ArticlePDF Available

Abstract and Figures

Ideomotor theory considers bidirectional action-effect associations to be the fundamental building blocks for intentional action. The present study employed a novel pupillometric and oculomotor paradigm to study developmental changes in the role of action-effects in the acquisition of voluntary action. Our findings suggest that both 7- and 12-month-olds (and adults) can use acquired action-effect bindings to predict action outcomes but only 12-month-olds (and adults) showed evidence for employing action-effects to select actions. This dissociation supports the idea that infants acquire action-effect knowledge before they have developed the cognitive machinery necessary to make use of that knowledge to perform intentional actions.
Content may be subject to copyright.
From outcome prediction to action selection: developmental
change in the role of actioneffect bindings
Stephan A. Verschoor,
Michiel Spap
Szilvia Biro
Bernhard Hommel
1. Leiden University Institute for Psychological Research & Leiden Institute for Brain and Cognition, The Netherlands
2. Helsinki Institute for Information Technology, Finland
3. Center for Child and Family Studies, Leiden University, The Netherlands
Ideomotor theory considers bidirectional actioneffect associations to be the fundamental building blocks for intentional action.
The present study employed a novel pupillometric and oculomotor paradigm to study developmental changes in the role of
action-effects in the acquisition of voluntary action. Our findings suggest that both 7- and 12-month-olds (and adults) can use
acquired actioneffect bindings to predict action outcomes but only 12-month-olds (and adults) showed evidence for employing
action-effects to select actions. This dissociation supports the idea that infants acquire actioneffect knowledge before they have
developed the cognitive machinery necessary to make use of that knowledge to perform intentional actions.
Some authors have suggested a rudimentary ability to
represent action goals to be present at birth (e.g.
Meltzoff & Moore, 1997; Rizzolatti & Craighero, 2004;
Rochat, 2001). But where do such representations come
from? Given the worlds complexity and the dramatic
changes the mind and body of infants undergo during
development, it is rather unlikely that they are innate and
permanent (Greenwald, 1970; Harless, 1861; Hommel &
Elsner, 2009; James, 1890; Lotze, 1852). Piaget (1936), in
his influential constructivist approach to cognitive
development, firstly suggested that goals should be
adaptive to the infants changing skills and abilities
and may derive from its own sensorimotor exploration
and experience motivated by a predisposition to adjust
to its environment. Although he states that such explor-
atory actions are integrated with their effects into
schemata necessary for perception and action, he does
not elaborate on the underlying cognitive mechanism
(Piaget, 1954).
Exactly such a mechanism was proposed in the late
nineteenth century, though largely neglected in psychol-
ogy for a century. Following the lead of Lotze (1852) and
Harless (1861), James (1890) suggested a cognitive
mechanism that does what Piaget (1936, [1963]) pro-
posed; it provides actors with action goals that are
rooted in their own sensorimotor experience. In his
ideomotor theory James stated that all actions are
necessarily involuntary when being carried out for the
first time. Indeed, if one defines action as goal-directed
movement, it presupposes some sort of anticipation of its
effect. This again implies knowledge on actioneffect
relationships, which needs to be acquired before the
action can be carried out in order toproduce the
outcome intentionally. Ideomotor theory suggests that
such knowledge is acquired on the fly: whenever people
move, they automatically and unintentionally create
bidirectional associations between the perceived effects
and the motor pattern producing them. This association
brings the movement under voluntary control: Once
acquired, the agent can now activate the motor pattern
producing a movement by thinking of(i.e. endoge-
nously activating the representation of) a perceptual
effect. Indeed, infants start to motor babble (i.e. produce
random movements) in utero (cf. Meltzoff & Moore,
Address for correspondence: Stephan A. Verschoor, Leiden University, Department of Psychology, Cognitive Psychology Unit, Wassenaarseweg 52,
2333 AK Leiden, The Netherlands; e-mail:
©2013 John Wiley & Sons Ltd
Developmental Science 16:6 (2013), pp 801–814 DOI: 10.1111/desc.12085
1997) which could explain the possible presence of goal
representations at birth and they are consistently
exploring their environment. This provides ample oppor-
tunity to acquire movement/actioneffect associations
and thus a steadily increasing pool of possible action
goals. Thus, James considered bidirectional movement/
actioneffect associations the fundamental building
blocks of intentional action and provides a mechanism
that could allow the emergence of goal-directed action in
Ideomotor theory was revived and refined by Green-
wald (1970), Prinz (1990, 1997), and Hommel (1996;
Elsner & Hommel, 2001) and is now part of a broader
theoretical movement stressing the interplay between
perception and action (Hommel, M
usseler, Aschersleben
& Prinz, 2001; Meltzoff, 2006; Meltzoff & Prinz, 2002).
This motivated numerous demonstrations of bidirec-
tional actioneffect acquisition in humans ranging from
4-year-olds (Eenshuistra, Weidema & Hommel, 2004;
Kray, Eenshuistra, Kerstner, Weidema & Hommel, 2006)
to adults (Elsner & Hommel, 2001). Actioneffect
acquisition was found for a wide range of actions and
effects (for a review see Hommel & Elsner, 2009),
suggesting a general actioneffect integration mecha-
nism. In addition, actioneffect acquisition has been
found after just one trial (Dutzi & Hommel, 2009),
suggesting that the mechanism is fast-acting and
implicit. Actioneffect acquisition is modulated by
the same factors that influence instrumental learning
(e.g. temporal contiguity and contingency of movement
and effect; Elsner & Hommel, 2004) and does not
depend on voluntary attention (Dutzi & Hommel, 2009;
Elsner & Hommel, 2001; Band, van Steenbergen,
Ridderinkhof, Falkenstein & Hommel, 2009). Together
with the fact that it was also found in animals (see Elsner
& Hommel, 2001), this suggests that actioneffect
integration is a fairly low-level and automatic process
(Elsner & Hommel, 2004).
If bidirectional actioneffect associations are indeed
the fundamental building blocks for intentional action,
the system that generates these associations should be
operative early in life. Especially since infants show
evidence of goal-directed behavior from a very early age
on: depending on the definition, goal-directed behavior
is thought to start somewhere between birth (Meltzoff &
Moore, 1997; von Hofsten, 2004) and about 9 months of
age (Hauf, 2007; Piaget, 1936, [1963]). Actioneffect
knowledge has been implicated to be operational in
higher order cognitive functions such as action under-
standing in 7-month-olds (e.g. Biro & Leslie, 2007; for a
review, see Hauf, 2007; Schneider, Eschman & Zuccol-
otto, 2002) and imitation in 9-month-olds (Hauf &
Aschersleben, 2008; Klein, Hauf & Aschersleben, 2006;
for a review, see Elsner, 2007; Meltzoff, 2006). Even
though these findings do not provide direct evidence for
bidirectional actioneffect acquisition, theories that
emphasize similar representational formats for first-
person experience and observed action (e.g. Fabbri-
Destro & Rizzolatti, 2008; Hommel et al., 2001; Meltz-
off, 2006; Tomasello, 1999), and conceptualize action
understanding as inverse planning (Meltzoff, 2006;
Baker, Saxe & Tenenbaum, 2009) consider them corrob-
orative. Other corroborating evidence was found in
studies that show very young infants to be sensitive to
actioneffect contingencies. For instance, newborns
actively adjust their sucking rate in response to their
mothers voice as ongoing conditional feedback (DeC-
asper & Fifer, 1980) and 2-month-olds pursue interesting
action effects by intentionally varying their sucking rate
(Rochat & Striano, 1999) or varying gaze direction
(Watson, 1967; for a review, see Gergely & Watson,
1999). Another line of research by Carolyn Rovee-Collier
shows that action effects aid memory retrieval for actions
from 2 months of age (Rovee & Rovee, 1969; for a review,
see Rovee-Collier, 1999). Telling as these studies may be
(they show that action contingent effects play an
important role in infant behavior and memory), they
were not designed to directly assess the bidirectionality
of actioneffect associations and their use for action
planning and may thus confound actual actioneffect
learning with simple operant conditioning. Nonetheless,
these findings may reflect a developing ability for
learning actioneffect contingencies.
The first study to find direct evidence for bidirectional
actioneffect associations in infants was undertaken by
Verschoor, Weidema, Biro & Hommel (2010) with 9-, 12-
and 18-month-old participants. A simplified version of
the free-choice design of Elsner and Hommel (2001) was
used. In the acquisition there were two types of trial: One
in which infants were permitted to touch a response key,
which resulted in a particular audiovisual effect, and
another in which they were prevented from touching the
key while another audiovisual effect was presented.
When the two effects were replayed in the test phase, all
infants were faster to touch the key again after the
previously self-produced effect than after the action-
independent effect. Moreover, the 18-month-olds
responded more often following the self-produced effect.
These results were taken as evidence for bidirectional
actioneffect learning, since the self-produced effect
activated the response that previously caused the effect.
Although this study shows direct evidence for action
effect learning in infants, the paradigm had some
drawbacks compared to the Elsner and Hommel (2001)
paradigm. The paradigm had only 10 acquisition trials
and two test trials which resulted in slightly noisy data,
©2013 John Wiley & Sons Ltd
802 Stephan A. Verschoor et al.
and due to the nature of the task the paradigm was
unsuited for comparison with adults. Thus it remains to
be seen whether bidirectional actioneffect learning
works similarly in infants and adults. Furthermore,
initial piloting showed that, due to difficulties with the
button pushing action, it was unsuitable for infants
younger than 9 months (Verschoor et al., 2010).
Overcoming these limitations calls for a more natural
type of action that is well established in very young
infants. Eye movements seem to be the ideal candidate:
infants have been reported to actively and accurately
control their eye movements from at least 4 months of
age (Scerif, Karmiloff-Smith, Campos, Elsabbagh,
Driver & Cornish, 2005) and, given that infants actively
control their gaze to gather information (Gredeb
ack &
Melinder, 2010; Falck-Ytter, Gredeb
ack & von Hofsten,
2006), to direct or follow attention (Perra & Gattis,
2010), and to engage in social behaviors (Senju & Csibra,
2008; Johnson, Ok & Luo, 2007), eye movements can be
considered truly goal-directed actions. Moreover, a study
by Herwig and Horstmann (2011) demonstrated
saccade-effect learning in adults in a paradigm concep-
tually very close to that of Elsner and Hommel (2001),
which indicates that actioneffect integration generalizes
to oculomotor action.
In the present study, participants made eye movements
towards visual stimuli appearing at the left or right of a
display, the two directions produced different auditory
effects analogous to the Elsner and Hommel (2001)
paradigm. In the subsequent test phase, participants
were presented with the effect tones and could freely
choose to make a saccade to the left or right of two
simultaneously presented visual stimuli. We expected
that the tone would prime the saccade that it was
produced by in the acquisition phase, so that this saccade
would be chosen more frequently and/or initiated more
quickly. This design allowed us to test both infants
younger than 9 months of age and adults, and to run
considerably more trials.
We tested 7- and 12-month-olds and adults. Seven-
month-olds were chosen because this group is known to
show understanding of goal-directed actions (e.g.
