Establishment and maintenance of socially learned conditioned reinforcement in young children: elimination of the role of adults and view of peers' faces.
In the initial study with young children. Greer and Singer-Dudek (2008) identified observational conditions under which new reinforcers were established with preschool-age children for learning new stimulus control and maintenance of previously mastered behavior. After determining that neutral stimuli (i.e., pieces of string and small plastic discs with no backup reinforcer) did not function to reinforce learning or performance when compared to primary reinforcers, the participants received an observational intervention. In the observational intervention, while both participants and peers were engaged in a previously mastered task, the participants observed peers receiving pieces of string or small plastic discs. The participants were denied access these stimuli. After experiencing these conditions in repeated 2- to 4-min sessions, the strings and discs functioned to reinforce responding and learning comparable to the children's learning and performance with primary reinforcers (i.e., preferred edibles). Singer-Dudek et al. (2008) replicated this study and effect with pieces of string. Greer et al. (2008) extended the effect to the conditioning of adult attention and approvals, showing that vocal verbal stimuli and adult attention could be conditioned as a function of the social learning conditions. In the latter studies, children for whom adult approvals did not reinforce learning or performance acquired approval as conditioned reinforcers for learning and performance as a function of observing other children receiving approvals while the participants were denied approvals. Singer-Dudek et al. (2011) found similar effects for opportunities to observe books. In this study, children who did not perform or learn when given opportunities to look at books as part of reinforcement operations did learn and perform when access to books was contingent on responding after the observational intervention described above.
In the case of the children in the studies described above, in addition to the peer, an adult experimenter was also involved in the intervention as it was the adult who dispensed the stimuli. One of the possibilities for the effect with children was that the presence of the adults and the adult experimenters' reactions to peers receiving the contingencies, or some other interaction between the presence of the peer and the experimenter, may be instrumental in establishing the neutral stimuli as reinforcers. In the following experiment, we sought to test whether use of the components of the observational intervention in prior experiments (Greer & Singer Dudek, 2008; Greer et al., 2008; Singer-Dudek et al., 2008) minus the presence of adult experimenters would also result in establishing neutral stimuli as conditioned reinforcers. Also, would the reinforcing effect maintain under classroom instructional conditions several weeks after the intervention?
Three 4-year-old males with language delays were the target participants (see Table 1), and four other 4-year-old children in the school with similar repertoires served as rotated peer confederates and were naive to the purpose of the study. They were recruited from the population of a private but publicly funded preschool of 90 children with and without disabilities. Informed consent for participants as well as peer confederates was properly obtained prior to participation in the study. The peer confederates did not have close contact with the participants; they were from different classrooms. The school used a behavior analytic approach to all teaching, curriculum, and behavior management. All responses to instruction in the school were measured, and the students in the school were experienced with one-to-one instruction; small-group instruction; and the use of instructional trials, pairing trials, and probe trials in which consequences were not delivered. Therefore, participants and peer confederates did not require habituation to receiving probe or instructional trials. The three target participants were chosen to participate because they were the first three children for whom pre-intervention assessments showed that the neutral stimuli (metal washers) did not function to reinforce these tasks. Table 1 shows the children's test scores as well as their existing repertoires at the time of the experiment as determined by the CABAS[R] International Curriculum and Inventory of Repertoires for Children From Pre-School Through Kindergarten (Greer & McCorkle, 2009; Reed, Osborne, & Corness, 2007; Waddington & Reed, 2009). Similar to participants, none of the peer confederates had an official diagnosis; rather, at the preschool level, they were classified as "Preschoolers With a Disability," consistent with state regulations, and functioned at the beginning reader and writer levels of verbal function. The peer confederates did not receive training prior to participating in the intervention because the task completed during the intervention was a task already in repertoire. Moreover, the peer confederates did not receive a backup reinforcer for washers earned during the intervention but performed readily for the washers during the intervention.
Table 1 Description of Participants Participant Age Level of verbal Standardized test Skills in capability scores repertoire at onset of experiment A 4 Listener, Vineland Adaptive Follows speaker, early Behavior multistep reader, early Scales-Interview directions writer Edition; Generalized Communication, 86 match to sample Daily Living Uses mands and Skills, 109 tacts with Socialization, 87 autocides Motor Skills, 100 Preschool Has appropriate Language play skills Scale-Fourth Uses social Edition: Auditory verbal Comprehension, 69 exchanges, Expressive including Communication, 71 sequelics and Total Language conversational Score, 67 units B 4 Listener, Vineland Adaptive Follows speaker, early Behavior multistep reader, early Scales-Interview directions writer Edition: Adaptive Generalized Behavior, 126 match to sample Daily Living Uses mands and Skills, 78 tacts with Socialization, 70 autocides Motor Skills, 88 Communication, 86 Preschool Has appropriate language play skills Scale-Fourth Uses social Edition: Auditory verbal Comprehension, exchanges, 126 Expressive including Communication, sequelics and 133 conversational units C 4 Listener, Vineland Adaptive Follows speaker. early Behavior multistep reader, early Scales-Interview directions writer Edition; Adaptive Generalized Behavior, 76 match to sample Communication, 76 Uses mands and Daily Living tacts with Skills, 78 autoclitics Socialization, 74 Motor Skills, 88 Preschool Has appropriate Language play skills Scale-Fourth Uses social Edition: Auditory verbal Comprehension, 75 exchanges, Expressive including Communication, sequelics and 79 conversational units
Materials and Setting
The pre-and post-observational intervention assessments for performance and learning tasks were collected in a small room in the school (2.438 x 3.048 m) devoted to the experiment. See Table 2 for description of each participant's performance and learning tasks. During the pre-and post-observational intervention assessments, the room was equipped with one child-sized table, one child-sized chair for the participant, and one adult-sized chair for the experimenter. All toys and other potential distracters were removed from the room so that the table and chairs were the only items in the room. Only the experimenter and the independent observer were present in the room. No peers were present, to avoid possible social facilitation effects (Heyes & Galef, 1996). Social facilitation refers to the tendency of humans to perform better in the presence of other humans (Guerin, 1994).
