Lately, I’ve been exploring the concept of complex adaptive systems and their relevance to human performance. Every coach or sports scientist understands that we cannot break down performance into isolated parameters and expect that a single parameter will account for the performance as a whole. The function of any complex system depends on the interaction of its components. As Russell Ackoff once said, you cannot decompose a car and assume that the engine alone will take you to your destination.

Every system (we can assume that an athlete is a system) has a primary component known as an attractor, which is a highly stable component of a system and typically does not fluctuate much over time (Balague et al, 2013). Whenever the system is perturbed (due to training or environmental stressors), it seeks an attractor state — a stable position within the system where the system feels most comfortable, in very basic terms.

For instance, an attractor could be something as straightforward as an athlete with kyphosis. The attractor state in which the athlete is situated is typically reflected in their history, training regimen specificity, and length of exposure. It can even extend to emotional and psychological burdens, which may manifest in the athlete’s overall posture. However, for the sake of this discussion, let’s limit ourselves to the physiological and anthropometric domains.

Figure 1: The visualisation shows the attractor state. The attractor landscape is best explained via ‘hilly terrain’ in which we want to move from one state to another. In this particular scenario — the attractor is fixed in the valley and due to the force of gravity it is very hard to escape from. To leave the attractor state, it is necessary to form an adequate landscape, which will facilitate the transition.

When this particular athlete trains using a specific attractor, it is only a matter of time before it begins to affect other physiological systems. For instance, kyphosis can result in inadequate breathing mechanics, suboptimal biomechanical execution, increase risk of injury and so on. Although it is possible to transition from this state to another, doing so requires modifications to the attractor landscape. This change can be brought about by any interventions that specifically target the issue, such as training, rehabilitation, or cultural changes (e.g., joining a new team or adjusting one’s mindset). The strength of an attractor can be attributed to the duration during which it received attention and grew stronger. In my experience, when capacity or intensity is built upon a deficiency, it only reinforces the attractor state and results in stronger compensatory mechanisms that ultimately impede performance.

When we effectively transform the landscape, we can escape the dead valley. Remaining in one place for an extended period can lead to a stiff pattern that is challenging to modify. This could apply to any physiological, biomechanical, or psychological factor that we can think of. Additionally, it’s important to note that attractors are not only negative; we must intentionally develop attractors that align with the demands of our sport or health-related goals while remaining aware of any negative ones that may harden patterns that are difficult to escape from. Simultaneously, we must recognize that the attractor landscape may shift without even knowing — there could be unintentional psychological load, changes in school timetable, new coach, family issues, and so on, which all contribute to the dynamic reorganization of the system.

Figure 2: The visualisation shows the attractor state in which the landscape has already been modified (training intervention) to facilitate an easier transition from one state to another.

From theory to real-world relevance

The emerging field of complex science in sports has already impacted the conceptualization of skill acquisition, motor control, rehabilitation, team dynamics, and adaptation (Montull et al., 2021; Pol et al., 2020; Pol et al., 2018; Torrents et al., 2016). Simultaneously, there have been efforts to establish a possible direction for the emergence of a new field, such as network physiology, using coordination principles rather than isolated measures of separated physiological systems (Zebrowska et al., 2020; Balagué et al., 2016; Balague et al., 2020).

So far, the focus has been largely theoretical, with no concrete evidence to illustrate the conceptualization. To begin with, let’s examine a straightforward investigation carried out by Den Hartigh et al. (2016) on rowers. The research team organized a sequence of rowing competitions which effectively demonstrated that athletes who were defeated in the first three matches experienced a decline in positive momentum once they began to lose in the fourth competition. This strongly underscores that negative experiences act as a powerful regulator (attractor) of performance, and an athlete’s performance history can offer significant explanatory value.

Furthermore, the possible relevance can be found in multiple physiological mechanisms and phenomena. As Dr Andy Galpin reminds us, physiology offers no free passes; everything matters, but time is the most critical domain. Attractor states can be beautifully illustrated by the phenomenon of functional vs non-functional overreaching. A recent comprehensive review of overtraining syndrome demonstrates the complexity of this issue (Armstrong et al., 2022). To make progress, we require a stimulus that will initially decrease our capabilities but followed by sufficient recovery; it will enhance our performance beyond the previous level. However, if the stimulus persists for too long, recovery is insufficient, or the signal-to-noise ratio is too high, we can fall into a “trapped attractor” state of non-functional overreaching, which is not the desired outcome (see Figure 3). From my personal experience working with both commercial and elite athletes, I have observed that training can become in most cases an environmental stressor rather than an adequate stimulus after a few years of training. This is often due to a suboptimal attractor landscape resulting from poor recovery, inadequate training load management, progression, non-specific training stimuli and other life factors.

