Content uploaded by Giovanni Galfano
Author content
All content in this area was uploaded by Giovanni Galfano on May 14, 2014
Content may be subject to copyright.
This article was downloaded by: [Universita di Padova]
On: 02 October 2012, At: 07:23
Publisher: Psychology Press
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office:
Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
The Quarterly Journal of Experimental
Psychology
Publication details, including instructions for authors and subscription
information:
http://www.tandfonline.com/loi/pqje20
Eye gaze cannot be ignored (but neither can
arrows)
Giovanni Galfano a , Mario Dalmaso a , Daniele Marzoli a , Giulia Pavan a ,
Carol Coricelli a & Luigi Castelli a
a Department of Developmental and Social Psychology, University of Padua,
Padua, Italy
Accepted author version posted online: 09 Feb 2012.Version of record first
published: 18 Apr 2012.
To cite this article: Giovanni Galfano, Mario Dalmaso, Daniele Marzoli, Giulia Pavan, Carol Coricelli & Luigi
Castelli (2012): Eye gaze cannot be ignored (but neither can arrows), The Quarterly Journal of Experimental
Psychology, 65:10, 1895-1910
To link to this article: http://dx.doi.org/10.1080/17470218.2012.663765
PLEASE SCROLL DOWN FOR ARTICLE
Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions
This article may be used for research, teaching, and private study purposes. Any substantial
or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or
distribution in any form to anyone is expressly forbidden.
The publisher does not give any warranty express or implied or make any representation that the
contents will be complete or accurate or up to date. The accuracy of any instructions, formulae,
and drug doses should be independently verified with primary sources. The publisher shall not
be liable for any loss, actions, claims, proceedings, demand, or costs or damages whatsoever or
howsoever caused arising directly or indirectly in connection with or arising out of the use of this
material.
Eye gaze cannot be ignored (but neither can arrows)
Giovanni Galfano, Mario Dalmaso, Daniele Marzoli, Giulia Pavan, Carol Coricelli,
and Luigi Castelli
Department of Developmental and Social Psychology, University of Padua, Padua, Italy
Recent studies have tried to shed light on the automaticity of attentional shifts triggered by gaze and
arrows with mixed results. In the present research, we aimed at testing a strong definition of resistance
to suppression for orienting of attention elicited by these two cues. In five experiments, participants
were informed with 100% certainty about the future location of a target they had to react to by presen-
tation of either a direction word at the beginning of each trial or instructions at the beginning of each
block. Gaze and arrows were presented before the target as uninformative distractors irrelevant for the
task. The results showed similar patterns for gaze and arrows—namely, an interference effect when the
distractors were incongruent with the upcoming target location. This suggests that the orienting of
attention mediated by gaze and arrows can be considered as strongly automatic.
Keywords: Attention; Automaticity; Gaze; Arrows; Spatial cueing.
Over recent decades, the study of visuospatial atten-
tion mechanisms has focused on the attempt to
identify the stimuli that are potentially able to elicit
reflexive attention shifts by means of the spatial-
cueing paradigm (e.g., Posner & Cohen, 1984). In
this paradigm, a peripheral target requiring some
kind of response is preceded by a cue with a spatially
congruent or incongruent spatial vector in order to
test whether performance is enhanced when the
target appears at a location in which attention has
been oriented. Initially, it was thought that only per-
ipheral abrupt onset cues were able to trigger reflexive
attention shifts (Jonides, 1981). However, in recent
years, evidence has been provided showing that
arrows (i.e., a central cue) can trigger a shift of atten-
tion even when uninformative with respect to target
location, thus suggesting that arrow-mediated
orienting can be defined as automatic (Eimer,
1997; Ristic, Friesen, & Kingstone, 2002; Tipples,
2002). Moreover, it has been shown that participants
also tend to shift attention following the direction
signalled by the gaze of a face presented at fixation
(e.g., Driver et al., 1999; Friesen & Kingstone,
1998; Frischen, Bayliss, & Tipper, 2007; Galfano
et al., 2011; Hietanen, 1999; Langton & Bruce,
1999; S. P. Tipper, 2010), a phenomenon known
as gaze-mediated orienting.
Although it is well established that both arrows
and gaze can indeed elicit exogenous attention
shifts, the extent to which attention orienting eli-
cited by these stimuli is automatic is still debated.
In this regard, Sato, Okada, and Toichi (2007)
Correspondence should be addressed to Giovanni Galfano, Dipartimento di Psicologia dello Sviluppo e della Socializzazione,
Università di Padova, Via Venezia 8, 35131, Padua, Italy. E-mail: giovanni.galfano@unipd.it
We would like to thank Steve Tipper and three anonymous reviewers for the many helpful suggestions on a previous version of the
paper.
#2012 The Experimental Psychology Society 1895
http://www.psypress.com/qjep http://dx.doi.org/10.1080/17470218.2012.663765
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY
2012, 65 (10), 1895–1910
Downloaded by [Universita di Padova] at 07:23 02 October 2012
have demonstrated that gaze-mediated orienting
can occur without awareness of the cue stimulus.
Automaticity is a multifaceted concept, and
Jonides (1981) has proposed and made popular
three criteria in the specific domain of attention
shifting. One such criterion is related to insensitiv-
ity to capacity demands and states that orienting
of attention can be defined as automatic if it is
relatively unaffected by a concomitant task. In
this regard, Law, Langton, and Logie (2011) have
recently shown a significant gaze-mediated orient-
ing effect independently of whether participants are
asked to perform a concomitant secondary task.