Woodward, 1998; Csibra, 2008; Verschoor & Biro,
2012). Since some actionperception theorists (Hommel
et al., 2001; Meltzoff, 2006; Meltzoff & Prinz, 2002;
Woodward, 2009; Rizolatti & Craighero, 2004) stress
that the same representational format is used for
observed and self-initiated action, even 7-month-olds
should be able to pick up actioneffect associations. A
first study that shows actioneffect acquisition in
infants younger than 9 months was recently published
by Paulus, Hinnius, Elk and Beckering (2012). They
found electrophysiological evidence indicating that
infants at 8 months of age show stronger motor
resonance when listening to previously self-produced
action-related sounds than when hearing other sounds.
It remains to be seen whether the underlying action
effect associations are bidirectional in the sense that
they can be reversed to generate overt action. Dissoci-
ations between acquired action knowledge and the use
of such knowledge are by no means new; for instance
Keen (2005) found a similar dissociation in her work on
reaching and looking for occluded objects which shows
that infantslooking behavior exhibits knowledge while
actions do not conform to this knowledge. In a more
general sense, such dissociations are apparent in look-
ing-time studies wherein infants are reported to possess
knowledge on actions they cannot perform themselves
(e.g. Verschoor & Biro, 2012; Csibra, Gergely, Biro &
os, 1999).
To maximize the chance of finding developmental
changes in this study, we included the older of the two
youngest groups in Verschoor et al. (2010) since no
differences were found between 9- and 12-month-olds in
that study. They were also included to replicate Ver-
schoor et al.s (2010) finding that infants at this age
show bidirectional actioneffect learning and to contrast
this ability with the suggested inability of the 7-month-
olds to initiate true intentional action (Hauf, 2007). We
included adults to confirm that the same principles of
bidirectional actioneffect learning can be shown in
12-month-old infants and adults.
Using an eye-tracker enabled us to measure not only
the choice of actions and the time to initiate them
(reaction time), but pupil size as well. This is a relatively
new measure in developmental studies (e.g. Falck-Ytter,
2008; Jackson & Sirois, 2009; Gredeb
ack & Melinder,
2010; for a review, see Laeng, Sirois & Gredeb
ack, 2012)
but has been extensively used in psychological research
on adults since the early 20th century (Hess, 1975).
Pupils have the interesting characteristic of reacting not
only to luminance, they reliably dilate with superimposed
sympathetic activation (Libby, Lacey & Lacey, 1973;
Beatty & Lucero-Wagoner, 2000). Although these dila-
tions are not directly causally related to central process-
ing load, they empirically reflect variations in central
processing load with extraordinary precision (Beatty &
Lucero-Wagoner, 2000). Task-Evoked Pupillary
Responses (TEPRs) can indicate motivational phenom-
ena such as arousal (Bradley, Miccoli, Escrig & Lang,
2008; Laeng & Falkenberg, 2007), attention allocation
(e.g. Hess & Polt, 1960), cognitive load (Kahneman &
Beatty, 1966), and mental effort (Kahneman, 1973; Hess
& Polt, 1964). TEPRs are pre-conscious and mediated by
the locus coeruleus (Laeng et al., 2012). Whatever the
exact interpretation of this measure, using it enables us
©2013 John Wiley & Sons Ltd
Spontaneous actioneffect binding in infants and adults 803
to contrast acquisition contingent vs. non-contingent
responses: whether TEPRs are taken to reflect differ-
ences in sympathetic activation in general or arousal,
attention allocation, cognitive load or mental effort, all
interpretations suggest that dilations should be larger for
incongruent responses.
We assumed that in our task two types of process
might play a role and affect TEPRs. For one, even in
studies that were successful in demonstrating that action
effects bias the choice of actions (e.g. Elsner & Hommel,
2001), participants did not always choose the action that
was previously associated with the present trigger stim-
ulus. To some degree, this might be due to random noise
but it may also reflect a strategy to create some
variability and exert active control. Exerting control
calls for the investment of more cognitive resources,
which would suggest that selecting and/or performing an
action that is not associated with, and thus primed by,
the trigger stimulus is more effortful. If so, one would
expect that choices of tone-incongruent actions (i.e. of an
action different from the one that previously produced
the present trigger stimulus) are accompanied by
(greater) pupil dilation.
For another, there is evidence that actioneffect
associations not only affect the choice of actions but
also their evaluation. In the study of Band et al. (2009),
participants performed a probabilistic learning task, in
which key-presses triggered tones of a particular pitch in
80% of the trials and of another pitch in the remaining
trials. The presentation of a less frequent action effect
generated an electrophysiological component that is
known as feedback-related negativity (Miltner, Braun
& Coles, 1997), which is commonly observed when
negative feedback is presented. This suggests that
actioneffect associations are used to generate particular
expectations about effects given the execution of a
particular action. In infant studies, TEPRs have been
used as an index of the violation of expectations (Jackson
& Sirois, 2009; Gredeb
ack & Melinder, 2010). Accord-
ingly, it is possible that carrying out a tone-incongruent
action results in (more) pupil dilation reflecting the
violation of a tone-induced expectation regarding the
action outcome (i.e. the location of the action end point
and/or the targeted stimulus).
Although both processes would predict greater pupil
dilations in incongruent responses, we considered that
these might be distinguished in terms of their temporal
dynamics: whereas a choice-related process would be
likely to affect pupil responses briefly before or after
response execution, an expectation/evaluation-related
process would be more likely to affect pupil responses
after response execution (Band et al., 2009).
Two groups of infants were tested: 15 7-month-olds
(mean: 7.15 months, SD =.21, SE = .05, 8 female) and
20 12-month-olds (mean: 12.11 months, SD = .26,
SE = .05, 9 female). The infants were recruited through
direct mail. Informed consent and answers to a ques-
tionnaire regarding their general health were obtained
from all caretakers. The infants were all healthy full-term
and without any pre- or perinatal complications. In
addition, 24 undergraduate students (mean age: 23.8
years, SD = 2.47, SE = .50, 14 female) participated in
exchange for course credits. All reported being healthy
and having normal or corrected-to-normal vision and
hearing. Two additional 12-month-olds and one
adult were excluded due to technical error, and two
more 7- and four 12-month-olds were excluded due to
fussiness. In addition, two 7- and two 12-month-olds
were excluded for not meeting the criterion for the
minimal amount of test trials.
Test environment and apparatus
During the experiment participants sat in a specially
designed stimulus-poor curtained booth (infants on the
lap of their caretaker) in front of the monitor/eye-tracker
apparatus. The distance between eyes and apparatus was
approximately 70 cm (the screens viewing angle was
34.1°by 21.8°). Participant behavior was monitored
online by means of a camera located above the appara-
tus. The experimenter controlled the experiment from a
separate control room. A 17-inch TFT screen, equipped
with an integrated Tobii T120 eye-tracker operating at 60
Hz, was used for visual and auditory data presentation,
and for data collection. The Tobii T120 has an average
accuracy of .5 visual degrees and allows for a reasonable
amount of free head movement by the subject
(30 922 930 cm). It recorded gaze direction and pupil
size. Stimulus presentation was controlled by a PC
running E-prime
software (Schneider, Eschman &
Zuccolotto, 2002).
Infants were tested at a time of day when they were likely
to be alert and in a good mood. Caretakers and
participants were given instructions prior to the exper-
iment. Adults were given no instruction with regard to
the task. The caretakers were instructed not to move
after calibration and gently fixate the infant against their
©2013 John Wiley & Sons Ltd
804 Stephan A. Verschoor et al.
tummy to maintain the eye-tracker alignment and to
entertain the infant during the 1-min interruption
between calibration and the experiment. The eye-tracker
was calibrated using a 9-point calibration consisting of
an animated dancing infant accompanied by music. The
calibration was accepted with a minimum of eight points
acquired successfully. The experimenter could play an
attention-grabbing sound during the experiment to
regain attention. If the attention grabbing sound did
not work, caretakers were encouraged to direct the
infants attention to the middle of the screen by pointing
to it. Lighting conditions were kept constant during
testing and across subjects. Furthermore, the luminance
levels were controlled for by presenting the stimuli in a
random fashion. After completion of the experiment,
further information on the rationale was provided.
Acquisition Phase
The experiment began with an acquisition-phase of 48
trials (see Figure 1). The background color of the screen
was gray. An acquisition trial started with a brightly
colored dot with a superimposed line drawing (4.3°by
4.3°) being displayed at the center of the screen (Snod-
grass & Vanderwart, 1980). The dot served as start signal
and fixation mark. To keep the display interesting to the
subjects, the color of the dot changed randomly from
trial to trial (selected from eight bright colors) and the
superimposed line drawing was randomly selected (with-
out replacement) from a selection of 50 drawings. The
dot disappeared after the subject fixated properly for an
interval that varied from trial to trial (so to remove any
bias or habituation that might be caused by fixed
intervals between trials) between 150 and 350 ms.
Immediately after the dot disappeared, photographs of
two different faces (randomly selected without replace-
ment from 100 grayscale pictures from the Nottingham
scansemotional faces database,
uk, displaying emotionally neutral faces of 50 men and
50 women from a frontal perspective) appeared left and
right from the dot. Faces were chosen to elicit sponta-
neous saccades as they are known to attract infants
attention (Goren, Sarty & Wu, 1975; Johnson, Dziura-
wiec, Ellis & Morton, 1991). The 5.3°by 5.3°pictures
appeared at 9.7°, center to center, to the left and right of
the center of the screen. To avoid perseverance to either
left or right across acquisition trials, the images imme-
diately started to pulsate. One of the faces started
shrinking to 4.1°while the other started growing to 6.5°
(which picture started shrinking was randomized); one
cycle from intermediate size to small, to intermediate, to
large and back to intermediate, took 2 s.
The faces evoked spontaneous saccades and thus
served as response locations. When a saccade towards
one of the two face locations was detected, the face at the
other location disappeared. The targeted face stopped
pulsating and, depending on the targeted side, one of two
distinct 200-ms effect sounds (tringor piew) was
presented. Each effect-sound was consistently designated
to either the left or the right response area (RA) during
the entire acquisition phase (the mapping was balanced
across participants); RAs were defined as the maximum
size of the pulsating images: 6.5°by 6.5°. A saccadic
response was defined as an eye movement to the left or
Acquisition trial Test trial
Figure 1 Acquisition trial T1: Each trial starts with an intertrial interval of 500 ms. T2: A fixation dot is displayed at screen center.
T3: After succesful fixation, faces appear at either side of the screen where they started to pulsate. T4: Depending on the saccade
target, the face at the other side disappears and an effect sound is played for 200 ms. Test trial T1: Each trial starts with an intertrial
interval of 500 ms. T2: A fixation dot is displayed at screen center. After succesful fixation one of the previous action effects is
played. T3: The dot disappears and thereafter the same face appears on both sides. T4: The participant freely chooses where to
©2013 John Wiley & Sons Ltd
Spontaneous actioneffect binding in infants and adults 805
right response area (minimal amplitude 4.3°). Reaction
Times (RTs) were defined as the time it took from the
disappearance of the central dot to the time one of the
RAs was entered. The maximum allowable RT was
defined as 2000 ms; when subjects did not respond within
this time the same trial was repeated. After each trial, an
inter-trial interval of 500 ms was presented. If during the
acquisition phase the subject showed declining attention
to the screen or was otherwise distracted, the acquisition
phase could be shortened (minimum amount of acqui-
sition trials was set at 30).
