Kind Extinction: A Procedural Variation on Traditional Extinction

Abstract

Operant extinction has substantial evidence to support its effectiveness across a variety of populations and behaviors. However, extinction procedures may be less-preferred by learners, caregivers, other community stakeholders, and the staff implementing them. In the current study, we evaluated the effectiveness of a “kind extinction” procedural modification, in which we provided a functionally arbitrary reinforcer in the form of genuine positive regard and validation, contingent on interfering behavior, while implementing escape and tangible extinction. The procedure produced large and rapid decreases in interfering behavior, accompanying increases in alternative behavior, and was rated as acceptable by caregivers and staff. Implications for increasing the social validity of behavioral procedures, as well as contributing to a more kind and compassionate future for the field of applied behavior analysis are discussed.

Full text

Vol.:(0123456789)
Behavior Analysis in Practice
https://doi.org/10.1007/s40617-023-00833-w
SI: COMPASSION INAPPLIED BEHAVIOR ANALYSIS
Kind Extinction: AProcedural Variation onTraditional Extinction
CourtneyTarbox1· JonathanTarbox1,2· TairaLanaganBermudez1· ErinSilverman1· LaurenServellon1
Accepted: 18 June 2023
© The Author(s) 2023
Abstract
Operant extinction has substantial evidence to support its effectiveness across a variety of populations and behaviors.
However, extinction procedures may be less-preferred by learners, caregivers, other community stakeholders, and the staff
implementing them. In the current study, we evaluated the effectiveness of a “kind extinction” procedural modification, in
which we provided a functionally arbitrary reinforcer in the form of genuine positive regard and validation, contingent on
interfering behavior, while implementing escape and tangible extinction. The procedure produced large and rapid decreases
in interfering behavior, accompanying increases in alternative behavior, and was rated as acceptable by caregivers and staff.
Implications for increasing the social validity of behavioral procedures, as well as contributing to a more kind and compas-
sionate future for the field of applied behavior analysis are discussed.
Keywords extinction· kindness· compassion· applied behavior analysis
A substantial body of research has supported the effective-
ness of operant extinction (Lerman & Iwata, 1996) and
extinction is considered one of many critical components of
comprehensive evidence-based behavioral intervention for
autistic individuals (Huete etal., 2014). Despite the over-
whelming scientific evidence in favor of the effectiveness
of extinction procedures, extinction is not without limita-
tions. Extinction can be difficult to implement, especially
for escape-maintained behavior. Early procedural versions
of escape extinction prescribed immediate physical guidance
contingent on interfering behavior (Iwata etal., 1990). Phys-
ical guidance can be challenging to implement with a learner
that is larger and/or stronger than the implementer but, more
important, physically limiting a learner’s movement for
therapeutic purposes is disconcerting from an ethical stand-
point. Behavior analysts have an ethical responsibility to use
the least restrictive procedures that are likely to be effective
(Behavior Analyst Certification Board, 2020, Code 2.15), of
minimizing or eliminating physical guidance (Association
for Behavior Analysis International, 2010), and of treating
others with compassion (Behavior Analyst Certification
Board, 2020, Core Principle 2).
Taylor etal. (2019) defined compassion as action that is
aimed at ameliorating another person’s suffering or distress.
All applied behavior analytic procedures might usefully be
reevaluated in terms of whether they evoke distress and it
seems possible that extinction may be a productive place
to start. Extinction can cause extinction bursts, which can
include a temporary escalation of the behavior, as well as
the occurrence of negative emotional responding (Fisher
etal., 2022). Behavioral escalation and negative emotional
responding could reasonably be interpreted as distress, so
extinction procedures may be one context in which behavior
analysts have an opportunity to take action to ameliorate
that distress in the human beings who we are entrusted to
support.
In addition to extinction potentially increasing learner
distress in some cases, extinction may have low social valid-
ity (Pemberton & Borrego, 2007). That is, extinction pro-
cedures may have low social acceptability to learners, their
caregivers, and other community stakeholders, and interven-
tions not incorporating extinction may be more preferred by
those actually experiencing the interventions (Owen etal.,
Highlights
• Extinction as a procedure may be less-preferred by stakeholders
• Evaluated a modification to extinction that included emotional
validation
• The procedure effectively reduced tantrums for four autistic children
* Courtney Tarbox
ctarbox@firststepsforkids.com
1 FirstSteps forKids, LosAngeles, CA, USA
2 University ofSouthern California, LosAngeles, CA, USA

Behavior Analysis in Practice
2021). From a purely functional, process-based standpoint,
there is nothing cold or unkind about a basic principle of
learning; extinction is a basic learning process that applies
to all operant behavior. But the specific ways in which
extinction procedures are implemented can send negative
messages to community stakeholders and to learners them-
selves. Although most ABA practitioners could tell you that
extinction for behavior maintained by tangible reinforcers
consists of withholding the tangible reinforcer contingent
on the behavior, many practitioners also withhold attention
or other forms of interaction when implementing extinction.
For example, if a child engages in a tantrum that is motivated
by access to a toy, ABA practitioners may implement extinc-
tion by withholding tangible reinforcement (i.e., access to
the toy) during the tantrum while also refraining from pro-
viding attention to the behavior, effectively ignoring the
child. Journal articles that describe extinction or differential
reinforcement procedures commonly describe procedures
that, by default, involve not interacting with the learner,
other than to reinforce specific target alternative behaviors
(Petscher etal., 2009). That is, target interfering behavior
results in no interaction with the learner, target replacement
behavior results in access to the functional reinforcer, and
no other interaction between the therapist and learner is to
occur, other than preventing physical safety concerns.