Table 2 Description of Dependent Variables Participant Learning tasks selected as dependent Performance variables task selected as dependent variable A Tact pictures of food items Point to Tact pictures clothing of community helpers Point lo textual and items on a intraverbal responses to numbers page in a field of 10 B Tact and point to coin values Point to Give number of objects upon request clothing State ordinal position items on a page in a field of 10 C Tact and point to coin values Point to Rhyme words clothing Tact activities associated with time of day items on a page in a field of 10
The intervention was conducted in the same room as the pre- and post-intervention assessments. However, during the observational intervention, two child-sized desks and chairs were arranged so that the participant and peer confederate were seated facing a Masonite partition that separated them from the experimenter. The participant and peer confederate could not see or hear the experimenter, who was located behind the partition, at any point during the intervention, nor were they informed of the presence of the experimenter behind the Masonite partition. The partition that separated the children from the experimenter was 1.27 m wide and 1.22 m high. Figure 1 shows the configuration of the intervention environment.
Two white chutes (15.2 cm long and 3.8 cm in diameter) constructed of PVC pipe were set at a 90-degree angle in holes in the partition that connected the experimenter's side of the partition to the participant and peer's side of the partition (see Figure 2). The chutes made it possible for the delivery of the washers to the peer without the peer seeing the experimenter. The chutes were assembled so that the metal washers could easily slide down them into the transparent plastic cups located on the participants' and peer confederates' tables. The plastic cups (11 cm high and 8.5 cm in diameter) located on either side of the partition were secured to their respective tables using a heavy-duty magnet, which prevented either child from moving the cups from their positions on the table. Moreover, the positioning of the cup directly underneath the chute disallowed the peer confederates from removing the washers that were dispensed into their cups, as there was not sufficient space for them to insert their hands or fingers to access to washers. Figure 2 shows the location of the cups.
In addition to the partition that separated the participant and the peer confederate from the experimenter, there was also a Masonite partition that separated the participant and peer confederate so that they could not see each other's task performance. This partition was 1.016 m long and 1.22 m high. This partition was constructed high enough that the participant and the peer confederate could not view each other (i.e., they could not view each other's faces or the performance of tasks). The partition also had outside walls that were located on the left and right sides of the participant and peer confederate, respectively (see Figure 1). All partition walls were attached to each other using door hinges. A transparent plastic cup was located on the far corner of each child's desk, positioned in front of each child. A piece of clear Plexiglas (20.3 x 25.4 cm) was built into the partition that separated the participant from the peer confederate so that both children could view the plastic cups located on the other's desk (see Figure 2). The arrangement allowed the participant to see and hear the delivery of the metal washer into the peer confederate's cup, but the participant could not observe the task that the peer confederate was engaged in at his desk or the peer's face. The performance task that both the participants and peer confederates engaged in was a sorting task, which is described later.
An adult assistant brought the participant and the peer confederate from their respective classrooms to the experimental room. After seating the participant and peer confederate and providing instructions to both, the adult assistant left the room and observed the participant and peer through a baby monitor, via a camera that was mounted above and behind the heads of the participant and peer confederate. The tasks that the participant and peer confederate were engaged in were also recorded using a wireless D-Link Securicam that was connected to the experimenter's laptop computer on the experimenter's side of the partition. The camera, which was connected to an approximately 2-m-long pole mount made of PVC pipe, was positioned 40.6 cm above the children's heads in order to capture their responses to tasks.
Pre-and post-intervention response assessments consisted of the following:
1. A counterbalanced pre-intervention ABAB (or BABA) functional analysis, where B represents washers and A represents primary reinforcers for learning and performance for each child. This involved comparisons of correct and incorrect responses per minute to performance tasks under repeated sessions using the metal washer as a consequence for responding and repeated sessions of edible consequences for performing.
2. Multiple baselines across participants for three acquisition tasks in which only metal washers were delivered contingent upon correct responses to blocks of 20 instructional trial presentations before and after the intervention.