Test phase
After acquisition, the test phase followed directly (32
trials) (see Figure 1). A test trial started with a similar
dot with superimposed line drawing as in the acquisition
phase, again serving as a start and fixation stimulus.
However, after the subjects fixated on the dot (fixation
time again varied randomly between 150 and 350 ms),
the dot stayed on the screen for another 200 ms during
which one of the effect sounds that was previously
triggered by one of the two eye movements, was played
after which the dot immediately disappeared. Then two
identical 5.3°by 5.3°images of the same face (again
randomly selected without replacement from the Not-
tingham scans emotional faces database) appeared 9.7°
to the left and right of the center of the screen. The
images were identical to avoid any influence on the
subjects gaze preference. To further minimize influence
on preference, the faces now pulsated in synchrony; they
both either started growing or shrinking (randomized
and with the same motion parameters as in the acqui-
sition). Again, this was expected to evoke a spontaneous
saccade and the question of interest was whether the
direction of this saccade would be biased by the tone.
Saccades towards the location that previously produced
the tone were considered congruent, while saccades
towards the alternative location were considered incon-
gruent. The minimum number of test trials to enter
analysis was 21. Except for the absence of the effect after
the saccade, the remaining procedure was as in the
acquisition phase.
After the experiment, adults were asked if they noticed
any regularity in the sound mapping in the experiment. If
so, they were asked what it was (e.g. When I looked to
the right I heard sound X, when looking to left I heard
sound Y). Then, all subjects were asked whether they
noticed that there were two parts to the experiment. If
they did notice, they were asked more specifically if they
noticed any regularity in the sounds during the first
(acquisition) phase; if not, they were scored as unaware.
If they noticed two phases but no regularity in the sound
mapping, they were asked specifically if they had noticed
that during the first phase there was a mapping between
sounds and direction of looking. If they did they were
considered aware, otherwise unaware.
Data acquisition
was used to collect RTs during acquisition and
test phases, the number of left and right responses during
acquisition, and the number of congruent and incongru-
ent responses during test. Furthermore, the E-gaze data
files produced by E-prime
were imported into BrainVi-
sion Analyzer software (Version 1.05, BrainProducts,
Germany) to analyze gaze position and pupillary data.
First, pupil sizes of both eyes were averaged to create
more stable data. Artifacts and blinks as detected by the
eye-tracker were corrected by using a linear interpolation
algorithm. After this a 10 Hz low-pass filter was used,
commonly used for pupil data (e.g. Hupe, Lamirel &
Lorenceau, 2009). To ensure that there were no erroneous
pupil data we then rejected artifacts using the parameters
of a minimal pupil size of 1 mm and a maximum of 5 mm;
furthermore, the maximum allowed change in pupil size
was defined as .03 mm in 17 ms.
Given that the acquisition of actioneffect associa-
tions is sensitive to the same factors as stimulusresponse
learning (Elsner & Hommel, 2004), the bias to respond
in either direction during acquisition was calculated
(Acquisition Bias,AB=the number of leftward sac-
cades minus the number of rightward saccades). As the
size of this bias represents the degree to which partic-
ipants were selectively exposed to one of the action
effects, the AB variable was used as covariate in the
analyses when appropriate.
Acquisition phase
RT and response frequency
All ANOVAs were performed with age group as a
between-subjects factor. ANOVAs on the percentage of
left responses and number of completed acquisition trials
showed no effects, ps>.3 (see Table 1).
Another ANOVA, on mean RT, revealed a significant
effect of age group, F(2, 56) =55.30, p<.001,
©2013 John Wiley & Sons Ltd
806 Stephan A. Verschoor et al.
p=.66: post-hoc Tukey HSD comparisons revealed
that adults responded significantly faster than infants,
ps<.001, (see Table 1).
A repeated-measures ANOVA on RTs with right-vs.-
left as within-subjects factor showed no effects, ps>.8.
Only four out of 24 adult participants reported being
aware of the actioneffect mapping in the acquisition
Test phase
Response frequency
Again all ANOVAs were performed with age group as a
between-subjects factor. An ANOVA on the number of
completed test trials showed a significant effect of age
group, F(2, 56) =7.71, p=.001, g
p=.22. Post-hoc
Tukey HSD comparisons revealed that adults and
7-month-olds completed more test trials than 12-
month-olds (mean adults =32, 7-month-olds =31.6,
12-month-olds =29.6, ps<.02). This is probably due
to increased agility and fussiness in the 12-month-olds
and increased motivation in the adults (see Table 2).
A repeated-measures ANOVA on percentage of
responses with left vs. right as within-subjects factor
revealed that, overall, participants showed no tendency
to saccade more often to either side, p>.05; however, it
did show an interaction with age group, F(2, 52) =4.11,
p=.02, g
p=.14. Separate comparison showed that
while adults made more leftward saccades, F(1,
23) =4.23, p=.05, g
p=.16 (18 left vs. 14 right), 12-
month-olds did the opposite, F(1, 18) =5.94, p=.03,
p=.25 (12 left vs. 18 right) (see Table 2).
More importantly for our purposes, a repeated-mea-
sures ANOVA on response frequency with congruency as
within-subjects factor showed that the percentage of
acquisition-congruent vs. incongruent responses did not
differ and congruency did not interact with age group,
In addition, we performed a median split on RTs on
each subject, classifying trials as either fast or slow, and
calculated the percentage of fast congruent responses vs.
the percentage of slow congruent responses. We then
performed a repeated-measures ANOVA on the percent-
age of congruent responses, with fast vs. slow as within-
subjects factor. We found no main effect of fast vs. slow
on percentage (F>1), but we did find a significant
interaction of fast vs. slow with age group, F(1, 56)
=3.58, p=.03, g
p=.11. We then tested the age groups
separately, showing that adults had a higher percentage
of congruent responses in their fast responses compared
to their slow responses (54% vs. 45%), F(1, 23) =11.5,
p=.003, g
p=.33, while the infants showed no such
effect (see Table 2).
We also performed a repeated-measures ANOVA on
percentage of congruent reactions with Time (dividing
the responses into three bins; trial 110, 1121 and 22
32) as within-subjects factor which did not yield any
effects, ps>.2.
Reaction times
Again all ANOVAs were performed with age group as a
between-subjects factor. Since the test phase was self-
paced we also performed an ANOVA on inter-trial
interval (ITI) and found a significant effect, F(2,
56) =36.53, p<.001, g
p=.57. Post-hoc Tukey HSD
comparisons revealed that adults responded significantly
faster than infants, ps<.001 (see Table 3).
As in the acquisition phase, an ANOVA on RTs
showed that adults responded faster than the two infant
age groups, F(2, 56) =89.07, p<.001, g
p=.76; all
HSD ps<.001.
Table 1 Mean scores of acquisition phase (standard
deviation in parentheses)
Age group scores
Number of acquisition
Percentage of
left responses RT in ms
7-month-olds 47.1 (3.36) 60.3 (37) 441 (49)
12-month-olds 47.15 (2.30) 48.9 (30) 440 (57)
Adults 48 (0) 52.4 (11) 293 (50)
Table 2 Mean frequency scores of test phase (standard deviation in parentheses)
Age group
test trials
of left
Percentage of
in fast reactions
Percentage of
responses in slow
7-month-olds 31.6 (1.55) 61.5 (34.9) 48.8 (6.9) 46.3 (12.4) 51.3 (12.6)
12-month-olds 29.6 (3.44) 38.1 (28.7) 49.8 (6.6) 51.8 (6.7) 47.6 (13.3)
Adults 32 (0) 56.5 (15.5) 49.7 (10.9) 54.4 (13.5) 45.1 (12.4)
©2013 John Wiley & Sons Ltd
Spontaneous actioneffect binding in infants and adults 807
Another repeated-measures ANOVA on RTs with left
vs. right as within-subjects factor showed no effect,
p>.1 (see Table 3).
More importantly for our purposes, we performed a
repeated-measures ANOVA on RTs with congruent vs.
incongruent as within-subjects factor, using AB as a
covariate which showed that acquisition-congruent
responses were initiated 14 ms faster than incongruent
responses, F(1, 55) =4.20, p=.05, g
p=.07, and this
effect interacted with age group, F(2, 55) =4.38, p=.02,
p=.14. Separate comparisons showed that the con-
gruency effect was significant in adults, F(1, 22) =10.60,
p=.004, g
p=.33, and 12-month-olds, F(1, 18) =8.51,
p=.009, g
p=.32, but not in 7-month-olds, F(1, 13)
=2.51, p=.14 (see Figure 2). Additional non-paramet-
ric analysis in the 7-month-olds also failed to show an
effect of congruency on RTs in this group (see Table 3).
We also performed a repeated-measures ANOVA on
RTs with Time (dividing the responses in three bins; trial
110, 1121 and 2232) and congruence as within-
subject factors using AB as a covariate. We found an
overall tendency regarding the main factor of Time,
(F(2, 102) =2.64, p=.08, g
p=.05) with slower
responses as the test progressed which interacted with
age group, F(4, 102) =3.72, p=.01, g
p=.13, and
further separate testing revealed that only the 12-month-
olds showed a significant slowing as the test progressed,
F(2, 32) =7.02, p=.003, g
p=.31. No further interac-
tions with Time were found, ps>.3. The main effect of
congruency on RTs was significant, F(1, 51) =10.85,
p=.002, g
p=.18, and showed that congruent
responses were initiated 23 ms faster. The interaction
of congruency with age group on RTs also reached
significance, F(2, 51) =5.70, p=.006, g
p=.18. Sepa-
rate testing for the age groups revealed that the effect was
significant in adults, F(1, 21) =7.45, p=.013,
p=.26, and in the 12-month-olds, F(1, 16) =13.90,
p=.002, g
p=.47, but not in 7-month-olds, F(1, 12)
=.95, p=.35 (see Table 3).
Pupil dilation
TEPRs were sorted according to congruency of the
response and the stimulus- and response-locked time
functions were averaged. Segments were created,
depending on the analysis, from 2000 ms before the
presentation of the sound or RT to 8000 ms after while
allowing for overlapping segments. Following the
method used by Bradley et al. (2008), pupil diameter
measurement began after the initial pupil reflex caused
by the fixation stimulus. Visual inspection showed the
light reflex to end around 500 ms after effect presenta-
tion (see Figure 3). To accommodate for the variable
RTs across age groups and conditions, we considered
both stimulus-locked and response-locked TEPRs. TEP-
Rs were calculated as the percentage of dilation relative
to the baseline to make the data more comparable across
age groups.
