Skip to main content
NCBI home page
Behavior Analysis in Practice logoLink to Behavior Analysis in Practice
. 2021 Jan 29;14(1):272–282. doi: 10.1007/s40617-020-00509-9

Stimulus Control Research and Practice: Considerations of Stimulus Disparity and Salience for Discrimination Training

Mary E Halbur 1,✉, R Kyle Caldwell 1, Tiffany Kodak 1,✉
PMCID: PMC7900359  PMID: 33732597

Abstract

Stimulus control plays a prominent role in behavior-analytic service delivery, as many discrimination skills are necessary for daily interactions. Clarification and standardization of terminology are necessary for the advancement of research and practice on stimulus control. The purpose of the present article is to provide an overview of stimulus control and discrimination training as they relate to the disparity and salience of stimuli. An overview and examples of stimulus disparity and stimulus salience are provided, followed by recommendations for efficacious service delivery.

Keywords: Stimulus control, Discrimination, Disparity, Salience

Discrimination Training

The identification and arrangement of sources of stimulus control in behavior-analytic treatment are critical for effective service delivery for individuals with autism spectrum disorder (ASD), intellectual disability (ID), and developmental disability (DD). For example, proper arrangement of sources of stimulus control is necessary for the development of language training programs, such as teaching children with ASD to answer questions and respond to stimuli in their environment (Axe, 2008; Pilgrim, 2015). If behavior comes under faulty sources of stimulus control, attaining important treatment goals may be hindered until this problem is resolved (e.g., Grow et al., 2011; Walpole et al., 2007). Stimulus control is established through discrimination training, and many everyday interactions with the environment require discriminations.

Discrimination is conceptualized as a fundamental behavioral process and is of critical importance to behavior analysts (Skinner, 1969). Antecedent stimuli acquire their controlling function by occurring with specific consequences that follow behavior, such as reinforcement. A discriminative stimulus (SD, also referred to as S+) is associated with the availability of reinforcement for a response in its presence but not in its absence, whereas an S-delta (S∆, also referred to as S−) is associated with the unavailability of reinforcement for a response in its presence (Skinner, 1938). Stimulus control occurs when a response is occasioned by these antecedent stimuli (Dinsmoor, 1995). For example, the response “Mary” has come under the control of relevant antecedents when this response occurs following the question “What is your name?” and does not occur when asked of people who do not share this same name. Discrimination is shown by an individual’s differentiated responding with respect to such stimuli, and stimulus control has been established when a response reliably occurs in the presence of an SD, but not in its absence.

Simple and complex discriminations, as well as stimulus control more generally, are of interest to basic, applied, and translational researchers and clinicians. Due to the prominence of discrimination training in both clinical practice and research, careful arrangement of procedures to establish stimulus control is warranted. Thus, it is important to understand the components that are present within discrimination training. A simple discrimination consists of an antecedent, response, and consequence. For example, a simple discrimination may consist of selecting a green card from a two-card array (e.g., red card and green card) on repeated trials in which the therapist says “go.” The red card is not targeted for instruction, nor does responding to the red card produce a reinforcer. Alternatively, a conditional discrimination involves the inclusion of a sample stimulus (i.e., the conditional stimulus, such as the spoken words “go” and “stop”), which alters the comparison stimulus to which a response will be reinforced. Responding to a comparison (e.g., the green card) in the presence of one sample stimulus (e.g., the spoken word “go”) and not the other sample, and responding to another comparison (e.g., the red card) only in the presence of the other sample stimulus (e.g., the spoken word “stop”), demonstrates conditional control (Saunders & Spradlin, 1990). Learning conditional discriminations is necessary for success in school and daily living for skills such as reading, daily grooming, and social interactions.

Discrimination training is complicated by the many properties of stimuli included in training. For example, a green square on a piece of paper has properties of hue, brightness, saturation, intensity, shape, and size, among others. Typical stimuli included in conditional discrimination training for children with ASD likely have many more properties. Variations in these properties make it difficult to ensure that the learner’s responding comes under the control of the specific properties of the targeted stimulus instead of other, irrelevant properties (i.e., faulty stimulus control). Thus, when conducting discrimination training, it is important to consider how certain variables affect the rate of acquisition of targeted discriminations (Nevin et al., 2005). As such, discrimination training requires careful considerations and may require manipulations in order to isolate specific components that may influence the rate of acquisition.

Two variables that may affect the efficacy and efficiency of discrimination training are stimulus disparity and salience (Dinsmoor, 1995). Stimulus disparity refers to the difference between stimuli and is often measured by magnitude or intensity. For example, see Figure 1 for a representation of a match-to-sample training trial with high-disparity and low-disparity comparison stimuli (also refer to Hannula et al., 2020). The left portion of the diagram shows a representation of high disparity in visual stimuli, in which there is a high magnitude of the difference in color of these comparison stimuli. The right portion of the diagram shows a representation of low disparity in visual stimuli, in which there is a low magnitude of difference in the color of these comparison stimuli. This is contrary to stimulus salience, which refers to the magnitude of the difference in the targeted stimulus and background stimulation that influences attention to the target (Dinsmoor, 1995). For example, a small variation in hair length following a haircut may not be a salient stimulus for a friend to attend to and make a comment about this change. When presented with two stimuli (i.e., an S+ and S−), salience refers to the magnitude of the difference between the S+ and the background stimulus (i.e., how dissimilar they are), rather than the difference between the S+ and S−, which is instead a measure of stimulus disparity (Dinsmoor, 1995).

Figure 1.

Figure 1

Open in a new tab

Stimulus Disparity of Comparison Stimuli for Two Sets of Stimuli in a Match-to-Sample Arrangement. Note. High-disparity comparisons are shown in the left panel, and low-disparity comparisons are shown in the right panel. Solid lines represent correct responses, and dashed lines represent incorrect responses

The roles of stimulus disparity and stimulus salience have been investigated together (e.g., Dinsmoor et al., 1983; Dinsmoor et al., 1982; Fields, 1978) and in isolation (e.g., Alsop & Davison, 1991; Davison & Nevin, 1999; Fields, 1979; Rodríguez-San Juan & Rodríguez, 2017). However, disparity and salience can be easily confused because both terms relate to differences within and between stimuli. Additionally, similarities and differences in salience and disparity of stimuli may impact the acquisition of discrimination. The distinction between these terms is important for behavior analysts because clinical applications of discrimination training should rely on behavior analysts being able to understand and manipulate each of these variables in isolation to improve treatment outcomes. In clinical practice, behavior analysts will need to be able to discriminate between disparity and salience in order to identify faulty sources of stimulus control (i.e., during problem solving). For example, behavior analysts will need to identify whether overlapping components of targets in an array are not disparate enough or whether the antecedent verbal stimuli selected for inclusion are not salient to the learner.