A second important criterion is related to resist-
ance to suppression and states that attention shifts
can be defined as automatic if they occur even
when they are not functional for performing the
task. One way to address this criterion has been to
make the central cue uninformative as regards to
target location. There is consistent evidence
showing that both gaze and arrows elicit reliable
attention shifts even when the target is equally
likely to appear at congruent or incongruent
locations (e.g., Eimer, 1997; Friesen & Kingstone,
1998; Tipples, 2002).
A more stringent test of the resistance to sup-
pression criterion is to make cues counterpredictive
with respect to target location (Posner, Cohen, &
Rafal, 1982). The logic underlying this manipu-
lation was to motivate participants in shifting atten-
tion in the direction opposite to that signalled by
the cue. Driver et al. (1999) showed that gaze-
mediated orienting took place even when partici-
pants were informed that gaze correctly indicated
the target location in only 20% of total trials.
Moreover, Friesen, Ristic, and Kingstone (2004)
presented participants with counterpredictive gaze
and arrow cues so that the target appeared in the
location opposite to the cued location in the
majority of trials and only rarely (8%) in the cued
location. Because counterpredictive gaze cues, but
not counterpredictive arrow cues, were found to
elicit attention shifts to the cued location, the
authors concluded that gaze-mediated orienting
can be considered as more strongly automatic
than arrow-mediated orienting, as the latter was
sensitive to the manipulation of expectancies.
However, Tipples (2008) found that when partici-
pants were presented with counterpredictive gaze
and arrows, a significant cueing effect was exhibited
for both cue types, suggesting that, even though
participants knew that the target was more likely
to appear in the opposite direction to that signalled
by the cue, they were unable to ignore the direc-
tional information provided by both cues. It is
worth reiterating that by making the cue counter-
predictive as to target location, the location in
which the target is more likely to appear is the
opposite to that indicated by the cue. This means
that, with such manipulation, participants are
somehow forced to process the directional infor-
mation provided by the cue in order to extract
information about the location of the upcoming
target stimulus. Hence, any significant attention
shifting in the cued location could result from a
voluntary (not automatic) processing of the cue,
necessary for establishing the most likely location
for the target to appear (i.e., the cue becomes task
relevant).
Other studies addressing resistance to suppres-
sion focused on paradigms investigating voluntary
oculomotor responses (e.g., Hermens & Walker,
2010; Kuhn & Benson, 2007; Kuhn & Kingstone,
2009, Ricciardelli, Bricolo, Aglioti, & Chelazzi,
2002). In these studies, gaze and arrows were task-
irrelevant distractors because participants relied on
a different cue to perform the task. For instance,
Kuhn and Benson (2007) instructed participants to
make a saccade in the direction signalled by an
instruction: For half of the participants, a red fixation
point indicated a saccade towards left and a green fix-
ation point a saccade towards right, with the reverse
association for the other half of the participants. In
addition to the presentation of the coloured fixation
point, either an arrow ora face with avertedgaze was
presented as a distractor given that the directional
information conveyed by the cues had no relation
to the instructed saccade. Results showed that par-
ticipants were slower and made more directional
errors in executing the saccade when the distractor
was incongruent with the required saccade direction,
irrespective of whether it was a gaze or an arrow. This
pattern appears to be robust, and it has been con-
firmed by other studies using a similar paradigm
1896 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
(e.g., Hermens & Walker, 2010; Kuhn & Kingstone,
2009, but see Ricciardelli et al., 2002). These studies
represent a significant advance in the assessment of
the automaticity of gaze- and arrow-mediated orient-
ing, in that participants had no need to process gaze
and arrow cues because these stimuli did not convey
any information about target location. However, par-
ticipants in these studies had to associate a specific
directional information to a colour cue and retrieve
this information each time they saw the cue. Recent
evidence suggests that arbitrary spatial associations
between nonspatial cues and spatial vectors take
time to develop and require robust training
(Guzzon, Brignani, Miniussi, & Marzi, 2010,
Experiment 3). In light of these arguments, one
cannot rule out the possibility that the distracting
effect exerted by both gaze and arrows resulted from
using cues of different strength, with distractor
stimuli signalling an explicit, unequivocal spatial
vector overcoming task-relevant cues characterized
by an arbitrary spatial meaning.
The aim of the present research was to address
resistance to suppression of gaze- and arrow-
mediated orienting, trying to shed light on the con-
troversial results obtained so far. In the experiments
presented in the next paragraphs, we relied on a
modified version of the classic covert orienting para-
digm. The key difference was that participants were
informed with 100% probability where the target
would have appeared on each trial by means of a
direction word, and they had to press a key each
time they detected the target. Thus, arrows and
gaze were distractors presented before the target,
not informative as to target location and totally irre-
levant for the task. In this way, participants were
encouraged to voluntary shift attention in advance
to the location of the upcoming target, and the
ability of gaze and arrow cues to elicit an involuntary
attention shift and to interfere with voluntary orient-
ing was tested. Unlike previous studies using coun-
terpredictive cues, participants had no need to
process gaze and arrow cues in that these stimuli
did not convey any information about target location.
In addition, unlike studies focusing on voluntary
oculomotor responses, in the present research, the
information about the location of the upcoming
target stimulus was conveyed by means of a direction
word, a self-evident cue that does not require an arti-
ficial mapping in order to be interpreted. Previous
research demonstrated that direction words are very
effective in pushing attention in the corresponding
direction, even when not informative as to target
location (Hommel, Pratt, Colzato, & Godijn,
2001). Thus, the choice of a cue (i.e., direction
words) that has been shown to elicit robust attention
shifts even when not informative about target
location represents a much stronger test of the resist-
ance to suppression criterion with respect to previous
studies. This latter feature makes the present set of
studies specifically suited to provide a more reliable
and unambiguous test concerning the automaticity
of gaze- and arrow-mediated orienting through the
combined use of cues possessing a less unbalanced
capability of pushing attention.
