==== Front Open Access J Sports MedOpen Access J Sports MedOpen Access Journal of Sports MedicineOpen Access Journal of Sports Medicine1179-1543Dove Medical Press 10.2147/OAJSM.S169758oajsm-9-139PerspectiveHow much further for the sub-2-hour marathon? Sousa Caio Victor 1Sales Marcelo Magalhães 2Nikolaidis Pantelis Theodoros 3Rosemann Thomas 4Knechtle Beat 45 1 Graduate Program in Physical Education, Universidade Católica de Brasília, Brasília, Brazil 2 Physical Education Department, Universidade Estadual de Goiás, Quirinópolis, Goiás, Brazil 3 Exercise Physiology Laboratory, Nikaia, Greece 4 Institute of Primary Care, University of Zurich, Zurich, Switzerland, beat.knechtle@hispeed.ch 5 Medbase St. Gallen Am Vadianplatz, St. Gallen, Switzerland, beat.knechtle@hispeed.chCorrespondence: Beat Knechtle, Medbase St. Gallen Am Vadianplatz, Vadianstrasse 26, 9001 St. Gallen, Switzerland, Tel +41 71 226 9300, Fax +41 71 226 9301, Email beat.knechtle@hispeed.ch2018 31 7 2018 9 139 145 © 2018 Sousa et al. This work is published and licensed by Dove Medical Press Limited2018The full terms of this license are available at https://www.dovepress.com/terms.php and incorporate the Creative Commons Attribution – Non Commercial (unported, v3.0) License (http://creativecommons.org/licenses/by-nc/3.0/). By accessing the work you hereby accept the Terms. Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed.The sub-2-hour marathon is a new barrier in endurance running performance, and it has been widely debated in the scientific community. In this review we present a mathematical model to estimate the possible year when a male could break through the sub-2-hour barrier, and also an estimation of when a female could break Paula Radcliffe’s marathon running record. Further, we present several aspects (ie, physiology, nationality, age, biomechanics, pacing, and drafting) that are associated with marathon running performance in elite runners and, finally, the possible characteristics of the male to break the sub-2-hour barrier. In summary, with the results of the developed equations, it is possible that a male athlete can break through the sub-2-hour barrier in the next decade (with Nike® Breaking2 performance 1920–2018 [NBP]: y =0.0417x2–14.18x +3,128; year of 2026; without NBP 1920–2018: y =0.045x2–15.12x +3,194; year of 2027). This marathoner will possibly have a maximal oxygen uptake >85 mL∙kg−1∙min−1 and should perform the race at a pacing higher than 85% of maximal oxygen uptake. In addition, this runner should pay more attention to strength training, endurance strength, speed training, and focus on running training at an intensity above the anaerobic threshold. Most likely, this runner originates from East Africa (especially from Ethiopia) and will have an age of ~27 years. For the females, there is poor evidence regarding the physiological profile of the female marathoner who will break Radcliffe’s record, but the available literature suggests that it will not happen any time soon. Keywords endurancerunningperformancepacingdrafting ==== Body Introduction The marathon run is the most classic Olympic running event. Since it was introduced as a central part of the modern Olympic Games in 1896, it takes place in countless cities worldwide, mostly because of its popular appeal,1 with the participation rates increasing continuously during the last years.2,3 Furthermore, the record times have been progressively improved since the standardization of the peculiar 42.195 km marathon distance.4,5 Sports sciences have significantly evolved since the 1920s, especially regarding exercise physiology, training methods, and thus affecting physical performance.6 In relation to this, a breakpoint regarding endurance performance were the three Olympic gold medals achieved by Emil Zátopek, who won the 5,000 m and the 10,000 m and set a new time record for the marathon at the 1952 Olympic Games in Helsinki. Zátopek was one of the first athletes to use high-intensity interval trainings in his routines, a method that has been exhaustively studied in the last decades.7 Although Zátopek was an Olympic and World Record holder, nowadays his marathon race time would not be enough fast even to qualify to the Olympic Games. Even though the marathon records have progressively improved over the years for both females and males, a time barrier has been put to the marathon distance for males, the sub-2-hour performance (Figure 1A). Whereas some athletes and scientists consider a sub-2-hour marathon as a physiological impossibility,8,9 some other authors believe that it may happen any time soon.4,10,11 In addition, females have also improved their marathon performance since it was included in the Olympic Games in 