Autistic self-stimulatory behaviors (stims): Useless repetitive behaviors or nonverbal communication?


Leif Ekblad1 , Gerit Pfuhl2

1.Leif Ekblad, Independent researcher, Nöbbelöv 1005, 241 93 Eslöv, Sweden.

2.Gerit Pfuhl, Department of Psychology, Uit The Arctic University of Norway

Keywords: Emotion, Repetitive behaviors, Stims, Autism, Neurodiversity, Communication


Abstract

Background: Some of us engage in repetitive behaviors. Is this just a tick or do these repetitive
behaviors, also knowns as stims serve a function? We here set out to test whether stims are part
of non-verbal communication.

Methods: The study used the neurodiversity concept previously defined in Aspie Quiz (Ekblad,
2013) to research whether and how stims are related to emotions and feelings. A variety of stims
were presented to 5,000 people and the relation to emotion words assessed.

Results: We extracted 22 stims that related to a wide range of emotions, such as being nervous,
anxious, stressed, frustrated, anticipating, excited, angry, upset, closed state of mind and even
pretending to be uninterested. Furthermore, the higher the score on neurodiversity (autistic
tendencies) the more likely a person indicated using stims for expressing emotional states.

Conclusion: Since the study shows that some stims are predictably related to emotions, we
recommend that caregivers of nonverbal autistic children learn these associations which could
facilitate to understand the child.


Introduction

Repetitive behaviors are defined as part of the Autism Spectrum Disorders (ASD) and used as
one of the two diagnostic criteria in DSM-V (American Psychiatric Association, 2013). Lam et
al. (2008) found by using an explorative factor analysis that repetitive behaviors could be broken
down into three categories: Repetitive Motor Behaviors, Insistence on Sameness, and
Circumscribed Interests. The primary target of this study was Repetitive Motor Behaviors. A
review of the literature by Leekam et al. (2011) concluded that restricted and repetitive behaviors
had been relatively neglected in research studies. Gabriels et al. (2008) claimed, using
correlations only, that repetitive and stereotyped behaviors were related to an unusual sensory
profile. In the autistic community, Repetitive Motor Behaviors are typically called stims, a term
that will be used in this study.

Rajagopalan & Göcke (2014) proposed a visual method for identifying stims based on
video sequences, with an aim to assist in autism diagnosis. Abowd et al. (2005) created a
measuring device that could detect stims with an overall system accuracy of 69%, although, the
almost perfect detection accuracy of stimming events could be achieved by allowing for minor
frame-level insertion errors. Bakan (2014) explored the link between stimming and music. Given
that stims are a central part of autism, it is surprising that no hypothesis for the function of stims
has been published, and that no empirical research for a possible function exists to date. Instead,
the available research is focused on detecting and eliminating stims, and treatment plans typically
try to extinguish stims without really knowing their potential function, something that could
cause adverse long-term effects.


Neurodiversity and Aspie Quiz

Neurodiversity sees “atypical neurological development as a normal human difference”
(Jaarsma&Welin, 2012), and hence the right to be different but not treated as ill or needing a cure
(Jaarsma&Welin, 2012, Griffin&Pollak, 2009). Neurodiversity has been defined with factor
analysis using the Aspie Quiz online questionnaire (Ekblad, 2013). The primary factor in Aspie
Quiz was named neurodiversity factor while the secondary factor was named neurotypical factor.
In the construction phase, 1,800 items were evaluated using a population of 550,000 individuals.
Scores were calculated by using the factor loadings as weight factors. The difference between the
neurodiverse score and the neurotypical score was used to provide classifications. A score
difference above or equal to 35 was classified as neurodiverse (ND), a score difference below or
equal to -35 was classified as neurotypical (NT), while scores in between were classified as
mixed. The score difference had a .83 correlation with the AQ test measuring autistic traits
(Ekblad, 2013).

Aims and hypothesis

In Aspie Quiz, factor analysis of stims and neurodiverse communication differences like staring,
unusual emotion expression and tunes or rhythms constantly replaying themselves, cluster on the
same factor. Unlike what some researchers have assumed (Gabriels et al. 2008), stims did not
cluster with sensory traits (Ekblad 2013). This, though, makes it unlikely that stims can be
explained by sensory differences. Hence, our hypothesis is that stims are a means of
communication, i.e. the play a role in neurodiverse communication.

