Skip to main content
NCBI home page
Behavior Analysis in Practice logoLink to Behavior Analysis in Practice
. 2021 Jan 19;14(1):141–160. doi: 10.1007/s40617-020-00511-1

The Effect of the Establishment of Conditioned Reinforcement for Reading Content on Second-Graders’ Reading Achievement

Lara M Gentilini 1,✉, R Douglas Greer 1
PMCID: PMC7900349  PMID: 33732584

Abstract

There remains a gap in the current literature as to how to reliably measure and increase students’ “voluntary reading,” based on research suggesting a relation between reading amount and reading achievement. We tested the effect of the establishment of conditioned reinforcement for reading via a collaborative shared reading (CSR) conditioning procedure on eight 2nd-grade students with and without learning disabilities and developmental disorders. This conditioning procedure was composed of opportunities for reciprocal reading and collaboration on comprehension and vocabulary tasks related to the reading content, such that partners (teacher–participant or participant–participant) were required to work together. We utilized a combined small-n experimental-control simultaneous-treatment design with a single-case multiple-probe design nested within each small group in order to compare within- and between-group differences for participants in the CSR procedure with a teacher or peer. All participants for whom conditioned reinforcement for reading was established (n = 7) demonstrated gains in reading achievement after a maximum of nine sessions (412 min), with grade-level increases between 0.2 and 2.5 on measures of reading comprehension and between 0.3 and 3.1 on measures of vocabulary. The students in the teacher-yoked condition (n = 3) demonstrated more significant gains in their average increases in achievement, although the peer-yoked procedure was also effective and possibly more viable in a classroom setting. These results suggest that a CSR procedure with a teacher or peer should be considered as a means of increasing the reading achievement of early elementary students via increases in the reinforcement value of reading.

Keywords: conditioned reinforcement, conditioned seeing, derived relational responding, reading comprehension, reinforcement value, vocabulary

The national push (National Reading Panel, 2000) for increased voluntary reading among elementary-age students is supported by extensive research demonstrating positive relations between reading motivation, self-reported reading amount, and reading achievement (Becker et al., 2010; Guthrie & Wigfield, 2000; Stutz et al., 2016; Wigfield et al., 2016). For example, there are positive relations between reading amount and reading achievement (Heyns, 1978), as well as between intrinsic reading motivation and later reading achievement (Becker et al., 2010; Gottfried et al., 2001; Wang & Guthrie, 2004). Children who read in their leisure time perform better on measures of reading achievement when controlling for cognitive abilities (Cipielewski & Stanovich, 1992), with reading amount outside of school as the best predictor of reading achievement in elementary school (Anderson et al., 1988).

Although instructional reading programs cite improvements in voluntary reading as a major goal, a significant percentage of elementary-age students do not spend their leisure time reading for pleasure or information (Kurth & Kurth, 1987; Ysseldyke & Algozzine, 1983). Furthermore, there remains a gap in the current literature as to how to reliably measure and increase students’ voluntary reading, with much of the present research on voluntary reading relying on ad hoc instruments and nonoperational constructs (Stutz et al., 2016; Wigfield et al., 2016). Whereas voluntary reading may be associated with constructs of reading “interest” or “motivation” from some psychological perspectives, other theoretical models—such as the theory of verbal-behavior development—frame the moment-to-moment selection of the student’s attention to the text in terms of one’s reinforcement value of reading (Greer, 1983; Greer & Ross, 2008; Greer & Speckman, 2009; Moore, 2008; Skinner, 1987). We operationally define reinforcement value of reading as the student emitting observing responses to textual stimuli, or demonstrating reinforcement stimulus control, as supported by decades of research (Fantino & Romanowich, 2007; Greer, 2020).

Verbal behavior research on the reinforcement value of reading is precipitated by studies of children’s conditioned reinforcement for the observation of various stimuli, with faster and more immediate learning as one effect of such increased reinforcement value (e.g., Du et al., 2015; Greer & Han, 2015; Greer et al., 2011; Maffei et al., 2014). For example, previous research has demonstrated accelerated rates of learning of textual responses to words for preschool-age students as a collateral effect of increased reinforcement for looking at books (Buttigieg & Greer, 2020; Tsai & Greer, 2006). Such studies have focused on reinforcement stimulus control for looking at print stimuli, which is relevant for beginning readers learning to read by seeing and saying the word, or basic textual responding (Buttigieg & Greer, 2020; Singer-Dudek et al., 2011; Tsai & Greer, 2006).

However, a second category of reinforcement stimulus control has to do with students who are in the “read to learn” stage, where the goal is to establish reinforcement for the content of reading. Children in early elementary school begin to shift from learning to read to reading for the purpose of learning, which requires different reinforcement stimulus control. The types of texts beginning readers encounter are also less content based than more advanced texts, as early elementary reading levels may not allow for storylines that are as complex. The reinforcement for see-and-say therefore serves as a bridge until the complexity of content meets stronger reading skills, creating a continuum of reinforcement stimulus control for looking at textual stimuli, to seeing and saying text, to reading for continued content. To facilitate students’ development along this continuum as reading skills develop, teachers as strategic scientists can arrange instructional consequences to increase the reinforcement value of reading content for the purposes of promoting reading achievement (Greer, 2002; Vargas, 2013).

One method for conditioning sustained attention to reading content as a reinforcer is a collaborative shared reading (CSR) procedure, which is composed of opportunities for reciprocal reading and collaboration on comprehension and vocabulary tasks (Cumiskey Moore, 2017; Gentilini & Greer, 2020). Studies evaluating a teacher–student CSR intervention (e.g., teacher yoked) demonstrated significant increases on various measures of reading achievement, with increases in reinforcement value of reading among second-grade students (i.e., increases up to 2.2 grade levels; Gentilini & Greer, 2020). Researchers have found comparable, if not more significant, gains (i.e., increases up to 4.5 grade levels) with older students who acquired content as a conditioned reinforcer through a CSR procedure involving peer interaction paired with reading-related stimuli (e.g., peer yoked), which is significantly less intensive for teachers (Bly & Greer, 2019; Cumiskey Moore, 2017).

Given the intensive amounts of one-on-one teacher time required by the teacher-yoked pairing procedure (Gentilini & Greer, 2020), there is a need for an empirical investigation of the effectiveness of a peer-yoked CSR procedure among early elementary students—especially at an age when peer interaction begins to outweigh teacher attention as a reinforcer. This study therefore had two purposes: (a) to test the effect of the establishment of conditioned reinforcement for reading via a CSR conditioning procedure on second-grade students with and without learning disabilities and developmental disorders and (b) to compare a peer-yoked and teacher-yoked CSR procedure on students’ gains in reading achievement. Given the empirical support for the contribution of reading amount and motivation to the expansion of reading repertoires, the study of the reinforcement value of reading can inform both theory and practice in terms of the establishment of conditioned reinforcement for reading as a means of enhancing literacy achievement for early elementary students.

Method

Participants

Participants were second-grade students in a general education classroom of 19 students at the onset of the study. The classroom was situated within a Title I (kindergarten through second grade) public elementary school located in a suburb outside a major metropolitan area in the northeast United States that utilizes the Comprehensive Application of Behavior Analysis to Schooling Accelerated Independent Learner model of education (www.cabasschools.org). We identified and enrolled eight eligible students. In order to be eligible for the study, students could not have conditioned reinforcement for reading in their repertoires (Table 1).

