English
中文Introduction
The concept of learning refers to any processes which result in either permanent storage of new information or modification of previously stored information (Schweizer and Koch 2001). Different learning tasks require various demands, e.g. in a canine, the use of a litter tray as observed in conditioning may require the formation of a new association. Animals with central nervous systems such as dogs can learn through more complex processes that allow them to adapt to a larger variety of environmental circumstances. The main types of learning of this nature include classical conditioning, operant conditioning, imitation, and imprinting. Durkovic (1975) and Cammaerts (2004) have shown that other types of learning including sensitization and habituation are generally regarded as simple forms of learning however this can also be observed in unicellular organisms. Similarly learning by association follows the same mechanisms as habituation where the animal establishes a link between behaviour and a stimulus or a particular situation. Kolb has defined Experiential Learning Theory as an interactive process which involves four clear learning modes which form two bipolar dimensions of learning (Metallidou and Platsidou 2008). However this theory is limited in animal training due to the fact that it relies on active experimentation and reflective observation. In animal training these two facilities are unavailable and it has yet to be proven that animals have the ability to reflect.
Classical Conditioning
Classical conditioning also called Pavlovian conditioning involves the creation of a conditioned reflex (Buser 2006). In his most famous experiment, a neutral stimulus (S), in this case a bell is rung just before meat powder (unconditioned stimulus US) is squirted into a dog’s mouth. The powder produces the reflexive response of saliva production (unconditioned response or UR) which was measured through a cannula inserted in the dog’s mouth as seen in figure 1 below. If the bell is rung and shortly followed by the meat powder many times in succession, however with a rest period between presentations, in due course salivation will occur to the sound of the bell before the meat powder is delivered. The bell is now a conditioned stimulus (CS), and saliva production to the bell is now called a conditioned response (CR). This will result in the formation of a new conditioned reflex. According to Bitterman (2006), the conditioned response will be sustained as long as the ringing of the bell continues to be linked with the presentation of the meat powder. As in this example, the conditioned response is adaptive because it prepared the dog for the impending unconditioned stimulus.
This type of conditioning is seen in animal training be they domestic or show but with the use of a clicker or a whistle. Animal trainers refer to this as a bridge due to the fact that it bridges the time between when the animal performs a desired behaviour and when it gets its reward. A clicker can also be used for classical conditioning by clicking it and delivering some desirable treat, many times in a row. In such cases the clicker becomes the conditioned stimulus (CS).
In a real life situation, classic or Pavlovian conditioning can also be developed without intent. It has been observed by personal experience with a dog that when a rain coat is taken off the peg (S) the dog gets excited (UR) as he associates this with going for a walk (US) whilst the opening of a refrigerator door can be associated with presentation of dog food in a similar manner. In the first examples the coat becomes the conditioned stimulus whereas the excitation of the dog becomes the conditioned response. Similarly in the second example opening the refrigerator becomes the (S) and associated feeding becomes the (US), which leads to salivating (UR) and the tin of food becomes the(CS). Ultimately the salivating becomes the (CR).
Operant Conditioning
Whilst classical conditioning forms an association between two stimuli, operant conditioning forms an association between certain behaviour and the consequence (Cammaerts 2004). Kirsch et al. (2004) have referred to this as response-stimulus (RS) conditioning. This is due to the formation of an association between the animal’s response or behaviour and the stimulus that follows, which is the consequence.
To attain operant conditioning in a canine, it must be placed in a situation allowing it to give a response whenever it wants. Each time it responds it receives a reward. Skinner (1938) showed that after a certain time the animal can be observed responding in that situation if it ‘wants’ to be rewarded. For the animal, the consequence has to be immediate and this can be undertaken by using a bridge as mentioned above.
As operant conditioning relies on response to a behaviour, there are four possible consequences to any behaviour as shown in table 1 below.
Litter training for domestic animals can be carried out by utilizing this type of learning by providing positive punishment if the litter tray is not used e.g. a tap on the nose. However negative re-enforcement can be applied once the litter training is complete e.g. the tap on the nose is removed. Training a domestic dog to roll over can be done by positive reinforcement whereby a treat can be presented if the dog rolls over as directed. Each time the dog rolls over a treat is presented.
Therefore it can be summarized that pleasant events increase or reinforce voluntary (operant) behaviour, and unpleasant events weaken or punish operant behaviour. New behaviour can be created through operant conditioning and is referred to as shaping, or the reinforcement of successive approximations of a target behaviour (Cardinal et al. 2003). For example, a young dog can learn to roll over if provided with food and praise (the reinforcement).
Imitation and Imprinting
Many species including young dogs also learn through imitation. In general, it is a fast and efficient way of learning functional new behaviours. For example a puppy will imitate its mother, in such tasks as feeding, using the litter tray and sleeping in a certain location.
Imprinting is the development of an attachment to the mother or a moving object close by during a certain brief period in the life of a young animal. This behaviour can be observed in many mammal species such as sheep, deer, and dogs. Young puppies follow their owners around in a similar manner.
There is disagreement over whether imitation and imprinting are special cases of operant conditioning or a unique type of learning.
Speed of Learning
Many different studies have engaged a number of task types to investigate the relationship between processing speed and intelligence, (Neubauer 1997; Roberts and Stankov 1999). The main advantage of a high processing speed as discussed by Neubauer (1997) implies that it increases the prospect of temporarily stored information to be processed suitably before decay occurs.
Conclusion
In conclusion, learning produces a relatively lasting change in behaviour as a result of experience, thus allowing animals to adapt and cope with variable environments. Animals with central nervous systems such as dogs can learn through more complex processes that allow them to adapt to a larger variety of environmental circumstances. The main types of learning of this nature include classical conditioning, operant conditioning, imitation, and imprinting.