Social validity is a core value of the field of ABA (Wolf,
1978). According to Wolf, “By giving the same status to
social validity that we now give to objective measurement
and its reliability we will bring society into our science,
soften our image, and make more sure our pursuit of social
relevance” (p. 207). Wolf also emphasized considering
whether the ends justify the means. For example, do the
participants, caregivers and other consumers consider the
treatment procedures acceptable? Wolf proposed this call
to action 45 years ago and some have wondered if we, as a
field, have adequately heeded the call (Callahan etal., 2017).
If we have not adequately addressed the social validity of our
extinction procedures, it may have practical consequences.
When we ask caregivers to implement extinction with their
children, we are expecting a great deal from them. Often, we
are asking caregivers to commit to withholding a reinforcer
that they have a long history of providing to their child. In a
real sense, the behavior–reinforcer relation in that child–car-
egiver interaction may be an important part of their relation-
ship, even if it may be viewed as maladaptive by behavior
analysts who are trying to help the child learn alternative
behaviors. In other words, from the caregivers’ perspective,
reinforcing their child’s behavior may be perceived as a way
of caring for or showing love to their child. In essence, when
we ask caregivers to implement extinction, we are asking
them to do something terribly difficult now, in hopes that
it will lead to a better outcome later. Of course, extinction
may take time, and even a short period can be difficult for
caregivers to adhere to. We are asking caregivers to forgo the
short-term negative reinforcement they get by discontinuing
their child’s discomfort, in order to access the long-term
positive reinforcement of their child having less interfering
behavior and therefore a better life in the future. Although
it is entirely possible that this choice could be “worth it” to
caregivers, it raises the question: What if behavior analysts
could be equally effective in decreasing interfering behavior,
while making the procedures less aversive to implement?
One treatment option might be to exclude extinction
from behavioral intervention packages aimed at decreasing
interfering behavior, and a growing amount of research has
evaluated this. For example, Ingvarsson etal. (2009) evalu-
ated functionally arbitrary noncontingent reinforcement in
the form of preferred tangibles with and without extinction
for escape-maintained interfering behavior and found it was
effective for two of three participants. However, stronger
results were obtained with the addition of escape-extinction.
Other research has produced mixed results, with some stud-
ies showing that extinction may be required in order to sub-
stantively thin out reinforcement for alternative behavior to
clinically reasonable levels (Piazza etal., 1997), and other
studies showing that treatment effects can be produced and
maintained even in the absence of extinction, so long as the
magnitude and quality of positive reinforcers are manipu-
lated sufficiently (Briggs etal., 2019).
A recent systematic review of research on reducing
escape-maintained behavior without escape extinction iden-
tified 38 peer-reviewed articles including 79 participants that
omitted escape extinction (Chazin etal., 2021). Overall,
across the studies, procedures that omitted escape extinction
produced lower rates of target behavior than baseline phases
and comparisons between procedures that included versus
omitted escape extinction produced equivocal results. The
authors interpreted these findings as supporting the poten-
tial utility of behavioral intervention procedures that do not
include escape extinction. Although the results of this review
are highly encouraging and intervention without extinction
is certainly possible under some circumstances, it seems
plausible that extinction may not be able to be eliminated
entirely for some learners and some behaviors.
One alternative to omitting extinction would be to
modify the way in which it is implemented to increase its
acceptability. Piazza etal. (1996) evaluated differential
positive reinforcement for compliance with escape extinc-
tion, with and without physical guidance, in an 11-year-old
boy with autism. During escape extinction without physi-
cal guidance, demands were presented vocally and gestur-
ally and physical guidance was not used. If the participant
did not comply with instructional requests, researchers
continued to present vocal and gestural requests continu-
ously until compliance occurred, at which point the par-
ticipant accessed positive reinforcement. If the participant

Behavior Analysis in Practice
left his seat, instructional demands were not continued, but
he was reminded vocally every 2 min that he could earn
positive reinforcement if he returned to his seat and com-
pleted his work. The number of instructional trials required
before positive reinforcement was earned was gradually
increased from 1 to 28. The procedure was highly effective
and was compared to a subsequent escape extinction pro-
cedure that included physical guidance; escape extinction
without physical guidance was more effective. Although
this study with a single participant was promising, little
additional research has attempted to replicate it and little
research has been published that has attempted to modify
extinction to make it more socially valid.
In popular psychology media, such as blogs or websites
created for parents and other caregivers, caregivers are
often given advice to respond to their child’s tantrums by
validating the child’s feelings, which consists of delivering
attention in the form of statements of concern (Bernstein,
2013). On one hand, responding to an upset child by try-
ing to comfort them is likely a compassionate approach,
and yet some behavior analysts may be concerned that
responding to a tantrum with a preferred consequence,
such as attention, may inadvertently reinforce the behav-
ior. However, if the maintaining function of the behavior
is not attention, then it may still be possible to imple-
ment functional extinction, while also providing a warm,
empathetic, emotionally validating response to the child’s
emotions and experience. If such attempts at comfort and
emotional validation have the effect of alleviating suffer-
ing or distress for the child, then this procedure might be
considered a more compassionate approach to extinction,
in line with the definition of compassion offered by Taylor
etal. (2019). To date, no research of which we are aware
has attempted to evaluate the effectiveness of such a pro-
cedure. The connect and shape model of behavior manage-
ment describes procedures similar to these (Whittingham,
2015), but no known research has been published that has
evaluated the efficacy of that model.
