Spatial working memory in a disappearing object task is impaired in female but not male dogs with chronic osteoarthritis

doi:10.21203/rs.3.rs-3000891/v1

Chronic pain in humans is associated with impaired working memory but it is not known whether this is the case in long-lived companion animals, such as dogs, who are especially vulnerable to developing age-related chronic pain conditions. Pain-related impairment of cognitive function could have detrimental effects on an animal’s ability to engage with its owners and environment or to respond to training or novel situations, which may in turn affect its quality of life. This study compared the performance of 20 dogs with chronic pain from osteoarthritis and 21 healthy control dogs in a disappearing object task of spatial working memory. Female neutered osteoarthritic dogs, but not male neutered osteoarthritic dogs, were found to have lower predicted probabilities of successfully performing the task compared to control dogs of the same sex. This shows that chronic pain from osteoarthritis may be associated with impaired working memory in dogs, a result which parallels findings from studies of human chronic pain disorders. That female dogs may be particularly prone to these effects warrants further investigation.

Spatial working memory

Disappearing object task

Osteoarthritis

Chronic pain

Dog

In humans, chronic pain is often associated with cognitive deficits (Moriarty et al. 2011) including in facets of executive function such as working memory (Berryman et al. 2013) and spatial working memory (Antepohl et al. 2003; Luerding et al. 2008). There are several hypothesised mechanisms by which this association could arise (Moriarty et al. 2011). Chronic pain may increase the allocation of attention and working memory to pain-related processing, thus reducing the availability of these capacity-limited resources for other cognitive tasks (Apkarian et al. 2004a). Chronic pain is also associated with affective disorders (Leino and Magni 1993; McWilliams et al. 2003; Ohayon and Schatzberg 2003) and impaired sleep (Drewes et al. 2000; Nicassio and Wallston 1992; Riley et al. 2001), both of which have been shown to impair working memory (Christopher and MacDonald 2005; Harvey et al. 2004; Miyata et al. 2013; Rose and Ebmeier 2006; Steenari et al. 2003). Chronic pain can cause structural and functional changes in brain areas such as the medial prefrontal cortex, amygdala and posterior cingulate cortex (Baliki et al. 2008; Moriarty et al. 2011), all of which play roles in executive function and other cognitive and decision-making processes (Apkarian et al. 2004b).

In contrast to research in humans, the extent to which chronic pain influences cognitive function in non-human animals has received relatively little study. Most work in this area has been done on laboratory rodents, mainly using models of induced pain (Mor et al. 2017; Moriarty et al. 2011; Negrete et al. 2017). However, an understanding of the relationship between spontaneous and naturally occurring chronic pain conditions and cognitive function in species kept and managed by humans is important from both a scientific and animal welfare perspective. For example, long-lived companion animals such as domestic dogs and cats are especially prone to the development of chronic pain conditions as they age. If these lead to cognitive deficits, for example in executive functions such as working memory, they could impair the animals’ ability to engage with their owners and environment or to respond to training or novel situations, and this in turn may affect their quality of life. Additionally, recognition of links between cognitive deficits and chronic pain conditions could assist with the diagnosis of these conditions, as well as preventing changes in cognitive function from being wrongly attributed to other disorders such as age-related cognitive decline.

One common disorder of dogs likely to induce chronic pain is osteoarthritis which affects approximately 200,000 dogs in the UK (Anderson et al. 2018). Dogs with osteoarthritis show behavioural (Hielm-Björkman et al. 2003; Wiseman et al. 2001) and gait (Conzemius et al. 2003; Moreau et al. 2003) changes which can be reversed with nonsteroidal analgesia (Hielm-Björkman et al. 2009; Moreau et al. 2003) or surgical treatment (Conzemius et al. 2003), suggesting osteoarthritis causes pain in dogs, as it does in humans (Arendt-Nielsen et al. 2010). Furthermore, dogs with osteoarthritis also display lower mechanical and thermal nociceptive thresholds than healthy control dogs (Hunt et al. 2018; Knazovicky et al. 2016), suggesting hyperalgesia, as well as increased temporal summation and impaired descending noxious inhibitory control (Hunt et al. 2018). These physiological signs indicate central sensitisation (Woolf 2011) and imply that the chronic pain observed in dogs with osteoarthritis shares many of the physiological changes observed in human chronic pain (Arendt-Nielsen et al. 2010).

Here we investigate whether naturally occurring osteoarthritis in dogs owned by the public is associated with changes in working memory as is the case in humans (Antepohl et al. 2003; Berryman et al. 2013; Luerding et al. 2008). Specifically, we compare the spatial working memory of dogs with and without chronic osteoarthritis using the disappearing object task devised by Fiset et al. (2003), which is well suited for use in dogs owned by the public as it can be performed in a single session lasting only a few hours (Smith et al. 2021). If dogs, like humans, show working memory deficits in chronic pain disorders, this also increases the validity of using spontaneous canine osteoarthritis as a model for human chronic pain, thus offering a 3Rs Replacement and Reduction alternative to models of induced pain used in laboratory-housed animals (Guhad 2005).

