Pacing Behavior Research Articles

The influence of facility and home pen design on the welfare of the laboratory-housed dog

We have an ethical and scientific obligation to Refine all aspects of the life of the laboratory-housed dog. Across industry there are many differences amongst facilities, home pen design and husbandry, as well as differences in features of the dogs such as strain, sex and scientific protocols. Understanding how these influence welfare, and hence scientific output is therefore critical. A significant proportion of dogs' lives are spent in the home pen and as such, the design can have a considerable impact on welfare. Although best practice guidelines exist, there is a paucity of empirical evidence to support the recommended Refinements and uptake varies across industry. In this study, we examine the effect of modern and traditional home pen design, overall facility design, husbandry, history of regulated procedures, strain and sex on welfare-indicating behaviours and mechanical pressure threshold. Six groups of dogs from two facilities (total n=46) were observed in the home pen and tested for mechanical pressure threshold. Dogs which were housed in a purpose-built modern facility or in a modern design home pen showed the fewest behavioural indicators of negative welfare (such as alert or pacing behaviours) and more indicators of positive welfare (such as resting) compared to those in a traditional home pen design or traditional facility. Welfare indicating behaviours did not vary consistently with strain, but male dogs showed more negative welfare indicating behaviours and had greater variation in these behaviours than females. Our findings showed more positive welfare indicating behaviours in dogs with higher mechanical pressure thresholds. We conclude that factors relating to the design of home pens and implementation of Refinements at the facility level have a significant positive impact on the welfare of laboratory-housed dogs, with a potential concomitant impact on scientific endpoints.

Is noise reactivity reflected in auditory response variables, including those that measure cognition, in dogs? Initial findings

Noise reactivity and noise phobia are anxiety- and panic-related conditions in dogs that may affect up to 50% of dogs across their lifetime. Affected dogs show a range of signs of distress including trembling, freezing, panting, social withdrawal, pacing, salivating, and escape behaviors. Noise reactivity and phobia have been shown to be comorbid conditions, and their presence may increase the risk and severity of other anxiety-related conditions. Anxiety disorders may interfere with dogs' abilities to perform problem-solving tasks or to interpret information that could be useful in such tasks, including tasks involving or affected by noise. The extent to which noise reactivity or phobia is related to auditory dysfunction or impairment is not known. In this study, we asked whether known noise reactivity in dogs was reflected in any of several measures of auditory function: the brainstem auditory-evoked response, auditory middle-latency response, and mismatch negativity. Most noise-reactive dogs in this study were mildly affected (mean anxiety intensity rank [noise] score 17.65, maximum = 64). Comparison of the major auditory measures of dogs who were noise reactive with those of dogs who were non–noise reactive revealed a significant difference in only one variable of the brainstem auditory-evoked response test, right ear wave-V (RE-V) (Welch's t = 2.55, df = 22.41, P < 0.02). Auditory middle- and late-latency responses were present in all dogs that allowed for the completion of this test, providing initial evidence of higher order auditory-cognitive function. Behaviorally, the group of noise-reactive dogs was significantly different from the group of non–noise-reactive dogs with respect to their ability to undergo this testing: 5 of the 17 noise-reactive dogs were too reactive to undergo or complete the test but none of the 14 non-noise dogs were unable to undergo and complete testing. This study suggests that the underlying pathology resulting in noise reactivity may not influence auditory middle-latency response or mismatch negativity related variables, but the study should be expanded to a larger and more severely affected population of dogs.

Pacing Behavior and Tactical Positioning in 500- and 1000-m Short-Track Speed Skating.

To explore pacing behavior and tactical positioning during the shorter 500- and 1000-m short-track competitions. Lap times and intermediate rankings of elite 500- and 1000-m short-track-skating competitors were collected over the 2012-13 season. First, lap times were analyzed using a MANOVA, and for each lap, differences between sex, race type, final ranking, and stage of competition were determined. Second, Kendall tau-b correlations were used to assess relationships between intermediate and final rankings. In addition, intermediate rankings of the winner of each race were examined. Top-placed athletes appeared faster than bottom-placed athletes in every lap in the 500-m, while in the 1000-m no differences were found until the final 4 laps (P < .05). Correlations between intermediate and final rankings were already high at the beginning stages of the 50-m (lap 1: r = .59) but not for the 1000-m (lap 1: r = .21). Although 500- and 1000-m short-track races are both relatively short, fundamental differences in pacing behavior and tactical positioning were found. A fast-start strategy seems to be optimal for 500-m races, while the crucial segment in 1000-m races seems to be from the 6th lap to the finish line (ie, after ± 650 m). These findings provide evidence to suggest that athletes balance between choosing an energetically optimal profile and the tactical and positional benefits that play a role when riding against an opponent, as well as contributing to developing novel insights in exploring athletic behavior when racing against opponents.

