Functional MRI in Awake Dogs Predicts Suitability for Assistance Work

Area of Science:

• Neuroscience
• Animal Behavior
• Veterinary Medicine

Background:

• Service dogs undergo extensive training to assist individuals with disabilities.
• Predicting training success early is crucial for resource allocation and animal welfare.
• Neuroimaging offers a potential tool for assessing canine cognitive and emotional states.

Purpose of the Study:

• To evaluate the efficacy of functional magnetic resonance imaging (fMRI) in predicting service dog training success.
• To identify neural correlates associated with successful service dog outcomes.
• To develop a predictive classifier for service dog potential.

Main Methods:

• fMRI scans were conducted on 49 dogs aged 17-21 months at the start of service training.
• Brain responses were measured while dogs observed hand signals from familiar and unfamiliar individuals.
• A classifier was developed using fMRI data from regions of interest (ROIs) including the caudate, amygdala, and visual cortex.

Main Results:

• The fMRI-based classifier achieved a positive predictive value of 94% and a negative predictive value of 67%.
• Area under the ROC curve was 0.91, demonstrating significant predictive capability (0.80 with cross-validation, P=.01).
• Caudate response magnitude correlated positively with success; amygdala response interacted with visual cortex and correlated negatively with success.

Conclusions:

• fMRI shows significant potential for predicting service dog success.
• Successful service dogs exhibit caudate activity, suggesting generalized learning of hand signals.
• Reduced amygdala arousal, particularly in interaction with visual cortex during stranger-based signals, is associated with training failure.