The purpose of the current program evaluation was to
empirically evaluate the effectiveness of real-life proce-
dural modifications to extinction procedures that were being
implemented at a community-based ABA provider. The pur-
pose of the procedural modifications was to make extinction
more “kind” by offering individualized authentic comfort
and positive regard to the learner, while still functionally
implementing extinction. The specific procedural modifica-
tion evaluated in this study consisted of providing function-
ally arbitrary reinforcers in the form of positive attention,
contingent on interfering behavior, while still implementing
extinction for the identified function of the behavior (i.e.,
escape and tangible). In addition, we assessed the accept-
ability of the procedure to caregivers and staff.
Method
Participants andSetting
Four children with autism spectrum disorder, ages 3–4
years, participated in the study. All were receiving com-
prehensive ABA treatment from a community-based ser-
vice provider, with a weekly intensity of 22, 27, 26, and
17 hr of one-on-one treatment for Agnes, Gilbert, Mildred,
and Ingrid, respectively. All were referred for participation
in this program evaluation because they engaged in behav-
iors that their clinical team identified as interfering with
learning. All program evaluation sessions were conducted
by the participants’ regular clinical team, in the context
of their regular behavioral intervention services, which
consisted of center-based services for Agnes, Gilbert, and
Ingrid, and home-based services for Mildred.
Response Measurement
Agnes engaged in a variety of behaviors that made it
challenging for her to benefit from instruction, including
vocal protests and tantrums (i.e., yelling or screaming in
a voice volume above conversational level, crying, falling
to the floor), tensing body, and throwing objects. Gilbert
engaged in tantrums that interfered with learning, includ-
ing screaming, crying, and falling to the ground. Ingrid
and Mildred engaged in tantrums and vocal protesting that
made it difficult for them to benefit from instruction.
Data were also collected on replacement behaviors for
all participants. Replacement behaviors for all participants
included mands for escape or tangible items. During all
sessions, frequency data were collected on tantrums and
manding using pen-and-paper data collection. Data were
also collected on learner attempts to respond to clinician-
initiated instruction. Learner responding to instructional
trials was defined as the learner responding to a clinician
instruction in a contextually appropriate manner within 5 s
of the instruction, excluding mands or interfering behavior
(which were measured separately) or other noncontextu-
ally related behaviors (e.g., repetitive behavior). Responses
did not need to be correct to be scored as responses to
instruction. For example, if a clinician asked a learner to
name an object and they emitted the incorrect name, that
response would still be scored as a response to the clini-
cian’s instruction. Learner responding to instructional tri-
als was scored as present or absent for each instructional
trial and summarized as the percentage of instructional
trials in which the learner responded to the instruction.
Social validity was measured by asking the caregiv-
ers of participants and the direct care staff working with

Behavior Analysis in Practice
the participants questions with Likert-type scales post-
treatment. Table1 depicts the questions and scores for
each participant.
Interobserver Agreement
A second independent observer collected data to assess
interobserver agreement (IOA). The total method was used
to calculate IOA, by dividing the smaller frequency recorded
by the larger frequency recorded for each session, dividing
the smaller frequency by the larger frequency, and multi-
plying by 100. Table2 depicts the percentage of sessions
with IOA data, the mean IOA, and the range of IOA, for all
measures and all participants.
Procedures
All participants received functional assessments, followed
by evaluations of their kind extinction treatment programs.
Indirect Functional Assessment
An indirect functional assessment was conducted for
Agnes’s behavior by using the Questions About Behavioral
Function (Paclawskyj etal., 2000).
Functional Analysis
Interview-informed synthesized contingency analyses
(IISCA), currently referred to as “practical functional assess-
ments,” were conducted for Gilbert, Ingrid, and Mildred.
IISCA procedures were similar to those described in Hanley
etal. (2014).