Working on something else for a while: Pacing in creative design projects

In an interview study among 25 architects we investigated how activities were allocated over time in the design phase of an architectural project. Specifically, linking the literatures about pacing behavior and incubation, we set out to identify patterns related to how the interviewees paced their activity before a deadline. We used two types of materials to stimulate the answers in the interview: 1) standard graphs, developed in previous research; 2) a timeline. Five main themes emerged, that were termed pacing; milestones; multiple projects; deadlines; and quality–time trade-offs. Based upon the results we propose a model that includes overlapping U-shapes of activity, that is, most activity at the start of a project and right before a deadline. A second study provided a comparison of the pacing patterns among 85 respondents in jobs that required different levels of creativity. Not only U-shape pacing, but also deadline pacing was more common in creative jobs, whereas early starting patterns were less common. The two studies provide insight into how professionals in creative jobs deliberately use pacing that allows for incubation.

Brief time-based activity pacing instruction as a singular behavioral intervention was not effective in participants with symptomatic osteoarthritis.

Osteoarthritis (OA) of the lower extremities is a prevalent cause of disability in which symptoms interfere with mobility and activity participation. Behavioral self-management for OA symptomatology is commonly recommended; but these interventions are underutilized, unstandardized in application, and at times, unavailable in the context of clinical care. For people with chronic pain, rehabilitation professionals may select to apply activity pacing instruction as one behavioral strategy to manage symptoms. Activity pacing is widely used in combination with other pharmacological and behavioral interventions but has not been studied as a singular behavioral intervention for people with OA. The purpose of this study was to evaluate the effectiveness of an occupational therapist-delivered, time-based activity pacing program for treatment of pain, fatigue, and physical function in people with symptomatic knee or hip OA. A 3-arm randomized controlled trial was conducted in which 193 people were randomized into tailored activity pacing, general activity pacing, or usual care arms. Assessments were done at 10 weeks and 6 months after baseline. Using linear mixed models, Western Ontario and McMaster Universities Osteoarthritis Index pain scores changed over time, decreasing the most in the general and usual care groups; only the usual care group had decreased pain over 6 months. The tailored and general activity pacing groups reported higher frequency of pacing behaviors than the usual care group at 10 weeks, but pacing was not sustained at 6 months. This trial does not support the use of time-based pacing as a singular behavioral strategy for people with knee or hip OA.

Development and Initial Validation of the Activity Patterns Scale inPatients With Chronic Pain.

Several self-report measures were used to identify 6 activity patterns in chronic pain patients: pain avoidance, activity avoidance, task-contingent persistence, excessive persistence, pain-contingent persistence, and pacing. Instruments for assessing pacing should include 3 pacing behaviors (breaking tasks into smaller tasks, taking frequent short rests, slowing down), each of which relate to a single goal (increasing activity levels, conserving energy for valued activities, and reducing pain). This article presents the Activity Patterns Scale (APS), which assesses these 6 activity patterns. Study 1 included 291 participants with chronic pain, and tested 3 structures using confirmatory factor analyses. The structure with the best fit had 8 factors corresponding to the hypothesized scales. High correlations in the expected direction were found between the APS subscales and the "Patterns of Activity Measure-Pain." Study 2 included 111 patients with chronic pain, and aimed at examining the association between the APS subscales and adjustment to pain. It was found that that activity avoidance was associated with daily functioning and impairment. Negative affect was positively associated with activity avoidance and excessive persistence, and negatively associated with task-contingent persistence, which was also positively associated with positive affect. This study showed that the APS is a valid and reliable instrument for clinical practice and research. This article presents a valid and reliable instrument to assess activity patterns in patients with chronic pain. The findings suggest that avoidance, persistence, and pacing are multidimensional constructs. Distinguishing between these dimensions sheds light on previous contradictory results and has direct clinical implications regarding recommending the most advisable activity patterns.

The Influence of Collective Behavior on Pacing in Endurance Competitions.

