Re: Errorless learning & Prompts

[Guest guest]

Dear Sue,

We use " input " prompts to model correct responses when intro'ing a new

target. These could be verbal or physical prompts. We assessed which prompt

levels were the most effective and created a prompt hierarchy or a sequence

of prompts from most to least intrusive.

We also track the prompt levels by assigning a number in the hierarchy and

placing it in a column on the data sheet and this ensures that we are moving

to very clean and less directed responses quickly without boring her.

To keep her successful, we will end on a mastered and completely generalized

response. Indeed, we will often create a sequence of responses that typically

end with a response we know she will get. Many people already use mastered

NVI's like this on Lovaas programmes.

We do this to ensure we are not always ending on a prompted trial when she is

learning new concepts. I suppose that the best thing is to think of keeping

learning positive and successful - Think about what your child does well and

enjoys. I have one friend who uses animal and other noises to end tacting

tasks and another who will give a direction (receptive command) to " paint " or

another creative task she knows her son enjoys.

We do " maintenance " in this way now. So, in essence if we have a task that we

feel needs to be highly maintained we might add it to a verbal module.

Example:

Target is FUNCTIONS in RFFC and the item is cooker (stove), but we would

like to also maintain ROOM LABELS...we have done this.

SD1: Can you show me a cooker?

R1: Child indicates correctly

SD2: What do you do on a cooker?

R2: Cook

SD3: What room is a cooker in?

R3: The kitchen

You could keep this up by saying:

Go to the kitchen - Object Retrieval program

Find a cookie on the table - Prepositions/Object location

Use item as functional reinforcer

and then ask

Do you like cookies? - Yes/No

Another idea if your child enjoys puzzles is to use a mastered puzzle related

to the theme of your earlier target. So, if you were teaching animal tacts

(labels) you could end by doing a mastered animal puzzle and modelling the

animal noise playfully whilst the child completed it. The configurations are

endless and specific to the child. Keep it functional and fun are our

programme keywords - So, when programming we look for ways to use stimuli

that act as a pure SD

Example of pure SD:

An SD can be as simple as a phone ringing. The sound of the ring is an SD to

pick it up and talk. The phone NOT ringing is an SD NOT to pick it up.

How does the stimuli help reduce error in response is worth a thought. Hope

some of this helps -

UK