A systematic review of behaviour analytic processes and procedures for conditioning reinforcers among individuals with autism, developmental or intellectual disability

Figures & data

Table 1. Pairing studies

Study	Participant characteristics Age/gender/diagnosis	Purpose Research question Stimulus to be conditioned	Method Study outline	Main outcome Results Response measured	Quality of evidence
+ Ardoin et al., 2004 A method for conditioning reinforcer preference in students with moderate mental retardation	3 boys 12 years Mild to moderate mental retardation	Assess the accuracy of a pictorial choice preference assessment and examine two conditioning procedures for low preference items. Stimulus to be conditioned: VISUAL	ABA design but with only 2 data points in baseline. Neutral stimulus (NS) to be conditioned: white paper with a blue + sign on it. Conditioning was carried on through the simultaneous presentation of the NS and the picture of a highly preferred stimulus to consume upon completion of individualised tasks. The test consisted of probe sessions during which only the NS was presented as a consequence for work completion.	MIXED: NS and reinforcement presented together upon work completion resulted in an increase that was also maintained in sessions where only the NS was presented as a consequence for task completion. A failure to return to baseline in performance was registered that the authors attributed to a general learning effect derived from increased fluency in the academic tasks.	Weak
+ Axe & Laprime, 2017 The effects of contingent pairing on establishing praise as a reinforcer with children with autism	2 boys 5 and 9 years Autism	Condition praise as a reinforcer through a pairing procedure similar to the one described by Dozier et al. (2012). Stimulus to be conditioned: SOCIAL (praise)	Three consecutive experiments. In experiment 1 a reinforcer and a pairing analysis were conducted utilising an AB design. During pairing sessions praise was presented together with a known reinforcer and contingent upon a button pressing response. Aim of the second experiment was to gain experimental control in the context of a reversal design, while experiment 3 tested maintenance.	POSITIVE: praise became an effective reinforcer, but maintenance of responding measured in terms of button pressing was short-lived and reduced after a few sessions.	Moderate
Carroll & Klatt, 2008 Using stimulus- stimulus pairing and direct reinforcement to teach vocal verbal behavior to young children with autism	1 boy and 1 girl 22–23 months Autism	Two experiments: the first aimed to increase vocalisations in both participants through SSP procedure. The second examined direct reinforcement as a means to bring speech sounds under echoic control for one of the two participants. Stimulus to be conditioned: SPEECH SOUNDS	Multiple baseline design. The intervention targeted one new and one known, low in frequency, sound. The target sound was presented three times and consequently twice more while delivering the preferred item for 20 seconds. If the participant emitted the target sound, the presentation of the successive trial was delayed 20 seconds to control for adventitious reinforcement.	MIXED: for one child there was an immediate increase in the frequency of unprompted known target sound but not in the new one. No result was observed for the second child.	Moderate
+ Delgado et al., 2009 Effects of conditioning reinforcement for print stimuli on match-to-sample responding in preschoolers	2 boys and 1 girl 3–5 years Pervasive Developmental Disorder (PDD)	Demonstrate an accelerated rate of learning through the conditioning of previously neutral images as reinforcers for looking. Conditioned reinforcement for observing is considered a developmental cusp that allows for the emergence of matching to sample (MTS) responding. Stimulus to be conditioned: VISUAL	Time lagged multiple probe design across participants. Pre and post-treatment measures were: number of seconds participants looked at novel visual stimuli and ratio of learn unit to criteria for 2D MTS programs. Conditioning procedure consisted of presenting the visual stimulus for 5-second intervals and pairing it with conditioned or unconditioned reinforcers. If the participants stopped looking the interval was started over again; otherwise, the interval was immediately followed with a test trial where no reinforcement was provided for looking responses.	POSITIVE: for all three participants, the treatment phase resulted in conditioned reinforcement for observing print stimuli as measured in seconds spent observing them. A functional relationship was identified between the acquisition of conditioned reinforcement for visual stimuli and a decrease in the mean number of learn units to criteria in MTS teaching programs. For two participants, generalised matching emergence was observed.	Moderate
Esch et al., 2005 Evaluating stimulus-stimulus pairing and direct reinforcement in the establishment of an echoic repertoire of children diagnosed with autism	2 girls and 1 boy 6–8 years Autism	Produce lasting effects supplementing the pairing procedure with an echoic training consisting of direct reinforcement of the vocal responses acquired through the pairing procedure. Stimulus to be conditioned: SPEECH SOUNDS	Three experiments, the first of which is an AB multiple baseline design across topographies with a constant series control. Procedure tested consists in the pairing of a speech sound with a reinforcing item and subsequent, direct reinforcement of echoic responding. The second study is a systematic replication of Miguel et al. (2002), while the third one examines a shaping procedure for one of the three participants.	NEGATIVE: no increase in child emitted speech sounds after pairings, no increase in the subsequent echoic training, and no effect observable on the rate of production of target speech sounds in the participants. Differential reinforcement procedure led to equivocal results in the two participants: and vowel production increased only for one of them.	Moderate
+ Esch et al., 2009 Evaluation of an enhanced stimulus-stimulus pairing procedure to increase early vocalisations of children with autism	3 boys 2–5 years Autism	Evaluate procedural variations in SSP to increase stimulus salience. Stimulus to be conditioned: SPEECH SOUNDS	Non concurrent multiple baseline design combined with a reversal design. The study aims to evaluate the effects of the modified SSP procedure and reinforcement on the frequency of children’s vocalisations as compared to a non-contingent reinforcement (NCR) control condition. Pairing trials were presented contingently upon an observing response.	MIXED: acceptable but moderate effect on the production of target speech sounds, suggesting a possible automatic reinforcement effect of SSP. Listener skills are suggested as a possible prerequisite for optimally benefiting from SSP procedures.	Moderate
+ Greer & Han, 2015 Establishment of conditioned reinforcement for visual observing and the emergence of generalised visual identity matching	3 boys 5 years Autism	Establish 2D print stimuli as conditioned reinforcers for the observing response through direct reinforcement of prolonged observing behaviour and consequently test the acquisition of generalised MTS. Stimulus to be conditioned: VISUAL	Non-concurrent multiple probe design aiming to test generalised MTS after conditioning of printed 2D stimuli as reinforcers for observing responses. Reinforcement was provided for observing 2D print pages for intervals of five seconds. Reinforcement consisted of alternating praise (conditioned reinforcement) and edibles (non-conditioned reinforcement).	POSITIVE: after conditioning of observing responses of printed 2D stimuli, probes for MTS tasks showed a dramatic increase without direct training. Preference for looking at books during free time also increased after treatment. Preference was measured in terms of the number of 5-second intervals spent observing books.	Moderate