Kind Extinction Treatment Evaluation
During all sessions in the treatment evaluation, participants
were engaged with their regular learning activities, consist-
ing of structured or naturalistic discrete trial instruction,
depending on the individual participant’s regular treatment
format. Sessions were 10 min in duration for Agnes and 5
Table 1 Social validity data
Questions and scores for caregivers and staff involved in the study. Questions were answered from 1 (strongly disagree) to 5 (strongly agree)
Respon-
dent 1
Respon-
dent 2
Respon-
dent 3
Caregivers
Changes made to my child’s extinction intervention made the intervention more kind 4 5 5
I believe the extinction intervention modeled by the team was effective in decreasing interfering behaviors 4 4 5
I prefer the modified extinction that was modeled over traditional extinction procedures 5 5 4
I would like to implement the modified extinction intervention with my child 4 4 3
I believe that giving comfort as a consequence for interfering behavior made the behavior worse for my child 2 1 1
I value the goal of making ABA procedures feel more kind to my child 5 5 5
Staff
I approve of the goal of the modifications we implemented to make extinction procedures more kind 5 4 5
I believe the extinction procedures we used were effective in decreasing the clients’ interfering behaviors 5 4 5
I prefer the modified extinction procedures we implemented over traditional extinction procedures 5 4 5
I would like to implement the modified extinction procedures we used with other clients in the future 5 4 5
I believe that giving attention as a consequence for interfering behavior made the behavior worse for my client 1 2 3
Table 2 Interobserver Agreement (IOA) data
Percentage of sessions with IOA data, mean IOA, and range of IOA,
for all measures and all participants
% of Sessions Mean IOA Range of IOA
Agnes
Tantrums 27 100 100–100
Mands 32 98 86–100
Response to instruc-
tions
29 100 100–100
Ingrid
Tantrums 29 98 86–100
Mands 17 95 90–100
Response to instruc-
tions
177 100 100–100
Gilbert
Tantrums 31 97 83–100
Mands 28 96 83–100
Response to instruc-
tions
29 100 100–100
Mildred
Tantrums 32 100 100
Mands 32 94 88–100
Response to instruc-
tions
32 97 80–100

Behavior Analysis in Practice
min in duration for all other participants. Before each ses-
sion began, a brief multiple stimulus preference assessment
was conducted, consisting of a single choice trial among
two or three highly preferred stimuli that had been identi-
fied as tangible reinforcers functionally related to the target
interfering behavior. The following contingencies were in
place for replacement behaviors during all baseline, kind
extinction, and follow-up sessions: Learner responding to
instructional trials and mands for the functional reinforc-
ers, in the absence of target interfering behaviors, resulted
in immediate access to the functional reinforcers for 30 s.
Compliance with instructions was not physically prompted.
Individual prompting and prompt fading hierarchies were
determined by each participant’s clinical case supervisor,
but generally included most-to-least prompting for skills that
were on acquisition and least-to-most prompting for skills
that were being maintained.
Baseline Baseline sessions were identical to the test sessions
of the IISCA, with the exception that replacement behav-
iors resulted in access to functional reinforcers, as described
above.
Kind Extinction During kind extinction sessions, replace-
ment behaviors continued to be reinforced with the same
synthesized contingency of escape-to-tangible on a fixed
ratio 1 schedule, identical to baseline. Kind extinction proce-
dures consisted of two components: (1) breaking the contin-
gency between the target behaviors and access to escape and
tangibles; and (2) providing immediate positive attention
and validation (not the functional reinforcer) to the partici-
pant, contingent on the target behavior and any accompany-
ing emotional responding. For example, contingent on tan-
trum behavior, the clinician may have offered their hand to
hold and stated, “Cleaning up can feel really hard. I can see
you’re frustrated.” The specific topographies of comfort and
validation were customized for each individual participant,
based on caregiver report and on direct observations by the
clinical team of what acts of kindness resulted in positive
affect (e.g., smiling, laughing) and movement toward clini-
cians in the past. In addition, during all sessions, if clinicians
observed that an act of kindness resulted in a worsening of
affect (e.g., grimacing, frowning) or moving away from the
clinician, then the clinician chose a different act of kindness
for the rest of the session. In addition, acts of kindness were
offered, not forced. For example, hugs were offered gently,
not physically forced.
To break the contingency between the target behavior and
escape from work, clinicians implemented another instruc-
tion after delivering attention contingent on the target behav-
ior. The same prompting hierarchies were implemented as
were done in baseline (e.g., partial or full verbal models,
gesture prompts, modeling prompts) again omitting physical
prompting. If the learner eloped from the instructional area,
the instructional materials were brought to the learner. To
break the contingency between the target behavior and tan-
gible reinforcers, after the clinician delivered validation con-
tingent on the target behavior, they withheld the tangible
reinforcer until the learner engaged in a replacement behav-
ior in the absence of the target behavior.
Follow‑up After the kind extinction phase, the learner’s
regular ongoing treatment team were trained in kind extinc-
tion and the treatment plan was then implemented across the
remainder of the learner’s regular behavioral technicians.
Formal evaluation data were discontinued until follow-up
data were collected to evaluate if treatment gains had main-
tained after 2 weeks for Agnes, 3 weeks for Ingrid, 4 weeks
for Gilbert, and 4 weeks for Mildred. Contingencies dur-
ing follow-up sessions were identical to the kind extinction
phase.
Results
The results of the indirect functional assessment for Agnes
suggested escape and tangible functions for her behavior
(data available from authors upon request). Figure1 depicts
the results of the IISCA for Mildred, Ingrid, and Gilbert.
The results of the IISCA suggested a synthesized function of
escape-to-tangible for Mildred, Ingrid, and Gilbert.
Figure2 depicts responses per minute of tantrums and
mands during the kind extinction treatment evaluation for all
participants. During the baseline phase, increasing or stable
rates of tantrums were observed for all participants, with a
mean of .47, 1.25, 1.17, and 1.09 response per minute for
Agnes, Ingrid, Gilbert, and Mildred, respectively. Substan-
tial decreases in tantrums were observed during the kind
extinction phase for all participants, with a mean of .18, .2,
.22, and .13 response per minute for Agnes, Ingrid, Gilbert,
and Mildred, respectively. The mean percentage decrease
compared to baseline in the kind extinction condition was
62%, 84%, 81%, and 88% for Agnes, Ingrid, Gilbert, and
Mildred, respectively.
