ABLA-R Test: The 6 Levels and What They Predict

The ABLA-R tests six basic discrimination skills with a yellow can and red box. See each level, how it is scored, and how BCBAs use the results.

Key takeaway

The ABLA-R is a short, hands-on test of how well a learner can tell simple things apart. Its full name is the Assessment of Basic Learning Abilities, Revised.

genArete: Learner-Centered Skill Assessment
Watch the full CEU recording

genArete: Learner-Centered Skill Assessment

Mark Malady, BCBA · 1 CEU · 62 min
Watch on openceu.com →

Free learning path on this topic

Child Development for Behavior Analysts

Typical development for behavior analysts. Kristen Byra's overview, then Kelly Brzak's deep dives on ages 2 to 5, 6 to 8, and 9 to 11, with milestones, common problem behaviors, and programming tips for each band.

Start the path →4 courses · 4 CEUs · Watch at your own pace

Experts who cover this

2 recordings on openceu cover abla-r.

The ABLA-R is a short, hands-on test of how well a learner can tell simple things apart. Its full name is the Assessment of Basic Learning Abilities, Revised. It uses a yellow can, a red box, and a few small objects. In a single short session, it shows which of six basic levels a learner can pass.

Those results matter because they help predict what a learner can learn, and how fast. A learner who fails a level usually struggles with other tasks of that same type. That lets a BCBA (board certified behavior analyst) pick teaching formats the learner can succeed with. It also helps the team avoid targets that set the learner up for repeated failure.

What the ABLA-R measures#

The ABLA-R measures discrimination skills. In behavior analysis, a discrimination means responding one way to one thing and a different way to another. Picking the cup when someone says "cup" is a discrimination. So is putting a sock in the sock drawer instead of the shirt drawer.

Most teaching depends on discriminations like these. Matching, receptive labels, and following directions all need them. A receptive label means pointing to or picking the item someone names. If a learner cannot yet make a basic discrimination, harder goals built on it will stall.

The test does not measure talking, social skills, or self-care. It answers a narrower question. It shows which basic kinds of discrimination a learner can already make and which ones are not there yet.

Where the ABLA-R came from#

Kerr, Meyerson, and Flora published the original test in 1977. It was first developed with people who had severe intellectual disabilities.

Over the next decades, Garry Martin, Dickie Yu, and their colleagues in Winnipeg ran many studies on the test. Their work checked its reliability, its level order, and what it predicts. They also tested it with children with autism. That research led to the revised version, the ABLA-R. In 2011, DeWiele and colleagues from that group published a self-instructional manual for training testers.

The ABLA-R kept the same basic idea and the same simple materials. It also revised parts of the test, including some of the middle levels. Before testing anyone, check the current manual for the exact level list and steps.

Materials and the testing format#

The materials are simple and cheap. There is a yellow can and a red box, each big enough to hold a small object. There is a small, oddly shaped piece of yellow foam to place inside. Level 4 adds a small yellow cylinder and a small red cube.

Each level follows the same teaching and testing routine.

  1. The tester shows the correct response.
  2. The tester guides the learner's hand through it.
  3. The learner tries on their own.

A correct response earns praise and a small reward the learner likes. An error leads to a short correction. The tester shows the right response again, guides the learner, and then gives a new trial.

The pass rule is 8 correct responses in a row. The fail rule is 8 errors in total on that level. These fixed rules keep scoring consistent across testers. Testers usually work up one level at a time, starting at Level 1.

The original six levels, at the table#

Below are the six levels as they appeared in the original ABLA. The ABLA-R changed some of the middle levels, so use the current manual when you test.

Level 1: Imitation. There is one container on the table. The tester puts the foam piece in it, then hands the foam to the learner. The learner copies the action and puts the foam in the container.

Level 2: Position discrimination. The yellow can and red box sit side by side. The can always stays in the same spot. The learner has to put the foam in the can each time. Position alone tells the learner where to go.

Level 3: Visual discrimination. The same two containers are used. Now the tester swaps their positions at random between trials. The learner has to find the yellow can wherever it is. They must follow what the container looks like, not where it sits.

Level 4: Visual quasi-identity match-to-sample. The containers still swap places. The tester hands the learner either the yellow cylinder or the red cube. The learner puts the cylinder in the can and the cube in the box. This is a matching task. The objects look like their containers but are not exact copies, which is why it is called quasi-identity.

Level 5: Two-choice auditory discrimination. The containers stay in fixed spots. The tester says "red box" or "yellow can" in random order. The learner puts the foam in the one named. Here the only cue that changes is the spoken word.

Level 6: Auditory-visual combined discrimination. The tester again names a container. Now the containers also swap places at random. The learner has to use both what they hear and what they see. This level looks a lot like receptive labeling, such as "touch dog" when pictures move around.

Why the results predict learning#

The levels form a rough ladder. Higher levels are generally harder than lower ones. A learner who passes a high level usually passes the lower ones too.

The useful part is what a fail tells you. A learner who fails a level often has great trouble learning new tasks of that same type. They may take many trials or not learn the task at all. A learner who passes the level tends to learn similar tasks much faster.

Here are two examples. A learner who fails Level 4 is likely to find matching tasks hard. A learner who fails Level 6 is likely to struggle with "point to the ___" tasks when the choices move around.

Mark Malady teaches on learner-centered skill assessment. He names the original ABLA from Kerr and colleagues in 1977 and singles it out for its evidence.