+ Greer et al., 2011 Effects of conditioning voices as reinforcers for listener responses on rate of learning, awareness, and preferences for listening to stories in preschoolers with autism	2 boys 1 girl 4–5 years Autism	Evaluate changes before and after voice conditioning protocol (VCP) in: rate of acquisition of listener instructional objectives, frequency of observing responses to voices and adults’ presence, listening to stories activity selection during free play. Stimulus to be conditioned: SOCIAL (recorded voices)	Delayed non concurrent pre and post intervention multiple probe design. Intervention consists of the VCP aiming to the acquisition of conditioned reinforcement for listening to recorded voices. During training sessions, known preferred items (edibles) were delivered contingently on pressing the button that resulted in recorded voices of familiar adults reading a story. Efficacy was measured via preference probes.	MIXED: VCP appears to affect participants’ repertoires in different ways. All participants demonstrated a significant decrease in learn units required to master listener objectives while observing responses and listening to stories increased for two participants. Decrease in stereotypy while listening to stories was observed in two participants.	Moderate
Jerome & Sturmey, 2014 The effects of pairing non-preferred staff with preferred stimuli on increasing the reinforcing value of non-preferred staff attention	3 adult males 42–56 years Mild to moderate Intellectual Disability	Two studies, the second of which is relevant to the conditioning procedure since it explores the applicability of pairing procedures to staff attention. Stimulus to be conditioned: SOCIAL (staff attention)	Experimental design: alternating treatment design embedded in a multiple baseline across participants design to demonstrate experimental control. After evaluation of staff preference, non-preferred staff attention was conditioned through repeated pairings with the delivery of preferred stimuli.	POSITIVE: SSP procedure was effective in changing the value of previously non-preferred staff. After the conditioning procedure staff attention became an effective reinforcer as shown in terms of task engagement measured through progressive ratio and in terms of the rate of approach to previously non-preferred staff members.	Strong
+ Longano & Greer, 2006 The effects of a stimulus-stimulus pairing procedure on the acquisition of conditioned reinforcement on observing and manipulating stimuli by young children with autism	Experiment 1 1 boy 5 years Autism Experiment 2 2 boys 6–7 years Autism	Evaluate the applicability of SSP procedure to independent and seatwork. Stimulus to be conditioned: VISUAL	Multiple baseline across behaviours. In the first experiment, appropriate toy play, play at the computer and looking at books were the behaviours observed. In the second experiment, the behaviour of interest was independent work. During SSP trials, students were asked to complete the intervals without stereotypy or passivity. During the pairing interval, the experimenter delivered vocal praise two or three times per interval while during the successive test interval no pairing was conducted.	POSITIVE: experiment one shows an increase in appropriate play in all the behaviours measured, as well as a decrease in stereotypy and passivity in varying degrees. In experiment two, a significant increase in independent work was observed. The authors claim these results support the notion that SSP procedure was effective in increasing appropriate behaviours and decreasing stereotypy.	Weak
Miguel et al., 2002 The effects of a stimulus – stimulus pairing procedure on the vocal behavior of children diagnosed with autism	3 boys 3–5 years Autism	Evaluate effects of SSP procedure on the vocal utterances of three boys with limited vocal repertoire. Stimulus to be conditioned: SPEECH SOUNDS	Multiple baseline design across vocal behaviours. Target sounds were the two lowest frequency syllable utterances as measured in free operant observation. Pairing sessions consisted of 20 SSP trials defined as five consecutive presentations of the target sound with preferred food delivery on the third one.	MIXED: Increase in child emitted speech sounds observed in two participants, only one of whom showed a significant increase in both sounds.	Moderate
+ Moher et al., 2008 Non generalised and generalised conditioned reinforcers: establishment and validation	3 boys 2 girls 9–21 years Autism or PDD in four participants, mental retardation, ADHD and bipolar disorder for one participant.	Establishing and validating the use of tokens. First and second of four experiments focused on procedures for conditioning tokens and satiation effects. The third and fourth experiment studied satiation and generalised tokens respectively. Stimulus to be conditioned: TOKEN	The first experiment studied the conditioning procedure through an AB design in which two tokens were used for each participant. In stage one, the token and the preferred edible were delivered concurrently, in stage two the token was exchanged for the backup item. No conditioning procedure was necessary for one participant, who was simply verbally informed of the token exchange rule.	POSITIVE: Conditioned tokens functioned as reinforcers on a hand raising response and mirrored the reinforcing effectiveness of the edible they were paired with. Novel and unconditioned tokens, however, did not function as reinforcers for hand-raising responses for any student.	Moderate
Normand & Knoll, 2006 The effects of a stimulus – stimulus pairing procedure on the unprompted vocalisation of a young child diagnosed with autism	1 boy 3 years Autism	Test the effects of SSP procedure on the spontaneous vocalisations. Stimulus to be conditioned: SPEECH SOUNDS	Multiple baseline design across two target phonemes. Conditions were: baseline, control, SSP. A follow-up condition is added to test for the duration of effects, reported in the literature as short-lived.	NEGATIVE: No increase in the vocal production of target sounds was observed, neither in the control nor in the pairing or follow-up conditions.	Moderate
+ Petursdottir et al., 2011 Analysing stimulus-stimulus pairing effects on preferences for speech sounds	3 boys 3,5 and 4 years Autism	Identify variables influencing the success of SSP in establishing a preference for specific auditory stimuli. Stimulus to be conditioned: SOCIAL (speech syllables adult female voice).	Reversal design in which target and control sounds were presented alternately, in the context of a multiple baseline design across participants. An observing response was required before stimulus presentation.	NEGATIVE: the concurrent operant procedure failed to establish a reliable preference for speech sound as measured via button pressing frequency despite several procedural variations.	Moderate
+ Rader et al., 2014 Stimulus-stimulus pairing of vocalisations: A systematic replication	2 boys 1 girl 4–7 years Autism	Systematic replication of SSP study by Esch et al. (2009). Stimulus to be conditioned: SPEECH SOUNDS	Multiple baseline across participants. Treatment consisted of presenting three repetitions of the target sound together with high preference items. Sounds were presented in motherese, interspersed with non-target sounds. An observing response was required.	MIXED: a substantial increase only in target vocalisations was shown in two children, demonstrating that exposure alone was not responsible for the increase.	Moderate
Ward et al., 2007 The effects of a delay of non-contingent reinforcement during a pairing procedure in the development of stimulus control of automatically reinforced vocalisation.	2 girls 3 years Developmental and language delays	Evaluate the efficacy of a procedure that included both automatic and socially mediated reinforcement in the development of vocal repertoire. Stimulus to be conditioned: SPEECH SOUNDS	Multiple probes across sounds. Treatment consisted of a multi-component package: the pairing of target sound with a variety of reinforcers, reinforcement of vocal play and reinforcement of occasional echoic responses.	MIXED: Important increase in speech sounds production and echoic behaviour for one participants and significant gain in echoic control of one sound for the other.	Moderate