Ethical approval

Ethical approval for animal use was obtained from the University of Bristol AWERB (VIN/17/005), and ethical approval for recruitment and participation of members of the public (dog owners) was obtained from the University of Bristol Faculty of Health Sciences Research Ethics Committee (Application Ref: 31623)

Animals

Forty-one dogs were recruited via a social media campaign and poster and leaflet placement within veterinary clinics in Bristol and North Somerset. All recruited dogs had been neutered and 20 (12 females, 8 males) were assigned to the osteoarthritis group with a further 21 (6 females, 15 males) making up the control (non-osteoarthritis) group. Allocation to groups was based on clinical histories and clinical examination of the dogs as described below. Further information on the subjects is presented in Table 1.

Apparatus

The disappearing object task apparatus comprised four identical cuboidal open-backed wooden boxes (30cm high, 20cm wide, 15cm deep) each containing one of four identical carrier bags of 800g aquarium gravel for weighting. A transportable barrier to prevent the dog seeing the boxes at certain times during the task was constructed from a folding two-panel laundry airer (each panel 102cm high*62cm long) covered with opaque black plastic sheeting. These pieces of apparatus were arranged as shown in Fig. 1.

The object to be hidden consisted of either a tennis ball or squeaky rubber ring toy attached to a 125cm long, 1mm thick transparent nylon thread. All equipment was cleaned between uses with F10 disinfectant spray (Health and Hygiene (Pty) ltd., Florida Hills, South Africa).

Procedure

The task was performed in each owner’s own home by two experimenters, E1 and E2. A clinical examination including an orthopaedic examination of all appendicular joints was performed on each dog by a veterinary surgeon (E1: MS). A clinical history was also taken from each owner regarding any clinical signs of osteoarthritis or potential signs of other health problems. This information was used to assign each dog to the osteoarthritic or healthy control group. Both the dog’s owner and veterinary surgeon were independently asked to score the severity of the dog’s clinical signs of osteoarthritis as “none”, “mild”, “moderate” or “severe”.

The disappearing object task apparatus was set up and the dog was then shown both object types (tennis ball and rubber ring). The object that was most preferred by the dog was selected for use, according to E1’s subjective opinion (based on initial object approached, relative duration of interaction with each object and presence of tail-wagging and jumping behaviours). One osteoarthritic dog showed no interest in either object and so their own preferred toy (a blue plastic ring) was used instead.

The test procedure was adapted from that of Fiset et al. (2003; see also Smith et al. 2021) and consisted of three phases, shown in Table 2. In each trial, E2 held the dog by the collar whilst E1 placed the object in the required position (see Table 2) and returned to the position indicated in Fig. 1, before E2 released the dog. In the shaping phases, the dog received a reward (food, verbal praise and petting) when they touched the object with their nose, mouth or paw, and were then led back to the start point by E2 ready for the next trial. In the training and testing phases, the object was held by a string and moved in front of all boxes and then behind the target box (i.e. if the target box was box 3 or 4, the object was moved in front of boxes 1, 2, 3 and 4 and then behind the target box; if the target box was 1 or 2 the object was moved in front of boxes 4, 3, 2 and 1 and then behind the target box). The dog was able to observe the object being positioned and then the opaque barrier was placed in front of the dog (as shown in Fig. 1) for a specified memory retention interval (0s during training; 0, 60 or 120s during testing). The barrier was then removed and the dog was released with E1 looking towards the back of the test area (away from the boxes) and E2 looking towards E1 so as not to provide any visual cues as to the location of the object. If, following release, the dog visited (looked behind) the target (correct) box, they received a reward and were then led back to the starting position by E2 ready for the next trial. If the dog visited a box other than the target box or did not visit any box within 60s of release, they received no reward and were led by E1 back to the starting position ready to begin the next trial. Dogs were not allowed to visit more than one box per trial.

The testing phase involved three trials for each box, one per memory retention interval of 0, 60 and 120 seconds (based on Smith et al. (2021)), in order to vary the difficulty of the task and assess whether and how the retention interval affected the success rates of osteoarthritic and control dogs. The target box and interval used on each trial was pseudorandomly generated using the RAND() function in Microsoft Excel, such that each combination of box and interval occurred once and the same target box or interval did not occur in two consecutive trials. On each trial, the first box visited was recorded and noted as either a ‘success’ (correct box) or ‘fail’ test outcome.

During each home visit, owners were instructed to complete four questionnaires: the Helsinki Chronic Pain Index (HCPI) (Hielm-Björkman et al. 2009) and Canine Brief Pain Inventory (CBPI) (Brown et al. 2008) to assess the owner’s subjective opinion of their dog’s chronic pain; the Sleep and Night Time Restlessness Evaluation (SNoRE) to assess their dog’s sleep quality (Knazovicky et al. 2015) (as impaired sleep is common in human chronic pain disorders (Menefee et al. 2000) and may impair working memory (Chee and Choo 2004; Steenari et al. 2003)); the Canine Cognitive Dysfunction Rating Scale (CCDR) (Salvin et al. 2011) to screen dogs for clinical signs of age-related cognitive decline which could adversely affect task performance.