Thus, the purpose of this article is to help behavior analysts (a) differentiate between stimulus disparity and salience within the context of previous research and discrimination training, (b) identify examples of stimulus disparity and salience as they relate to research and practice, and (c) arrange sources of stimulus control more effectively in practice. First, we provide an overview of research on stimulus disparity and salience. Thereafter, we provide recommendations for manipulating stimulus disparity and salience in practice by selecting and arranging targets for discrimination training, as well as remediating error patterns during behavior-analytic service delivery. Although there are basic and translational studies on stimulus disparity and salience that will be described in the following sections, there is little applied research on these topics. Nonetheless, due to the strength of the analyses of these variables in the basic literature and the importance of these findings for efficacious service delivery, the remainder of this article explores potential recommendations for practice that consider stimulus disparity and salience within discrimination training.

Stimulus Disparity

Research on Stimulus Disparity

Basic research studies with pigeons (e.g., Hodos & Bonbright, 1972; McCarthy & Davison, 1980a, 1980b), rats (e.g., Lawrence, 1952), and humans (e.g., Swets, 1959) suggested discrimination performance depends highly on the level of disparity of physical differences between stimuli (Davison & Nevin, 1999). When the disparity (magnitude of physical differences between target stimuli) is low, accuracy within training is low (Alsop & Davison, 1991). For example, Lawrence (1952) suggested that discriminations are more efficiently acquired if the discrimination is first trained with highly disparate stimuli than with less disparate stimuli. Within his investigation, rats were taught to jump from a starting box to an array of gray compartments that varied in intensity of the saturation of gray. Lawrence investigated the easy-to-difficult continuum by comparing highly similar conditions from the start (i.e., similar gray compartments; low-disparity condition) to highly disparate conditions (i.e., less similar grays) that were faded across training. Results suggested the efficiency of the training procedure was impacted by the disparity of stimuli from the onset of training.

Research on stimulus disparity has also been conducted with humans. For example, Gallagher and Alsop (2001) manipulated disparity with the sample stimulus and the comparison stimuli for humans in signal-detection tasks. The sample stimuli presented to participants (i.e., tones) ranged from low-disparity durations (e.g., 525 ms vs. 475 ms) to high-disparity durations (e.g., 550 ms vs. 450 ms). The comparison stimuli were manipulated by varying the density in pixels between the two alternatives (e.g., 96 dots vs. 84 dots). Participants were played two tones, then shown two patterns on the screen. They were told, “When the second tone is longer, press the pattern with more dots. When the second tone is shorter, press the pattern with less dots” (p. 187). Results of this investigation suggested that the disparity of the comparison stimuli is equally as important as the disparity of sample stimuli, because comparison stimuli are presented simultaneously. Thus, if the client cannot engage in a simultaneous discrimination of the comparisons (e.g., the client cannot differentiate between pictures shown in an array), then the presentation of samples is irrelevant.

Applied researchers have also considered stimulus disparity when designing studies, although direct manipulations of disparity are limited (Eikeseth & Hayward, 2009). Studies that use an adapted alternating-treatments design attempt to control for disparity when assigning targets to conditions (e.g., Cariveau et al., 2020; Grow et al., 2011). For example, researchers may use the logical analysis procedure (Campanaro et al., 2020; Nottingham et al., 2017; Wolery et al., 2014) to equate training sets by taking into account the physical similarity of pictures, the number of phonemes in the targets, the starting and ending sounds, and scores during probe sessions. This procedure involves assigning targets to different conditions if they contain similar sounds or begin with the same letter (e.g., “cat” and “hat”) or have visual similarities (e.g., two targets contain red circles). This procedure is recommended to prevent the assignment of low-disparity targets to a training set, which is hypothesized to slow the rate of acquisition (Grow & LeBlanc, 2013).

Recommendations for Research and Practice on the Disparity of Target Selection

The selection of stimuli and targets for any behavior-analytic program is an important component of instructional design. Previous researchers have recommended simultaneous training of multiple targets, presenting concise and specific auditory instructions, and counterbalancing the visual and auditory stimuli during conditional discrimination training (Green, 2001; Grow et al., 2011; Grow & LeBlanc, 2013). A portion of these recommendations have been individually shown to impact acquisition and efficiency of programming (i.e., simultaneously training of multiple targets; Grow et al., 2011). In addition to these recommendations, it is valuable for practitioners to consider other variables that could impact acquisition with regard to stimulus disparity when selecting targets for use in practice.

Visual Stimuli Selection

Highly disparate visual stimuli are beneficial to include at the onset of discrimination training (e.g., McCarthy & Davison, 1980b; Swets, 1959). The differences between selected stimuli should account for variables such as orientation, wavelength, and color (White, 1986). This is not a new recommendation, as previous researchers (e.g., Grow & LeBlanc, 2013) have suggested starting with highly disparate stimuli as initial targets. Although there is not a current body of applied research to guide the selection, recommendations can be based on basic research. Future researchers should also continue to investigate these areas.

Practitioners should avoid including stimuli that have overlapping components (e.g., colors, shapes) within early discrimination training programs at the onset, especially if they will be used as comparison stimuli during a listener response task (Lawrence, 1952). For example, when teaching visual–visual conditional discriminations (e.g., matching) of vegetables, practitioners and researchers should consider variables that might influence stimulus disparity. Rather than selecting targets such as kale, lettuce, broccoli, and asparagus that are not disparate due to their overlapping colors, sizes, and shapes, they should teach visually disparate targets such as carrots, tomatoes, brussels sprouts, and eggplant; these visual stimuli have greater disparity due to variation in color, size, and shape. However, after initial conditional discriminations are taught, practitioners should continue discrimination training with stimuli with high disparity, then gradually incorporate stimuli with lower disparity (i.e., more similarity in color, shape, and size) in discrimination training.

