DTT in ABA: What Discrete Trial Training Is

Discrete trial training is the format most people picture when they picture ABA: instructor and learner at a table, the same question asked again and again, a reinforcer after the correct answers. The picture is accurate but it misses why the format exists. Discrete trial training is built for one job, which is teaching a discrimination the learner cannot yet make.

The four parts of a trial

"Discrete" means each trial is a separate, bounded event. You can point at where it starts and where it stops, which is exactly what makes the data usable. Every trial has the same skeleton:

1. The instruction. The cue that signals what to do, usually called the SD. "Touch red."
2. The prompt, if one is needed. Physical guidance, a gesture, a model, or a spoken hint. Prompts are planned to fade, not to stay.
3. The response. What the learner does, scored correct or incorrect against a definition written in advance.
4. The consequence. Reinforcement for a correct response, or a correction procedure for an incorrect one.

Then a brief pause, the inter-trial interval, and the next trial begins. That pause is what keeps trials countable.

A worked example

Target: touch the red card when asked, with a blue card also on the table.

• Instructor places red and blue cards down and says "touch red."
• Learner touches red within five seconds, no prompt needed. Scored correct.
• Instructor delivers the reinforcer and marks a plus.
• Two-second pause. Cards are swapped left to right so position is not the cue.
• Next trial.

Ten trials of that take a few minutes and produce ten data points. That density is the whole argument for DTT: a skill that would come up twice an hour in ordinary life comes up ten times in five minutes, and the learner gets ten chances to be right about it.

Note the card swap. If the red card sat on the left every time, a learner could score ten out of ten by learning "touch the left one" and never look at color at all. Controlling for that is part of running the format properly.

What discrete trial training is good at, and what it is not

Discrete trial training is strongest when a learner needs to acquire a discrimination from scratch: colors, letters, matching, imitation, early receptive language. The repetition is the point, and there is no faster way to get a learner from zero to reliable on a new distinction.

It is weakest at the thing that comes next. A learner who is perfect on "touch red" at a table may not hand you the red cup in the kitchen, because the table, the cards, the instructor and the phrasing were all part of what they learned. That gap is why programs move skills into natural environment teaching once acquisition is solid. The two formats are sequential, not opposed.

The criticism of discrete trial training worth taking seriously

Discrete trial training has a reputation problem, and some of it is earned by how the format has been used rather than by the format itself. Long tables of massed trials with reinforcers unrelated to the task can produce a learner who performs on cue and initiates nothing. Running DTT well means keeping sessions short, keeping reinforcement genuinely reinforcing, mixing acquisition targets with mastered ones so the learner is not failing constantly, and moving on as soon as the skill is acquired.

The modern position in most programs is that DTT is one tool with a specific job, not a description of what ABA is. If you are evaluating a provider and every session is table-based trials, that is worth asking about.

Why it shows up in your training

Both formats appear together in the BCBA content outline under a single task, which tells you how the field frames them: trial-based and free-operant procedures are two arrangements a behavior analyst is expected to design and evaluate, and choosing between them is part of the job. RBTs implement both, which is why the 40-hour RBT training covers discrete-trial teaching and naturalistic teaching in the same section.