=Paper=
{{Paper
|id=Vol-1419/paper0091
|storemode=property
|title=The Capacity for Implicit Social Learning in Relation to Autistic Traits.
|pdfUrl=https://ceur-ws.org/Vol-1419/paper0091.pdf
|volume=Vol-1419
|dblpUrl=https://dblp.org/rec/conf/eapcogsci/MacinskaKJ15
}}
==The Capacity for Implicit Social Learning in Relation to Autistic Traits.==
https://ceur-ws.org/Vol-1419/paper0091.pdf
The Capacity For Implicit Social Learning In Relation To Autistic Traits
Sylwia Macinska (S.T.Macinska@2014.hull.ac.uk)
Department of Psychology, University of Hull
HU6 7RX, United Kingdom
Manon Krol (M.Krol@2013.hull.ac.uk)
Department of Psychology, University of Hull
HU6 7RX, United Kingdom
Dr Tjeerd Jellema (T.Jellema@hull.ac.uk)
Department of Psychology, University of Hull
HU6 7RX, United Kingdom
Abstract such as subtle facial expressions and bodily gestures. People
learn social cue-outcome relationships and use this
People learn social cue-outcome relationships and use this
knowledge to guide their behavior, even in the absence of knowledge to guide their behavior, even in the absence of
explicit knowledge of social contingencies. The aim of the explicit knowledge of social contingencies (Lewicki, Hill,
current study was to investigate variations in implicit social Czyzewska, 1992). Social cues may implicitly acquire value
learning abilities between typically developed individuals on a short timescale, sometimes within the space of a single
with low and high number of autistic traits, as assessed by the interaction (Heerey & Velani, 2010).
AQ questionnaire. In the learning phase, participants Given the importance of social intuition, an impaired
repeatedly observed two different identities whose gaze ability to implicitly encode and decode social cues might
direction and facial expression were manipulated to convey result in difficulty adjusting behavior to the demands of
either a pro- or anti-social disposition toward the observer.
social situations. Indeed, such an impairment has been
These dispositions were determined by specific contingencies
between these cues. Crucially, the participants were not aware proposed to be a crucial factor underlying the social
of these contingencies (as confirmed in the debrief session). deficiencies in autism spectrum disorders (ASD; Frith &
In the test phase, the participants showed specific biases in Frith, 1999). ASD is characterized by deficits in social
their perceptual report of morphs of the two identities, which interaction and social communication, including nonverbal
reflected that they had implicitly learned the identities’ communicative behaviors. As a result, individuals with
dispositions. The results indicated that in the current paradigm ASD are likely to misinterpret social cues and engage in
others’ dispositions can be learned implicitly (i.e. without socially inappropriate behavior. It has been argued that the
awareness of the cue contingencies), and that this ability is social impairments in ASD are linked to an impaired theory
correlated with AQ scores; participants with higher AQ scores
of mind; the ability to understand others as intentional
showed less implicit social learning than those with lower AQ
scores. agents, that is, to interpret their minds in terms of intentional
states such as desires and beliefs (Baron-Cohen, 2000).
Keywords: Implicit learning; Autism spectrum disorder; There are indications that social deficits seen in individuals
Intuition; Social cognition; Emotional facial expression. with ASD are especially related to failure in automation of
social cue processing rather than in the ability to understand
Introduction intentions per se (Jellema et al., 2009; Senju, Southgate,
People exchange large numbers of nonverbal cues, which White & Frith, 2009). This may be compensated for through
are typically interpreted in a seemingly effortless manner. deliberate reasoning about others’ intentions. However, the
The ability to encode and decode social information is interpretation of social cues using effortful cognitive
crucial to successfully navigate the social world. This social processes would be much slower and possibly less accurate
competence seems to be employed automatically and than in involuntary processing.
involuntarily; people have a tendency to attribute Nevertheless, a recent meta-analysis looking at implicit
dispositional causes to non-verbal behavior in order to learning in individuals with ASD did not find any anomalies
obtain a quick impression of other’s thoughts and feelings, on a number of implicit learning tasks (Foti et al., 2014).