Data analysis

Data were analysed using mixed-effects logistic regression in R, using the lme4::glmer() function with the ‘binomial’ family specification. Model estimates were based on an adaptive Gaussian Hermite approximation with 100 integration points.

The success/failure of the dog on each trial during the testing phase was the binary outcome variable in all models. Initial univariable models are commonly performed for selection of factors to include in a final model (for example: Alves et al. 2002; Bogaert et al. 2005; Kooby et al. 2003). However, in this study, the interaction of each variable (shown in Supplementary Table S1) with group (osteoarthritis/control) was considered of greater importance than the main effect of each variable. Therefore, each initial model contained not only the variable of interest but also group and the interaction between variable and group, such that variables with a significant group*factor interaction would be detected regardless of whether the main effect was significant. Vet and owner-assigned severity scores and analgesia-related factors were added to initial models alone without assessing an interaction with group, since all control dogs and no osteoarthritic dogs had a vet-assigned severity score of ‘none’, all control dogs and only one osteoarthritic dog had an owner-assigned severity score of ‘none’, and no control dogs were receiving analgesia. The Quality of Life score on the CBPI was transposed to a number with “poor” given a score of 0 and “excellent” given a score of 4. ‘Dog id’ was included as a random effect. All models were performed on trial-level data (success vs failure).

All fixed-effect variables where p<0.1 for the main effect or effect of interaction between that variable and group were selected for addition to the final multivariate model along with their interaction with group, and all fixed-effect variables where p<0.05 for the final model were considered statistically significant. Where significant results were obtained for interactions between factors, Mann-Whitney U tests were used to compare combinations of individual factor levels because these data were non-normal. For statistically significant continuous variables, predicted probabilities of success were estimated using the multivariate model and the equations of regression lines obtained. Mann-Whitney U-tests were used to compare continuous signalment variables between groups.

The signalments of dogs recruited in each group are shown in Table 1. Analysis of continuous variables showed that osteoarthritic dogs had significantly higher HCPI (U=39, p=1.39*10^-5), CBPI Severity (U=56, p=2.57*10^-5), and CBPI Interference (U=49.5, p=1.84*10^-5) scores, and lower CBPI Quality of life (U=250, p=0.01497) scores than control dogs, but the two groups did not differ in SNoRE score (U=181, p=0.8109; Fig. 2A). We investigated whether osteoarthritis was more severe in females than males (as is the case in humans: Affleck et al. 1999; Keefe et al. 2000; Srikanth et al. 2005) but whilst median scores for all clinical questionnaire components except for the CBPI QOL score were higher in female osteoarthritic dogs, this effect was not significant (Holm-Bonferroni-adjusted α-thresholds) for any questionnaire score (Fig. 2B) nor was there any significant difference between the vet-assigned severity scores of male and female osteoarthritic dogs (Fisher’s exact test, p= 0.796). There were no significant differences between groups in age (U=186, p=0.5284), sex (Fisher’s exact test, p=0.0616) or body condition score (U=140.5, p=0.0554). No dogs had a CCDR score of 50 or more (the threshold for probable cognitive dysfunction (Salvin et al. 2011).

Results from initial models are shown in Supplementary Table S1. Results for the final multivariate model are shown in Table 3. Overall, osteoarthritic dogs performed less well on the disappearing object task (mean predicted probability of success = 0.673, 95% CI = 0.612-0.730) compared to control dogs (mean predicted probability of success = 0.727, 95% CI = 0.668-0.779), but a significant (p<0.05) group by sex interaction indicated that this was principally due to female (U=58, p=0.0412), but not male (U=46, p=0.375) osteoarthritic dogs showing lower proportions of successful visits than control dogs of the same sex (Fig. 3).

There was a significant effect of memory retention interval, with predicted probability of success decreasing with increasing interval duration as expected (Fig. 4; equation of regression line based on predicted data: y=0.836 – 0.00216x), and a significant age by group interaction, with control (equation of regression line based on predicted data: y=1.19 - 0.0630x) but not osteoarthritic dogs (equation of regression line based on predicted data: y=0.612 + 0.0162x) showing decreased probability of success with increasing age.

There was also a significant box by group interaction for boxes 2 and 3, with control dogs (box 2 mean proportion of successful visits = 0.617, 95% CI=0.490-0.729, U=328, p=0.00372; box 3 mean proportion of successful visits = 0.557, 95% CI=0.433-0.675, U=325, p=0.00459), but not osteoarthritic dogs (box 2 mean proportion of successful visits = 0.690, 95% CI=0.562-0.794, U=180, p=0.579; box 3 mean proportion of successful visits= 0.550, 95% CI=0.425-0.669, U=237, p=0.294), showing a lower proportion of successful visits when the object was hidden behind these boxes compared to when it was hidden behind box 1 (control group mean proportion of successful visits = 0.887, 95% CI=0.785-0.944; osteoarthritic group mean proportion of successful visits = 0.644, 95% CI=0.517-0.754). A significant breed class by group interaction was also seen for terrier-type dogs, with control group terriers (mean proportion of successful visits = 0.957, 95% CI=0.790-0.998, U=0, p=0.0378) but not osteoarthritic group terriers (mean proportion of successful visits = 0.667, 95% CI=0.467-0.820, U=9, p=1) showing increased predicted probability of success compared to gundogs of the same group (mean proportion of successful visits for control gundogs = 0.690, 95% CI=0.601-0.767; mean proportion of successful visits for osteoarthritic gundogs = 0.670, 95% CI=0.576-0.752). There were also significant main effects found for age, terrier breed class, and target boxes 2 and 3, but these were considered subsidiary to the interaction effects involving these variables.