Method

Initially, a pilot study was set up in Aspie Quiz that presented 45 stims to neurodiverse
participants (beta version 8, 2006). For each stim a drop-down menu offered 32 emotion words,
or “I don’t do this stim”, or “non-emotional reason”. Each participant could select up to three of


the 32 possible emotions for each stim, or could select “I don’t do this stim”, or “non-emotional
reason”. Results from this pilot study indicated that to account for 50% of the answers for the
preferred emotion, on average 4.6 emotions were required. For 80%, on average 11.4 emotions
were required and for 90%, on average 16.0 emotions were required. That indicated that answers
were not just random. Some of the emotions were associated with many stims (especially
thinking, bored and anxious). This seemed to indicate that there were sometimes two reasons for
doing a stim: a generic one and a specific one.

Driven by the success of the pilot study, we started an iterative process to find optimal
stim – emotion associations. Here, we recruited a few hundred participants1. Stims that mostly
appeared to relate to generic reasons, and for which no specific reason could be found, were
dropped. To check the validity of the method, a few of the proposed associations were
deliberately selected to be less likely.

Driven by the success of the pilot study, we started an iterative process to find optimal
stim – emotion associations. Here, we recruited a few hundred participants1. Stims that mostly
appeared to relate to generic reasons, and for which no specific reason could be found, were
dropped. To check the validity of the method, a few of the proposed associations were
deliberately selected to be less likely.

After the iterative pilot studies, a list of 22 promising stim – emotion pairs remained, see
Table 1. These 22 stims – emotion pairs and one control with no known or suspected association
(whistling when upset) were tested on a larger population (final version 2 items; H23, 2013). This
stim survey was presented before the regular items in Aspie Quiz, and hence not to bias the
answers by participants knowing their neurotype. Hence, this study is foremost about the function


of stims, irrespective of whether one has few or many autistic traits and defines oneself as
neurodiverse or neurotypical, respectively.

Participants

A total of 4,990 responses were collected (2,890 male and 2,100 female). Mean age was 31 years,
SD 14 years. 54 % of the participants were from the US, 13% were from the UK, and 6% were
from Australia.

Material

The 22 stims from the iterative procedure and their emotion words were presented sequentially to
the participants through the online survey system on rdos.net, see Table 1 for the stim – emotion
pairs. The stims were presented with four response options: a) “I don’t do this”, b) a particular
reason from the iterative pilot studies, c) “I do it independent of mood”, and d) “I do it for other
reasons”. As an example: I flap my hands when I’m excited. The following analysis and results
relate to this study.

Data analysis

First, the prevalence of “I don’t do it” answers was compared to the prevalence of the three
answers that stated that one does a particular stim. Second, the prevalence of “I do it independent
of mood” was compared with the prevalence of the two emotional answer options “I do it for this
reason” and “I do it for another reason”. This is referred to as validity analysis. Third, to measure
how representative the chosen emotion was, the prevalence of “I do it for another reason” was
compared with “I do it for this reason” while excluding “I don’t do it” and “I do it independently
of mood” answers. This is an analysis of the emotional function. For all three analyses, we used a
logistic regression model with the score difference from the Aspie Quiz as the predictor


Results

The distribution of Aspie Quiz answers resulted in a grouping of 904 male as neurodiverse (ND),
1,161 male as neurotypical (NT), 949 female as neurodiverse and 602 female as neurotypical.
The overall prevalence of stims was significantly higher in NDs (60%, range 29% to 92%) than
in NTs (31%, range 8% to 76%), t21=5.26, p < .0001, Cohen’s d = 1.6. For those who do the
stim, it expressed an emotion in both NDs and NTs and to a similar degree. There were 16 out of
the 22 stims where NDs more often did it for the suggested reason, e.g. “jumping up and down
when happy” than NTs did. The stims “pacing”, “rubbing hands together”, and “clenching fists”
were statistically more often done for emotional reasons by NTs than NDs (though, note that
these stims had a vast prevalence difference) but compared to prevalence the validity estimates
had low to negligible effect sizes.

Table 1. Relation between emotions and stims.