During the pre-and post-intervention assessments for the performance task, Participants A and C began with metal washer reinforcement in the first phase of the pre-intervention condition and then received alternating phases of edibles, metal washers, and edibles (in a BABA sequence, where B was the metal washer condition and A the preferred edibles condition). Participant B began with edibles in the first stage followed by alternating phases of metal washers, edibles, and metal washers (in an ABAB design). Participants received metal washers or edibles in this same manner of alternating phases following the observational intervention.
The baselines of the instructional tasks occurred prior to the observational intervention and were stopped when the participants demonstrated that they were not learning when reinforcement operations for correct responses using the metal washers alone and corrections were in effect. Corrected responses did not receive reinforcement operations. After the observational intervention, the same reinforcement operations and corrections were resumed with the instructional tasks. For the learning tasks, the participants' pre-and post-intervention sessions were staggered to control for instructional history. That is, Participants B and C remained in baseline while Participant A completed the intervention condition until washer reinforcement was shown to have become a conditioned reinforcer for Participant A's learning tasks. When Participants B and C's correct responding remained at low levels while Participant A's level of correct responding increased following the intervention, Participant B was introduced to the intervention. Similarly, in staggered fashion, Participant C remained in baseline while Participant B underwent the intervention condition and subsequently returned to post intervention sessions for learning tasks. When Participant B's levels of correct responses to learning tasks increased following the intervention condition, Participant C entered the intervention condition.
Pre-and post-intervention performance task. Data during the pre-and postintervention assessments for the performance task were collected as correct and incorrect responses per minute to 10 trial sessions that were timed. Correct responses were recorded as a plus, and incorrect responses were recorded as a minus and converted to rate (number per minute correct and incorrect).
Pre-and post-intervention learning tasks. Data during the pre-and postintervention assessments for each participant's learning task were collected as responses to blocks of 20 instructional trials. Correct responses were recorded as a plus, and incorrect responses were recorded as a minus.
Observational intervention. Data during the observational intervention were converted to correct and incorrect responses per minute to 10 trial sessions. Correct sorting responses were recorded as a plus, and incorrect sorting responses were recorded as a minus. In addition to data collection of correct and incorrect sorting responses, data were also recorded on the participants' vocal and non-vocal mands during the intervention sessions at a later time by viewing the recorded sessions on the experimenter's laptop computer. Vocal mands were defined as the participants' vocally requesting access to the metal washers they observed the peer confederates receive. Examples of vocal mands included statements such as "Put some in here!" "I'm doing it too!" and "I want some." Non-vocal mands consisted of the participants' non-vocally indicating that they wanted to gain access to the washers. Examples of non-vocal mands included pulling on the chute located on their side of the partition, pulling on their transparent plastic cup, putting their fingers in their cup, and pointing to their peer's cup through the Plexiglas built into the center partition. Data were also collected on peer confederate statements and physical postures associated with receiving the washers throughout the intervention condition. Examples of these statements included "I got one!" or "I'm winning!" and examples of physical postures emitted by the peer confederates included pointing to and/or putting their fingers into their cup following the delivery of a washer.
Test of maintenance of reinforcement effect. Instruction across the initial three learning tasks for each participant were delivered and measured as correct and incorrect responses to instructional trial presentations consistent with those used in the experimental analyses of learning before and after the initial intervention. Correct responses were recorded as a plus (+), and incorrect responses and refusals to respond were recorded as a minus (-).
Pre-and post-intervention performance task. The performance task used for each participant was one that all of the participants could readily perform (i.e., they were maintenance tasks). This performance task was the same for each participant and is described in Table 2. Performance sessions consisted of blocks of 10 trials that were timed using a digital timer in order to determine the number of correct and incorrect responses emitted per minute. For each trial, the experimenter presented the correct matching stimulus with one incorrect match along with the vocal antecedent "point to [clothing iteml" and either (a) delivered a food item or a metal washer into a transparent plastic cup on the table in front of the participant for correct responses or (b) removed the stimuli following an incorrect response or after 3 s in which no response was emitted. A new trial was immediately presented.
The observational conditioning intervention was implemented when performance task data showed that the metal washers did not function to reinforce responding. This was determined by the number of correct responses emitted by the participants showing a descending trend or extinction of correct responses whereby participants ceased responding altogether or ceased emitting correct responses. Following the intervention condition, the performance task was reintroduced comparing metal washer reinforcement versus edible reinforcement for the performance task.
Pre-and post-intervention learning tasks. We also tested the effects of metal washer reinforcement operations in instructional trials with three tasks the participants did not have in repertoire before and after the intervention (see Table 2). Each learning task was presented as responses to blocks of 20 instructional trials, or sessions, and was not timed. For each task, the participant received a metal washer delivered into his transparent plastic cup following correct responses to instructional trials. Incorrect responses or refusals to respond resulted in the correction procedure, in which the experimenter repeated the instructions and modeled the correct response for the participant, who then repeated the correct response. The participants' corrected responses were not reinforced. Metal washers were not delivered for incorrect responses or refusals to respond.