1984 and increased their participation rates relatively more than males during the last decades.2 In 2003, the British athlete Paula Radcliffe ran the marathon in 2 hours 15 minutes and 25 seconds, setting a new and impressive World Record (Figure 1B). Since then, no females have been able to break it, some authors even consider this female race time of 2 hours and 15 minutes as the equivalent for the male sub-2-hour marathon.12 In this update of the report written by Joyner et al,4 we aimed to summarize the key factors identified in the literature that have been argued as important in improving a marathon performance. We further performed a mathematical analysis to predict when the sub-2-hour marathon would happen for males, and also propose an speculative quantification of the possible aspects that could delay (or be delaying) this breakthrough in human running performance and the sex difference in marathon running performance. Analysis All procedures used in the study were approved by the Institutional Review Board of Kanton St. Gallen, Switzerland, with a waiver of the requirement for informed consent of the participants given the fact that the study involved the analysis of publicly available data. We searched literature (PubMed, Web of Science, and Scopus; reverse search) for keywords related to all aspects that could be related to marathon performance in elite runners: marathon, sub-two, endurance performance, elite run, etc. We obtained data from two publicly available databases (www.olympic.org and www.iaaf.org). The best official marathon race times were recorded from males and females, being: 1999–2018 (up to April 2018) World Championships, Olympics and Annual Record for males and females both; and the best official Olympic Marathon race times since 1920 for males and 1984 for females, which was the year when the marathon running distance was standardized and involved females, respectively. In 2017, an event to reach the sub-2-hour time record was performed (Nike® Breaking2, Nike Inc., Beaverton, OR, USA), in which an athlete reached the closest performance ever since (2 hours and 25 seconds), but not yet enough to break through the barrier (https://news.nike.com/news/breaking2-results). Since the marathon record times have been progressively decreasing over the years (Figure 1), we performed different regressions analyses to determine the possible year of the sub-2-hour performance. We calculated two nonlinear regressions using second-order polynomial adjustment, one with and one without the Nike Breaking2 record (Table 1). We used a polynomial regression because the data seem to be curvilinear (Figure 1). Moreover, polynomial adjustments are also applied if the curvilinear response function is unknown, since a polynomial function is a good approximation to the true function13 and because linear regressions are also limited for performance progressions, since the performance never attenuates or decreases. A further analysis from 1920 to 2012 was also performed to test the accuracy of the model, with the data from 2013 to 2018 to be applied in the formula. All data analyses were performed using the Microsoft Excel (MS Excel 2011 for Mac, Google, Mountain View, CA, USA) and GraphPad Prism (v. 6.0, GraphPad Software, La Jolla, CA, USA). Perspective Data analyses indicate that if performance improvement would be linear over the years, the sub-2-hour marathon performance is quite very near (2018), considering or not the Nike Breaking2 event. However, the limits of human performance have been more difficult to break through every year.14 In the marathon running, every minute in the final race time is achieved by a combination of factors including a great athlete with unique biomechanical10 and physiological15 qualities, running in a good day, with favorable environmental and race course conditions. Previous data analyses to predict marathon indicate contradictory results.4,5 Joyner et al4 analyzed male’s World Records and suggested that the sub-2-hour marathon will break around 2020–2022, based on a 10–20 seconds of performance improvement every year. In this update, that includes important recent records by Kimetto and Kipchoge, the results indicate that it may happen in 2026–2027 (non-linear analysis). On the other hand, using a velocity-based analysis, Weiss et al5 report that the sub-2-hour marathon will not be broken in the century; the authors further discuss that the female’s marathon record are not supposed to be beaten any time soon. Another study reported that the sub-2-hour marathon is unlikely to happen with an analysis based on the World Records gap between males and females.16 The biomechanical and environmental effects that could affect human performance in marathon running have been widely discussed in a review by Hoogkamer et al.10 In summary, cooperative drafting, tailwind, downhill, and footwear modifications seem to be fundamental to reduce the metabolic cost of running, possibly leading to a great physical performance. However, although downhill courses and cooperative strategies with fresh runners to overcome the air resistance are not allowed by the International Association of Athletics Federations (IAAF), the Nike Breaking2 performed by Eliud Kipchoge (Kenyan, 33 years old) had the best conditions possible in a non-downhill course, but the best marathon performance of all time still was not enough to beat the sub-2-hour barrier. Despite the fact that the marathon performance of Kipchoge is unofficial, it may give some enlightenment regarding what else could be done to break through the sub-2-hour barrier. The physiological aspects that could lead an athlete to achieve such a performance have been widely discussed from different points of view.15,17–21 Physiological aspects The physiological aspects that have been considered as the determinants of performance in endurance athletes are maximum oxygen consumption (VO2max), running velocity at the anaerobic threshold, and running economy.15,17,22,23 Champions in endurance running sports (ie, half-marathon, marathon, ultra-marathon) have a VO2max of 70–85 mL∙kg−1∙min−1.15 In view of the above, although VO2max is not the most determinant physiological factor for running performance, in order to overcome the barrier of running a marathon below 2 hours, the athlete must have a VO2max higher than the endurance champions have today. In addition, running a marathon corresponds to an intensity (average pace) equal to 75%–85% of the VO2max.15 Thus, perhaps, to overcome the barrier of running a marathon below 2 hours, athletes may need to maintain an average running pace higher than literature has often been pointing out that such distance has been realized.15 On the other hand, the running economy, which is traditionally expressed as mL∙kg−1∙km−1, seems to be a better predictor of performance than VO2max.18 An individual with a lower VO2max but with better running economy may have an enhanced endurance performance than an individual with a higher VO2max.18 When comparing the running economy of Spanish runners with Eritrean runners, Lucia et al18 showed that there was no significant difference in VO2max, but Eritrean runners had a better running economy. The Eritrean runner, who became World Cross Country champion over 12 km in 2007 (Zersenay Tadese; born 1982; also participated in the Nike Breaking2), used only 150 mL∙kg−1∙km−1 when he ran at speeds of 17, 19, and 21 km∙h−1. This is the lowest VO2 reported at these speeds,24 suggesting that running economy can be an important predictor of running performance. Strength training, strength endurance, and also speed training have been suggested as an important intervention for the improvement of the running economy. Johnston et al19 examined the effect of 10-week strength training on a range of physiological variables. The training did not result in any significant increase in VO2max, but the running economy at a given submaximal load improved by 4%. Berryman et al25 in a meta-analysis assessed the effects of strength training in middle- and long-distance runners and concluded that endurance athletes could benefit from it by improving their energy cost of running, locomotion, maximal power, and maximal strength. Thus, strength training and sprint training seem to be important interventions to allow runners, at some point, to be able to run a marathon in <2 hours. In addition, it appears that training at the anaerobic threshold intensity is also effective in improving running economy, and in turn, favoring performance in endurance trials.20 Tjelta et al21 demonstrated that both VO2max and running economy explain 89% of the anaerobic threshold velocity. Moreover, Maffulli et al26 reported strong (r =0.80–0.90) and significant (p<0.05) correlations between running velocity at the anaerobic threshold and performance over running distances from 5,000 m to the marathon. In order to increase running velocity at the anaerobic threshold and, in turn, running performance, studies demonstrated that running training above the anaerobic threshold velocity may be more effective.27,28 Taken together, the three main physiological factors affecting a marathon performance in elite runners could be summarized as running economy, anaerobic threshold, and VO2max, in this order of importance (Figure 2). It is elementary that each one of these aspects affects the other, but evidence seems clear pointing that the sub-2-hour