Elsevier

Journal of Veterinary Behavior

Volume 24, March–April 2018, Pages 62-68
Journal of Veterinary Behavior

Canine Research
A preliminary investigation into personality and pain in dogs

https://doi.org/10.1016/j.jveb.2018.01.005Get rights and content

Abstract

Adherence to basic animal welfare standards involves effective monitoring and control of pain, especially in a veterinary setting. Assessment relies on behavioral and physiological indicators. However, individual differences in physiology mediate consistent individual differences in behavior, referred to as “personality”. Therefore, personality may confound measurements of pain. The current work is a preliminary investigation into whether extraversion and neuroticism are associated with differences in individual behavioral and physiological responses to pain. Twenty dogs were observed during recovery from routine castration in a clinical setting. Core temperature was recorded using infrared thermography on admission, 15 minutes after extubation and every 30 minutes thereafter, until the subject was collected by their owner. Behavior during recovery was scored using Short-Form Glasgow Composite Measure Pain Scale at the same intervals as infrared thermography readings. Personality was measured using Monash Canine Personality Questionnaire–Revised, and owners rated their dog's tolerance to pain on a 5-point Likert scale. Pain score did not have an association with eye temperature discrepancy or core temperature changes from control, indicating it may not predict affective response to pain. More highly “extravert” subjects had significantly higher pain scores (P = 0.031), despite experiencing similar tissue damage. More “extravert” subjects showed significantly greater right eye temperature (P = 0.035), suggesting hemispheric dominance. “Neuroticism” had no association with physiological or behavioral responses to pain. Finally, owners were not able to predict their dog's behavioral or physiological response to pain. These results indicate that personality may be a useful clinical tool for assessing individual differences in response to pain, whereas owners' ratings of their dog's response are not reliable.

Introduction

Animals are unable to verbally convey their emotions to human caregivers, which makes the measurement of pain difficult (Reid et al., 2013). Therefore, behavior-based scales are used to quantify pain levels in animals, assisting the administration of the correct dosage of analgesic drugs and informing decisions on humane end points (Ashley et al., 2005). Consequently, it is vital that these scales are both sensitive and valid, to reduce the welfare implications that could occur through the incorrect assessment of pain (Rutherford, 2002). However, “personality”—defined as individual differences in behavior that are stable over time and across contexts (Koolhaas et al., 1999)—may confound assessment of rating scales. For example, human subjects scoring more highly for “extraversion” express their experiences of pain more clearly (Harkins et al., 1989), although they may experience pain less intensively (Ramírez-Maestre et al., 2004, Soriano et al., 2012). “Extraversion” is characterized by traits such as energetic behavior, assertiveness, and the tendency to seek stimulation (Costa and McCrae, 1985). Furthermore, highly neurotic people have a higher emotional stress response to pain when compared to those who have a low score for neuroticism (Goubert et al., 2004, Koenig et al., 2015). “Neuroticism” is associated with the tendency to experience unpleasant emotions easily and a low degree of emotional stability (Costa and McCrae, 1985).

The association between personality and pain response has recently been investigated in animals in a clinical setting (Ijichi et al., 2014). This study provides preliminary evidence that extraversion correlates with behavioral expressions of pain in horses, while “neuroticism” is associated with reduced tolerance to pain. However, it is not known whether personality affects the emotional experience of pain, as well as its behavioral expression in animals, as it does in humans (e.g., Asghari & Nicholas 2006). Furthermore, this study used a variety of naturally occurring tissue damage, making comparison across individuals more complex. In addition, it is not known whether the link between personality and pain is a species-specific phenomenon or whether it is seen in other nonhuman mammals.

In dogs, personality and pain can be measured using validated questionnaires. The Monash Canine Personality Questionnaire–Revised (MCPQ-R) has been validated as having good interrater and test-retest reliability for 5 factors which include “extraversion” and “neuroticism” (Ley et al., 2009b). On this scale, extravert dogs are typically active, excitable, and restless, whereas neurotic dogs are characterized as fearful, submissive, and timid. Canine pain can be measured using the Short-Form Glasgow Composite Measure Pain Scale (CMPS-SF) (Reid et al., 2007) as it has been shown to be both more sensitive and have less interobserver variability when compared to other tests (Guillot et al., 2011). It is designed as a clinical tool for dogs in acute pain and uses 30 descriptors within 6 categories to inform decisions about pain management (Reid et al., 2007).