When the data for each participant across each learning task showed that the metal washers did not function to reinforce acquisition as determined by the absence of an ascending trend, the observational conditioning intervention was implemented. Following the intervention condition, the learning tasks were reintroduced under conditions in which metal washers were delivered contingent upon correct responses and corrections were delivered for incorrect responses, as was done in the baseline conditions.
Observational intervention. The intervention condition consisted of a sorting task that the participants had in repertoire. Each participant was seated next to a peer confederate separated by a partition so that neither could view the other or the task in which the other was engaged as described previously. Due to the Plexiglas built into this partition, the participant was able to view the white chute leading to the transparent plastic cup on the peer confederate's side of the partition. The participant could observe the delivery of the metal washers into the peer confederate's cup. In addition to the partition separating the participant from the peer confederate, both the participant and peer confederate were seated facing a partition that separated them from the experimenter.
Upon entering the experimental room, the adult assistant who delivered the participant and peer confederate to the room instructed both children to sit at their respective tables and to engage in the sorting task already set up on their table. The participants and peer confederates were told the following by the assistant:
Please sit at your desk, and when I leave the room you can begin sorting the different colored shapes into one of these four containers. Put the red shapes into the container that has a picture of the red shapes, put the blue shapes into the container that has a picture of the blue shapes, put the orange shapes into the container that has a picture of the orange shapes, and put the tan shapes into the container that has a picture of the tan shapes. I will be back in a few minutes.
The adult assistant then exited the room, and the children began responding to the sorting task. Correct and incorrect sorting responses emitted by the peer confederate resulted in the experimenter delivering a metal washer down the peer confederate's white chute into his transparent plastic cup. The participant could see and hear the delivery of the metal washers into the peer confederate's cup. However, at no time did the participant receive metal washers in his own cup.
Intervention sessions continued based on each individual's responding in the intervention and ranged from 20 to 28 sessions. Criterion for termination of the procedure occurred under one of two conditions: (1) correct responding demonstrating extinction effects (Participants A and B) or (2) a mean number of sessions was met, calculated from previous studies in which the observational intervention was used (Participant C; Greer & Singer-Dudek, 2008; Greer et al., 2008; Singer-Dudek et al., 2008; Singer-Dudek, Oblak, Greer, & Pereira-Delgado, 2007). One to two intervention sessions per day over a period of 1 to 2 weeks were implemented during the participants' and peer confederates' free time.
Test for maintenance of reinforcing effect. Instructional trials to criteria constituted a measure of the participants' rates of learning. This was a measure of the number of instructional trials the student required to achieve the mastery criterion of 900/0 correct responses in 20 instructional trial sessions. The numbers of instructional trials for each of the learning tasks for each participant were calculated by summing the number of objectives achieved for 400 instructional trials presented to the participant before and after the implementation of washer reinforcement. The 400 trials, under each reinforcement operation condition, were then divided by the number of objectives achieved for each participant, resulting in the participants' instructional trials to mastery under each condition.
During the primary reinforcer condition, across each of the three learning tasks for each participant, the participant and the participant's teacher sat across from or beside each other at a child-sized table. The teacher presented instruction in the form of instructional trials in which correct responses to instructional trials were reinforced with a preferred reinforcer that the participant typically worked for. Incorrect responses or refusals to respond resulted in the correction procedure, in which the teacher repeated the instruction and modeled the correct response for the participant to repeat. Corrected participant responses were not reinforced.
Prior to the washer reinforcer condition, participants were not given access to the metal washers during their respective time periods after the initial intervention. During the washer reinforcer condition, the teacher presented instruction in the form of instructional trials in which correct responses were reinforced with the teacher dispensing a metal washer into the transparent plastic cup in front of the participant. No other form of reinforcement was delivered. At no time were the metal washers exchanged for a backup reinforcer; rather, the metal washers themselves were the reinforcement. Incorrect responses and refusals to respond resulted in the same consequences as described during baseline conditions.
The participants' correct and incorrect responses on both the performance and learning tasks in the pre- and post-intervention conditions were recorded immediately by the experimenter and, for many sessions, by an independent observer who observed recordings of the sessions. This second observer collected data on the participants' responses for 48% of performance tasks with 100% interobserver agreement recorded, for 50% of the acquisition task sessions with 100% interobserver agreement recorded, and for 100% of the intervention sessions with a mean of 95% interobserver agreement (range 85%-98%). The second observer viewed the recorded intervention sessions and collected data on the participants' correct and incorrect sorting responses at a later time. Interobserver agreement was taken for 47% of total tests for the maintenance of conditioned reinforcers' effect sessions, with 100% agreement recorded.
Pre-and Post-Intervention Performance Conditions
Figure 3 shows per-session response rates during the performance tasks for Participants A and B across the pre-and post-intervention conditions. Figure 4 shows per-session response rates for Participant C's performance task. In the pre-intervention condition, all three participants emitted higher numbers of correct responses, with corresponding rates of incorrect responses during each of the edible reinforcement conditions. During both metal washer phases, all three participants' correct responses demonstrated extinction effects following several sessions with metal washer reinforcement in place. During the second metal washer phase, there were immediate decreases in correct responses from the primary reinforcer condition that immediately preceded it. These results were consistent across all three participants.
