Answers are grounded in this paper and cite their sources.
Physiologists strive to decipher the mysteries of the human body. Our curiosity extends beyond merely answering queries concerning the “ifs” and “buts”—we aim to comprehend the “hows” and “whys.” Amid the labyrinthine sphere of medical research, the realm of physiology emerges as a sovereign, distinguished by its conceptual complexity. Given the inherent human limitations in rationalizing and understanding, the implementation of models becomes indispensable, serving as a tool to distil the intricate realities of physiological systems into manageable units. Models can serve as a bridge between the theoretical domain and the tangible world of empirical data, thereby providing a conduit for exploration and understanding. Effectively, models facilitate the deconstruction of the vast array of physiological inquiries into manageable, distinct questions. This, in consequence, enables motivated and bright scientists to devise pragmatic methods of seeking answers to these unresolved mysteries. The spectrum of physiological models is vast: ranging from proteomics1 to animal models2 and extending into the realm of laboratory3 and applied human physiology.4 They assist in deconstructing complex biological phenomena into constituent parts. In the most minuscule realm of physiological exploration, proteomic and genomic models assume an integral role in representing protein and gene expression profiles of isolated tissues, extending even to individual cells.5 These models aid in unraveling the diversity within a cell population, and forge the path to a superior understanding of submicroscopic (patho-)physiology. Simultaneously, a few rungs higher on the ladder of functional model complexity, animal models such as mice and zebrafish serve as invaluable tools for studying the effects of drug treatments or non-pharmacological interventions.6 All approaches, from modeling the smallest molecular dependencies to full living animal research, illuminate our understanding of human health and disease. Frequently, it is our models that are the catalysts for ground-breaking discoveries. Advancing toward applied human physiology, we encounter another series of models. Not only do our human study subjects serve as models for a broader, generalizable population of humans—our measurement technology itself, which we commonly utilize for physiological examination, operates based on models and assumptions about reality. These models, though seemingly simple, play a pivotal role in converting complex and heterogeneous data into interpretable and ostensibly direct physiological measures. In many instances, these models offer a critical bridge between theoretical physiology and practical health care.7 Nevertheless, while these models are potent tools, they are not devoid of their unique set of limitations and challenges, particularly concerning their proper use and justifiable generalization. For instance, while animal and cellular models are informative, they do not perfectly represent the complex intricacies of human physiology. We acknowledge that mice are not humans and any observed effect in mice merely indicates potential implications for human physiology.8, 9 This disparity can stir controversy among researchers when diverse approaches toward a question yield varying results: For example, whether there is neocytolysis after returning from hypoxic to normoxic environments remains an unresolved question, with conflicting answers derived from human and animal studies.10-12 Similar issues pervade all domains of physiological research, as all are reliant on models for measurement. For example, in-vitro studies using cell lines often overlook intricate cellular interactions and the microenvironment found in living organisms. Genomic and proteomic analyses can be exceedingly precise and reliable if conducted with extreme caution and expertise. However, they become futile if the query is not sufficiently answered by the analysis itself. To state it succinctly, even an impeccable method falters when the model it is used upon fails to generalize, or provide reliable information, to the question at hand. Therefore, extrapolating results from scientific models to a broader context demands caution, as misinterpretation can lead to fallacious conclusions. In applied human physiology, the waters need to be navigated with even more caution: Experiments on humans themselves are based on assumptions and modeled simplifications and there are very few measurement modalities in human physiology that do not rest heavily on extensive presupposition. This culminates in an experimental environment where not only the experiment itself needs to be meticulously evaluated for appropriate generalizability, but also the measurement devices themselves often harbor their own set of limitations. Physiological measurement devices are growing increasingly advanced and offer a wealth of scientifically and clinically meaningful information. However, in many instances, they provide surrogate data, often failing to capture the full reality of the physiological condition under scrutiny. In clinical reality, for instance, oxygen delivery to peripheral capillary beds is a critical clinical marker for evaluating the stability and adequacy of perfusion. Regrettably, this is not what our common devices are genuinely measuring. The difference between an infrared blood oxygenation sensor and the actual delivery of oxygen to the body's cells is striking. While the former provides an immediate snapshot of blood oxygen saturation levels based on the, sometimes, erroneous presupposition that the light absorption of a certain wavelength perfectly reflects the oxygenation status of hemoglobin, the latter is a highly complex process. The actual rate of oxygen delivery to peripheral organs is influenced by a myriad of factors including cardiac output, hemoglobin concentration, pH, 2,3-bisphosphoglyceric acid concentration, local perfusion, and more.13-15 Another rather surprising example is the measurement of arterial blood pressure. Despite being one of the most common medical diagnostic procedures,16, 17 blood pressure measurement, especially when automated, depends heavily on a plethora of assumptions.18 This dependency results in poor measurement reproducibility and a significant propensity to be affected by undetected measurement artifacts.19 The concept of “core body temperature” also highlights the considerable variation in clinical interpretation, both in terms of measurement techniques and the temporal context of these measurements. It is crucial to measure core temperature at the specific location at which the physiological phenomenon is being investigated: for example, the hypothalamus in cases of hypothalamic thermoregulation or the thoracoabdominal region when studying the temperature's impact on the cardiovascular system. Furthermore, it is essential to consider the etiology and the resulting local and temporal spread of temperature changes.20 Altogether, human physiology encompasses the challenges in measurement: Human subjects may behave unpredictably and the experiment itself needs to generalize to a broader understanding of the question at hand. Beyond that, measurement devices need to measure what they purport to, or their measurement of surrogate parameters needs to be interpreted with supreme care. As if all these factors were not enough, we need to ensure the reliability of our measurement results and evaluate whether we require secondary confirmation or even redundant parallel measurement to assure high sensitivity and specificity for whatever effect we wish to examine. This concept is one we became all too familiar with during the COVID-19 pandemic and remains a significant issue in all fields of physiological research.21 Despite these challenges, the field of physiology continues to advance, fueled largely by these very limitations. Issues such as the reproducibility crisis in research, the struggle to find the appropriate statistical approach, setting up a valid experiment,22 and the challenge in differentiating between effect and significance may complicate the picture.23 Yet, it is through confronting and overcoming these difficulties that we stimulate scientific progress. Conscientious researchers and diligent publishing standards are the antidote to overwhelming complexity and act as a safeguard for our field.24 The continuous refinement of our models, the enhancement of our measurement techniques, and the fostering of a culture of rigorous and transparent research can indeed bring us closer to understanding reality. The microscopic and molecular methods have led the way in the decades, progressing at astonishing rates and enabling evermore insights.25 As we bridge this gap in all fields of physiology, we concurrently advance our comprehension of the human body. This bridging has consistently laid the foundation for a multitude of breakthrough discoveries, benefitting millions of patients every day. In a sense, the dichotomy between measurement and reality often sparks the initial interest to pursue a career in science. In physiology, this represents a challenge—one that we collectively embrace. None.
No takes yet. Share an insight, caveat, or question.
Bothe et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: