1. Introduction
The physiological (Baláš et al., Reference Baláš, Panáčková, Strejcová, Martin, Cochrane, Kaláb, Kodejška and Draper2014) and psychological (Mangan et al., Reference Mangan, Andrews, Miles and Draper2025) demands of climbing can vary in several ways including the route, means of fall protection, difficulty and foreknowledge of the route (Draper et al., Reference Draper, Dickson, Fryer, Blackwell, Winter, Scarrott and Ellis2012; Fryer et al., Reference Fryer2013; Hardy & Hutchinson, Reference Hardy and Hutchinson2007) which can alter the chance and consequences of failure. If a lead climber cannot finish a route, they will fall, a potential physical threat, or must ask to be taken tight on the rope, a potential ego threat (Hurni, Reference Hurni2003). If a climber responds inappropriately, stressors can significantly negatively affect their state and performance (Garrido-Palomino & España-Romero, Reference Garrido-Palomino and España-Romero2019; Hardy & Hutchinson, Reference Hardy and Hutchinson2007; Pijpers et al., Reference Pijpers, Oudejans and Bakker2005). Consequently, a key to climbing is managing stress (Hörst, Reference Hörst2010), thus being in a challenge state, rather than a threat state (Hase et al., Reference Hase, O’Brien, Moore and Freeman2019), while lead climbing may facilitate performance.
The findings of Giles et al. (Reference Giles, Mangan and Draper2025) support coaches’ assertions of lead climbing’s often limiting psychological challenges (Hurni, Reference Hurni2003; Macleod, Reference Macleod2010). Somatic anxiety is thought to stem from an inability to differentiate between perceived and actual risk as well as rational and irrational fear (Morris et al., Reference Morris, Davis and Hutchings1981), and must be overcome for optimal climbing performance (Binney & McClure, Reference Binney and McClure2005), notably on-sight climbing (Vasile et al., Reference Vasile, Stănescu, Pelin and Bejan2022). Somatic anxiety has been linked with a threat of physical harm while cognitive anxiety being more associated with social or performance evaluation (Morris et al., Reference Morris, Davis and Hutchings1981; Morris & Liebert, Reference Morris and Liebert1973). A potential lead fall, particularly, appears to be associated with perceived injury risk, even indoors, where most injuries result from chronic overuse, not acute trauma (Backe et al., Reference Backe, Ericson, Janson and Timpka2009). Considerable skill and experience are needed for climbers to appropriately respond to potentially evocative stimuli allowing for maintained performance (Gajdošík et al., Reference Gajdošík, Baláš and Draper2020). Incorrectly managed stress may adversely affect performance (Garrido-Palomino & España-Romero, Reference Garrido-Palomino and España-Romero2019; Sanchez et al., Reference Sanchez, Boschker and Llewellyn2010), including movement control (Pijpers et al., Reference Pijpers, Oudejans, Holsheimer and Bakker2003). This again speaks to lead climbing in a threat versus challenge state (Hase et al., Reference Hase, O’Brien, Moore and Freeman2019).
Unlike ascent style, route difficulty alters physical demands, with unchanged consequences of failure (Janot et al., Reference Janot, Steffen, Porcari and Maher2000; Mermier et al., Reference Mermier, Robergs, McMinn and Heyward1997). Greater route difficulty increases physiological demands (Baláš et al., Reference Baláš, Panáčková, Strejcová, Martin, Cochrane, Kaláb, Kodejška and Draper2014; Janot et al., Reference Janot, Steffen, Porcari and Maher2000; Mermier et al., Reference Mermier, Robergs, McMinn and Heyward1997), and lowers the likelihood of the climber finishing the route. Therefore, relative route difficulty (RRD) is likely a significant climbing stressor.
Fall potential may induce anxiety, limiting ability to complete a route (Hague & Hunter, Reference Hague and Hunter2011). Increasing RRD creates a state where perceived task demands may exceed coping resources (Lazarus, Reference Lazarus1991). Cognitive evaluation theory proposes two levels of cognitive appraisal, which mediate stressors and stress response (Lazarus, Reference Lazarus1991). Primary appraisal establishes the situation’s importance to the individual and potential endangerment. These demand evaluations are made about climbers’ perceptions of danger (physical or esteem), uncertainty, and effort (physical and psychological). Previous research has explored differences in perceptions of uncertainty and effort, but only Giles et al. (Reference Giles, Mangan and Draper2025) has examined perceived danger and found demands appraisals of perceived danger for intermediate climbers while leading would yield threat cognitive evaluations. However, this was true only of the subjective self-report, not cardiovascular reactivity. Differences in RRD are another way to explore such danger appraisals.
This study, with differences in RRD of on-sights and conceivably differences in the likelihood of participants’ success, is also hypothesized to produce differences in demand evaluations from perceived danger. Participants attempting a route at or below their ability may be expected to experience a ‘challenge’ cognitive appraisal state when secondary appraisal perceives they have sufficient coping potential to meet the situational demands (indicated by a positive cognitive evaluation score). Conversely, those climbing a route beyond their ability experience a ‘threat’ state when secondary appraisal indicates insufficient coping potential, which in the case of lead climbing may make harm seem potentially imminent (Lazarus, Reference Lazarus1991). Perceived coping ability, performance outcomes and climbers’ resultant behavior and emotional response to state are likely dictated by: perceived control (Jones, Reference Jones1995), self-efficacy (Bandura, Reference Bandura1997) and achievement disposition (Duda, Reference Duda, Elliot and Dweck2005), as described in the Theory of Challenge and Threat States in Athletes (TCTSA; Hase et al., Reference Hase, O’Brien, Moore and Freeman2019; Jones et al., Reference Jones, Meijen, McCarthy and Sheffield2009). The TCTSA provides a theoretical framework for the objective quantification of participants’ demand and resource evaluations of such factors (Seery, Reference Seery2011) and a sports-focused framework for the biopsychosocial model of Challenge and Threat (Blascovich & Tomaka, Reference Blascovich and Tomaka1996), which has its origins in the work of Lazarus and Folkman (Reference Lazarus and Folkman1984) and Dienstbier (Reference Dienstbier1989). Challenge states have been hypothesized to have a small, but stable effect to improve performance outcomes through adaptive physiological and cognitive responses and threat states to hinder performance through maladaptive responses (Behnke & Kaczmarek, Reference Behnke and Kaczmarek2018; Hase et al., Reference Hase, O’Brien, Moore and Freeman2019). This has previously been seen in other sports (Moore et al., 2012; Turner et al., Reference Turner, Jones, Sheffield and Cross2012) but no challenge and threat differences were seen with different ascent styles (ie lead vs top rope) in intermediate climbers (Giles et al., Reference Giles, Mangan and Draper2025), possibly because lead climbing was not a sufficient stressor for experienced, intermediate climbers. Simply, if a climber doubts their ability to cope with the demands of a stressor arising due to perceived danger, uncertainty and effort, anxiety is likely to result and affect performance. This is like the related concepts of approach and avoidance and such motivational states can have implications for many biological systems (Mendes & Park, Reference Mendes, Park and Elliot2014).
Research on psychological demands and behavioral changes’ effect with altered route difficulty is limited (Mangan et al., Reference Mangan, Andrews, Miles and Draper2025). Unlike intermediate climbers, advanced and expert climbers have not been found to have increased mental anxiety or psychophysiological stress while lead climbing compared with top roping (Dickson et al., Reference Dickson, Fryer, Draper, Winter, Ellis and Hamlin2012b; Fryer et al., 2013b). However, no studies have yet examined smaller ability group differences in indoor lead climbing. Hardy and Hutchinson (Reference Hardy and Hutchinson2007), in line with Hardy and Whitehead (Reference Hardy and Whitehead1984), reported increased anxiety may be expected when a climber attempts a route at, or near, their ability limit. Participants in Hardy and Hutchinson (Reference Hardy and Hutchinson2007) climbed routes at their lead limit, one and two grades below on traditional outdoor climbs. With increased difficulty, cognitive and somatic anxiety and activation were significantly greater. There was also a significant difference in effort measures (heart rate (HR), perceived exertion and perceived mental effort). Interestingly, the belayer’s rating of the climber’s performance also increased, with superior performance during the harder ascents. Since these ascents were completed with the greater objective danger of traditional climbing, participants were not tasked with routes above their ability. Climbing performance has also been assessed using geometric entropy (GE), an analysis of the climber’s center of mass during the duration of their ascent (Cordier et al., Reference Cordier, France, Bolon and Pailhous1993). Geometric entropy has been found to decrease (improve) with increased route knowledge and climbing experience (Cordier et al., Reference Cordier, France, Pailhous and Bolon1994; Watts et al., Reference Watts, Espan͂a-Romero, Ostrowski and Jensen2021). Additionally, the Climber’s Movement Performance Assessment Tool (CM-PAT; Taylor et al., Reference Taylor, Giles, Panáčková, Mitchell, Chidley and Draper2020) uses climbing-specific considerations of movement quality as assessed by experienced coaches. The CM-PAT has shown excellent reliability and explains a much greater variance in climbing performance than GE.
Increased route difficulty, relative to the ability of a climber, creates a situation where the consequences of a failure are identical, however, the likelihood of the climber failing to complete the climb increases. The purpose of this study was to investigate the role of ability on the resultant psychophysiological, emotional and behavioral changes to an on-sight of a route below, at, or above the climber’s self-reported indoor on-sight ability. Assessment of participants’ cognitive evaluation in response to task instructions were made via cardiovascular reactivity and self-report inventories. Climbers’ perception of their emotional state and psychophysiological responses were also assessed, along with climbing performance. Therefore, the aim of this study was to examine differences between climber ability relative to the difficulty of a route, and their subjective psychological, objective physiological and behavioral responses.
1.1. Hypotheses
The greater the difficulty of the route, relative to the ability of the climber:
H1: the lower the likelihood of the climber successfully completing the ascent, the lower the coaches’ assessment of performance and the less fluid the displacement of mass will be.
H2: the greater cognitive and somatic anxiety and the lower self-confidence will be.
H3: the greater the rise in anticipatory heart rate will be. Given previous psychophysiology research, it is unlikely that cortisol will differ significantly.
Further:
H4: An ascent of a route above a climber’s on-sight ability will be considered ego-threatening, resulting in threat reactivity, greater stress response and reduced performance when compared to that of a route set at a climber’s on-sight ability.
H5: A climber’s challenge and threat index will be positively correlated to climbing performance, lower and more positive interpretations of cognitive anxiety, greater positive self-confidence, greater self-efficacy, challenge interpretation and greater demand resources.
H6: The coaches’ assessment of performance scores will be correlated to lower and more positive cognitive anxiety, greater positive self-confidence and somatic anxiety, lower anticipatory heart rate; greater self-efficacy, challenge interpretation and greater demand resources.
2. Methods
The same assessment techniques as Giles et al. (Reference Giles, Mangan and Draper2025) were largely employed in this study. Please consult the relevant appendices for additional methodological detail where provided.
2.1. Participants
Sixty-one intermediate to advanced climbers (Draper et al., Reference Draper, Giles, SchöFfl, Fuss, Watts, Wolf, BaláŠ, EspañA Romero, Gonzalez, Fryer, Fanchini, Vigouroux, Seifert, Donath, Spoerri, Bonetti, Phillips, Stöcker, Bourassa-Moreau and Abreu2016) volunteered for this study. Participants were included based on being active, competent lead climbers, self-reporting on-sight abilities of French 6a+ to 6c+ (f6a+ to 6c+; International Rock Climbing Research Association (IRCRA) 12-16; Yosemite Decimal System (YDS) 5.10c-5.11c) and recruited via posters, targeted social media advertisements and word of mouth from the climbing community of the North of England. All participants were healthy, normotensive adults. Refer to Table 1 for experience, anthropometric, ability and fitness characteristic data. Ethical approval was granted for this study by the University of Derby’s College of Life and Natural Sciences Research Ethics Committee. All participants gave informed written consent prior to participation in the study and after an explanation of the methods.
Participants anthropometric and climbing experience characteristics (mean ± SD)
Table 1. Long description
The table consists of three main sections: anthropometrics, experience, and grade. From top to bottom, anthropometric variables include age, height in metres, and mass in kilogramme. For age, combined participants average 33.4 plus or minus 10.1 years, males 34.0 plus or minus 10.6, females 31.6 plus or minus 8.2. Height is 1.75 plus or minus 0.07 metres combined, 1.77 plus or minus 0.07 for males, 1.68 plus or minus 0.05 for females. Mass is 70.0 plus or minus 9.3 kilogramme combined, 72.7 plus or minus 8.2 for males, 61.1 plus or minus 6.9 for females. The experience section lists years climbing and sessions a week. Years climbing are 9.9 plus or minus 9.3 combined, 10.5 plus or minus 10.1 for males, 7.7 plus or minus 5.9 for females. Sessions a week are 2.6 plus or minus 1.1 combined, 2.6 plus or minus 1.2 for males, 2.6 plus or minus 1.1 for females. The grade section includes indoor on-sight and indoor red-point scores using I R C R A standards. Indoor on-sight is 13.6 plus or minus 1.7 combined, 13.8 plus or minus 1.6 for males, 12.9 plus or minus 1.8 for females. Indoor red-point is 15.7 plus or minus 1.8 combined, 15.8 plus or minus 1.7 for males, 15.2 plus or minus 1.9 for females. Abbreviations are defined as m for metres, kg for kilogramme, O S for on-sight, R P for red-point, and I R C R A for international rock climbing research association.
Note: m metres; kg kilogramme; OS on-sight; RP red-point; IRCRA international rock climbing research association.
Participants were sorted by on-sight ability relative to the difficulty of the study route (f6b) being at lead limit (climber’s on-sight lead level; CLL; n = 20; f6b to 6b+; YDS 5.10d-5.11a), above on-sight grade (CLLabove; n = 17; f6a to 6a+; YDS 5.10b-5.10c) or below on-sight grade (CLLbelow; n = 24; f6c to 6c+; YDS 5.11b-5.11c). This classification into the three groups remained constant regardless of participants’ success or failure in ascending the study route. Refer to Table 2 for participants’ experience, anthropometrics and ability data sorted by group.
Participants’ climbing experience, anthropometrics and ability by group (mean ± SD)
Table 2. Long description
Starting from the top, the table is divided into three columns for C L L above (n equals 17), C L L (n equals 20), and C L L below (n equals 24). Under Experience, Years Climbing are 6.4 plus or minus 9.4, 6.7 plus or minus 5.1, and 15.0 plus or minus 9.9 respectively. Sessions a week are 2.3 plus or minus 1.1, 2.1 plus or minus 0.8, and 3.3 plus or minus 1.0. Anthropometrics include Gender: 8 female and 9 male for C L L above, 2 female and 18 male for C L L, 4 female and 20 male for C L L below. Age in years is 32.8 plus or minus 11.6, 32.7 plus or minus 9.3, and 34.5 plus or minus 10.0. Height in meters is 1.72 plus or minus 0.09, 1.75 plus or minus 0.07, and 1.77 plus or minus 0.06. Mass in kilograms is 66.6 plus or minus 10.8, 71.4 plus or minus 8.7, and 71.3 plus or minus 8.3. The note at the bottom states C L L refers to climber's on-sight lead level.
Note: CLL climber’s on-sight lead level.
2.2. Experimental design
All participants attended a single session at a commercial climbing gym in the North of England. Participants were not provided with exact route details until they received route instructions. Participants were asked not to alter their training prior to the study and avoid strenuous exercise to allow adequate rest. Additionally, participants were asked to avoid alcohol and caffeine within 24 hours, and consuming food within two hours of the sessions. As salivary samples were collected, participants were instructed not to brush their teeth 30 minutes before attendance, nor consume water five minutes before a sample (Gonzalez et al., Reference Gonzalez, Del Mar Bibiloni, Pons, Llompart and Tur2012). Guideline adherence was confirmed verbally before commencing testing.
Each climb was videoed (Panasonic HD V720 digital video camera) for GE analysis (see Appendices A & B for detailed video and GE method), climb time and coaches’ performance assessment (Taylor et al., Reference Taylor, Giles, Panáčková, Mitchell, Chidley and Draper2020). Increased GE resulted in a less fluent, noisier path of the center of mass. Climbing performance variables (base of support, transitioning movement, coordination, technique, and tactics) were used for CM-PAT scoring, ranging from 1 (poor/non-existent) to 5 (flawless) with unique descriptors (see Appendix C). Mean category scores and an overall score were calculated. Two investigators independently produced CM-PAT scores, with an interrater coefficient of variation of 0.82 (95% confidence interval: 0.42-0.94), demonstrating good reliability of the scale.
Anthropometrics were recorded in a quiet classroom. Afternoon sessions were conducted to minimize circadian rhythm’s influence, particularly on salivary cortisol (Thuma et al., Reference Thuma, Gilders, Verdun and Loucks1995). Prior to climbing HR, blood pressure reactivity and salivary cortisol were assessed in response to task instructions. A separate section of the climbing gym was utilized to control for noise, temperature, lighting and ensure against the presence of others (except the belayer/researchers) for each trial.
2.3. Climbing wall and route setting
A single route graded French 6b (12 IRCRA; YDS 5.10c), confirmed by four expert climbers, was set for the study. The route had a 6a first half and 6b+ top half, giving an overall grade of 6b. The easier beginning reduced ground fall potential. Participants were not provided with information on the route they were to attempt prior to starting to climb, except the grade, hold color and which quickdraws to use. The first quickdraw was at 3.1 meters, the next nine quickdraws were 1.2 ± 0.1 meters apart. In the event of a fall, the climber travelled a short distance before being arrested by the belayer and the rope.
2.4. Procedure – pre, warm-up & climbing
Figure 1 represents the testing session procedures. After explaining procedures, ascertaining health history, fitness to participate and informed consent, participants were then fitted with climbing shoes, harness and optional chalk-bag. Participants completed a thorough warm-up adapted from Binney and McClure (Reference Binney and McClure2006), Gresham (Reference Gresham2007) and Tenke and Higgins (Reference Tenke and Higgins1999) and used by Draper et al. (Reference Draper, Dickson, Blackwell, Priestley, Fryer, Marshall, Shearman, Hamlin, Winter and Ellis2011b) and Dickson et al. (Reference Dickson, Fryer, Draper, Winter, Ellis and Hamlin2012b) followed by ten minutes of seated recovery, then a pre-climb data collection period comprising of: five minutes of rest, one minute of instructions and three minutes of mental preparation. During this time participants were equipped with a Polar H7 chest strap and V800 HRM (Polar, Finland) and, once seated, the Finapres Portapres Model-2 (Finapres Medical Systems B.V., Amsterdam, The Netherlands). Finger arterial pressure was recorded continuously using an appropriately sized cuff on the mid-phalanx of the left middle finger. HR reactivity to the presented task instructions was assessed to determine task engagement, an important prerequisite for challenge and threat cardiovascular reactivity analysis (Blascovich et al., Reference Blascovich, Vanman, Mendes and Dickerson2011). Although pre-ejection period and HR are both considered markers of task engagement (increased HR and decreased pre-ejection period), only HR was used as the Portapres cannot measure pre-ejection period (Seery, Reference Seery2011). Participants were required to sit still and quiet, upright, with their arm supported at heart level and legs facings forwards, bent at a 90-degree angle. After five minutes of data collection, participants were presented with audio instructions detailing the upcoming climbing task, via headphones (QC 25, Bose). HR and blood pressure were recorded during this time.
Timeline for anthropometrics, baselines and climbing for each participant.
Figure 1. Long description
From left to right, the timeline is divided into six labeled blocks: Explanation of procedure, consent and health history for 15 minutes; Warm-up for 15 minutes subdivided into Pulse Raising, Mobilising, and Gentle Climbing, each 5 minutes; Rest for 10 minutes; C V recording for 9 minutes subdivided into Rest for 5 minutes, Instructions for 1 minute, and Mental Preparation for 3 minutes; Climb for 10 minutes subdivided into Visual Inspection for 5 minutes and Climbing for 5 minutes; Recovery for 15 minutes. Above each block, colored shapes indicate data collection points: pale green square for Baseline C V recording, yellow square for Post-instruction C V recording, orange square for Post-instruction self-report, blue square for Pre-Climb Emotions, red circle for Salivary Cortisol, and blue pentagon for Video. The key at the bottom defines each symbol and C V as cardiovascular. Arrows below the timeline indicate the duration of each phase.
The audiotaped instructions described the upcoming climb (Appendix D Audio transcripts), in which high task demands were promoted, typical of motivated performance situations. Participants were informed the task assessed performance and of route difficulty. The final part of the instructions was to prepare by thinking about their climbing performance for three minutes. After presenting the task instructions, participants completed the combined inventory assessing self-efficacy, cognitive evaluation, perceived control, task importance and demand resource evaluation (see Appendix E for more detail on psychological methods). Finally, a salivary cortisol sample was collected.
Participants were then shown the study route. Following a route inspection, participants completed the Immediate Anxiety Measurement Scale (IAMS) in relation to the upcoming climb to collect somatic and cognitive anxiety, self-confidence intensity and direction data. A ‘thermometer’ style scale was used to measure both the valence and intensity of participants’ perceived emotion on an x and y scale (Houtman & Bakker, Reference Houtman and Bakker1989). The IAMS’ validity has been determined by Thomas et al. (Reference Thomas, Hanton and Jones2002) compared to the CSAI-2 with good agreement (r = .61 to .70) and has been preferred with time limitations. Finally, participants attempted the route. Participants began climbing when ready, self-paced, protected from falling with a harness, rope and an experienced belayer. Climbing HR was recorded continuously. Salivary cortisol was sampled on returning to the ground after a 15-minute passive recovery period.
2.5. Data and statistical analysis
Baseline to pre-climb and pre- to post-climb delta values were calculated for several dependent variables. All statistical analysis was conducted using SPSS (version 22; Chicago IL) and Microsoft Excel (Version 2015; Redmond WA). Univariate outliers were first identified as more than 3.3 SD from the mean (Tabachnick et al., Reference Tabachnick, Fidell and Osterlind2001). After identification, outliers were winsorized to a value 1% larger or smaller than the next most extreme score (Moore et al. (2012), this was necessary for two participants’ cardiovascular data. Normal distribution and variance homogeneity were then assessed through visual inspection of the frequency histogram, Shapiro-Wilk’s tests and examining variance around the mean using box plots (if the maximum variance was less than three times the mean then equal variance was assumed). This is the normally accepted rule to determine whether the analysis of variance (ANOVA) test is reliable. Descriptive statistics were calculated for all variables and reported as mean ± SD.
Blood pressure signals were analyzed with BeatScope (Version 1.1a) to calculate HR, cardiac output and total peripheral resistance which were used to differentiate challenge and threat states (see Table 3). Cardiovascular reactivity scores were calculated for cardiac output and total peripheral resistance by subtracting the raw cardiovascular responses for the last minute of baseline from the average raw cardiovascular responses across one-minutes post presentation of task instructions. Average cardiac output and total peripheral resistance reactivity were combined into the challenge and threat index (Moore et al., 2013; Turner et al., Reference Turner, Jones, Sheffield, Slater, Barker and Bell2013; Turner et al., Reference Turner, Jones, Sheffield, Barker and Coffee2014). The challenge and threat index was calculated by converting average cardiac output and average total peripheral resistance reactivity values into z scores and summing them. Cardiac output was assigned a weight of +1 while total peripheral resistance was assigned a weight of -1 so that larger values reflected challenge reactivity.
Cardiovascular indices/Challenge and Threat Index calculations
Table 3. Long description
From top to bottom, the table has three columns: Measure, Calculation, and Unit. First row: Heart Rate, no calculation, unit is b dot min to the minus one. Second row: Heart Rate reactivity, calculation is H R sub P O S T minus H R sub P R E, unit is b dot min to the minus one. Third row: Cardiac Output, calculation is C O equals S V times H R, unit is L per min. Fourth row: Cardiac Output reactivity, calculation is C O sub P O S T minus C O sub P R E, unit is L per min. Fifth row: Total Peripheral Resistance, calculation is open parenthesis M A P forward slash C O close parenthesis times 80, unit is dyn dot s dot c m to the 5th power. Sixth row: Total Peripheral Resistance reactivity, calculation is T P R sub P O S T minus T P R sub P R E, unit is dyn dot s dot c m to the 5th power. Seventh row: Challenge and Threat Index, calculation is C O sub R E A C T I V I T Y Z dash S C O R E plus open parenthesis T P R sub R E A C T I V I T Y Z dash S C O R E times minus one close parenthesis, unit is blank. Table footnotes define abbreviations: H R is heart rate, C O is cardiac output, S V is stroke volume, T P R is total peripheral resistance, M A P is mean arterial pressure.
Notes: HR heart rate; CO cardiac output; SV stroke volume; TPR total peripheral resistance MAP mean arterial pressure.
Analyses of covariance (ANCOVA) were conducted for each independent variable. For dependent variables significantly affected by the covariate, results of the analysis of covariates were presented (including adjusted means and standard errors (SE) for the dependent variable). To control for increasing error rate due to multiple testing, conceptually linked independent variables were tested together using a multivariate analysis of variances (MANOVA or MANCOVA). ANOVA, ANCOVA or two-way ANOVA tests were used for all other comparisons, except pre-post task instruction HR, which was assessed with a paired samples t-test. Post-hoc least significant difference (LSD) tests were used to explore the source of the mean differences between groups for each significant ANOVA, while controlling for the error rate. Pearson’s correlations were used to assess the relationship between the weighted challenge and threat index and performance with psychophysiological, performance, emotions and self-report characteristics.
For H1, MANOVA was used to assess across climbing performance variables/GE and ANOVA within each CM-PAT subscale and GE. For H2, ANOVA was used to assess cognitive/somatic anxiety and self-confidence between groups. For H3, anticipatory HR and cortisol were considered together using MANOVA and independently using ANOVA across groups. For H4, threat reactivity, stress response and climbing performance were assessed using MANOVA (cardiovascular reactivity, GE/CM-PAT subscales) and ANOVA (all other variables in Tables 8 & 9). For H5 and H6, correlation was used to examine the relationships between challenge/threat, climbing performance and other relevant variables.
For all analysis, the critical α-level was set at 0.05; corrections for multiple comparisons were made using Benjamini and Hochberg (Reference Benjamini and Hochberg1995) false-discovery rate (FDR) method, which has been supported for use instead of Bonferroni adjustments (Glickman et al., Reference Glickman, Rao and Schultz2014). Unadjusted and adjusted p values were calculated and, where necessary, presented to provide an indication of the likelihood of type I and type II error rates; given it has been previously argued that reducing the possibility of type II error is preferable in exploratory research (Hoad & Monks, Reference Hoad and Monks2011; Perneger, Reference Perneger1998). Effects sizes were determined using η p 2 for multiple comparisons and interpreted with the following effect size demarcations: small effect: η p 2 = .01; medium effect: η p 2 = .06; large effect: η p 2 = .14 (Richardson, Reference Richardson2011).
3. Results
3.1. Pre-climb measurements
Subjective assessment of the intensity and direction of somatic and cognitive anxiety and self-confidence data can be seen in Table 4. The results highlighted significant differences with large effect sizes in somatic anxiety’s intensity and direction. Post-hoc LSD were significant and indicated meaningful differences occurred between CLLabove and both CLL (MD = 21.1, 95% CI [14.7, 27.4]) and CLLbelow (MD = 25.4, 95% CI [19.3, 31.5]), while the direction was significant between CLLabove and both CLL (MD = 21.2, 95% CI [12.8, 29.6]) and CLLbelow (MD = 14.5, 95% CI [6.4, 22.5]). Post-hoc LSD were significant and indicated meaningful differences between CLLabove and both CLL (MD = 11.1, 95% CI [0.9, 21.3]) and CLLbelow (MD = 16.8, 95% CI [7.0, 26.6]). Finally, self-confidence intensity and direction were both significant with large effect sizes. Post-hoc LSD were significant and revealed meaningful differences between CLLbelow and both CLLabove (MD = 17.6, 95% CI [8.9, 26.3]) and CLL (MD = 19.1, 95% CI [10.7, 27.5]); and the direction between CLLabove and both CLL (MD = 17.6, 95% CI [8.9, 26.3]) and CLLbelow (MD = 19.1, 95% CI [10.7, 27.5]).
Pre-climb emotional state, assessed immediately prior to climbing for the three ability groups (mean ± SD)
Table 4. Long description
Beginning at the top, the table lists somatic anxiety, cognitive anxiety, and confidence, each subdivided into intensity and direction. For somatic anxiety intensity, C L L above group (n equals 17) shows 40.1 plus or minus 11.0, C L L (n equals 20) shows 19.0 plus or minus 9.5, and C L L below (n equals 24) shows 14.7 plus or minus 8.5. ANOVA F value is 37.715, p is less than .005, eta p squared is .565, F D R is .008. Somatic anxiety direction: C L L above is minus 8.5 plus or minus 13.1, C L L is minus 1.7 plus or minus 16.7, C L L below is 12.7 plus or minus 9.7, F is 13.951, p less than .005, eta p squared .325, F D R .017. Cognitive anxiety intensity: C L L above is 35.1 plus or minus 15.9, C L L is 24.0 plus or minus 17.3, C L L below is 18.3 plus or minus 13.4, F is 5.917, p .005, eta p squared .169, F D R .042. Cognitive anxiety direction: C L L above is minus 5.8 plus or minus 14.1, C L L is minus 2.3 plus or minus 16.9, C L L below is 4.0 plus or minus 16.5, F is 1.963, p .150, eta p squared .063. Confidence intensity: C L L above is 26.2 plus or minus 13.7, C L L is 39.2 plus or minus 19.0, C L L below is 48.1 plus or minus 13.7. Adjusted mean (S E): C L L above is 27.1 (3.7), C L L is 39.2 (3.4), C L L below is 47.4 (3.1), F is 8.773, p less than .005, eta p squared .235, F D R .025. Confidence direction: C L L above is minus 2.71 plus or minus 14.05, C L L is 14.9 plus or minus 14.43, C L L below is 16.44 plus or minus 11.05, F is 12.105, p less than .005, eta p squared .298, F D R .033. Notes clarify C L L as climber’s on-sight lead level, ANOVA as analysis of variance, F D R as false-discovery rate, and star as sport climbing significant covariate.
Notes: CLL climber’s on-sight lead level. ANOVA analysis of variance; FDR false-discovery rate; * % sport climbing significant covariate.
Anticipatory HR and delta cortisol were considered together to assess pre-climb psychophysiological alterations (Table 5). The difference between groups on the combined dependent variables was significant (p = .037; η p 2 = .089). Follow-up one-way ANOVAs showed significance with a large effect size in pre-climb HR, but no significant change in cortisol. Post-hoc LSD showed a significant difference in mean pre-climb HR between CLLabove and CLLbelow (MD = 20.9%, 95% CI [9.4, 32.3]).
The psychological components of anticipatory heart rate and delta cortisol concentrations for the three ability groups (mean ± SD)
Table 5. Long description
From left to right, the first column lists psychological components: Anticipatory Heart Rate (percent) and Cortisol percent Pre-Post. The next three columns show group means plus or minus standard deviations for C L L above (n equals 17), C L L (n equals 20), and C L L below (n equals 24). For Anticipatory Heart Rate, values are 36.3 plus or minus 22.7, 24.7 plus or minus 18.3, and 15.4 plus or minus 19.6, respectively. For Cortisol percent Pre-Post, values are 12.9 plus or minus 35.6, 15.0 plus or minus 33.4, and 8.8 plus or minus 31.1. The MANOVA column for Anticipatory Heart Rate reports p equals .037 and eta sub p squared equals .089. The ANOVA columns for Anticipatory Heart Rate show F sub 2, 58 equals 6.687, p equals .002, eta sub p squared equals 0.187, and F D R equals .002. For Cortisol percent Pre-Post, MANOVA is .204, ANOVA F sub 2, 58 is .816, p is 0.007, and F D R is blank. Table footnotes define C L L as climber's on-sight lead level, MANOVA as multivariate analysis of variance, b dot min super minus 1 as beats per minute, and F D R as false-discovery rate.
Notes: CLL climber’s on-sight lead level. MANOVA multivariate analysis of variance; b.min-1 beats per minute; FDR false-discovery rate.
Self-report measures
Self-efficacy, cognitive evaluation, perceived control and demand resources can be seen in Table 6 with self-efficacy and demand resources showing significant differences with large effect sizes. Post-hoc LSD indicated self-efficacy was significantly greater for both CLL (MD = 1.4, 95% CI [0.6, 2.2]) and CLLbelow (MD = 2.3, 95% CI [1.5, 3.1]) compared with CLLabove. Self-efficacy was also found to be significantly greater for CLLbelow (MD = 0.9, 95% CI [0.1, 1.6]) than CLL. Post-hoc LSD showed significant demand resource evaluation differences between CLLabove and both CLL (MD = 2.4, 95% CI [1.2, 3.6]) and CLLbelow (MD = 3.6, 95% CI [2.5, 4.8]) as well as between CLL and CLLbelow (MD = 1.2, 95% CI [0.2, 2.3]).
Self-report measures assessed post instructions for each of the three ability groups (mean ± SD). For cognitive evaluation, more +ve score indicates a challenge state, -ve more threatening; for demand resources, more +ve indicates coping resources outweighing demands
Table 6. Long description
Starting from the top row, the table lists four self-report measures: Self-Efficacy, Cognitive Evaluation, Perceived Control, and Demand Resource. For each measure, three columns display mean plus or minus standard deviation for C L L above (n equals 17), C L L (n equals 20), and C L L below (n equals 24). Self-Efficacy scores are 3.3 plus or minus 1.1, 4.7 plus or minus 1.4, and 5.6 plus or minus 1.2 respectively. Cognitive Evaluation scores are 1.5 plus or minus 1.7, 1.1 plus or minus 1.2, and 1.2 plus or minus 1.4. Perceived Control scores are 5.3 plus or minus 1.6, 5.9 plus or minus 1.0, and 5.8 plus or minus 1.5. Demand Resource scores are negative 2.1 plus or minus 1.5, 0.4 plus or minus 1.7, and 1.6 plus or minus 2.0. To the right, ANOVA columns show F values, p values, eta sub p squared, and F D R for each measure. Self-Efficacy has F equals 6.847, p less than .005, eta sub p squared .367, F D R .013. Cognitive Evaluation has F equals 0.345, p .710, eta sub p squared .012. Perceived Control has F equals 0.866, p .426, eta sub p squared .029. Demand Resource has F equals 20.818, p less than .005, eta sub p squared .418, F D R .025. Significant differences are observed for Self-Efficacy and Demand Resource across groups.
Notes: CLL climber’s on-sight lead level. MANOVA multivariate analysis of variance; ANOVA analysis of variance; FDR false-discovery rate.
3.2. Task Instructions
Task engagement
HR reactivity to the task instructions was calculated to assess task engagement (Table 7), a key prerequisite for challenge and threat cardiovascular reactivity analysis (Blascovich et al., Reference Blascovich, Vanman, Mendes and Dickerson2011). HR differences were assessed with a two-way mixed ANOVA for the main effects of ‘time’ (pre- or post-instruction), ‘group’ (CLLabove, CLL, CLLbelow), as well as the interaction effect ‘time*group’. There was no statistically significant interaction for ‘time*group’ or for the main effect ‘group’. However, a significant and large effect existed for ‘time’ (p < .005; η p 2= .259). Post-hoc paired samples t-test showed HR to have increased significantly for all groups pre-post instructions. Additionally, participants indicated success on the route was important to them, with task importance scores of ≥ 4.3, a one-way ANOVA did not find a significant difference between ability groups. Finally, the manipulation check indicated all participants engaged in task-relevant thoughts while considering the upcoming climbing task, supporting the HR and task engagement data in asserting all groups experienced a motivated performance situation.
Engagement with task instructions assessed with task importance and pre-post heart rate (b.min-1) (mean ± SD)
Table 7. Long description
The table presents two main measures: heart rate and task importance, each subdivided into specific rows. For heart rate, pre-instruction values are 83.6 plus or minus 12.0 for C L L above (n equals 17), 80.2 plus or minus 12.8 for C L L (n equals 20), and 79.2 plus or minus 13.9 for C L L below (n equals 24). Post-instruction heart rate is 85.4 plus or minus 12.8 for C L L above, 82.4 plus or minus 12.5 for C L L, and 81.0 plus or minus 14.5 for C L L below. The two-way mixed ANOVA for heart rate reports p equals .906 and eta sub p squared equals .003. For task importance, the row ‘How important is doing well?’ shows 4.5 plus or minus 1.4 for C L L above, 4.3 plus or minus 1.5 for C L L, and 4.5 plus or minus 1.9 for C L L below. ANOVA statistics for this row are F sub 2, 58 equals 0.143, p equals .867, and eta sub p squared equals .005. All values are reported as mean plus or minus standard deviation. C L L refers to climber's on-sight lead level; b dot min to the minus 1 means beats per minute.
Notes: CLL climber’s on-sight lead level. ANOVA analysis of variance; b.min-1 beats per minute.
Cardiovascular markers
Alterations in the cardiovascular markers of cardiac output and peripheral resistance were assessed continuously at rest prior to, during and post presentation of the task instructions (Table 8). Cardiovascular reactivity was assessed with a one-way MANOVA (cardiac output and total peripheral resistance reactivity). The difference between groups on the combined dependent variables was statistically significant. Follow-up one-way ANOVAs did not indicate significance for the increase in cardiac output but were significant with a medium effect size for the decrease in total peripheral resistance reactivity. Post-hoc LSD revealed significant differences in the total peripheral resistance reactivity between CLLabove and CLL (MD = 60.9 dyn.s.cm5, 95% CI [12.4, 109.4]) and CLLbelow (MD = 51.9 dyn.s.cm5, 95% CI [1.5, 102.4]). The combined challenge and threat index did not differ significantly, although it was negative (towards threat) in CLLabove. Resting post instruction cortisol alterations were measured post presentation of the instructions, while it was lower in CLLbelow, a one-way ANOVA was not significant.
Cardiac output and total peripheral resistance cardiovascular markers pre-post task instructions (mean ± SD)
Table 8. Long description
Starting from the top row, the table lists cardiovascular markers for three C L L groups: C L L above (n equals 17), C L L (n equals 20), and C L L below (n equals 24). Cardiac output (liters per minute) is shown pre-instruction as 5.2 plus or minus 0.8, 5.3 plus or minus 1.0, and 5.1 plus or minus 1.3, and post-instruction as 5.5 plus or minus 0.9, 5.7 plus or minus 1.1, and 5.4 plus or minus 1.4. Peripheral resistance (dyn dot s dot cm to the 5 power) is pre-instruction 1397 plus or minus 287, 1372 plus or minus 296, and 1545 plus or minus 404, and post-instruction 1404 plus or minus 298, 1327 plus or minus 312, and 1481 plus or minus 382. Reactivity rows show cardiac output (liters per minute) as 0.25 plus or minus 0.25, 0.38 plus or minus 0.29, and 0.34 plus or minus 0.25, with peripheral resistance (dyn dot s dot cm to the 5 power) as 7.3 plus or minus 70.7, negative 44.6 plus or minus 61.0, and negative 53.6 plus or minus 90.4. The MANOVA column for cardiac output reactivity reports p less than .005 and eta sub p squared equals .617. ANOVA columns show F sub 2 comma 58 equals 1.215, p equals .304, eta sub p squared equals .040 for cardiac output reactivity, and F equals 3.466, p equals .038, eta sub p squared equals .107 for peripheral resistance reactivity. Challenge and threat index values are negative 0.81 plus or minus 1.53, 0.35 plus or minus 1.74, and 0.28 plus or minus 1.73, with ANOVA F equals 2.755, p equals .072, eta sub p squared equals .087. Cortisol (nanomoles per liter) is 3.22 plus or minus 3.33, 2.52 plus or minus 1.86, and 2.18 plus or minus 1.63, with ANOVA F equals 0.997, p equals .376, eta sub p squared equals .034. Notes clarify abbreviations: C L L is climber’s on-sight lead level, ANOVA is analysis of variance, nmol per liter is nanomoles per liter, C O is cardiac output, dyn dot s dot cm to the 5 power is vascular resistance, F D R is false-discovery rate.
Notes: CLL climber’s on-sight lead level. ANOVA analysis of variance; nmol/L nanomoles per litre; CO cardiac output; dyn.s.cm5 vascular resistance; FDR false-discovery rate.
3.3. Climbing Task Demands
Fourteen CLLabove, seven CLL and zero CLLbelow did not complete the route. Data for successful and unsuccessful participants are presented by group in Table 9. Differences were assessed with two-way ANOVAs for the main effects of ‘group’ and ‘success’, as well as the interaction effect ‘group*success’. There was no statistically significant interaction for ‘group*success’ for climbing time nor for ‘success’. However, differences in the main effect of ‘group’ were significant with a medium effect size (p = .034; η p 2 = .113). Post-hoc LSD showed significant differences between CLLbelow and both CLLabove (MD = 40.0 sec, 95% CI [5.4, 74.6]) and CLL (MD = 44.8 sec, 95% CI [16.4, 73.3]). There were no significant HR differences (see Figure 2). There was no significant interaction for ‘group*success’ for mean HR, furthermore no significant main effect was indicated for ‘success’ or ‘group’. There was no statistically significant interaction for ‘group*success’ for peak HR and no significant effect was indicated for ‘group’. However, differences in the main effect of ‘success’ were significant a with medium effect size (p = .034; η p 2 = .083). Post-hoc LSD indicated unsuccessful participants had significantly greater peak HR than successful participants (MD = 18.6 b.min-1, 95% CI [3.8, 33.4]).
Group differences in success, climbing time and heart rate (mean ± SD)
Table 9. Long description
From left to right, the columns are grouped as C L L above, C L L, and C L L below. Each group contains columns for unsuccessful (F) and successful (S) ascents, with sample sizes: C L L above F n equals 14, S n equals 3; C L L F n equals 7, S n equals 13; C L L below F n equals 0, S n equals 24. The rows list climb time in seconds, average heart rate in beats per minute, and peak heart rate in beats per minute. For climb time, values are: C L L above F 202.6 plus or minus 68.9, S 213.9 plus or minus 28.7; C L L F 224.0 plus or minus 43.3, S 202.2 plus or minus 36.5; C L L below S 168.3 plus or minus 33.8. For average heart rate: C L L above F 157.1 plus or minus 11.7, S 157.5 plus or minus 20.8; C L L F 158.8 plus or minus 12.0, S 151.0 plus or minus 12.1; C L L below S 148.2 plus or minus 18.4. For peak heart rate: C L L above F 174.4 plus or minus 22.4, S 156.1 plus or minus 20.3; C L L F 181.8 plus or minus 14.7, S 160.9 plus or minus 31.3; C L L below S 161.6 plus or minus 19.0. Unsuccessful ascents are absent in C L L below. Notes clarify abbreviations: C L L is climber’s on-sight lead level, S is successful ascent, F is unsuccessful ascent, A N O V A is analysis of variance, sec is seconds, b dot min to the minus 1 is beats per minute, F D R is false-discovery rate.
Notes: CLL climber’s on-sight lead level. S successful ascent; F unsuccessful ascent; ANOVA analysis of variance; sec seconds; b.min-1 beats per minute; FDR false-discovery rate.
shows mean HR by group immediately before climbing and at each quickdraw. Differences in climbing HR were greater for CLLabove than CLL and CLLbelow.
Figure 2. Long description
The y-axis is labeled Heart Rate in beats per minute, ranging from 80 to 180. The x-axis is Climb Phase, with points labeled Pre-Climb, Q D 1 through Q D 9, and Top. Three lines represent C L L above in blue circles, C L L in orange triangles, and C L L below in gray squares. All groups start with lower heart rates pre-climb, then rise sharply to Q D 1, and increase more gradually through Q D 9 and Top. C L L above consistently has the highest heart rate, peaking near 170 beats per minute at Q D 6 to Top. C L L is intermediate, peaking near 160 beats per minute, while C L L below remains lowest, peaking near 150 beats per minute. The gap between groups widens as the climb progresses.
3.4. Climbing Performance
Performance data by group are in Table 10. The MANOVA across CM-PAT subscales and ANOVAs within GE and each CM-PAT subscale were significant with large effect sizes. Post-hoc LSD were significant and showed greater GE in CLLabove (less fluent movement) than both CLL (MD = 0.099, 95% CI [0.013, 0.186]) and CLLbelow (MD = 0.126, 95% CI [0.043, 0.210]). Like GE, total CM-PAT score was greatest in CLLbelow, with significant differences between CLLabove and both CLL (MD = 9.3, 95% CI [3.6, 15.0]) and CLLbelow (MD = 18.5, 95% CI [13.0, 24.0]), and between CLL and CLLbelow (MD = 9.2, 95% CI [4.0, 14.5]).
Geometric entropy, coaches’ score and movement analysis for the three ability groups (mean ± SD)
Table 10. Long description
From the top row, geometric entropy values are 0.91 plus or minus 0.15 for C L L above (n equals 17), 0.81 plus or minus 0.12 for C L L (n equals 20), and 0.78 plus or minus 0.11 for C L L below (n equals 24). MANOVA p equals .006, eta sub p super 2 equals .357. ANOVA F equals 4.746, p equals .013, eta sub p super 2 equals .147, FDR equals .050. Next, total coaches’ score increases from 40.1 plus or minus 10.0 to 48.2 plus or minus 8.2 to 57.4 plus or minus 8.2 across the groups. ANOVA F equals 23.085, p less than .005, eta sub p super 2 equals .443, FDR equals .005. Base of support rises from 2.7 plus or minus 0.6 to 3.3 plus or minus 0.6 to 3.9 plus or minus 0.7. ANOVA F equals 19.946, p less than .005, eta sub p super 2 equals .408, FDR equals .010. Transitioning scores are 2.4 plus or minus 0.7, 2.9 plus or minus 0.6, and 3.9 plus or minus 0.6. ANOVA F equals 29.639, p less than .005, eta sub p super 2 equals .505, FDR equals .015. Co-ordination scores are 2.6 plus or minus 0.8, 3.2 plus or minus 0.6, and 3.9 plus or minus 0.7. ANOVA F equals 17.491, p less than .005, eta sub p super 2 equals .376, FDR equals .020. Technique scores are 2.0 plus or minus 0.6, 2.7 plus or minus 0.6, and 3.5 plus or minus 0.8. ANOVA F equals 24.005, p less than .005, eta sub p super 2 equals .453, FDR equals .025. Tactics scores are 3.0 plus or minus 0.8, 3.6 plus or minus 0.7, and 4.4 plus or minus 0.5. ANOVA F equals 21.765, p less than .005, eta sub p super 2 equals .429, FDR equals .030. Notes clarify C L L as climber’s on-sight lead level, MANOVA as multivariate analysis of variance, ANOVA as analysis of variance, and FDR as false-discovery rate.
Notes: CLL climber’s on-sight lead level. MANOVA multivariate analysis of variance; ANOVA analysis of variance; FDR false-discovery rate.
3.5. Correlations: Challenge and threat index and performance
The challenge and threat index allows reactivity comparison to be assessed with a single measure. Pearson’s correlations (Table 11) revealed several significant relationships. Greater challenge and threat index was associated with greater self-confidence interpretation and perceived control. Greater CM-PAT score was associated with lower somatic anxiety and greater self-confidence intensity and direction, greater self-efficacy and greater demand resources evaluation.
Mean ± SD and correlation (r) for performance, psychological variables, and the challenge and threat index
Table 11. Long description
Beginning at the top, the table lists psychophysiological variables: Heart Rate percent (24.3 plus or minus 19.6, correlation with challenge and threat index negative point zero one zero, with coaches’ assessment negative point one three one), and Cortisol percent (11.9 plus or minus 32.5, correlation negative point zero five five, coaches’ assessment point three one three). Performance variables include Coaches’ Score (49.2 plus or minus 11.4, correlation point two two zero). Geometric Entropy (0.82 plus or minus 0.13, correlation negative point one two two, coaches’ assessment negative point four zero seven, double asterisk for significance). Emotional variables are divided into Somatic Anxiety Intensity (23.2 plus or minus 14.3, correlation negative point zero nine nine, coaches’ assessment negative point five six zero, double asterisk), Somatic Anxiety Direction (2.1 plus or minus 15.9, correlation point zero one five, coaches’ assessment point two four seven), Cognitive Anxiety Intensity (24.9 plus or minus 16.7, correlation point zero one eight, coaches’ assessment negative point two four nine), Cognitive Anxiety Direction (negative 0.8 plus or minus 16.2, correlation point zero four one, coaches’ assessment point zero one one), Self-Confidence Intensity (39.0 plus or minus 17.8, correlation point two one eight, coaches’ assessment point four eight one, double asterisk), Self-Confidence Direction (10.5 plus or minus 15.4, correlation point two six six, challenge and threat index single asterisk, coaches’ assessment point four six zero, double asterisk). Self-report variables include Self-Efficacy (4.7 plus or minus 1.5, correlation point one five six, coaches’ assessment point five zero seven, double asterisk), Cognitive Evaluation (1.2 plus or minus 1.4, correlation negative point zero three five, coaches’ assessment negative point zero one four), Perceived Control (5.7 plus or minus 1.4, correlation point three one two, challenge and threat index single asterisk, coaches’ assessment point two two nine), and Demand Resource (0.2 plus or minus 2.3, correlation point one nine seven, coaches’ assessment point five one nine, double asterisk). Statistical significance is indicated by single asterisk for p less than point zero five and double asterisk for p less than point zero one.
Notes: *p < .05; **p < .01.
4. Discussion
The main findings of this study, investigating differences based on route difficulty relative to the climber were: a) significant and meaningful improvements in climbing performance with increased ability relative to the route; b) greater intensity of somatic and cognitive anxiety and lower self-confidence between CLLabove and CLL and CLLbelow; c) a significant difference in anticipatory HR between CLLabove and CLLbelow, but no significant differences in salivary cortisol; d) cardiovascular reactivity indicative of a threat evaluation in CLLabove and challenge evaluation in CLL and CLLbelow, although variation obscured any significance. The results indicated a predictable difference in performance and success with RRD. Furthermore, the results show increased fall likelihood consequently increases psychophysiological and emotional responses and is speculated to also be partially responsible for performance differences. The results highlight the importance of relative difficulty in psychophysiology research; as well as the challenges faced by intermediate climbers attempting routes at and above their on-sight ability indoors given what appears to be a genuine threshold effect at on-sight level with performance and psychophysiological implications.
To our knowledge, this is the first study to explore climbers’ responses with participants of varying ability attempting a single on-sight ascent of the same indoor route. This was also one the largest climbing studies completed to date. Finally, as with Giles et al. (Reference Giles, Mangan and Draper2025), it was also among the first studies to use detailed performance, psychophysiological and emotional response measures.
Increased RRD increases physiological demands on a climber (Baláš et al., Reference Baláš, Gajdošík, Krupková, Chrastinová, Hlaváčková, Bačáková and Giles2021; Baláš et al., Reference Baláš, Panáčková, Strejcová, Martin, Cochrane, Kaláb, Kodejška and Draper2014; Mermier et al., Reference Mermier, Robergs, McMinn and Heyward1997; Watts & Drobish, Reference Watts and Drobish1998) and decreases the chance of success. Falling potential and associated perceived injury risk is often a potent and intractable climbing stressor (Baláš et al., Reference Baláš, Gajdošík, Krupková, Chrastinová, Hlaváčková, Bačáková and Giles2021; Hurni, Reference Hurni2003; Macleod, Reference Macleod2010). Giles et al. (Reference Giles, Mangan and Draper2025) found lead climbing to be a considerable stressor, with greater anxiety, reduced confidence and lower performance compared to top-roping the same route. These findings support Hardy and Hutchinson (Reference Hardy and Hutchinson2007) who concluded the physical threat accompanying harder routes, especially lead routes, is a major determinant of climbing performance anxiety.
Discussed studies have highlighted route difficulty’s implications on physiological demands (Baláš et al., Reference Baláš, Panáčková, Strejcová, Martin, Cochrane, Kaláb, Kodejška and Draper2014; Janot et al., Reference Janot, Steffen, Porcari and Maher2000; Mermier et al., Reference Mermier, Robergs, McMinn and Heyward1997; Watts & Drobish, Reference Watts and Drobish1998) and psychological responses (Hardy & Hutchinson, Reference Hardy and Hutchinson2007; Hardy & Whitehead, Reference Hardy and Whitehead1984). Additionally, expertise, even with a relatively small ability difference between groups may have played a role. Given the CLLbelow group’s greater experience, cognitive functioning may have played a role (Marczak et al., Reference Marczak, Ginszt, Gawda, Berger and Majcher2018). Decision-making and problem solving are key factors separating climbing ability levels (Medernach & Memmert, Reference Medernach and Memmert2021) and may have impacted this on-sight ascent. Expertise’s effect on route preview may also have been a factor (Sanchez et al., Reference Sanchez, Lambert, Jones and Llewellyn2012; Sanchez et al., Reference Sanchez, Torregrossa, Woodman, Jones and Llewellyn2019). Whitaker et al. (Reference Whitaker, Pointon, Tarampi and Rand2020) reported that as climbing skill increases, accuracy of climbing movement possibility and sequence recall increases with route preview. The following section focuses on the analysis of physiological, psychological, and behavioral data.
4.1. Cardiovascular reactivity to task instructions
Indoor, on-sight lead climbing offers salient task demand: perceived danger and uncertainty of an unknown route. Assessing participants’ cardiovascular responses to task instructions allows objective insight into the participant’s evaluation of the task instructions (Jones et al., Reference Jones, Meijen, McCarthy and Sheffield2009).
The cardiovascular data (Table 8) somewhat support the prediction that increased RRD creates a state where perceived task demands may exceed coping resources (Lazarus, Reference Lazarus1991) with significant differences in total peripheral resistance, but not cardiac output. The weighted challenge and threat index did not reach significance, with CLLabove possessing a negative index, while both CLL and CLLbelow reported a positive score. From the cardiovascular reactivity results, CLLabove may be experiencing a threat state, with smaller-than-challenge HR and cardiac output increases output and no change or a small increase in total peripheral resistance. Conversely, CLL and CLLbelow appear to exhibit a challenge state, with increased, cardiac output and reduced total peripheral resistance, though our cardiac output data was not significant (Behnke & Kaczmarek, Reference Behnke and Kaczmarek2018; Jones et al., Reference Jones, Meijen, McCarthy and Sheffield2009) and with a smaller effect size (η p 2 = .107 vs .278 than seen for different ascent styles in Giles et al. (Reference Giles, Mangan and Draper2025). Together, these results suggest that ascent style is a much more salient factor in challenge/threat states than relative route difficulty.
This study’s results appear in line with those hypothesized and are supported by the differences in the intensity of somatic and cognitive anxiety, self-confidence, anticipatory HR, and assessed performance. There were also significant differences in participants’ appraisal of the task demands and their coping resources. Differences in demand resource evaluation were found between all groups and were larger (η p 2 = .418 vs .301) than seen across ascent styles in Giles et al. (Reference Giles, Mangan and Draper2025). The cardiovascular reactivity results suggest an imbalance between CLLabove’s cognitive appraisal of task demands and their coping resources. There was also a significant positive correlation between CM-PAT score and demand resources evaluation. CLLabove participants likely perceived the task demands as exceeding their coping resources, per self-report and reactivity data.
Challenge states may result in superior performance through more apt emotional responses and interpretation of emotions (Moore et al., 2012; Moore et al., 2013; Turner et al., Reference Turner, Jones, Sheffield, Barker and Coffee2014). While it has been impossible to determine the contribution of performance occurring in this study because of the technical difficulty and the evaluation of the participants, the greater threat evaluation indicates the performance decrements in CLLabove may have, in part, occurred because of their appraisal of route difficulty. Several associated mechanisms have been explored in previous challenge and threat research (Behnke & Kaczmarek, Reference Behnke and Kaczmarek2018; Hase et al., Reference Hase, O’Brien, Moore and Freeman2019). Performance differences may be from more effective attention than a threat state (Hase et al., Reference Hase, O’Brien, Moore and Freeman2019; Jones et al., Reference Jones, Meijen, McCarthy and Sheffield2009). In particular, quiet eye durations (Mann et al., Reference Mann, Williams, Ward and Janelle2007) and muscular tension are possibly greater in a threat state (Blascovich et al., Reference Blascovich, Mendes, Tomaka, Salomon and Seery2003; Moore et al., 2012; Moore et al., 2013). In climbing, performance disruption has been observed in reduced degrees of freedom of movement, increased movement entropy, climbing time (Pijpers et al., Reference Pijpers, Oudejans, Holsheimer and Bakker2003), explorative movements (Pijpers et al., Reference Pijpers, Oudejans and Bakker2005), and eye fixation duration (Nieuwenhuys et al., Reference Nieuwenhuys, Pijpers, Oudejans and Bakker2008). These factors may have contributed to the reduced movement fluency and lower CM-PAT scores seen in this study.
There were no significant group differences in perceived control in this study. Self-efficacy perceptions have been associated with lower fear of failure (Kontos, Reference Kontos2004), climbers taking greater calculated risks, attempting harder climbs and having more confidence in their ability (Llewellyn et al., Reference Llewellyn, Sanchez, Asghar and Jones2008). Self-efficacy is a key aspect of the resource appraisals and contributes significantly to climbers’ perception they can cope with the tasks (Bandura, Reference Bandura1997). This study’s results appear to support this with differences in participants’ self-reported self-efficacy between all three ability groups, in fact these differences were even stronger in terms of effect size than seen (η p 2 = .367 vs .165) with lead vs top rope in Giles et al. (Reference Giles, Mangan and Draper2025). Additionally, there was a significant positive correlation between CM-PAT score and self-efficacy across the three groups (r = .507 corresponds to r 2 = .257, showing a 25.7% associated variance, a greatly interesting effect).
As hypothesized, in line with many other significant differences, CLLabove were found to experience cardiovascular reactivity indicative of a threat appraisal, compared to CLL and CLLbelow, though there was no significant difference in challenge & threat index between groups. Furthermore, the CLLabove climbers’ evaluation of the task demands, relative to their coping resources was also significantly lower than both CLL and CLLbelow. It is possible, as speculated in Giles et al. (Reference Giles, Mangan and Draper2025), that a more threatening state occurred in CLLabove because attempting a route above on-sight ability for an intermediate climber is a more novel task than leading a submaximal route. Further, it may be that the threat state caused some performance differences, although, again, there was no significant challenge & threat difference between groups and it is impossible to directly attribute the changes to psychological factors or simply because of increased route difficulty.
Unlike different ascent styles (Giles et al., Reference Giles, Mangan and Draper2025), RRD brought about demand-resource evaluations and cardiovascular reactivity in line with TCTSA predictions (Behnke & Kaczmarek, Reference Behnke and Kaczmarek2018; Hase et al., Reference Hase, O’Brien, Moore and Freeman2019; Jones et al., Reference Jones, Meijen, McCarthy and Sheffield2009). Specifically, greater RRD resulted in demands outweighing resources and threat cardiovascular reactivity. Demand appraisals likely occurred due to uncertainty about the route (on-sight), and perceived danger (potential lead fall). While TCTSA predictions have been explored with differences in perceived uncertainty and effort, only Giles et al. (Reference Giles, Mangan and Draper2025) has studied situations eliciting differences in perceived danger. Different ascent styles and RRD, both hypothesized to bring about danger demand appraisals, produced conflicting results. Further research is needed to understand the antecedents of demand appraisals in climbing, particularly differences in stressors. However, unlike ascent style, RRD may prove to be a valid means of instigating danger evaluations in climbers; but more research is needed.
4.2. Pre-climb psychophysiology and emotion
Considering this study and previous studies (Baláš et al., Reference Baláš, Giles, Chrastinová, Kárníková, Kodejška, Hlaváčková, Vomáčko and Draper2017; Dickson et al., Reference Dickson, Fryer, Blackwell, Draper and Stoner2012a; Draper et al., Reference Draper, Dickson, Fryer, Blackwell, Winter, Scarrott and Ellis2012; Fryer et al., Reference Fryer2013), it may be said that while cortisol increased in response to climbing tasks, these differences unlikely resulted solely due to physical effort (Hill et al., Reference Hill, Zack, Battaglini, Viru, Viru and Hackney2008; Jacks et al., Reference Jacks, Sowash, Anning, McGloughlin and Andres2002). As such, it is impossible to differentiate between ascent styles (Giles et al., Reference Giles, Mangan and Draper2025), or between the participant’s ability in this study with either plasma or salivary cortisol.
Elevated cortisol has been found when situational factors important to performance, stress or competition have been manipulated (Fernandez-Fernandez et al., Reference Fernandez-Fernandez, Boullosa, Sanz-Rivas, Abreu, Filaire and Mendez-Villanueva2015; Villavicencio et al., Reference Villavicencio, Bravo, Ibarz and Solé2021). However, reduced cortisol reactivity has been found with habituation from repeated stressor exposures (Giles et al., Reference Giles, Fryer, Dickson, Moore and Draper2020; Kirschbaum et al., Reference Kirschbaum, Prussner, Stone, Federenko, Gaab, Lintz, Schommer and Hellhammer1995). Cortisol simply appears too insensitive to differentiate between the responses of experienced climbers completing a habituated task.
Anticipatory HR was significantly elevated in the CLLabove group versus CLLbelow (MD = 21%, 95% CI [9, 32]), representing elevated sympathetic activation, caused by physiological and psychological readiness for the task (Malmo, Reference Malmo1957). Anticipatory HR rise can be beneficial, as seen in experienced climbers, or maladaptive. Heightened sympathetic activation in more advanced climbers may be from improved baroreceptor sensitivity (Sheel, Reference Sheel2004), which may, in turn, elevate anticipatory HR (Fryer, Reference Fryer2013). This study’s results also show relative difficulty appears to play a larger role in anticipatory HR change than ascent style with a much larger effect size seen across ability groupings than across top rope, lead and lead run-out in Giles et al. (Reference Giles, Mangan and Draper2025), however it must be noted that study’s participants had less experience and lower OS ability than this sample. Collectively, these findings suggest higher anticipatory HR before climbing is associated with greater anxiety and poorer performance in less experienced intermediate climbers and, unlike Fryer (Reference Fryer2013), do support greater anticipatory HR increases enhancing psychophysiological preparation in more experienced climbers.
Analysis of participants’ responses showed significant differences in the intensity of participants’ somatic and cognitive anxiety and self-confidence (Table 4). The findings for CLLabove likely occurred due to their perception of the task demands being beyond their ability. Furthermore, unlike intensity, CLL and CLLabove also interpreted their somatic and cognitive anxiety as more debilitative than CLLbelow.
Somatic anxiety intensity was significantly greater for CLLabove than both CLL and CLLbelow. As with Giles et al. (Reference Giles, Mangan and Draper2025), these findings support the idea of a threat of physical harm, or at least situations where there was an increased likelihood of a perceived threat of harm, being a major determinant of performance anxiety when leading (Morris et al., Reference Morris, Davis and Hutchings1981). The much larger effect size for somatic anxiety intensity (ηp2 = .565 vs .136) but very similar effect sizes for cognitive anxiety and self-confidence in this study than Giles et al. (Reference Giles, Mangan and Draper2025) suggests relative difficulty plays a larger role than ascent style for intermediate climbers for somatic anxiety. Participants’ interpretation of the somatic anxiety, as with the intensity, were also significant, however unlike intensity, differences also occurred between CLLabove and CLL groups compared to the CLLbelow, in line with performance. Only CLLbelow interpreted their lower levels of somatic anxiety as facilitative.
There were significant group differences in somatic anxiety, and CM-PAT score also had a significant negative correlation (Table 11; r = –.560 corresponds to r 2 = .314, showing a very relevant associated variance). Such findings contrast with studies that found higher somatic anxiety being associated with improved performance (Draper et al., Reference Draper, Dickson, Fryer and Blackwell2011c; Hardy & Hutchinson, Reference Hardy and Hutchinson2007; Sanchez et al., Reference Sanchez, Boschker and Llewellyn2010) but agree with Vasile et al. (Reference Vasile, Stănescu, Pelin and Bejan2022) that found low somatic anxiety significantly correlated with on-sight ability. Successful climbers in Sanchez et al. (Reference Sanchez, Boschker and Llewellyn2010) reported higher pre-climb somatic anxiety. Similarly, Hardy and Hutchinson (Reference Hardy and Hutchinson2007) found higher somatic anxiety and climber performance ratings during a harder ascent; although, that study involved traditional, not indoor sport climbing. Finally, Draper et al. (Reference Draper, Dickson, Fryer and Blackwell2011c) found successful climbers had greater somatic anxiety than unsuccessful climbers. Differences between these studies and the present study may be due to climber ability. Draper et al. (Reference Draper, Dickson, Fryer and Blackwell2011c) found successful climbers had more climbing and leading experience, and the climbers in Sanchez et al. (Reference Sanchez, Boschker and Llewellyn2010) were of a world cup level. It appears more experienced participants can interpret elevated somatic anxiety as facilitative for performance (Vasile et al., Reference Vasile, Stănescu, Pelin and Bejan2022; Woodman & Hardy, Reference Woodman and Hardy2003). However, for this study’s intermediate climbers, particularly, CLLabove, somatic anxiety occurred alongside elevated cognitive anxiety and reduced performance.
Cognitive anxiety, like somatic, differed significantly between groups. While not significantly different between groups, both the CLLabove and CLL interpreted their cognitive anxiety as debilitative, unlike CLLbelow. These results are like Hardy and Hutchinson (Reference Hardy and Hutchinson2007) who found both somatic and cognitive anxiety were significantly greater when participants attempted an on-sight lead at their limit, versus two grades below. However, unlike Hardy and Hutchinson (Reference Hardy and Hutchinson2007), values were only significant in CLLabove, who attempted a route beyond their on-sight ability. This may have occurred because the aforementioned difference in methodology of a traditional route greater objective danger, with the increased cognitive demands of placing gear (Bisharat, Reference Bisharat2009). Climbing has been shown to be a considerably high load cognitive task. In a dual-task study of climbing and word recall, Green and Helton (Reference Green and Helton2011) reported significant disruption of word recall and climbing performance, despite differences in the nature of both tasks. This is likely due to the high cognitive demands of climbing. Unsurprisingly, elevated levels of cognitive anxiety in this study were associated with reduced climbing performance.
Finally, self-confidence was relatively lower in CLLabove than both CLLbelow and CLL. Self-confidence direction was also significantly greater in CLLbelow than CLLabove and CLL. There was also a significant positive relationship with climbing performance for both the intensity and direction of self-confidence. Significantly lower self-confidence prior to lead climbs has been reported by several authors versus top rope (Aras & Akalan, Reference Aras and Akalan2011; Dickson et al., Reference Dickson, Fryer, Blackwell, Draper and Stoner2012a; Fryer, Reference Fryer2013). Draper et al. (Reference Draper, Dickson, Fryer and Blackwell2011c) found significantly greater self-confidence in successful versus unsuccessful climbers. Equally, Sanchez et al. (Reference Sanchez, Boschker and Llewellyn2010) found successful climbers had a small, but insignificantly greater self-confidence and showed an individual may experience both positive and negative emotions in response to a stressor, but more successful athletes are able to maintain a more positive affective state, in line with findings in other sports (Jones et al., Reference Jones, Swain and Hardy1993; Lox, Reference Lox1996). Such findings are also typical of other sports: a meta-analysis of anxiety and self-confidence found athletes achieved greater performances when both anxious and self-confident (Woodman & Hardy, Reference Woodman and Hardy2003). Increased relative self-confidence before climbing may improve technique and tactics (Draper et al., Reference Draper, Dickson, Fryer and Blackwell2011c); consequently, such differences maybe responsible for the performance of CLL and CLLbelow and conversely the lower performance of CLLabove (Draper et al., Reference Draper, Dickson, Fryer and Blackwell2011c).
4.3. Climbing task demands and performance
This study’s success rates across ability groups appears to reinforce the validity of self-reported climbing grades as suggested by Draper et al. (Reference Draper, Dickson, Blackwell, Fryer, Priestley, Winter and Ellis2011a) and Draper et al. (Reference Draper, Giles, SchöFfl, Fuss, Watts, Wolf, BaláŠ, EspañA Romero, Gonzalez, Fryer, Fanchini, Vigouroux, Seifert, Donath, Spoerri, Bonetti, Phillips, Stöcker, Bourassa-Moreau and Abreu2016).
There were no significant differences in mean or peak climbing HR between groups, or an interaction between conditions, or success. There was a significant difference between groups for climbing time, though RRD appears to play a much smaller role (η p 2 = .113 vs .672) than ascent style compared to Giles et al. (Reference Giles, Mangan and Draper2025). Unsuccessful participants had significantly greater peak HR than successful climbers. This study’s observed lead climbing values are comparable to previous studies (Aras & Akalan, Reference Aras and Akalan2011; Draper et al., Reference Draper, Jones, Fryer, Hodgson and Blackwell2010; Fryer et al., Reference Fryer2013). The results are also similar to Mermier et al. (Reference Mermier, Robergs, McMinn and Heyward1997), Watts and Drobish (Reference Watts and Drobish1998) and Janot et al. (Reference Janot, Steffen, Porcari and Maher2000) with increased difficulty eliciting greater HR, but, not always significantly. However, as with Giles et al. (Reference Giles, Mangan and Draper2025), while this study’s results support HR as a physiological load indicator, they do not support mean or peak HR as a psychological challenge indicator, as the results are obscured by other factors.
Performance was assessed with GE and CM-PAT score (Table 10) and were found to differ significantly between groups. Greater GE has been shown to also occur with increased anxiety with intermediate climbers in Giles et al. (Reference Giles, Mangan and Draper2025) and non-climbers participating in a climbing task in Pijpers et al. (Reference Pijpers, Oudejans, Holsheimer and Bakker2003). The effect of ascent style seen in Giles et al. (Reference Giles, Mangan and Draper2025) for GE was much larger than seen for RRD in this study (η p 2 = .540 vs .147). However the total GM-PAT effect size between the ascent style of Giles et al. (Reference Giles, Mangan and Draper2025) and RRD of this study were quite similar and both large (η p 2 = .513 vs .443), with the subscales base of support (η p 2 = .156 vs .408) and transitioning (η p 2 = .217 vs .505) appearing to be more affected by RRD, technique (η p 2 = .453 vs .436) affected similarly by both and tactics (η p 2 = .667 vs .429) and coordination (η p 2 = .476 vs .376) being more affected by ascent style. It is impossible to determine the exact cause of the deterioration in performance seen, this could be due to the task’s physical demands, anxious disruption of performance (Nieuwenhuys & Oudejans, Reference Nieuwenhuys and Oudejans2012) and/or a threshold effect with CLLabove climbers’ forearm flexors more likely to be above a maximal metabolic steady state/muscle oxygen breakpoint/critical force of their forearm flexors (Baláš et al., Reference Baláš, Gajdošík, Giles and Fryer2022; Baláš et al., Reference Baláš, Gajdošík, Javorský, Berta and Feldmann2024; Giles et al., Reference Giles, Chidley, Taylor, Torr, Hadley, Randall and Fryer2019). As the physical demands assessed with HR across all three groups were similar, the climbing performance of the present study was possibly limited to a greater extent by technical performance, than physiological. This may be characteristic of the formative nature of intermediate climbers’ performance and may not be true of more experienced climbers.
To summarize, performance was significantly affected by attempting routes of differing RRD. While there were no significantly different physical responses, there were meaningful differences in GE and CM-PAT score between groups. Conceivably, differences in climbing performance occurred due to physical climbing ability and ability to manage anxiety with a potential fall. Not only may a potential fall induce anxiety, but that anxiety may also limit a climber’s ability to complete a route (Hague & Hunter, Reference Hague and Hunter2011). To explore this, objective physiological and subjective psychological measures were made in response to the task instructions, prior to the climber’s attempt at the route itself and in response to the climb. Differences in these factors will be discussed in the following sections.
4.4. Climbing anxiety
Given the equivocal results of previous psychophysiological ascent style studies (Draper et al., Reference Draper, Dickson, Fryer, Blackwell, Winter, Scarrott and Ellis2012; Fryer et al., Reference Fryer2013; Hardy & Hutchinson, Reference Hardy and Hutchinson2007), the significance observed in Giles et al. (Reference Giles, Mangan and Draper2025), particularly the run-out condition, combined with this study’s responses of CLLabove participants, speculation about sources of anxiety in climbing is possible. Habituated situations for climbers, such as a submaximal lead route, may not be anxiety-inducing. However, run-out routes with greater consequences of a fall (Giles et al., Reference Giles, Mangan and Draper2025), or routes beyond on-sight ability, elicit a greater response. The results of non-climbers versus experienced climbers in Giles et al. (Reference Giles, Fryer, Dickson, Moore and Draper2020) also support this, with significant differences in anxiety, HR and salivary cortisol in response to a ladder climb. It is suggested that these situations induce anxiety, as they are unusual and atypical of their usual recreational sessions.
Situations decreasing the actual or perceived chance of climbing success appear to induce anxiety. Failure can be associated with a threat of physical harm (Backe et al., Reference Backe, Ericson, Janson and Timpka2009), or a threat to self-efficacy from a climber’s perceived evaluation by others (Bandura, Reference Bandura1997). A fear of failure is associated with anticipation of shame in evaluative situations and a tendency to appraise situations as threatening (Sagar & Lavallee, Reference Sagar and Lavallee2010). Climbing success is largely dictated by reaching the top of a route, and is often described as self-esteem and self-confidence enhancing (Stiehl & Ramsey, Reference Stiehl and Ramsey2005), but failure can negatively impact esteem and confidence (Sagar & Lavallee, Reference Sagar and Lavallee2010; Stiehl & Ramsey, Reference Stiehl and Ramsey2005). Based on Giles et al. (Reference Giles, Mangan and Draper2025) and this study it is possible to describe responses to the climbs investigated, but not an explanation why certain factors disrupt performance.
To summarize climbers’ responses respecting pre-climb psychophysiology and emotion, the ability for salivary cortisol concentrations to differentiate between participants’ ability were limited. While cortisol concentrations increased for all participants, in line with previous research (Dickson et al., Reference Dickson, Fryer, Blackwell, Draper and Stoner2012a; Draper et al., Reference Draper, Dickson, Fryer, Blackwell, Winter, Scarrott and Ellis2012; Fryer, Reference Fryer2013; Fryer et al., Reference Fryer2013) the values obtained did not differentiate the psychophysiological responses of climbers based on RRD and ability. Conversely, anticipatory HR changes did increase from rest by a significantly greater amount between CLLabove and CLLbelow. Clearer than the psychophysiological changes, were the significant and meaningful differences in cognitive, somatic anxiety and self-confidence between CLLabove and both CLL and CLLbelow. Interestingly, while differences in the intensity of emotions were not significant between CLL and CLLbelow, the interpretation was, supporting task differences in the positive affective state before climbing (Jones et al., Reference Jones, Swain and Hardy1993; Lox, Reference Lox1996). Differences in the psychological and physiological responses to climbing tasks may occur because of situations unusual to the participant, including those resulting in physical harm, or ego-threatening; but not those typical of intermediate climber’s recreational sessions.
4.5. Limitations
This study must be considered in light of its limitations. The setting of a public climbing wall may have affected participants’ responses. Participants who did not ascend the entire route had a shorter climbing time and distance which may have influenced GE, behavioral and psychophysiological outcomes. The female participants’ menstrual cycle may have affected cortisol levels in response to stressful tasks (Montero-López et al., Reference Montero-López, Santos-Ruiz, García-Ríos, Rodríguez-Blázquez, Rogers and Peralta-Ramírez2018). Findings are only representative of the individual indoor, on-sight route used in the study and of intermediate/advanced climbers. The five-minute period for route preview may have adjusted perceptions of challenge/threat based on a participant’s route reading efficacy.
The CLLbelow group had much more climbing experience both in terms of years of experience and current climbing frequency, which could be a meaningful contextual factor that may have affected relevant variables including self-confidence/anxiety, climbing movement proficiency and threat appraisal/coping resources. No previous climbing literature has considered experience separately from ability in examining climbers of different ability levels while lead climbing. While experienced climbers are not immune to sensing emotional, self-esteem and social risk associated with failing on a route (West & Allin, Reference West and Allin2010), more experienced climbers are typically more task-oriented and goal-oriented (Gonzalez, Reference Gonzalez2019; Kiewa, Reference Kiewa2001; Sarrazin et al., Reference Sarrazin, Roberts, Cury, Biddle and Famose2002), which could make them more interested in self-actualization (Mangan et al., Reference Mangan, Andrews, Miles and Draper2025) and worrying less about the evaluations of others which would make them less prone to cognitive anxiety (Morris et al., Reference Morris, Davis and Hutchings1981; Morris & Liebert, Reference Morris and Liebert1973). Additionally, some studies have found increased climbing and safety self-efficacy which are correlated with experience, increased comfort with climbing risk and riskier styles of climbing (Jakus & Shaw, Reference Jakus and Shaw1996; Llewellyn & Sanchez, 2008; Llewellyn et al., Reference Llewellyn, Sanchez, Asghar and Jones2008; Martha et al., Reference Martha, Sanchez and Gomà-i-Freixanet2009; Raue et al., Reference Raue, Kolodziej, Lermer and Streicher2018) and could have led to an increased comfort with lead climbing among the more experienced CLLbelow group in this study.
4.6. Suggestions for future research
The findings support that performance is limited both psychologically and physiologically. As with ascent style changes in Giles et al. (Reference Giles, Mangan and Draper2025), performance decrements were associated with greater cognitive anxiety. For climbers and coaches two factors are apparent; first, there would be a benefit to exploring sources of cognitive anxiety in intermediate climbers, in order to understand the antecedents of their performance; second, interventions addressing anxiety resulting from leading (Garrido-Palomino & España-Romero, Reference Garrido-Palomino and España-Romero2023) and leading beyond the on-sight ability of climbers should concentrate on reducing cognitive anxiety and improving self-confidence to improve enjoyment and performance.
Future climbing psychophysiology studies should consider chronic/baseline stress, trait anxiety and personality which may affect cognitive appraisals of challenge and threat. Additionally, climbing experience should be considered as a covariate to avoid conflating ability and experience (Mangan et al., Reference Mangan, Andrews, Miles and Draper2025).
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/SJP.2026.10036.
Data sharing
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Authorship credit
DG: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Software, Project administration. KM: Project administration, Writing – review & editing. ND: Conceptualization, Supervision, Project administration, Resources, Writing – review & editing.
Funding statement
No funding was required for the study.
Competing interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Open access