+ Studies in which a response was required by the participant to start the pairing trial

Table 2. Discrimination studies

Study	Participant characteristics Age/gender/diagnosis	Purpose Research question Stimulus to be conditioned	Method Study outline	Main outcome Results Response measured	Quality of evidence
Carbone et al., 2013 Teaching eye contact to children with autism: A conceptual analysis and single case study	1 boy 3 years Autism	Extend the analysis of development of social skills to the teaching of eye contact as a language pragmatic skill. Stimulus to be conditioned: SOCIAL (eye contact)	AB design: at baseline manding accompanied by eye contact was infrequent. Treatment consisted of extinction for mands without eye contact and differential reinforcement for mands with eye contact.	POSITIVE: Listener face and eye contact became a conditioned reinforcer for looking and orienting toward the listener in the context of a chain of behaviours in manding. Reinforcing properties of eye contact were not directly tested.	Moderate
Isaksen & Holth, 2009 An operant approach to teaching joint attention skills to children with autism	2 boys 2 girls 3.5–5.5 years Autism	Develop a behavioural training protocol to establish responding to joint attention (JA) bids, engaging in turn-taking activities and initiating JA. Stimulus to be conditioned: SOCIAL (nods and smiles)	Multiple baseline design across children. Training consisted in 3 phases: teach the child to respond to JA bids through prompt and prompt fading, establishing nods and smiles as SD for the availability of preferred items and teaching turn-taking and switch between responding and initiating JA behaviours.	POSITIVE: Post-intervention and follow up measures show an increase in both respondent and initiating JA behaviours as measured on the Early Social Communication Scale (ESCS-m).	Moderate
Lugo et al., 2017 Operant discrimination training to establish praise as a reinforcer	3 girls 1 boy 2–3 years Autism Spectrum Disorder	Evaluate operant discrimination training as a teaching procedure to condition praise statements as reinforcers. Stimulus to be conditioned: SOCIAL (praise statement)	Non-concurrent multiple baseline design to evaluate the effects of operant discrimination training on the conditioning of unfamiliar praise statement as reinforcers. Replication of Holth et al. (2009) and Taylor-Santa et al. (2014).	MIXED: All participants demonstrated a reinforcing effect on arbitrary touching a card response, although these effects were short-lived since responding decreased to zero levels in one or two sessions after training	Moderate
Taylor-Santa et al., 2014 A discrimination training procedure to establish conditioned reinforcers for children with autism	2 boys 1 girl 6 years Autism	Identify the optimal method for establishing conditioned reinforcement and test in the applied setting a procedure in which stimuli trained as discriminative were then presented as consequence. Stimulus to be conditioned: VISUAL	Multiple probe design across stimulus/response to evaluate the effects of discrimination training procedure on responding. During discrimination training, SD and S-delta were established. The same stimuli were then tested for their acquired reinforcing properties during post-training.	POSITIVE: Same pattern of responding in all three participants: responding increased during post-test only in the SD condition, remaining low in the S-delta condition. Previously neutral stimuli were tested in their newly acquired reinforcing properties. Responses measured comprised basic switches operations.	Moderate