First we analyzed whether the percentage of trials
rejected due to erroneous data points differed across age
groups. An ANOVA on the percentage of kept trials
yielded a reliable main effect of age group, F(2, 56)
=4.30, p=.02, g
p=.13 (average percentage of kept
trials: 7-month-olds 92%, 12-month-olds 92%, adults
99%). Post-hoc Tukey HSD comparisons showed that in
adults significantly fewer segments were rejected than in
the 12-month-olds, p=.03, and the same tendency was
visible in the comparison of adults and 7-month-olds,
Table 3 Mean RT scores of test phase (standard deviation in parentheses)
Age group scores in ms ITI RT RT Congruent RT Incongruent RT trial 110 RT trial 1121 RT trial 2232
7-month-olds 2761 (842) 432 (71.9) 437.6 (77.9) 425.0 (78.0) 410.1 (71.1) 427.8 (74.7) 426.9 (109.5)
12-month-olds 2807 (902) 449 (62.3) 427.4 (60.5) 470.6 (83.4) 420.9 (85.7) 431.2 (53.8) 493.6 (104.7)
Adults 1218 (255) 237.6 (42.9) 230.4 (47.4) 244.0 (39.2) 245.9 (47.6) 234.8 (49.4) 229.2 (49.2)
7-month olds 12-month olds Adults
Reaction times in ms
Congruent Incongruent
Figure 2 Mean reaction times (+SE) for adults (N =24)
7-month-olds (N =17) and 12-month-olds (N =22) in
congruent and incongruent test trials.
©2013 John Wiley & Sons Ltd
808 Stephan A. Verschoor et al.
p=.08, an unsurprising observation given the differ-
ences in attentional resources between infants and adults.
The stimulus-locked analysis of TEPRs in congruent
and incongruent trials used a 500 ms pre-effect baseline
(Beatty & Lucero-Wagoner, 2000). A repeated-measures
ANOVA on pupil dilations with congruency as within-
subjects factor revealed no a priori effects of congruence
on baselines (500 to 0 ms), ps>.10. TEPRs start from
200 to 300 ms after stimulus onset and peak around 1200
ms post-stimulus (Beatty & Lucero-Wagoner, 2000) in
the range of 500 ms to 2000 ms (Beatty, 1982). We
therefore calculated the mean TEPRs for congruent and
incongruent responses as the mean percentage of change
from baseline to 5002000 ms post effect onset.
A repeated-measures ANOVA revealed that, overall,
participants showed larger relative dilations in incon-
gruent trials, F(1, 56) =6.80, p=.01, g
p=.11, and
this effect was not modulated by age group, p>.10 (see
Figure 3). To take a closer look into developmental
changes, we then analyzed the infant data separately. On
average, infants showed larger relative dilations in
incongruent trials, F(1, 33) =6.78, p=.02, g
and this effect was not modulated by age group, p>.10.
Of particular importance (given the reaction time
results), the congruency effect remained significant when
the 7-month-olds were tested separately, F(1, 14) =12.0,
p=.004, g
p=.46 .
For the response-locked analysis, we calculated the
percentage of dilation from a 700-ms time window from
saccade onset on, to a 200-ms pre-response baseline. A
repeated-measures ANOVA showed no a priori effects of
congruence on baselines (200 to 0 ms), ps>.10. The
analysis of these data yielded a significantly larger
relative dilation in incongruent than congruent trials,
F(1, 56) =7.82, p=.007, g
p=.12, while the interac-
tion with age group was not significant, p>.10 (see
Figure 4). A separate analysis of the infant data showed
a main effect for congruency, F(1, 33) =8.41, p=.007,
p=.20, that was not modulated by age group,
Another version of this analysis with a 1000-ms pre-
response baseline produced a different pattern (a
repeated-measures ANOVA revealed no a priori effects
of congruence on baselines, ps>.10): a congruency
effect, F(1, 56) =10.19, p=.001, g
p=.17 (see Fig-
ure 4), but also an interaction of congruency with age
group, F(2, 56) =3.99, p=.02, g
p=.13. Separate
testing showed that the effect was only reliable in the
7-month-olds, F(1, 14) =10.59, p=.006, g
while the other two groups did not reach significance.
The aim of the current study was to directly compare
actioneffect learning in infants and adults using a novel
paradigm that relies on oculomotor actions that occur
spontaneously and do not require verbal instruction. We
succeeded in developing an eye-tracking paradigm that
was equally suitable for both very young infants and
adults. Moreover, the paradigm allowed for concurrently
investigating the impact of actioneffect learning on
biases in, and the efficiency of, action selection as
measured by response choice and RT, respectively, and
on action effort and/or monitoring, as indicated by pupil
As expected from ideomotor theory (James, 1890;
Hommel et al., 2001), adults and 12-month-olds were
faster in carrying out responses that were congruent with
the present trigger tone (i.e. responses that produced this
tone in the acquisition phase) than incongruent
responses. The only difference between congruent and
incongruent responses was their past relationship with
the tones, which indicates that the congruency effect
reflects associative knowledge acquired during the
–500 0 500 1000 1500 2000
Time in ms
Pupil size in mm
Pupil Congruent
Pupil Incongruent
Figure 3 Mean relative pupil sizes for congruent and
incongruent responses, stimulus-locked.
–1000 –500 0 500 1000 1500
Time in ms
Pupil size in mm
Pupil Congruent
Pupil Incongruent
Figure 4 Mean relative pupil sizes for congruent and
incongruent responses, response-locked.
©2013 John Wiley & Sons Ltd
Spontaneous actioneffect binding in infants and adults 809
acquisition phase. Moreover, the fact that the tones now
primed the response they previously had followed
suggests that the underlying association was bidirec-
tional in nature. Both observations are consistent with
ideomotor theory and fit well with the observations of
Herwig and Horstmann (2010), who reported oculomo-
tor actioneffect learning in adults. Interestingly, these
authors used visual action effects while the present study
employed auditory effects. This confirms that the mech-
anism underlying actioneffect learning is general and is
not bound to a particular modality, as long as the effects
are contingent on, and temporally close to, the corre-
sponding actions (Elsner & Hommel, 2004). We also
found that most adults were unaware of the saccade
effect mapping, which is in line with the idea that action
effect acquisition is a low-level, fast and automatic
process that does not require attention. It seems reason-
able to assume that the same holds for the infants.
The present findings fit with observations from
manual actions in a developmental study of action
effect acquisition by Verschoor et al. (2010). In this
study, reliable RT effects were found in 9-, 12- and 18-
month-olds, indicating actioneffect acquisition in these
age groups. We obtained a similar RT effect in 12- but
not 7-month-olds. We take this to imply that, although
ongoing contingent actioneffects can influence behavior
and memory at this age (Gergely & Watson, 1999; Rovee-
Collier, 1999), actioneffect associations cannot yet be
reversed to play an active role in prospective action
control. This is in line with the dissociation between
acquired action knowledge and the use of such knowl-
edge found by Keen (2005) and Sommerville, Woodward
and Needham (2005). It also fits with similar dissocia-
tions in looking-time studies wherein infants are reported
to possess knowledge of actions they cannot perform
themselves (e.g. Verschoor & Biro, 2012; Csibra et al.,
Although we found an effect in the adults indicating
that fast responses were more likely to be acquisition-
congruent than slow responses (this can be taken as
further evidence that actioneffect learning relies on a
fast and automatic mechanism, at least in adults),
congruency effects were restricted to RTs and did not
affect response choice. One might assume that the lack of
frequency effects suggests different developmental path-
ways with respect to manual and oculomotor actions.
There are several arguments against this interpretation.
For one, although manual free-choice studies have shown
that presenting an action effect can bias response choice
towards the response that had previously produced that
effect in adults (e.g. Elsner & Hommel, 2001; Eenshu-
istra et al., 2004; Kray et al., 2006), even in free-choice
studies that did find a reliable effect on response
frequency, frequency turned out to be less sensitive to
actioneffect learning than reaction time. For instance,
Verschoor et al. (2010) obtained a congruency effect on
response frequency in 18-month-olds, but not in younger
infants, while congruency affected reaction time in 12-
and 9-month-olds as well. Since Verschoor et al. (2010)
used only very few test trials, one might suggest that in
their study extinction, which younger infants are more
susceptible to (e.g. Hartshorn, Rovee-Collier, Gerhard-
stein, Bhatt, Wondoloski, Klein, Gilch, Wurzel & Cam-
pos-de-Carvalho, 1997), could not have played a major
role. In the present study the test phase contained
considerably more test trials (which were necessary to get
sufficiently clean pupil dilation data). Our paradigm thus
provided more opportunity for extinction since action
effects were no longer presented during test trials.
However, we tested whether the effect of congruency
on RTs and response frequency declined over time and
found no such effect. Even though actioneffect learning
can be demonstrated under extinction conditions in
principle, extinction does make the effect weaker (Elsner
& Hommel, 2001) and it may have weakened it enough
to selectively annihilate the frequency effect altogether.
Moreover, Herwig and Horstmann (2011) showed under
extinction conditions a reliable reaction time effect in
their very similar, albeit forced-choice occulomotor
paradigm using even more test trials (32 vs. 96). This
indicates long-lasting, extinction-resistant bidirectional
Further, since the current paradigm is conceptually
very close to that of Herwig and Horstmann (2011), and
the adults failed to show frequency effects, it is more
likely that in the manual version of the actioneffect task
of Verschoor et al. (2010), the action effects affected
response choice differently from the current paradigm.
In manual actioneffect paradigms the only attention-
drawing events in the test phase are the presented action
effects. Their mere presence is unlikely to affect action
choice directly, so that all possible response biases can be
attributed to the degree to which the action effect
reactivated a previously acquired association, which then
spread activation to the corresponding response repre-
sentation. In other words, even though action effects
attract exogenous attention, they eventually impact
action selection in an entirely endogenous fashion.
Indeed, neuroimaging studies have shown that the
presentation of previously acquired action effects acti-
vates the supplementary motor area, which underlies
endogenously driven but not exogenously driven action
selection (Elsner, Hommel, Mentschel, Drzezga, Prinz,
Conrad & Siebner, 2002; Melcher, Weidema, Eenshuis-
tra, Hommel & Gruber, 2008; Paulus et al., 2011). In
contrast, in our oculomotor version of the task, the
©2013 John Wiley & Sons Ltd
810 Stephan A. Verschoor et al.
endogenous impact of the action effect competes with
the direct, exogenous impact of the saccade goals the
faces in our case. It is possible that this exogenous impact
is so strong that it outweighs the impact of the
endogenous bias to a degree that the latter is too weak
to determine which response is being chosen, even
though it can still speed up congruent and/or slow down
incongruent responses. Accordingly, the present findings
do not necessarily require the assumption that action
effect learning is different in, or follows different
developmental pathways with respect to, manual and
oculomotor actions. One could even speculate that the
exogenous attention evoked had a stronger impact on the
7-month-olds, thus resulting in a lack of RT effect in this
As expected, we found reliable effects of congruency
on pupil dilation with incongruent saccades resulting in
larger relative dilations. These findings need to be
interpreted with caution, as there are no shared stan-
dards regarding the handling of TEPRs. TEPRs can start
from 200300 ms after stimulus onset and peak around
1200 ms post-stimulus (Beatty & Lucero-Wagoner,
2000), sometimes even later (e.g. Bradly et al., 2008;
Beatty, 1982). The effects found do fit within these
temporal dynamics. On the other hand, little is known
about developmental and aging-related changes in these
biometric variables across the lifespan. Since our exper-
iment is self-paced (to ensure infant cooperation) ITIs do
vary systematically between age groups, with shorter
ITIs for the adults. What is more, in the 12-month-olds
and adults RTs also vary with congruency. These timing
factors could reduce pupillary effects in these age groups.