Table 1.

Demographic Characteristics of Participants at Study Onset

Interventioncondition Intervention group Participant Age(years) Sex Race/ethnicity Free/reducedlunch status English as asecond language Diagnosis/educationalclassification Grade-level equivalentfor i-Ready Readinga Scholastic GuidedReading Program Levelb
Teacher CSR Dyad 1 1 8.4 Male Black Not eligible No Attention-deficit/hyperactivitydisorder Mid 2 L
2 8.3 Male White Not eligible No Autism spectrum disorder Grade 1 K
Peer CSR Dyad 2 3 7.7 Male White Not eligible No Mid 2 L
4 7.7 Female Hispanic Eligible Yes Early 2 L
Teacher CSR Dyad 3 5 7.4 Female White Not eligible No Other health impaired Grade 1 I
6 8.2 Female Hispanic Eligible Yes Early 2 L
Peer CSR Dyad 4 7 8.2 Female Other Not eligible No Grade 1 L
8 7.9 Female White Not eligible Yes Communication impaired Early 2 L
Open in a new tab

Note. All demographic information was gathered from the school’s database, with free/reduced lunch based on the local familial income criterion. CSR = collaborative shared reading. Blank cells indicate the absence of a formal diagnosis of educational classification.

aBased on the participants’ overall scores on the i-Ready K–12 Adaptive Reading Diagnostic immediately prior to the onset of the study, which includes the domains of phonological awareness, phonics, high-frequency words, literature comprehension, informational text comprehension, and vocabulary.

bBased on the participants’ performance on the Developmental Reading Assessment, Second Edition, immediately prior to the onset of the study. Participant 5 scored within a first-grade reading level; all other participants performed on a second-grade reading level.

Design

We utilized a combined small-n experimental-control simultaneous-treatment design with a single-case multiple-probe design nested within each small group, in order to compare within- and between-group differences (Dugard et al., 2012; Greer et al., 2007). The two groups differed in that one group completed the CSR procedure utilizing a teacher-yoked contingency, and one group completed the CSR procedure utilizing a peer-yoked continency. There was no change in the participants’ typical instructional environment outside of the context of the intervention, and the intervention replaced—rather than supplemented—participants’ daily reading instruction.

Measures and Procedure

Reinforcement Value of Reading

We individually video-recorded participants during 10-min periods of class-wide silent reading utilizing 5-s whole-interval recording, for a total of 120 intervals per session (Charlesworth & Spiker, 1975; Gentilini & Greer, 2020). We utilized recording devices, as opposed to real-time data collection by the experimenter, to decrease the possibility of instructional control as a confounding variable, and habituated students to the presence of the recording devices in the weeks prior to the study. Previous research has demonstrated that observations of participants’ responses utilizing whole-interval recording are highly correlated with automatically recorded continuous responses under laboratory conditions (Greer, 1981; Greer et al., 1973). Prior to every silent-reading session, we provided each participant with two to seven books within the range of their reading level (Pearson Education, 2006). We covered the pictures in all books to control for the presence of picture stimuli, as the presence of pictures moderates comprehension (Mercorella, 2017). We placed a video-recording device on the participant’s desk, such that it was 1–2 ft (0.3–0.6 m) from the participant and the participant’s eyes were visible for the entirety of the silent-reading session. We utilized a data sheet divided into 5-s intervals across 10 min for the purposes of scoring the video recordings. We recorded the participant as “reading” any time their eyes moved across the page from left to right, then returned to the leftmost side of the page. If a participant selected a new book, we coded their behavior as reading if they emitted observing responses for reading immediately prior, continued to visually attend to the book stimuli while selecting a new book, and turned to the first page of a new book within 20 s of closing the previous book.

Covert Reading Comprehension

We utilized a multiple-measures approach in our assessment of covert reading comprehension to increase the rigor of our research. We included participants’ grade-level equivalencies on Test 4: Passage Comprehension of the Woodcock–Johnson Tests of Achievement, Fourth Edition (WJ-IV; Schrank et al., 2014), and the Gray Silent Reading Tests (GSRT; Wiederholt & Blalock, 2000). Both of these measures have demonstrated excellent test–retest reliability among diverse populations of second-grade students (McGrew et al., 2014; Wiederholt & Blalock, 2000). We utilized Form A of both assessments in the preintervention probe sessions and Form B in the postintervention sessions. We administered WJ-IV Test 4: Passage Comprehension with participants in a one-to-one setting and conducted the GSRT in small groups of three to six participants.

Comprehension Drawing

We measured conditioned seeing, or the reflex of “seeing” what is being described in a text in the absence of a visual stimulus (Skinner, 1953, 1957). We calculated participants’ percentage accuracy in response to a comprehension drawing task, in which they were required to read, comprehend, and then visualize a given series of sentences. We displayed a series of five sentences via a presentation on a laptop in small groups of two to five participants. Each sentence included one nonsense word: The reading difficulty of this sentence was based on lists of first- and second-grade high-frequency words and the average reading level of the participants. We created two sets of stimuli and randomly assigned the participants to engage with one of the two sets. We counterbalanced the presentation of stimuli sets across probe sessions.

Vocabulary Knowledge

We also utilized a multiple-measures approach in our measure of vocabulary. Participants completed Test 17: Reading Vocabulary of the WJ-IV (Schrank et al., 2014). Previous research has indicated that this measure has excellent test–retest reliability among diverse groups of second-grade students (McGrew et al., 2014). To measure participants’ accuracy in deriving the meaning of unknown words in context—referred to in other theoretical models as inference comprehension—we conducted a derived relational responding task. We instructed the participants to turn their pictures from the comprehension drawing task face down and write the meaning of the nonsense word for each of the given sentences. The participants were assigned to the same groups of stimuli as in the comprehension drawing task.

Preference Assessment for Teachers Versus Peers

As children progress from early to upper elementary school around third grade, their reinforcement shifts from teacher attention to peer interaction. Because there are idiosyncratic differences in the reinforcement value of adults and peers for children in second grade, we conducted a verbal preference assessment at the onset of the study to possibly explain potential variation in the effectiveness of the conditions (Northup et al., 1996). We presented each participant with the written instruction “Please write down the name of a teacher or classmate in our class with whom you’d like to . . . ,” followed by one of five designated activities. We counterbalanced the order in which “teacher” and “classmate” were presented across sentences and participants. This would allow us to determine whether the peer-yoked condition would only be effective for students who were more reinforced by peers than by teachers, and vice versa.

CSR Conditioning Procedure

The teacher provided the participants with written steps of the procedure and a trade-in menu with a corresponding game board, which had a character that students moved contingent on correct responding in various intervention tasks. The teacher also provided one to three books within the participants’ reading level. If the participants in a peer-yoked dyad were at different reading levels, the teacher presented books within the higher range of the participant with the lower reading level (i.e., one to two levels above this participant’s reading level). The picture stimuli in each book were covered, and there was one contextualized nonsense word approximately every 10 sentences. The teacher also presented each participant with the vocabulary worksheets for the given books, the answer keys for those worksheets, and the comprehension worksheet. For the purposes of recording treatment fidelity, a second observer utilized a fidelity checklist with the steps of the intervention (Figure 1).

