Abstract (eng)
Clicker training is a training method based on positive reinforcement that can be used in an optimal and neutral way (independent of the person). In this technique, the animal is first conditioned to the "click" with positive reinforcement (usually food reward as the primary reinforcer), followed by marking the display of a desired behavior with this "click" (secondary reinforcer). To maintain the positive association of the click, the animal is rewarded immediately after the click, for example with food. Opinions differ on the right way to reward in this context: Is it more useful to feed continuously or occasionally after the click? In the following study, a review of the basics of the conditioning of behavior was first written in the form of a literature review with focus on the advantages of intermittent or continuous reward. Subsequently, it was analyzed which method of reward is used more frequently in horse training. For this purpose, 50 YouTube videos were evaluated regarding the time of administration of the reward in connection with the click. The evaluation of 50 videos showed that in only one case the reinforcement method was irregular. So, the strategy of continuous reward (food after each click) has become very popular. According to the author, the training went well and without major errors in 84 % of the videos. In 8 % of the videos, mistakes were made in training by the trainer (for example clicked at the wrong moment). In another 8 % of the videos, the training was neither good nor bad, according to the author. Thus, the reward schedule is only one aspect in the clicker training. The general correct use of the clicker (e.g., timing) is very important, too. The literature shows that, in general, continuous reinforcement leads to rapid learning success. In contrast, occasional reinforcement shows a slower learning curve, but what is learned is retained longer without rewards. Basically, both strategies are successful. Possible reasons for the proliferation of continuous reinforcement after each click were dis-cussed at the end. According to the knowledge of learning physiology a continuous reward does not seem to be necessary, especially if the animal already knows its tasks. There may be other arguments that favor continuous reinforcement. A new study examined how regular or irregular reinforcement affects the well-being of the animal. It was assumed that the animal expects the continuous reward (the click is the announcement of the reward) and that omission of the reinforcer resembles negative punishment and leads to frustration. In fact, irregular reinforced dogs were less motivated. The personality of the animal had an influence on its stress resistance. Therefore, solely considering the learning psychology of the animals does not seem to be sufficient. Aspects of applicability should also be considered. The procedure of always giving food after the click is easy to convey to trainers. It is also conceivable that the trainer considers it unfair not to always reward the animal when it has performed well. These aspects, especially that of animal welfare, justify the use of regular reinforcement.