It is important that practitioners carefully select exemplars of a concept to teach based on critical and variable attributes of stimuli included in sets. A concept has been defined as a set of common features or attributes that make a stimulus part of a group (Layng, 2019; Tiemann & Markle, 1990). Layng (2019) recommended that practitioners begin by classifying components of concepts that are critical; components are characterized as “must have” attributes (i.e., critical features that define the concept) and “can have” attributes (i.e., variable attributes that do not define the concept but may co-occur with critical attributes). The goal of instruction is to teach learners to identify exemplars of a concept by the critical attributes rather than by variable attributes. For example, a chair has the critical attributes of seating one person, having a back, and supporting a person’s legs at a 90° angle while seated, as well as variable attributes such as color, shape, number of legs, and material. Stimuli may share some but not all critical attributes (e.g., a chair and a stool both seat one person, but a stool does not have a back). In addition, stimuli may share variable attributes but not any critical attributes (e.g., a chair and a pair of shoes may share materials such as being made of suede), whereas other stimuli may share neither critical nor variable attributes (e.g., grapes have neither shape nor material in common with a chair).

Problems may occur in the form of faulty stimulus control if an insufficient number of attributes is varied within stimulus sets. Therefore, appropriate stimulus control is often only achieved if there are some similarities between trained stimuli, but the disparity is systematically manipulated with “close-in” examples (i.e., one or more similar components that lack a defining feature). Layng (2019) suggested that for learners to acquire a concept or groups of items, a sequence of requirements is necessary: (a) The learner must discriminate examples of the concept from close-in nonexamples that do not have at least one of the critical attributes, (b) the learner must discriminate a range of examples of the concept that have varying attributes, and (c) the learner should engage in a generalized repertoire with stimuli that were not part of training (i.e., including examples and nonexamples; p. 348).

Although there is a minimum number of examples and nonexamples that can be included in a stimulus set to establish a concept (see Layng, 2019, for several examples), teaching children with ASD could begin with a set of stimuli with high disparity, meaning that only one exemplar contains the critical attribute(s) and the remaining stimuli have one or no variable attributes. As teaching progresses, the disparity of stimuli could be gradually reduced by including stimuli with one but not all critical attributes, as well as varying exposure to variable attributes. Future researchers should continue to investigate the optimal number of varied attributes and methods to systematically manipulate the disparity of stimuli in sets, as the research supporting this does not emphasize a sequence of training along a continuum of highly disparate targets to targets with low disparity.

Selection of Discriminative Stimuli or Antecedent Verbal Stimuli

The disparity of the sample stimuli, SD, and antecedent verbal stimuli should also be considered. In clinical practice, the similarity and length of auditory stimuli presented throughout a session may be variables that affect acquisition rate (Pierrel & Sherman, 1962; Pierrel et al., 1970). The use of disparate and concise auditory instructions during conditional discrimination training reduces the risk of other irrelevant features exerting stimulus control over responses (Green, 2001; Grow & LeBlanc, 2013). For example, the antecedent verbal stimuli “point to the square” and “point to the oval” have nearly identical auditory stimuli (i.e., they both begin with “point to the”). However, if “point to the” is removed from the auditory sample stimulus, then the remaining components of the auditory stimulus are more disparate (i.e., “square” and “oval”). Alternatively, if a practitioner selected the targets “circle” and “oval,” there are also several overlapping components: They end with the same sound and have two syllables each. These may be critical variables that are often overlooked in stimulus selection; thus, researchers have recommended assigning stimuli with overlapping sounds (i.e., low disparity) to different stimulus sets (e.g., Carroll et al., 2016; Grow & LeBlanc, 2013).

The type of auditory stimuli to include in discrimination training should also be considered during training. Some researchers have investigated the effect of environmental sounds (e.g., instruments like a piano) and spoken names of stimuli (i.e., the clinician saying “piano”) as the SD on acquisition during discrimination training (Eikeseth & Hayward, 2009; Uwer et al., 2002). For example, Eikeseth and Hayward (2009) found young children with ASD learned to select pictures of musical instruments more efficiently when the sound of the instrument was the SD in comparison to presenting the spoken instrument name as the SD. Thus, the environmental sounds may be more disparate than spoken words, especially if the words contain overlapping vowels or consonants. Results of studies like these demonstrate the importance of the disparity of auditory sample stimuli in practice and show the necessity for continued applied research on auditory stimulus disparity and acquisition. Further, errorless transfer of control procedures may be beneficial. For example, stating the name of a musical instrument while a recording of the sound it makes is played, and then gradually fading the sounds from being played, may provide helpful initial guidelines for future researchers and practitioners when teaching discriminations of low-disparity stimuli.

Recommendations for Research and Practice on the Disparity of Behavior–Consequence Relations

It is important for practitioners to consider the disparity of the behavior–consequence relation when evaluating persistent error patterns in acquisition tasks. Previous researchers have proposed that error patterns could be a result of low disparity between contingencies (i.e., low disparity between the consequences of correct responding and incorrect responding). Similarly, these researchers also considered the magnitude of the consequence, thus manipulating salience (described in more detail in the next section of the article). For example, Fisher et al. (2014) described a strategy to manipulate behavior–consequence relations in the context of conditional discrimination training. Following a differential reinforcement baseline, researchers implemented a second-order reinforcement schedule so that a correct response produced praise and a small edible item that was placed in one of three clear bins on the table in front of participants. Once three consecutive correct responses occurred and three edible items accumulated in the bins, the participant received praise and those three edibles. However, if an error occurred, all edibles from the clear bins were removed. By using a visual aid, the salience of the contingency for engaging in correct responses was increased, thus affecting the disparity between consequences for correct and incorrect responses for learners who demonstrated error patterns with common training procedures (e.g., prompt delays, error correction procedures, blocked trial arrangements). This visual aid resulted in the acquisition of the targeted conditional discriminations. Arranging schedule manipulations such as those in Fisher et al. (2014) can be implemented to reduce these errors and to promote discriminability of consequences when arranging differential reinforcement.