Stim

Reason/emotion

Prevalence

Validity/ Is
emotion

Suggested
emotion

Talking to yourself

When hyperfocused

ND

NT

Z

R2

92%

76%

13.2

.042

54%

63%

-4.9

.004

69%

57%

5.1

.008

Walking on toes

For pleasure

ND

NT

Z

R2

48%

22%

18.5

.058

67%

70%

-1.6

.001

62%

38%

7.9

.040

Tapping fingers

When nervous

ND

NT

Z

R2

85%

64%

16.3

.051

55%

59%

-2.9

.002

75%

57%

8.0

.024

Biting lip, cheek or tongue

When anxious

ND

NT

Z

R2

82%

56%

18.3

.045

62%

67%

-3.1

.002

84%

70%

6.7

.020

Grinding teeth

When stressed

ND

NT

Z

R2

60%

34%

17.0

.080

72%

81%

-5.1

.010

79%

69%

4.6

.010


Watching a spinning,
blinking or glittering object

For pleasure

ND

NT

Z

R2

71%

27%

26.6

.012

74%

80%

-2.9

.003

82%

76%

3.3

.006

Pacing

When thinking

ND

NT

Z

R2

77%

42%

21.8

.080

77%

86%

-6.3

.014

76%

70%

3.4

.004

Spinning an object

For pleasure

ND

NT

Z

R2

70%

42%

18.6

.055

74%

78%

-2.3

.002

87%

84%

1.6

.001

Twirling hair

For pleasure

ND

NT

Z

R2

56%

31%

16.5

.042

53%

61%

-3.1

.003

63%

62%

1.6

.002

Pulling hairs from head,
face or body

When frustrated

ND

NT

Z

R2

53%

28%

16.4

.043

57%

57%

-0.2

0

56%

30%

8.2

.046

Rubbing arms or thighs

When anxious

ND

NT

Z

R2

67%

25%

25.6

.112

73%

75%

-1.7

.001

81%

59%

7.5

.031

Rubbing hands together

When anticipating

ND

NT

Z

R2

62%

27%

22.5

.084

64%

77%

-6.8

.018

65%

59%

2.5

.003

Flapping hands

When excited

ND

NT

Z

R2

36%

8%

22.6

.124

76%

84%

-2.7

.007

77%

69%

2.9

.009

Sticking tongue out

When hyperfocused

ND

NT

Z

R2

35%

15%

16.0

.050

81%

88%

-4.1

.017

71%

54%

4.5

.016

Pressing eyes

When stressed

ND

NT

Z

R2

61%

24%

23.0

.089

81%

86%

-3.2

.006

83%

74%

4.6

.012

Clenching and unclenching
fists

When angry

ND

NT

Z

R2

64%

26%

23.4

.091

71%

82%

-6.3

.016

73%

73%

0.0

0

Rocking back-&-forth

For comfort

ND

NT

60%

18%

77%

77%

88%

75%


Z

R2

26.3

.130

0.1

0

5.6

.024

Jumping up and down

When happy

ND

NT

Z

R2

38%

12%

19.1

.076

83%

93%

-4.1

.018

83%

69%

4.2

.017

Biting yourself

When anxious

ND

NT

Z

R2

35%

9%

20.2

.093

75%

81%

-1.0

.001

69%

56%

3.6

.012

Banging head

When upset

ND

NT

Z

R2

29%

6%

19.5

.101

89%

91%

-2.2

.010

85%

66%

4.5

.027

Crossing arms

When in closed state
of mind

ND

NT

Z

R2

82%

61%

15.3

.042

60%

63%

-1.8

.001

75%

65%

4.7

.009

Pullling grass

Pretending to be
uninterested

ND

NT

Z

R2

56%

27%

18.6

.056

51%

51%

0.0

0

47%

33%

4.2

.013

First column is the self-stimulatory behavior, second column is the presented emotion, fifth
column indicates whether subjects did the behaviour for emotional reasons i.e. validity whereas
last column shows inasmuch it is the suggested emotion. Proportion per group shown, Z value
based on score difference, statistically significant p (Bonferroni-Holm corrected) in bold. ND =
neurodiverse, NT = neurotypical, R2 = pseudo R2

Discussion

It was possible to extract 22 stims that related to a wide range of emotions, such as being
nervous, anxious, stressed, frustrated, anticipating, excited, angry, upset, closed state of mind and
even pretending to be uninterested. We found that stims had a higher prevalence in NDs than NTs
also seen in medium to large effect sizes. Neurodiverse people agreed more that the presented
emotion or internal state was related to the specific stim described. Our study cannot distinguish
whether stims are unintentional emotional expressions or whether they are intentionally done for
communicating one’s emotional state. Goldman et al. (2009) found a high prevalence of stims in


low-functioning ASD. This group has to express themselves more non-verbally than verbally
compared with high-functioning autistics. Note, that many of those stims by Goldman et al. were
related to stress and anxiety. In other words, low functioning autistics have these stims because
they are stressed and anxious, not because these stims are diagnostic of neurodiversity or ASD. It
would be interesting to compare it with deaf and blind people. They might as well rely on the
similar stims to express anxiety and stress. We also found that sometimes two stims represent a
similar emotional state, preventing any unambiguous mapping of a stim to emotion. It needs to be
seen whether these are generational, cultural are purely individual differences.

