Assessment of cognitive dysfunction in companion dogs.

Detection of cognitive dysfunction syndrome (CDS) in pet dogs is largely based on owner-reported alterations in behavior, providing only subjective data for diagnosis. The aim of this study was to validate standardized tests that permit the assessment of canine cognitive function in a clinical setting, possibly providing an objective method for diagnosing CDS. Privately owned dogs (N = 12, 5.9 ± 3.7 years) were administered a modified version of a neuropsychological test, originally validated on laboratory beagle dogs (Milgram et al., 1994). Learning rates, measured by the number of errors before achieving a learning criterion, were measured in two tasks of increasing cognitive demand. Simple discrimination learning was more readily acquired (9.57 ± 6.43 errors) than reversal learning (34.0 + 13.05; P<0.001). The same procedure in a sample of laboratory beagles (N = 7, 4.9 ± 2.2 years) produced a similar pattern; fewer errors were required to achieve discrimination than reversal (9.61 + 8.69 vs. 53.57 + 10.79; P < 0.001). Nonetheless, pet dogs outperformed beagles in reversal learning (P = 0.004). Only one owner reported behavioral abnormalities in his dog, which was also the oldest dog in the clinical population tested (16 years). This dog had the worst performance in reversal learning (65 errors), supporting the hypothesis that clinical based testing may be useful for diagnosis. In accordance with the theoretical framework of the model proposed by Milgram, the results confirm that objective neuropsychological tests can be used to assess cognitive functions in both laboratory and pet dogs. A larger sample, including an expanded number of aged dogs, will help determine the sensitivity of these tests to objectively assess age-related cognitive deficits.