Stimulus-Stimulus Pairing to Increase Vocalizations: What the Research Tells Clinicians

Some learners make very few sounds, which can make it hard to start teaching mands, tacts, or echoics. Stimulus-stimulus pairing (SSP) has been proposed as a low-demand way to build early vocal behavior—but past studies have shown mixed results. A recent systematic review examined the evidence from 2015 to 2024, helping clinicians understand when SSP might be worth trying, what to watch for, and when to switch strategies. See also: stimulus-stimulus pairing for increasing vocalizations.

What Is the Research Question and Why Does It Matter?

The question is whether stimulus-stimulus pairing tends to increase vocalizations in children with language delays. In SSP, an adult says a sound (like “ba”) and then gives something the child likes—without requiring the child to copy the sound first. See also: Behavior Analysis in Practice.

Clinicians care because SSP could offer a low-demand way to build early speech behavior while keeping sessions positive and choice-based.

This matters because past SSP studies have shown mixed results. Sometimes vocalizations go up, sometimes they barely change, and sometimes they don’t change at all. If outcomes vary that much, clinicians need to know when SSP is worth trying and when to stop and switch strategies. The review also checks how strong the newer evidence is and highlights where the research remains thin.

What Did the Researchers Do?

The authors searched for studies from 2015 to 2024 that tested SSP as the main treatment and measured learner vocalizations as the outcome. They included peer-reviewed papers plus theses and dissertations, as long as the study manipulated SSP and tracked changes in vocal sounds. They found 12 studies with 61 total participants. Most participants were diagnosed with autism and did not have functional spoken language.

They coded procedural details that affect how SSP works in real clinics: whether the target sound was “new” or already in the child’s repertoire, how many times the adult said the sound per trial, what pairing timing was used (simultaneous, delayed, or trace), whether the child had to orient first, whether there was a control for accidental reinforcement, what items were paired, and how much SSP the child received.

For a smaller set of studies, they ran a meta-analysis to estimate the average effect size while accounting for differences across participants. One study was excluded because baseline data were too limited. In the end, 6 studies with 17 participants were included.

On average, baseline vocalizations were very low, and the overall effect showed an increase during SSP—but with meaningful differences from child to child.

How to Use This in Day-to-Day Clinical Practice

Treat SSP as a reasonable “try it and see” option for learners who have very low speech sounds and are not yet learning well from direct echoic teaching. The research trend suggests SSP can increase vocalizations for some children, but not everyone.

Your clinical job is not to add SSP to every plan. It’s to decide whether SSP matches this learner’s needs, run it with clean data, and make a fast decision based on that learner’s response.

Be Clear About Your Goal and Measurement

In these studies, the outcome was usually rate of vocalizations—not correct words. If you run SSP, plan to measure simple things: frequency of any vocal sounds, frequency of specific target sounds, and whether vocalizations happen outside SSP trials too.

Decide ahead of time what would count as “enough change” to keep going—for example, a steady rise across several sessions, not just one good day. This keeps you from running SSP for months when it’s not helping.

Pick Target Sounds Thoughtfully

Many studies targeted sounds children already made sometimes; others targeted sounds not seen in baseline. The review doesn’t give a clear rule about which is better, so individualize.

A practical starting point: pick one or two simple consonant-vowel sounds that fit the child’s current motor abilities (like “m,” “b,” or “p” plus a simple vowel). If the child already makes a similar sound rarely, you can target that to try increasing it. If the child has almost no sounds, test a very simple sound that’s easy to produce and easy to hear.

Protect Learner Dignity and Choice

SSP is often used because it doesn’t require compliance or forced imitation. Keep it that way.

If the learner moves away or shows clear refusal, pause and reset or offer a different activity. Pairing should not look like constant demands or blocking. Think of it as creating pleasant sound experiences around the learner—not “making speech happen.”

Use Strong Preference Information

Most studies used some form of preference assessment. Items paired included toys, edibles, and social play, often in combination.

In practice, do a quick preference check before sessions and re-check often—preferences shift fast. If you pair with edibles, be careful that chewing doesn’t compete with vocalizing. If you pair with play, make sure it’s truly fun for that learner, not just adult-led attention.

When possible, mix social and non-social reinforcers so the adult and interaction can become part of what the learner enjoys. But don’t assume social attention is reinforcing.

Decide How to Structure Trials

Some studies required orienting (like “look”), and some had learners initiate trials (like reaching for an item or pressing a button). The review suggests learner initiation and orienting may matter, but the evidence isn’t clear enough to identify one best method.

A good low-pressure option: arrange chances for the learner to initiate. Place a preferred item in view but not in hand, then pair the sound right as the learner reaches. This keeps motivation high and reduces “random pairing” when the learner isn’t attending.

Keep Your Procedure Simple and Consistent

Many studies used one adult-emitted sound per trial; others used three or five. Since there’s no clear winner, start with one clear model to reduce confusion and make data easier to interpret.

Use a steady rhythm—about one to two pairings per minute. Avoid “machine-gun pairing” where you repeat the sound too quickly. Fast massed trials may increase problem behavior, reduce reinforcer value, or just turn the session into noise.

Watch for Accidental Reinforcement

Some studies used an “omission” approach—withholding the item briefly if the child made the target sound right then—to avoid teaching an operant response by mistake. Others did not.

Decide what you want SSP to be. If your goal is purely pairing, delivering the item right after a child vocalization could turn it into response-contingent reinforcement. Then you’re not really testing SSP.

One practical approach: pause briefly if the child vocalizes right as you’re about to deliver, then deliver after a short delay. Document that rule in your protocol. If you prefer to move toward operant teaching sooner, you can choose response-contingent methods—but label it clearly and track it as a different intervention.

Plan Dosage Like a Pilot, Not a Forever Program

Across studies, total trials varied widely—some children got a few dozen pairings, others got hundreds. The field doesn’t yet know the “right dose.”

Run SSP as a short-term test with a clear review date. For example: try daily brief sessions for one to two weeks, graph vocalizations, and decide whether to continue.

If vocalizations rise, your next step isn’t to keep pairing forever. It’s to use those sounds inside functional teaching—like shaping mands with strong motivation and respecting communication attempts, including AAC.

Don’t Assume Generalization

The outcome in these studies was increased vocalizations, not functional spoken communication. If SSP increases sound-making, that’s only a starting point.

You still need to teach meaningful communication in ways that reduce frustration and increase autonomy. That often means combining speech work with AAC so the learner has a reliable way to communicate right now, even while you’re building vocal behavior.

Recognize the Research Gaps

Many included studies were dissertations. Many did not report who implemented SSP or describe learners in enough detail to know who SSP fits best. No studies reported participant race or ethnicity, which limits how well we can judge generality across communities. The meta-analysis included only a small number of studies and participants.

Use SSP as one tool, not a guarantee. Keep your decision-making tied to your learner’s data, family priorities, and the learner’s comfort in sessions.

Works Cited

Oliveira, J. S. C. D., Barall, R. J., Moeyaert, M., Khan, S. U., & Shillingsburg, M. A. (2025). Systematic review and meta-analysis of stimulus-stimulus pairings to increase vocalizations in children with language delays: 2015–2024. The Analysis of Verbal Behavior, 41, 151–178. https://doi.org/10.1007/s40616-025-00219-9