Table 3. Observational conditioning studies

Citation Author/title/yr.	Participant characteristics Age/gender/diagnosis	Purpose Research question	Method Study outline	Main outcome Results Response measured	Quality of evidence
Greer & Singer-Dudek, 2008 The emergence of conditioned reinforcement from observation	5 boys 1 girl 3 to 5 years Mild to moderate language delays and behaviour disorders	Evaluate if observation of a peer receiving disks or strings as consequences of performance could condition these previously neutral stimuli as reinforcers. Stimulus to be conditioned: OBJECTS (string/disk)	Multiple baseline across participants. Pre-intervention assessment demonstrated absent reinforcing properties of the objects to be conditioned. Observational conditioning consisted of having the target child observe the confederate peer receive the token or the string as a consequence of correct responding, while observation of the performance and attempting to reach for the object were prevented. After two sessions in which the target child asked or tried to reach for the object, the post-intervention test was run.	POSITIVE: All children either mastered or significantly improved their performance in both acquisition and maintenance tasks during the post-intervention condition, when disks/strings were delivered as reinforcers after the observational conditioning.	Moderate
Singer-Dudek & Oblak, 2013 Peer presence and the emergence of conditioned reinforcement from observation.	2 boys Developmental delays 1 girl Typically developing 3–5 years	Evaluate the acquisition of reinforcing values for previously neutral items after an observational treatment. Stimulus to be conditioned: OBJECTS (strings and toothpicks)	Delayed multiple baseline design across participants combined with a reversal design. Pre-intervention assessed the non-reinforcing effect of the items to be conditioned as compared to an effective reinforcer (food). Treatment was in two phases: no peer present and peer present. The peer’s response was out of the participant sight, and participants could only see the delivery of the object, with no additional stimulus (e.g., praise).	POSITIVE: Non-reinforcing items can become reinforcers for MTS tasks after an observational treatment. Reinforcing properties were acquired only when the peer was present during the observational conditioning phase.	Moderate
Singer-Dudek et al., 2011 Establishing books as conditioned reinforcers for preschool children as a function of an observational conditioning	2 female 1 male 4 years Mild language and developmental delays	Evaluate the feasibility and efficacy of a procedure based solely upon observational learning in conditioning stimuli as reinforcers. Stimulus to be conditioned: OBJECTS (books)	Delayed multiple baseline design across participants that consisted of measuring participants’ responses before and after observational conditioning of books as reinforcers. During the intervention, participants were exposed to peers performing a task (non-observed) and receiving/observing books, for which peers already had a preference. Simultaneously participants were asked to perform the same task at the same time while receiving no consequence for completion.	POSITIVE: For all participants, books, previously neutral, acquired reinforcing properties. Reinforcing effects were observed on maintenance and acquisition tasks. Time spent looking at books during free time also increased after treatment.	Moderate
Leaf et al., 2012 Observational effects on the preferences of children with autism	First experiment: 3 boys 5–6 years Autism Second experiment: 2 out of the three original study participants.	Modify preferences for toys through an observational procedure. Stimulus to be conditioned: OBJECTS (toys)	Two experiments: the first to alter children’s preferences for toys, the second to test for maintenance and replicate results with different items. Conditioning procedure consisted of observation of a preferred adult playing with non-preferred toys in exciting ways. ABABA and ABACA design and a multiple baseline design were utilised, ABA and multiple baseline across participants in the second experiment.	POSITIVE: In the first experiment, all three children switched their preference for toys after the intervention, results replicated with a different set of materials. Reinforcing properties were not tested.	Moderate
Leaf et al., 2016 Changing preference from tangible to social activities through an observation procedure	2 boys 1 girl 5 years Autism	Test the feasibility of an observational procedure to shift preferences from tangible items to initially non-preferred social activities. Stimulus to be conditioned: SOCIAL (activities)	Reversal design: preference for tangible items, social activities, and control items were measured by calculating the percentage of trials in which the participant selected the item. During baseline, all three participants selected the item that was preferred at the previous preference assessment but after observing five trials during which a peer confederate chose and experienced a social game all of them shifted their preference to the social activity.	POSITIVE: All three participants demonstrated a change in their preferences after the observational procedure, and started to select the social activity as a reinforcer after engaging in a simple task.	Strong