Dogs were assigned to the osteoarthritic and control groups on the basis of clinical examination and clinical history. Differences in the Helsinki Chronic Pain Index (HCPI) (Hielm-Björkman et al. 2009) and Canine Brief Pain Inventory (CBPI) (Brown et al. 2008) scores indicate that dogs in the osteoarthritis group were indeed experiencing more severe pain, functional impairment and decreased quality of life compared to control dogs, hence supporting the assignments made. Analysis of the disappearing object task detected poorer memory performance in osteoarthritic compared to control dogs, but this was principally due to female osteoarthritic dogs making fewer successful visits to the target boxes than female control dogs, whilst the same relationship was not observed for male dogs. Thus, female but not male neutered osteoarthritic dogs exhibited impaired spatial memory in the disappearing object task compared to control dogs of the same sex and neutering status. It seems unlikely that this effect was mediated by the interplay between sex-specific hormones (cf. (Frye et al. 2004; Gouchie and Kimura 1991) and osteoarthritis (or the pain it induced), because all dogs in the study were neutered. However, osteoarthritis is more common and severe in human women than men (Affleck et al. 1999; Keefe et al. 2000; Srikanth et al. 2005), especially following the menopause (Lawrence et al. 1966; Silman and Newman 1996; Wluka et al. 2000), with ovariectomised rats (Hoegh-Andersen et al. 2004) and sheep (Cake et al. 2005) being used in previous studies to model this effect. Therefore, it is possible that ovariectomised female dogs in this study may also have had more severe osteoarthritis than male dogs. However, whilst the median HCPI and CBPI Interference scores were slightly higher for female than male osteoarthritic dogs, this difference was not statistically significant, implying that female osteoarthritic dogs did not have markedly more severe osteoarthritis than male dogs in this study.

Female mammals often perform less well in spatial memory tasks than male mammals (Jones et al. 2003; Jones and Healy 2006), so it is possible that the combination of this effect and spatial working memory impairment from (the pain of) osteoarthritis was sufficient to cause a significant decrease in performance in osteoarthritic female dogs, but neither factor alone was sufficient to cause decreased performance in healthy female dogs or male osteoarthritic dogs. A related finding was reported by (Harris et al. 2008b) who found that acute stress (as may co-occur with a pain state) resulted in behavioural changes and poorer performance in a spatial memory task in female but not male adult rats, although there was no evidence that chronic stress, as may be induced by longer-term osteoarthritic pain, had a similar effect (Harris et al. 2008a; Harris et al. 2008b).

The decrease in probability of success with increasing memory retention interval is similar to that observed by (iset et al. (2003) and Smith et al. (2021) and suggests that dogs were indeed using working memory to locate the object rather than relying on cues such as scent, in which case they would likely have shown similar success rates at all intervals. Control dogs also showed decreased probabilities of success with increasing age, supporting previous findings that spatial working memory decreases in older dogs (Snigdha et al. 2012; Tapp et al. 2003). However, osteoarthritic dogs did not show this effect. One potential explanation is that because both chronic pain and age are associated with changes in prefrontal cortex structure and function (Apkarian et al. 2004b; Rypma and D'Esposito 2000) that may be linked to working memory deficits (Rypma and D'Esposito 2000), the effect of osteoarthritic chronic pain on working memory may have masked any changes with age in this ability.

Control dogs, but not osteoarthritic dogs, were less likely to find the object if it was behind box 2 or 3 than box 1 (see also Smith et al. 2021). This may have been because outer boxes (1 and 4) were nearer to environmental cues such as walls or furniture compared to inner boxes (2 and 3), which may have made them easier for dogs to distinguish. It is possible that spatial working memory deficits affected osteoarthritic dogs’ ability to use such environmental cues and thus they did not show such an effect. Finally, we observed that control terriers were more likely to successfully find the object than control gundogs. However, since there were only two terriers per group this finding is unlikely to be representative of the general population.

In summary, our study illustrates the potential for field investigations of cognitive abilities, including working memory components of executive function, in dogs owned by the public. Such in-home investigations are likely to be less stressful for dogs than those that require them to be tested in an unfamiliar laboratory setting, and hence will provide data that is relatively uncontaminated by incidental effects of acute stress on cognitive function (Harris et al. 2008b; Mendl 1999). They can also start to reveal the effects of natural ageing and spontaneous changes in health on cognitive ability in this key companion animal species, which are important in their own right from an animal welfare perspective and may also be of use in helping to understand and model human health and ageing (Hoffman et al. 2018). To this end, our study is the first to indicate that some dogs with osteoarthritis and associated chronic pain may have working memory deficits, a finding which has clear parallels in human chronic pain disorders (Berryman et al. 2013). That these may be particularly pronounced in females is of interest and warrants further investigation.