Experiments 1a and 2a addressed gaze-
mediated orienting, whereas Experiments 1b and
2b addressed arrow-mediated orienting. In
Experiments 1a and 1b, automaticity was tested
using a trial-by-trial manipulation: On each trial,
participants were presented with a direction word
(i.e., “left”or “right”) signalling with 100% prob-
ability the location of the upcoming target. After
the direction word, a gaze (Experiment 1a) or an
arrow (Experiment 1b) was presented at fixation
as a distractor, since they were uninformative as
to target location. In Experiments 2a and 2b, auto-
maticity was assessed using a block-by-block
manipulation: For the whole duration of a block,
the target always appeared on the left or on the
right, and participants were informed with 100%
probability about the future location of the target
before the beginning of each block. In this way,
participants did not have to change attentional
control settings depending on the word displayed
on each trial, and the spatial location of the target
stimulus was constant for a whole block of trials.
Experiment 3 was conducted to rule out possible
alternative explanations for the results of the pre-
vious studies by using a discrimination task rather
than a simple detection task.
In all the experiments, eye position was moni-
tored and recorded by means of eye-tracking equip-
ment. At the beginning of the experiment,
participants were asked to maintain fixation in the
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10) 1897
AUTOMATIC ORIENTING BY GAZE AND ARROWS
Downloaded by [Universita di Padova] at 07:23 02 October 2012
centre of the screen for the whole duration of a trial,
trying not to make any eye movement. They were
also informed that their eye movements were
recorded and that the experimenter could see
whether they were maintaining fixation or not.
This was done in order to ensure that participants
did not shift the eyes in advance towards the
location in which the target was expected. Indeed,
this behaviour would have probably resulted in a
decreased probability of processing the distractor
stimulus, because gaze and arrows would have not
been foveated. Hence, if participants had shifted
their eyes in the expected target location in
advance, it would be no surprise to find no interfer-
ence when the direction word and the distractor
stimulus have incongruent spatial vectors. In sum,
we recorded eye position in order to remove
offline trials in which participants failed to main-
tain fixation within 2° of the fixation point in case
the results did not show an interference effect (for
a similar procedure, see Christ & Abrams, 2006).
As regards general predictions, three possible out-
comes could have been expected. If both arrow and
gaze distractors were able to trigger an automatic
orienting of attention, then we should have observed
slower response times (RTs) in detecting the target
when the distractor was incongruent with target
location. Indeed, this would have meant that partici-
pants were unable to ignore the direction cue that
pushed the participants’attention in a genuinely
automatic way, thus interfering with target detection.
If only gaze was able to trigger an automatic shift of
attention, then we should have expected an interfer-
ence when participants were presented with the
gaze distractor but not with the arrow distractor.
Finally, if participants were able to suppress the
spatial meaning of gaze and arrow cues, thenno inter-
ference should have been predicted.
EXPERIMENT 1A
Method
Participants
Twenty-eight students (14 females) from the
University of Padua participated in the experiment
on a voluntary basis. Their mean age was 22.20
years. All participants had normal or corrected-
to-normal vision.
Apparatus, stimuli, and procedure
Presentation of the stimuli and registration of
responses were controlled with E-Prime 1.2 soft-
ware. Stimuli were presented on a 17-inch compu-
ter monitor, with a resolution of 1,152 ×864
pixels. As anticipated earlier, testing was conducted
on a Tobii T120 (Tobii Technology, Stockholm,
Sweden), to monitor participants’eye movements
for the whole duration of the experiment. Before
the experiment started, a standard calibration pro-
cedure was conducted. Participants were instructed
to fixate a central red dot on a grey background and
to follow this dot as it moved around the screen in
nine different positions. After successful cali-
bration, the experimental trials began.
The screen background was set to black. Each
trial began with a fixation cross of 0.82° in height
and length, remaining in the centre of the screen
for 1,000 ms (see Figure 1, Panel a). Two white-
contoured boxes were placed on the right and on
the left of the fixation point on the same horizontal
meridian at a distance of 10.8° from the fixation
point. Each box subtended a visual angle of 2.4°
in height and width. The two boxes were visible
throughout the trial. Then, the word “destra”(i.e.,
right in Italian) or “sinistra”(i.e., left in Italian)
appeared in the centre of the screen. Words
appeared in 24-point Arial bold font. After 1,000
ms, the word was replaced by a schematic face
with gaze averted leftwards or rightwards with the
same probability. The schematic face was the
same as that used by Galfano, Rusconi, and
Umiltà (2006, Experiment 3). The face subtended
a visual angle of 3.7° in width and 3.2° in height.
After either 100 ms or 1,200 ms, depending on
the stimulus onset asynchrony (SOA), a target
appeared in one of the two boxes. The target was
a white dot subtending 1° of visual angle, appearing
at a distance of 12° from the centre of the screen.
These two different SOAs were used in order to
investigate the time course of the interference
effect, if any. Indeed, it might be that the shorter
SOA did not allow participants to disengage
1898 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
attention from the location signalled by the distrac-
tor and reorient it to the location signalled by the
word on spatially incongruent trials. In contrast,
when participants had 1,200 ms between distractor
onset and target onset, they could have enough time
to reorient attention towards the location signalled
by the word on spatially incongruent trials.
