When teaching skills, whether these are new skills or maintenance and generalisation of mastered targets, it is essential that the reinforcement level is correct to keep the student motivated and learning. Reinforcement happens when a consequence is applied after a behaviour that cause that behaviour to increase in frequency in future.
Many people believe that Applied Behaviour Analysis programs involve delivering reinforcement in the form of small pieces of edibles or other preferred items for each skill that the child is asked to do. Many people cite this as an issue with ABA as they believe children should want to participate in learning for social praise alone. For many children with autism, initially social praise does not actually function as a reinforcer, it does not increase the target behaviour. However ABA programs should look to use socially acceptable reinforcing items and also look to fade reinforcement to appropriate levels. Examples of fading of reinforcement can include using a token economy.
At Skybound, we always try to use appropriate reinforcement and ideally like to use reinforcers that are naturally motivating within activities. However, for some children, the use of additional, extrinsic reinforcers are needed. We particularly like the Early Start Denver Model (ESDM) decision tree to help us think about the different levels of reinforcement. This decision tree is designed to be used for when a child is not making adequate progress within the ESDM program and includes looking at the instructional programs and the level of reinforcement. We feel that the decision tree can also be advantageous to consider generally for ABA programs.
When considering reinforcement it suggests using the following sequence of reinforcement, ideally using level 1, however if progress is not seen, moving up through the levels may be needed.
1. The natural activity with added social reinforcement – Add social praise such as cheers and high fives for completing parts of the activity. Or sing a song that the student enjoys as you play, these can also be silly songs that you make up on the spot to a tune that the student enjoys.
2. Extrinsic reinforcement but related to the activity and social reinforcement – bring in other reinforcing items that are related to the activity. For example, adding glitter to a painting or using a massaging bug toy to tickle a child while singing incy wincy spider.
3. Reinforcing toys not related to the activity and social reinforcement – use other toys to pair the activity. For example, running cars through paint and creating patterns using the wheels while painting or using a favourite doll in the swimming pool.
4. Non-social toys such as electronics and social reinforcement – Electronic toys such as iPads are harder to incorporate into activities but can be used to contrive motivation. Token economy systems can also be used during NET activities where the “non-social” toys are the terminal reinforcer.
5. Edibles and social reinforcement – Edibles are also a way of contriving reinforcement and small reinforcing edibles such as raisins or nuts can be delivered quickly during an activity without having to interrupt the play too much as they are consumed quickly.
It is important that social reinforcement is paired with the delivery of any extrinsic reinforcement in order to pair the social reinforcement with existing preferred items.
For extra reading please see “Early Start Denver Model for Young Children with Autism” by Sally Rogers and Geraldine Dawson.