Increasing Confident Thoughts in an Adolescent With Autism: A Pilot Study

Abstract

Private events such as thoughts and feelings occur within the individual and are inaccessible to outside observers. Creating interventions for troublesome private events, therefore, is challenging. Precision teaching has a number of studies where participants self-count targeted private events and intervene by engaging in 1-min timings of positive thoughts. The present pilot study extended the 1-min timing procedure to a 15-year-old girl with autism spectrum disorder. After the intervention, the participant’s level of despondent thoughts dropped by ÷4.55 in level, with lower levels of anxiety and depression as measured by the Beck Anxiety Inventory and the Beck Depression Inventory.

Skinner (1953) acknowledged that not only do people engage in publicly observable behaviors, but there also exists a wealth of private behaviors within the skin of an individual. These unseen behaviors, called private events, are unobservable to people outside of the organism but are equally important to publicly observable behaviors. Radical behaviorism suggests that private events should be researched as a topic in their own right and in terms of their relationship to public behaviors (Moore, 2011).

Beyond radical behaviorism’s urging, other behaviorally inspired methods, such as acceptance and commitment therapy, rely on the self-monitoring and exploration of private events to help people lead active lives despite the presence of harmful thoughts or feelings (Woidneck et al., 2013). Applied practitioners can effectively alter both the private behaviors (i.e., thoughts, feelings, and urges) and actions associated with covert behaviors by identifying covert thoughts or feelings as behavior (Calkin, 2009; Moore, 2000). Therefore, the necessity for a mechanism to classify and count private events has theoretical and practical significance.

A system devoted to measuring public and private events is precision teaching (PT). PT has a robust literature demonstrating the effectiveness of applied strategies that encourage self-monitoring of private events, also known as “inner behavior” (Kostewicz et al., 2000). Self-monitoring strategies used in PT specifically help an individual become aware of feelings and the events that affect them (McCrudden, 1990). Charting inner behavior has been a successful procedure to recognize and change inner behavior associated with aggression (Kostewicz et al., 2000); thoughts associated with what many have labeled as low self-esteem (Calkin, 2002); negative inner behaviors defined by senior citizens that included thoughts and feelings of loneliness, sadness, and gloom (Kubina et al., 1994); and depression (Clore & Gaynor, 2006; Patterson & McDowell, 2009).

An important consideration for an intervention involving self-counting thoughts and feelings for students with autism spectrum disorder (ASD) is the individual’s ability to observe their own behavior. Research has noted substantial differences between males and females with ASD in terms of recognizing visual cues and their perception of events. A study found that males with ASD performed worse on perceptual attention to detail in go/no-go tasks than males without ASD. Females with ASD, however, were able to perform at the same level of success as their peers without ASD (Oswald et al., 2016). Thus, individuals with ASD are capable of correctly identifying thoughts and feelings and accurately assessing their own inner behaviors.

The present case study extends the PT literature on inner behavior to a teenager with ASD who needed immediate help. Given the literature base on the usefulness of 1-min timings for the reduction of harmful thoughts, the following research question was asked: To what extent will a daily 1-min counting procedure of novel confident thoughts affect despondent thoughts (DTs) of an adolescent girl with ASD?

Method

Participant and Setting

The participant was an adolescent girl, age 15, diagnosed with ASD and clinically diagnosed with posttraumatic stress disorder (PTSD) and depression. The participant was assessed by a local university psychology clinic, which diagnosed PTSD after the participant self-reported feelings of depression, anxiety, and suicidal ideation to a school counselor. The participant also reported frequent thoughts reflecting a negative opinion of herself. Due to her diagnosis of ASD, outside clinicians voiced concern that she would need a specific intervention to interrupt perseverative thoughts and to help her acquire new inner behaviors that reflected confidence. The participant in the present case study demonstrated within-bounds levels of perception for girls with ASD (Lai et al., 2012) and an ability to recognize distinct thoughts and feelings. She attended a public high school in the central Pennsylvania region and had no other concomitant diagnoses. All baseline, training, intervention, and generalization sessions took place in the participant’s home.

Dependent Variables

The dependent variable was DTs. DTs were private events or inner behavior that created fear or worry about future events, accompanied by physiological changes such as increased blood pressure noticeable to the participant. Examples of DTs included “I’m stupid,” “I am a failure,” “I hate myself,” and “I’m ugly.” Prior to the intervention, the participant generated a list of approximately 30 thoughts that were measured via the participant’s self-report; the participant referred to these exemplars until she was sure she could identify a DT. Any DTs not on the list but recognized by the participant as despondent were recorded.

Independent Variable

The confident-thoughts intervention was an extension of the 1-min counting procedure used with various inner-behavior studies (Calkin, 2002). The current intervention consisted of the participant counting confident thoughts within a 1-min time trial within 1 hr of the end of the school day every day during the intervention condition. The participant listed approximately 30 thoughts she self-identified as confident, such as “I’m smart,” “I can do this,” “I’m a good friend,” “I like who I am,” and “I’m a success.” Similar to DTs, the list of self-generated confident thoughts was used to teach the participant to recognize a confident thought. Any other confident thoughts not on the list of 30 but identified by the participant during the intervention were also included in her count.