Acknowledgements

This research was funded by the UK Biotechnology and Biological Sciences Research Council South West Biosciences Doctoral Training Partnership (BBSRC SWBio DTP) programme, grant number BB/J014400/1. We are grateful to the dogs and their owners for their enthusiastic participation in the study, and to Dr Lauren Harris and Mr Juan Carlos Arenas-Salcedo for taking the role of E2 when assisting with data collection.

Supplementary information Supplementary material associated with this article can be found, in the online version, at https://doi.org/...'

Funding This research was funded by the UK Biotechnology and Biological Sciences Research Council South West Biosciences Doctoral Training Partnership (BBSRC SWBio DTP) programme, grant number BB/J014400/1.

Ethical approval Ethical approval for animal use was obtained from the University of Bristol AWERB (VIN/17/005), and ethical approval for recruitment and participation of members of the public (dog owners) was obtained from the University of Bristol Faculty of Health Sciences Research Ethics Committee (Application Ref: 31623)

Author’s contributions All authors contributed to study conceptualisation, design and methodology. Data collection was carried out by MS who also curated, analysed and visualised the data, and wrote the first draft with input from JCM and MM who also provided supervision. A second draft was prepared by MM with input from JCM and MS. All authors read and approved the final manuscript.

Conflict of interest The authors have no competing interests to declare.

Availability of data and materials Data are at: https://figshare.com/s/d735b73229158c2f7487