rather than using effortful cognitive processes and deliberate However, the reported studies examined implicit learning in
reasoning. non-social domains, with tasks such as artificial grammar
Implicit (social) learning is widely assumed to play a learning, serial reaction task or contextual cueing. There is
central role in social cognition and is regarded as a cognitive an emerging body of research suggesting that implicit
substrate of social intuition (Lieberman, 2000). According learning is not a global impairment, but rather may be found
to Lieberman, social intuition involves making rapid either intact or impaired depending on the type of the
judgments about the emotions, intentions or attitudes of information to be learned (Travers et al., 2013).
others on the basis of learned sequences of nonverbal cues,
554
The autism spectrum condition could be considered as a Sharpsburg, PA), at a distance of approximately 50cm from
scale of social aptitude with autistic traits found throughout the participant.
the population to differing degrees, with clinical levels of
autism at the high end of this continuum (Constantino & Procedure
Todd, 2003). The Autistic-spectrum Quotient (AQ)
questionnaire is one of the measures that has been
Learning Phase The learning phase used in the current
developed to gauge autistic traits in the population at large
study was broadly consistent with the learning phase of the
(Baron-Cohen et al., 2001). It is a self-report questionnaire
experiment conducted by Hudson et al. (2012). Participants
with a maximum score of 50, where a score of 32 or higher
viewed 64 clips of two different identities, 32 for each
may meet the diagnostic criteria for ASD (Baron-Cohen et
identity. Each clip consisted of 20 frames, the first and last
al., 2001). Although not currently used as a diagnostic tool,
frame were displayed for 750 ms and the other 18 frames for
the questionnaire has been found to be a sustainable
30 ms each. Two facial features were manipulated: gaze
indicator of ASD with 83% of accuracy (Woodbury-Smith,
direction and facial expression. Each clip started with the
Robinson, Wheelwright, & Baron-Cohen, 2005). As
gaze either directed towards the participant (direct gaze
typically developed individuals with relatively high levels
direction) and then gradually averted horizontally until at a
of autistic traits are likely to share some social and cognitive
30º angle away from the observer at the end of the clip
deficits with individuals with ASD, it should be possible to
(indirect gaze direction), or began with a 30º aversion and
test typically developed individuals with varying AQ scores
ended with direct gaze direction (the clip played
to enhance our understanding of ASD (Hudson, Nijboer, &
backwards). To manipulate facial expressions, the clip
Jellema, 2012).
began either with an identity displaying a happy expression,
which gradually morphed into an angry expression, or began
Current Study with an angry expression which morphed into a happy
The aim of the current study was to investigate whether expression (clip played backwards).
implicit social learning would take place in a simple Crucially, each identity portrayed a different combination
computer-based task, and whether there are variations in of gaze direction and emotional expression, such that each
implicit social learning abilities in typically developed could be said to have a different disposition toward the
individuals that vary in autistic traits (AQ). It was observer (Figure 1). For identity A, gaze averting away from
hypothesized that higher AQ scores would be correlated the observer was accompanied by a change from a happy to
with poorer implicit social learning. In the current study, an angry expression, while gaze movement towards the
two types of nonverbal social cues were manipulated to observer was accompanied by a change from angry to happy
affect social judgments: emotional expression and gaze (played backward). This identity can therefore be inferred to
direction. Participants were hypothesized to implicitly learn hold a pro-social disposition toward the observer; smiling
the agent’s disposition toward them on the basis of specific when looking at the person and frowning when looking
combinations of these two cues. The experiment consisted away. For identity B, the reverse cue combinations were
of an initial learning phase, followed by a test phase, in used and this identity can therefore be inferred to hold an
which the extent of learning was measured. antisocial disposition toward the observer.
Each identity displayed happy and angry expressions for
Methods exactly the same amount of time and looked at and away
from the observer for exactly the same amount of time. This
was to ensure that a social disposition could only be learnt
Participants on the basis of the specific combination of two cues; each
Fifty-one undergraduate students (20 women) from the cue on itself could not cause any social learning effects. The
University of Hull participated in the experiment in social disposition of the character used in the study was
exchange for course credit. All participants provided written counterbalanced across participants; for half of the
informed consent prior to the experiment. After exclusion of participants James was holding a pro-social disposition
one participant (see below) the mean age of participants was while Simon was holding an anti-social disposition
21.3 (SD=2.77). (Experiment 1), while for the other half Simon was the pro-
social and James the anti-social identity (Experiment 2).
