Introduction
This study examines whether repeated assessment probes can teach vocabulary on their own, or whether mixed-operant instruction is needed to build true bidirectional naming skills. The findings have direct implications for how clinicians interpret progress data and design treatment plans. Understanding when probing itself may be doing the teaching helps prevent misattributing gains to other interventions.
What is the research question being asked and why does it matter?
This study asked a straightforward but important question: Can repeated probes alone help a child learn new words bidirectionally, or is mixed-operant instruction (MOI) required? Bidirectional learning means the child can both point to an item when they hear its name (listener skill) and say the name when they see the item (speaker skill). If probing alone builds these skills, then what we call “assessment” may actually function as teaching—and that matters for treatment planning and progress monitoring.
This is relevant to daily ABA practice because many children with autism or language delays do not pick up new vocabulary just from hearing adults label things. Clinicians often hope for “incidental learning,” where a child hears a label during play and later understands and uses it without direct teaching. If MOI works better than probing, clinicians should prioritize it when this goal matters. If probing itself builds skills, clinicians need to avoid assuming an intervention worked when the real change came from repeated probes.
The study also examined whether probe order matters. Some probes tested listener skills first (pointing) and then speaker skills (tacts), while others tested speaker first. This matters because probe order might inadvertently help the child by providing extra chances to hear the word right before testing production.
What did the researchers do to answer that question?
Nine boys ages 3–6 with autism or language delays participated. All had basic prerequisite skills: matching pictures, some echoic ability, and some existing vocabulary. The team used sets of five new picture cards at a time. For each set, they first conducted a “naming experience” using a matching task where the adult said the name while the child matched pictures, with correct matching reinforced. Afterward, they ran probe blocks under extinction to test three skills separately: pointing to the named picture (listener), saying the name when seeing the picture (tact), and saying the name when asked “What is this?” (manded tact).
Some children received only a short baseline with 2–3 rounds of naming experience plus probes before moving to MOI. Three children received an extended baseline with up to five rounds before MOI was added, allowing researchers to observe what repeated probing alone might produce. MOI taught several operants in rapid rotation across trials (matching with echoic, tact, manded tact, and listener responding), with reinforcement and prompting. After MOI, they re-ran the same probes with the original baseline sets.
They also conducted a “generative” probe with a brand-new set to test whether skills carried over, and a one-month follow-up probe (for 8 of 9 children) to assess maintenance. A key limitation: the probe format included a teaching-like naming experience right before testing, so “probing” was not a clean test of pure incidental learning.
How you can use this in your day-to-day clinical practice
If you use Inc-BiN probes in clinic, treat them as a possible teaching event—not a neutral test. In this study, the probe package always included a reinforced matching “naming experience” right before the extinction probes. Your “assessment” can change the skill you are measuring, especially with repetition over days or weeks. If you probe often, you may be building listener and speaker responses through exposure plus the child’s echoic behavior, even without directly teaching those targets.
When you want to know if an intervention truly worked, control your exposure. Do fewer repeated probes with the same stimuli, or rotate to new stimuli more often to separate practice effects from real treatment effects. If you must repeat probes, factor this into your interpretation. If a child shows growth after several probe sessions, do not automatically credit a new procedure you started right before the jump. Consider whether the child had multiple recent chances to hear the labels during the naming experience and probe instructions.
Repeated probing alone helped some children, but not all, and progress took time. Two children in the extended baseline reached the study’s emergence criterion without MOI, while others still needed it. For clinical decision-making, a reasonable “wait and watch” period may be appropriate for some learners if goals and risks fit the case. If a child shows steady improvement across repeated probe opportunities, you might continue exposure-based sessions briefly while monitoring the trend, rather than switching plans after one or two low probe scores. But do not stretch this too far—in the study, they capped baseline probes for ethical reasons, and some children still did not reach criterion without MOI.
MOI appeared useful for building the full package of skills with trained sets. After MOI, children showed much higher listener and speaker responding on probes with the original baseline stimuli. MOI is a good option when your goal is to link listening and speaking across the same items and you want faster progress than incidental exposure provides. Use MOI when you see this pattern: the child can match and echo, but does not start tacting or responding as a listener from exposure alone.
If you use MOI, match the teaching to the learner’s dignity and motivation. The study used preferred items, breaks, and adjusted task demands as needed. MOI can feel drill-like, so build in choice, brief sessions, and meaningful items when possible. The authors noted that using items the child cares about may matter, even though research often uses uncommon items for control. You can maintain clinical relevance by teaching names of real-life items the child wants to talk about while still rotating operants.
Be careful about what “generalization” means here. Many children showed strong responding on post-MOI probes with trained sets, but the generative probe with a new set was weaker for speaker skills for many participants. Listener responding generalized better than tacting for most children. Do not assume that once a child “has Inc-BiN” with one set, they will automatically demonstrate it with new items. Plan for generalization by programming multiple sets over time and checking new sets without over-teaching right before the test.
Probe order may matter somewhat, but do not over-rely on it. The group probed listener-first showed somewhat better combined listener-plus-speaker outcomes on the generative probe than the speaker-first group. If your probe includes you saying the word (like “Point to the ___”), that can act like a prompt or extra exposure right before testing tacting. To avoid giving extra help, consider separating listener and speaker probe sessions, using different staff, or using fewer spoken models right before a tact test. To support learning, listener-first sequences may be a reasonable, low-effort way to give extra exposure—as long as you document that the probe is not a pure test.
Plan for maintenance directly. At one month, most children retained listener gains more than tact gains. If tacting is an important goal, schedule spaced review, embed use across daily routines, and verify functional use—not just table-top performance. If a child maintains pointing but not saying, your next step might be more practice using the word in natural moments with choice and real outcomes, not additional probe blocks.
Finally, apply this only to learners with the basic prerequisites the study required. These children already had some echoic skills, could do matching, and had some existing tacts and listener responses. If your learner has limited echoics, weak attending, or limited joint attention, this approach may not work the same way. In those cases, build the foundations first, then decide whether exposure-based probes, MOI, or another language-teaching plan is the best fit.
Works Cited
Olaff, H. S., & Holth, P. (2025). Acquisition of incidental bidirectional naming: Isolating the effects of probing and mixed-operant instruction. The Analysis of Verbal Behavior, 41, 200–234. https://doi.org/10.1007/s40616-025-00221-1