Research on Stimulus Salience

Stimulus salience is the extent to which an SD differs from background stimulation (Dinsmoor, 1995). Thus, stimuli with more intense and easily identifiable features are more likely to exhibit salient properties in occasioning responding (Cooper et al., 2007). Considering the continuous nature of intensity and magnitude, stimulus salience is not a present-versus-absent assessment but, instead, a continuum along which multiple variables, such as reinforcement history or sensory sensitivity, can be manipulated to influence rates of discrimination acquisition (Dinsmoor et al., 1983; Fields, 1979; Madden & Perone, 1999; Pierrel et al., 1970). Put more simply, Pilgrim (2015) stated that “a stimulus property does not control behavior because it is salient; we describe it as salient because it controls responding” (p. 30).

Dinsmoor et al. (1982) differentially reinforced pigeons’ key pecks on a white key bisected with a horizontal black line, alternating between variable-interval and extinction schedules. If the pigeons engaged in an alternative response (pressing a perch bar), additional stimuli were presented that were associated with each component. Researchers changed the position of the horizontal line when the interval schedule was in effect so that the angle of the black line varied between stimulus sets by either 30° or 60°. This manipulation permitted a measure of the disparity of these stimuli. However, researchers also made modifications to evaluate stimulus salience. Independent of tilt level, the experimenters manipulated the contrast intensity (brightness) of the black line on the key by reducing the power of the illuminating lamps; they made it increasingly more difficult to see. Because the brightness was modified across all stimulus sets, regardless of tilt level, the experimenters were able to evaluate how the line’s intensity influenced discriminability; this provided a measure of salience. As the black line’s intensity was reduced (as salience decreased), discrimination acquisition decreased almost immediately regardless of the line’s tilt/position. Thus, until stimulus properties are salient and occasion responding to them, it is possible that variations in disparity will not affect behavior. Said another way, ensuring that stimuli are salient to the learner is critically important, as the salience of the targets may override disparity initially.

Pierrel et al. (1970) also evaluated stimulus salience by arranging a multiple schedule to examine the effects of intensity on the acquisition of rats’ bar pressing. This followed a disparity comparison in which the experimenters manipulated the decibel range between the schedule-correlated auditory stimuli (S+ and S−). Results showed the more disparate the stimuli, the quicker the discrimination was acquired. However, to evaluate salience, the experimenters maintained a consistent decibel difference between the S+ and S− and exposed subjects to conditions in which decibel values were on the low or higher end of the total range of decibel values included. This was done to assess how the intensity of the stimuli (salience) influenced discrimination. Results indicated that, regardless of disparity, sets that were less salient (e.g., lower decibel values) were slower to produce discrimination than high-intensity stimuli. Similar to Dinsmoor et al. (1982), if stimuli are not salient, it is less likely that discriminations will be acquired differentially based on the magnitude of the differences between them (i.e., stimulus disparity).

A few applied studies have also shown that stimulus salience affects responding in clinical populations. For example, Dube and McIlvane (1999) evaluated a method for increasing the salience of stimuli. Participants were individuals with ID who displayed stimulus overselectivity (i.e., restricted stimulus control by one component of a stimulus) during a matching procedure. To respond accurately to the match-to-sample task, participants had to attend to two sample stimuli presented side by side (i.e., a compound stimulus), and then select a one-stimulus comparison that contained one of the two samples from an array of three pairs of stimuli when the sample stimuli were still present. Participants’ results suggested that they were only attending to one of the two stimuli in the compound sample (i.e., only one of the two samples was salient and acquired stimulus control). To reduce overselective responding, participants first engaged in a differential observing response (DOR), which required them to match the compound sample to the comparison stimulus that contained the exact same two stimuli. Other comparison pairs in the array contained one of the two sample stimuli and a second, irrelevant form (e.g., if the sample was XI, the array for the DOR may have consisted of AI, XI, and XB). After engaging in a correct DOR, participants were more accurate in matching the compound sample to a one-stimulus comparison that contained one of the two samples. The results showed that requiring a DOR increased the salience of both samples in the compound stimulus and reduced selective attending to only one of the two samples.

Mitteer et al. (2020) also investigated whether stimulus salience affects tact acquisition by evaluating the effects of stimulus background on acquisition. Specifically, experimenters compared acquisition when target stimuli did or did not have a background (i.e., the picture consisted of the target stimulus with scenery behind or a picture with a blank white background, respectively). The majority of participants’ responding was not influenced by the presence or absence of a background stimulus, although other aspects of the methods (e.g., the inclusion of multiple exemplars of each target) may have hindered their analysis of stimulus salience on acquisition. Although differences in acquisition were marginal across participants, researchers suggested that preparing stimuli without backgrounds can be beneficial in preventing faulty stimulus control. However, removing stimulus backgrounds alone does not necessarily ensure that the target’s relevant features are salient to the learner, and should continue to be an area for future investigation.

Recommendations for Research and Practice on the Salience of Target Stimuli

Behavior may come under the control of irrelevant antecedent stimuli, such as portions of instructional materials, that hinder the development of control by critical features. This is likely to occur if some irrelevant aspect of a stimulus is more salient, such as having a greater intensity, hue, or size. For example, faulty stimulus control may occur if an instructor is teaching a learner to discriminate photos of different emotions such as happy, sad, and angry, and the learner attends to irrelevant features of the stimuli (e.g., the person’s nose in each picture, which remains unchanged). By attending to irrelevant features of the stimuli, the learner is unlikely to acquire the targeted discrimination. Therefore, some consideration is needed to remediate or prevent these types of faulty stimulus control. Furthermore, if multiple stimuli are presented together, and only one stimulus is salient, then the other stimuli (or a portion of them) will not come to control responding.

Practitioners may consider the backgrounds of stimuli that will be placed together within training sets (Lie & Alsop, 2010). A practitioner preparing a set of targets for a program should assign stimuli to sets that control for the background, especially if it is not the critical feature being investigated (Mitteer et al., 2020). To increase salience, practitioners and researchers could start with stimuli that have no background so that learners do not attend to an irrelevant feature of the stimulus (i.e., some aspect of the background). Salience can also be manipulated by including multiple exemplars of each target so that the critical features for the discrimination are consistent across exemplars (Layng, 2019; Mitteer et al., 2020). In later stages of training, it is likely that including stimuli with backgrounds would be beneficial, so the learner attends to critical features of the stimulus even though the salience of the respective targets is slightly reduced. Controlling for these variables will help ensure that the learner is acquiring the targeted discrimination and not attending to other irrelevant variables, such as the background color. Future researchers should replicate and extend Mitteer et al. (2020) by comparing the use of varied versus consistent backgrounds and the types of prompts used in these procedures to make the target (and critical attributes) salient.