Figure 1.

Figure 1

Open in a new tab

Sequence of the Design. Note. The figure shows the sequence of the design to test the effect of the establishment of conditioned reinforcement for reading via a collaborative shared reading (CSR) intervention with a teacher or peer on participants’ reading achievement. We utilized a small-n experimental-control simultaneous-treatment design, with a single-case multiple-probe delayed design nested within each small group (Dugard et al., 2012; Greer et al., 2007).

We paired participants in dyads based on their average (i.e., mean) reinforcement value of reading and their grade-equivalent placement on the i-Ready K–12 Adaptive Reading Diagnostic (Curriculum Associates, 2017). We then randomly assigned the dyads to the teacher or peer CSR condition. We video-recorded two silent-reading sessions per participant and conducted additional sessions if there was not a stable level of responding (i.e., difference in reading of more than 25 intervals between two consecutive sessions for a given participant). We calculated the participants’ reinforcement value of reading within each silent-reading session and probe period (e.g., baseline, preintervention, postintervention), reporting it as a percentage of total intervals in which each participant read. We also measured the categorical variable of whether conditioned reinforcement for reading was in the repertoires of each of the participants; we considered the participants to have conditioned reinforcement for reading in their repertoire if they read for at least 75% of the 5-s whole intervals across sessions based on previous research (Buttigieg & Greer, 2020; Cumiskey Moore, 2017; Gentilini & Greer, 2020; Nuzzolo-Gomez et al., 2002; Tsai & Greer, 2006).

CSR With a Peer

Prior to the onset of the intervention, the teacher vocally explained the steps of the procedure while demonstrating a model of each step (Fig 2). The teacher provided prompting on the steps of the intervention for the first session of the intervention for each dyad. The teacher did not sit at the table with the dyad in successive sessions and only intervened to provide vocal prompts for the purposes of procedural fidelity or to answer questions.

Figure 2.

Figure 2

Open in a new tab

Sequence of the Collaborative Shared Reading Conditioning Procedure With a Peer or Teacher for Establishing Conditioned Reinforcement for Reading. Note. The duration of the overt and covert reading periods was dependent on the average number of intervals in which the participant(s) engaged in observable reading behavior across preintervention probe sessions immediately prior to entering the intervention. The criterion for each session of a given duration of reading was at least 80% accuracy in the derived relational responding tasks and 100% accuracy in the comprehension task. Once the participants met the criterion for a given duration of overt and covert reading, the experimenter increased the duration of the reading periods by 2 min; once the participant met the criterion for three 2-min increases in session duration, the experimenter conducted the probe procedures for the reinforcement value of reading with all participants who were entered into the intervention at that time.

In Step 1, participants engaged in reciprocal overt reading. The duration of reading was dependent on participants’ average number of intervals in which they were engaged in reading during preintervention probe sessions immediately prior to entering the intervention, with a minimum starting duration of 2 min.

During Step 2, participants wrote a synonym for each nonsense word they had encountered during the reading period. Then, partners discussed their answers, agreed on one, and checked their response against an answer key. If their response matched the answer key, the participants advanced their character one space on the game board. If their answer did not match, participants went back to the nonsense word in the text and read the sentence aloud with the correct word from the answer key, replacing the nonsense word. Participants rotated reading the corrected sentence aloud for each incorrect response.

In Step 3, participants read silently one at a time for the same duration of time as the reciprocal-reading period. Finally, during Step 4, participants began a comprehension task that involved (a) selecting a scene from the pages read by both partners during the silent-reading period, (b) writing the context of that scene on the comprehension worksheet, and (c) drawing a picture of that scene. Participants were not permitted to refer back to the text during the task. Then, participants used their partner’s drawing to guess the scene that was depicted. If both participants guessed correctly, they advanced three spots on the game board. If either or both of the participants guessed incorrectly, the participant(s) who guessed incorrectly drew their own version of the scene after being told the correct response. Following completion of the comprehension worksheet, participants utilized the provided trade-in menu and jointly selected a reinforcer based on their progress on the game board.

The criterion for each session of a given duration of reading was at least 80% accuracy in the vocabulary task and both participants correctly guessing the scene drawn by their partner in the comprehension task (i.e., 100% accuracy). Upon meeting the criterion, we increased the duration of reciprocal and silent reading by 2 min until the participants met the criterion for three consecutive increases in reading durations (e.g., 2, 4, 6 min). Once any of the participants met the criterion for a phase of the intervention, we conducted the probe procedures for the reinforcement value of reading with all participants who were entered into the intervention at that time, with two or three silent-reading sessions per set (e.g., Postprobe Set 1). We began the postintervention probe procedures for the reading-achievement measures if all of the participants who were entered into the intervention at a given time met the criterion for conditioned reinforcement for reading. If none, or only some, of the participants met the criterion for conditioned reinforcement for reading, all participants continued in the intervention until any of the participants met the criterion for a given phase of the intervention—at which point we repeated the probe procedures for the reinforcement value of reading.

CSR With a Teacher

The four steps of the CSR conditioning procedure with a teacher were identical to those in the CSR procedure with a peer. The teacher and participant engaged in reciprocal reading (Step 1), followed by the vocabulary task in which the participant completed the vocabulary worksheet (Step 2). The teacher and participant discussed each nonsense word after the participant had written down a response, and checked the answer key after deciding on one response.

Interobserver Agreement and Treatment Fidelity

Agreement was 99.3% (range 96.7%–100%) for a random sample of 62.5% of preintervention probe sessions and 97.7% (range 95.8%–100%) for a random sample of 31.7% of postintervention probe sessions for the reinforcement value of reading. We calculated 99.2% (range 97.3%–100%) interobserver agreement across 100% of preintervention probe sessions for comprehension drawing and 98.2% (range 96.0%–100%) interobserver agreement across 100% of postintervention probe sessions for comprehension drawing. For derived relational responding, we calculated 100% interobserver agreement across 100% of preintervention and postintervention probe sessions. In terms of treatment fidelity, the second independent observer reported high treatment fidelity for both the teacher-yoked and peer-yoked CSR procedures (Table 2).

Table 2.

Treatment Fidelity Reported by the Second Independent Observer

Interventioncondition Participant/interventiongroup Average treatmentfidelity Range Sessionsobserved Interventioncomponents prompted Range
Teacher CSR Participant 1, Dyad 1 91.2% 73.6%–100% 37.5%
Participant 2, Dyad 1 92.5% 77.4%–100% 37.5%
Peer CSR Participants 3 and 4, Dyad 2 92.8% 85.0%–100% 37.5% 23.7% 7.7%–38.5%
Teacher CSR Participant 5, Dyad 3 100% 55.6%
Participant 6, Dyad 3 100% 33.3%
Peer CSR Participants 7 and 8, Dyad 4 83.5% 72.7%–100% 44.4% 25.0% 15.4%–38.5%
Open in a new tab

Note. We calculated the treatment fidelity of the teacher’s or peer’s implementation of the collaborative shared reading (CSR) procedure by dividing the number of point-by-point components accurately implemented within each video-recorded or in-person session of the intervention by the total number of components included on a fidelity checklist, and multiplying that number by 100 to obtain the percentage. Blank cells indicate that the participant did not enter a second phase of intervention.