Participants went through 288 trials divided into
four different blocks of 72 trials each. One third of
all the trials (i.e., 96 trials, 24 for each block) were
target-absent catch trials. There were potentially 48
data points for each relevant condition of the exper-
imental design—namely, for each SOA (100 ms
versus 1,200 ms) and congruency level (congruent
versus incongruent). Gaze direction was random-
ized, and it pointed left or right with the same
probability. The target had the same probability
of appearing on the right or on the left throughout
each block. Catch trials and experimental trials
were randomly presented, but their proportion
was constant in each block. Participants were expli-
citly told that the direction word was 100% predic-
tive of target location. They were instructed to
ignore gaze distractors and to press the spacebar
using their dominant hand as fast as possible each
time they saw the target on the screen. Finally, par-
ticipants were required to refrain from responding
and to wait for the next trial to begin when a
catch trial was displayed. If the target was shown,
and participants’response was not provided in
2,000 ms, the next trial began.
Results and discussion
In this and all the subsequent experiments, data
were collapsed across gaze direction and target
location, and a new variable called congruency
was obtained. On congruent trials, targets appeared
in locations that were congruent with gaze direc-
tion. On incongruent trials, targets appeared in
the opposite location. Trials in which an error
was committed were discarded from the analyses
(0.2%). Both trials in which participants omitted
to press the space bar in response to target appear-
ance and responses on catch trials were considered
as errors. Accuracy was not analysed given the low
percentage of errors committed by participants.
Participants failed to maintain fixation within 2°
of the fixation point on 7% of the total trials.
A 2 (congruency: congruent trials versus
incongruent trials) ×2 (SOA: 100 versus 1,200
ms) repeated measures analysis of variance
(ANOVA) was performed on mean RTs. A main
effect of congruency emerged, F(1, 27) =7.894,
Figure 1. Trial sequence for (a) Experiment 1a and (b) Experiment 1b. The word “destra”means right in Italian. A spatially incongruent trial
is illustrated in both panels. Stimuli are not drawn to scale.
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10) 1899
AUTOMATIC ORIENTING BY GAZE AND ARROWS
Downloaded by [Universita di Padova] at 07:23 02 October 2012
MSE =117.05, p=.009, showing that partici-
pants responded faster to congruent targets (M=
393 ms; SE =15) than to incongruent targets
(M=398 ms; SE =15; see Figure 2, Panel a). A
main effect of SOA emerged, F(1, 27) =22.727,
MSE =1,470.13, p,.001, probably reflecting
the classic warning effect (e.g., Sanders, 1975).
No interaction between congruency and SOA
was found, F(1, 27) =0.523, MSE =176.25.
Nonetheless, one-tailed ttests were performed to
establish whether the interference effect was signifi-
cant at each SOA. These revealed that interference
was close to significance at the 100-ms SOA, t
(27) =1.376, p=.09, and statistically significant
at the 1,200-ms SOA, t(27) =2.114, p=.02.
This pattern suggests that gaze triggered an
automatic attention shift in the signalled direction:
Indeed, participants were faster in detecting the
target on congruent trials than on incongruent
trials, probably because attention was already
engaged in the upcoming target location.
Moreover, the nonsignificant interaction between
congruency and SOA suggests that attention
shifts triggered by gaze were early rising and long
lasting.
In order to ascertain whether failures to main-
tain fixation showed a spatial bias congruent with
gaze direction, a ttest was performed on the
number of eye movements with congruency
(congruent trials versus incongruent trials) as
factor. SOA was not included as factor in this
analysis and in those of the subsequent exper-
iments, as failures to maintain fixation at the 100-
ms SOA were rare. The results showed no signifi-
cant spatial bias, t(27) =0.235.
EXPERIMENT 1B
Method
Participants
Thirty students (20 females) from the University of
Padua participated in the experiment on a voluntary
basis. Their mean age was 24.77 years. All partici-
pants had normal or corrected-to-normal vision.
None of them had taken part in the previous
experiment.
Apparatus, stimuli, and procedure
Everything was the same as that in Experiment 1a,
except for the schematic face that was replaced by
an arrow pointing to the left or to the right. The
arrow was drawn so as to occupy the same area as
the schematic face—namely, 3.7° in width and
3.2° in height, with a symmetric tale and head in
order to be comparable to the two eyes conveying
directional information (see Figure 1, Panel b).
Figure 2. Mean RTs (response times) for congruent and incongruent trials as a function of SOA (stimulus onset asynchrony) in (a) Experiment
1a and (b) Experiment 1b. Error bars indicate standard errors.
1900 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
Results and discussion
Participants committed no errors. Eye movement
data from one participant were lost due to a hard-
ware failure. Participants failed to maintain fixation
on 20.2% of the total trials.
A 2 (congruency: congruent trials versus incon-
gruent trials) ×2 (SOA: 100 ms versus 1,200 ms)
repeated measures ANOVA was performed on
mean RTs. A main effect of congruency emerged,
F(1, 29) =27.555, MSE =78.31, p,.001,
showing that participants were faster in detecting
congruent targets (M=403 ms; SE =10) than
incongruent targets (M=412 ms; SE =10; see
Figure 2, Panel b). A main effect of SOA emerged,
F(1, 29) =15.677, MSE =666.15, p,.001, prob-
ably reflecting the classic warning effect (e.g.,
Sanders, 1975). No interaction between congruency
and SOA was found, F(1, 29) =0.239, MSE =
142.04. The ttests confirmed that interference was
significant at both the 100-ms SOA, t(29) =3.808,
p,.001, and the 1,200-ms SOA, t(29) =2.531,
p=.008. The results basically mirrored those
obtained with a gaze distractor in Experiment 1a.
Indeed, participants were faster detecting targets at
congruent locations than at incongruent locations.
