Dog Cognition

Some statistics on dogs as pets 40% of U.S. households have at least 1 dog 70+ million dogs in U.S. alone Estimated $45 billion/year industry Why dogs? Behavior appears to be attuned to humans As companion animals, show “loyalty”, “compassion”, and other cognitive-emotional behaviors

Why study dog cognition? Unlike other domestic animals, wild conspecific is readily available Data are available domestication of a wild canine species (Fox) which demonstrates evolution of many “dog-like” traits. Appear to have cognitive abilities that are more advanced than most non-human primates. Question is why DO dogs have good cognitive abilities? Two competing theories to explain evolution of cognition in dogs: Domestication hypothesis (Hare & Tomasello, 2005) Two-Stage hypothesis (Wynne)

What does research say? Dogs can count: Dogs can problem solve: Numerical competence and ability to discriminate more and less Dogs about as good at numerical competence as the great apes! Dogs can problem solve: Get a stick through an obstacle Solve simple puzzles Their BEST skill? Social Skills Greater degree of socialization = better skills Follow human gestures Can be fooled by humans pointing wrong direction Dogs may perform actions that are counterproductive for themselves if humans ask for that action

Sensitivity to human social cues Dogs show distinct sensitivity to human social stimuli Reliably alter behavior in presence of such stimuli to obtain reinforcement These behaviors emerge after instruction or mediation of human companion Theory of Mind and Dogs: Heyes (1998): “…an animal with a theory of mind believes that mental states play a causal role in generating behavior and infers the presence of mental states in others by observing their appearance and behavior under various circumstances”. Several sets of data point to such an ability in dogs

Dogs can….. Selectively avoid forbidden food, but grab it when the owner is not looking Beg from an individual that can see them, rather than their owner who cannot. Learn via Social learning and Imitation Watch human for cues to obtain food/toy Can be taught to imitate: “do it” Follow a human point: sensitive to Arm point Head turning Nodding Bowing Glancing in direction of target Show word learning

Lupfer-Johnson and Ross Dogs, along with just a few other species, are able to learn from conspecifics Human children Red winged blackbirds Dwarf (Siberian) but not Syrian hamsters Rats What is common element: All are social species Social behaviors important for feeding Even in dogs! Pavlov’s work showed that feeding can be conditioned Socializing while searching for food is advantageous Help one another All more likely to eat when work together Working together increases likelihood of survival for individual and the group Examined whether food preferences could be acquired from interacting with recently fed conspecifics

Method 22 dogs in boarding facility (doggie day care) 1 dog served as demonstrator for 12 total demonstrator-observer pairs All other dogs served once as either demonstrator or observer Used flavored food: basil or Thyme to dog food Procedure Demonstrator dog ate basil or thyme food in separate room Then, entered group room and allowed to interact with observer dog for 20 minutes Then observer dog offered both thyme and basil food; had to choose one to eat Food weighed to determine how much they ate of each food.

Results One way ANOVA on the data Dogs were significantly more likely to eat the flavor the demonstrator dog ate Dogs with basil demonstrators ate significantly more basil food than those with thyme demonstrators (apparently thyme is icky)

Scent Detection as Cognition Dogs have great noses, obviously Can we use this talent for human good? Bomb detection Contraband detection; drugs and fruit Disease Detection Wildlife counting? Scat detection of endangered wildlife?

Training Scat Detection Oldenburg, Schoon and Heitkonig, 2016 Malinois (that is a Belgian Shepherd) Scat from 3 groups of otters Used a radial arm maze to teach Decoy scents of fox, rabbit, hare, cattle Careful handling Criteria: 5/6 samples correct Examined generalization from one sample of otters to another (with discrimination from decoy scents)

Training Scat Detection Could the dog do it? Originally, dog could not generalize Went back and retrained After retraining, found she could do it Part of problem (as someone who does scent work) was their training procedure Did not use errorless learning Made it WAY too difficult Simon Gabois was mastered training program and Canadian Wildlife programs use dogs to find and detect presence of many endangered species

Dogs show Perspective Taking Can do perspective taking Change reaction to forbidden food (Call, et a, 2003; Tomasello, 2008) Change where drop ball depending on position of human Begging responses change depending on actions of human Attempt to communicate with humans: Move objects closer Indicate location of items Ask for help with problem Occurs as early as 8 weeks Service dogs are better! Miklosi, et al, 2003; Viranyi, et al, 2006; Topal, et al, 2006

Is your choice my Choice? Study by Prato-Previde, Marshall-Pescini and Valsecchi (Italians!). Interested in how dogs’ owners may influence how dogs choose between bigger and smaller choice Food choice is particularly strong Most dogs food driven Choose bigger (evolutionary drive, too!) But, also want to “please” their owners

Method 54 dog-owner dyads Three different tasks: Mostly pure breeds Some mixed breeds Three different tasks: Bigger smaller choice Bigger smaller choice with human pointing to smaller 1:1 choice with human pointing to a particular choice Also gave the CBARQ assessment Several subscales on aggression, excitation, separation anxiety, general fears Did not feed dogs for several hours before study

