Reduced ability of dogs to maintain eye contact with their owner is a sign of cognitive decline due to ageing  – as rescue dogs performed worse than non-rescue dogs in the present study regardless of age, this could possibly indicate cognitive impairment. Rescue dogs also tended to be more stressed during the eye contact test than non-rescue dogs. Making eye contact with humans increases oxytocin levels in dogs hence, as oxytocin reduces stress reactivity in dogs, the higher durations of eye contact may account for fewer stress behaviours displayed by non-rescue dogs. 

There was no difference between rescue and non-rescue dogs in the Inferential Reasoning Test. Stewart et al., (2015) found that an average of 50% of dogs, or fewer, choose the correct cup during this test, possibly due to the dogs’ misinterpreting the owner lifting the empty cup as a communication signal. However, in the present study, 71% of rescue dogs chose the correct cup, while 63% of control dogs chose the correct cup so it is not clear that the dogs were misinterpreting the test. Several studies have demonstrated a strong positive link between absolute brain size and executive functions. Horschler et al., (2019) found that absolute brain size is a predictor of executive function in dogs, showing that dogs with larger brains performed better in several cognitive tests. However, they also found that the inferential reasoning test did not fit their model – brain size did not predict performance in this particular test. Hence, the lack of a difference between the rescues and controls in the present study may have been due to the nature of the test itself rather than a true reflection of cognitive abilities of the respective groups. Stress behaviours during this test were positively correlated with both the simple and complex and memory tests, indicating that rescue dogs that found any one of these tests stressful were likely to find the others stressful too. 

Rescue dogs were better than control dogs at remembering where the treat was hidden in Memory Test A, which differed from the expected outcome. 

Rescue dogs had shorter durations of eye contact than control dogs which indicates lower impulse control, and also scored higher in “Fearfulness” on the DPQ. Dogs with higher levels of impulsivity and fearfulness were more likely to display resource guarding behaviour. In addition, the fact that rescue dogs maintained eye contact for shorter durations than non-rescue dogs may indicate that human social contact is less important for them than for non-rescues. Most dogs find praise equally or more rewarding than food, but those that find food more rewarding are less likely to seek contact with their owner. While the life histories of the rescue dogs are unknown, it is possible that they may have come from situations where food was less plentiful than their current home. Therefore, rescue dogs may be more highly motivated than control dogs to keep track of potentially “scarce” food resources which may explain why they perform better than controls in remembering where food is hidden. 

Percentage time spend displaying stress behaviours during Memory Test A was negatively correlated with eye contact duration for both rescue dogs and control dogs – dogs that performed worse in the Eye Contact Test were more likely to find the memory test stressful. This may indicate that dogs that maintain less social engagement with their owners and have lower impulse control are also more stressed in a situation where they have to keep track of hidden food. 

This test differed from Memory Test A in that it was more complex and challenged the dogs in both working memory capacity and spatial memory. Impairment in spatial memory can often occur even before impairment in executive functions, especially in older dogs. Poor performance in this test has been shown to be a sign of cognitive decline in ageing dogs, therefore, the results of this test in the present study do not support the hypothesis of cognitive impairment in rescue dogs. 

There was a negative correlation between number of tries to find the correct cup and number of training classes the dog had attended in the previous two years, indicating that dogs that had attended more training sessions were better at finding the treat in fewer attempts. Many dog training classes are centred around “nose-work” where the dog is trained to independently search for a particular scent in an area. While the type of training classes the dogs attended is not known, this type of class is particularly popular in Sweden, so it is possible that these dogs have attended nose-work classes. While the dogs may have been using scent to locate the treat, Duranton and Horowitz (2019) argue that training dogs in nose-work allows them to act autonomously with their own initiative and helps in learning to analyse their environment and develop problem solving skills. This may account for the correlation observed.