Results

Reinforcement Value of Reading

Table 3 shows changes in participants’ reinforcement value of reading throughout the course of the study, with comparable gains in reinforcement value across conditions for participants who met the criterion for conditioned reinforcement. None of the participants demonstrated the criterion for conditioned reinforcement prior to entering the intervention (Table 3, Fig 3). The reinforcement value of reading remained relatively stable from the baseline probe sessions to the probe sessions immediately prior to the onset of the intervention for all participants, with a mean change of 11.8% (range 1.6%–22.1%) in reinforcement value between probe periods in the absence of the CSR intervention (Table 3).

Table 3.

Participants’ Average Reinforcement Value of Reading per Probe Period

Intervention condition Intervention group Participant Baseline probe average Probe average immediately prior to intervention Postprobe Set 1 average Postprobe Set 2 average Postprobe Set 3 average
Teacher CSR Dyad 1 1 70.5 58.8 58.0 77.5
2 55.8 63.4 75.0 75.4
Peer CSR Dyad 2 3 46.7 68.8 68.8 86.3
4 69.6 68.0 78.8 81.7
Teacher CSR Dyad 3 5 35.9 17.1 26.3 25.8 20.0
6 38.4 22.1 77.5 67.6 79.6
Peer CSR Dyad 4 7 28.8 20.5 52.5 82.9 78.8
8 37.1 45.4 74.2 73.8 87.5
Open in a new tab

Note. Based on the average percentage of 5-s whole intervals of observable reading across two or three 10-min silent-reading sessions per probe period (i.e., baseline, preintervention, each postintervention set). We defined conditioned reinforcement for reading as an average of at least 75% of 5-s whole intervals of reading across 10-min probe sessions in a given probe period. CSR = collaborative shared reading.

Figure 3.

Figure 3

Open in a new tab

Implementation of the Independent Variable. Note. The figure shows the implementation of the independent variable, as measured by the percentage of 5-s whole intervals of reading within each 10-min probe session, or the reinforcement value of reading. We defined conditioned reinforcement for reading as an average of at least 75% of intervals of reading across probe sessions within a given probe period. We delayed the design by entering the next set of dyads into the intervention only once the effect of the intervention was present in the first set of dyads. We conducted the probe procedures at the onset of the study, directly before entering the participants into the intervention, and after any participant who was entered into the intervention met the criterion for a phase of the collaborative shared reading procedure.

The majority of participants had increases in reinforcement value for reading as they progressed through the intervention, and seven of the participants acquired conditioned reinforcement for reading by the conclusion of the intervention (excluding Participant 5, who decreased in reinforcement value throughout the intervention; Table 3, Fig 3). Table 4 shows the average increase in reinforcement value from the probe sessions immediately prior to the onset of the intervention to the probe sessions immediately following the conclusion of the intervention for all participants who met the criterion for conditioned reinforcement for reading by the conclusion of the study (i.e., excluding Participant 5). Participants in the teacher condition demonstrated an increase of an average of 29.4% (range 12.0%–57.5%) in reinforcement value, and participants in the peer condition demonstrated an increase of an average of 31.9% (range 13.7%–54.1%; Table 4).

Table 4.

Average Increase in Dependent Measures for Teacher Condition Versus Peer Condition

CSR condition
Measure Teacher condition
(n = 3)		 Peer condition
(n = 4)
Reinforcement value of reading

29.4%

(range 12.0%–57.5%)

31.9%

(range 13.7%–54.1%)
WJ-IV Test 4: Passage Comprehension

1.5 grade levels

(range 0.6–2.5)

0.6 grade levels

(range 0.2–1.2)
GSRT

0.7 grade levels

(range 0–1.5)

0.5 grade levels

(range 0–1.5)
Comprehension drawing

19.1%

(range 4.0%–36.0%)

8.3%

(range 5.3%–9.3%)
WJ-IV Test 17: Reading Vocabulary

1.3 grade levels

(range 0.3–3.1)

0.7 grade levels

(range 0.3–1.0)
Derived relational responding

17.8%

(range 0%–40.0%)

18.4%

(range 0%–53.4%)
Open in a new tab

Note. Average increase across measures of reinforcement value and reading achievement for participants in the teacher condition versus peer condition between probe sessions immediately prior to and following intervention. Data only include participants who met the criterion for conditioned reinforcement for reading by the conclusion of the study (i.e., excluding Participant 5). We rounded any decreases to 0. WJ-IV = Woodcock–Johnson Tests of Achievement, Fourth Edition; GSRT = Gray Silent Reading Tests.

Reading Achievement

Participants who acquired conditioned reinforcement in the teacher condition demonstrated larger gains across all measures of reading achievement besides derived relational responding, for which gains were comparable (Table 4, Figs 4 and 5). Tables 5 and 6 show grade-level increases in measures of reading comprehension and vocabulary between probe sessions immediately prior to and following the intervention. Of the participants who met the criterion for conditioned reinforcement for reading (i.e., excluding Participant 5), the majority increased beyond the amount expected based solely on the elapsed time between probe periods for the standardized measures of reading achievement (WJ-IV Test 4: Passage Comprehension, GSRT, and WJ-IV Test 17: Reading Vocabulary; Table 5, Figures 6 and 7). Similarly, the majority of participants who met the criterion for conditioned reinforcement for reading demonstrated increases in the measures of comprehension drawing and derived relational responding (Table 6, Figs 7 and 8).

Figure 4.

Figure 4

Open in a new tab

Average Grade-Level Increase. Note. The figure shows the average grade-level increase on the Woodcock–Johnson Tests of Achievement, Fourth Edition (WJ-IV), Test 4: Passage Comprehension; Gray Silent Reading Tests (GSRT); and WJ-IV Test 17: Reading Vocabulary between probe sessions immediately prior to and following the collaborative shared reading procedure for participants who met the criterion for conditioned reinforcement for reading in the teacher or peer condition. The error bars represent the range of increases for students in each condition, with any decreases in score rounded to 0 and comparable numbers of participants who decreased across conditions (i.e., Participant 2 in the teacher-yoked condition and Participant 4 in the peer condition for the GSRT).

Figure 5.

Figure 5

Open in a new tab

Average Increase in Percentage Accuracy on the Comprehension Drawing and Derived Relational Responding Tasks. Note. The figure shows the average increase in percentage accuracy on the comprehension drawing and derived relational responding tasks between probe sessions immediately prior to and following the intervention for participants who met the criterion for conditioned reinforcement for reading in the teacher or peer condition. The error bars represent the range of increases for students in each condition, with any decreases in score rounded to 0 and comparable numbers of participants who decreased across conditions (i.e., Participant 1 in the teacher condition and Participants 3 and 7 in the peer condition for derived relational responding).

Table 5.