Following the observational intervention, the data showed high rates of correct responses with corresponding low rates or no incorrect responses for all three participants under the primary reinforcer and metal washer conditions. Following the intervention, correct responding during metal washer and edible phases were comparable for all participants and comparable to responding in the prior studies (Greer & Singer-Dudek, 2008; Greer et al., 2008; Singer-Dudek et al., 2008; Singer-Dudek et al., 2011).
Pre-and Post-Intervention Learning Conditions
Figures 5-7 show the numbers of correct responses for all three participants across Learning Tasks 1-3, presented before and after the intervention. Participant A's correct responses across the three learning tasks ranged from 2 to 13 across all three tasks when the metal washers were delivered as reinforcement for responses to instructional trial presentations. Following the intervention, the numbers of correct responses with metal washer reinforcement ranged from 15 to 20 correct responses. Mastery level responding (90% or higher correct responses for two consecutive sessions) was achieved in two or three sessions for all three learning tasks for Participant A (see Figures 5-7). Participant B emitted low numbers of correct response (range 0 to 7 correct responses in 20 instructional trial blocks) across his three learning tasks prior to the intervention. Following the intervention, mastery level responding occurred within six sessions, two sessions, and three sessions for Learning Tasks 1-3, respectively. Correct responses ranged from 15 to 20 in 20 instructional trial blocks (see Figures 5-7). Participant C's pre-intervention data showed a steadily descending trend in correct responses across all three learning tasks. Correct responses ranged from 0 to 8 for blocks of 20 instructional trials. Following the intervention, Participant C's correct responses increased and he achieved the mastery criterion for Learning Tasks 1 and 3 following eight and five sessions, respectively. Participant C did not achieve mastery level responding for Learning Task 2, as it was later determined by the participant's classroom teacher that a prerequisite skill that may have allowed him to acquire the skill more quickly may have been missing; however, an ascending trend was shown, indicating that learning occurred (see Figure 6).
Test for Maintenance of Reinforcing Effect
The effects for each participant are described in terms of their instructional trials to criteria prior to and during the implementation of washer reinforcement for correct responses to three learning tasks. Figures 8-10 show instructional trials to criteria for each individual learning task, respectively, for each participant. Overall, participants' rate of learning remained at relatively stable levels when metal washer reinforcement was in place compared to the participants' rate of learning under reinforcement with preferred items.
Learning Task 2 for Participant 13 (see Figure 9) and all three learning tasks for Participant C (see Figures 8-10) showed that instructional trials to criteria remained constant when a known reinforcer was in place as well as when washer reinforcement was in place. In some cases, instructional trials to criteria decreased during the implementation of washer reinforcement. This was shown for Learning Task 1 for Participant A (see Figure 8), and Learning Task 3 for Participant B (see Figure 10). Finally, in some cases, participant rate of learning decreased slightly, as was shown in Participant A's second and third learning tasks (see Figures 9 and 10).
Similar to prior studies (Greer & Singer-Dudek, 2008; Greer et al., 2008; Singer-Dudek et al., 2008; Singer-Dudek et al., 2007), the observational intervention acted to convert the neutral stimuli to conditioned reinforcers. The contribution of the current study shows that the experimenter need not be present in order for this conditioning effect to occur at levels consistent with the prior studies where an experimenter was present. As was shown in prior studies, increases in correct response rates to a performance task as well as increases in the number of correct responses across three learning tasks were found with metal washer reinforcement in place for all participants following the intervention.
Several of the learning tasks showed immediate mastery or almost mastery responding following the intervention, as found in prior studies (Greer & Singer-Dudek, 2008; Greer et at., 2008), which may have been instances of "latent learning."
Therefore, prior to the intervention, the participants may have learned from the corrections they received; however, because the washers were not functioning to reinforce correct responding, participants did not emit correct responses. These effects were shown for all of Participants A and B's learning tasks (see Figures 5-7). However, there were several tasks where a more gradual learning trend was shown following the intervention. This was evident for all of Participant C's learning tasks (see Figures 5-7).
Over the course of the intervention, particularly during Participants A and B's intervention conditions, the peer confederates also began emitting statements such as "I got one!" or "I'm winning!" upon receiving the metal washers in their cups. As sessions continued, Participants A and B's correct responding went into extinction. Following several sessions of observing their peer receive the washer, Participants A and B emitted some vocal and non-vocal mands for washers, and a corresponding lower rate of correct responses was shown. Participants began emitting such phrases as "Put some in here!" as they pointed to their own cup, or "I'm doing it too!" Participants A and B emitted non-vocal mands as well, in the form of pulling on their own white chute and trying to pull their plastic cups off of the table upon seeing their peers receive the metal washers. Variable numbers of these behaviors continued until correct responding went into extinction. Participant C's responding, however, continued at relatively stable levels and was not shown to decrease over time. He also emitted fewer vocal and non-vocal mands during the intervention. The peer confederate vocal and non-vocal statements and posturing about receiving the washers were lowest for Participant C's intervention condition compared to these responses for Participants A and B. The lack of mands appeared to correlate with his continued stable responding. Some participants in the previous studies (Greer & Singer-Dudek, 2008; Singer-Dudek et al., 2008) also demonstrated stable responding during the intervention. Although correct responding did not cease for these participants in the prior studies and Participant C in this study, the return to performance and learning tasks showed that the stimuli did acquire reinforcing properties. It is possible that participants who emit stable responding in the intervention have stronger histories of verbally governed responding (Catania & Mathews, 1982; Catania, Mathews, & Shimoff, 1990).