During follow-up, mean rates of tantrums were 0, .05,
.07, and 0 responses per minute for Agnes, Ingrid, Gilbert,
and Mildred, respectively. Follow-up rates represent a mean
reduction from baseline of 100%, 96%, 94%, and 100% for
Agnes, Ingrid, Gilbert, and Mildred, respectively.
Manding was generally low and/or variable during base-
line for all participants, with mean rates of .2, .08, .14, and
.7 responses per minute for Agnes, Ingrid, Gilbert, and Mil-
dred, respectively. During the kind extinction phase, their
average mands per minute increased to .3, 1.5, 1.24, and
1.07, respectively. During follow-up, mands occurred at a
mean rate of .4, 1.15, 1.2, and 1.6, respectively.

Behavior Analysis in Practice
Figure3 depicts the percentage of trials in which all
participants responded to clinician-initiated teaching trials
for all participants. Agnes, Gilbert, and Mildred displayed
low percentages of response-to-instruction during baseline,
with mean percentages of 20%, 0%, and 4%, respectively.
All three demonstrated substantial increases in the behavior
during the kind extinction phase, with means of 84%, 84%,
and 88%, respectively. During follow-up, they responded
to clinician-initiated instructional trials during 100%, 83%,
and 100% of opportunities. Ingrid responded to clinician-
initiated teaching trials during 100% of trials in the baseline
phase and her responding remained high throughout the rest
of the evaluation, with 94% during kind extinction and 100%
during follow-up.
Discussion
Rapid and substantial decreases in interfering behavior were
observed across all four children who participated. Although
not directly targeted in the intervention, replacement behav-
iors in the form of mands and responding to instruction also
substantially improved. Furthermore, follow-up data were
collected and the effects of treatment maintained at 2–4
weeks. These results provide preliminary evidence sug-
gesting that providing genuine positive attention and emo-
tional validation while implementing escape and/or tangible
extinction may be an effective procedure for decreasing
interfering behavior in some children with autism.
In addition to being effective, the data on social validity
provide initial evidence that the kind extinction procedure
was found acceptable to both caregivers and staff. In an
open-ended written comment to their child’s clinical staff,
one caregiver wrote that “I think it helps the kids feel loved
and if we just ignore them they may feel unloved.” It is inter-
esting to note that one staff member, respondent 3 in Table1,
provided a score of 3 (neither agree nor disagree) to the
question of whether providing verbal attention during kind
extinction made the behavior worse for their child. How-
ever, they also followed up with open-ended typed feedback
that stated the following, “Vocal attention didn’t work but
rubbing his back seemed to calm him down during a task.
Overall, I found modified extinction to be successful.”
In response to the question asking parents and staff
whether they preferred kind extinction over traditional
extinction, all participants responded positively, with either
a score of 4 or 5. These social validity data provide pre-
liminary evidence that kind extinction may be more socially
acceptable than traditional extinction. This finding could be
important because people may be more likely to adhere to
treatments that they find more acceptable (Milosevic etal.,
2015). However, these data should be interpreted with
caution because the current study only implemented kind
extinction. Parents had previous experience with traditional
Fig. 1 Practical Functional
Assessment (PFA) Results.
Note. Black circles depict rate
of tantrums and precursor
behavior in the synthesized
tangible and escape condition,
whereas white circles depict
rate of tantrums and precursor
behavior during the control
condition

Behavior Analysis in Practice
extinction before participating in this program evaluation but
their experience with kind extinction was more recent at the
time that they answered the social validity questions. Staff
had previous experience with traditional extinction and were
concurrently implementing traditional extinction with other
learners who did not participate in this program evaluation.
Future research could compare the social validity of tradi-
tional versus kind extinction in people who have an equal
duration of experience with each.
In addition to the effectiveness and social validity of
kind extinction, BCBAs are ethically bound to consider
how intrusive and how compassionate our procedures are
(Behavior Analyst Certification Board, 2020). We might
elaborate on the core principle of treating others with com-
passion by suggesting that creating nurturance and support
for our learners, even during the most difficult of times,
is the purpose of what we do in ABA. Behavioral princi-
ples and procedures, in themselves, are capable of being
implemented more-ethically or less-ethically, as well as
more-compassionately and less-compassionately. It may be
reasonable to argue that the core principle of treating others
with compassion calls on us to consider how compassionate
a procedure is as equally important to how effective it is and
how socially valid it is. Much more future research is needed
to innovate and refine ABA procedures to make a variety of
our procedures more compassionate.
It should be noted that compassion, as defined by Taylor
etal. (2019), consists of actions that alleviate the suffering
or distress of others. We did not directly evaluate whether
the extinction procedures used in this program evaluation
decreased distress for the children who participated. Future
research might include measures of affect so that clini-
cians may quantify whether implementing kind extinction
increases, decreases, or does not change client distress (Reid,
2016). Such data might help evaluate the extent to which
offering emotional validation and kindness during extinction
Fig. 2 Kind Extinction Treat-
ment Evaluation. Note. Black
circles depict rate of tantrums
and white circles depict rate of
mands during baseline, kind
extinction, and follow-up condi-
tions, across all participants

Behavior Analysis in Practice
may have the effect of being a more compassionate approach
to extinction.