Affleck G et al. (1999) Everyday life with osteoarthritis or rheumatoid arthritis: independent effects of disease and gender on daily pain, mood, and coping. Pain 83:601-609
Alves A, Panis Y, Trancart D, Regimbeau J-M, Pocard M, Valleur P (2002) Factors associated with clinically significant anastomotic leakage after large bowel resection: multivariate analysis of 707 patients. World Journal of Surgery 26:499-502 doi:10.1007/s00268-001-0256-4
Anderson KL et al. (2018) Prevalence, duration and risk factors for appendicular osteoarthritis in a UK dog population under primary veterinary care. Scientific Reports 8:5641 doi:10.1038/s41598-018-23940-z
Antepohl W, Kiviloog L, Andersson J, Gerdle B (2003) Cognitive impairment in patients with chronic whiplash-associated disorder–a matched control study NeuroRehabilitation 18:307-315
Apkarian AV et al. (2004a) Chronic pain patients are impaired on an emotional decision-making task Pain 108:129-136 doi:10.1016/j.pain.2003.12.015
Apkarian AV, Sosa Y, Sonty S, Levy RM, Harden RN, Parrish TB, Gitelman DR (2004b) Chronic back pain is associated with decreased prefrontal and thalamic gray matter density. J Neurosci 24:10410-10415 doi:10.1523/JNEUROSCI.2541-04.2004
Arendt-Nielsen L, Nie H, Laursen MB, Laursen BS, Madeleine P, Simonsen OH, Graven-Nielsen T (2010) Sensitization in patients with painful knee osteoarthritis. Pain 149:573-581 doi:10.1016/j.pain.2010.04.003
Baliki MN, Geha PY, Apkarian AV, Chialvo DR (2008) Beyond feeling: chronic pain hurts the brain, disrupting the default-mode network dynamics. J Neurosci 28:1398-1403 doi:10.1523/JNEUROSCI.4123-07.2008
Berryman C, Stanton TR, Jane Bowering K, Tabor A, McFarlane A, Lorimer Moseley G (2013) Evidence for working memory deficits in chronic pain: a systematic review and meta-analysis. Pain 154:1181-1196 doi:10.1016/j.pain.2013.03.002
Bogaert D et al. (2005) Epidemiology of nasopharyngeal carriage of neisseria meningitidis in healthy dutch children. Clinical Infectious Diseases 40:899-902 doi:10.1086/428351
Brown DC, Boston RC, Coyne JC, Farrar JT (2008) Ability of the Canine Brief Pain Inventory to detect response to treatment in dogs with osteoarthritis. Journal of the American Veterinary Medical Association 233:1278-1283 doi:10.2460/javma.233.8.1278
Cake MA, Appleyard RC, Read RA, Smith MM, Murrell GA, Ghosh P (2005) Ovariectomy alters the structural and biomechanical properties of ovine femoro-tibial articular cartilage and increases cartilage NOS. Osteoarthritis Cartilage 13:1066-1075 doi:10.1016/j.joca.2005.07.001
Chee MWL, Choo WC (2004) Functional imaging of working memory after 24 hr of total sleep deprivation. Journal of Neuroscience 24:4560-4567
Christopher G, MacDonald J (2005) The impact of clinical depression on working memory. Cognitive neuropsychiatry 10:379-399
Conzemius MG, Aper RL, Corti LB (2003) Short‐term outcome after total elbow arthroplasty in dogs with severe, naturally occurring osteoarthritis. Veterinary Surgery 32:545-552
Drewes AM, Nielsen KD, Hansen B, Taagholt SJ, Bjerregard K, Svendsen L (2000) A longitudinal study of clinical symptoms and sleep parameters in rheumatoid arthritis. Rheumatology 39:1287-1289 doi:10.1093/rheumatology/39.11.1287
Fiset S, Beaulieu C, Landry F (2003) Duration of dogs' (Canis familiaris) working memory in search for disappearing objects. Animal Cognition 6:1-10
Frye CA, Edinger KL, Seliga AM, Wawrzycki JM (2004) 5 alpha-reduced androgens may have actions in the hippocampus to enhance cognitive performance of male rats. Psychoneuroendocrinology 29:1019-1027 doi:10.1016/j.psyneuen.2003.10.004
Gouchie C, Kimura D (1991) The relationship between testosterone levels and cognitive-ability patterns. Psychoneuroendocrinology 16:323-334 doi:10.1016/0306-4530(91)90018-o
Guhad F (2005) Introduction to the 3Rs (Refinement, Reduction and Replacement). Contemporary Topics in Laboratory Animal Science 44:58-59
Harris AP, D'Eath RB, Healy SD (2008a) Sex differences in spatial cognition are not caused by isolation housing. Behaviour 145:757-778 doi:10.1163/156853908783929142
Harris AP, D'Eath RB, Healy SD (2008b) Sex differences, or not, in spatial cognition in albino rats: acute stress is the key. Animal Behaviour 76:1579-1589 doi:10.1016/j.anbehav.2008.07.016
Harvey PO, Le Bastard G, Pochon JB, Levy R, Allilaire JF, Dubois B, Fossati P (2004) Executive functions and updating of the contents of working memory in unipolar depression. Journal of psychiatric research 38:567-576
Hielm-Björkman AK et al. (2003) Evaluation of methods for assessment of pain associated with chronic osteoarthritis in dogs. Journal of the American Veterinary Medical Association 222:1552-1558 doi:10.2460/javma.2003.222.1552
Hielm-Björkman AK, Rita H, Tulamo R-M (2009) Psychometric testing of the Helsinki chronic pain index by completion of a questionnaire in Finnish by owners of dogs with chronic signs of pain caused by osteoarthritis. American Journal of Veterinary Research 70:727-734 doi:10.2460/ajvr.70.6.727
Hoegh-Andersen P et al. (2004) Ovariectomized rats as a model of postmenopausal osteoarthritis: validation and application. Arthritis Res Ther 6:R169-180 doi:10.1186/ar1152
Hoffman JM, Creevy KE, Franks A, O'Neill DG, Promislow DEL (2018) The companion dog as a model for human aging and mortality. Aging Cell 17 doi:10.1111/acel.12737
Hunt JR et al. (2018) Electrophysiological characterisation of central sensitisation in canine spontaneous osteoarthritis pain. Pain 159:2318-2330 doi:10.1097/j.pain.0000000000001336
Jones CM, Braithwaite VA, Healy SD (2003) The evolution of sex differences in spatial ability. Behavioral Neuroscience 117:403-411 doi:10.1037/0735-7044.117.3.403
Jones CM, Healy SD (2006) Differences in cue use and spatial memory in men and women. Proc Biol Sci 273:2241-2247 doi:10.1098/rspb.2006.3572