Grade-Level Increase in Comprehension and Vocabulary Between Probe Sessions Immediately Prior to and Following Intervention

Participant
Dyad 1
(teacher condition)		 Dyad 2
(peer condition)		 Dyad 3
(teacher condition)		 Dyad 4
(peer condition)
Measure 1 2 3 4 5 6 7 8
WJ-IV Test 4: Passage Comprehension 2.5a 1.4a 0.2a 0.4a — 0.6a 1.2a 0.4
GSRT 0.5a — 1.5a — — 1.5a 0.5a —
WJ-IV Test 17: Reading Vocabulary 3.1a 0.5a 0.8a 1.0a — 0.3 0.5a 0.4a
Open in a new tab

Note. A dash denotes that the participant did not demonstrate an increase on that measure. WJ-IV = Woodcock–Johnson Tests of Achievement, Fourth Edition; GSRT = Gray Silent Reading Tests.

atThe participant demonstrated a grade-level increase beyond the amount expected based solely on the elapsed time between probe periods (i.e., approximately 1 month, or 0.1 grade levels), and there was a greater increase in scores between the probe session immediately prior to and following the intervention, compared to that from the first to second preintervention probe sessions (if applicable). We utilized the same test versions in both preintervention probe sessions, whereas we used the alternate versions in the postintervention probe sessions.

Table 6.

Percentage Increase in Comprehension Drawing and Derived Relational Responding Between Probe Sessions Immediately Prior to and Following Intervention

Participant
Dyad 1
(teacher condition)		 Dyad 2
(peer condition)		 Dyad 3
(teacher condition)		 Dyad 4
(peer condition)
Measure 1 2 3 4 5 6 7 8
Comprehension drawing 4.0 36.0 9.3 5.3 42.7 17.3 12.0 6.6a
Derived relational responding — 40.0 — 20.0 20.0 13.3a — 53.4
Open in a new tab

Note. The dash denotes that the participant did not demonstrate an increase on that measure.

aThe participant improved on that measure from preintervention to postintervention, but these gains were smaller than those from the first to second preintervention probe sessions.

We utilized the same stimuli sets in both preintervention probe sessions, whereas we used the alternate stimuli sets in the postintervention probe sessions.

Figure 6.

Figure 6

Open in a new tab

Grade Equivalences for the Teacher Condition. Note. The figure shows the grade equivalences on the Woodcock–Johnson Tests of Achievement, Fourth Edition (WJ-IV), Test 4: Passage Comprehension; Gray Silent Reading Tests (GSRT); and WJ-IV Test 17: Reading Vocabulary in the preintervention and postintervention probe sessions for participants in the teacher condition. Participants 1, 2, and 6 demonstrated conditioned reinforcement for reading following the intervention. The asterisks (*) represent participants who met two criteria: (a) a grade-level increase beyond the amount expected based solely on the elapsed time between probe periods (i.e., approximately 1 month, or 0.1 grade levels) and (b) a greater increase in scores between the probe session immediately prior to and following the intervention, compared to that from the first to second preintervention probe sessions (for participants in the delay). Note that we utilized the same test versions for both preintervention probe sessions (i.e., Form A), whereas we used the alternate versions in the postintervention probe sessions (i.e., Form B).

Figure 7.

Figure 7

Open in a new tab

Grade Equivalences in the Peer Condition. Note. The figure shows the grade equivalences on the Woodcock–Johnson Tests of Achievement, Fourth Edition (WJ-IV), Test 4: Passage Comprehension; Gray Silent Reading Tests (GSRT); and WJ-IV Test 17: Reading Vocabulary in the preintervention and postintervention probe sessions for participants in the peer condition. All four participants demonstrated conditioned reinforcement for reading following the intervention. The asterisks (*) represent participants who met two criteria: (a) a grade-level increase beyond the amount expected based solely on the elapsed time between probe periods (i.e., approximately 1 month, or 0.1 grade levels) and (b) a greater increase in scores between the probe session immediately prior to and following the intervention, compared to that from the first to second preintervention probe sessions (for participants in the delay).

Figure 8.

Figure 8

Open in a new tab

Percentage Accuracy in the Comprehension Drawing and Derived Relational Responding Preintervention and Postintervention Probe Sessions for the Teacher Condition. Note. The figure shows the percentage accuracy in the comprehension drawing and derived relational responding preintervention and postintervention probe sessions for participants in the teacher condition. Participants 1, 2, and 6 demonstrated conditioned reinforcement for reading following the intervention. Participant 6 demonstrated smaller gains in derived relational responding from the preintervention to postintervention probe sessions, as compared to those of the first to second preintervention probe sessions. Note that we utilized the same stimuli sets in both preintervention probe sessions, whereas we used the alternate stimuli sets in the postintervention probe sessions.

Of the participants in the delayed design who acquired conditioned reinforcement for reading (Participants 6, 7, and 8) and demonstrated increases in reading achievement, gains in comprehension and vocabulary from the first to second preintervention probe sessions were smaller than gains between probe sessions immediately prior to and following the intervention. The exceptions are for Participant 6 for WJ-IV Test 17 and derived relational responding, and Participant 8 for WJ-IV Test 4 and comprehension drawing (Tables 5 and 6, Figs 6, 7 and 8).

Participant 5 (teacher condition) did not meet the criterion for conditioned reinforcement for reading and did not demonstrate gains on any standardized measure of reading comprehension or vocabulary, including WJ-IV Test 4: Passage Comprehension, the GSRT, and WJ-IV Test 17: Reading Vocabulary (Table 5, Figure 6). Participant 5 increased in both comprehension drawing and derived relational responding from the assessments immediately prior to and following the intervention (Table 6, Figs 8 and 9).

Figure 9.

Figure 9

Open in a new tab

Percentage Accuracy in the Comprehension Drawing and Derived Relational Responding Preintervention and Postintervention Probe Sessions for the Teacher Condition. Note. The figure shows the percentage accuracy in the comprehension drawing and derived relational responding preintervention and postintervention probe sessions for participants in the teacher condition. All four participants demonstrated conditioned reinforcement for reading following the intervention. Participant 8 demonstrated smaller gains in comprehension drawing from the preintervention to postintervention probe sessions, as compared to those of the first to second preintervention probe sessions.

Progression Through the Intervention

Table 7 shows the number and duration of intervention sessions across participants in the teacher and peer CSR conditions. The mean total duration across all sessions for participants who exited the intervention was 291:51 min (range 209:36–356:10) in the teacher-yoked condition and 361:02 min (range 311:36–412:16) in the peer-yoked condition (Table 7). Progression through the intervention varied across participants (Figs 10 and 11).

Table 7.