In prior studies, participants could observe their peers' facial reactions, whereas in the present study, they could not. This would appear to rule out the possible effects of the peer confederates' expressions on the intervention outcomes.
Following the reintroduction of the washers to the participants contingent upon correct responses to learning tasks, participants' rates of learning under metal washer reinforcer were comparable to their rates of learning when their known reinforcers were in place. In some cases, rates of learning were slightly faster with washer reinforcement in place, and in some cases, the opposite effect occurred; however, overall rates of learning across all three learning tasks and participants were comparable (see Figures 8-10). Results provide evidence that neutral stimuli conditioned as a reinforcer as a function of the observational intervention implemented in previous studies (Greer & Singer-Dudek, 2008; Greer et al., 2008; Singer-Dudek et al., 2008; Singer-Dudek et al., 2007), as well as in the current study, do maintain their reinforcing effectiveness for at least 6 to 10 weeks.
Emitting behavior or acquiring new behaviors as a function of observation of others is regarded as social learning. Experiments with children have identified types of social learning, including the capability to learn new operants as a function of observation of reinforcement and corrections received by others (Catania, 2007; Greer et al., 2004), the capability for observational learning of new operants (Pereira-Delgado & Greer, 2009), acquiring language incidentally where no consequences received by others is observed (Greer & Longano, 2010; Greer, Stolfi, & Pistoljevic, 2007; Horne & Lowe, 1996), and conditioned reinforcement by observation. The acquisition of conditioned reinforcers by observation has been reported in non-humans (Dugatkin, 1996, 2000) and in humans as reported herein.
In the case of the social learning of conditioned reinforcers by young children, the presence of the experimenter or an adult in the observational conditions does not appear to be a necessary component of the conditioning outcome that has now been identified in several experiments (Greer & Singer-Dudek, 2008; Greer et al., 2008; Singer-Dudek et al., 2008; Singer-Dudek et al., 2011). Nor does it seem necessary for the observer to see peers' reaction to receiving the stimuli. Clearly, the role of the peer in the conditioning process needs to be studied. In a recent unpublished dissertation, Oblak (2010) found that having a participant deliver neutral stimuli (metal bolts) to a peer when the participant did not receive the bolts acted to condition the bolts as reinforcers for learning and performance. This suggests that the peer is an important component, although no condition was implemented in which the participants dropped the bolt into a box when a peer was not present. Thus, although the peer seems to be a critical component, we simply do not know, to date, that this is the case.
If the effect is the result of vicarious Pavlovian second-order conditioning or stimulus contingency relations, the adult does not appear to be key to the effect. This leaves the peer as the possible conditioned reinforcer in the pairing process, if the source of the effect is indeed attributable to a kind of vicarious Pavlovian pairing or stimulus contingency relation. For example, in the present study, we recorded statements by the peer confederates such as "I'm winning!" This may have had a stimulus enhancement effect, where the statements focused the target students' attention on the washers, and this may have enhanced the potential of the stimuli to acquire reinforcing attributes. Thus, such statements may have played a role in the establishment of the washers as reinforcers. While we think the presence of the peer and perhaps the peers' statements may have been a likely variable, it is not the only possibility. For example, social facilitation, or conditions in which the presence of others facilitates behavior, may have played a role (Guerin, 1994). However, it should be noted that no peers were present in the pre- and post-performance tests, and no other children in the classroom were receiving the washers. Thus, in the first experiment, social enhancement did not appear to play a role. However, it could have in the second experiment, had the other children in the classroom received washers also. They did not; hence, social facilitation does not seem likely.
While other factors may play a role, the conditions in which the effect occurs are inherently of a social nature. The utility of social learning in many cases, but not all, is superior to trial-and-error learning (Rendell et al., 2010). Indeed, a recent study reports that acquiring a social learning capability--in this case, the ability to acquire language incidentally, or naming--led to significantly faster rates of learning by kindergarteners and first graders as a function of experimenter demonstrations prior to instructional trials, whereas, prior to the participants' acquiring naming, demonstrations did not accelerate learning (Greer, Corwin, & Buttigieg, 2011). This suggests, as do the observational learning studies cited earlier, that learning by observation is in some cases more efficient for children with naming than direct instruction involving reinforcement and corrections alone, while for children without naming, it is not.