Table 4. Comparison studies

Citation Author/title/yr.	Participant characteristics Age/gender/diagnosis	Purpose Research question	Method Study outline	Main outcome Results Response measured	Quality of evidence
Cividini-Motta et al., 2017 An assessment of three procedures to teach echoic responding	5 males 1 female 7–17 years Autism	Evaluate a protocol to assess the most effective teaching procedure for increasing participants’ vocalisations and echoic responding. Stimulus to be conditioned: SPEECH SOUNDS	Within subjects alternating treatment design comparing the effects of three treatment conditions. Vocal imitation training, stimulus-stimulus pairing (SSP-RIP) and mand model were compared.	MIXED: assessment protocol identified effective teaching procedures in five out of six children, but in terms of evaluating conditioning procedures, no single procedure was superior, and there was considerable variability in their efficacy.	Moderate
Dozier et al., 2012 A comparison of two pairing procedures to establish praise as a reinforcer	12 adults 17–56 years Mild to severe mental retardation associated with other diagnoses for some participants	Evaluate if the pairing of neutral stimuli such as praise with primary reinforcers such as food would be more effective in establishing praise alone as a reinforcer in a contingent or in a non-contingent preparation Stimulus to be conditioned: SOCIAL (praise).	ABC, reversal or multiple baseline designs were applied depending on subjects. In experiment 1, praise statements were presented and immediately followed by a preferred edible. In experiment 2, praise statements and food were presented contingently on a simple motor response. Procedures compared: RIP vs RCP.	MIXED: Experiment 1 results demonstrated that RIP was ineffective in conditioning praise statement as reinforcers for simple target behaviours in three participants. Experiment 2 results demonstrated that praise acquired reinforcing properties in four participants through RCP. Effects of praise on simple motor responses measured newly acquired reinforcing efficacy.	Moderate
Holth et al., 2009 An operant analysis of joint attention skill and the establishment of conditioned social reinforcers	8 males 2–12 years 4 Autism 2 Down Syndrome 2 Typically developing (1 did not complete the study)	Compare SSP with ODT in the establishment of two different visual or auditory neutral stimuli as reinforcers. Stimulus to be conditioned: SOCIAL (computer-administered applause/yay sound/smiles).	Single-subject design with pre-test and post-test. Four phases starting with a pre-test to identify potential reinforcers. ODT was run for one stimulus and SSP for the other. Sequence effect was controlled. Procedures compared: ODT vs. SSP.	MIXED: Five children clearly emitted more responses after the ODT procedure than after the SSP procedure, for one child no difference was found, while one child emitted more responses after SSP than after ODT. Responses tested were basic motor responses such as moving an object or hand clapping.	Moderate
Lepper et al., 2013 Effects of an operant discrimination training on the vocalisations of nonverbal children with autism	3 boys 2-3-4 years Autism	Evaluate the feasibility of ODT (Holth et al., 2009) to increase vocal production while comparing it to SSP. Stimulus to be conditioned: SPEECH SOUNDS.	Alternating treatment design to compare ODT, SSP and control conditions. In each condition, target and non-target speech sounds were measured. Procedures compared: ODT vs SSP	MIXED: ODT increased two participants’ target vocalisations only, showing similar results to SSP condition, but ODT was preferred over SSP by all participants.	Moderate
Lepper & Petursdottir, 2017 Effects of response-contingent stimulus pairing on vocalisation of non-verbal children with autism	3 boys 4-5-6 years Autism	Compare the effects of RCP and RIP on the development of new vocalisations. Stimulus to be conditioned: SPEECH SOUNDS.	Two experiments, the first of which focused on comparing RIP and RCP to condition new language sounds as reinforcers through a multielement design with target sounds randomly assigned to each procedure. Procedures compared: RCP vs. RIP	POSITIVE: RCP was superior to RIP. RCP proved to be more effective in developing new vocalisations in all participants. A subsequent treatment also proved effective in maintaining gains and increasing the durability of RCP.	Moderate
Miliotis et al., 2012 An evaluation of the number of presentations of target sounds during stimulus stimulus pairing trials	1 girl 1 boy 6–8 years Autism	Identify optimal number of repetitions in SSP trials. Stimulus to be conditioned: SPEECH SOUNDS.	Alternating treatment design comparing different NUMBER OF PRESENTATIONS in SSP trials.	POSITIVE: the enhanced SSP procedure (Esch et al., 2009) showed the best results when one sound presentation was associated with the delivery of the preferred items. An attending response was required.	Moderate
Rodriguez & Gutierrez, 2017 A comparison of two procedures to condition social stimuli to function as reinforcers for children with autism	6 males 18–36 months Autism	Compare procedure based on operant and respondent paradigms to increase the value of social stimuli. Stimulus to be conditioned: SOCIAL (smiles, yay sounds, clapping)	Multiple baseline design with an embedded multiple treatment with reversal design. The respondent pairing trials were presented contingently on a simple motor response. Procedures compared: RCP vs ODT	NEGATIVE: Both conditioning procedures proved to have limited and overall not durable effects on basic motor response rates. RCP was found to have clear conditioning effects for two out of the six participants. ODT proved effective for two participants but with unstable data.	Moderate
Stock et al., 2008 A comparison of stimulus-stimulus pairing, standard echoic training and control procedures on the vocal behavior of children with autism	2 females 1 male 2–4 years Autism	Compare echoic and SSP procedures to increase vocal production in three preschoolers, early learner with autism. Stimulus to be conditioned: SPEECH SOUNDS.	Alternating treatment design to compare the effects of SSP, standard echoic training and control condition. Procedures compared: SSP vs ECHOIC vs CONTROL (enriched environment)	NEGATIVE: Results showed an immediate but short-lived effect in vocal production for only one participant out of the three. In the echoic condition, children emitted many avoidance behaviours. This study failed to demonstrate any superiority of SSP on control or standard echoic training conditions.	Moderate

Table 5. Summary of comparison studies

Stimulus to be conditioned	SOCIAL STIMULI			SPEECH SOUNDS
Response measured	MOTOR RESPONSE			CHILD EMITTED SPEECH SOUNDS
Short reference/conditioning procedure	Dozier et al., 2012	Holth et al., 2009	Rodriguez & Gutierrez, 2017	Cividini-Motta et al., 2017	Lepper et al., 2013	Lepper & Petursdottir, 2017	Miliotis et al., 2012	Stock et al., 2008
RIP	-	-	N/A	+ 1 of 6 participants Named SSP	+	-	N/A	-Named SSP
RCP	+ Named response stimulus pairing	N/A	+ Named respondent procedure	N/A	N/A	+	3vs1 pairing	N/A
ODT	N/A	+	-	N/A	+ Preferred	N/A	N/A	N/A
Other condition (e.g., echoic or control)	N/A	N/A	N/A	+	-	N/A	N/A	-

RCP: Response-contingent pairingSSP: Stimulus-stimulus pairingRIP: Response independent pairingODT: Operant discrimination trainingN/A: Not-Applicable+: the procedure was deemed effective by the authors-: the procedure was deemed ineffective by the authors

