Improving Performance Covertly and Remotely with Tactile Stimulation

Abstract

Tactile stimulation can be used to privately prompt or provide feedback for a variety of behaviors. However, technological limitations have primarily resulted in narrow investigations of tactile stimulation delivered on a time-based schedule by a device like a MotivAider or WatchMinder. Recent advances in smartphone and watch technologies have created new ways for practitioners to deliver tactile stimulation privately and remotely to improve behavior. The purpose of this Brief Report is to provide a case study and a description of the advantages and applications of utilizing this technology, including preserving the privacy and dignity of clients by covertly improving performance; prompting behavior based on environmental events independent of scheduled times; delivering feedback quickly, quietly, and remotely without interrupting interactions; and fading dependencies on tactile prompting and feedback.

Tactile stimulation involves the production of haptic sensations on the skin in the form of a light touch, vibration, pressure, shape, or texture (Poletto, 2006). Over the last decade, numerous technologies have adopted tactile stimulation as a form of feedback to shape a user’s interactions with a device or to prompt a specific reaction. For example, surgical training tools, virtual-reality equipment, and even modern video game controllers are now programmed to deliver tactile stimulation in the form of haptic feedback to a user. Similarly, many modern cell phones deliver tactile stimulation in the form of vibrations when a text message, notification, or call is received, prompting users to attend to their device.

In the behavior-analytic literature, tactile stimulation has been successfully used to prompt a wide range of behaviors, including appropriate school behaviors (Anson, Todd, & Cassaretto, 2008; Shih, Wang, & Wang, 2014), completion of daily tasks (Mechling, 2007), on-task behavior (Lancioni et al., 1991), breaks in sedentary time (Green, Sigurdsson, & Wilder, 2016), positive interactions (Mowery, Miltenberger, & Weil, 2010), behavior-specific praise (Markelz et al., 2019), and social initiations (Taylor & Levin, 1998). A variety of technologies have been utilized to prompt behavior using tactile stimulation, including a pager (Taylor & Levin, 1998), MotivAider (Mowery et al., 2010), WatchMinder (Green et al., 2016), Apple iPhone and Watch (Markelz et al., 2019), Invisible Clock (Anglesea et al., 2008), wireless door chime (Anson et al., 2008), mini vibrator attached to the earmolds of a hearing aids (Lancioni et al., 1991), microphone (Roll, 1973), and Nintendo Wii remote (Shih et al., 2014).

The use of tactile stimulation to prompt behaviors or provide haptic feedback has shown clear benefits. However, one area that has received less attention is tactile stimulation delivered remotely from a therapist or teacher as a form of prompting or feedback to a learner or client. The lack of research in this area is likely due, at least in part, to the limitations of the technology that was previously available. Up until recently, commercially available devices capable of delivering remote vibrations from a secondary user have come in the form of vibrating pagers, dog collars, or sex toys. These devices are often too delayed, too stigmatizing, too large, or too loud to produce immediate covert feedback to consumers of behavior-analytic services. Fortunately, recent advances in smartphone and smartwatch technologies have created opportunities for users to incorporate covert haptic feedback in new ways. It is now possible to program these devices so a secondary user can deliver immediate covert tactile stimulation remotely.

There are numerous benefits to delivering tactile stimulation remotely contingent on environmental events as opposed to a time-based schedule. First, tactile stimulation can be multifunctional. For example, tactile stimulation can be used to prompt, reinforce, or punish multiple behaviors. Second, tactile stimulation can be discriminated. For example, distinct forms of tactile stimulation can be used to prompt or provide feedback for different behaviors. Third, tactile stimulation can be covert. For example, feedback can be delivered quietly without interrupting instruction or embarrassing the learner. Fourth, tactile stimulation can be delivered immediately. For example, feedback can be delivered as soon as the target response occurs without waiting for a break or for the behavior chain to end. Fifth, tactile stimulation can be delivered anywhere. For example, feedback can be delivered to a learner even when he or she is not in the same physical location as the instructor. Finally, tactile stimulation can be systematically faded. For example, the intensity of the tactile stimulation can be reduced as the natural contingencies begin to take over, or it can be used on a more intermittent schedule.

The purpose of the following case study was to investigate the utility of covert tactile stimulation as a conditioned reinforcer to increase the percentage of time a 6-year-old with autism would engage in coloring during free time in the classroom.