Practitioners should also maintain consistency in the size of the targets, especially within an array during listener-responding programs. For example, if the practitioner is attempting to teach listener responses to vehicles, and the picture of the police car is smaller than the pictures of the firetruck and boat, the learner may select the police car due to the size of the stimulus and not the relevant components of the picture (e.g., Doran & Holland, 1979; Lorah et al., 2014; Schreibman, 1975). That is, the size of the stimulus may be more salient than the features of the picture itself. If a learner is attending to an irrelevant feature, such as the size of the stimulus rather than the picture, the response will not come under the control of the relevant stimuli, and this discrimination will not be acquired. As discriminations are acquired through training, practitioners and researchers should consider systematically adding or varying irrelevant components of stimuli in training to reduce future restricted stimulus control by irrelevant variables.

Differential Observing Responses

Stimulus overselectivity or restricted stimulus control occurs when isolated stimulus features or components are controlling behavior (Dube & McIlvane, 1999; Lovaas et al., 1979). For example, a learner may select a picture based on its position in a comparison array rather than on the auditory stimulus presented by the instructor, which shows the behavior was under faulty stimulus control. One strategy to remediate restricted stimulus control is the use of a DOR. Practitioners can arrange a verbal DOR (e.g., echoing the sample stimulus; Kisamore et al., 2013) or nonverbal DOR (e.g., matching stimuli based on one component of the stimulus; Dube & McIlvane, 1999), depending on the learner’s echoic repertoire and the targeted skill.

The inclusion of a DOR is particularly useful for increasing the salience of stimuli, because the DOR draws the learner’s attention to relevant aspects of the stimuli that are necessary to acquire the discrimination, as well as highlights for the practitioner why errors may be occurring. For example, if the learner echoes “catch” instead of “cat” when teaching an auditory–visual conditional discrimination, and there is also a baseball glove in the array, then the salience of the auditory stimulus “cat” must be addressed. This error pattern could potentially be resolved by presenting the auditory sample stimulus first (in the absence of the comparison array; Petursdottir & Aguilar, 2016; Schneider et al., 2018), emphasizing the difference between the samples when presenting the auditory sample stimulus (e.g., emphasizing specific sounds in the sample; Fisher et al., 2019), requiring the participant to engage in a correct DOR (i.e., echoing “cat” rather than “catch”), and then presenting the array of comparison stimuli. These modifications increase the salience of the auditory sample by removing the visual comparisons that might block attending to the auditory sample and increasing the intensity of the auditory stimulus (e.g., saying “cat” more clearly or loudly). Only after the critical features of the auditory stimuli are salient should discrimination training be expected to be successful.

It is also important for practitioners to consider increasing the salience of the other stimuli that are not being attended to by the learner. This could also be accomplished by altering the stimulus, such as making it more intense (e.g., brighter, louder, bigger), to try to increase its salience. Behavior analysts should consider doing this when teaching verbal conditional discriminations. For example, if a learner is only attending to one component of an antecedent verbal stimulus—such as saying “water” when asked “What do you drink?” and “What do you drink from?”—the salience of other portions of the antecedent verbal stimuli could be altered. The practitioner could manipulate stimulus salience by elongating or emphasizing the relevant portion of the antecedent verbal stimulus “from” (Axe, 2008). Once the client begins to attend to both stimuli (i.e., attends to “drink” and “from”), the disparity of consequences for responding to those stimuli may establish control by both stimuli. Thus, once the antecedent stimulus becomes more salient, the disparity of consequences for responses to both antecedent stimuli should assist in discrimination training.

Stimulus-Fading Procedures

Another extension of outcomes from research on stimulus salience involves the use of intensity in facilitating the transfer of stimulus control through stimulus-fading procedures (Cook, 1960; Dinsmoor, 1995). By arranging conditions that highlight the critical features of the S+ and S− so they are more salient to the observer, practitioners can subsequently transfer stimulus control to the targeted SD by gradually fading the other aspects of that stimulus. For example, if a learner is not acquiring a discrimination of letters with overlapping components (such as E and F), the practitioner could increase the salience of the unique features of the letter by adding a colored line to each stimulus. Refer to Figure 2 for an example of stimulus-fading steps to increase stimulus salience and then gradually remove these stimulus modifications while establishing this letter discrimination. As the learner begins to respond differentially, the bottom line is gradually faded. By implementing stimulus-fading procedures, practitioners could establish discriminations among stimuli that gradually become less discrepant by manipulating stimulus salience during discrimination training, as previously described (Cook, 1960; Dinsmoor et al., 1983).

Figure 2.

Figure 2

Open in a new tab

Stimulus Fading From Letter E to Letter F. Note. Changes to the intensity of the black/gray lines represent fading across steps

Recommendations for Research and Practice on the Salience of Behavior–Consequence Relations

Previous researchers have manipulated the salience of behavior–consequence relations by using different magnitudes of reinforcement for independent versus prompted correct responses (e.g., Boudreau et al., 2015; Campanaro et al., 2020; Johnson et al., 2017; Leaf & McEachin, 1999; Sundberg & Partington, 1998; Vladescu & Kodak, 2010), such as providing a greater amount of reinforcers when the learner responds independently and fewer reinforcers for correct responses following prompts. For example, Campanaro et al. (2020) found that the immediate implementation of differential reinforcement (either higher quality, larger magnitude, or a dense schedule of reinforcement for independent correct responses) resulted in efficient acquisition for their participants. Thus, practitioners should consider that the practice of differential reinforcement may help increase the salience of the contingencies for each response within behavior-analytic training arrangements. Future studies should investigate the effects of various levels of differential reinforcement on salience of contingencies for learners.

Extensions of Stimulus Disparity and Salience for Practitioners

The salience and disparity of stimuli included in service delivery are also important considerations for practitioners who will be implementing interventions. Service delivery frequently involves the use of multiple intervention components (e.g., prompts, error correction, reinforcement schedules). Less experienced instructors who have not yet learned to rapidly differentiate between interventions may benefit from methods to enhance the salience of instructional materials. Practitioners could format and use data sheets from the onset of training that increase the salience of intervention components, as well as capture relevant variables to be included in quantitative analyses.