One potential concern with providing attention con-
tingent on interfering behavior is that we may reinforce
the behavior with attention, especially if the behavior is
already partially maintained by attention. If the behav-
ior is not already partially maintained by attention, then
one concern is that we may inadvertently “shape up” an
attention function. In either case, if providing attention
contingent on behavior while implementing tangible and/
or escape extinction did indeed reinforce the behavior
then, by definition, the behavior would either persist or
increase in rate. This was not observed in any of the four
participants in the current study. An additional concern
could be that contingent attention could somehow encour-
age attention-maintained interfering behavior outside of
study sessions but this concern does not seem plausible.
If a particular consequence is reinforcing a behavior then
the effect on that behavior is far more likely to be seen
in the context in which the consequence is delivered, as
compared to some other unrelated context in which that
reinforcer is not delivered. In short, if contingent attention
did indeed reinforce the target behaviors in this study, the
most logical place that would have been observed was dur-
ing the treatment sessions that the data reflect.
If future practitioners are concerned about potentially
“shaping up” an attention function, then implementing this
procedure and measuring the effects, as we did in the cur-
rent program evaluation, would give the practitioner the
tools they would need to detect that effect and discontinue
the procedure. Put simply, if a practitioner tried the kind
extinction procedure and the behavior did not decrease,
either because it did not functionally represent extinction
or because it “shaped up” an attention function, then that
would be clearly visible in the treatment evaluation data: The
rate of behavior would not decrease. The procedure could
Fig. 3 Participant Engage-
ment with Instruction. Note.
Data depict the percentage of
clinician-delivered instructional
trials in which participants
responded to instruction during
baseline, kind extinction, and
follow-up conditions, across all
participants

Behavior Analysis in Practice
then be terminated in favor of a more traditional extinction
procedure or another alternative.
When motivating operations are considered, it is perhaps
not surprising that providing contingent attention while
implementing kind extinction did not reinforce the behav-
ior. By definition, when tangible and/or escape-motivated
interfering behavior occurs, a tangible and/or escape estab-
lishing operation must be in place for those reinforcers to be
potent at that time (Laraway etal., 2003). Therefore, it seems
probable that a powerful establishing operation for attention
was lacking, which should predict the lack of a positively
reinforcing effect of attention on interfering behavior in that
context.
Some limitations to the current study warrant discussion.
First, we did not conduct an experimental functional analysis
for one of the four participants and instead used only indi-
rect assessment methods. Future research should consider
including experimental functional assessment methods for
all participants. One related potential concern is that because
we conducted synthesized functional analyses for three par-
ticipants, and did not rule out attention specifically, it is
possible that the interfering behavior of some or all of the
participants had attention functions. This possibility seems
particularly unlikely because, if the behaviors were sensitive
to reinforcement by attention, then the treatment procedure
should have maintained the behavior, as attention was always
given contingent on interfering behavior.
An additional potential limitation is that we did not
compare kind extinction to more traditional extinction pro-
cedures, so it is possible that more traditional procedures
would have produced larger or more rapid reductions in
interfering behavior. Although this is possible, we believe
that the reductions observed in the current study were amply
rapid and large to be clinically significant. In addition, the
social validity data suggest that the procedure was highly
acceptable and, one might argue, that kinder approaches
to extinction do not need to out-perform traditional extinc-
tion, as long as they are effective and stakeholders find them
socially valid. Still, future research might attempt to evaluate
whether kind approaches to extinction may actually outper-
form traditional approaches, such as leading to lesser extinc-
tion bursts.
Finally, one potential limitation is that we implemented
kind extinction on top of a baseline of differential reinforce-
ment; that is, we did not test kind extinction alone. From a
scientific perspective, it would be desirable to evaluate a
procedural modification of extinction alone. From a practi-
cal perspective, such an evaluation may be of little utility,
because extinction is almost never considered an adequate
behavior intervention plan in isolation anymore (Cooper
etal., 2019). The specific reason we chose to not evaluate
kind extinction in isolation is that the evaluation was done
in the context of real-life everyday treatment for learners in
our care and we had no ethically justifiable reason to imple-
ment extinction alone. Still, because we only evaluated kind
extinction in combination with differential reinforcement of
alternative behavior, future research could consider evalu-
ating kind extinction procedures in isolation, if ethical, or
in combination with other ethical intervention components.
One detail the clinicians in this treatment evaluation anec-
dotally noted was that they appreciated how the kind extinc-
tion procedure gave them something to actively do while
implementing extinction, in addition to not reinforcing the
behavior with the functional reinforcer. In other words, it
seems possible that the kind extinction procedure may func-
tion as a type of alternative behavior for staff or caregivers
to engage in, instead of merely omitting reinforcement for
the target behavior. Future researchers may consider evaluat-
ing whether giving caregivers an alternative behavior, such
as providing positive attention, increases the integrity with
which caregivers withhold the functional reinforcer when
being trained to implement extinction with their children.
It also seems possible that providing positive attention may
provide staff or caregivers with a behavior that allows them
to escape from implementing the potentially nonpreferred
procedure of following through with instructional requests.
However, if the kind extinction procedure is less aversive for
staff or parents to implement, then implementing it should
create less of an establishing operation that evokes staff
avoidance of implementing the procedure. In other words,
higher-preferred extinction procedures may be less motivat-
ing for staff to avoid and therefore could lead to higher pro-
cedural integrity on the part of staff or caregivers. This pos-
sibility remains purely speculative and should be addressed
by future research.