Method

Participant and Setting

The participant was a 6-year-old boy with a diagnosis of autism spectrum disorder enrolled in kindergarten at a local private elementary school. The student was assigned to an inclusive classroom where he received support from behavior technicians overseen by a Board Certified Behavior Analyst. The classroom consisted of 25 students, one teacher, and one aide. The classroom contained five rectangular tables and one semicircular table. Students could sit on ball chairs (crates with a ball in them), wobble stools (round stools that could rock in circles), and regular chairs. The classroom also had a carpet with seating for 24 students, cubbies where students kept their supplies, two bookshelves, and a teacher’s desk with speakers on it. In addition to the classroom, the participant’s home and a school administrator’s office were used to host pairing sessions to condition the tactile stimulation as a putative reinforcer for coloring.

Materials

A mobile application developed by the first author was used to send tactile stimulation immediately and remotely from an iPhone to an Apple Watch paired using Bluetooth technology. This was accomplished by opening and connecting companion applications on both devices. Tactile stimulation was sent to the Apple Watch when a button in the companion iPhone application was tapped. Additional materials included data sheets for recording performance, a task analysis with instructions for using the application, and paper and crayons needed to complete the coloring activity. Finally, preferred stimuli and activities (bed for jumping, karaoke machine for singing, bouncy ball, etc.) available in the home and the school administrator’s office were utilized to assist in pairing the tactile stimulation as a putative reinforcer for coloring.

Dependent Variable

The dependent variable in this study was the percentage of time engaged in coloring while in the classroom. This behavior was recorded when the teacher provided students in the classroom an opportunity to draw freely. These sessions usually occurred after the completion of a worksheet while students waited for the teacher to check their work. The behavior was not recorded during other activities that required the participant to follow directions when he was coloring (e.g., draw five apples and color them red, color a worksheet of a rabbit). Engagement in coloring was defined as the client within approximately 1 ft of the activity coloring the paper, touching the paper, choosing a crayon, or touching a crayon. If 5 s passed without engagement based on the operational definition or if the client engaged in behaviors targeted for reduction (spitting, physical aggression), he was not considered engaged. Recording sessions began as soon as the participant was presented with his paper and had his crayons and was told to color. Sessions ended when these materials were removed.

Independent Variable

The independent variable used in this study was the delivery of covert tactile stimulation from an Apple Watch worn by the participant delivered contingently after a predetermined amount of time engaged in coloring (i.e., 10, 20, and 30 s). The watch was worn by the participant around his ankle throughout the study after initial observations found him distracted when it was placed on his wrist. During intervention phases, tactile stimulation was delivered remotely by the behavior technician standing around 10 ft from the participant using the companion iPhone application. Additionally, immediately before sessions that occurred during the first week of the initial intervention phase, the participant was given the rule “If you keep coloring, you will get a buzz.” This rule was dropped in later sessions after the participant was observed repeating this rule (e.g., “I’m going to color to get a buzz.”).

Training Sessions

Before the study began, a single 1-hr pairing session was conducted in the participant’s home to condition tactile stimulation as a putative reinforcer for coloring. This was accomplished by delivering tactile stimulation to the Apple Watch a total of 150 times every 30 s while the participant was engaged in preferred activities (playing with the karaoke machine, jumping on the bed, etc.). Additional 10-min pairing sessions occurred throughout the study on Wednesday mornings in the office of the school administrator. This location was chosen for its privacy, convenience, and access to stimuli preferred by the participant (fans, lights, bouncy balls, etc.). During these sessions, the covert stimulation was delivered by the behavior technician on average every 30 s while the participant was engaged with a preferred activity.

Experimental Design

For this study, a withdrawal design composed of six alternating baseline and intervention phases was used. Regardless of the phase, throughout the study, the teachers reminded the participant that he would earn a sticker if he spent 80% or more of his time engaged in coloring during the session (i.e., “If you color for most of the time, you will get a sticker from the teacher.”). The initial baseline phase was started 1 week after the pairing session in the participant’s home. During baseline phases, the Apple Watch was placed on the participant’s ankle underneath his sock when the school day started and removed when the school day ended. During baseline sessions, no additional prompting or tactile stimulation from the teachers or behavior technicians occurred once the participant was provided with his materials and told to color. Intervention phases were introduced once stability in the percentage of time engaged in coloring was observed across sessions during the baseline phases. During the first intervention phase, the behavior technician delivered tactile stimulation to the Apple Watch worn by the participant every 10 s that he was engaged in coloring. Once stability in performance was observed, the second baseline phase was introduced. During the second intervention phase, the behavior technician delivered tactile stimulation every 20 s the participant was engaged in coloring. Again, once stability was observed, a third baseline phase was introduced. During the third, and final, intervention phase, tactile stimulation was delivered every 30 s the participant was engaged in coloring.