However, the utility of social learning in the above cases refers to the learning of operants. In the case of the study reported herein, and in related referenced studies, reinforcers were established that we and others (Williams, 1994) argue are conditioned reinforcers. It is possible that the conditions involve differentially correlated pairings or stimulus contingency relations in some way (Fantino, 2008). Perhaps the presence of others somewhere in the general vicinity (i.e., differentially correlated social conditions) is the conditional property, and that is a target for future study. For example, would participants for whom the neutral stimuli have been conditioned, via the observation conditions we describe, respond continually for the conditioned stimuli under automated conditions in which no other children or adults are present--conditions like those we used to eliminate the adult presence? However, in the pre- and post-performance tests in the current study, no peers were present, although adults were present. Even if sessions were automated as described previously, perhaps only some daily exposure to peers, even outside of the automated tests, is needed. How to test that has not been determined, although in the nonhuman literature, a control condition in which others do not receive the stimuli is suggested. This is in fact what we did.
Regardless of the unknowns, the finding does seem to be robust. Moreover, it may be more efficient than traditional pairing procedures in developing new reinforcers in applied settings for participants like those reported in this and prior related studies. However, whether the social learning procedure is more efficient is likely dependent on the individuals' having certain repertoires in place. Perhaps they need to have the capability to learn new operants by observation. Pereira-Delgado and Greer (2009) found that children who could not learn new operants from observation could do so after an intervention involving learning to monitor the accuracy of responses of others when reinforcement was contingent on accurate observations. This finding suggests that observational learning of operants is social. It may be that the capability of learning operants may need to be present in order for children to acquire conditioned reinforcers under the conditions we described. Alternately, the capability to learn conditioned reinforcers by observation may come first. The study by Pereira-Delgado and Greer did not include tests for the presence of conditioned reinforcement by observation. It is also possible that the two are independent. We simply do not know, and these are questions for further research.
Social learning is an important way in which humans and non-humans acquire critical repertoires. It is the primary way in which much of our verbal behavior is learned (Greer & Longano, 2010; Greer & Ross, 2008; Hart & Risley, 1995; Horne & Lowe, 1996). Indeed, at least one function of tacts is being reinforced by attention or approval, and approval has been conditioned as a reinforcer for learning and performance, as reported by Greer et al. (2008). Thus, conditioned reinforcement of adult approval or attention by observation may play a role in the emergence of tacts in young children. The social learning of conditioned reinforcers appears to play an important role in other human activities also. The effects of advertising on fashion and the expansion of individually useful and detrimental conditioned reinforcers (e.g., smoking) may be related to processes like those we report. Certainly, the social establishment of reinforcers in adults has not been tested; hence, it may be premature to suggest the spread of the phenomenon to adults and stimuli associated with fashion. Our data suggest that such analyses might prove useful in the analysis of the acquisition of socially learned addictions such as smoking. It is possible that observing peers smoke, even when the observer's first experience with smoking is aversive, enhances the reinforcing aspect, since the observed person has the cigarettes and the observer does not. Thus, the situation may be similar. Future research might test this possibility.
CATANIA, A. C. (2007). Learning (Interim 4th ed.). Cornwall-on-Hudson, NY: Sloan Publishing.
CATANIA, A. C., & MATTHEWS, B. A. (1982). Instructed versus shaped human verbal behavior: Interactions with nonverbal responding. Journal of the Experimental Analysis of Behavior, 38, 233-248.
CATANIA, A. C., MATHEWS, B. A., & SFIIMOFF, E. H. (1990). Properties of rule-governed behavior and their implications. In D. E. Blackman & H. Lejeune (Eds.), Behavior analysis in theory and practice (pp. 215-230). Hillside, NJ: Erlbaum.
COOK. M., MINEKA, S., WOLKSTEIN, B., & LAITSCH, K. (1998). Conditioning of snake fear in unrelated rhesus monkeys. Journal of Abnormal Psychology, 94, 591-610.
DORRANCE, B. R., & ZENTALL, T. R. (2002). Imitation of conditional discriminations in pigeons (Columbia livia). Journal of Comparative Psychology, 116, 277-285.
DUGATKIN. L. A. (1996). The interface between culturally-based preference and genetic preference: Female mate choices in Poecilia reticulata. Proceedings of the National Academy of Sciences, 93,2770-2773.
DUGATKIN, L. A. (2000). The imitation factor. New York, NY: Free Press.
DUGATKIN, L. A., & GODIN, J. G. (1992). Reversal of female mate choice by copying in the guppy (Poecilia reticulata). Proceedings of the Royal Society of London B, 249, 179-184.
FANTINO, E. (2008). Choice, conditioned reinforcement, and the Prius effect. The Behavior Analyst, 31, 95-111.
GREER, R. D., CORWIN, A., & BUTTIGIEG, S. (2011). The effects of the verbal developmental capability of naming on how children can be taught. Psychological Research Record, 1, 23-54.
GREER, R. D., KEOHANE, D., ME1NCKE, K., GAUTREAUX, G., PEREIRA, J. A., CHAVEZ--BROWN, M.. & YUAN, I.. (2004). Key instructional components of effective peer tutoring for tutors, tutees, and peer observers. In D. J. Moran & R. W. Malott (Eds.), Evidence-based educational methods (pp. 295-334). New York, NY: Elsevier/Academic Press.