marathon runner will have an outstanding running economy with a great running velocity at anaerobic threshold. Pacing, aging, and nationality Another major aspect for endurance performance is the pacing strategy, the ability to affectively expend energy to prevent premature fatigue prior to the completion of the event.29–31 In that regard, there are evidences suggesting that, in a marathon, older athletes run at a steadier pace (smaller changes) than younger athletes.30 It has been suggested that an “even” or “constant” pace is the best strategy for endurance performance.31,32 Moreover, the pacing strategy used by elite marathoners have been suggested as not being optimal,33 which means that coaches and athletes may still have room for improvement in their tactical methods. Noble and Chapman34 recently reported that African marathoners reach a peak of performance earlier than non-Africans counterparts (27.1±0.1 versus 30.1±0.1 years, respectively). Further, African marathoners have also the fastest marathon time compared with non-Africans (2:05:47±0:00:06 versus 2:08:41±0:00:06 h:min:s). These authors also showed that African marathoners specialize earlier and have an elevated rate of performance improvement and, finally, that there is a positive correlation between age of specialization and rate of improvement.34 More specifically, there are also reports that East Africans (ie, Kenyan, Ethiopian) are faster than others.14,35,36 Especially, Ethiopian males were faster than all other males in the marathon run when the performances of the world’s best runners in marathon from the IAAF during 1999–2015 were considered.14 Furthermore, Tam et al37 compared physiological variables in elite Kenyan marathoners with European controls. Although the authors reported no difference in VO2max (64.9±5.8 versus 63.9±3.7 mL·kg−1.min−1) and energy economy in submaximal running speed (18 km h−1: 178±16 mL·kg−1 km−1 for both), Kenyan runners were faster in half- and full marathon.37 For the females, it seems that they reach a peak of performance a little later than males, and most of the female athletes participating in marathons are from Ethiopia and Kenya.14 Although the current rankings are dominated by east African females, the World Record was achieved by a British runner, Paula Radcliffe. Furthermore, Joyner38 recently discussed the possible sex differences for marathon performance, and apart the social factor, the VO2max in females is the possible limiting factor. On the other hand, recent evidence shows that Mo Farrah, Olympic champion in the 5,000 and 10,000 m and recently retired from the tracks, started a career in marathon events. This athlete has the conditions and may be the one to run a <2 hour marathon. However, he would be an outlier by being a non-African runner and 35 years of age. Suggesting that it may be an exception to the athlete profile we have outlined previously to break this barrier. Taken together, although eastern Europeans have a good historical performance regarding marathon records, it is reasonable to assume that the male to break the sub-2-hour marathon is likely to be an East African (especially from Ethiopia) at around 27 years old with a steady pacing strategy. From a practical point of view, this would mean running the marathon at a steady pace of 4′34′′/mile or 2′50′′/km. Although difficult to imagine that a few seconds could affect a marathon performance, Kipchoge finish the Nike Breaking2 with 5 km partials with a mean pace variation from 2′49′′/km to 2′51′′/km, with the slowest partial being the last 2′53′′/km and missed the record by 25 seconds. Perhaps a steady pacing that makes it possible to stay a few seconds ahead of the deadline could help motivate the athlete within the last and most demanding kilometers. Drafting Performing an activity in a sheltered position is defined as “drafting,” which has been studied in many sports, such as cycling,39 swimming,40 and running.41 In summary, evidences show that drafting improves performance in endurance sports and that the benefit would likely increase with increasing speed of performance.42 In a marathon, the first half has an optimal performance possibly because of potential saving of energy due to drafting. However, the second half is more difficult because there are few runners that could provide drafting for more than the first half of a marathon race.10 As pointed out by Hoogkamer et al10 and performed in the Nike Breaking2, the best option would be to have several runners in a loop course, then when the first ones starts to fatigue, fresh runners would replace then. Nevertheless, this scenario is not allowed by the IAAF.43 For the females, little is known regarding the physiological