Figure 3: The visualisation shows the attractor states of functional X non-functional overreaching. The attractor landscape in A condition is optimized for a successful transition from the original state to the desired destination. In the B condition, the stimulus stayed there for too long, that the original intention to improve was lost and the system settled into a suboptimal state — non-functional overreaching. In other words, we can call it “attractor trapping”, where a system gets stuck in an attractor state that is not the desired outcome.

Moreover, attractor trapping is also one of the strongest arguments against early specialization models in youth athletes. If you have ever tried to have a conversation about why the early specialization is a bad model, you find out soon that it is extremely hard to bring the evidence against it on the physiological level. There is a growing body of literature that supports the idea of early diversification, but due to its extremely complex nature, the precise mechanisms are still very blunt (Mosher et al, 2022).

In my experience, early specialization is typically characterized by high intensity and volume of training, focused on a narrow range of specialized movements. Essentially, we establish a fixed attractor state in our performance landscape from an early age, which can be related to any domain such as skill acquisition, postural control, specific physiological development, autonomous nervous regulation etc. Narrowing the range of attractors limits our options in later stages. I have witnessed amazingly talented 15-year-old swimmers with an extremely rigorous training regimen, but they developed biomechanical, postural, and physiological deficiencies along the way, which resulted in being trapped in their own physiology. Dr Bondurchuk eloquently described this concept - whether using high or low intensity with younger athletes, the results will be the same. The difference is that with high-intensity work, there is no turning back, and the nervous system’s plasticity becomes rigid, making high-intensity and/or increased volume of high-intensity the only way to continue improving (Moyer, 2020). Continuing this path for too long can limit options, and the required change is usually drastic and difficult to accept, requiring significant lifestyle and training changes. Identifying crucial physiological, biomechanical, social and cultural attractors in the development framework of youth athletes might provide the needed conceptual clarity for designing optimal environments and limiting unnecessary damage.

Implications:

1. Coaches and sports scientists should consider athletes as complex adaptive systems and understand the interactions among the individual components that impact the athlete’s performance. They should identify attractors, which are stable states of the system, and work towards modifying the attractor landscape to facilitate a transition to a desired state. Training as well as the lifestyle, cultural and behavioural aspects are essential for transforming the attractor landscape and escaping the dead valley of a stiff pattern that is challenging to modify.

2. It’s vital to signify the non-linearity of the process and highlight the importance of history and previous training exposure, which can significantly impact the effectiveness of interventions. The inclusion of past exposure also invites the model of hysteresis, which has already been used as a marker of stress and tolerance (Montull et al., 2020). Therefore, the holistic overview can help design interventions that can effectively modify the athlete’s attractor landscape and improve performance.

3. Some studies have already utilized the complex system’s approach of “critical slowing down” and the examination of time-series data (such as detrended fluctuation analysis) to analyse tipping points, which could potentially indicate a shift from one state to another (Nazarimehr et al., 2020). Studies in psychology and neurosciences have already successfully identified that the concept of “critical slowing down” might predict depression or the possibility of an epileptic seizure. (Van de Leemput et al., 2014; Maturana et al., 2020). With the growth of artificial intelligence and machine learning, the analysis and prediction of transition phases may become easier.

Reference list:

Armstrong, L. E., Bergeron, M. F., Lee, E. C., Mershon, J. E., & Armstrong, E. M. (2022). Overtraining Syndrome as a complex systems phenomenon. Frontiers in Network Physiology, 1. https://doi.org/10.3389/fnetp.2021.794392

Balagué, N., González, J., Javierre, C., Hristovski, R., Aragonés, D., Álamo, J., Niño, O., & Ventura, J. L. (2016). Cardiorespiratory coordination after training and detraining. A principal component analysis approach. Frontiers in Physiology, 7, 35. https://doi.org/10.3389/fphys.2016.00035

Balagué, N., Hristovski, R., Almarcha, M., Garcia-Retortillo, S., & Ivanov, P. C. (2020). Network Physiology of exercise: Vision and perspectives. Frontiers in Physiology, 11, 611550. https://doi.org/10.3389/fphys.2020.611550