Use of Enhanced Data Sheets

Depending on the type of skill being taught, enhanced data sheets should capture variables that help the instructor correctly implement treatment, as well as record behavior that could guide treatment modifications. On an enhanced data sheet, intervention components, such as the target to present on each trial and the types of prompts to provide, could be prominently displayed to increase the salience of these components. LeBlanc et al. (2019) found the use of an enhanced data sheet led to higher procedural integrity (when compared to a nonenhanced data sheet), specifically with regard to rotating target stimuli across trials and counterbalancing comparison stimuli. Thus, the use of enhanced data sheets will assist with increasing procedural integrity along with the collection of data on other variables (e.g., position of selection, biases to stimuli). Furthermore, the information collected on enhanced data sheets will provide valuable information about the learner’s error patterns that may help identify faulty sources of stimulus control. For example, if a learner makes frequent errors during discrimination training but selects stimuli from all positions in the array and does not appear to have any specific type of bias, then the practitioner may suspect that low disparity between comparison stimuli is responsible for poor performance.

Error Analyses

Error analyses can be valuable for practitioners to highlight response patterns that are occurring to or away from specific stimuli. Once practitioners have data from enhanced data sheets, error analyses can be conducted to isolate and identify patterns of correct and incorrect performance. Error analyses identify the most common errors made by the learner during instruction, as well as the different types of errors that may be present (e.g., Grow et al., 2011; Sundberg & Sundberg, 2011). It has been recommended that results from error analyses can help identify faulty sources of stimulus control early in instruction, as well as identify treatments that have a higher chance of remediating faulty stimulus control. The error analyses from enhanced data sheets may point out stimuli in an array that are not disparate enough from each other. For example, the learner always selects the zebra every time “horse” is the sample stimulus. In another example, the learner alternates between selecting the horse and the zebra when either sample stimulus is presented but does not select the other animals in the array. When conducting an error analysis, practitioners should compile completed data sheets for review, define different types of possible errors, code error types by frequency, and finally calculate the percentage of each error type by dividing the number of each type of error by the total number of errors observed (e.g., Grow et al., 2011).

It is also possible that error analyses will show other learner response patterns (e.g., molar and molecular win/stay responses, position biases) that should be remediated with the consideration of the disparity and salience of the stimuli. For example, win/stay responses may occur if the learner responds to a target that was reinforced in a previous trial or step of training (Grow et al., 2011). To help with conducting error analyses, practitioners can circle the stimulus or component the learner selected across trials or sessions (e.g., selecting a specific target many trials in a row, or selecting a specific position in the array such as the left stimulus) in order to measure these error patterns. It is important to note that many error analyses would not be possible to conduct if discontinuous data collection methods (i.e., data collection on the first trial of a session or probes only; Cummings & Carr, 2009) are used because the response location or target stimulus position would not be available. Thus, the results of these error analyses from enhanced data sheets can be used to identify many types of error patterns prior to extended intervention exposure.

Conclusion

Discrimination training is commonplace in behavior-analytic practice. Because a proportion of learners with ASD, ID, and DD may fail to acquire conditional discrimination without specialized training procedures (Green, 2001; Grow et al., 2011; Grow & LeBlanc, 2013; Kodak et al., 2015), the development and use of procedures that manipulate stimulus disparity and salience are important topics to applied behavior analysts. Thus, recommendations based on applications of stimulus-disparity and salience research to applied practice will provide tools for practitioners and avenues of future research.

An important first step in the advancement of stimulus control methods in practice is the standardization and clarification of terms when discussing stimulus disparity and salience. Practitioners who do not consider the effects of stimulus disparity and salience on discrimination training may be unable to identify effective intervention modifications to address faulty stimulus control. This article sought to address these issues by defining these terms and relating them to studies that manipulated disparity and salience to improve discrimination training. The present article also provided recommendations for behavior analysts to consider when developing training procedures and selecting targets for discrimination training, with an emphasis on how recommendations differentially affect stimulus disparity and salience. It is our hope that these recommendations will assist practitioners in increasing the efficacy of discrimination training methods by considering critical variables such as stimulus disparity and salience. We also hope that the present article will lead to multiple avenues of future research on stimulus disparity and salience.

Compliance with Ethical Standards

Conflicts of interest

The authors declare that they have no conflicts of interest.

Funding

Not applicable.

Availability of data and material

Data sharing is not applicable to this article as no data sets were generated during the current study.

Code availability

Not applicable.

Footnotes

Publisher’s Note

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

Contributor Information

Mary E. Halbur, Email: mary.halbur@marquette.edu

Tiffany Kodak, Email: tiffany.kodak@marquette.edu.