A number of theoretical models have been proposed in recent years to explain pacing strategies observed in individual competitive endurance events. These have typically related to the internal regulatory processes that inform the making of decisions relating to muscular work rate. Despite a substantial body of research which has investigated the influence of collective group dynamics on individual behaviors in various animal species, this issue has not been comprehensively studied in individual athletic events. This is somewhat surprising given that athletes often directly compete in close proximity to one another, and that collective behavior has also been observed in other human environments including pedestrian interactions and financial market trading. Whilst the reasons for adopting collective behavior are not fully understood, collective behavior is thought to result from individual agents following simple local rules that result in seemingly complex large systems that act to confer some biological advantage to the collective as a whole. Although such collective behaviors may generally be beneficial, competitive endurance events are complicated by the fact that increasing levels of physiological disruption as activity progresses may compromise the ability of some individuals to continue to interact with other group members. This could result in early fatigue and relative underperformance due to suboptimal utilization of physiological resources by some athletes. Alternatively, engagement with a collective behavior may benefit all due to a reduction in the complexity of decisions to be made and a subsequent reduction in cognitive loading and mental fatigue. This paper seeks evidence for collective behavior in previously published analyses of pacing behavior and proposes mechanisms through which it could potentially be either beneficial, or detrimental to individual performance. It concludes with suggestions for future research to enhance understanding of this phenomenon.

Longitudinal mediation in the PACE randomised clinical trial of rehabilitative treatments for chronic fatigue syndrome: modelling and design considerations

Background Clinical trials require large monetary and time commitments and should provide information on both whether and how treatments work. Treatment mechanisms can be studied using mediation analysis, allowing refinement of treatments. Mediation studies often use only single contemporaneous measures of mediator and outcome limiting the conclusions that can be drawn. Longitudinally measured mediators and outcomes, such as those in the Pacing, Graded Activity, and Cognitive Behaviour Therapy: A Randomised Evaluation trial (PACE, ISRCTN 54285094) allow for more realistic estimates of mediated effects.

Social and environmental influences on pacing in a female Malayan sun bear (Helarctos malayanus)

Pacing behaviour is complex and identifying the motivational basis for pacing and designing an effective remedial strategy can be challenging. Details of the behaviour may provide insight into the motivational basis of behaviour and should be carefully examined. A long-term observational study of pacing by a female Malayan sun bear ( Helarctos malayanus ) was conducted to gather detailed information on the pacing behaviour and guide husbandry changes. Full-day behaviour observations were conducted (n=60; mean duration=5.9 hrs/observation). Pacing data were recorded using continuous sampling methodology with all-occurrences recording of events before and after pacing bouts. Solitary behaviour and social interaction between the focal female and a male and female exhibit partner were recorded using instantaneous sampling and all-occurrences methodologies, respectively. All pacing bouts occurred between tunnels to off-exhibit indoor holding areas, bouts were short in duration and variable, and pacing often began after the male bear entered a shift tunnel. Social interactions between the two females were frequently agonistic. Based on these findings, the non-focal female bear was removed from the group and the focal female and male were given access to indoor holding areas on most days. The focal female only paced once when provided access to indoor holding. When the bears did not have off-exhibit access, pacing characteristics appeared similar to bouts before modifications but pacing no longer appear to be influenced by the male’s behaviour. Our detailed analysis of pacing by the female sun bear provided insight regarding motivation, the need for husbandry changes, and evaluation of those changes.

Nocturnal behavior in captive giraffe (Giraffa camelopardalis)--A pilot study.

Captive giraffes (Giraffa camelopardalis) are known to perform oral and locomotor stereotypies. However, many studies do not consider the behavioral repertoire of these animals during the time when animals are confined to night quarters. At two zoological institutions, a total of six captive giraffes were observed via camera trap technology throughout six diurnal and nocturnal periods to record feeding, ruminating, and stereotypic behaviors. The effect of browse enrichment was assessed on alternate nights to determine how behaviors may be altered in the presence of natural forage. Results need to be interpreted with caution due to a high proportion of time when animals were out of camera range. For the observed time, stereotypical licking behavior was significantly higher at night compared to daytime at both facilities, while tongue play increased at the same time, but not significantly. The provision of browse enrichment during the night decreased the rate of tongue playing, but not significantly; however, browse did significantly reduce pacing behavior. Across treatments and institutions, observed oral stereotypies tended to correlate negatively with increased feeding behavior. Apart from a short-term effect of enrichment, this study indicates relevant differences in the frequencies of behaviors observed during the day and night, suggesting that assessing nocturnal behavior specifically may be important in many species.