The current investigation aims to investigate whether personality affects emotional and behavioral response to pain in dogs in a clinical setting. Castration was selected as it is a common routine procedure that causes moderate postoperative pain (Wagner et al., 2008) and is often conducted on healthy, young animals. In addition, the ability of owners to predict their dogs' response to pain was measured as horse owners' ratings have been shown to have high predictive accuracy (Ijichi et al., 2014). Canine “extraversion” and “neuroticism” was measured using the MCPQ-R (Ley et al., 2009a) and compared with pain behavior using the CMPS-SF (Reid et al., 2007). Emotional response to pain was measured using infrared thermography (IRT) as core temperature increases with arousal (Stewart et al., 2005, Travain et al., 2015) and decreases with pain in cattle (Stewart et al., 2008). Tympanic differences in temperature relate to lateralized cerebral blood flow (Riemer et al., 2016), reflecting emotional valence. Therefore, discrepancy between the right and left eye was explored as this may indicate lateralized cerebral blood flow.

Based on human and equine research, it was hypothesized that (1) extraversion will correlate positively with behavioral indicators of pain and may correlate with changes in physiology; (2) neuroticism will correlate negatively with owners' ratings of the subject's tolerance and positively with emotional response to pain; and (3) owners' tolerance ratings will correlate negatively with behavioral indicators of pain and emotional response. In addition, the association between behavioral and emotional response to castration will be investigated to determine if behavior is an accurate indicator of the emotional state of subjects.

Section snippets

Materials and methods

Subjects were assessed between 24th October, 2016 and 17th January, 2017 at 2 veterinary surgeries based in Gloucester and Surrey (UK). Subjects were admitted and treated as per standard protocol for each veterinary practice. Patients were premedicated with acepromazine and subcutaneous buprenorphine. General anesthesia was induced by intravenous propofol and maintained using inhaled isoflurane. Subjects were observed while pain was caused by a routine, voluntary procedure conducted in normal

Postcastration behavior and changes in physiology

Paired t-tests indicated significant differences in core temperature from control in observations 1, 2, 3, 5, 6, and 7 (Table 1; Figure 1). Paired t-tests and Wilcoxon signed-rank tests did not detect significant differences from control for eye temperature discrepancy (Table 2). The change in pain scores across observations can be seen in Figure 2.

Relationship between behavioral and physiological responses to pain

Spearman's correlation revealed that pain score did not have an association with eye temperature discrepancy (rs = −0.091, N = 164, P = 0.246) or

Discussion

Accurate pain assessment is essential for animal welfare and vital for correct pain management (Rutherford, 2002). Ijichi et al. (2014) provided preliminary evidence to suggest that behavioral indicators of pain in horses may not accurately indicate the level of damage sustained. Instead, this study found that the behavioral response to damage is associated with personality. This indicates that behavior-based pain assessment tools may not be accurate and the subsequent management of pain among

Conclusion

Our study provides preliminary evidence for individual variation in behavioral and physiological response to a moderately painful procedure. “Extraversion” was associated with differences in response to pain after castration as those scoring more highly for this factor presented with more prominent behavioral indicators of pain. The relationship between “extroversion” and emotional response to pain was more complex. More extrovert subjects had possible greater increases in core temperature and

Acknowledgments

The authors are indebted to the staff and clients from 2 anonymous veterinary practices without whom this research would not have been possible. In addition, the authors are very grateful for a rigorous review process that improved the article. The idea for the article was conceived by James Lush. The experiments were designed by James Lush and Carrie Ijichi. The study was performed by James Lush. The data were analyzed by James Lush and Carrie Ijichi. The article was written by James Lush and

References (39)

Cited by (0)

View full text