Keefe FJ, Lefebvre JC, Egert JR, Affleck G, Sullivan MJ, Caldwell DS (2000) The relationship of gender to pain, pain behavior, and disability in osteoarthritis patients: the role of catastrophizing. Pain 87:325-334
Knazovicky D, Helgeson ES, Case B, Gruen ME, Maixner W, Lascelles BDX (2016) Widespread somatosensory sensitivity in naturally occurring canine model of osteoarthritis. Pain 157:1325
Knazovicky D, Tomas A, Motsinger-Reif A, Lascelles BDX (2015) Initial evaluation of nighttime restlessness in a naturally occurring canine model of osteoarthritis pain. PeerJ 3:e772
Kooby DA et al. (2003) Influence of transfusions on perioperative and long-term outcome in patients following hepatic resection for colorectal metastases. Annals of Surgery 237:860
Lawrence JS, Bremner JM, Bier F (1966) Osteo-arthrosis. Prevalence in the population and relationship between symptoms and x-ray changes. Annals of the Rheumatic Diseases 25:1-24
Leino P, Magni G (1993) Depressive and distress symptoms as predictors of low back pain, neck-shoulder pain, and other musculoskeletal morbidity: a 10-year follow-up of metal industry employees. Pain 53:89-94
Luerding R, Weigand T, Bogdahn U, Schmidt-Wilcke T (2008) Working memory performance is correlated with local brain morphology in the medial frontal and anterior cingulate cortex in fibromyalgia patients: structural correlates of pain–cognition interaction. Brain 131:3222
McWilliams LA, Cox BJ, Enns MW (2003) Mood and anxiety disorders associated with chronic pain: an examination in a nationally representative sample. Pain 106:127-133 doi:http://dx.doi.org/10.1016/S0304-3959(03)00301-4
Mendl M (1999) Performing under pressure: stress and cognitive function. Applied Animal Behaviour Science 65:221-244 doi:10.1016/S0168-1591(99)00088-X
Menefee LA, Cohen MJM, Anderson WR, Doghramji K, Frank ED, Lee H (2000) Sleep disturbance and nonmalignant chronic pain: a comprehensive review of the literature. Pain Medicine 1:156-172
Miyata S, Noda A, Iwamoto K, Kawano N, Okuda M, Ozaki N (2013) Poor sleep quality impairs cognitive performance in older adults. Journal of Sleep Research 22:535-541 doi:10.1111/jsr.12054
Mor D, Kendig MD, Kang JWM, Gemikonakli G, Austin PJ, Kalman E, Corbit LH (2017) Peripheral nerve injury impairs the ability to maintain behavioural flexibility following acute stress in the rat. Behavioural Brain Research 328:123-129 doi:10.1016/j.bbr.2017.04.003
Moreau M, Dupuis J, Bonneau NH, Desnoyers M (2003) Clinical evaluation of nutraceutical, carprofen and meloxicam for the treatment. The Veterinary Record 152:323-329
Moriarty O, McGuire BE, Finn DP (2011) The effect of pain on cognitive function: A review of clinical and preclinical research. Progress in Neurobiology 93:385-404 doi:10.1016/j.pneurobio.2011.01.002
Negrete R, Gutierrez MSG, Manzanares J, Maldonado R (2017) Involvement of the dynorphin/KOR system on the nociceptive, emotional and cognitive manifestations of joint pain in mice. Neuropharmacology 116:315-327 doi:10.1016/j.neuropharm.2016.08.026
Nicassio PM, Wallston KA (1992) Longitudinal relationships among pain, sleep problems, and depression in rheumatoid-arthritis. Journal of Abnormal Psychology 101:514-520 doi:10.1037/0021-843x.101.3.514
Ohayon MM, Schatzberg AF (2003) Using chronic pain to predict depressive morbidity in the general population Archives of General Psychiatry 60:39-47
Riley JL, Benson MB, Gremillion HA, Myers CD, Robinson ME, Smith CL, Waxenberg LB (2001) Sleep disturbance in orofacial pain patients: Pain-related or emotional distress? Cranio-the Journal of Craniomandibular & Sleep Practice 19:106-113 doi:10.1080/08869634.2001.11746159
Rose EJ, Ebmeier KP (2006) Pattern of impaired working memory during major depression. J Affect Disord 90:149-161
Rypma B, D'Esposito M (2000) Isolating the neural mechanisms of age-related changes in human working memory. Nature Neuroscience 3:509 doi:10.1038/74889
Salvin HE, McGreevy PD, Sachdev PS, Valenzuela MJ (2011) The canine cognitive dysfunction rating scale (CCDR): A data-driven and ecologically relevant assessment tool. The Veterinary Journal 188:331-336 doi:http://dx.doi.org/10.1016/j.tvjl.2010.05.014
Silman AJ, Newman J (1996) Obstetric and gynaecological factors in susceptibility to peripheral joint osteoarthritis. Annals of the Rheumatic Diseases 55:671
Smith M, Murrell JC, Mendl M (2021) Two assays of working memory in companion dogs: The holeboard and disappearing object tasks. Applied Animal Behaviour Science 23:105179 doi:10.1016/j.applanim.2020.105179
Snigdha S, Christie LA, De Rivera C, Araujo JA, Milgram NW, Cotman CW (2012) Age and distraction are determinants of performance on a novel visual search task in aged Beagle dogs. Age (Dordr) 34:67-73 doi:10.1007/s11357-011-9219-3
Srikanth VK, Fryer JL, Zhai G, Winzenberg TM, Hosmer D, Jones G (2005) A meta-analysis of sex differences prevalence, incidence and severity of osteoarthritis. Osteoarthritis Cartilage 13:769-781 doi:10.1016/j.joca.2005.04.014
Steenari M-R, Vuontela V, Paavonen EJ, Carlson S, Fjällberg M, Aronen ET (2003) Working memory and sleep in 6-to 13-year-old schoolchildren. Journal of the American Academy of Child & Adolescent Psychiatry 42:85-92
Tapp PD, Siwak CT, Estrada J, Holowachuk D, Milgram NW (2003) Effects of age on measures of complex working memory span in the beagle dog (Canis familiaris) using two versions of a spatial list learning paradigm. Learning & Memory 10:148-160 doi:10.1101/lm.56503
Wiseman ML, Nolan AM, Reid J, Scott EM (2001) Preliminary study on owner-reported behaviour changes associated with chronic pain in dogs. The Veterinary Record 149:423-424
Wluka AE, Cicuttini FM, Spector TD (2000) Menopause, oestrogens and arthritis. Maturitas 35:183-199
Woolf CJ (2011) Central sensitization: Implications for the diagnosis and treatment of pain. Pain 152:S2-S15 doi:http://dx.doi.org/10.1016/j.pain.2010.09.030