Table 6. Procedures used across studies and reported results

Stimulus to be conditioned/ Experimental preparation	Speech sounds	Social and voices	Visual and Objects	Token
SSP: Stimulus-stimulus pairing Also called traditional procedure (Lugo et al., 2017) or response independent pairing (RIP) in Lepper and Petursdottir (2017). Procedural variations in terms of stimuli characteristics and number of presentations. In RIP the pairings are time-based and do not depend on any behaviour of the subject. Pairings can be simultaneous, delayed or trace, mirroring basic literature on respondent conditioning.	Carroll & Klatt, 2008 (mixed results) Cividini-Motta et al., 2017 (CMPR) Esch et al., 2005 (-) Lepper et al., 2013 (CMPR) Lepper & Petursdottir, 2017 CMPR Miguel et al., 2002 (mixed results) Normand & Knoll, 2006 Neg. Stock et al., 2008 (CMPR) Ward et al., 2007 (mixed results)	Dozier et al., 2012 (CMPR) Holth et al., 2009 (CMPR) Jerome & Sturmey, 2014 (+).
RCP: Response contingent pairing Also called response stimulus pairing (Dozier et al., 2012) and respondent pairing procedure (Rodriguez & Gutierrez, 2017). A response is either required from the participant to start the pairing trial or a prompt is provided for the participant to emit some sort of observing response before the pairing trial starts.	Esch et al., 2009 (mixed results) Miliotis et al., 2012 (CMPR) Lepper & Petursdottir, 2017 (CMPR) Rader et al., 2014 (mixed results)	Dozier et al., 2012 (CMPR) Axe & Laprime, 2017 (+) Greer et al., 2011 (mixed results) Rodriguez & Gutierrez, 2017 (CMPR) Petursdottir et al., 2011 (-)	Ardoin et al., 2004 (+) Delgado et al., 2009 (+) Greer & Han, 2015 (+) Longano & Greer, 2006 (+)	Moher et al., 2008 (+).
ODT: Operant discrimination training Also called discriminative stimulus procedure (Rodriguez & Gutierrez, 2017). The neutral stimulus becomes an SD for a reinforced response and can then be tested as a reinforcer.	Lepper et al., 2013 (CMPR)	Carbone et al., 2013 (+) Holth et al., 2009 CMPR Isaksen & Holth, 2009 (+) Lugo et al., 2017 (mixed results) Rodriguez & Gutierrez, 2017 (CMPR)	Taylor-Santa et al., 2014 (+).
Observational Conditioning Observation of participant’s confederate or familiar adult receiving neutral stimuli as a consequence for brief non-observed tasks shifts subsequent participants’ preference		Leaf et al., 2016 (+)	Greer & Singer-Dudek, 2008 Pos. Leaf et al., 2012 (+) Singer-Dudek & Oblak, 2013 (+) Singer-Dudek et al., 2011 (+)
Total number of studies	12	11	9	1

-: negative results+: positive resultsCMPR: Comparison study

Table 7. Quality of evidence ratings in pairing studies

Brief citation	Q. of E.	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Q9	Q10	Q11	Q12	Q13	Q14	Total
Ardoin et al. 2004	Weak	1	0	1	0	0	0	0	0	0	1	1	1	0	0	5
Axe & Laprime, 2017	Moderate	1	1	1	1	1	0	1	1	1	0	1	1	0	0	10
Carroll & Klatt 2008	Moderate	1	1	1	1	1	0	1	1	0	0	1	1	0	0	9
Delgado et al., 2009	Moderate	1	1	1	1	1	0	0	1	0	1	0	1	0	0	8
Esch et al., 2005	Moderate	1	1	1	1	1	0	1	1	0	0	1	1	0	0	9
Esch et al., 2009	Moderate	1	1	1	1	1	0	1	1	0	1	1	1	0	0	10
Greer & Han, 2015	Moderate	1	1	1	1	1	0	0	1	0	0	0	1	0	0	7
Greer et al., 2011	Moderate	1	1	1	1	1	0	0	1	0	1	0	1	0	0	8
Jerome & Sturmey, 2014	Strong	1	1	1	1	1	0	1	1	1	1	1	1	0	0	11
Longano & Greer, 2006	Weak	1	1	1	1	0	0	0	1	0	0	0	1	0	0	6
Miguel et al., 2002	Moderate	1	1	1	1	1	0	1	1	1	0	1	1	0	0	10
Moher et al., 2008	Moderate	1	1	1	1	1	0	1	0	0	1	1	1	0	0	9
Normand & Knoll, 2006	Moderate	1	1	1	1	1	0	1	1	1	0	1	1	0	0	10
Petursdottir et al., 2011	Moderate	1	1	1	1	1	0	0	0	0	1	1	1	0	0	8
Rader et al., 2014	Moderate	1	1	1	1	1	0	1	1	0	0	1	1	0	0	9
Ward et al., 2007	Moderate	1	1	1	1	1	0	1	1	0	0	0	1	0	0	8
Quality of evidence ratings in discrimination studies
Carbone et al., 2013	Moderate	1	1	1	1	1	0	0	1	1	0	1	1	0	0	9
Isaksen & Holth, 2009	Moderate	1	1	1	1	1	0	1	0	0	1	1	1	0	0	9
Lugo et al., 2017	Moderate	1	1	1	1	1	0	1	1	0	1	1	1	0	0	10
Taylor-Santa et al., 2014	Moderate	1	1	1	1	1	0	1	1	0	1	1	1	0	0	10
Quality of evidence ratings in observational conditioning studies
Greer & Singer-Dudek, 2008	Moderate	1	1	1	1	1	0	0	0	0	1	0	1	0	0	7
Singer-Dudek & Oblak, 2013	Moderate	1	1	1	1	1	0	0	1	0	1	0	1	0	0	8
Singer-Dudek et al., 2011	Moderate	1	1	1	1	1	0	1	1	1	1	1	1	0	0	10
Leaf et al., 2012	Moderate	1	1	1	1	1	0	1	1	0	1	1	1	0	0	10
Leaf et al., 2016	Strong	1	1	1	1	1	0	1	1	1	1	1	1	0	0	11
Quality of evidence ratings in comparison studies
Cividini-Motta et al., 2017	Moderate	1	1	1	1	1	0	1	1	0	0	1	1	0	0	9
Dozier et al., 2012	Moderate	1	1	1	1	1	0	1	1	0	1	1	1	0	0	10
Holth et al., 2009	Moderate	1	1	1	1	1	0	1	0	0	1	1	1	0	0	9
Lepper et al., 2013	Moderate	1	1	1	1	1	0	0	1	0	1	1	1	0	0	9
Lepper & Petursdottir, 2017	Moderate	1	1	1	1	1	0	1	1	0	1	1	1	0	0	10
Miliotis et al., 2012	Moderate	1	1	1	1	1	0	1	1	1	0	0	1	0	0	9
Rodriguez & Gutierrez, 2017	Moderate	0	1	1	1	1	0	1	1	1	0	1	1	0	0	9
Stock et al., 2008	Moderate	1	1	1	1	1	0	0	1	1	0	0	0	0	0	8