Number and Duration of Intervention Sessions Across Participants

Intervention condition Intervention group Participant(s) Initial target duration (min:s) Duration of sessions, Phase 1 (min:s) Duration of sessions, Phase 2 (min:s) Total duration of sessions
Teacher CSR Dyad 1 1 6

1. 35:21

2. 42:18

3. 40:55

4. 39:44

1. 51:08

2. 37:50

3. 50:22

4. 58:32

356:10 min

≈5.9 hr
2 6

1. 29:14

2. 30:27

3. 36:19

4. 26:08

5. 41:47

6. 42:06

7. 38:46

8. 43:47

288:34 min

≈4.8 hr
Peer CSR Dyad 2 3 and 4 7

1. 41:18

2. 54:19

3. 43:29

4. 53:04

5. 46:51

6. 39:10

7. 57:21

8. 76:44

412:16 min

≈6.9 hr
Teacher CSR Dyad 3 5 2

1. 19:15

2. 14:06

3. 24:02

4. 21:39

5. 22:03

6. 21:50

7. 27:08

8. 28:04

 1. 31: 29

209:36 min

≈3.5 hr
6 2

1. 11:32

2. 23:51

3. 31:56

4. 34:23

5. 31:52

1. 36:19

2. 42:07

3. 48:57

4. 52:08

313:05 min

≈5.2 hr
Peer CSR Dyad 4 7 and 8 3

1. 35:16

2. 36:15

3. 32:05

4. 29:18

5. 27:12

6. 38:27

7. 29:02

8. 39:30

9. 44:31

311:36 min

≈5.2 hr
Open in a new tab

Note. CSR = collaborative shared reading.

Figure 10.

Figure 10

Open in a new tab

Progression Through the Intervention for Participants in Dyads 1 and 2. Note. The criterion for each session of a given duration of reading was at least 80% accuracy in the vocabulary task and 100% accuracy in the comprehension task. The criterion for a given phase of the intervention was three consecutive increases in reading durations (e.g., 2, 4, 6 min).

Figure 11.

Figure 11

Open in a new tab

Progression Through the Intervention for Participants in Dyads 3 and 4. Note: Criterion for each session of a given duration of reading was at least 80% accuracy in the vocabulary task and 100% accuracy in the comprehension task. Criterion for a given phase of the intervention was three consecutive increases in reading durations (e.g., 2 min, 4 min, 6 min)

Preference Assessment

Table 8 shows participants’ relative preference for teachers versus peers in the verbal preference assessment at the onset of the study. All participants assigned to the teacher condition (i.e., Participants 1, 2, 5, and 6) demonstrated a stronger relative preference for peers prior to the onset of the study. For those assigned to the peer condition, Participants 3 and 4 demonstrated a stronger relative preference for peers, and Participants 7 and 8 demonstrated a stronger relative preference for teachers (Table 8).

Table 8.

Percentage of Relative Preference in a Verbal Preference Assessment at the Onset of the Study

Participant Relative preference for teachers Relative preference for peers
1 0% 100%
2 0% 100%
3 0% 100%
4 20% 80%
5 20% 80%
6 40% 60%
7 100% 0%
8 100% 0%
Open in a new tab

Discussion

The results of this study suggest that the establishment of conditioned reinforcement for reading via a CSR procedure with a teacher or peer is effective in increasing the reading achievement of second-grade students. This study is one of several replications demonstrating a functional relation between the establishment of conditioned reinforcement for reading and increases in reading achievement (Bly & Greer, 2019; Cumiskey Moore, 2017). However, this study fills a gap in the current literature in that it is the first empirical investigation of a CSR procedure with peers applied to second-grade students. In addition, it differs from previous applications of the CSR procedure (Cumiskey Moore, 2017; Gentilini & Greer, 2020), as it includes students with and without disabilities.

The stable trend in the reinforcement value of reading from the baseline probe sessions to the probe sessions immediately prior to the onset of the intervention suggests that increases from the preintervention probe sessions to the postintervention probe sessions were functionally related to the intervention. Participants for whom we delayed the intervention generally demonstrated greater gains in reading achievement between the probe session immediately prior to and following the intervention, compared to those from the first to second preintervention probe sessions. Gains between the first and second preintervention probe sessions likely improved simply as a result of repeated assessment, whereas the postintervention assessments utilized test forms and stimuli sets that were novel to the participants. Increases in reading achievement can therefore be functionally related to the increases in the reinforcement value resulting from the CSR intervention, as the delayed design controlled for threats to internal validity (e.g., maturation and instructional history). It is likely that the intervention conditioned content as a reinforcer by increasing the participants’ purpose of reading—that is, to acquire information and derive meaning from the text that can be used effectively for some aesthetic or technical function. The steps within the CSR procedure increased the participants’ stimulus control for observing, in part by incorporating tasks that required a high degree of attention to the text in order to contact the pertinent content. It is unlikely that 10 min of sustained silent reading with second-graders would be under the control of the teacher’s presence alone. It is also improbable that factors such as a history of social reinforcement of instruction following or a history of punishment for off-task behavior would have changed from pre- to postintervention, and therefore would explain the observed changes in silent reading and/or reading achievement.

Seven participants acquired conditioned reinforcement for reading through the CSR intervention with a teacher (n = 3) or a peer (n = 4), such that the students’ attention was consistently selected out by the content of the text by the conclusion of the study. Participants in the teacher condition who preferred peers acquired conditioned reinforcement, and vice versa, suggesting that students can accrue benefits in reading achievement from the CSR procedure, regardless of whether their relative preference for teachers versus peers aligns with their assigned condition. Participant 5 was the only participant to consistently demonstrate a decrease in the reinforcement value of reading throughout the intervention and did not demonstrate conditioned reinforcement for reading or increases on the majority of measures of reading achievement by the onset of the study. This participant entered the intervention with relatively low overall reading comprehension and vocabulary scores compared to the other participants, as well as the lowest relative reading level based on their performance on the Developmental Reading Assessment, Second Edition (Pearson Education, 2006). It is possible that students who are below a certain grade-level equivalency in reading do not benefit from the CSR procedure, further suggesting that skill and will are not mutually exclusive in the development of reading achievement (Watkins & Coffey, 2004). There likely exists a continuum of reinforcement from looking at text, to seeing and saying text, to reading text for content. It is therefore also likely that some students need to achieve a greater skill fluency to benefit from conditioning for content, whereas students with more advanced reading levels (such as Participant 1) possess both the skill and the content reinforcement potential. A larger sample size may provide greater generality of the procedure to students with various reading abilities and at different stages of reinforcement for reading.

The results of this study also demonstrate more educationally significant gains in reading achievement for participants in the teacher condition who met the criterion for conditioned reinforcement for reading, as opposed to those in the peer condition. Despite demonstrating comparable average increases in the reinforcement value of reading from the probe sessions immediately prior to and following the intervention, participants in the teacher condition who acquired conditioned reinforcement for reading increased to a greater degree than participants in the peer condition on the majority of dependent measures. However, all participants for whom conditioned reinforcement for reading was established demonstrated gains on at least some measures of reading comprehension and vocabulary after only nine sessions at most, totaling a maximum of 412 min (6.9 hr) across sessions per participant or dyad. Although less effective, the peer procedure did work in conditioning content as a reinforcer and increasing reading achievement among second-grade students. Hence, if the teacher procedure is not viable due to its intensive nature, students will likely benefit from the peer-yoked procedure. The peer-yoked procedure is more efficient than the teacher-yoked protocol and more effective than continuing to teach discriminations under prosthetic reinforcement control.

