- Pets and Animals»
- Dogs & Dog Breeds
Using Jackpots of Treats in Dog Training
What exactly are jackpots in dog training and how can they benefit your dog? Imagine going gambling at your favorite casino one evening and playing the slot machine. You insert a coin and press the button on the front panel or pull down the lever and watch the reels spin with high hopes. Lemons, cherries and bells flash quickly before your eyes. The reels soon start spinning more slowly as you hold your breath hoping for a winning combination. Unfortunately you do not win. You try again and again, until at some point you hear bells ringing and music playing as flashing lights surround you. After a few seconds of disbelief, your heart starts pumping faster as you realize you have just won the jackpot! The attendant stops by, ensures the machine is in good working order and congratulates you on the win. The win certainly caught you off guard and was a pleasant surprise wasn't it?
In a similar fashion, in dog training you can recreate a similar range of emotions where your dog will be caught off guard by a pleasant jackpot, but in this case you'll be using doggy currency-- in other words, treats! Giving jackpots in dog training can seem like the perfect way to reward those exceptional performances of behavior that deserve to leave a lasting impression that will have an impact on your dog. Giving a jackpot of treats may seem fairly easy, but there are some common misconceptions and mistakes dog owners and trainers make which may have an impact on the outcome of training. Let's take a critical view on jackpots and see what the experts say about them and take a glimpse into some studies.
How to Use Jackpots in Dog Training
As many other training techniques, the use of jackpots is subject to different interpretations. Some consider a jackpot as giving a series of treats one-by-one, others consider jackpots as access to a pile of treats and some others think that jackpots simply consist of the delivery of higher value treats.
A friend of mine who was getting ready to train her dog for competition obedience once told me that every time her dog sat quickly, she would deliver a jackpot of treats. Her interpretation of giving a jackpot of treats was giving several treats all at once. Her intent was to reward the fast sits with a remarkable reward that would have her dog sitting faster and with more frequently. As understandable as her approach was, she seemed to be missing a fundamental point: The notion that jackpots are used to reward those exceptional performances of behavior. What my friend was doing by rewarding fast sits over and over was simply feeding more treats, she was not giving jackpots, or at least, perhaps to her dog they felt like jackpots the first or second time, but giving it over and over again, for a trained behavior doesn't fit the exact definition of a jackpot. Indeed, according to Karen Pryor, "a true jackpot is delivered all at once and contingently, to both mark and strongly reinforce the first occurrence of a rare behavior, or the first achieving of a difficult move."
Notice how Karen Pryor uses the the term "mark", this means you give the jackpot the moment the desired behavior unfolds, not afterwards. Many make the mistake of using jackpots as a reward, that is, they mark the behavior with a verbal marker or a clicker, and then give access to the jackpot of treats. Karen Pryor seems to frown on this stating when it comes to the ability of reinforcing behavior, there's really no difference between giving one treat or an especially large, numerous, or a wonderful high-value treat. Yes, the high value or large treats may make the dog more interested in the training and may keep the clicker nice and strong, but what really provides information and strengthens a behavior over another is ultimately the clicker. A jackpot given after the click doesn't provide information on the behavior performed, it just tells the dog "sometimes my owner gives me more treats than usual." Therefore, it's not surprising when Karen Pryor mentions how people who reward one behavior with a click followed by a small amount of treats, and then reward another behavior with the click followed by a large amount of treats usually see no discernible differences between the results. Karen Pryor explains that this happens because they are really not using the jackpot as an event marker, but more as a reward.
Another important distinction Karen Pryor points out is differentiating jackpots from non-contingent rewards. While jackpots are used to mark an exceptional performance, non-contingent rewards are not associated with any particular behavior. If you are dealing with a very shy dog that is very tentative in trying new behaviors and is undergoing a moment of discouragement, you may find that giving several, good tasting treats may perk her up and get her going again. But in this case, you're not giving a jackpot as you're not marking any particular behavior, what you're really doing is giving a non-contingent reward.
What effect does the jackpot have on the dog? Get ready for another roller coaster ride as this is also a subject of controversy. You would expect it to reinforce behavior causing the dog to want to repeat it. Gail Fisher in her book: The Thinking Dog: Cross Over to Clicker Training explains that when it comes to training a new behavior, it's rare that a jackpot would work in having the dog repeat the jackpot earning behavior. British dog trainer Elizabeth Kershaw studied jackpots and found that the longer it took for the dog to eat the treats, the more likely the association between behavior and reward was lost. When training a new behavior, rapid rates of reinforcement where one treat was delivered quickly to reward the behavior seemed to work better. A possible solution to shorten the time span between behavior and reward is to stop doling out a series of treats one-by-one and rather allowing the dog to wolf down treats from a stuffed treat bag or opening your hand and unraveling a pile of treats. Your dog should gulp these down quicker than eating them one-by-one.
On the other hand, giving a series of treats one-by-one is often suggested by several trainers when training recalls. The purpose is to keep the dog with you longer, versus a dog who gets his cookie and then just dissipates in thin air seconds later. Pamela Dennison, in her book "The Complete Idiot's Guide to Positive Dog Training" explains that to a dog a wad of food is the same as one cookie, so spreading treats out when training the recall is helpful. To heavily reinforce a recall she suggests giving out 10 to 30 treats.
There are several other circumstances where a jackpot can turn handy. For instance, you may find it helpful giving a jackpot when your dog gets over a slump that seemed to have lasted forever, suggests Patricia McConnell. Don't fall into the mistake though of overdoing jackpots! Getting too many jackpots may cause treats to lose value, just as you would no longer be impressed if you got a pay raise every week. On top of that, consider that too many jackpots may at the same time cause disappointment when you'll deliver regular reinforcement.
As seen, different trainers have different approaches when it comes to delivering jackpots, but what do studies have to say? If you're a science junkie, you may enjoy reading this whole PDF file discussing the effects of jackpots on behavior. Interestingly, there seems to really be a connection between increasing duration in giving treats and seeing a decrease in behavior rather than an increase. According to Belke (1997) "when reinforcer duration was increased, rates of both lever pressing and running decreased." Also most seem to agree on not overdoing it to prevent satiation. Burch & Bailey (1999) theorize that a jackpot may cause decreased responses and even a diminishing effects on standard reinforcers. When it comes to the quality of reinforcers, flavor also seems to have an impact. It appears that repeated exposure to the same food can lead to a decrease in performance possibly due to habituation.
Overall, the study is quite disappointing when it comes to evaluating the efficacy of jackpots. It seems to suggest that they aren't effect as we hoped for. But there are perhaps too many variables to draw a good conclusion. Perhaps what trainer Melissa Alexander mentions goes a long way "Jackpots make the giver feel good, but they interrupt the flow of training and focus the dog on the food, rather than the task. " She also mentions that participants on the OC-Assist-Dogs mailing list suggest giving a regular treat coupled with effusive praise rather than changing the amount or type of treat. "Overall, it's clarity of criteria and a consistently high rate of reinforcement that leads to a solid behavior." But then again it's worth pondering if these people actually used the jackpot as an event marker or more as a reward, and if they were giving the treats one-by-one falling into the duration trap as mentioned above (Belke).
Yet,it can't be ignored that many trainers still find the use of jackpots helpful, if so, it means that they must be working otherwise they would be useless.What are your thoughts?