Balague, N., Torrents, C., Hristovski, R., Davids, K., & Araújo, D. (2013). Overview of complex systems in sport. Journal of Systems Science and Complexity, 26(1), 4–13. https://doi.org/10.1007/s11424-013-2285-0

Den Hartigh, R. J., Van Geert, P. L., Van Yperen, N. W., Cox, R. F., & Gernigon, C. (2016). Psychological momentum during and across sports matches: Evidence for interconnected time scales. Journal of Sport & Exercise Psychology, 38(1), 82–92. https://doi.org/10.1123/jsep.2015-0162

Maturana, M. I., Meisel, C., Dell, K., Karoly, P. J., D’Souza, W., Grayden, D. B., Burkitt, A. N., Jiruska, P., Kudlacek, J., Hlinka, J., Cook, M. J., Kuhlmann, L., & Freestone, D. R. (2020). Critical slowing down as a biomarker for seizure susceptibility. Nature Communications, 11(1), 2172. https://doi.org/10.1038/s41467-020-15908-3

Montull, L., Passos, P., Rocas, L., Milho, J., & Balague, N. (2021). Proprioceptive dialogue — interpersonal synergies during a cooperative slackline task. Nonlinear Dynamics, Psychology, and Life Sciences, 25(2), 157–177.

Montull, L., Vázquez, P., Hristovski, R., & Balagué, N. (2020). Hysteresis behaviour of psychobiological variables during exercise. Psychology of Sport and Exercise, 48(101647), 101647. https://doi.org/10.1016/j.psychsport.2020.101647

Mosher, A., Till, K., Fraser-Thomas, J., & Baker, J. (2022). Revisiting early sport specialization: What’s the problem? Sports Health, 14(1), 13–19. https://doi.org/10.1177/19417381211049773

Nazarimehr, F., Jafari, S., Perc, M., & Sprott, J. C. (2020). Critical slowing down indicators. EPL (Europhysics Letters), 132(1), 18001. https://doi.org/10.1209/0295-5075/132/18001

Pol, R., Balagué, N., Ric, A., Torrents, C. Hristovski, R., Kiely, J. (2020). Training or Synergizing? Complex Systems Principles Change the Understanding of Sport Processes. Sports Med –Open, 6, 28. doi: 10.1186/s40798–020–00256–9

Pol, R., Hristovski, R., Medina, D., Balagué, N. (2018). From micro- to macroscopic injuries: Applying the Complex Systems Dynamic Approach to Sports Medicine. British Journal of Sports Medicine, 0, 1–8.

Moyer, J. (2020). And then what? Understanding CNS sensitivity, plasticity and long term development in training. Just Fly Sports. https://www.just-fly-sports.com/understanding-cns-sensitivity-plasticity-and-long-term-development-in-training/

Teques, P., Araújo, D., Seifert, L., Del Campo, V. L., & Davids, K. (2017). The resonant system: Linking brain-body-environment in sport performance. Progress in Brain Research, 234, 33–52. https://doi.org/10.1016/bs.pbr.2017.06.001

Torrents, C., Ric, A., Hristovski, R., Torres-Ronda, L., Vicente, E., & Sampaio, J. (2016). Emergence of exploratory, technical and tactical behavior in small-sided soccer games when manipulating the number of teammates and opponents. PloS One, 11(12), e0168866. https://doi.org/10.1371/journal.pone.0168866

Van de Leemput, I. A., Wichers, M., Cramer, A. O. J., Borsboom, D., Tuerlinckx, F., Kuppens, P., van Nes, E. H., Viechtbauer, W., Giltay, E. J., Aggen, S. H., Derom, C., Jacobs, N., Kendler, K. S., van der Maas, H. L. J., Neale, M. C., Peeters, F., Thiery, E., Zachar, P., & Scheffer, M. (2014). Critical slowing down as early warning for the onset and termination of depression. Proceedings of the National Academy of Sciences of the United States of America, 111(1), 87–92. https://doi.org/10.1073/pnas.1312114110

Zebrowska, M., Garcia-Retortillo, S., Sikorski, K., Balagué, N., Hristovski, R., Javierre, C., Petelczyc, M. (2020). Decreased coupling among respiratory variables with effort accumulation. Europhysics Letters, 132: 28001.