Stimuli
Stimuli were created using Poser 7 animation software
(Curious Labs, Inc., Santa Cruz, CA, & e frontier, Inc.,
Scotts Valley, CA). Two male identities were used named
James and Simon. The characters were oriented facing the
observer throughout the presentation and were depicted
from the top of the shoulders upwards. The stimuli were
presented on a 21 inch monitor (100 Hz refresh rate) using
e-Prime software (Psychology Software Tools, Inc.,
555
Figure 2: Two test trials are shown (top row and bottom
row). The central panels show morphs consisting of 50%
Figure 1: Pro-social and anti-social dispositions of the two Simon and 50% James (M3), for angry (top) and happy
identities in the learning phase of Experiment 2 (upper (bottom) facial expressions. The morphs were flanked by
panels: Simon, bottom panels: James). James and Simon displaying neutral expressions, labeled as
1 and 2.
Debriefing Phase Directly following the learning phase,
participants were required to give verbal responses to a AQ Questionnaire At the end of the experiment,
series of questions checking whether they had picked up the participants completed an online version of the AQ (Baron-
cue-identity contingencies. The questions were: (1) Could Cohen et al., 2001).
you describe what you just have seen? (2) How many
different identities did you see? (3) What can you tell me
about their facial expressions? (4) What can you tell me
Results
about their gaze direction? (5) Did you detect certain The debriefing phase indicated that one participant had
patterns between facial expressions, gaze directions and discovered the cue-identity contingency and his data was
identities? Answers were manually recorded. removed from the analysis.
The dependent variable consisted of mean scores on the
Testing Phase In the testing phase, a morph of the two five levels of morphs used in the test phase. A score of 1
identities was presented in the centre of the screen, with the indicated that participants believed the morph resembled the
faces of James and Simon displaying neutral expressions on pro-social identity more than the anti-social identity, while a
either side of the morphed identity (Figure 2). The morphed score of 2 indicated that participants believed the morph
identity was either smiling or frowning, and was composed resembled the anti-social identity more than the pro-social
of different proportions of either, the smiling James and the one.
smiling Simon or the frowning James and the frowning Two separate experiments were conducted on two
Simon. Five morph levels (proportions) were used: M1 = participant groups, none of the participants participated in
60% pro-social and 40% anti-social, M2 = 55% pro-social both experiments. In Experiment 1, James had the pro-social
and 45% anti-social, M3 = 50% pro-social and 50% anti- disposition and Simon the anti-social disposition; in
social, M4 = 45% pro-social and 55% anti-social, M5 = Experiment 2 the dispositions were reversed. The data from
40% pro-social and 60% anti-social. There were in total 20 both experiments was analyzed using a 2x5x2 repeated
test trials (5 morph levels x 2 emotions x 2 repetitions). measures ANOVA with Morph Facial Expression (happy,
Participants were required to select whether the morphed angry) and Morph Level (M1, M2, M3, M4, M5) as within-
identity resembled more closely the pro-social identity (key subject factors and Experiment (Exp1, Exp2) as between-
1) or the anti-social identity (key 2) The rationale was that subjects factor.
when participants would have implicitly learned that The main effect of the factor Morph Facial Expression
identity A had a pro-social disposition and identity B an was significant (F1,48 = 6.7, p = .013, ηp2 = .12), with higher
anti-social disposition toward them, then they would be scores for the angry morph than for the happy morph. This
more likely to judge the smiling morph as more similar to indicated that implicit learning had taken place. The Morph
identity A, and the frowning morph as more similar to Facial Expression x Experiment interaction was non-
identity B, because intuitively identity A would be significant (F1,48 = .30, p = .59, ηp2 = .006). The main effect
associated with a ‘positive’ and identity B with ‘negative’ of Morph Level was significant (F4,192 = 33.4, p < .001, ηp2
attitude. = .41), indicating that participants were sensitive to the
different morph proportions. The main effect of Experiment
was non-significant (F1,48 = 3.28, p = .076, ηp2 = .064). The
remaining 2-way interactions, and the 3-way interaction,
were all non-significant (all p’s > .16) (Figure 3, top panel).
