From Foes to Friends: Using Positive Reinforcement to Teach a Cat and Dog to Happily Coexist
Si is an approximately 6-year-old domestic short haired tabby cat. He was found as a stray when he was about a year old. At the time, our home consisted of two dogs and two other cats. Si instantly snuggled with the cats but would only observe the dogs from atop the cat trees.
After a few months, he fully settled in and did not mind the dogs, except when he was on our bed and they went toward it. Si would run to the end of it and air swat toward the dogs. Seeing that the dogs were indifferent, we didn’t do anything about the situation.
Jeter came to us when he was approximately 8 years old. A Chihuahua/terrier mix weighing 30 pounds, he was visually impaired, plagued by separation anxiety, extremely nervous of people, and his history was unknown.
Jeter was kept separated from the main part of the household when he arrived, and slowly exposed to the environment and various stimuli over a two-week period. He would stay in his safe place, a room of his own with a baby gate in the doorway, while we worked on his separation anxiety. Our other two cats had since passed away, and our other dogs were kept on a different level of the house in order to decrease environmental stimuli during this time.
Prior to seeing Si, Jeter was introduced to a fake, realistic-looking stuffed cat at a distance. He approached the stuffed cat with loose body language and a play bow. Because of his calm and affiliative response to the stuffed cat, we were a little surprised to find that, from Day 1, Jeter was extremely focused on Si, though non-aggressively. Despite the intensity of his focus, his body language toward Si was similar to what he displayed toward the stuffed cat. If Jeter had charged the stuffed cat, stared at it nose to nose, remained still with tense body language, or attacked it, these behaviors would have been an indication that he could be aggressive toward Si.
Si was also extremely skittish and hesitant to even walk by Jeter’s room. Knowing the importance of having a home where they could both enjoy each other’s company, I laid out our goals for our family and then established a management and training plan.
The ultimate goal was for Jeter and Si to enjoy each other’s company and be able to relax while the other was in view. In order to achieve this, both pets needed to reach their own milestones. Jeter needed to develop a strong behavior in place of chasing Si. Teaching an animal to perform another behavior instead of something we don’t want them to do is known as “differential reinforcement of an alternate behavior,” or DRA.
My plan for Jeter actually involved three behaviors that together made up “relaxing on a mat”:
- Seeing Si
- Hold a stay on his mat
- Looking away from Si
Developing an effective behavior change plan for Si would prove to be more of a challenge. This is because we needed to change his emotions about Jeter from negative to a positive through creating a positive conditioned emotional response (+CER). Eventually, the goal would be for Si to see Jeter and feel happy due to Jeter’s association with reinforcers. We would then fade off of the reinforcers while maintaining the +CER. Some behaviors indicating a +CER included Si rolling on his side, walking up to Jeter and rubbing up against him, going to Jeter and lying down close by, purring with relaxed body language, and not looking for a quick escape.
I did have an advantage when working with these two. Si had already been trained using a clicker, and knew various tasks such as offering his paws for nail trims and nose targeting items. Jeter knew the hand signals for sit and down, was very food motivated, and was an extremely quick learner. I had high hopes that our goal would be obtainable.
Jeter’s two weeks of separation when he arrived at our home included no physical contact with Si, nor either of the other dogs. He was able to see them in passing at first, and we slowly increased their exposure to each other over the two-week timeframe. This is advice that I have given to my own clients many times when helping them transition a newly adopted dog into their home. Just like many of my clients, I found myself wanting to speed up the process.
The great thing about networking with other trainers is that we are there to support each other with cases and with our own pets. Even though Jeter was good with other dogs, Rich Allen from Wags to Rich’s encouraged me to stick to the training plan, and every good training plan starts with managing the environment so unwanted behaviors cannot be practiced.
Three baby gates were put in place. The first one was at the entrance to Jeter’s room. The second was at the entrance to a spare bedroom. This was so Si could easily access a room to get away from Jeter, but Jeter would not be able to follow. This is also where Si’s cat box, water, and meals were placed. The third gate was at the top of the stairway. This enabled me to have a door and a gate as a barrier to the downstairs where the other dogs stayed.