Table 1 Signalments of participant dogs

	Group
Variable	Control	Osteoarthritis
Sex
Female (neutered)	6	12
Male (neutered)	15	8
Breed Class:
Gundog	10	9
Crossbred	4	3
Hound	1	0
Pastoral	1	4
Terrier	2	2
Utility	1	0
Working	2	2
Analgesia provision
No	21	11
Yes	0	9
Analgesia frequency
None	21	11
Occasional	0	3
Daily	0	6
Analgesia type
None	21	11
Nonsteroidal analgesia only	0	6
Other analgesics provided	0	3
Season
Summer (May-August)	12	13
Winter (November-February)	9	7
Vet-assigned severity score
None	21	0
Mild	0	11
Moderate	0	7
Severe	0	2
Owner-assigned severity score
None	21	1
Mild	0	9
Moderate	0	9
Severe	0	1
Object choice for task
Tennis ball	11	8
Squeaky ring	10	10
Both objects used	0	1
Own toy	0	1
Continuous variables:
Age (years)	7.85+/- 0.50	0.81+/- 0.68
Body Condition Score	4.67+/- 0.42	5.45+/-0.63
SNoRE Score	19.95 +/- 3.93	20.42 +/-3.90
HCPI Score	4.14 +/- 1.97	14.90 +/- 3.31
CBPI Severity Score	0.0375 +/- 0.0557	2.04 +/- 0.974
CBPI Interference Score	0.0750 +/- 0.1058	1.86 +/- 0.774
CBPI Quality of Life (QOL) Score	3.70 +/- 0.42	3.33 +/- 0.32
CCDR Score	25.75 +/- 0.74	27.30 +/- 0.99

Table 2 Phases of the disappearing object task

Phase	Object location and barrier placement (if present)	Criterion for reaching next phase
Shaping Phase (Stage 1)	Object placed on floor next to dog, dog encouraged verbally and by pointing to touch the object (successful trial recorded when object is touched)	5 successful consecutive trials
Shaping Phase (Stage 2)	Object moved slightly closer to boxes with each successful trial	Object reaches boxes
Shaping Phase (Stage 3)	Object placed midway between two (randomly selected) boxes	10 successful trials
Shaping Phase (Stage 4)	Object placed behind box	8 trials (2 per box)
Training phase	Object placed behind box and barrier put in place momentarily (0s to habituate dog to presence of barrier) then removed before releasing dog	12 trials (3 per box)
Testing phase	Object placed behind box and barrier then put in place for either 0, 60 or 120s (memory retention interval) before releasing dog	12 trials (1 per box per retention interval)

Table 3 Results of the final multivariable model

	Main effect					Interaction with Group
Variable	Odds ratio	95% confidence limits		z	p	Odds ratio	95% confidence limits		z	p
Group
Control	Ref.	-	-	-	-
Osteoarthritis	0.01	0	0.86	-2.027	0.04268
Owner-assigned severity score
None	Ref.	-	-	-	-
Mild	0.399	0.065	2.45	-0.993	0.32085
Moderate	0.788	0.125	4.985	-0.253	0.80043
Severe	0.774	0.064	9.34	-0.201	0.84037
Box
1	Ref.	-	-	-	-	Ref.	-	-	-	-
2	0.173	0.064	0.472	-3.433	0.0006	7.203	1.902	27.277	2.907	0.00365
3	0.125	0.043	0.364	-3.814	0.00014	6.883	1.667	28.424	2.666	0.00768
4	0.643	0.216	1.916	-0.793	0.42801	4.112	0.971	17.414	1.92	0.05485
Sex
Female	Ref.	-	-	-	-	Ref.	-	-	-	-
Male	0.783	0.318	1.929	-0.532	0.5948	4.016	1.26	12.796	2.351	0.01873
Breed Class:
Gundog	Ref.	-	-	-	-	Ref.	-	-	-	-
Crossbred	0.559	0.23	1.36	-1.281	0.20012	2.514	0.681	9.278	1.383	0.16657
Hound	1.133	0.21	6.124	0.145	0.88432
Pastoral	2.699	0.275	26.54	0.852	0.39445	0.62	0.052	7.429	-0.377	0.70618
Terrier	16.094	1.899	136.383	2.548	0.01082	0.069	0.006	0.818	-2.119	0.03405
Utility	0.877	0.222	3.468	-0.187	0.85183
Working	1.504	0.392	5.767	0.596	0.55145	0.506	0.091	2.818	-0.777	0.43689
Continuous variables:
Interval duration	0.992	0.984	0.999	-2.283	0.02242	0.992	0.982	1.002	-1.527	0.12669
Trial number	0.999	0.88	1.135	-0.013	0.99002	0.908	0.764	1.078	-1.105	0.26899
Age	0.666	0.465	0.953	-2.221	0.02632	1.674	1.048	2.673	2.156	0.03112

Ref. = reference category

No competing interests reported.

smithetalsupplementarymaterial.docx

Spatial working memory in a disappearing object task is impaired in female but not male dogs with chronic osteoarthritis

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Methods

Results

Discussion

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1