556
The mean AQ score was 15.5 (SD = 5.6), ranging from 5 Discussion
to 32. A correlation analysis for the mean difference score
The main finding was that in the current paradigm typically
[mean score for Happy Morph across the 5 morph
developed individuals were capable of implicit social
proportions minus mean score for Angry Morph across the 5
learning. It suggests that the pro- or anti-social dispositions
morph proportions] and the individual AQ scores revealed a
of another individual can be implicitly understood on the
significant negative correlation (r = -.287, p = .043),
basis of specific social cue combinations conveyed in a
reflecting that an increase in the level of autistic traits
social encounter, without one being aware of the
corresponded with a decrease in the extent of implicit social
contingencies. The implicitly learned dispositions may serve
learning.
as useful predictors of an individual’s future behaviour,
affecting one’s subsequent responses to that individual. A
second important finding was that this effect is influenced
by individual differences in autistic traits. Those with less
autistic traits (low AQ score) implicitly learned the social
dispositions better than those with more autistic traits (high
AQ score). Presumably, those with lower AQ scores are
better in ‘picking up’ the meaning of social contingencies.
It is tempting to speculate about the implications of the
current findings for individuals with ASD. In line with the
theory of an autism spectrum continuum (Baron-Cohen et
al. 2001), extrapolation of the results of the high AQ group
to individuals with ASD would suggest more severe deficits
in the implicit social learning ability in ASD, which might
underpin core deficits in social understanding in ASD.
According to a recent account, sensitivity to social signals
such as direct eye gaze is essential for learning through
social interaction (Sodian, Schuwerk & Kristen, 2015).
Impaired processing of social cues in ASD could explain
insufficient learning from others and the same may be
expected of those with high AQ scores.
However, it is not quite clear whether individuals with
high AQ scores are less receptive to social cues per se, or
whether they are equally receptive, but have a diminished
ability to learn associations between emotions and specific
individuals. Face identity processing has been found to be
impaired in people with ASD for tasks that require face
memory (Weigelt, Koldewyn & Kanwisher, 2012). This
deficit is much stronger when a delay intervenes between
two presentations of an identity. However, in the current
study the morphs in the testing phase were presented
simultaneously with the faces of both identities, which
should have reduced memory demand for this particular
task.
It is important to note that the task employed in this study
is a simple computer task, which is only a human-artefact
representation of a social interaction. While the under-
complexity of the experimental design allows controlling
for extraneous/confounding variables, the results of the
Figure 3: The mean scores for all participants in the two current study need to be approached with caution when
experiments are shown (top panel). For illustrative generalising to naturalistic settings.
purposes, based on their scores on the AQ questionnaire, There is also a possibility that the variations in implicit
half of participants were allocated to a low AQ group learning between individuals low and high in autistic traits
(middle panel, M=11.5, SD=2.04) and the other half to a may be due to low-level visual learning, rather than to a
high AQ group (bottom panel, M=19.79, SD=4.90). form of social learning. The face configuration of a happy
morph involves eyes looking at the participant and a smiling
mouth, which is a closer visual match to the happy pro-
social face than the angry anti-social face, which has the
eyes averted. Similarly, the angry morph is a closer visual
557
match to the angry anti-social face. If the learning occurred Constantino, J. N., & Todd, R. D. (2003). Autistic traits in
on the basis of the closer matching of low-level features – the general population: a twin study. Archives of General
eye and mouth configuration – it would indicate that Psychiatry, 60, 524–30.
individuals high in autistic traits show deficits in implicit
learning that are not exclusive to the social domain. Foti, F., Crescenzo, F. De, Vivanti, G., Menghini, D., &
Subsequent studies, involving non-social stimuli, will Vicari, S. (2014). Implicit learning in individuals with
investigate whether the lower propensity for implicit autism spectrum disorders : a meta-analysis.
learning in the high AQ group is specific to the social PsychologicalMedicine,15, 1-14.
domain and does not apply to non-social learning, or
whether it is a more generalised impairment. Frith, C. D. (1999). Interacting Minds--A Biological Basis.