Results Dependent variable: what choice the dogs made Free choice condition 74% chose large quantity more often 18% chose smaller 8% showed no preference

Results 1:1 condition: Bigger/Smaller owners’ preference 82% chose owners choice 6% chose opposite plate 12% showed no preference Bigger/Smaller owners’ preference 32% chose larger 32% chose owner’s choice 36% chose both equally often

Other Effects Gender differences: no differences Age effects: older dogs were likely to be more accurate Training Effects: no effects Location effects: indoors better than outdoors CBARQ: dogs more likely to follow owner preference were more likely to have higher separation anxiety scores

Why choose owner’s preference? What has years of socialization selected dogs to do? Attend to owners “please” owners by obeying commands, doing what owners desire Dogs are selected to both Attend to humans Choose most food

Choice of Target when Begging (Povinelli and Eddy, 1996): Dogs were trained to beg from a human for food Offered choice of a blindfolded human or a human that could see them (for control, also a human with the blindfold over the mouth, nose, around the neck) Dogs preferred the human with no blindfold over the eyes; no difference between this an person with blindfold who could see Only chimps, bonobos also do this Povllelli, et al, 1990; Heyes, 1993 Dogs, like chimps, use human behavior for cues to food location Humans pointed, turned head or just turned eyes to look at location of hidden food Dogs could use all three cues to determine where the food was located

Expansion: NOT a Clever Hans Trick Held, et al Expansion: NOT a Clever Hans Trick Held, et al., 2001; Ashton and Cooper (in Cooper et al, 2003) Dogs could use errors as clues, as well Dogs were blindfolded or not Watched/not watched model get a hidden food Those who could watch did better Had other dogs watch the blindfolded dogs find the food Blindfolded dogs made many mistakes before found food Those dogs who watched avoided the areas that the food was not and went more directly to the final food location, avoiding the errors

Dogs show social modeling Can model other dogs Not as good as model humans Snout contact provides information (Lupfer-Johnson) Very good at modeling off of humans Action matching: Do as I do Topal, et al, 2006; Huber, et al, 2009; Range, et al, 20070

Very simple, but important, Study! Clear demonstration of dog following conspecific modeling! Claudia Fugazza and Adam Miklósi (Eötvös Loránd, University, Budapest, Hungary) Do As I Do: Using Social Learning to Train Dogs https://www.youtube.com/watch?v=cEHI2mHytNQ

Karen Overall, 2016 Dr. Overall is director of Veterinary Behavioral Medicine at Texas A&M and one of first board certified veterinary behaviorists Wants trainers and veterinarians to understand HOW memory works HOW medication and nutrition affects learning and memory in canines

Understanding memory LTP: long term potentiation When we learn, repeated stimulation of a neuron and the neural circuit results in structural changes in the neuron itself Dendritic growth Dopamine and Glutamate are critical neurotransmitters Takes repeated trials to form these circuits and get LTP Important reward and memory brain structures Amygdala Nucleus Accumbens Frontal lobes of cerebral cortex Yes, even in dogs!

Understanding memory Medications and environmental events may affect how learning and memory occurs Antidepressants and anxiolytics come in three general categories Benzodiazepines Tricyclic antidepressants SSRIs/SNRIs Yes, these are used in dogs! Affect how dogs learn and remember Can inhibit learning about “bad things”, reduce anxiety But can also inhibit reward learning Need to understand these effects as deal with dog behavior problems

Nutrition matters Several important amino acids for brain growth and learning, particularly long-chain polyunsaturated fatty acids (PUFAs) Arochidonic acid (ARA) Docosahexonoic acid (DHA) Eicosahexanoic acid (EHA) Important, if not critical, for early brain development ARA: protect cells in hippocampus from oxidative stress DHA: encourage development-stage-specific associational learning May be involved in some canine behavioral disorders E.g., Aggressive behavior in GSD: lower DHA PUFAs protect from oxidative stress, preventing overstimulation of anxiety circuits

Dogs get Dementia! Canine Cognitive Disorder Long term effects of oxidative stress Reduces ability of LTP Reduced astrocytes in brain Lower glutamate Increase in amyloid deposits (plaque) Shifts in oxygen and energy use in brain VERY similar to human dementias! Ketogenic diets and other dietary interventions may reduce canine dementia…may also help people

Why study dog cognition/behavior? Dogs are an integral part of life with humans 60% of households have a dog Estimated that Americans will spend $52.87 billion on pets in 2012 Pet spending has increased every year since it was $17 billion in 1994. More and more dogs used as assistance and therapy dogs People feel strong bond with dogs: Research suggests have same bond as a human family member

Why study dog cognition/behavior? Recent research suggests dogs may function cognitively as high or higher than the great apes (our closest genetic relative) Dog cognitive behavior appears to have “co-evolved” with similar behavior in humans Evidence of shared genes for some behaviors and disorders Disease processes are highly similar

Why study dog cognition/behavior? Research also shows that variables that affect human learning and stress affect dogs in very similar ways Understanding canine cognition, social perception and the disease processes that affect these behaviors gives us a window into human behavior! Dogs are a window into the human soul!