Limitations and Suggestions for Future Research

Considering the time-intensive nature of the teacher-yoked CSR procedure, the overall effectiveness and relative efficacy of the peer-yoked CSR procedure increase the applicability of this intervention in the classroom setting. Future studies should continue to measure the treatment fidelity of a CSR intervention with a peer, as the question remains as to whether this procedure can be implemented with a high degree of fidelity without the need for teacher mediation. Future research should focus on identifying tactics to enhance the viability of the peer procedure in terms of decreasing the need for instructional guidance from adults.

Another limitation of the current study is that the criterion for conditioned reinforcement for reading has not been identified. Our criterion was based on existing verbal-behavior developmental research (e.g., Buttigieg & Greer, 2020; Cumiskey Moore, 2017; Nuzzolo-Gomez et al., 2002; Tsai & Greer, 2006), but there remains the possibility that an alternative criterion may have yielded different gains in reading achievement. Experimenters might consider conducting a parametric analysis to evaluate the criterion for conditioned reinforcement for reading that is optimal in its effect on reading achievement (Fuller & Fienup, 2018). Future studies can also test alternative mastery criteria for the intervention. Such data can potentially lend themselves to increasing the efficiency and feasibility of the intervention.

It may also be effective to combine aspects of a teacher-yoked and peer-yoked procedure within one intervention, as both the type and variation of reinforcers have an effect on the degree to which the target behavior is increased (Becraft & Rolider, 2015; Milo et al., 2010). For instance, the derived relational responding task might include a peer-yoked contingency, and the comprehension component of the intervention might incorporate reinforcement from an adult. Another suggestion is to allow students to choose for themselves whether they would prefer working with a teacher or a peer, as this may increase the effectiveness and efficiency of the pairing procedure (Patall, 2013).

Implications

Considering that a wide range of studies have reported reading amount as a positive predictor of reading competence and growth (Anderson et al., 1988; Cipielewski & Stanovich, 1992; Cunningham & Stanovich, 1997; Taylor et al., 1990), the results of this study suggest that a peer-yoked CSR procedure is both an efficient and effective method by which to promote the development of higher order repertories of reading. Educators of early elementary-school students should consider this relatively simple and scalable procedure as a systematic method for establishing conditioned reinforcement for reading—with the acquisition of this reinforcer promoting successive gains in a variety of advanced reading repertories as students’ covert reading behavior comes under stimulus control. As B. F. Skinner proposed,

By scheduling reading material so that the student is reinforced at the right time, we can “hook” him so that he will go on reading more and more difficult things and continue to read throughout his life. We shouldn’t teach great books; we should teach a love of reading. Knowing the contents of a few works of literature is a trivial achievement. Being inclined to go on reading is a great achievement. (Evans, 1968, p. 73)

Author Note

Lara M. Gentilini completed part of the requirements for the degree of doctor of philosophy under the mentorship of R. Douglas Greer in the preparation of this article.

Availability of data and materials

Data can be made available upon request by contacting the first author.