It is as yet unknown whether the deficits in implicit social Science, 286, 1692–1695.
learning in the high AQ group are specific to implicit social
learning and does not apply to explicit social learning. The Heerey, E. a., & Velani, H. (2010). Implicit learning of
lack of automatic processing of social cues may be social predictions. Journal of Experimental Social
compensated for through deliberate reasoning about others’ Psychology, 46, 577–581.
intentions. For example, Senju and colleagues (2009) found
a striking dissociation between implicit and explicit theory Hudson, M., Nijboer, T. C. W., & Jellema, T. (2012).
of mind; while the performance of individuals with ASD on Implicit social learning in relation to autistic-like traits.
an implicit theory of mind task revealed intriguing Journal of Autism and Developmental Disorders, 42,
difference from that of controls, they showed intact explicit 2534–2545.
theory of mind. As a result, one may expect intact explicit
social learning in individuals with high AQ scores. Jellema, T., Lorteije, J., Van Rijn, S., Van T’Wout, M., De
However, the interpretation of social cues using effortful Haan, E., Van Engeland, H., & Kemner, C. (2009).
cognitive processes would be much slower and possibly less Involuntary interpretation of social cues is compromised
accurate than in involuntary processing, which may be the in autism spectrum disorders. Autism Research, 2, 192–
primary reason for difficulties in social interactions found in 204.
ASD. Future studies should contrast the ability for implicit
and explicit learning in relation to AQ scores. Lewicki, P., Hill, T., & Czyzewska, M. (1992).
In summary, this study provided evidence that people are Nonconscious aquisition of information. American
capable of learning contingencies between social cues and Psychologist, 47, 796-801.
use this knowledge to guide their behavior, even in the
absence of explicit knowledge. Furthermore, this effect is Lieberman, M. D. (2000). Intuition: a social cognitive
related to individual differences in autistic traits, with neuroscience approach. Psychological Bulletin, 126(1),
individuals low in autistic traits outperforming individuals 109–137.
high in autistic traits. If the present pattern of results would
be more profound in an ASD sample, then that might help to Senju, A., Southgate, V., White, S., & Frith, U. (2009).
explain their inefficiency in using social cue-outcome Mindblind eyes: An absence of spontaneous Theory of
relationships to modulate their behavior. Mind in Asperger syndrome. Science, 325, 883-885.
Acknowledgments Sodian, B., Schuwerk, T., & Kristen, S. (2015). Implicit and
Spontaneous Theory of Mind Reasoning in Autism
SM and MK were funded by Hull University studentships.
Spectrum Disorders. In M.Fitzgerald (Ed.). Autism
Spectrum Disorder - Recent Advances. InTech.
References
Baron-Cohen, S. (2000). Theory of mind and autism: a Travers, B. G., Powell, P. S., Mussey, J. L., Klinger, L. G.,
fifteen year review. In S. Baron-Cohen, H. Tager- Crisler, M. E., & Klinger, M. R. (2013). Spatial and
Flusberg & D. J. Cohen (Eds.), Understanding other identity cues differentially affect implicit contextual
minds: perspectives from autism and developmental cueing in adolescents and adults with autism spectrum
cognitive neuroscience (2nd ed.). Oxford: Oxford disorder. Journal of Autism and Developmental
University Press. Disorders, 43, 2393–2404.
Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., Weigelt, S., Koldewyn, K., & Kanwisher, N. (2012). Face
& Clubley, E. (2001). The autism-spectrum quotient identity recognition in autism spectrum disorders: a
(AQ): evidence from Asperger syndrome/high- review of behavioral studies. Neuroscience &
functioning autism, males and females, scientists and Biobehavioral Reviews, 36, 1060-1084.
mathematicians. Journal of Autism and Developmental
Disorders, 31, 5–17.
558
Woodbury-Smith, M. R., Robinson, J., Wheelwright, S., &
Baron-Cohen, S. (2005). Screening adults for Asperger
Syndrome using the AQ: a preliminary study of its
diagnostic validity in clinical practice. Journal of Autism
and Developmental Disorders, 35, 331–335.
559