Entia non sunt multiplicanda praeter necessitatem

Abstract

I have read with great interest the paper by Tucker et al. (2006) on the effect of ambient temperature on power output regulation during prolonged exercise at a fixed rating of perceived exertion (RPE). It is well established that high ambient temperature increases perceived exertion during prolonged exercise at a fixed power output (Nybo & Nielsen, 2001). Therefore, it is not surprising that Tucker et al. demonstrated a greater reduction in power output when exercising at a fixed RPE in the hot trial. These authors, however, went a step further and concluded that this study provides evidence that neural activation of locomotor muscles is regulated by a subconscious intelligent system that, based on afferent feedback from thermoreceptors and the rate of heat storage, calculates the reduction in power output necessary to prevent excessive heat accumulation during exercise at high ambient temperature. This strong conclusion is reflected in the title which I found misleading after reading the entire paper. The findings of Tucker et al. are certainly consistent with this inventive theory called the central governor model (St Clair Gibson & Noakes, 2004). However, they do not provide either direct or indirect evidence for the existence of such a subconscious regulatory system as the authors failed to rule out other possible explanations. For example, various sensations (e.g. leg effort, dyspnoea and warm discomfort) contributing to the overall conscious sensation of effort, stress or discomfort that is felt during exercise (Noble & Robertson, 1996) might directly motivate subjects to voluntary reduce their power output (Jones & Killian, 2000), i.e. to take a conscious decision to reduce their pedalling rate and/or load. This psychobiological model predicts that subjects exercising in the heat would consciously reduce power output to compensate with less leg effort and dyspnoea (Killian et al. 1992) the increased discomfort associated with high ambient temperature (i.e. warm discomfort). This voluntary behaviour would also reduce metabolic heat production and associated thermoregulatory responses (e.g. sweating), thus contributing to keeping warm discomfort and overall RPE within the prescribed limits (Gagge et al. 1969). A crucial point made by Tucker et al. in favour of their ‘feedback–feedforward’ control system is that the high rate of heat storage during the first few minutes of exercise in the hot trial preceded the significant reduction in power output. However, this delayed adjustment of power output is also consistent with the relatively slow increase in the sensation of warm discomfort at the beginning of constant workload exercise in the heat (Gagge et al. 1969). The data on ‘thermal comfort’ provided by Tucker et al. do not rule out this possibility as the scale they used seems to rate thermal sensation (from ‘much too cool’ to ‘much too hot’) rather than comfort. Unfortunately, a reference for this ‘modified Borg category scale’ was not provided. This is not a trivial detail as these two constructs, thermal sensation and thermal comfort, are separate and not always interrelated (Gagge et al. 1969; Cabanac, 1971). Considering this was a paper on perceptually regulated exercise, it is also regrettable that sufficient details on RPE procedures (e.g. standardized instructions given to the subjects and anchoring methods used during familiarization) were not provided either in the Methods or in the cited reference (Borg, 1982). In conclusion, the results of this study can be satisfactorily explained by a model in which conscious sensations play a direct role in the behavioural regulation of homeostasis during exercise without the need for any additional entity such as the subconscious intelligent system proposed by Tucker and colleagues. Indeed, Cabanac et al. (1992) stated that ‘sensation gave decisional advantages to the first animals which possessed it, by freeing them from the need for an infinitely complex hardware reflex network in their nervous system'. Therefore, until Tucker et al. provide some real evidence in favour of the central governor model (or some evidence against the alternative psychobiological model briefly described here), we should remember that ‘entia non sunt multiplicanda praeter necessitatem’ (Ockham's razor principle) and choose, among two equally predictive theories, the least complex.