Results

During the initial baseline phase, the percentage of time engaged in coloring was recorded at a low level (ranging from 23% to 48%) with a slight increasing trend (Fig. 1). Immediately after the introduction of the first intervention phase, there was an increase in the absolute level change from 36% to 82%, followed by an increasing trend in performance with a range between 64% and 100% of time engaged in coloring. An immediate drop in the absolute level from the first intervention phase to the second baseline phase from 85% to 51% was recorded, followed by a decreasing trend with a range of 33% to 51%. Following the change from the second baseline phase to the second intervention phase, another immediate increase in the absolute level change (33% to 95%) was observed, which preceded a stable, high level of responding with one outlier recorded on the 18th session (ranging from 59% to 95%). The change from the second intervention phase to the third baseline phase demonstrated another large absolute level change (91% to 39%), followed by high variability in performance throughout the phase (10% to 65%) occurring at low levels. The change from the third baseline phase to the third intervention phase indicated a final large absolute level change (39% to 92%), where responding remained high throughout the phase (ranging from 82% to 100%). Throughout the entirety of the study, less than 4% of the data recorded overlapped across baseline and intervention phases, indicating a very large degree of separation between the percentage of time engaged in coloring across baseline and intervention sessions. Interobserver agreement (IOA) data were collected for 17% of sessions using a total-duration recording method. Results indicated that IOA was 93% with a range of 84% to 100%. Procedural fidelity data were also recorded across 17% of sessions and were reported at 96% with a range of 80% to 100%.

Fig. 1.

The impact of tactile stimulation delivered contingently every 10, 20, and 30 s on a participant’s percentage of time engaged in coloring in the classroom

Discussion

Results from this case study demonstrate how tactile stimulation delivered remotely can be used to discreetly improve responding. There are several points that are worthwhile to note. First, the intervention was delivered remotely by an observer, which immediately helped to remove the presence of the therapist as a prompt for the required behavior. Second, the feedback was delivered privately, which helped to preserve the participant’s dignity and kept him from standing out from his peers during the coloring sessions. Third, the intervention did not interfere with the occurrence of the behavior. In other words, the participant was able to continue coloring while the tactile stimulation was being delivered. Finally, the gains in coloring behavior remained while the intervention was systematically faded across intervention phases.

There are several limitations to the case study reported. First, there was only one participant in the study. Additionally, this participant had some previous exposure to tactile stimulation a year prior when it was used to modify on-task behavior in the classroom. The data from this intervention were also promising but are not reported here due to multiple methodological limitations. Second, the experimental design did not include a phase where tactile stimulation was delivered prior to pairing. As a result, it is possible the stimulation itself could have functioned as a reinforcer without the necessity for pairing or the rule provided. Third, the study only investigated modifying a single response class. If the tactile stimulation served as a reinforcer for coloring behavior, it is feasible that it could have also served as a reinforcer for other skills that could have assisted in the classroom. Fourth, although the study demonstrated that gains remained even as the duration between delivery of the tactile stimulation was increased, the school year ended before the intervention could be fully faded from the activity. Fifth, the application used to send tactile stimulation to the watch from the phone is not yet publicly available. However, the application was fairly simple to develop and could possibly be re-created by another developer familiar with Apple’s Swift programming language. Finally, although social validity data were not formally collected, informally, the client, his caregivers, and his teachers made several positive comments about the intervention. The client requested the watch several times throughout the day, both during the intervention and as far as 1 year after the intervention was completed. The client’s parents noted that they appreciated that the intervention was private and did not make their child stand out from his peers. Finally, after seeing the success of the intervention, the client’s teachers requested that the watch be used throughout his day across projects and asked if a similar intervention could be done with several other students.

In conclusion, the ability to remotely and covertly deliver tactile stimulation to a learner has fantastic potential across numerous response classes. For example, it could be used to covertly eliminate filler phrases used by public speakers. Moreover, it could be used by supervisors to provide covert feedback to trainees without interrupting a treatment session. Additionally, it could be used to prompt social interactions without embarrassing a client. Finally, it could be used to replace the audio cues used in tag teaching to provide feedback in locations in which it is too loud to hear or too public to share. In other words, the possibilities for this technology are vast and exciting to consider.

Funding Information

The authors received no financial support for the research, authorship, or publication of this article.

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Ethical Approval

This project was submitted to the University of Louisville Human Subjects Protection Program and was deemed to not meet the “Common Rule” definition of human subjects’ research. As such, the project did not require institutional review board review.

Informed Consent

This case study relied on an archival clinical data set. Consent to use this data set was collected from the caregiver.