Unfortunately, due to the attentional abilities of infants
standardizing the ITIs was not an option. Furthermore,
since light adaptation dilations decrease in amplitude
and latency with increasing gestational age (Cocker,
Fielder, Moseley & Edwards, 2005), one could speculate
that the same could hold true for TEPRs. Thus ideal
intervals for measuring TEPRs and baselines may vary
accordingly. Therefore, pupillary effects should be
expected to be most pronounced in the 7-month-olds.
This is indeed what we find. In the current study we
chose intervals as suggested by the literature. However,
in the literature there is no standard for response-related
evaluative effects.
To accommodate for the variable RTs and ITIs in the
current experiment, we considered both stimulus-locked
and response-locked TEPRs, which, however, yielded
identical outcomes. Of particular interest, both analyses
revealed main effects of congruency but no interaction
with age. Moreover, the congruency effects remained
reliable when being tested in the 7-month-olds alone, the
only age group that did not show a congruency effect in
RTs. On the one hand, the fact that 7-month-olds are
sensitive to the congruency between their action and the
presented action effect demonstrates that they have
acquired information about the relationship between
their actions and the novel auditory effects these actions
produced in the experiment. Accordingly, we take this
observation to indicate that even the youngest group was
able to integrate some kind of information about actions
and their effects. On the other hand, the dissociation
between the dilation effect and the RT effect in these
infants suggests that the two measures do not assess the
same underlying processes.
As suggested by Band et al. (2009) and Blakemore,
Frith and Wolpert (1999), actioneffect associations may
not only serve as an informational basis for action
selection, the major theme of ideomotor theory, but also
for predicting the perceptual consequences of an action.
This prediction can be matched against the actually
produced consequences in order to evaluate whether the
action goal was reached. The late timing of our stimulus-
and response-locked congruency effects in the TEPRs
suggests that these effects were picking up processes
related to action evaluation rather than action selection
proper. Indeed, error-related negativity type patterns as
reported by Band et al. (2009) in connection with action
prediction have been found to be related to pupillary
responses as well (Wessel, Danielmeier & Ullsperger,
2011). These pupillary effects might thus indicate a
mismatch between expected and actual action effect in
incongruent trials and/or reflect the adaptive processes
thereby triggered that update the systems knowledge
about actioneffect relationships. The first possibility
would fit well with the violation-of-expectation approach
suggested by Jackson and Sirois (2009) and Gredeb
and Melinder (2010), while the interpretation in terms of
mismatch-induced control processes would be more
along the lines of the traditional TEPR literature
which focuses on arousal, attention allocation, cognitive
load and mental effort. In any case, the effect reflects
knowledge about actioneffect contingencies, and our
stimulus- and response-locked findings suggest that this
knowledge is equally present in all three age groups.
This dissociation between RT findings, which imply
action-selection effects in adults and 12-month-olds, and
TEPRs, which suggest action-evaluation effects in all
participants, allows for two important conclusions. First,
the processing of an actioneffect stimulus activates a
representation that creates particular expectations, with-
out necessarily activating the corresponding actions. This
means that actioneffect expectation may be correlated
with, and perhaps even functional for, action selection
uhn, Keizer, Rombouts & Hommel, 2011) in older
agents, but raising an expectation is not identical to
©2013 John Wiley & Sons Ltd
Spontaneous actioneffect binding in infants and adults 811
selecting an action. Second, infantsabilities to construct
actioneffect expectations develop earlier than their
abilities to use actioneffect representation for inten-
tional action selection. A similar dissociation between
acquisition of actioneffect knowledge versus use of
actioneffect knowledge was reported by Sommerville
et al. (2005). They showed that violation-of-expectation
to a change of goal was influenced by actioneffect
experience in 3-month-olds, while the observation of
actions and their effects did not influence action
There are several possible reasons why selection
abilities develop more slowly and why actioneffect
knowledge in the 7-month-olds affected expectation-
related effects only. One possibility is that associations
between motor patterns and novel action effects are
either not yet bidirectional, are too weak, or take too
much time to retrieve to affect performance under our
testing conditions. Another possibility is that novel
action effects are not yet directly associated with actions
but only with representations of already existing action
effects. Taken altogether, it seems safe to assume that, at
7 months of age, knowledge about relations between
actions and their effects has a stronger impact on the
prediction of action effects than on the selection of
intentional actions.
To conclude, the dissociation we obtained in 7-month-
olds suggests a developmental precedence of action
monitoring over intentional action selection. This again
suggests that infants acquire the knowledge necessary for
performing intentional actions sometime before they
have (fully) developed the cognitive machinery necessary
to make use of that knowledge to perform intentional
action (Keen, 2005; Sommerville et al., 2005). Combin-
ing the current data with those of Verschoor et al. (2010)
suggests a major change in actioneffect learning from
just action monitoring to action selection, just before the
ninth month of age. If one takes into account the
functional and representational equivalence of self-per-
formed and perceived actions as suggested by Theory of
Event Coding (Hommel et al., 2001), this pattern fits
with data suggesting that at 6 months of age infants can
understand goal-directed action (e.g. Woodward, 1998 ),
or more accurately, experience violation-of-expectation
to a change of goal, but are unable to perform true
intentional action (distinguishing means from ends) until
around 8 to 9 months of age (Goubet, Rochat, Maire-
Leblond & Poss, 2006; Hauf, 2007; Piaget, 1936, [1963]).
Our findings also fit with results from studies on action
perception, showing that infants at 9 but not 7 months of
age can use observed actioneffect relations to guide
behavior (Hauf & Aschersleben, 2008). In addition, our
data suggest that motor resonance when listening to
previously self-produced sounds in 8-month-olds, as
found by Paulus et al. (2011), might indeed reflect the
existence of knowledge about actioneffect relations; and
yet, we do not necessarily expect this knowledge to result
in overt behavior, at least not at 7 months of age. Similar
evidence for actionknowledge activation during action
observation has been obtained in infants as young as 6
months (Nystr
om, 2008). Some authors have argued that
it is lacking representational equivalence between self-
produced actions and observed actions that prohibits
infants younger than 9 months from imitation (Hauf,
2007). Our data, together with those of Paulus et al.
(2011), Verschoor et al. (2010), Nystr
om (2008), and
Sommerville et al. (2005), suggest that it is not repre-
sentational equivalence that is reached by 9 months of
age, but the ability to successfully use bidirectional
actioneffect associations, learned either by observation
or by experience, for voluntary action.
The authors declare no competing interests.
This research was supported by the Netherlands Orga-
nization for Scientific Research. We especially thank
Thijs Schrama for technical support and Henk van
Steenbergen for analytical support.
Baker, C.L., Saxe, R., & Tenenbaum, J.B. (2009). Action
understanding as inverse planning. Cognition,113 (3), 329
Band, G.P.H., van Steenbergen, H., Ridderinkhof, K.R.,
Falkenstein, M., & Hommel, B. (2009). Actioneffect nega-
tivity: irrelevant action effects are monitored like relevant
feedback. Biological Psychology,82, 211218.
Beatty, J. (1982). Task-evoked pupillary responses, processing
load, and the structure of processing resources. Psychological
Bulletin,91, 276292.
Beatty, J., & Lucero-Wagoner, B. (2000). The pupillary system.
In J. Caccioppo, L.G. Tassinary & G. Berntson (Eds.), The
handbook of psychophysiology (2nd edn., pp. 142162).
Cambridge: Cambridge University Press.
Biro, S., & Leslie, A. (2007). Infantsperception of goal-dir-
ected actions: development through cue-based bootstrap-
ping. Developmental Science,10, 379398.
Blakemore, S.-J., Frith, C.D., & Wolpert, D.W. (1999).
Spatiotemporal prediction modulates the perception of
©2013 John Wiley & Sons Ltd
812 Stephan A. Verschoor et al.
self-produced stimuli. Journal of Cognitive Neuroscience,11,
Bradley, M.M., Miccoli, L., Escrig, M.A., & Lang, P.J. (2008).
The pupil as a measure of emotional arousal and autonomic
activation. Psychophysiology,45, 602607.
Cocker, K.D., Fielder, A.R., Moseley, M.J., & Edwards, A.D.
(2005). Measurements of pupillary responses to light in
term and preterm infants. Neuro-Ophthalmology,29,95
Csibra, G. (2008). Goal attribution to inanimate agents by
6.5-month-old infants. Cognition,107, 705717.
Csibra, G., Gergely, G., Biro, S., & Ko
os, O. (1999). Goal
attribution without agency cues: the perception of pure
reasonin infancy. Cognition,72, 237267.
DeCasper, A.J., & Fifer, W.P. (1980). Of human bonding:
newborns prefer their mothersvoices. Science,1208, 1174
Dutzi, I.B., & Hommel, B. (2009). The microgenesis of action
effect binding. Psychological Research,73, 425435.
Eenshuistra, R.M., Weidema, M.A., & Hommel, B. (2004).
Development of the acquisition and control of actioneffect
associations. Acta Psychologica,115, 185209.
Elsner, B. (2007). Infantsimitation of goal-directed actions:
the role of movements and action effects. Acta Psychologica,
Elsner, B., & Hommel, B. (2001). Effect anticipation and action
control. Journal of Experimental Psychology: Human Percep-
tion and Performance,27, 229240.
Elsner, B., & Hommel, B. (2004). Contiguity and contingency
in actioneffect learning. Psychological Research,68, 138
Elsner, B., Hommel, B., Mentschel, C., Drzezga, A., Prinz, W.,
Conrad, B., & Siebner, H.R. (2002). Linking actions and
their perceivable consequences in the human brain. Neuro-
Image,17, 364372.
Fabbri-Destro, M., & Rizzolatti, G. (2008). Mirror neurons and
mirror systems in monkeys and humans. Physiology,23, 171
Falck-Ytter, T. (2008). Face inversion effects in autism: a
combined looking time and pupillometric study. Autism
Research,1, 297306.
Falck-Ytter, T., Gredeb
ack, G., & von Hofsten, C. (2006).
Infants predict other peoples action goals. Nature Neuro-
science,9, 878879.
Gergely, G., & Watson, J.S. (1999). Early social-emotional
development: contingency perception and the social biofeed-
back model. In P. Rochat (Ed.), Early social cognition (pp.