Research Design

An A-B design was implemented to explore the effects of the confident-thoughts intervention on the number of DTs self-recorded by the participant. Although A-B designs lack important elements of single-case experimental designs, such as verification and replication, they have contributed useful findings to research and practice (Cooper et al., 2020). In the present case study, the baseline phase was formatted as “business as usual” with the participant self-recording despondent thoughts and feelings.

Procedures

Prebaseline

The participant completed both the Beck Anxiety Inventory (BAI) and the Beck Depression Inventory (BDI) prior to baseline data collection; both are criterion-referenced tools to measure anxiety and depression in children and adolescents (Stockings et al., 2015). Michele M. Brown, who worked directly with the participant, assessed the participant’s current levels of anxiety and depression as either mild, moderate, or severe. The participant created two written lists—one included only DTs and the other all the confident thoughts she recognized. Michele M. Brown reviewed the list with the participant, and both agreed on the inclusion of thoughts as either despondent or confident. Afterward, the participant was taught to use a standard mechanical (manual) counter that she would “click” when each thought occurred. The participant was also given a data sheet to record her daily frequency counts of DTs during baseline.

Baseline

Baseline was conducted in the participant’s home and lasted for 14 days under a business-as-usual condition. The 14-day criterion was selected to acclimate the participant to self-recording her despondent thoughts and feelings. During baseline, the participant collected data on the number of despondent thoughts and feelings with a mechanical counter. She set the timer on her cell phone for 30 min and began self-recording within 15 min of waking in the morning. The frequency counts were then transcribed onto a data-tracking worksheet by the participant. At the end of the week, Michele M. Brown then transposed the counts into an electronic version of the standard celeration chart (SCC; CentralReach, 2019; Potts et al., 1993).

Intervention

The confident-thoughts intervention began with the participant thinking and recording the number of unique confident thoughts she thought to herself during a 1-min timing. The participant was instructed to count only discrete thoughts and not count the same instance of a confident thought multiple times. As the participant counted the number of thoughts, she used a mechanical counter to record the number during the interval and then recorded the cumulative number of thoughts illustrated on the mechanical counter on her data collection sheet. The intervention took place within 30 min of the end of the school day, or at 4:00 p.m. Eastern Standard Time on weekends. Although the participant was instructed to avoid counting duplicate confident thoughts during the same 1-min trial, the same unique confident thought did count if it was used during the next day’s trial. The intervention ended per the participant’s request; she reported feeling better and more confident. The participant was given the option to restart the intervention at any time, and Michele M. Brown conducted frequent check-ins with her counselor to make sure she felt supported. After completion of the intervention, the participant was again assessed using the BAI and BDI.

Procedural Integrity

During all phases, Michele M. Brown observed the process of the participant recording her self-counted DTs through direct observation. Michele M. Brown sat next to the participant where she could hear the distinctive “click” of the mechanical counter and the participant self-recording her tally. The audible click indicated that the participant recorded a thought. With the response being observable only to the participant, Michele M. Brown could only create a tally of audible clicks she heard. The resulting tally was compared to the participant’s tally on her data collection sheet for 20% of baseline sessions and for 50% of trials in the intervention condition, or 7 out of 14. Integrity was 100% for all trials across all conditions. It should be noted that the procedural integrity reported was process oriented; as such, we could not verify the content of thoughts as they were inaccessible to outside observers.

Results

Figure 1 presents the results of the study on an SCC segment. SCC segments take a portion of the full six-cycle, 20-celeration-period paper chart (Pennypacker et al., 2003) and faithfully reproduce a specific segment to conform with space requirements in most journals. The SCC segment has a ratio scaled vertical axis and equal-interval scaled horizontal axis. The specific engineering of the SCC permits quantified measures such as trend (aka celeration), level, and bounce (aka variability).

Fig. 1.

A standard celeration chart segment showing data and metrics for despondent thoughts

The SCC segment in Figure 1 has a vertical axis scaled to 200 and a horizontal axis scaled to 56. Those dimensions were selected to accommodate the data in terms of the maximum values and the calendar time. Each X represents the summative daily count of DTs self-counted by the participant per 30-min observation period. The intervention took place over 8 weeks, and the figure shows the baseline, a period of time when the participant became ill, and the intervention (Table 1).

Table 1. .

Data for Confident and Negative Thoughts

Date	Baseline Deceleration (Negative)
11/11	45
11/12	19
11/13	118
11/14	49
11/15	57
11/18	40
11/19	25
11/20	18
11/21	147
11/22	18
11/25	45
	Intervention Deceleration (Negative)
12/03	16
12/04	12
12/05	22
12/06	9
12/09	12
12/10	6
12/11	5
12/12	5
12/16	13
12/17	6
12/18	6
12/19	7
12/20	14
12/21	11
12/22	13
12/23	10
12/24	8
12/29	7
01/02	11
01/06	11
01/07	6

The bottom half of Figure 1 is a replication of the first chart but without data. The SCC segment has three signature within condition measures: level, celeration, and bounce. The level (i.e., gray line in Figure 1) was calculated in both conditions with the geometric mean and represents the average for the condition, or the condition average for DTs. The celeration line (i.e., solid black line) was fit by linear regression with the logarithmic transformation. The bounce lines (i.e., dashed lines) were drawn parallel to the celeration with a 90% confidence interval.