peculiarities of elite marathoners, but there is no reason to believe that regarding pacing strategies and drafting they would respond different. Paula Radcliffe’s record is a clear outline marathon performance, and the limited evidence available suggests that this record will endure a little longer.5 Conclusion In summary, although it is possible that a male athlete breaks through the sub-2-hour barrier in the next decade (2025–2028), we warn that empirical models may be intrinsically weak, but it could give us an idea that this breakout human performance is close. This marathoner will possibly have a VO2 max >85 mL∙kg−1∙min−1 and should perform the race at a running intensity higher than 85% of VO2 max. In addition, this runner should focus on running training at intensities above the anaerobic threshold. Most likely, this runner originates from East Africa (especially from Ethiopia) and will have an age of ~27 years. Regarding Radcliffe’s record, it could be beaten at any time; however, there is poor evidence regarding the physiological profile of the female marathoner who will achieve it. Author contributions CVS, MMS, PTN, TR, and BK have made substantial contributions to conception, design, data acquisition, data analysis, and interpretation of this manuscript; CVS, MMS, PTN, TR, and BK have also worked on the draft, critically revised for important intellectual content, and approved the final version of the manuscript. All authors agreed to all aspects of the work in ensuring that questions related to the accuracy or integrity of the work are appropriately investigated and resolved. Disclosure The authors report no conflicts of interest in this work. Figure 1 Data points of the best marathon performance of males (A) and females (B) including the breakthrough human performance for each category. Figure 2 Speculative estimates of key points in each fundamental factor affecting marathon performance in elite runners. Notes: Each fundamental factor was based on hierarchical importance as biomechanics,10,44 physiology,15,22 tactics,29,30,42 and genetic pool.4,14,30 Abbreviation: LT, lactate threshold. Table 1 Regression analyses to determine the possible year of breakthrough in the sub-2-hour marathon time Year R2 Equation 1920–2018  With NBP 2026 0.941 y =0.0417x2–14.18x +3128  Without NBP 2027 0.942 y =0.045x2–15.12x +3194 1920–2013  With NBP AL =99.1% 0.934 y =0.039x2–13.63x +3088 Note: R2: determination coefficient. Abbreviations: NBP, Nike® Breaking2 performance; AL, agreement level. ==== Refs References 1 Jones H The Expert’s Guide to Marathon Training: Includes a 16-week Programme to Finishe Those 26.2 Miles! London Carlton 2003 2 Knechtle B Nikolaidis PT Zingg MA Rosemann T Rüst CA Half-marathoners are younger and slower than marathoners Springerplus 2016 5 1 76 26844023 3 Martin D Marathon running as a social and athletic phenomenon: historical and current trends Pedoe DT Marathon Medicine London The Royal Society of Medicine Press Limited 2001 28 38 4 Joyner MJ Ruiz JR Lucia A The two-hour marathon: who and when? J Appl Physiol 2011 110 1 275 277 20689089 5 Weiss M Newman A Whitmore C Weiss S One hundred and fifty years of sprint and distance running – past trends and future prospects Eur J Sport Sci 2016 16 4 393 401 26088705 6 Ivy JL Exercise physiology: a brief history and recommendations regarding content requirements for the kinesiology major Quest 2007 59 1 34 41 7 Milanovic Z Sporis G Weston M Effectiveness of high-intensity interval training (HIT) and continuous endurance training for VO2 max improvements: a systematic review and meta-analysis of controlled trials Sports Med 2015 45 10 1469 1481 26243014 8 Hill AV The physiological basis of athletic records Lancet 1925 206 5323 481 486 9 Liu Y Schutz RW Prediction models for track and field performances Meas Phys Educ Exerc Sci 1998 2 4 205 223 10 Hoogkamer W Kram R Arellano CJ How biomechanical improvements in running economy could break the 2-hour marathon barrier Sports Med 2017 47 9 1739 1750 28255937 11 Boullosa DA de Almeida JA Simoes HG The two-hour marathon: how? J Appl Physiol 2011 110 1 292 294 12 Hunter SK Joyner MJ Jones AM The two-hour marathon: what’s the equivalent for women? J Appl Physiol 2015 118 10 1321 1323 25525211 13 Motulsky H Christopoulos A Fitting Models to Biological Data Using Linear and Nonlinear Regression: A Practical Guide to Curve Fitting Oxford Oxford University Press 2004 14 Nikolaidis PT Onywera VO Knechtle B Running performance, nationality, sex and age in 10 km, half-marathon, marathon and 100 km ultra-marathon IAAF 1999–2015 J Strength Cond Res 2017 31 8 2189 2207 28731980 15 Joyner MJ Coyle EF Endurance exercise performance: the physiology of champions J