101136). Hillsdale, NJ: Erlbaum Publishers.
Goren, C.C., Sarty, M., & Wu, P.Y. (1975). Visual following
and pattern discrimination of face-like stimuli by newborn
infants. Pediatrics,56, 544549.
Goubet, N., Rochat, P., Maire-Leblond, C., & Poss, S. (2006).
Learning from others in 9- to 18-month-old infants. Infant
Child Development,15, 161177.
ack, G., & Melinder, A. (2010). Infantsunderstanding
of everyday social interactions: a dual process account.
Cognition,114, 197206.
Greenwald, A.G. (1970). Sensory feedback mechanisms in
performance control: with special reference to the ideomotor
mechanism. Psychological Review,77,7399.
Harless, E. (1861). Der Apparat des Willens. Zeitschrift fuer
Philosophie und philosophische Kritik,38,5073.
Hartshorn, K., Rovee-Collier, C., Gerhardstein, P., Bhatt, R.S.,
Wondoloski, T.L., Klein, P., Gilch, J., Wurzel, N., &
Campos-de-Carvalho, M. (1997). The ontogeny of long-term
memory over the first year-and-a-half of life. Developmental
Hauf, P. (2007). Infantsperception and production of inten-
tional actions. Progress in Brain Research: From Action to
Cognition,164, 285301.
Hauf, P., & Aschersleben, G. (2008). Actioneffect anticipa-
tion in infant action control. Psychological Research,72,
Herwig, A., & Horstmann, G. (2011). Actioneffect associa-
tions revealed by eye movements. Psychonomic Bulletin &
Review,18, 531537.
Hess, E.H. (1975). The tell-tale eye: How your eyes reveal hidden
thoughts and emotions. New York: Van Nostrand Reinhold.
Hess, E.H., & Polt, J. (1960). Pupil size as related to interest
value of visual stimuli, Science,132, 149150.
Hess, E.H., & Polt, J. (1964). Pupil size in relation to mental
activity during simple problem-Solving. Science,13, 1190
Hommel, B. (1996). The cognitive representation of action:
automatic integration of perceived action effects. Psycholog-
ical Research,59, 176186.
Hommel, B., & Elsner, B. (2009). Acquisition, representation,
and control of action. In E. Morsella, J.A. Bargh & P.M.
Gollwitzer (Eds.), Oxford handbook of human action (pp.
371398). New York: Oxford University Press.
Hommel, B., M
usseler, J., Aschersleben, G., & Prinz, W. (2001).
The theory of event coding (TEC): a framework for
perception and action planning. Behavioral and Brain
Sciences,24, 849937.
Hupe, J.M., Lamirel, C., & Lorenceau, J. (2009). Pupil
dynamics during bistable motion perception. Journal of
with pupil dilation. Developmental Science,12, 670679.
James, W. (1890). The principles of psychology. New York:
Macmillan/Harvard University Press.
Johnson, M.H., Dziurawiec, S., Ellis, H., & Morton, J. (1991).
Newbornspreferential tracking of face-like stimuli and its
subsequent decline. Cognition,40,119.
Johnson, S.C., Ok, S.-J., & Luo, Y. (2007). The attribution of
attention: 9-month-oldsinterpretation of gaze as goal-dir-
ected action. Developmental Science,10, 530537.
Kahneman, D. (1973). Attention and effort. Englewood Cliffs,
NJ: Prentice Hall.
Kahneman, D., & Beatty, J. (1966). Pupil diameter and load on
memory. Science,154, 15831585.
Keen, R. (2005). Using perceptual representations to guide
reaching and looking. In J. Rieser, J. Lockman & C.A.
Nelson (Eds.), Action as an organizer of learning and
©2013 John Wiley & Sons Ltd
Spontaneous actioneffect binding in infants and adults 813
development: Minnesota Symposium on Child Psychology, 33
(pp. 301322). Mahwah, NJ: Lawrence Erlbaum Associates.
Kiraly, I., Jovanovic, B., Prinz, W., Aschersleben, G., &
Gergely, G. (2003). The early origins of goal attribution in
infancy. Consciousness and Cognition,12, 752769.
Klein, A., Hauf, P., & Aschersleben, G. (2006). The role of
action effects in 12-month-oldsaction control: a comparison
of televised model and live model. Infant Behavior and
Development,29, 535544.
Kray, J., Eenshuistra, R., Kerstner, H., Weidema, M., &
Hommel, B. (2006). Language and action control: the
acquisition of action goals in early childhood. Psychological
Science,17, 737741.
uhn, S., Keizer, A., Rombouts, S.A.R.B., & Hommel, B.
(2011). The functional and neural mechanism of action
preparation: roles of EBA and FFA in voluntary action
control. Journal of Cognitive Neuroscience,23, 214220.
Laeng, B., & Falkenberg, L. (2007). Womens pupillary
responses to sexually significant others during the hormonal
cycle. Hormones and Behavior,52, 520530.
Laeng, B., Sirois, S., & Gredeb
ack, G. (2012) Pupillometry: A
window to the preconscious? Perspectives on Psychological
Libby, W.L., Lacey, B.C., & Lacey, J.I. (1973). Pupillary and
cardiac activity during visual attention. Psychophysiology,10,
Lotze, R.H. (1852). Medicinische Psychologie oder Physiologie
der Seele. Leipzig: Weidmann.
Melcher, T., Weidema, M., Eenshuistra, R.M., Hommel, B., &
Gruber, O. (2008). The neural substrate of the ideomotor
principle: an event-related fMRI analysis. NeuroImage,39,
Meltzoff, A.N. (2006). The like meframework for recognizing
and becoming an intentional agent. Acta Psychologica,124,
Meltzoff, A.N., & Moore, M.K. (1997). Explaining facial
imitation: a theoretical model. Early Development and
Parenting,6, 179192.
Meltzoff, A.N., & Prinz, W. (Eds.) (2002). The imitative mind:
Development, evolution and brain bases. Cambridge: Cam-
bridge University Press.
Miltner, W.H.R., Braun, C.H., & Coles, M.G.H. (1997).
Event-related potentials following incorrect feedback in a
time-estimation task: evidence for a genericneural system for
error detection. Journal of Cognitive Neuroscience,9, 788798.
om, P. (2008). The infant mirror neuron system studied
with high density EEG. Social Neuroscience,3(34), 334337.
Paulus, M., Hunnius, S., Elk, M., & Beckering, H. (2012). How
learning to shake a rattle affects 8-month-old infants
perception of the rattles sound: electrophysiological evi-
dence for actioneffect binding in infancy. Developmental
Cognitive Neuroscience,2(1), 9096.
Perra, O., & Gattis, M. (2010). The control of social attention
from 1 to 4 months. British Journal of Developmental
Psychology,28, 891908.
Piaget, J. (1936 [1963]). The origins of intelligence in children.
New York: W.W. Norton & Company.
Piaget, J. (1954). The construction of reality in the child. New
York: Basic Books.
Prinz, W. (1990). A common coding approach to perception
and action. In O. Neumann & W. Prinz (Eds.), Relationships
between perception and action (pp. 167201). Berlin: Springer.
Prinz, W. (1997). Perception and action planning. European
Journal of Cognitive Psychology,9, 129154.
Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron
system. Annual Review of Neuroscience,27, 169192.
Rochat, P. (2001). The infants world. Cambridge, MA: Harvard
University Press.
Rochat, P., & Striano, T. (1999). Emerging self-exploration by
2-month-old infants. Developmental Science,2, 206218.
Rovee, C.K., & Rovee, D.T. (1969). Conjugate reinforcement of
infant exploratory behavior. Journal of Experimental Child
Rovee-Collier, C. (1999). The development of infant memory.
Current Directions in Psychological Science,8,8085.
Scerif, G., Karmiloff-Smith, A., Campos, R., Elsabbagh, M.,
Driver, J., & Cornish, K. (2005). To look or not to look?
Typical and atypical development of oculomotor control.
Journal of Cognitive Neuroscience,4, 591604.
Schneider, W., Eschman, A., & Zuccolotto, A. (2002). E-Prime
users guide. Pittsburgh, PA: Psychology Software Tools Inc.
Senju, A., & Csibra, G. (2008). Gaze following in human
infants depends on communicative signals. Current Biology,
18, 668671.
Snodgrass, G., & Vanderwart, M. (1980). A standardized set of
260 pictures: norms for naming agreement, familiarity, and
visual complexity. Journal of Experimental Psychology:
Human Learning and Memory,6, 174215.
Sommerville, J.A., Woodward, A.L., & Needham, A. (2005).
Action experience alters 3-month-old infantsperception of
othersactions. Cognition,96,B1B11.
Tomasello, M. (1999). The cultural origins of human cognition.
Cambridge, MA: Harvard University Press.
Verschoor, S., & Biro, S. (2012). Means selection information
overrides outcome selection information in infantsgoal
attribution. Cognitive Science,36, 714725.
Verschoor, S.A., Weidema, M., Biro, S., & Hommel, B. (2010).
Where do action goals come from? Evidence for spontaneous
actioneffect binding in infants. Frontiers in Cognition,1, 201.
von Hofsten, C. (2004). An action perspective on motor
development. Trends in Cognitive Sciences,8, 266272.
Watson, J.S. (1967). Memory and contingency analysisin
infant learning. Merrill-Palmer Quarterly,13,5576.
Wessel, J.R., Danielmeier, C., & Ullsperger, M. (2011). Error
awareness revisited: Accumulation of multimodal evidence
from central and autonomic nervous systems. Journal of
Cognitive Neuroscience,23, 30213036.
Woodward, A. (1998). Infants selectively encode the goal object
of an actors reach. Cognition,69,134.
Woodward, A.L. (2009). Infantsgrasp of othersintentions.
Current Directions in Psychological Science,18,5357.
Received: 31 August 2011
Accepted: 26 March 2013
©2013 John Wiley & Sons Ltd
814 Stephan A. Verschoor et al.
... A crucial advantage of focusing on gaze when testing very young infants is that infants as young as four months are already capable of visuo-attentional control (Johnson, Posner, & Rothbart, 1991), while motor control is still limited at that age. Infants as young as seven months are able to acquire oculomotor action-effect associations, while the use of such action-effect associations for action control has been shown to occur at an age of around one year (Verschoor, Spape, Biro, & Hommel, 2013). 3 ...
... However, more relevant for the present work are studies focusing on effect-based oculomotor control in adults. These studies have extended the ideomotor approach to the oculomotor domain using saccade latencies, the oculomotor counterpart to manual RTs, to measure action-effect learning (e.g., Herwig & Horstmann, 2011;Huestegge & Kreutzfeldt, 2012;Verschoor et al., 2013;see Herwig, 2015, for a recent review). Among them, the study by Huestegge and Kreutzfeldt (2012) is of particular relevance for the presented research. ...