This finding seems to indicate that participants
shifted attention in the direction signalled by the
arrow, and, for this reason, they were slower at detect-
ing the target when it appeared in the opposite
location to that signalled by the arrow. As for
Experiment 1a, no interaction between SOA and
congruency emerged, and, thus, participants were
not able to suppress the attentional shift triggered
by the arrow irrespective of whether the target
appeared 100 or 1,200 ms after distractor onset.
In order to ascertain whether failures to keep fix-
ation showed a spatial bias congruent with arrow
direction, a ttest was performed on the number
of eye movements, with congruency as factor.
Participants made more erroneous eye movements
on spatially congruent (M=31.4; SE =7.4) than
on incongruent trials (M=26.9; SE =7.1), t
(28) =2.41, p,.05. This pattern is in line with
that reported by Kuhn and Kingstone (2009),
showing that arrow distractors tend to activate
automatic oculomotor responses.
In order to compare the RT effects obtained in
Experiments 1a and 1b, a 2 (congruency: congruent
trials versus incongruent trials) ×2 (SOA: 100 ms
versus 1,200 ms) ×2 (experiment: 1a versus 1b)
mixed-design ANOVA was conducted on mean
RTs. Neither the main effect of experiment nor
the interactions involving congruency were signifi-
cant, ps..3, demonstrating that the ability of gaze
and arrow cues to elicit an automatic attentional
shift was similar in the two experiments.
Importantly, the absence of a main effect of exper-
iment suggests that there was no significant differ-
ence in RTs between Experiments 1a and 1b.
Therefore, even though participants committed
no errors in Experiment 1b, their performance in
terms of RTs was comparable to that in
Experiment 1a, which seems to suggest that
response strategies were similar in the two
experiments.
In the next two experiments, one involving gaze
and the other arrows, a different manipulation was
used with respect to Experiments 1a and 1b.
Indeed, the information about the future location
of the target was provided using a block-by-block
manipulation. The future location of the target
was kept constant within each block (i.e., the
same spatial set was used throughout a block of
trials), and, thus, participants did not have to
update the information about the location of the
target on every trial as indicated by the direction
word. This means that participants could shift
attention in advance to a particular location
without the need to wait for the direction word.
Gaze distractors were used in Experiment 2a,
whereas arrow distractors were used in
Experiment 2b.
EXPERIMENT 2A
Method
Participants
Thirty-four students (18 females) from the
University of Padua participated in the experiment
on a voluntary basis. Their mean age was 22.18
years. All participants had normal or corrected-
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10) 1901
AUTOMATIC ORIENTING BY GAZE AND ARROWS
Downloaded by [Universita di Padova] at 07:23 02 October 2012
to-normal vision. None of them had taken part in
the previous experiments.
Apparatus, stimuli, and procedure
Apparatus and stimuli were the same as those in
Experiment 1a. Procedure was changed in that
the indication about the future location of the
target was given on a block-by-block basis. The
experiment comprised 288 trials, divided into four
blocks of 72 trials each. One third of all the trials
(i.e., 96 trials, 24 for each block) were catch trials.
Considering the whole experiment, there were
potentially 48 data points for each critical con-
dition—namely, for each SOA and congruency
level. Before beginning the task, participants were
informed about the future location of the target
(i.e., left or right) for the next 72 trials. Half of
the participants completed the blocks in this
sequence: target on the right, target on the left,
target on the right, and target on the left. The
other half of participants completed the four
blocks in the reverse order.
Results and discussion
Trials in which an error was committed were dis-
carded from the analyses (0.2%). Accuracy was
not analysed given the low percentage of errors
committed by participants. Eye movement data
from one participant were lost due to a hardware
failure. Participants failed to maintain fixation on
5.9% of the total trials.
A 2 (congruency: congruent trials versus incon-
gruent trials) ×2 (SOA: 100 ms versus 1,200 ms)
repeated measures ANOVA was performed on
mean RTs. A main effect of congruency emerged,
F(1, 33) =7.760, MSE =157.93, p=.009,
showing that participants responded faster to con-
gruent targets (M=419 ms; SE =11) than to
incongruent targets (M=425 ms; SE =11; see
Figure 3, Panel a).
A main effect of SOA emerged, F(1, 33) =
5.804, MSE =1,020.32, p=.022, probably
reflecting a warning effect (e.g., Sanders, 1975).
No interaction between congruency and SOA was
found, F(1, 33) =0.933, MSE =315.32. The t
tests showed that interference was significant only
for the 1,200-ms SOA, t(33) =2.063, p=.02,
although means for congruent (M=427 ms,
SE =13) and incongruent trials (M=430 ms,
SE =12) at the 100-ms SOA were in the expected
direction. Overall, similar to previous experiments,
participants were faster in detecting congruent
targets than incongruent targets. The results indi-
cate that, even when participants were informed
that the location of the upcoming target was con-
stant for a whole block, a gaze distractor was able
to trigger an automatic attentional shift towards
the signalled location. Moreover, such a pattern
was independent of SOA to a large extent.
Figure 3. Mean RTs (response times) for congruent and incongruent trials as a function of SOA (stimulus onset asynchrony) in (a) Experiment
2a and (b) Experiment 2b. Error bars indicate standard errors
1902 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
Participants were unable to ignore the information
provided by the distractor when the target appeared
1,200 ms after distractor onset, when they were
likely to have sufficient time to ignore the infor-
mation conveyed by the distractor and shift atten-
tion towards the upcoming target location.
In order to ascertain whether failures to keep fix-
ation showed a spatial bias congruent with gaze
direction, a ttest was performed on the number
of eye movements with congruency as factor. No
significant spatial bias was observed, t(32) =0.295.