1: “yes” response to question0: “no” response to question

Figure 1. Flowchart depicting inclusion process

Figure 2. Summary of participants’ age

Figure 3. Lines of research

Figure 4. Stimuli to be conditioned across lines of research

Figure 5. Dependent variable measures across lines of research

Figure 6. Reported effectiveness across lines of research: Pairing with and without response, discrimination and observational conditioning

Figure 7. Reported effectiveness per stimuli to be conditioned across lines of research

Figure 8. Quality of evidence across lines of research

Ardoin, S. P., Martens, B. K., Wolfe, L. A., Hilt, A. M., & Rosenthal, B. D. (2004). A method for conditioning reinforcer preferences in students with moderate mental retardation. Journal of Developmental and Physical Disabilities, 16(1), 33–51. https://doi.org/10.1023/B:JODD.0000010038.69725.88

 Web of Science ®Google Scholar

Axe, J. B., & Laprime, A. P. (2017). The effects of contingent pairing on establishing praise as a reinforcer with children with autism. Journal of Developmental and Physical Disabilities, 29(2), 325–340. https://doi.org/10.1007/s10882-016-9526-9

 Web of Science ®Google Scholar

Dozier, C. L., Iwata, B. A., Thomason-Sassi, J., Worsdell, A. S., & Wilson, D. M. (2012). A comparison of two pairing procedures to establish praise as a reinforcer. Journal of Applied Behavior Analysis, 45(4), 721–735. https://doi.org/10.1901/jaba.2012.45-721

 PubMed Web of Science ®Google Scholar

Delgado, J. A. P., Greer, R. D., Speckman, J. M., & Goswami, A. (2009). Effects of conditioning reinforcement for print stimuli on match-to-sample responding in preschoolers. The Journal of Speech and Language Pathology – Applied Behavior Analysis, 3(2–3), 198–216. https://doi.org/10.1037/h0100245

 Google Scholar

Esch, B. E., Carr, J. E., & Michael, J. (2005). Evaluating stimulus-stimulus pairing and direct reinforcement in the establishment of an echoic repertoire of children diagnosed with autism. The Analysis of Verbal Behavior, 21(1), 43–58. https://doi.org/10.1007/BF03393009

 PubMedGoogle Scholar

Miguel, C. F., Carr, J. E., & Michael, J. (2002). The effects of a Stimulus-Stimulus Pairing procedure on the vocal behavior of children diagnosed with autism. The Analysis of Verbal Behavior, 18(1), 3–13. https://doi.org/10.1007/BF03392967

 PubMedGoogle Scholar

Esch, B. E., Carr, J. E., & Grow, L. L. (2009). Evaluation of an enhanced stimulus-stimulus pairing procedure to increase early vocalisations of children with autism. Journal of Applied Behavior Analysis, 42(2), 225–241. https://doi.org/10.1901/jaba.2009.42-225

 PubMed Web of Science ®Google Scholar

Greer, R. D., & Han, H. S.-A. (2015). Establishment of conditioned reinforcement for visual observing and the emergence of generalised visual-identity matching. Behavioral Development Bulletin, 20(2), 227–252. https://doi.org/10.1037/h0101316

 Google Scholar

Greer, R. D., Pistoljevic, N., Cahill, C., & Du, L. (2011). Effects of conditioning voices as reinforcers for listener responses on rate of learning, awareness, and preferences for listening to stories in preschoolers with autism. The Analysis of Verbal Behavior, 27(1), 103–124.

 PubMedGoogle Scholar

Jerome, J., & Sturmey, P. (2014). The effects of pairing non-preferred staff with preferred stimuli on increasing the reinforcing value of non-preferred staff attention. Research in Developmental Disabilities, 35(4), 849–860. https://doi.org/10.1016/j.ridd.2014.01.014

 PubMed Web of Science ®Google Scholar

Longano, J. M., & Greer, R. D. (2006). The effects of a stimulus-stimulus pairing procedure on the acquisition of conditioned reinforcement on observing and manipulating stimuli by young children with autism. Journal of Early and Intensive Behavior Intervention, 3(1), 62–80. https://doi.org/10.1037/h0100323

 Google Scholar

Moher, C. A., Gould, D. D., Hegg, E., & Mahoney, A. M. (2008). Non-generalised and generalised conditioned reinforcers: Establishment and validation. Behavioral Interventions, 23(1), 13–38. https://doi.org/10.1002/bin.253

 Web of Science ®Google Scholar

Normand, M. P., & Knoll, M. L. (2006). The effects of a stimulus-stimulus pairing procedure on the unprompted vocalisations of a young child diagnosed with autism. The Analysis of Verbal Behavior, 22(1), 81–85. https://doi.org/10.1007/BF03393028

 PubMedGoogle Scholar

Petursdottir, A. I., Carp, C. L., Matthies, D. W., & Esch, B. E. (2011). Analysing stimulus-stimulus pairing effects on preferences for speech sounds. The Analysis of Verbal Behavior, 27(1), 45–60. https://doi.org/10.1007/BF03393091