The data show that during baseline, the level for DT was 41. The confident-thoughts intervention produced a level of 9 for DT. A between-condition impact metric called the level multiplier expresses the change between one condition and the next (Kubina, 2019). The level multiplier of ÷4.55 indicates a substantial drop in the average DTs (41 ÷ 9 = 4.55; the division sign was used to indicate a drop, thus ÷4.55).

In baseline, DT data decelerated at ÷1.1 per week (i.e., a 9% weekly decay rate). During the intervention condition, the celeration did not change from ÷1.1 per week. A ÷1.1 suggests the rate of change is very slow and unlikely to be meaningful (Kubina & Yurich, 2012). The closer a celeration value is to X1, the closer it is to not changing. The bounce, or variability, for DTs in baseline was X4.5. In the intervention condition, the bounce for DTs remained at X4.5.

The last set of results concern the BAI and BDI. Postintervention administration of the BDI showed a reduced score from 17 to 16, and the BAI showed a 9-point decrease from 15 to 6. The participant also noted that she felt less depressed and that the intervention was worth doing on the social validity questionnaire.

Social Validity

After completing the experiment, the participant completed an 11-item questionnaire that asked her to rate her performance and the effectiveness of the intervention; it also asked for long-response answers reflecting any potential impacts the intervention had on her feelings. The participant ranked the intervention as a 5 out of 5 in the areas of having access to all necessary materials and in the ability to recognize both despondent and confident thoughts and feelings. She also noted that the intervention helped her feel more confident and less depressed, that it was an intervention she would recommend to other students who needed help with reducing negative thoughts and feelings and increasing confident thoughts, and that the intervention was a worthwhile experience.

Discussion

This case study explored the effects a confident-thoughts intervention would have on the DTs of an adolescent girl with ASD. The results indicate a significant drop in the level of DTs during the intervention. Namely, in baseline the level of 41 meant the participant had at least one or more DTs every second of the 30-min observation period. After the intervention, the level dropped to 9, meaning the participant had, on average, one DT every 3 min 30 s. The reduction in DTs was a relief for the participant.

Further context for the meaningfulness of the changes appears in the quantitative values of celeration and bounce. Both celerations in each baseline and the intervention were unchanged and very slow moving, close to x1. The current study falls in line with the findings of Calkin (1992) and Kubina et al. (1994), in that the 1-min timing procedure had a greater effect on the frequency of inner behaviors than celeration. For DTs, the variance and bounce stayed the same from baseline to intervention.

The measures in the A-B design are very promising, as were the lower levels of depression and anxiety as measured by the BDI and BAI. The changes translate from borderline clinical depression at the beginning of the study to a mild mood disturbance after the intervention ended, leading to a change in the clinical diagnosis from borderline clinical depression to mild mood disturbance (Brouwer et al., 2013). Perhaps the most significant outcome was the participant’s report that she no longer had suicidal ideation after the intervention.

The results further systematically extend the 1-min timing procedure for inner behavior with its inclusion of a teenager with ASD. The current results were similar to Cobane and Keenan’s (2002), who demonstrated noteworthy increases in confident thoughts (x6.25 per week) and a corresponding deceleration of negative thoughts (÷4.5 per week) during a free/tally procedure. The participant with ASD had similar effects regarding her targeted negative inner behaviors, defined as DTs; the 1-min confident-thoughts procedure reduced the overall daily frequency even though the rate of decline as measured by celeration did not change from the baseline.

Another closely related study had participants with mild and moderate depression self-count depressive thoughts and feelings (Patterson & McDowell, 2009). After a timed intervention focusing on positive thoughts, the participants had a lower frequency of depressive thoughts and feelings. Furthermore, they demonstrated lower scores on the BDI after the study ended. However, the participants’ depressive thoughts ranged from 2 to 20, whereas the participant in the present study had a range of 18 to 147 DTs. It would appear the 1-min counting procedure affects negative thoughts at different frequencies.

There are several limitations to this case study. First, an A-B design cannot establish a functional relation; therefore, the results should be interpreted with caution. Additionally, the study ended without the researchers being able to collect maintenance data that would have demonstrated a continued high level of confident thoughts. Future studies should focus on maintenance data, increase the number of participants, and include people of other genders with ASD.

Funding

The research was not supported by any external funding.

Declarations

Conflicts of interest/Competing interests

Michele M. Davidson declares no conflict of interest. Richard M. Kubina Jr. owns equity in CentralReach software. The financial interest has been reviewed by the Pennsylvania State University’s Individual Conflict of Interest Committee and is currently being managed by the university.

Ethics approval

Ethics approval was not required for this pilot study. The study took place at the participant’s home, and the data presented are preexisting data.

Consent to participate

Consent from the study was gained through the participant and the participant’s parent.