EXPERIMENT 2B
Method
Participants
Twenty-one students (11 females) from the
University of Padua participated in the experiment
on a voluntary basis. Their mean age was 22.29
years. All participants had normal or corrected-
to-normal vision. None of them had taken part in
the previous experiments.
Apparatus, stimuli, and procedure
Apparatus and stimuli were the same as those in
Experiment 1b. Procedure was the same as that
in Experiment 2a in that the information about
the future position of the target was given to par-
ticipants on a block-by-block basis.
Results and discussion
Trials in which an error was committed were dis-
carded from analyses (0.2%). Accuracy was not ana-
lysed given the low percentage of errors committed
by participants. Participants failed to maintain fix-
ation on 4% of the total trials.
A 2 (congruency: congruent trials versus incon-
gruent trials) ×2 (SOA: 100 ms versus 1,200 ms)
repeated measures ANOVA was performed on
mean RTs. A main effect of congruency emerged,
F(1, 20) =13.280, MSE =354.99, p=.002,
showing that participants were faster in detecting
congruent targets (M=422 ms; SE =17) than
incongruent targets (M=437 ms; SE =18; see
Figure 3, Panel b).
A main effect of SOA emerged, F(1, 20) =6.322,
MSE =2,329.29, p=.021, probably reflecting a
warning effect (e.g., Sanders, 1975). No interaction
between congruency and SOA was found, F(1,
20) =2.149, MSE =272.28. The ttests confirmed
that interference was significant at both the 100-
ms SOA, t(20) =2.98, p=.003, and the 1,200-ms
SOA, t(20) =2.64, p=.008. Consistent with the
results of the previous experiments, participants
responded faster to congruent targets than to incon-
gruent targets. This suggests that arrows triggered a
shift of attention in the signalled direction even
though they were totally task irrelevant, and, thus,
on incongruent trials, participants were slower at
detecting the target because attention was engaged
to the opposite location. Importantly, this happened
even though participants were informed about the
location of the upcoming target at the beginning of
each block. Furthermore, the shift of attention trig-
gered by the distractor was early rising and long
lasting, in that it was observable both with 100-ms
and 1,200-ms SOAs.
In order to ascertain whether failures to keep fix-
ation showed a spatial bias congruent with arrow
direction, a ttest was performed on the number of
eye movements with congruency as factor. No sig-
nificant spatial bias emerged, t(20) =0.066.
In order to compare results obtained in
Experiments 2a and 2b, a 2 (congruency: congruent
trials versus incongruent trials) ×2(SOA:100ms
versus 1,200 ms) ×2 (experiment: 2a versus 2b)
mixed-design ANOVA was conducted on mean
RTs. The only significant effectinvolving experiment
was the Experiment ×Congruency interaction, F(1,
53) =4.506, MSE =232.29, p=.038, indicating
that, even though the interference effect proved to
be significant for both gaze and arrow distractors, it
was larger for arrows (Experiment 2b) than for gaze
(Experiment 2a). This finding is somewhat surpris-
ing in light of previous results showing that gaze-
mediated orienting, a presumably innate phenom-
enon (e.g., Farroni, Massaccesi, Pividori, &
Johnson, 2004), is more automatic than arrow-
driven orienting, a learned phenomenon (e.g.,
Friesen et al., 2004), and it seems to undermine the
view of eyes as a special stimulus (e.g., Friesen &
Kingstone, 1998). However, this pattern is not
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10) 1903
AUTOMATIC ORIENTING BY GAZE AND ARROWS
Downloaded by [Universita di Padova] at 07:23 02 October 2012
entirely novel, in that it has already been reported in a
sample of healthy participants used asa control group
in a study aimed to address gaze- and arrow-mediated
orienting in schizophrenic patients (Akiyama et al.,
2008). Moreover, Kuhn et al. (2011) have recently
investigated differences in oculomotor responses eli-
cited by gaze and arrow distractors between children
and adults and reported stronger interference for
arrows than for gaze. This pattern seems to be con-
sistent with the one that emerged in the analyses of
failures to maintain fixations in the present exper-
iments. Indeed, the only case in which a significant
directional bias on involuntary saccades was observed
is related to arrow distractors (Experiment 1b).
1
In
sum, it seems that the stronger interference effect
for arrows that emerged in the present research is
not an isolated case. Future studies will need to
address this issue in more detail to identify exper-
imental conditions in which arrows may prove to be
more powerful than gaze.
A possible criticism to the whole pattern emer-
ging from the experiments reported so far is that
because we used a simple detection task with a
large number of catch trials (33% of total trials), par-
ticipants may have decided to disregard the infor-
mation conveyed by the direction word (or
instruction cue), which was 100% predictive of
target location only on 66% of the total trials (i.e.,
only on target-present trials). Had participants
used this strategy, then the current pattern of
robust gaze- and arrow-mediated orienting would
not be particularly innovative with respect to pre-
vious existing evidence (e.g., Ristic et al., 2002;
Tipples, 2002). In this regard, however, our partici-
pants were not informed about the exact frequency
of catch trials, which makes it unlikely that they
deliberately chose to ignore the direction word (or
the provided instructions). Nevertheless, in order
to definitely rule out this alternative interpretation
of the data, we carried out an additional experiment
combining the procedure used in Experiments 1a
and 1b with a discrimination task in which no
catch trials were present. This manipulation made
the word cue genuinely predictive in all trials. In
addition, both gaze and arrow distractors were
used, allowing for a within-participants comparison.
EXPERIMENT 3
Method
Participants
Sixteen students (8 females) from the University of
Padua participated in the experiment on a voluntary
basis. Their mean age was 25 years. All participants
had normal or corrected-to-normal vision. None of
them had taken part in the previous experiments.