 PubMedGoogle Scholar

Rader, L., Sidener, T. M., Reeve, K. F., Sidener, D. W., Delmolino, L., Miliotis, A., & Carbone, V. (2014). Stimulus-stimulus pairing of vocalisations: A systematic replication. The Analysis of Verbal Behavior, 30(1), 69–74. https://doi.org/10.1007/s40616-014-0012-0

 PubMedGoogle Scholar

Ward, S. J., Osnes, P. J., & Partington, J. W. (2007). The effects of a delay of noncontingent reinforcement during a pairing procedure in the development of stimulus control of automatically reinforced vocalisations. The Analysis of Verbal Behavior, 23(1), 103–111. https://doi.org/10.1007/BF03393050

 Google Scholar

Carbone, V. J., O’Brien, L., Sweeney-Kerwin, E. J., & Albert, K. M. (2013). Teaching eye contact to children with autism: A conceptual analysis and single case study. Education & Treatment of Children, 36(2), 139–159. https://doi.org/10.1353/etc.2013.0013

 Web of Science ®Google Scholar

Isaksen, J., & Holth, P. (2009). An operant approach to teaching joint attention skills to children with autism. Behavioral Interventions, 24(4), 215–236. https://doi.org/10.1002/bin.292

 Web of Science ®Google Scholar

Lugo, A. M., Mathews, T. L., King, M. L., Lamphere, J. C., & Damme, A. M. (2017). Operant discrimination training to establish praise as a reinforcer. Behavioral Interventions, 32(4), 341–356. https://doi.org/10.1002/bin.1485

 Web of Science ®Google Scholar

Holth, P., Vandbakk, M., Finstad, J., Grønnerud, E. M., Mari, J., & Sørensen, A. (2009). An operant analysis of joint attention and the establishment of conditioned social reinforcers. European Journal of Behavior Analysis, 10(2), 143–158. https://doi.org/10.1080/15021149.2009.11434315

 Google Scholar

Taylor-Santa, C., Sidener, T. M., Carr, J. E., & Reeve, K. F. (2014). A discrimination training procedure to establish conditioned reinforcers for children with autism. Behavioral Interventions, 29(2), 157–176. https://doi.org/10.1002/bin.1384

 Web of Science ®Google Scholar

Greer, R. D., & Singer-Dudek, J. (2008). The emergence of conditioned reinforcement from observation. Journal of the Experimental Analysis of Behavior, 89(1), 15–29. https://doi.org/10.1901/jeab.2008.89-15

 PubMed Web of Science ®Google Scholar

Singer-Dudek, J., & Oblak, M. (2013). Peer presence and the emergence of conditioned reinforcement from observation. Journal of Applied Behavior Analysis, 46(3), 592–602. https://doi.org/10.1002/jaba.72

 PubMed Web of Science ®Google Scholar

Singer-Dudek, J., Oblak, M., & Greer, R. D. (2011). Establishing books as conditioned reinforcers for preschool children as a function of an observational intervention. Journal of Applied Behavior Analysis, 44(3), 421–434. https://doi.org/10.1901/jaba.2011.44-421

 PubMed Web of Science ®Google Scholar

Leaf, J. B., Oppenheim-Leaf, M. L., Leaf, R., Courtemanche, A. B., Taubman, M., McEachin, J., … Sherman, J. A. (2012). Observational effects on the preferences of children with autism. Journal of Applied Behavior Analysis, 45(3), 473–483. https://doi.org/10.1901/jaba.2012.45-473

 PubMed Web of Science ®Google Scholar

Leaf, J. B., Oppenheim-Leaf, M. L., Townley-Cochran, D., Leaf, J. A., Alcalay, A., Milne, C., Kassardjian, A., Tsuji, K., Dale, S., Leaf, R., Taubman, M., & McEachin, J. (2016). Changing preference from tangible to social activities through an observation procedure. Journal of Applied Behavior Analysis, 49(1), 49–57. https://doi.org/10.1002/jaba.276

 PubMed Web of Science ®Google Scholar

Cividini-Motta, C., Scharrer, N., & Ahearn, W. H. (2017). An assessment of three procedures to teach echoic responding. The Analysis of Verbal Behavior, 33(1), 41–63. https://doi.org/10.1007/s40616-016-0069-z

 PubMed Web of Science ®Google Scholar

Lepper, T. L., Petursdottir, A. I., & Esch, B. E. (2013). Effects of operant discrimination training on the vocalisations of nonverbal children with autism. Journal of Applied Behavior Analysis, 46(3), 656–661. https://doi.org/10.1002/jaba.55

 PubMed Web of Science ®Google Scholar

Lepper, T. L., & Petursdottir, A. I. (2017). Effects of response-contingent stimulus pairing on vocalisations of nonverbal children with autism. Journal of Applied Behavior Analysis, 50(4), 756–774. https://doi.org/10.1002/jaba.415

 PubMed Web of Science ®Google Scholar

Miliotis, A., Sidener, T. M., Reeve, K. F., Carbone, V., Sidener, D. W., Rader, L., & Delmolino, L. (2012). An evaluation of the number of presentations of target sounds during stimulus-stimulus pairing trials. Journal of Applied Behavior Analysis, 45(4), 809–813. https://doi.org/10.9101/jaba.2012.45-809

 PubMed Web of Science ®Google Scholar

Rodriguez, P. P., & Gutierrez, A. (2017). A comparison of two procedures to condition social stimuli to function as reinforcers for children with autism. Behavioral Development Bulletin, 22(1), 159–172. https://doi.org/10.1037/bdb0000059

 Google Scholar

Stock, R. A., Schulze, K. A., & Mirenda, P. (2008). A comparison of stimulus-stimulus pairing, standard echoic training, and control procedures on the vocal behavior of children with autism. The Analysis of Verbal Behavior, 24(1), 123–133. https://doi.org/10.1007/BF03393061

 Google Scholar