References

  1. Alsop B, Davison M. Effects of varying stimulus disparity and the reinforcer ratio in concurrent-schedule and signal-detection procedures. Journal of the Experimental Analysis of Behavior. 1991;56:67–80. doi: 10.1901/jeab.1991.56-67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Axe JB. Conditional discrimination in the intraverbal relation: A review and recommendations for future research. The Analysis of Verbal Behavior. 2008;24:159–174. doi: 10.1007/BF03393064. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Boudreau BA, Vladescu JC, Kodak TM, Argott PJ, Kisamore AN. A comparison of differential reinforcement procedures with children with autism. Journal of Applied Behavior Analysis. 2015;48:918–923. doi: 10.1002/jaba.232. [DOI] [PubMed] [Google Scholar]
  4. Campanaro, A. M., Vladescu, J. C., Kodak, T., DeBar, R. M., & Nippes, K. C. (2020). Comparing skill acquisition under varying onsets of differential reinforcement: A preliminary analysis. Journal of Applied Behavior Analysis, 53, 690–706. 10.1002/jaba.615. [DOI] [PubMed]
  5. Cariveau, T., Batchelder, S., Ball, S., & La Cruz Montilla, A. (2020). Review of methods to equate target sets in the adapted alternating treatments design. Behavior Modification. Advance online publication. 10.1177/0145445520903049 [DOI] [PubMed]
  6. Carroll RA, Kodak T, Adolf KJ. Effect of delayed reinforcement on skill acquisition during discrete-trial instruction: Implications for treatment integrity errors in academic settings. Journal of Applied Behavior Analysis. 2016;49:176–181. doi: 10.1002/jaba.268. [DOI] [PubMed] [Google Scholar]
  7. Cook DA. On vanishing stimuli in instructional material. Journal of the Experimental Analysis of Behavior. 1960;3:292–292. doi: 10.1901/jeab.1960.3-292. [DOI] [PubMed] [Google Scholar]
  8. Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis (2nd ed.). Pearson. 
  9. Cummings AR, Carr JE. Evaluating progress in behavioral programs for children with autism spectrum disorders via continuous and discontinuous measurement. Journal of Applied Behavior Analysis. 2009;42:57–71. doi: 10.1901/jaba.2009.42-57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Davison M, Nevin JA. Stimuli, reinforcers, and behavior: An integration. Journal of the Experimental Analysis of Behavior. 1999;71:439–482. doi: 10.1901/jeab.1999.71-439. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dinsmoor JA. Stimulus control: Part II. The Behavior Analyst. 1995;18:253–269. doi: 10.1007/BF03392712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Dinsmoor JA, Bowe CA, Dout DL, Martin LT, Mueller KL, Workman JD. Separating the effects of salience and disparity on the rate of observing. Journal of the Experimental Analysis of Behavior. 1983;40:253–264. doi: 10.1901/jeab.1983.40-253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dinsmoor JA, Mueller KL, Martin LT, Bowe CA. The acquisition of observing. Journal of the Experimental Analysis of Behavior. 1982;38:249–263. doi: 10.1901/jeab.1982.38-249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Doran J, Holland JG. Control by stimulus features during fading. Journal of the Experimental Analysis of Behavior. 1979;31:177–187. doi: 10.1901/jeab.1979.31-177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dube WV, McIlvane WJ. Reduction of stimulus overselectivity with nonverbal differential observing responses. Journal of Applied Behavior Analysis. 1999;2:25–33. doi: 10.1901/jaba.1999.32-25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Eikeseth S, Hayward DW. The discrimination of object names and object sounds in children with autism: A procedure for teaching verbal comprehension. Journal of Applied Behavior Analysis. 2009;42:807–812. doi: 10.1901/jaba.2009.42-807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Fields L. Fading and errorless transfer in successive discriminations. Journal of the Experimental Analysis of Behavior. 1978;30:123–128. doi: 10.1901/jeab.1978.30-123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Fields L. Acquisition of stimulus control while introducing new stimuli in fading. Journal of the Experimental Analysis of Behavior. 1979;32:121–127. doi: 10.1901/jeab.1979.32-121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Fisher WW, Pawich TL, Dickes N, Paden AR, Toussaint K. Increasing the saliency of behavior-consequence relations for children with autism who exhibit persistent errors: Reducing persistent discrimination errors. Journal of Applied Behavior Analysis. 2014;47:738–748. doi: 10.1002/jaba.172. [DOI] [PubMed] [Google Scholar]
  20. Fisher WW, Retzlaff BJ, Akers JS, DeSouza AA, Kaminski AJ, Machado MA. Establishing initial auditory-visual conditional discriminations and emergence of initial tacts in young children with autism spectrum disorder. Journal of Applied Behavior Analysis. 2019;52(4):1089–1106. doi: 10.1002/jaba.586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gallagher S, Alsop B. Effects of response disparity on stimulus and reinforcer control in human detection tasks. Journal of the Experimental Analysis of Behavior. 2001;75:183–203. doi: 10.1901/jeab.2001.75-183. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Green G. Behavior-analytic instruction for clients with autism: Advances in stimulus control technology. Focus on Autism and Other Developmental Disorders. 2001;16:72–85. doi: 10.1177/108835760101600203. [DOI] [Google Scholar]
  23. Grow L, LeBlanc L. Teaching receptive language skills. Behavior Analysis in Practice. 2013;6:56–75. doi: 10.1007/BF03391791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Grow LL, Carr JE, Kodak TM, Jostad CM, Kisamore AN. A comparison of methods for teaching receptive labeling to children with autism spectrum disorders. Journal of Applied Behavior Analysis. 2011;44:475–498. doi: 10.1901/jaba.2011.44-475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hannula, C., Jimenez-Gomez, C., Wu, W., Brewer, A. C., Kodak, T., Gilroy, S. P., et al. (2020). Quantifying errors of bias and discriminability in conditional-discrimination performance in children diagnosed with autism spectrum disorder. Learning and Motivation,71,. 10.1016/j.lmot.2020.101659.
  26. Hodos W, Bonbright JC., Jr The detection of visual intensity differences by pigeons. Journal of the Experimental Analysis of Behavior. 1972;18:471–479. doi: 10.1901/jeab.1972.18-471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Johnson KA, Vladescu JC, Kodak T, Sidener TM. An assessment of differential reinforcement procedures for learners with autism spectrum disorder. Journal of Applied Behavior Analysis. 2017;50:290–303. doi: 10.1002/jaba.372. [DOI] [PubMed] [Google Scholar]
  28. Kisamore AN, Karsten AM, Mann CC, Conde KA. Effects of a differential observing response on intraverbal performance of preschool children: A preliminary investigation. The Analysis of Verbal Behavior. 2013;29:101–108. doi: 10.1007/BF03393127. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kodak T, Clements A, Paden AR, LeBlanc B, Mintz J, Toussaint KA. Examination of the relation between an assessment of skills and performance on auditory-visual conditional discriminations for children with autism. Journal of Applied Behavior Analysis. 2015;48:52–70. doi: 10.1002/jaba.160. [DOI] [PubMed] [Google Scholar]