Pacing Strategy, Muscle Fatigue, and Technique in 1500-m Speed-Skating and Cycling Time Trials.

To evaluate pacing behavior and peripheral and central contributions to muscle fatigue in 1500-m speed-skating and cycling time trials when a faster or slower start is instructed. Nine speed skaters and 9 cyclists, all competing at regional or national level, performed two 1500-m time trials in their sport. Athletes were instructed to start faster than usual in 1 trial and slower in the other. Mean velocity was measured per 100 m. Blood lactate concentrations were measured. Maximal voluntary contraction (MVC), voluntary activation (VA), and potentiated twitch (PT) of the quadriceps muscles were measured to estimate central and peripheral contributions to muscle fatigue. In speed skating, knee, hip, and trunk angles were measured to evaluate technique. Cyclists showed a more explosive start than speed skaters in the fast-start time trial (cyclists performed first 300 m in 24.70 ± 1.73 s, speed skaters in 26.18 ± 0.79 s). Both trials resulted in reduced MVC (12.0% ± 14.5%), VA (2.4% ± 5.0%), and PT (25.4% ± 15.2%). Blood lactate concentrations after the time trial and the decrease in PT were greater in the fast-start than in the slow-start trial. Speed skaters showed higher trunk angles in the fast-start than in the slow-start trial, while knee angles remained similar. Despite similar instructions, behavioral adaptations in pacing differed between the 2 sports, resulting in equal central and peripheral contributions to muscle fatigue in both sports. This provides evidence for the importance of neurophysiological aspects in the regulation of pacing. It also stresses the notion that optimal pacing needs to be studied sport specifically, and coaches should be aware of this.

Pacing Behavior and Tactical Positioning in 1500-m Short-Track Speed Skating.

To gain more insight in pacing behavior and tactical positioning in 1500-m short-track speed skating, a sport in which several athletes directly compete in the same race. Lap times and intermediate rankings of elite 1500-m short-track- skating competitors were collected over the season 2012-13 (N = 510, 85 races). Two statistical approaches were used to assess pacing behavior and tactical positioning. First, lap times were analyzed using a MANOVA, and for each lap differences between sex, race type, final rankings, and stage of competition were determined. Second, Kendall tau b correlations were used to assess relationships between intermediate and final rankings. In addition, intermediate rankings of the winner of each race were examined. In 1500 m (13.5 laps of 111.12 m), correlations between intermediate and final ranking gradually increased throughout the race (eg, lap 1, r = .05; lap 7, r = .26; lap 13, r = .85). Moreover, the percentage of race winners skating in the leading position was over 50% during the last 3 laps. Top finishers were faster than bottom-place finishers only during the last 5 laps, with on average 0.1- to 1.5-s faster lap times of the race winners compared with the others during the last 5 laps. Although a fast start led to faster finishing times, top finishers were faster than bottom-placed finishers only during the last 5 laps. Moreover, tactical positioning at 1 of the foremost positions during the latter phase of the race appeared to be a strong determinant of finishing position.

Defining pacing quantitatively: A comparison of gait characteristics between pacing and non-repetitive locomotion in zoo-housed polar bears

Pacing is a commonly investigated behavior in zoo animals, however, little data have been published to support a precise definition of pacing. High speed video of pacing behavior in 11 zoo-housed polar bears (Ursus maritimus) was compared with goal-oriented locomotion in the same bears. Specifically, step cycle duration and head height ratio were measured, as was the variability of these two dimensions. The median step cycle duration while pacing was found to be 17.0% (±26.7%) shorter than step cycle duration while locomoting (P<0.01). Step cycle duration while pacing also displayed a significantly lower coefficient of variation than step cycle duration while locomoting (9.4±3.5% versus 22.7±10.0%; P<0.01). Median head height was found to be 15.3% (±10.5%) higher while pacing than while locomoting (P<0.05) and was also found to have a significantly reduced coefficient of variation as compared to locomoting (5.6±2.6% versus 16.6±8.7%; P<0.05). The results of this study confirm that pacing is quantifiably different than non-repetitive locomotion and may reflect a state in which animals are disengaged with their environments.