Compliance with Ethical Standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (Teachers College IRB, Protocol 19-078) and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. This article does not contain any studies with animals performed by any of the authors.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Anderson RC, Wilson PT, Fielding LG. Growth in reading and how children spend their time outside of school. Reading Research Quarterly. 1988;23:285–303. doi: 10.1598/RRQ.23.3.2. [DOI] [Google Scholar]
  2. Becker M, McElvany N, Kortenbruck M. Intrinsic and extrinsic reading motivation as predictors of reading literacy: A longitudinal study. Journal of Educational Psychology. 2010;102:773–785. doi: 10.1037/a0020084. [DOI] [Google Scholar]
  3. Becraft JL, Rolider NU. Reinforcer variation in a token economy. Behavioral Interventions. 2015;30:157–165. doi: 10.1002/bin.1401. [DOI] [Google Scholar]
  4. Bly, B. D., & Greer, R. D. (2019). The result of enhancing the value of careful reading on reading achievement in fourth graders [Manuscript submitted for publication]. Teachers College, Columbia University.
  5. Buttigieg, S. F., & Greer, R. D. (2020). Establishment of preference for books in free-play settings accelerates preschoolers’ rates of learning to read their first words [Manuscript submitted for publication]. Teachers College, Columbia University.
  6. Charlesworth WR, Spiker D. An ethological approach to observations in learning settings. In: Weinberg RA, Wood FH, editors. Observation of pupils and teachers in mainstream and special education settings: Alternative strategies. The Council for Exceptional Children: Reston, VA; 1975. pp. 161–170. [Google Scholar]
  7. Cipielewski J, Stanovich KE. Predicting growth in reading ability from children’s exposure to print. Journal of Experimental Child Psychology. 1992;54:74–89. doi: 10.1016/0022-0965(92)90018-2. [DOI] [Google Scholar]
  8. Cumiskey Moore, C. (2017). The effects of conditioned reinforcement for reading on reading comprehension for 5th graders [Unpublished doctoral dissertation]. Columbia University, New York.
  9. Cunningham AE, Stanovich KE. Early reading acquisition and its relation to reading experience and ability 10 years later. Developmental Psychology. 1997;33:934–945. doi: 10.1037/0012-1649.33.6.934. [DOI] [PubMed] [Google Scholar]
  10. Curriculum Associates, LLC. (2017). i-Ready K–12 Adaptive Diagnostic for New Jersey. North Billerica, MA.
  11. Du L, Broto J, Greer RD. The effects of the establishment of conditioned reinforcement for observing responses for 3D stimuli on generalized match-to-sample in children with autism spectrum disorders. European Journal of Behavior Analysis. 2015;16:82–98. doi: 10.1080/15021149.2015.1065655. [DOI] [Google Scholar]
  12. Dugard P, File P, Todman J. Single-case and small-n experimental designs: A practical guide to randomization tests. New York, NY: Routledge; 2012. [Google Scholar]
  13. Evans RI. B. F. Skinner: The man and his ideas. New York, NY: E. P. Dutton & Co., Inc.; 1968. [Google Scholar]
  14. Fantino E, Romanowich P. The effect of conditioned reinforcement rate on choice: A review. Journal of the Experimental Analysis of Behavior. 2007;87:409–421. doi: 10.1901/jeab.2007.44-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fuller JL, Fienup DM. A preliminary analysis of mastery criterion level: Effects on response maintenance. Behavior Analysis in Practice. 2018;11:1–8. doi: 10.1007/s40617-017-0201-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gentilini LM, Greer RD. Establishment of conditioned reinforcement for reading content and effects on reading achievement for early-elementary students. The Psychological Record. 2020;70(2):327–346. doi: 10.1007/s40732-020-00382-6. [DOI] [Google Scholar]
  17. Gottfried AE, Fleming JS, Gottfried AW. Continuity of academic intrinsic motivation from childhood through late adolescence: A longitudinal study. Journal of Educational Psychology. 2001;93:3–13. doi: 10.1037/0022-0663.93.1.3. [DOI] [Google Scholar]
  18. Greer R, Dorow L, Wachhaus G, White E. Adult approval and students’ music selection behavior. Journal of Research in Music Education. 1973;21:345–354. doi: 10.2307/3344908. [DOI] [Google Scholar]
  19. Greer, R. D. (1981). An operant approach to motivation and affect: Ten years of research in music learning. In Documentary report of the Ann Arbor symposium: National symposium on the applications of psychology to the teaching and learning of music. Reston, VA: Music Educators National Conference.
  20. Greer RD. Contingencies of the science and technology of teaching and pre-behavioristic research practices in education. Educational Researcher. 1983;12(1):3–9. [Google Scholar]
  21. Greer RD. Designing teaching strategies: An applied behavior analysis systems approach. San Diego, CA: Academic Press; 2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Greer RD. The selector in behavior selection. The Psychological Record. 2020;70:543–558. doi: 10.1007/s40732-020-00385-3. [DOI] [Google Scholar]
  23. Greer RD, Han HSA. Establishment of conditioned reinforcement for visual observing and the emergence of generalized visual-identity matching. Behavioral Development Bulletin. 2015;20:227–252. doi: 10.1037/h0101316. [DOI] [Google Scholar]
  24. Greer RD, Pistoljevic N, Cahill C, Du L. 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. 2011;27:103–124. doi: 10.1007/BF03393095. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Greer RD, Ross DE. Verbal behavior analysis: Inducing and expanding new verbal capabilities in children with language delays. Boston: Allyn and Bacon; 2008. [Google Scholar]
  26. Greer RD, Speckman J. The integration of speaker and listener responses: A theory of verbal development. The Psychological Record. 2009;59:449–488. doi: 10.1007/BF03395674. [DOI] [Google Scholar]
  27. Greer RD, Stolfi L, Pistoljevic N. Emergence of naming in preschoolers: A comparison of multiple and single exemplar instruction. European Journal of Behavior Analysis. 2007;8:109–131. doi: 10.1080/15021149.2007.11434278. [DOI] [Google Scholar]
  28. Guthrie JT, Wigfield A. Engagement and motivation in reading. In: Kamil ML, Mosenthal PB, Pearson PD, Barr R, editors. Handbook of reading research. 3. Mahwah, NJ: Lawrence Erlbaum; 2000. pp. 403–422. [Google Scholar]
  29. Heyns B. Summer learning and the effect of schooling. New York, NY: Academic Press; 1978. [Google Scholar]
  30. Kurth, R., & Kurth, L. (1987). The use of time informal reading instruction in elementary schools [Paper presentation]. Annual Meeting of the American Educational Research Association, April 20–24, Washington, DC.
  31. Maffei J, Singer-Dudek J, Keohane DD. The effects of the establishment of adult faces and/or voices as conditioned reinforcers for children with ASD and related disorders. Acta de Investigación Psicológica. 2014;4:1621–1641. doi: 10.1016/S2007-4719(14)70970-6. [DOI] [Google Scholar]
  32. McGrew KS, LaForte EM, Schrank FA. Technical manual: Woodcock–Johnson IV. Rolling Meadows, IL: Riverside; 2014. [Google Scholar]
  33. Mercorella, K. A. (2017). The effects of sequencing and producing narrative components of a story on reading comprehension [Unpublished doctoral dissertation]. Columbia University, New York.
  34. Milo JS, Mace FC, Nevin JA. The effects of constant versus varied reinforcers on preference and resistance to change. Journal of the Experimental Analysis of Behavior. 2010;93:385–394. doi: 10.1901/jeab.2010.93-385. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Moore J. Conceptual foundations of radical behaviorism. Cornwall-on-Hudson, NY: Sloan; 2008. [Google Scholar]
  36. National Reading Panel. (2000). Teaching children to read: An evidence-based assessment of the scientific research literature on reading and its implications for reading instruction. National Institute of Child Health and Human Development.
  37. Northup J, George T, Jones K, Broussard C, Vollmer TR. A comparison of reinforcer assessment methods: The utility of verbal and pictorial choice procedures. Journal of Applied Behavior Analysis. 1996;29:201–212. doi: 10.1901/jaba.1996.29-201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Nuzzolo-Gomez R, Leonard MA, Ortiz E, Rivera CM, Greer RD. Teaching children with autism to prefer books or toys over stereotypy or passivity. Journal of Positive Behavior Interventions. 2002;4:80–87. doi: 10.1177/109830070200400203. [DOI] [Google Scholar]
  39. Patall EA. Constructing motivation through choice, interest, and interestingness. Journal of Educational Psychology. 2013;105(2):522–534. doi: 10.1037/a0030307. [DOI] [Google Scholar]
  40. Pearson Education. (2006). Developmental Reading Assessment (2nd ed.). New York, NY.
  41. Schrank FA, Mather N, McGrew KS. Woodcock–Johnson IV Tests of Achievement. Rolling Meadows, IL: Riverside; 2014. [Google Scholar]
  42. Singer-Dudek J, Oblak M, Greer RD. Establishing books as conditioned reinforcers for preschool children as a function of an observational intervention. Journal of Applied Behavior Analysis. 2011;44(3):421–434. doi: 10.1901/jaba.2011.44-421. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Skinner, B. F. (1953). Science and human behavior. New York, NY: Macmillan.
  44. Skinner, B. F. (1957). Verbal behavior. New York, NY: Prentice-Hall.
  45. Skinner BF. Whatever happened to psychology as the science of behavior? American Psychologist. 1987;42:780–786. doi: 10.1037/0003-066X.42.8.780. [DOI] [Google Scholar]
  46. Stutz F, Schaffner E, Schiefele U. Relations among reading motivation, reading amount, and reading comprehension in the early elementary grades. Learning and Individual Differences. 2016;45:101–113. doi: 10.1016/j.lindif.2015.11.022. [DOI] [Google Scholar]
  47. Taylor BM, Frye BJ, Maruyama GM. Time spent reading and reading growth. American Educational Research Journal. 1990;27:351–362. doi: 10.3102/00028312027002351. [DOI] [Google Scholar]
  48. Tsai H, Greer RD. Conditioned preference for books and faster acquisition of textual responses by preschool children. Journal of Early and Intensive Behavioral Interventions. 2006;3:35–60. doi: 10.1037/h0100322. [DOI] [Google Scholar]
  49. Vargas JS. Behavior analysis for effective teaching. New York, NY: Routledge; 2013. [Google Scholar]
  50. Wang JH, Guthrie JT. Modeling the effects of intrinsic motivation, extrinsic motivation, amount of reading, and past reading achievement on text comprehension between U.S. and Chinese students. Reading Research Quarterly. 2004;39:162–186. doi: 10.1598/RRQ.39.2.2. [DOI] [Google Scholar]
  51. Wiederholt, J. L., & Blalock, G. (2000). GSRT: Gray Silent Reading Tests. Pro-Ed.
  52. Wigfield A, Gladstone JR, Turci L. Beyond cognition: Reading motivation and reading comprehension. Child Development Perspectives. 2016;10:190–195. doi: 10.1111/cdep.12184. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Ysseldyke, J. E., & Algozzine, R. (1983). Where to begin in diagnosing reading problems. Topics in Learning and Learning Disorders, 2, 60–68.
  54. Watkins MW, Coffey DY. Reading motivation: Multidimensional and indeterminate. Journal of Educational Psychology. 2004;96:110–118. doi: 10.1037/0022-0663.96.1.110. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data can be made available upon request by contacting the first author.

Articles from Behavior Analysis in Practice are provided here courtesy of Association for Behavior Analysis International

RESOURCES