... While there is already a number of relevant studies on ideomotor principles in gaze control in adults (e.g., Herwig & Horstmann, 2011;Huestegge & Kreutzfeldt, 2012;Verschoor et al., 2013), there are still some open questions that have not yet been answered empirically, especially with respect to gaze control within animate versus inanimate environments. More specifically, a clear research gap exists with respect to the control of goal-oriented eye movements that serve to elicit changes in the gaze behavior of the interaction partner, such as when humans move their eyes in order to get the other person to look at a particular place. ...
Full-text available
Humans use their eyes not only as visual input devices to perceive the environment, but also as an action tool in order to generate intended effects in their environment. For instance, glances are used to direct someone else's attention to a place of interest, indicating that gaze control is an important part of social communication. Previous research on gaze control in a social context mainly focused on the gaze recipient by asking how humans respond to perceived gaze (gaze cueing). So far, this perspective has hardly considered the actor’s point of view by neglecting to investigate what mental processes are involved when actors decide to perform an eye movement to trigger a gaze response in another person. Furthermore, eye movements are also used to affect the non-social environment, for instance when unlocking the smartphone with the help of the eyes. This and other observations demonstrate the necessity to consider gaze control in contexts other than social communication whilst at the same time focusing on commonalities and differences inherent to the nature of a social (vs. non-social) action context. Thus, the present work explores the cognitive mechanisms that control such goal-oriented eye movements in both social and non-social contexts. The experiments presented throughout this work are built on pre-established paradigms from both the oculomotor research domain and from basic cognitive psychology. These paradigms are based on the principle of ideomotor action control, which provides an explanatory framework for understanding how goal-oriented, intentional actions come into being. The ideomotor idea suggests that humans acquire associations between their actions and the resulting effects, which can be accessed in a bi-directional manner: Actions can trigger anticipations of their effects, but the anticipated resulting effects can also trigger the associated actions. According to ideomotor theory, action generation involves the mental anticipation of the intended effect (i.e., the action goal) to activate the associated motor pattern. The present experiments involve situations where participants control the gaze of a virtual face via their eye movements. The triggered gaze responses of the virtual face are consistent to the participant’s eye movements, representing visual action effects. Experimental situations are varied with respect to determinants of action-effect learning (e.g., contingency, contiguity, action mode during acquisition) in order to unravel the underlying dynamics of oculomotor control in these situations. In addition to faces, conditions involving changes in non-social objects were included to address the question of whether mechanisms underlying gaze control differ for social versus non-social context situations. The results of the present work can be summarized into three major findings. 1. My data suggest that humans indeed acquire bi-directional associations between their eye movements and the subsequently perceived gaze response of another person, which in turn affect oculomotor action control via the anticipation of the intended effects. The observed results show for the first time that eye movements in a gaze-interaction scenario are represented in terms of their gaze response in others. This observation is in line with the ideomotor theory of action control. 2. The present series of experiments confirms and extends pioneering results of Huestegge and Kreutzfeldt (2012) with respect to the significant influence of action effects in gaze control. I have shown that the results of Huestegge and Kreutzfeldt (2012) can be replicated across different contexts with different stimulus material given that the perceived action effects were sufficiently salient. 3. Furthermore, I could show that mechanisms of gaze control in a social gaze-interaction context do not appear to be qualitatively different from those in a non-social context. All in all, the results support recent theoretical claims emphasizing the role of anticipation-based action control in social interaction. Moreover, my results suggest that anticipation-based gaze control in a social context is based on the same general psychological mechanisms as ideomotor gaze control, and thus should be considered as an integral part rather than as a special form of ideomotor gaze control.
... For these and further reasons, we concluded that anticipatory saccades reflected processes that were dissociable from effects of action effect anticipation on manual action selection (Pfeuffer et al., 2016). Instead, we suggested that anticipatory saccades reflected an anticipatory preparation for evaluating whether the actual effect matched the expected effect (for further theoretical disseminations of the idea that goal-directed action control consists not only of processes related to action selection, but also of processes related to outcome evaluation/effect monitoring see e.g., Band et al., 2009;Chambon & Haggard, 2013;Hommel, 2015Hommel, , 2017Verschoor et al., 2013;for cybernetic comparator models of movements including control loops based on the comparison of expected and actual effects see e.g., Frith & Wolpert, 2000;Wolpert & Flanagan, 2001;Wolpert & Ghahramani, 2000). That is, anticipatory saccades, spontaneously occurring during goal-directed action control, reflect a proactive effect monitoring process which prepares a later comparison of expected and actual effect. ...
Full-text available
When our actions yield predictable consequences in the environment, our eyes often already saccade towards the locations we expect these consequences to appear at. Such spontaneous anticipatory saccades occur based on bi-directional associations between action and effect formed by prior experience. That is, our eye movements are guided by expectations derived from prior learning history. Anticipatory saccades presumably reflect a proactive effect monitoring process that prepares a later comparison of expected and actual effect. Here, we examined whether anticipatory saccades emerged under forced choice conditions when only actions but not target stimuli were predictive of future effects and whether action mode (forced choice vs. free choice, i.e., stimulus-based vs. stimulus-independent choice) affected proactive effect monitoring. Participants produced predictable visual effects on the left/right side via forced choice and free choice left/right key presses. Action and visual effect were spatially compatible in one half of the experiment and spatially incompatible in the other half. Irrespective of whether effects were predicted by target stimuli in addition to participants' actions, in both action modes, we observed anticipatory saccades towards the location of future effects. Importantly, neither the frequency, nor latency or amplitude of these anticipatory saccades significantly differed between forced choice and free choice action modes. Overall, our findings suggest that proactive effect monitoring of future action consequences, as reflected in anticipatory saccades, is comparable between forced choice and free choice action modes.
... La TEC postule qu'un mécanisme d'intégration capte toutes les conséquences sensorielles de l'action (i.e., les événements perçus) et les associe au pattern moteur actuellement actif (i.e., l'événement produit) (Hommel & Elsner, 2009 ;Stoet & Hommel, 1999 ;Verschoor, Spapé, Biro, & Hommel, 2013). Par conséquent, réaliser une action devrait mener à lier des codes moteurs (e.g., la vitesse du mouvement, son orientation, son amplitude, etc.) aux codes représentant le contexte perceptif (i.e., le stimulus pour lequel l'action est déclenchée, la situation dans laquelle l'action se déroule, etc.). ...
Full-text available
Acting close to other people requires to take into account their actions, and their potential impact on the current situation. One of the main consequence related to this natural tendency to (co-)represent others’ actions is the need for discriminating between self- and other-related action representations. Today, despite major advances in the joint action literature, it remains some theoretical and mechanistic oppositions regarding the impact of certain situational and dispositional factors on the emergence and strength of the self-other discrimination problem. In the present manuscript, the nature and the resolution of this phenomenon, but also the typical behavioral measure used to illustrate it are being discussed. This thesis work supports an embodied social cognition approach: we highlighted the core role of sensorimotor experiences and the redundant coding of information resulting of, in resolving the self-other discrimination problem. It is hypothesized that the typical measure used to illustrate the self-other discrimination problem is, above all, an index of the ease with which self from the other.
... La TEC postule qu'un mécanisme d'intégration capte toutes les conséquences sensorielles de l'action (i.e., les événements perçus) et les associe au pattern moteur actuellement actif (i.e., l'événement produit) (Hommel & Elsner, 2009 ;Stoet & Hommel, 1999 ;Verschoor, Spapé, Biro, & Hommel, 2013). Par conséquent, réaliser une action devrait mener à lier des codes moteurs (e.g., la vitesse du mouvement, son orientation, son amplitude, etc.) aux codes représentant le contexte perceptif (i.e., le stimulus pour lequel l'action est déclenchée, la situation dans laquelle l'action se déroule, etc.). ...
Full-text available
Il est communément accepté que le simple fait d’agir à proximité d’autrui nous amène naturellement à considérer son action, et ses potentielles conséquences sur l’environnement. L’une des principales conséquences liée à cette tendance à (co-)représenter les actions d’autrui se caractérise par le besoin de discriminer entre des représentations référant à soi et à autrui. Malgré des avancées majeures, aujourd’hui encore, des oppositions théoriques et mécanistiques demeurent. Ces dernières se cristallisent autour de l’interprétation de l’incidence de certains facteurs censés affecter la genèse et la force de ce problème de discrimination. En cela, la nature de ce phénomène, sa résolution, et ce à quoi réfère vraiment la mesure typiquement utilisée pour l’illustrer sont discutés tout du long de ce document. Ce travail de thèse soutient une perspective incarnée de la cognition sociale en démontrant le rôle déterminant des expériences sensori-motrices, et du codage redondant des informations qui en résultent, dans le processus de résolution du problème de discrimination. Nous faisons ainsi l’hypothèse que la mesure utilisée est avant tout un indicateur de la facilité avec laquelle nous résolvons le problème de discrimination entre soi et autrui.
... Recent studies maintain that, apart from an affectation in gaze following, there exists an atypical regulation in the autonomic nervous system (ANS) in children with ASD, which may be contributing to the difficulties that they show in social processing. A reliable measure for studying this atypical regulation would be pupil dilation, given that babies are capable of controlling eye movements from four months of age [13]. ...
(1) Background: Children with autism spectrum disorder (ASD) show certain characteristics in visual attention which generate difficulties in the integration of relevant social information to set the basis of communication. Gaze following and pupil dilation could be used to identify signs for the early detection of ASD. Eye-tracking methodology allows objective measurement of these anomalies in visual attention. The aim is to determine whether measurements of gaze following and pupillary dilation in a linguistic interaction task, captured using eye-tracking methodology, are objective for early diagnosis of ASD. (2) Methods: 20 children between 17 and 24 months of age, made up of 10 neurotypical children and 10 children with ASD were paired together according to chronological age. A human face on a monitor pronounced pseudowords associated with pseudo-objects. Gaze following and pupil dilation was registered during the task. (3) Results: Significant statistical differences were found in the time of gaze fixation on the human face and on the object, as well as in the number of gazes. Also, there were significant differences in the maximum peak of pupil dilation, this being found in the neurotypical group at the moment of processing of the pseudoword, and in the ASD group in the baseline prior to the task (4) Conclusions: The registration and the duration of gaze, and the measurement of pupil dilation with ‘eye-tracker’ are objective measures for early detection of ASD.
... C'est l'histoire de l'erreur A-B : un événement moteur peut être complètement intégré à une situation de telle façon que l'absence de changement sensoriel suffisant ne permet pas une modification de l'équilibre mis en place par le système cognitif Smith, 2005a). Finalement, au cours de la phase d'acquisition de cette expérience, les enfants, apprennent un comportement moteur associé à l'effet A alors qu'aucun mouvement n'accompagne l'effet B. Or, pour utiliser une métaphore liée à l'enfance, c'est plutôt l'histoire du bonbon que l'on veut raconter : je perçois les conséquences sensorielles avant de manger (ou prendre) le bonbon, c'est-à-dire une agréable sensation gustative (ou une tape sur la main !). Verschoor et al. (2013) ont remédié à ce problème en utilisant une procédure impliquant le même type d'action pour les deux effets. En outre, ils montrent de manière intéressante que des enfants de 7 mois prédisent également les effets de l'action sans toutefois les sélectionner. ...