Apparatus, stimuli, and procedure
These were the same as those in Experiments 1a
and 1b (i.e., adopting a trial-by-trial manipulation),
with the following exceptions. The peripheral boxes
were placed at a distance of 4° from the fixation
point. Gaze and arrow distractors were tested in
different blocks of trials. The target was a white
letter (L or T), subtending 1.2° in height and 1°
in length. The participants were asked to perform
a letter discrimination task. They responded using
their index fingers positioned on a standard key-
board. Response buttons were arranged orthog-
onally (i.e., up versus down) to the position of the
target letters (i.e., left versus right) to minimize
effects related to stimulus–response mapping (e.
g., Galfano, Mazza, Tamè, Umiltà, & Turatto,
2008). Response button assignment was counter-
balanced across participants. The intertrial interval
was 1,000 ms. There were six blocks of 48 trials
each. Three blocks contained gaze distractors, and
three blocks contained arrow distractors.
Participants always completed three blocks with
the same distractor type before switching to the
other distractor. The order of distractor types was
counterbalanced across participants.
1
One possibility to account for this spatial bias is that, following the premotor theory of attention (e.g., Rizzolatti, Riggio, Dascola,
& Umiltà, 1987), while participants perceive the word cue and shift attention accordingly, saccadic eye movements to the attended
location are inhibited. This inhibition of saccades towards the attended location may result in the oculomotor system being vulnerable
to other cues (such as the arrow distractor) activating responses to the unattended location.
1904 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
Results and discussion
Participants failed to maintain fixation on 31.5% of
the total trials. In addition, data from one participant
were also excluded from the analyses because of dif-
ficulties in complying with the instructions of main-
taining fixation and responding as fast as possible
(mean RTs were more than 2 standard deviations
above the mean of the remaining sample).
A 2 (congruency: congruent trials versus incon-
gruent trials) ×2 (SOA: 100 ms versus 1,200
ms) ×2 (distractor type: gaze versus arrow) repeated
measures ANOVA was performed on mean RTs for
correct responses. A main effect of congruency
emerged, F(1, 14) =10.951, MSE =716.63,
p=.005, due to an overall interference effect reflect-
ing faster responses to congruent targets (M=608
ms, SE =28) than to incongruent targets (M=
625 ms, SE =30). SOA also yielded a significant
main effect, F(1, 14) =40.720, MSE =1,673.01,
p,.001, probably reflecting a warning effect
(Sanders, 1975). In line with previous experiments,
no significant Congruency ×SOA interaction was
observed, F(1, 14) =1.335, MSE =487.528.
Critically, neither the Distractor Type ×
Congruency interaction nor the triple interaction
were statistically significant, F(1, 14) =0.003,
MSE =730.894, and F(1, 14) =0.778, MSE =
793.794, respectively. The ttests confirmed that
interference was significant at both the 100-ms
SOA, t(14) =2.937, p=.005, and the 1,200-ms
SOA, t(14) =2.639, p=.009, for gaze (see Figure
4, Panel a). In the case of arrows, interference was
significant only at the 1,200-ms SOA, t(14) =
1.806, p=.04, although means for congruent
(M=653 ms, SE =36) and incongruent trials
(M=660 ms, SE =33) at the 100-ms SOA were
in the expected direction (see Figure 4, Panel b).
An ANOVA with the same factors as those
above was conducted on the percentage of incorrect
keypresses (5.06% of the total trials). No source of
variance was statistically significant. Yet, partici-
pants tended to commit more errors on incongruent
(M=5.52%, SE =1.11) than on congruent trials
(M=4.61%, SE =1.15), although the main effect
of congruency was not significant, F(1, 14) =
2.005, MSE =12.196. Importantly, this pattern
makes it unlikely that the present results can be
accounted for by any speed–accuracy trade-off.
In order to ascertain whether failures to keep fix-
ation showed a spatial bias congruent with distrac-
tor direction, separate ttests for each distractor type
were performed on the number of eye movements
with congruency as factor. No significant spatial
bias emerged for either gaze, t(14) =0.956, or
arrows, t(14) =1.144.
The RT pattern that emerged in the present
experiment confirms and extends the results
obtained in Experiments 1a and 1b, demonstrating
that our participants were not able to prevent shift-
ing attention following the spatial vector indicated
by both gaze and arrows, even when they were fully
aware that the best strategy would be to ignore
these distracting cues. The present findings
strongly support the view that the use of a high per-
centage of catch trials in the simple detection
experiments is unlikely to have influenced partici-
pants’strategies while performing the task.
Finally, these results are important in suggesting
that gaze and arrows push attention in a strongly
automatic manner, and to a comparable extent, in
a within-participants experimental design.
2
2
Although not strictly relevant to our hypotheses, in order to further test possible differences between the effects exerted by gaze
and arrow distractors, for all studies we performed additional analyses on RTs including the between-participants factor of gender (the
number of female and male participants was roughly balanced in our experiments). Bayliss, di Pellegrino, and Tipper (2005) have pro-
vided evidence suggesting that, compared to female participants, males show a reduced cueing effect for both gaze and arrows.
According to the authors, this pattern may reflect sex differences in responding to communicative signals. In brief, the only significant
interaction involving gender as factor was observed in Experiment 1a: Gender ×Congruency, F(1, 26) =5.957, p,.02. The ttests
indicated that female participants were faster in detecting congruent targets (M=420 ms, SE =25) than incongruent targets (M=
431 ms, SE =25), t(13) =3.864, p=.002. In contrast, RTs for congruent targets (M=364 ms, SE =13) and incongruent targets
(M=366 ms, SE =14) were not statistically different for male participants, t(13) =0.438. This pattern is partially consistent with
that reported by Bayliss et al. (2005), because a difference in attention shifting between female and male participants was clearly
visible at least for gaze distractors in Experiment 1a. However, it should be noted that the lack of other gender differences in the
present experiments may simply reflect a lack of power, in that Bayliss et al. (2005) tested much larger samples.