"we see that the ABLA has really strong support. It's a very basic assessment that's used for discrimination skills."From the talk — Mark Malady

In his review of skill assessments, few tools earn that rating.

How BCBAs use ABLA-R results#

Picking teaching formats. The highest level passed shapes how to teach. A learner who passes Level 4 but not Level 6 may do best with visual matching tasks. Spoken instructions with moving choices may fail for now. Visual cues, models, and matching can carry more of the teaching.

Choosing how to offer choices. Preference assessments find out what a learner likes. They can use real items, pictures, or spoken options. ABLA-R results can help predict which of these formats a learner can actually respond to.

Avoiding failure-prone targets. Repeated failure is hard on learners. It can lead to problem behavior and less interest in teaching. If a learner fails Level 6, a long program of receptive labels from moving arrays is a risky first goal. Easier formats can build success first.

Setting realistic timelines. Results help explain why some goals move slowly. That helps with caregiver expectations and treatment planning.

Re-testing. The test is short, so it is easy to repeat. Re-testing after a stretch of teaching can show whether a learner has moved up a level.

Teams should not guess. Staff often feel sure a learner "knows" a discrimination. A quick test can confirm it or show a gap.

Limitations and critiques#

It is narrow on purpose. Six discriminations do not make a full curriculum. The ABLA-R does not cover language production, play, social skills, or daily living. It works best next to a broader assessment.

Ceiling effects. A ceiling effect means a test is too easy to show what a learner can really do. Many learners with some spoken language may pass all six levels quickly. For them, the test adds little new information.

One sitting, one set of materials. A result reflects one session. Low motivation, a weak reward, or a bad day can lower a score. Vision, hearing, and motor problems can also affect results. Testers should rule these out.

How finely the levels sort people. Malady urges care here. He notes that experts do not fully agree on how cleanly the test separates one learner from another.

"there's a lot of disagreement on how that differentiation actually plays out across people."From the talk — Mark Malady

The safest use is to treat the levels as a strong guide. Watch real learning data too, and let those data adjust the plan.

A strong tool that gets used less#

Malady makes a pointed observation about the field. The tools with the best research support are not the ones used most often.

"when we look at utilization in the field, we almost have an inverse relationship happening."From the talk — Mark Malady

Utilization means how often a tool gets used. An inverse relationship means one goes up while the other goes down. Malady explains that tools with weak or emerging research support get used more. Tools with strong support get used less. He puts the ABLA in the strong group. He contrasts it with tools like the VB-MAPP and the ABLLS-R, which are used far more but have had weaker support.

Teams that skip the ABLA-R lose a cheap, fast check on readiness. One short session can show which kinds of targets a learner is ready for. For more on picking and combining assessments, see Generate: Learner Centered Skill Assessments.

What the research says#

The original test and its levels were described by Kerr and colleagues (Kerr, N., Meyerson, L., & Flora, J. A. (1977). The measurement of motor, visual and auditory discrimination skills. Rehabilitation Psychology). Martin and Yu later summarized decades of studies on the test (Martin, G. L., & Yu, D. C. T. (2000). Overview of research on the Assessment of Basic Learning Abilities test. Journal on Developmental Disabilities). That work supports the test's reliability, its level order, and its power to predict learning.

Other studies keep testing that predictive power. They are small, so read them as early support.

One study worked with nine adults with developmental disabilities. Researchers first found each person's high and low preferred work tasks. They then offered choices three ways: the real tasks, pictures, and spoken descriptions. For five of the nine adults, ABLA results predicted whether their choices stayed consistent.

A second study worked with six people with autism who scored at ABLA-R Levels 4, 5, or 6. It trained visual matching, then checked for equivalence class formation. That means a learner relates items that were never directly taught together. Only the participants at Level 6 showed it. Those at Levels 4 and 5 did not.

A third study tested receptive language. Five people who passed Level 6 and five who failed it tried two-choice picture tasks. Each task asked them to point to a picture of a common object after hearing its name. All five who passed Level 6 passed every task. Four of the five who failed Level 6 failed every task.

Together, these studies point the same way. Passing Level 6 seems to go with readiness for receptive language work.

FAQ#

What does the ABLA-R measure? It measures six basic discrimination skills. These run from simple imitation up to picking a named container while the containers move. The results show which kinds of tasks a learner is ready to learn.

How long does the ABLA-R take? It usually fits in one short session. Learners who need more trials at a level may take longer. Because it is short, teams can repeat it after a period of teaching.

Is the ABLA-R the same as the ABLLS-R? No. The names sound alike, but the tools do different jobs. The ABLA-R is a short test of six discrimination levels. The ABLLS-R is a long skills inventory with about 544 tasks across 25 areas. Many teams use the ABLA-R to check readiness and the ABLLS-R to plan a curriculum.

Turn this topic into a CEU

You just studied this. Now get credit for it.

Watch genArete: Learner-Centered Skill Assessment with Mark Malady, BCBA and earn 1 free BCBA CEU. Detailed certificate, delivered the moment you finish.

Watch and earn the CEU →Free account · No card · detailed cert

Keep going with a free learning path

Child Development for Behavior Analysts

Typical development for behavior analysts. Kristen Byra's overview, then Kelly Brzak's deep dives on ages 2 to 5, 6 to 8, and 9 to 11, with milestones, common problem behaviors, and programming tips for each band.

Start the path →4 courses · 4 CEUs · Watch at your own pace