We do not suggest that providing positive regard as a
consequence of interfering behavior is the only or best way
to implement extinction more kindly. Perhaps having a dis-
cussion with the learner before implementing extinction,
getting input from them on the form or format of extinc-
tion they would prefer, or empowering the learner to choose
or control some aspect of the extinction procedure could
also make extinction more kind, while maintaining equal
or better effectiveness. In particular, future research could
consider evaluating the social validity of kind extinction
from the learner’s perspective. Treatment choice procedures
(e.g., Hanley, 2010;Rajaraman etal., 2022) could be used
to assess whether learners, even ones who do not yet have
a functional manding repertoire, consistently choose kind
extinction over traditional extinction.
Future research is clearly needed on procedures for
implementing attention extinction with more kindness. At
the most fundamental level, attention extinction can eas-
ily be labeled as “ignoring” a person (even if it is ration-
alized by specifying that the behavior, not the person, is
being placed on extinction) and it is plainly apparent that

Behavior Analysis in Practice
ignoring another human being can appear rude, cold, and
uncaring. Future research might evaluate an attention extinc-
tion procedure where, contingent on the target behavior,
the learner is briefly and kindly told something like, “I’m
really sorry, I know you need my attention right now but I
can’t give it to you,” followed by then providing no atten-
tion until the interfering behavior ceases for a specified unit
of time. Such a procedure may not actually be extinction at
all, but rather, a small amount of attention reinforcement
contingent on the target behavior. Some previous research
has manipulated the parameters of attention reinforcement
for target behavior versus alternative behaviors to make the
attention reinforcement for alternative behaviors more potent
than the attention reinforcement for the target behavior. For
example, Athens and Vollmer (2010) implemented higher
preferred attention (praise) contingent on alternative behav-
ior and lower-preferred attention (reprimands) contingent on
the target behavior and successfully produced a reduction in
the target behavior.
In conclusion, if and when extinction is necessary, the
manner in which it is commonly implemented may appear
cold and uncaring to some. Especially for caregivers of
children with interfering behavior, traditional extinction
may be perceived and described as “The behavior analyst
won’t let me comfort my child when they are upset.” If such
procedures are indeed the only way to decrease behavior in
a socially meaningful manner, and research has been pub-
lished to substantiate that, then it may indeed be worth it to
ask caregivers to make this sacrifice. But if it is possible to
do what we do with greater kindness and compassion, one
might consider it an ethical imperative to at least try.
As the field of ABA takes stock of our daily practices
and considers them from a perspective of compassion and
kindness, we might ask ourselves questions such as, Would
I want a professional ignoring my child when they are genu-
inely upset? Would I want someone to ignore me when I am
genuinely upset, even if I am behaving maladaptively? Can
I remember a time in my own life when I was really strug-
gling and genuinely felt scared or unsupported by others?
Are we okay with potentially allowing our learners and their
caregivers to feel that way? As the field of ABA grapples
with decades-long challenges with public perceptions of our
field, it may be more important than ever to consider ways
in which we can be equally effective at what we do, but
more humane in how we do it. The fundamental question
that spurred the treatment evaluation project that this article
describes was: Can we be more kind while also holding fast
to the foundational dimensions that define our science and
make our practical work so powerful? We fundamentally
believe that the answer to this question is a resounding yes
and, at the same time, much work is left to be done in this
direction. The current article describes just one small step
toward reevaluating foundational ABA procedures with an
aim toward increasing compassion and kindness in every-
thing we do. We hope that this small step will encourage
other researchers and practitioners to continue this journey
with creativity and humility.
Funding Open access funding provided by SCELC, Statewide Califor-
nia Electronic Library Consortium.
Data Availability The datasets generated during and analyzed during
the current study are available from the corresponding author on rea-
sonable request.
Declarations
Conflict of Interest The authors have no conflicts of interest to declare.
Research Ethics The procedures described in this article were per-
formed in accordance with the ethical standards as laid down in the
1964 Declaration of Helsinki and its later amendments or comparable
internationally recognized ethical standards for research.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article's Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.
References
Association for Behavior Analysis International. (2010). Statement
on restraint and seclusion. https:// www. abain terna tional. org/
about- us/ polic ies- andpo sitio ns/ restr aint- and- seclu sion,- 2010.
aspx.Accessed30 Aug 2022
Athens, E. S., & Vollmer, T. R. (2010). An investigation of differential
reinforcement of alternative behavior without extinction. Journal
of Applied Behavior Analysis, 43(4), 569–589.
Behavior Analyst Certification Board (2020). Ethics code for behav-
ior analysts. https:// bacb. com/ wp- conte nt/ ethics- code- for- behav
ior- analy sts/
Bernstein, J. (2013). Five easy, powerful ways to validate your child's
feelings. Psychology Today. https:// www. psych ology today. com/
us/ blog/ liking- the- child- you- love/ 201309/ five- easy- power ful-
ways- valid ate- your- childs- feeli ngs
Briggs, A. M., Dozier, C. L., Lessor, A. N., Kamana, B. U., & Jess, R. L.
(2019). Further investigation of differential reinforcement of alterna-
tive behavior without extinction for escape-maintained destructive
behavior. Journal of Applied Behavior Analysis, 52(4), 956–973.
Callahan, K., Hughes, H. L., Mehta, S., Toussaint, K. A., Nichols, S.