Le travail de thèse que nous avons réalisé se proposait d’étudier l’anticipation des effets reliés au corps. La théorie idéomotrice de James (1890) qui soutient que c’est l’idée du mouvement qui déclenche une action constitue le sous bassement théorique de cette thèse. Dans le prolongement des rares travaux portant sur l’anticipation des effets reliés au corps, nous avons émis l’hypothèse que les effets tactiles et proprioceptifs étaient anticipés en fonction de la manière dont était réalisée l’action. Plus précisément, nous avons étudié dans une première expérience l’effet d’une contingence d’intensité entre une réponse et un effet tactile; une seconde expérience a porté sur le phénomène d’atténuation proprioceptive à partir d’un effet tactile; une troisième expérience a mis l’emphase sur l’anticipation de la fluence motrice, comprise comme un effet proprioceptif fournit au cours du mouvement. Ces expériences ont été enrichies de deux autres études portant sur (1)l’anticipation d’un effet tactile concomitant avec l’action et (2) le jugement perceptif d’un effet tactile. Enfin, nous avons entrepris d’appliquer ces recherches dans le développement de l’enfant et dans le domaine de la déficience motrice suite à une lésion cérébrale. Pris ensemble, les données empiriques de ces travaux mettent en lumière le rôle des effets reliés au corps dans l’anticipation de l’action. Ces résultats sont discutés au regard des théories récentes sur la prédiction des effets de l’action d’où émerge la notion de temporalité et fournissent une contribution supplémentaire à l’idée motrice de James. En outre, elles offrent un cadre théorique pertinent afin d’étudier les effets reliés au corps dans le champ de la lésion cérébrale. Enfin, en nous appuyant sur les données obtenues ainsi que sur leur discussion en termes de prédiction, nous proposons des perspectives de recherche auprès de personnes présentant une déficience motrice, des troubles d’utilisation des objets ou des enfants ayant eu un accident vasculaire cérébral artériel néonatal.
Learning about actions requires children to identify the boundaries of an action and its units. Whereas some action units are easily identified, parents can support children's action learning by adjusting the presentation and using social signals. However, currently little is understood regarding how children use these signals to learn actions. In the current study we investigate the possibility that communicative signals are a particularly suitable cue for segmenting events. We investigated this hypothesis by presenting 18-month-old children (N = 60) with short action sequences consisting of toy animals either hopping or sliding across a board into a house, but interrupting this two-step sequence either (a) using an ostensive signal as a segmentation cue, (b) using a non-ostensive segmentation cue, and (c) without additional segmentation information between the actions. Marking the boundary using communicative signals increased children's imitation of the less salient sliding action. Imitation of the hopping action remained unaffected. Crucially, marking the boundary of both actions using a non-communicative control condition did not increase imitation of either action. Communicative signals might be particularly suitable in segmenting non-salient actions that would otherwise be perceived as part of another action or as non-intentional. These results provide evidence of the importance of ostensive signals at event boundaries in scaffolding children's learning.
Full-text available
The temporal duration of world events is subjective. For example, for those who perform or observe it, an intentional action is shorter than an involuntary one. This phenomenon is called Temporal binding (TB) and shows that the action timing is shorter not only for intentional actions, but also if I consider something as the cause of something else. In fact, causality and time are linked in our mind. In the light of the link between perception of time and causality, TB can be used to understand if an individual is aware of acting, provided that this awareness is defined through the understanding that our actions cause consequences. The awareness of acting thus becomes a corollary of causal cognition and influences the time perception, creating the illusion that our goals are closer in time than they really are. In this article, I analyze the relationship between intentional actions, perception of time and causality to explore the origin of action awareness. I will then try to predict when humans and other animals acquire this awareness, offering concrete proposals in mental development and evolution.
Full-text available
La durata temporale degli eventi del mondo è soggettiva. Ad esempio per chi la compie o la osserva, un’azione compiuta intenzionalmente dura meno di una involontaria. Questo fenomeno si chiama Temporal binding (TB) e mostra che il tempo si accorcia nella nostra percezione non solo per le azioni intenzionali, ma anche se considero qualcosa come la causa di qualcos’altro. La causalità e il tempo infatti sono legati nella nostra mente. Alla luce del legame tra percezione del tempo e causalità, si può usare il TB per capire se un individuo abbia consapevolezza di agire, a patto però che si definisca tale consapevolezza attraverso la comprensione che le nostre azioni causino conseguenze. La consapevolezza di agire diventa così un corollario della cognizione causale ed influenza la percezione del tempo creando l’illusione che i nostri obiettivi siano più vicini nel tempo di quello che in realtà sono. In questo articolo, analizzo il rapporto tra azioni intenzionali, percezione del tempo e causalità per esplorare l’origine della consapevolezza di agire. Provo poi a prevedere quando gli esseri umani e altri animali acquisiscono tale consapevolezza, offrendo proposte concrete per esplorarla nello sviluppo psicologico e nell’evoluzione.
Full-text available
Background: Children with autism spectrum disorder (ASD) show certain characteristics in visual attention. These may generate differences with non-autistic children in the integration of relevant social information to set the basis of communication. Reliable and objective measurement of these characteristics in a language learning context could contribute to a more accurate early diagnosis of ASD. Gaze following and pupil dilation are being studied as possible reliable measures of visual attention for the early detection of ASD. The eye-tracking methodology allows objective measurement of these biomarkers. The aim of this study is to determine whether measurements of gaze following and pupillary dilation in a linguistic interaction task are potential objective biomarkers for the early diagnosis of ASD. Method: A group of 20 children between 17 and 24 months of age, made up of 10 neurotypical children (NT) and 10 children with an increased likelihood of developing ASD were paired together according to chronological age. A human face on a monitor pronounced pseudowords associated with pseudo-objects. Gaze following and pupil dilation were registered during the task These measurements were captured using eye-tracking methodology. Results: Significant statistical differences were found in the time of gaze fixation on the human face and on the object, as well as in the number of gazes. Children with an increased possibility of developing ASD showed a slightly higher pupil dilation than NT children. However, this difference was not statistically significant. Nevertheless, their pupil dilation was uniform throughout the different periods of the task while NT participants showed greater dilation on hearing the pseudoword. Conclusions: The fixing and the duration of gaze, objectively measured by a Tobii eye-tracking system, could be considered as potential biomarkers for early detection of ASD. Additionally, pupil dilation measurement could reflect differential activation patterns during word processing in possible ASD toddlers and NT toddlers.
Full-text available
The measurement of pupil diameter in psychology (in short, "pupillometry") has just celebrated 50 years. The method established itself after the appearance of three seminal studies (Hess & Polt, 1960, 1964; Kahneman & Beatty, 1966). Since then, the method has continued to play a significant role within the field, and pupillary responses have been successfully used to provide an estimate of the "intensity" of mental activity and of changes in mental states, particularly changes in the allocation of attention and the consolidation of perception. Remarkably, pupillary responses provide a continuous measure regardless of whether the participant is aware of such changes. More recently, research in neuroscience has revealed a tight correlation between the activity of the locus coeruleus (i.e., the "hub" of the noradrenergic system) and pupillary dilation. As we discuss in this short review, these neurophysiological findings provide new important insights to the meaning of pupillary responses for mental activity. Finally, given that pupillary responses can be easily measured in a noninvasive manner, occur from birth, and can occur in the absence of voluntary, conscious processes, they constitute a very promising tool for the study of preverbal (e.g., infants) or nonverbal participants (e.g., animals, neurological patients). © Association for Psychological Science 2012.
This chapter traces the gradual emergence of action control from the experience of action-produced events. It begins by reviewing and integrating fi ndings on the acquisition of action effects, that is, on the learning of associations between movements and perceivable outcomes in infants, children, and adults. Second, it discusses what is actually acquired by these learning processes, that is, how actions and action plans are cognitively represented. Third, it outlines how the acquired knowledge is employed in action control, that is, in the planning and production of goal-directed movement.
A physiological measure of processing load or "mental effort" required to perform a cognitive task should accurately reflect within-task, between-task, and betweenindividual variations in processing demands. This article reviews all available experimental data and concludes that the task-evoked pupillary response fulfills these criteria. Alternative explanations are considered and rejected. Some implications for neurophysiological and cognitive theories of processing resources are discussed.
Two-month-olds and newborns were tested in a situation where they had the opportunity to experience different auditory consequences of their own oral activity on a dummy pacifier. Modulation of oral activity was scored and analyzed relative to two types of contingent auditory feedback, either analog or non-analog to the effort exerted by the infant on the pacifier. The dummy pacifier was connected to an air pressure transducer for recording of oral action. In two different experimental conditions, each time the infant sucked above a certain pressure threshold they heard a perfectly contingent sound of varying pitch. In one condition, the pitch variation was analog to the pressure applied by the infant on the pacifier (analog condition). In another, the pitch variation was random (non-analog condition). As rationale, a differential modulation of oral activity in these two conditions was construed as indexing some voluntary control and the sense of a causal link between sucking and its auditory consequences, beyond mere temporal contingency detection and response-stimulus association. Results indicated that 2-month-olds showed clear signs of modulation of their oral activity on the pacifier as a function of analog versus non-analog condition. In contrast, newborns did not show any signs of such modulation either between experimental conditions (analog versus non-analog contingent sounds) or between baseline (no contingent sounds condition) and experimental conditions. These observations are interpreted as evidence of self-exploration and the emergence of a sense of self-agency by 2 months of age.
Presents a standardized set of 260 pictures for use in experiments investigating differences and similarities in the processing of pictures and words. The pictures are black-and-white line drawings executed according to a set of rules that provide consistency of pictorial representation. They have been standardized on 4 variables of central relevance to memory and cognitive processing: name agreement, image agreement, familiarity, and visual complexity. The intercorrelations among the 4 measures were low, suggesting that they are indices of different attributes of the pictures. The concepts were selected to provide exemplars from several widely studied semantic categories. Sources of naming variance, and mean familiarity and complexity of the exemplars, differed significantly across the set of categories investigated. The potential significance of each of the normative variables to a number of semantic and episodic memory tasks is discussed. (34 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).
A long-standing puzzle in developmental psychology is how infants imitate gestures they cannot see themselves perform (facial gestures). Two critical issues are: (a) the metric infants use to detect cross-modal equivalences in human acts and (b) the process by which they correct their imitative errors. We address these issues in a detailed model of the mechanisms underlying facial imitation. The model can be extended to encompass other types of imitation. The model capitalizes on three new theoretical concepts. First, organ identification is the means by which infants relate parts of their own bodies to corresponding ones of the adult's. Second, body babbling (infants' movement practice gained through self-generated activity) provides experience mapping movements to the resulting body configurations. Third, organ relations provide the metric by which infant and adult acts are perceived in commensurate terms. In imitating, infants attempt to match the organ relations they see exhibited by the adults with those they feel themselves make. We show how development restructures the meaning and function of early imitation. We argue that important aspects of later social cognition are rooted in the initial cross-modal equivalence between self and other found in newborns.