The door to Jeter’s room was left open when we were training, and closed when we were not. This was to prevent Jeter from barking and running at Si as he passed by the baby gate. I used a harness and leash for Jeter when he was out of his room to protect his neck if he pulled, and to prevent him from chasing Si. When I’m working with an animal on a DRA plan, I make sure they are not able to practice any of the behaviors that I’m trying to develop a DRA for—in Jeter’s case that was charging and barking.
I also wanted to make sure that I was able to work with Si every time he saw Jeter. I have found that an animal I’m working with can take a long time to develop a +CER if reinforcement is only offered some of the times they can see the thing I’m counter-conditioning to, compared to if I offer the animal consistent access to reinforcement all the times the “scary thing” is present. This is especially important in the beginning stages of training.
Training for Jeter
The beginning of Jeter’s first week with us consisted of him learning foundation behaviors, which included a verbal marker of “yes,” recognizing and immediately responding to his name, a cue of “leave it,” so I would be able to tell him to leave Si alone, and mat work where he would go to his mat and hold a stay with minimal distractions.
While we worked on the foundation behaviors, Si was not in view. It was important to break down the tasks into easy and fun sessions with little distractions.
When I found that Jeter was responding to my cues most of the time and understanding the activities, I started to work Jeter and Si together. Jeter was positioned on his mat and Si was free to roam of the rest of the upstairs. Every time Jeter looked at Si as he peeked around the corner, I immediately used my verbal marker of “yes.” This prompted Jeter to then look back to me for a treat. After taking the treat, his head immediately oriented back toward Si and I said “yes” again. Instead of distracting Jeter, this activity was to teach Jeter that he could look at Si, disengage from him, and then check in with me. We did this activity for about a minute, making sure that Jeter was staying on his mat and quickly responding to me.
Si would tuck his head back out of view, and I would then release Jeter to get off of his mat. As soon as Jeter saw Si reappear, I used my verbal marker, rewarded, and then sent him to his mat. Because Jeter was a very fast learner, I was able to quickly move to the second step: allowing Jeter to look at Si for up to five seconds without using a marker, and having him choose what he should do.
Sure enough, Jeter looked back to me in anticipation of a treat. This process is called “scans and check-ins.” He was scanning the environment and seeing Si, then choosing to check in with me without my needing to prompt him by saying “yes” as soon as he looked at Si. Instead, the “yes” was withheld until Jeter completed checking in with me. I then placed a special treat on his mat. At this time Si started moving around more.
Training for Si
Positioned on the outside of the gate, Si was allowed to control the situation by having the ability to increase or decrease the distance between himself and Jeter as he pleased. I did this so I did not accidentally cause Si to have a fight or flight reaction where he felt the need to attack Jeter to defend himself or run away to seek safety.
I made sure to give Si special treats every single time he looked at Jeter. Good timing as I tossed him treats was critical—the food needed to come after he looked at Jeter, so he could learn to associate Jeter with rewards. If I had given Si food before he looked at Jeter, I could inadvertently teach him that the presence of food was bad, because it meant that Jeter, who he was still afraid of at this point, would be appearing.
With the food aiding in creating a +CER, the training plan was well on its way.
Jeter was able to complete his activities with ease, and Si was becoming more relaxed. I decided to move our sessions to my bedroom and bring Jeter’s mat out with him. I placed the mat across from the bed at about 5 feet away, and I asked Jeter to go to and stay on it. Jeter was kept on leash and Si was able to access my bed, the dresser, the night stand, and one exit into the hallway.
Jeter had been successfully looking at Si and disengaging on his own when in his room. Because I had made the activity harder by moving to another room, I took a step back with Jeter’s training so I did not expect too much from him. I marked with “yes!” as soon as he looked at Si. When Jeter looked back to me, I placed a treat on the mat at his side so he was eating it as he looked toward me.
If Si made a quick movement, I rewarded Jeter with a jackpot—a handful of small treats on his mat. After one session that lasted approximately five minutes, I moved on to marking the behavior of looking away from Si, at me, and reinforced that choice to disengage.