Conclusion

Neurodiverse people express internal states non-verbally with specific stims more so than do
neurotypical people. Awareness of stims as a mean to express emotional states can facilitate
social interactions between NDs and NTs and reduce felt distressed in social interactions. We
especially encourage investigating stims in a natural setting to advance how accurate and reliable
the link is between stims as a mean for emotional communication in neurodiverse people.

Compliance with Ethical Standards

All procedures performed in studies involving human participants were in accordance with the
the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The
Aspie Quiz survey was constructed in such a way that participants could not be identified.
Informed consent was obtained from all individual participants included in the study. Research
items in Aspie Quiz could be skipped by selecting the “?” alternative (checked by default), and
skipping them had no effect on scores. Participants were informed that some items might be for
research only.


References

1.Abowd G.D., Vadas K., Westeyn T., Starner T., Bian X. (2005) Recognizing Mimicked
Autistic Self-Stimulatory Behaviors Using HMMs, Wearable Computers, 2005. Proceedings.
Ninth IEEE International Symposium doi:10.1109/ISWC.2005.45

2.American Psychiatric Association. Diagnostic and statistical manual of mental disorders.
5th edn, text rev. Washington, DC: APA; 2013

3.Bakan M.B. (2014) The Musicality of Stimming: Promoting Neurodiversity in the Ethnomusicology
of Autism MUSICultures . 2014 Vol. 41 Issue 2, p133-161. 29p.

4.Ekblad L. (2013) Autism, Personality, and Human Diversity: Defining Neurodiversity in
an iterative process using Aspie Quiz. SAGE Open 3(3): 2158244013497722.

5.Gabriels R.L., Agnew J.A., Miller L.J, Gralla J., Pan Z., Goldson E., Ledbetter J.C.,
Dinkins J.P., Hooks E. (2008) Is there a relationship between restricted, repetitive, stereotyped
behaviors and interests and abnormal sensory response in children with autism
spectrum disorders? Research in Autism Spectrum Disorders Volume 2, Issue 4, October
2008, Pages 660–670 https://doi.org/10.1016/j.rasd.2008.02.002

6.Goldman S., Wang C., Salgado MW et al. (2009). Motor stereotypies in children with
autism and other developmental disorders. Developmental Medicine & Child Neurology
51: 30–38.

7.Griffin E., Pollak D. (2009). Student experiences of neurodiversity in higher education:
Insights from the BRAINHE project. Dyslexia 15(1), 23-41

8.Jaarsma P, Welin S (2012). Autism as a natural human variation: reflections on the claims
of the neurodiversity movement. Health Care Analysis 20: 20-30


9.Lam K.S., Bodfish J.W., Piven J. (2008) Evidence for three subtypes of repetitive behavior
in autism that differ in familiarity and association with other symptoms. Journal Child
Psychology Psychiatry 49(11): 1193-1200.

10.Leekam S. R., Prior M. R., Uljarevic, M. (2011) Restricted and repetitive behaviors in
autism spectrum disorders: A review of research in the last decade. Psychological Bulletin
Vol 137(4), Jul 2011, 562-593. http://dx.doi.org/10.1037/a0023341

11.Rajagopalan S.S, Göcke R. (2014) Detecting self-stimulatory behaviours for autism diagnosis
IEEE International Conference on Image Processing 2014, DOI:
10.1109/ICIP.2014.7025294

Deixe um comentário

Preencha os seus dados abaixo ou clique em um ícone para log in:

Logotipo do WordPress.com

Você está comentando utilizando sua conta WordPress.com. Sair /  Alterar )

Foto do Google

Você está comentando utilizando sua conta Google. Sair /  Alterar )

Imagem do Twitter

Você está comentando utilizando sua conta Twitter. Sair /  Alterar )

Foto do Facebook

Você está comentando utilizando sua conta Facebook. Sair /  Alterar )

Conectando a %s