Synchrony Preference: Why Some People Go With the Flow and Some Don't

As work in organizations becomes more fluid and fast paced the effective execution of tasks often requires people to be temporally adaptable in working with others. This paper brings to light the importance of understanding how people relate to time in the context of social interactions. By integrating the research on time and social motives, we develop a relational perspective about time. We introduce the construct of synchrony preference, an individual difference that describes a willingness to adapt one's pace and rhythm within social interactions for the purpose of creating synchrony with others. We develop a measure of synchrony preference that we validate using multiple methods across 4 studies and 7 samples, which include over 1,400 individuals. In particular, we establish a nomological network and show that synchrony preference predicts flexible pacing behaviors, interpersonal facilitation, contribution to team synchrony, contribution to team performance, and job dedication. Our results reveal that both scholars and practitioners can benefit from considering people's preference for synchrony when working with others.

Activity pacing in daily life: A within-day analysis

Activity pacing is a widely used self-management strategy, but we lack a clear understanding of its nature and usefulness. One source of confusion is a lack of clarity about the use of pacing in everyday life (ie, naturalistic pacing) in people not trained on how to pace activities. It is unknown whether people engage in more pacing when pain is high (pain-contingent) or when fatigue is high (fatigue-contingent). Conversely, it is not known whether naturalistic pacing results in reduced symptoms. We conducted a multilevel daily process study in which participants with osteoarthritis (N=162) reported pain and fatigue severity and frequency of pacing behaviors 5times per day over 5days. We hypothesized that increased pain and fatigue would predict increased pacing and that pacing would have a short-term benefit in terms of decreased pain and fatigue. Multilevel modeling results showed that, as expected, both momentary pain and fatigue were positively associated with subsequent pacing behaviors. Contrary to our hypothesis, increased pacing was associated with higher subsequent levels of pain and fatigue. Naturalistic pacing seems symptom-contingent and not reinforced by symptom reduction. Naturalistic pacing may be distinct from trained or programmatic pacing in terms of outcomes, and further research into naturalistic pacing may provide an important foundation for how best to deliver activity pacing interventions.

A Content Analysis of Activity Pacing in Chronic Pain

Activity pacing is a common intervention for patients with chronic pain. Over the past decade a number of instruments have been developed to measure this construct, but their comparative psychometric properties have not been examined. To review the psychometric properties of existing measures of activity pacing, and provide suggestions for future research in this emerging area of pain research. A narrative review of current measures of activity pacing followed by a discussion of the conceptual and psychometric challenges in this area. Although there is evidence supporting activity pacing as a unitary construct, important differences remain among the various measures in terms of their item content and assumptions. All existing activity pacing measures include items that assess activity regulation, but vary in their specific content. Most importantly, questionnaire items often reflect different purposes of pacing behaviors. Current measures of activity pacing are inadequate. New measures are needed that are based on specific theoretical models; these measures should also make the goal or intent of pacing behaviors explicit. Improvements in the assessment of activity pacing will likely lead to a better understanding of the pacing construct and the effects of pacing interventions.

Study on activity pattern and incidence of stereotypic behavior in captive tigers

The present study examined daytime behavior patterns of 19 captive tigers (Panthera tigris tigris) housed in Nandankanan Zoological Park, Odisha, India. Behavioral observations of 1254 hours were analyzed for target behaviors using instantaneous sampling and 1-minute sample periods. We found that these captive tigers spent about 23% of the daytime exhibiting stereotypic behavior, that is, pacing, with a biphasic peak at 10:00 to 11:00 am and 16:00 to 17:00 pm. The incidence of stereotypic pacing behavior appears high, warranting further investigation of the cause of this stereotypic behavior and whether an effect of behavioral enrichment is required for a better understanding of welfare implications on the animals.

Clitoral anesthesia disrupts paced copulation in the female rat

Clitoral stimulation produced by sexual contact with a partner or during manual stimulation is associated with pleasure in humans, and produces conditioned place preference in rats. The present experiment investigated the effect of blocking genitosensory stimulation of the clitoris with lidocaine during copulation in female rats on a measure of female sexual motivation: pacing behavior. Sexually naïve, ovariectomized female rats were treated with 10μg estradiol benzoate 48h and 500μg progesterone 4h prior to a 30-min copulatory trial with a sexually vigorous stimulus male scheduled every 4days. A total of 10 copulatory sessions were divided into two phases of 5 trails each. In the first phase, females received an injection (0.05ml) of either 2% lidocaine, saline, or no injection to the clitoral sheath under isoflurane anesthesia immediately prior to the start of a copulatory session, and were then placed on one side of a paced mating chamber and allowed to copulate for 30min. In the second phase, females previously injected with lidocaine were switched to saline and vice versa, and the no injection group remained the same. Variables measured included overall time spent with the males, number of solicitations, contact–return latencies following male mounts, intromissions, and ejaculations; the frequency of entrances and exits from the male chamber, and frequency of mounts, intromissions, ejaculations. Sexual behavior was examined at session 1, session 5, and session 10. At test 5, females that received LID had a greater number of entrances/exits but spent significantly less time in the presence of the male during the copulatory bout than CNTL animals. These females also displayed a trend for longer contact return latencies s after ejaculations than VEH and CNTL groups. On session 10, females that received LID and subsequently switched to VEH treatment no longer differed from controls in entrance/exit numbers, time spent with males or ejaculation contact return latency. They did however, receive a greater number of intromissions and displayed shorter inter intromission intervals compared to CNTLs. We suggest that clitoral stimulation in the rat serves as both a reward signal and may contribute to the detection of differences in copulatory stimuli that are critical to pacing and potentially, the initiation of pregnancy.