Over the next week, I worked with Si and Jeter together daily for sessions that lasted between five and 15 minutes. I used their meals as opportunities to work and hand-fed them a few pieces at a time. Si’s comfort increased and he was also reinforced for any relaxation that he showed. The behaviors that he voluntarily offered that indicated relaxation included sitting, lying down, and rolling on his side while purring.
I made sure to end the sessions before Si decided that he had had enough of the training and walked away. I wanted to leave him wanting more so he was eager to train the next time. I continued to work with Jeter on his foundation behaviors in other rooms so he would be fluent with them no matter where he was.
By the beginning of the third week since arriving at our home, Jeter was spending more time out of his room. He stayed on leash and chose my bed as one of his favorite places to relax. One evening, Si entered the room and jumped on the bed without realizing Jeter was there—the two were suddenly nose to nose! I immediately gave Jeter the “leave it” cue, which he followed.
Si jumped off the bed and disappeared into another room, and Jeter immediately received a jackpot.
I thought a disaster had been averted until I went to sleep that night and discovered cat urine soaked through my sheets…
I removed all sheets from my bed unless I was sleeping in it. Next to the bed I placed an additional cat box with a top entry so Jeter could not stick his head in. A second cat tree was purchased and put within jumping range from my bed. My hope was that the modifications that I made to the environment would help Si work through the setback. I believe that they did, as Si did not pee on the bed again.
Christmas was approaching and I found wrapping presents to be the perfect opportunity for training. With Jeter on his mat, Si on the bed, and me stationed between them, Si began to decrease the distance and come closer to where I was. Jeter handled staying on his mat like a pro, and I finally started to see the playful side of Si as he batted around the bows. I began reducing the amount of rewards each pet got in their training sessions. Instead of being rewarded each time they performed a behavior that I liked, I moved on to a variable-ratio reinforcement schedule, where neither pet would know exactly how many times they needed to offer behaviors before being rewarded.
Finally, by the end of week three, the early signs of a +CER from Si started to show. Jeter had entered the room with a toy in his mouth and trotted past Si. I looked to Si and saw him go from a sit to a roll on his side. When Jeter lay down a few feet from Si, Si proceed to bat Jeter’s tail around in a playful manner as he purred.
I then allowed Jeter to start to drag his leash around instead of my holding it, and soon I removed the leash altogether. When Jeter would approach the bed as Si relaxed on it, Si did not charge the end of it like he did with the other dogs. At this time, I felt comfortable enough to put sheets back on my bed.
The turning point
Jeter went for emergency exploratory surgery in early January of 2018. While recovering on my bed, Si displayed a full +CER when he jumped next to Jeter and relaxed just a few feet away. Si rolled on his side, closed his eyes, and began to purr. Si continued to do this even when Jeter moved around slightly.
Over the next few days, Si got closer and closer to Jeter.
The two continued to bond as Jeter healed. Once Jeter had fully healed, Si remained relaxed no matter how close Jeter was to him, and they could both approach each other without Si fleeing and Jeter pursuing. I then decided to place pet steps next to the bed, so Jeter could easily access it at any time.
Fast forward four months
Si still has a baby gate that blocks off Jeter’s access to the spare bedroom so that he can have a safe place to escape to if he feels the need for it. A second cat box remains upstairs, and the cat tree is positioned within jumping distance from the bed.
The training process was slow, and our management plan was strict. I believe that if it was not for the care I took to go at the right pace in a methodical manner, instead of rushing things or trying a lot of different approaches and just hoping for the best, Jeter would still be chasing Si, and Si would still be peeing on my bed. Today, Si and Jeter can be found regularly napping and playing together.
What is even better to see is that Si is the one who most often seeks out Jeter for his company.
Tori Ganino CDBC, CPDT-KA, is a Certified Dog Behavior Consultant through the IAABC, Certified Professional Dog Trainer-Knowledge Assessed through the CCPDT, and a member of the International Companion Animal Network (ICAN). She owns Calling All Dogs located in Batavia, New York, where she teaches group classes and private lessons for obedience and behavior modification. More information about her services can be found at www.CallingAllDogsNY.com.