  30. Lawrence DH. The transfer of a discrimination along a continuum. Journal of Comparative and Physiological Psychology. 1952;45:511–516. doi: 10.1037/h0057135. [DOI] [PubMed] [Google Scholar]
  31. Layng TJ. Tutorial: Understanding concepts: Implications for behavior analysts and educators. Perspectives on Behavior Science. 2019;42:345–363. doi: 10.1007/s40614-018-00188-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Leaf, R. B., & McEachin, J. (1999). A work in progress: Behavior management strategies and a curriculum for intensive behavioral treatment of autism. Drl Books.
  33. LeBlanc LA, Sump LA, Leaf JB, Cihon J. The effects of standard and enhanced data sheets and brief video training on implementation of conditional discrimination training. Behavior Analysis in Practice. 2019;13:53–62. doi: 10.1007/s40617-019-00338-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lie C, Alsop B. Stimulus disparity and punisher control of human signal-detection performance. Journal of the Experimental Analysis of Behavior. 2010;93:185–201. doi: 10.1901/jeab.2010.93-185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lorah ER, Crouser J, Gilroy SP, Tincani M, Hantula D. Within stimulus prompting to teach symbol discrimination using an iPad® speech generating device. Journal of Developmental and Physical Disabilities. 2014;26:335–346. doi: 10.1007/s10882-014-9369-1. [DOI] [Google Scholar]
  36. Lovaas OI, Koegel RL, Schreibman L. Stimulus overselectivity in autism: A review of research. Psychological Bulletin. 1979;86:1236–1254. doi: 10.1037/0033-2909.86.6.1236. [DOI] [PubMed] [Google Scholar]
  37. Madden GJ, Perone M. Human sensitivity to concurrent schedules of reinforcement: Effects of observing schedule-correlated stimuli. Journal of the Experimental Analysis of Behavior. 1999;71:303–318. doi: 10.1901/jeab.1999.71-303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. McCarthy D, Davison M. Independence of sensitivity to relative reinforcement rate and discriminability in signal detection. Journal of the Experimental Analysis of Behavior. 1980;34:273–284. doi: 10.1901/jeab.1980.34-273. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. McCarthy D, Davison M. On the discriminability of stimulus duration. Journal of the Experimental Analysis of Behavior. 1980;33:187–211. doi: 10.1901/jeab.1980.33-187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mitteer DR, Luczynski KC, McKeown CA, Cohrs VL. A comparison of teaching tacts with and without background stimuli on acquisition and generality. Behavioral Interventions. 2020;35:3–24. doi: 10.1002/bin.1702. [DOI] [Google Scholar]
  41. Nevin JA, Davison M, Shahan TA. A theory of attending and reinforcement in conditional discriminations. Journal of the Experimental Analysis of Behavior. 2005;84:281–303. doi: 10.1901/jeab.2005.97-04. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Nottingham CL, Vladescu JC, Kodak T, Kisamore AN. Incorporating multiple secondary targets into learning trials for individuals with autism spectrum disorder. Journal of Applied Behavior Analysis. 2017;50:653–661. doi: 10.1002/jaba.396. [DOI] [PubMed] [Google Scholar]
  43. Petursdottir AI, Aguilar G. Order of stimulus presentation influences children’s acquisition in receptive identification tasks. Journal of Applied Behavior Analysis. 2016;49:58–68. doi: 10.1002/jaba.264. [DOI] [PubMed] [Google Scholar]
  44. Pierrel R, Sherman JG. Generalization and discrimination as a function of the SD-S∆ intensity difference. Journal of the Experimental Analysis of Behavior. 1962;5:67–71. doi: 10.1901/jeab.1962.5-67. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Pierrel R, Sherman JG, Blue S, Hegge FW. Auditory discrimination: A three-variable analysis of intensity effects. Journal of the Experimental Analysis of Behavior. 1970;13:17–35. doi: 10.1901/jeab.1970.13-17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Pilgrim, C. (2015). Stimulus control and generalization. In F. D. D. Reed & D. D. Reed (Eds.), Autism service delivery: Bridging the gap between science and practice (1st ed, pp. 25-74) Springer.
  47. Rodríguez-San Juan F, Rodríguez G. Assessment of learned changes in stimulus salience as a consequence of preexposure to a single stimulus: Use of a change blindness task. Learning and Motivation. 2017;57:15–21. doi: 10.1016/j.lmot.2017.01.001. [DOI] [Google Scholar]
  48. Saunders KJ, Spradlin JE. Conditional discrimination in mentally retarded adults: The development of generalized skills. Journal of the Experimental Analysis of Behavior. 1990;54:239–250. doi: 10.1901/jeab.1990.54-239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Schneider KA, Devine B, Aguilar G, Petursdottir AI. Stimulus presentation order in receptive identification tasks: A systematic replication. Journal of Applied Behavior Analysis. 2018;51:634–646. doi: 10.1002/jaba.459. [DOI] [PubMed] [Google Scholar]
  50. Schreibman L. Effects of within-stimulus and extra-stimulus prompting on discrimination learning in autistic children. Journal of Applied Behavior Analysis. 1975;8:91–112. doi: 10.1901/jaba.1975.8-91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Skinner, B. F. (1938). The behavior of organisms. Appleton-Century-Crofts.
  52. Skinner, B. F. (1969). Contingencies of reinforcement. Appleton-Century-Crofts.
  53. Sundberg, M. L., & Partington, J. W. (1998). Teaching language to children with autism and other developmental disabilities. Behavior Analysts. [DOI] [PMC free article] [PubMed]
  54. Sundberg ML, Sundberg CA. Intraverbal behavior and verbal conditional discriminations in typically developing children and children with autism. The Analysis of Verbal Behavior. 2011;27:23–44. doi: 10.1007/BF03393090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Swets JA. Indices of signal detectability obtained with various psychophysical procedures. Journal of the Acoustical Society of America. 1959;31:511–513. doi: 10.1121/1.1907744. [DOI] [Google Scholar]
  56. Tiemann, P. W., & Markle, S. M. (1990). Analyzing instructional content: A guide to instruction and evaluation. Morningside Press.
  57. Uwer R, Albrecht R, von Suchodoletz W. Automatic processing of tones and speech stimuli in children with specific language impairment. Developmental Child Neuroscience. 2002;44:527–532. doi: 10.1017/S001216220100250X. [DOI] [PubMed] [Google Scholar]
  58. Vladescu JC, Kodak T. A review of recent studies on differential reinforcement during skill acquisition in early intervention. Journal of Applied Behavior Analysis. 2010;43:351–355. doi: 10.1901/jaba.1984.17-175. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Walpole CW, Roscoe EM, Dube WV. Use of a differential observing response to expand restricted stimulus control. Journal of Applied Behavior Analysis. 2007;40:707–712. doi: 10.1901/jaba.2007.707-712. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. White KG. Conjoint control of performance in conditional discriminations by successive and simultaneous stimuli. Journal of the Experimental Analysis of Behavior. 1986;45:161–174. doi: 10.1901/jeab.1986.45-161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Wolery, M., Gast, D. L., & Ledford, J. R. (2014). Comparison designs. In D. L. Gast & J. R. Ledford (Eds.), Single case research methodology: Applications in special education and behavioral sciences (pp. 297–345). Routledge.

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

RESOURCES