Physiological and Psychological Effects of Deception on Pacing Strategy and Performance: A Review

The aim of an optimal pacing strategy during exercise is to enhance performance whilst ensuring physiological limits are not surpassed, which has been shown to result in a metabolic reserve at the end of the exercise. There has been debate surrounding the theoretical models that have been proposed to explain how pace is regulated, with more recent research investigating a central control of exercise regulation. Deception has recently emerged as a common, practical approach to manipulate key variables during exercise. There are a number of ways in which deception interventions have been designed, each intending to gain particular insights into pacing behaviour and performance. Deception methodologies can be conceptualised according to a number of dimensions such as deception timing (prior to or during exercise), presentation frequency (blind, discontinuous or continuous) and type of deception (performance, biofeedback or environmental feedback). However, research evidence on the effects of deception has been perplexing and the use of complex designs and varied methodologies makes it difficult to draw any definitive conclusions about how pacing strategy and performance are affected by deception. This review examines existing research in the area of deception and pacing strategies, and provides a critical appraisal of the different methodological approaches used to date. It is hoped that this analysis will inform the direction and methodology of future investigations in this area by addressing the mechanisms through which deception impacts upon performance and by elucidating the potential application of deception techniques in training and competitive settings.

Individual and environmental factors associated with stereotypic behavior and fecal glucocorticoid metabolite levels in zoo housed polar bears

Polar bears (Ursus maritimus) are known to exhibit repetitive pacing behaviors, usually described as stereotypic, in zoo environments. However, little quantitative information exists about the prevalence of pacing in the zoo population. Similarly, large, multi-institutional studies conducted to determine the relationship between stereotypic behavior in zoo polar bears and environmental/husbandry variables using corticoids as a measure of stress are lacking. The study reported here includes data from 55 bears housed in 20 North American zoos. Individual and zoo characteristics were collected and behavior and fecal glucocorticoid metabolites (FGM) were measured over a one-year period. Using an epidemiological approach, individual and facility level multiple linear regression models were constructed to determine the nature, strength and significance of environmental/husbandry variables and temperament (measured using a standardized novel object behavior test) on stereotypic pacing and FGM. We found zoo polar bears performed stereotypic pacing behavior during 14% of the day; the proportion rose to 22% when expressed as a percentage of time engaged in locomotory behavior. However, considerable variation in proportion of stereotypy was observed. Variables associated with reduced pacing at zoos were: enrichment, number of bears in the group, and bears having a view out of their exhibit with a strong suggestion that the existence of a positive reinforcement training program may also be important. Among individuals, bears whose temperament measured high on the “interest” axis (defined in terms of behavior directed toward the novel object) tended to display less stereotypic behavior and those that scored high on the “slow to approach” axis displayed more pacing. We found higher FGM levels were associated with higher proportions of stereotypic pacing, lower levels of the temperament variable “interest” and smaller dry land exhibit area. These results support other studies suggesting polar bears are particularly prone to stereotypic pacing behavior in zoos and that there is a link between stress (measured as FGM) and pacing in zoo polar bears. These findings also suggest that some easily available tools, namely environmental enrichment and possibly positive reinforcement training, can effectively reduce the incidence of these behaviors. Exhibit designers should take note that providing bears with a view out of their exhibit and larger land areas are associated with both behavioral and physiological benefits. Finally, certain temperaments are associated with elevated levels of both stereotypic behavior and corticoids. This information may provide a tool for proactively identifying the individuals most likely to develop pacing behaviors and providing appropriately enhanced care before the behavior becomes established.