Physiol 2008 586 1 35 44 17901124 16 Tucker R Santos-Concejero J The unlikeliness of an imminent sub-2-hour marathon: historical trends of the gender gap in running events Int J Sports Physiol Perform 2017 12 8 1017 1022 27967280 17 Tjelta LI Shalfawi SA Physiological factors affecting performance in elite distance runners Acta Kinesiologiae Universitatis Tartuensis 2016 22 7 19 18 Lucia A Esteve-Lanao J Olivan J Physiological characteristics of the best Eritrean runners-exceptional running economy Appl Physiol Nutr Metab 2006 31 5 530 540 17111007 19 Johnston RE Quinn TJ Kertzer R Vroman NB Improving running economy through strength training Strength Cond J 1995 17 4 7 13 20 Sjodin B Jacobs I Svedenhag J Changes in onset of blood lactate accumulation (OBLA) and muscle enzymes after training at OBLA Eur J Appl Physiol Occup Physiol 1982 49 1 45 57 6213407 21 Tjelta LI Tjelta AR Dyrstad SM Relationship between velocity at anaerobic threshold and factors affecting velocity at anaerobic threshold in elite distance runners Int J Appl Sports Sci 2012 24 1 8 17 22 Joyner MJ Modeling: optimal marathon performance on the basis of physiological factors J Appl Physiol 1991 70 2 683 687 2022559 23 di Prampero PE Atchou G Bruckner JC Moia C The energetics of endurance running Eur J Appl Physiol Occup Physiol 1986 55 3 259 266 3732253 24 Lucia A Olivan J Bravo J Gonzalez-Freire M Foster C The key to top-level endurance running performance: a unique example Br J Sports Med 2008 42 3 172 174 discussion 174 18048445 25 Berryman N Mujika I Arvisais D Roubeix M Binet C Bosquet L Strength training for middle- and long-distance performance: a meta-analysis Int J Sports Physiol Perform 2018 13 1 57 63 28459360 26 Maffulli N Capasso G Lancia A Anaerobic threshold and performance in middle and long distance running J Sports Med Phys Fitness 1991 31 3 332 338 1798301 27 Billat V Sirvent P Lepretre PM Koralsztein JP Training effect on performance, substrate balance and blood lactate concentration at maximal lactate steady state in master endurance-runners Pflugers Arch 2004 447 6 875 883 14740217 28 Tanaka K Watanabe H Konishi Y Longitudinal associations between anaerobic threshold and distance running performance Eur J Appl Physiol Occup Physiol 1986 55 3 248 252 3732252 29 Skorski S Abbiss CR The manipulation of pace within endurance sport Front Physiol 2017 8 102 28289392 30 Nikolaidis PT Knechtle B Effect of age and performance on pacing of marathon runners Open Access J Sports Med 2017 8 171 180 28860876 31 Abbiss CR Laursen PB Describing and understanding pacing strategies during athletic competition Sports Med 2008 38 3 239 252 18278984 32 Thomas K Stone MR Thompson KG Clair Gibson A St Ansley L The effect of self- even- and variable-pacing strategies on the physiological and perceptual response to cycling Eur J Appl Physiol 2012 112 8 3069 3078 22194003 33 Angus SD Did recent world record marathon runners employ optimal pacing strategies? J Sports Sci 2014 32 1 31 45 23879745 34 Noble TJ Chapman RF Elite African marathoners specialize earlier than elite non-African marathoners Int J Sports Physiol Perform 2017 13 1 102 106 35 Cribari M Rust CA Rosemann T Onywera V Lepers R Knechtle B Participation and performance trends of East-African runners in Swiss half-marathons and marathons held between 2000 and 2010 BMC Sports Sci Med Rehabil 2013 5 1 24 24289794 36 Onywera VO East African runners: their genetics, lifestyle and athletic prowess Med Sport Sci 2009 54 102 109 19696510 37 Tam E Rossi H Moia C Energetics of running in top-level marathon runners from Kenya Eur J Appl Physiol 2012 112 11 3797 3806 22382667 38 Joyner MJ Physiological limits to endurance exercise performance: influence of sex J Physiol 2017 595 9 2949 2954 28028816 39 Olds T Norton K Craig N Olive S Lowe E The limits of the possible: models of power supply and demand in cycling Aust J Sci Med Sport 1995 27 2 29 33 8521030 40 Chollet D Hue O Auclair F Millet G Chatard JC The effects of drafting on stroking variations during swimming in elite male triathletes Eur J Appl Physiol 2000 82 5–6 413 417 10985595 41 Pugh LG Oxygen intake in track and treadmill running with observations on the effect of air resistance J Physiol 1970 207 3 823 835 5532903 42 Zouhal H Ben Abderrahman A Prioux J Drafting’s improvement of 3,000-m running performance in elite athletes: is it a placebo effect? Int J Sports Physiol Perform 2015 10 2 147 152 24912074 43 International Association of Athletics Federations Competition Rules 2016–2017 Monaco International Association of Athletics Federations 2015 44 Arellano CJ Kram R Partitioning the metabolic cost of human running: a task-by-task approach Integr Comp Biol 2014 54 6 1084 1098 24838747