of vision. Using specially adapted mobile phones, we ‘score’ the video for P/N, I/A and S/O comments.
    The Dialogue Life software then produces a minute-by-minute timeline graph of the three relevant behaviours. We can then present this back to the team to show them what they actually did and to bring out relationships and connectivity between the actions (e.g. it is interesting how questions in a positive environment lead to more positive comments a few moments later and that negative comments seem to have a faster effect – raising some interesting evolutionary neuro-biology musings).
    It is early days yet, but we have had some very successful results in Dialogue Life’s first year and next year will bring some new very valuable new partnerships, I think. The tool is extraordinarliy powerful at raising teams’ self-awareness, a core platform for future development,

    I agree with Marcial that it can be an exciting challenge to develop engaging, customised development for each team, but a beauty of the tool and his principles is that it inspires so many different possibilities. There are so many ways to be positive (or negative), to ask questions, to advocate views that stimulate more questions. When, before we begin, people ask what the programme will be after the assessment, I explain that until we have done it, we don’t know, just as I would not expect the doctor to prescribe the cure until the diagnosis is complete.

    I also think it is very helpful to explain that this is not just about positivity. It is the ratios and the systemic interactions between the three behaviours that generate the complex emergence and connectivity that integrates the team to each other, other teams and the wider reality and the team’s core purpose.

    If you would like, please visit our new blog at or visit our website

    Thank you again

    Jonathan Wilson
    Dialogue Life UK

  5. Sharon McGann

    Love the research and from a practical perspective the Solutions Focused approach offers many tools to encourage teams to focus on “what is working and how to get more of that”.

  6. Mafalda Almeida Bruno

    Dear Sir,

    My name is Mafalda Almeida Bruno, from Portugal, and I am doing my PhD in Positive Psychology. I am going to analyse 40 session of psychoterapy using your Model (Losada Line), but I need to know if is there any form to operationalise each one of positive/negative responses.

    I send my abstract so that you can understand my work.

    Thank you very much for your attention.

    Mafalda Almeida Bruno

    Positive Communication in Therapy
    Micro-Analysis of Therapist-Client Communication: Using the Lens of Positive Psychology
    Mafalda de Almeida Bruno & Helena Águeda Marujo
    Lisbon University, Portugal
    Contacts: [email protected] & [email protected]

    This research adopts the Positive Psychology approach for the study of the therapeutic process. It is qualitative, descriptive and exploratory in nature. Its principal aim is to use the Micro-Analysis of Communication (McGee and colleagues, 2004), to analyze the type of communication and dialogues between therapists and clients. These data is being analyzed considering two models: Shelly Gable (2006)’s Active Constructive Response (ACR), whose research is based on understanding how the responses to the sharing of positive events enhance the quality and strength of close relationships, and the Meta Learning Model and the Losada Line, of Marcial Losada (1999), in particular, the Positive / Negative ratio. Both models are used to explore the content of the communication during therapy sessions and assess the impact on the efficacy of therapy, the therapeutic alliance and the quality of therapy, based on clients and therapists perceptions, of the use, by the therapist, of the 3:1 ratio and the ACR.
    Participants: Data is being collected on the Military Psychiatric Hospital in Lisbon, with a population that involves adults in psychotherapy processes (N = 4 therapists X two clients each, collecting data from 5 sessions from each client, in a total of 40 sessions under analysis).
    Instruments: After the audio recording of the 40 therapeutic sessions, data is content analyzed and scrutinized using the Micro-Analysis of Communication. An Appreciative Inquiry (AI) (Cooperrider & Whitney, 2005) interview is used to help characterize the quality of the therapeutic process, as perceived subjectively by each of the clients and therapists enrolled in the study.
    Procedure: data is gathered recording the therapeutic sessions of voluntary therapists and respective clients. After five sessions of each therapeutic pair, the therapist and client – first individually, and afterwards conjointly, are interviewed using an AI –based script, created for this study. The study gathers longitudinal assessment of the therapeutic process during the four months following the first collection of data (recording the sessions), in order to assess the temporal duration of the therapeutic process and to measure any differences in the duration of the process. The data collected in these longitudinal procedures will be subject to two types of evaluation: (1) the quality of the communication adopted during the therapeutic process, according to the ACR and 3:1 Ratio; and (2) the subjective perspectives on the current therapeutic process as assessed by the AI interviews with clients and therapists.
    During the 4 months Follow-up, data gathering includes the number of sessions that do occur following the last record, and any eventual ending of therapy.
    Research and theoretical questions under analysis

    1. Will a higher proportion of positive speech acts, when compared with negative ones, influence the perception of satisfaction, efficiency and duration of the therapeutic process, either from the therapist or client?
    2. How is the “Losada Line” applicable to the flourishing of the clinical process and is it useful in assessing the satisfaction, effectiveness and efficacy of the therapeutic process?
    3. Does the intensive use of an Active Constructive Response by the therapist after the sharing of successes by the client facilitate the clinical process, and make it more effective and subjectively satisfactory from the standpoint of the therapist and the client?

    4. Does the focus and attention upon problems and difficulties, when compared with that of successes, during the therapeutic conversation, impact differently the clinical process?

    5. How can the use of the micro-analysis of the therapeutic communication shed light upon the dynamics, sequences and consequences of that communication process, to understand if there is a different impact on the quality of therapy, namely will more positive, constructive, language and dialog, facilitate more the duration, satisfaction, efficacy and effectiveness of the process, from the perspectives of both client and therapist?
    Data is still under analysis, and will be published soon.

  7. Edmundo Cruz

    How can I get the questionaire to get the losada ratio between two persons?



  1. Meta Learning | Krista's Tech Tidbits - [...] learning focuses on the importance of teams and was part of a research study by Marcial Losada.  Losada’s study…
  2. “The five dysfunctions of a team and how to solve them” - [...] on the ways to bring teams from a “languishing” to a “flourishing” state. We have seen ( that, during…
  3. Positive Discussions Encourage Creativity | METACOGNITION - [...] his groundbreaking research on group dynamics, researcher Marcial Losada identified 15 high performing teams based on profitability, customer satisfaction,…

Submit a Comment

Your email address will not be published. Required fields are marked *

Mobile: +41 79 435 1660

Skype: didiermarlier

5 Route du Village

1884 Villars-sur-Ollon

CH - Switzerland

“Engaging Leadership” has been written for leaders who are about to engage their organisations in change."

10 + 11 =