M., Kutlu, M., & Wang, H.-T. (2017). Social validity of evidence-
based practices and emerging interventions in autism. Focus on
Autism & Other Developmental Disabilities, 32(3), 188–197.
Chazin, K. T., Velez, M. S., & Ledford, J. R. (2021). Reducing escape
without escape extinction: A systematic review and meta-analysis

Behavior Analysis in Practice
of escape-based interventions. Journal of Behavioral Education,
31, 186–215.
Cooper, J. O., Heron, T. E., & Heward, W. L. (2019). Applied behavior
analysis (3rd ed.). Pearson Education.
Fisher, W. W., Greer, B. D., Shahan, T. A., & Norris, H. M. (2022). Basic
and applied research on extinction bursts. Journal of Applied Behav-
ior Analysis, 56(1), 4–28. https://https:// doi. org/ 10. 1002/ jaba. 954
Hanley, G. P. (2010). Toward effective and preferred programming: A
case for the objective measurement of social validity with recipi-
ents of behavior-change programs. Behavior Analysis in Practice,
3(1), 13–21.
Hanley, G. P., Jin, C. S., Vanselow, N. R., & Hanratty, L. A. (2014).
Producing meaningful improvements in problem behavior of chil-
dren with autism via synthesized analyses and treatments. Journal
of Applied Behavior Analysis, 47(1), 16–36.
Huete, J., Schmidt, J., & López-Arvizu, C. (2014). Behavioral disorders
in young children with autism spectrum disorder. In J. Tarbox, D.
Dixon, P. Sturmey, & J. Matson (Eds.), Handbook of early inter-
vention for autism spectrum disorders (pp. 563–590). Springer.
Ingvarsson, E. T., Hanley, G. P., & Welter, K. M. (2009). Treatment of
escape-maintained behavior with positive reinforcement: The role
of reinforcement contingency and density. Education & Treatment
of Children, 32, 371–401.
Iwata, B. A., Pace, G. M., Kalsher, M. J., Cowdery, G. E., & Cat-
aldo, M. F. (1990). Experimental analysis and extinction of self-
injurious escape behavior. Journal of Applied Behavior Analysis,
23(1), 11–27.
Laraway, S., Snycerski, S., Michael, J., & Poling, A. (2003). Motivating
operations and terms to describe them: Some further refinements.
Journal of Applied Behavior Analysis, 36, 407–414.
Lerman, D. C., & Iwata, B. A. (1996). Developing a technology for the use
of operant extinction in clinical settings: An examination of basic and
applied research. Journal of Applied Behavior Analysis, 29, 345–382.
Milosevic, I., Levy, H. C., Alcolado, G. M., & Radomsky, A. S. (2015).
The treatment acceptability/adherence scale: Moving beyond the
assessment of treatment effectiveness. Cognitive Behaviour Ther-
apy, 44(6), 456–469.
Owen, T. M., Luczynski, K. C., Rodriguez, N. M., & Fisher, W. W.
(2021). Assessment of children's and caregivers' preferences for
treatments for escape-maintained problem behavior. Journal of
Applied Behavior Analysis, 54(3), 946–965.
Paclawskyj, T. R., Matson, J. L., Rush, K. S., Smalls, Y., & Vollmer,
T. R. (2000). Questions about behavioral function (QABF): A
behavioral checklist for functional assessment of aberrant behav-
ior. Research in Developmental Disabilities, 21(3), 223–229.
Pemberton, J. R., & Borrego, J., Jr. (2007). Increasing acceptance of
behavioral child management techniques: What do parents say?
Child & Family Behavior Therapy, 29(2), 27–45.
Petscher, E. S., Rey, C., & Bailey, J. S. (2009). A review of empiri-
cal support for differential reinforcement of alternative behavior.
Research in Developmental Disabilities, 30(3), 409–425.
Piazza, C. C., Fisher, W. W., Hanley, G. P., Remick, M. L., Contrucci,
S. A., & Aitken, T. L. (1997). The use of positive and negative
reinforcement in the treatment of escape-maintained destructive
behavior. Journal of Applied Behavior Analysis, 30(2), 279–298.
Piazza, C. C., Moes, D. R., & Fisher, W. W. (1996). Differential
reinforcement of alternative behavior and demand fading in the
treatment of escape-maintained destructive behavior. Journal of
Applied Behavior Analysis, 29(4), 569–572.
Rajaraman, A., Hanley, G. P., Gover, H. C., Staubitz, J. L., Staubitz, J.
E., Simcoe, K. M., & Metras, R. (2022). Minimizing escalation by
treating dangerous problem behavior within an enhanced choice
model. Behavior analysis in practice, 15(1), 219–242.
Reid, D. H. (2016). Promoting happiness among adults with autism
and other severe disabilities: Evidence-based strategies. Habilita-
tive Management Consultants.
Taylor, B. A., LeBlanc, L. A., & Nosik, M. R. (2019). Compas-
sionate care in behavior analytic treatment: Can outcomes be
enhanced by attending to relationships with caregivers? Behav-
ior Analysis in Practice, 12(3), 654–666.
Whittingham, K. (2015). Connect and shape: A parenting meta-strat-
egy. Journal of Contextual Behavioral Science, 4(2), 103–106.
Wolf, M. M. (1978). Social validity: The case for subjective meas-
urement or how applied behavior analysis is finding its heart 1.
Journal of applied behavior analysis, 11(2), 203–214.
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.