A secondary reinforcer is an event or action that predicts a primary reinforcer. A click is a predictor, very often, of food. I have seen many dogs begin to salivate where they hear the click, and lick their lips.

If the click is used to mark the onset of play, the dog will become quite aroused when hearing the click. There is an emotional connection between the marker and the subsequent reinforcer.

For any secondary reinforcer, the connection can only be maintained through regular use, if the clicker is used and no reinforcer follows the effect of the click will diminish.

When training puppies I whistle as I take their food to their pen, from 4 weeks. Initially they don’t look up but begin to scan the floor for the food, after about 5-7 days they look up for their individual meals or treats. At any time I can whistle to arouse them and they will run to the source of the whistle. This is always reinforced with food when they arrive, and at 6 weeks their mouths start to “nibble” in anticipation of the meal when they hear the whistle.

The pairing between the marker and the reinforcer, or punisher, can happen between any predictable (and unpredictable) event and any reinforcer. It very often happens without our realising it. As a training tool it is invaluable to be able to reinforce away from the behaviour. If the dog has moved or jumped in a way I want to reinforce, I can click, and give the food when the dog has moved to a new or appropriate location. When the Gordon turns towards you on the whistle – you click, interrupting their browsing is a seriously good achievement, and the reinforcer can be delivered where it suits. In the case of a recall over 100 yards, the marker can be delivered for leaving the rabbit and when the dog looks at you, you can visually begin to reinforce through getting the treat bag out, rummaging around and by the time the dog arrives have chosen the best sausage. If the dog is familiar with your routine of selecting sausages, then as they see you begin the pattern, the behaviour of moving towards you will be being reinforced.

When training my dogs in a group I use eye contact as the marker. I give the cue for a sit, as each dog arrives and sits, I look directly at them and follow with the food. If we are walking on lead, the dog that moves to my side gets marker with a touch, and the walk sets off. We may be standing watching some activity. If my dog remains calmly at my side, with no need to commentate, another touch, followed by food or affection.

Secondary reinforcers are often referred to as “bridges”, between the behaviour and the reinforcer, and can work similarly with punishers.

Learned Reinforcers

Additionally, new reinforcers can be built by pairing with primary reinforcers. Some dogs find stroking difficult to accept, but when paired with the delivery of food, it can become a pleasant, and eventually reinforcing activity. To counter the punishing effect of stroking will depend on to what degree the stroking is unpleasant. It will be a strong reinforcer, used on many hundreds of occasions to make a difference. In some cases it will need pairing for life, since in the absence of the primary reinforcer the dog moves away from the touch. It would suggest that the activity is a primary punisher.

But we can associate many different activities with primary reinforcers to enable easy interaction with the dogs. We are inclined to use the term “good boy” or “good girl” to reward the dog, but unless the way you use that expression has been paired with a primary reinforcer it is meaningless to the dog. It may make us feel better as we believe we have rewarded the dog, but unless the behaviour increases, we haven’t.

Unfortunately we accidentally pair many incorrect behaviours with correct behaviours. The dog jumps up to greet us and then sits, and the sit gets reinforced by our response. If the jump-up keeps repeating itself, it has been reinforced in some way, quite possibly because the successful “sit” originated in the jumping up. Sits that orginate in not jumping up must also be reinforced. We are faulty for only noticing the correct behaviour after an undesired behaviour has occurred.

If the unwanted behaviour continues, it is getting reinforced, your intention is irrelevant. In the situation of this incorrect/correct pattern the whole pattern is reinforced.

This is oftenly seen with “yo-yo” walking on a lead. The dog goes forward to the end of the lead, may even begin to pull, the walk stops, the dog returns to the desired position and gets reinforcement: either the walk continues or a treat is delivered. If this is continually re-occurring then the undesired/desired pattern has been set.

The first few times this occurs to explain to the dog the behaviour we desire is sufficient, from that point forward the dog should return to the correct position, maintain it for a short period and then be reinforced. We are seeking to teach the dog to maintain this position, not just pass through occasionally. It needs to be maintained before walking together commences.