GREER, R. D., & LONGANO, J. (2010). Naming a rose: How we may learn to do it. The Analysis of Verbal Behavior, 26, 73-106.
GREER, R. D., & MCCORKLE, N. P. (2009). CABAS[R] international curriculum and inventory of repertoires for children from pre-school through kindergarten (4th ed.). Yonkers, NY: CABAS[R]/Fred S. Keller School.
GREER, R. D., & ROSS, D. E. (2008). Verbal behavior analysis: Developing and expanding verbal capabilities in children with language delays. Boston, MA: Allyn 84 Bacon/Merrill.
GREER, R. D., & SINGER-DUDEK, J. (2008). The emergence of conditioned reinforcement from observation. Journal of the Experimental Analysis of Behavior, 89, 15-39.
GREER, R. D., SINGER-DUDEK, J., LONGANO, J., & ZRINZO, M. (2008). The emergence of praise as conditioned reinforcement as a function of observation in preschool and school age children. Revista Psychologie Mexico, 25, 5-26.
GREER, R. D., STOLFI, L., & PISTOLJEVIC, N. (2007). Emergence of naming in preschoolers: A comparison of multiple and single exemplar instruction. European Journal of Behavior Analysis, 8, 119-131.
GUERIN, B. (1994). Analyzing social behavior: Behavior analysis and the social sciences. Reno, NV: Context Press.
HART, B., & RISLEY, T. R. (1995). Meaningful differences in the everyday experiences of young American children. Baltimore, MD: Paul H. Brookes.
HEYES, C. M., & GALEF, B. G. (1996). Social learning in animals: The roots of culture. San Diego, CA: Academic Press.
HORNE, P. J., & LOWE, C. F. (1996). On the origins of naming and other symbolic behavior. Journal of the Experimental Analysis of Behavior, 65, 185-241.
OBLAK, M. (2010). The effects of the delivery of a neutral item to preschool peers on the acquisition of conditioned reinforcers (Unpublished doctoral dissertation). Columbia University, New York, NY.
PEREIRA-DELGADO, J. A., & GREER, R. D. (2009). The effects of peer monitoring training on the emergence of the capability to learn by observing instruction received by peers. The Psychological Record, 59, 407-434.
REED, P., OSBORNE, L. A., & CORNESS, M. (2007). Brief report: The effectiveness of different home-based behavioral approaches to early teaching intervention. Journal of Autism and Developmental Disorders, 37(9), 1815-1921. doi:10.1007/s10803-006-0306-8
RENDELL, L., BOYD, R., COWNDEN, D., ENQUIST, M., ERIKSOSON, K., FELDMAN, M.
LALAND, K. N. (2010). Why copy others? Insights from social learning strategies tournament. Science, 328(5975), 208-213.
SINGER-DUDEK, J., GREER, R. D., & SCHMELZKOPF, J. (2008). The effects of an observational intervention on the acquisition of reinforcing properties of a previously neutral stimulus. Journal of Early and Intensive Behavior Intervention, 5, 57-74.
SINGER-DUDEK, J., OBLAK, M., & GREER, R. D. (2011). The emergence of books as conditioned reinforcers for acquisition and maintenance as a function of an observation intervention with preschool children: An extension of conditioned reinforcement by observation. Journal of Applied Behavior Analysis, 44, 421-434.
SINGER-DUDEK, J., OBLAK. M., GREER, R. D.. & PEREIRA-DELGADO, J. (2007, March). Books as a conditioned reinforcer for performance and acquisition tasks as a function of observation. Paper presented at the Second Annual International CABAS Conference, Suffern, NY.
WADDINGTON, E. M., & REED, P. (2009). The impact of using the "Preschool Inventory of Repertoires for Kindergarten" (PIRKO) on school outcomes on children with autism spectrum disorders. Research in Autism Disorders, 3, 809-827.
WILLIAMS, B. A. (1994). Conditioned reinforcement: Neglected or outmoded explanatory construct? Psychonomic Bulletin and Review, 1(4), 457-475.
Michelle Zrinzo Is now at Hawthorne Foundation, Inc., New York, New York.
The first author completed these studies as part of the requirements for the PhD under the sponsorship of the second author.
Correspondence concerning this article should be sent to Michelle Zrinzo, Hawthorne Foundation, 156 William St., New York, NY 10038. E-mail: email@example.com
Michelle Zrinzo and R. Douglas Greer
Columbia University Teachers College and Graduate School of the Arts and Sciences
|Printer friendly Cite/link Email Feedback|
|Author:||Zrinzo, Michelle; Greer, R. Douglas|
|Publication:||The Psychological Record|
|Date:||Jan 1, 2013|
|Previous Article:||How sex and college major relate to mental rotation accuracy and preferred strategy: an electroencephalographic (EEG) investigation.|
|Next Article:||Go/no-go procedure with compound stimuli: effects of training structure on the emergence of equivalence classes.|