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10) 1905
AUTOMATIC ORIENTING BY GAZE AND ARROWS
Downloaded by [Universita di Padova] at 07:23 02 October 2012
GENERAL DISCUSSION
In recent years, a flourishing literature (e.g., Driver
et al., 1999; Friesen & Kingstone, 1998; Tipples,
2002) has challenged the traditional view, accord-
ing to which central cues elicit voluntary orienting
whereas peripheral cues trigger automatic orienting
of attention. On the one side, evidence has been
reported showing that gaze triggers a reflexive
shift of attention. On the other side, arrows have
also been shown to elicit automatic shifts of atten-
tion (e.g., Tipples, 2002), even though some
studies suggested that arrow-mediated orienting is
less automatic than gaze-mediated orienting (e.g.,
Friesen et al., 2004).
The aim of the present experiments was to
compare orienting of attention triggered by gaze
and arrows in an attempt to test a strong version
of the resistance to suppression criterion. In par-
ticular, in Experiments 1a and 1b, participants
were informed with 100% certainty about the
upcoming target location on a trial-by-trial basis
by means of a direction word presented at the
beginning of each trial. In Experiments 2a and
2b, the location of the target was constant for
the whole duration of a block, and participants
were informed about the target location at the
beginning of each block. Gaze (Experiments 1a
and 2a) and arrows (Experiments 1b and 2b)
were distractors that were not informative as to
target location and completely irrelevant for the
task. The aim was to test whether the distractors
were able to influence participants’allocation of
attention under conditions in which participants
were strongly motivated to ignore them. The
results consistently showed that participants were
faster in detecting the target when the location
signalled by the distractor and the upcoming
target location were congruent than when they
were incongruent. This finding suggests that
both gaze and arrow distractors triggered an auto-
matic shift of attention in the cued location:
When the target appeared in the location indi-
cated by the distractor (i.e., congruent trials), par-
ticipants were faster in detecting the target
because their attention was already set in that
location. When the target appeared in the
opposite location with respect to the directional
information provided by the distractor (i.e., incon-
gruent trials), participants were slower in that they
had to disengage attention from the incongruent
location and reorient it to the target location.
Thus, the distractors triggered a shift of attention,
which probably interfered with efficient target
detection. This effect was significant for both
gaze and arrows, thus suggesting that neither
cue can be ignored. In Experiment 3, we used a
within-participants design and combined a trial-
by-trial manipulation with a letter discrimination
task. This was done in order to rule out an
alternative account for the observed pattern that
would ascribe the presence of significant
Figure 4. Mean RTs (response times) for congruent and incongruent trials as a function of SOA (stimulus onset asynchrony) in Experiment 3
for (a) gaze distractors and (b) arrow distractors. Error bars indicate standard errors.
1906 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
interference effects to participants adopting the
strategy to ignore the direction word (or instruc-
tion cues). According to this explanation, partici-
pants might have decided not to use the
information about target location because of the
relatively high number of catch trials. In other
words, although both word and instruction cues
were 100% predictive of target location, the pres-
ence of 33% of catch trials would have resulted in
making them predictive of target location on 66%
of total trials. Results from the control exper-
iment, in which no catch trials were used, ruled
out this alternative account and confirmed the
presence of robust interference effects for both
gaze and arrow distractors.
Interestingly, in all five experiments, the inter-
ference effect was not modulated by the SOA
between distractor and target onset. Two different
SOAs were used: a very brief SOA of 100 ms and
a long SOA of 1,200 ms. The consistent lack of
interaction between congruency and SOA probably
reflects the fact that attention was automatically
pushed to the location signalled by the distractor,
and this effect was both early rising and long
lasting. Indeed, when the SOA between distractor
and target onset was 100 ms, participants probably
did not have enough time to disengage attention
from the location indicated by the distractors and
reorient it to the location in which the target was
expected to appear. Surprisingly, the interference
effect was observable also when the SOA was
1,200 ms. The ttests seem to indicate that it was
even stronger than at the 100-ms SOA. In this
case, participants had sufficient time to reorient
attention towards the location of the upcoming
target. Yet, even after 1,200 ms from distractor
onset, our results suggest that participants’atten-
tion was probably still engaged in the location sig-
nalled by the distractor. This latter pattern is not
consistent with evidence reported by Frischen and
Tipper (2004). In their study, they addressed
inhibitory after-effects of gaze-mediated orienting.
To this end, they used several types of task and
observed that, unlike the present results, RTs for
congruent and incongruent trials showed no sig-
nificant differences with a 1,200-ms SOA. One
possibility is that this discrepancy is due to the
different experimental paradigm implemented in
the current research. Future studies will address
this issue more thoroughly.
The finding that the information conveyed by
distractors interfered with the task indicates that
orienting of attention mediated by both gaze and
arrows resists suppression and can be defined as
strongly automatic. Indeed, even though partici-
pants were informed with 100% certainty about
the upcoming target location, and, thus, they
were motivated to voluntarily attend to that
location, results showed that they could not
ignore the information conveyed by the distractors.
However, it is important to point out that the
similar behavioural effect obtained for gaze and
arrows—namely, the elicitation of an automatic
attention shift—does not necessarily imply that
the underlying processes are the same for these
two cue types. Indeed, arrows are able to elicit an
automatic attention shift probably because they
are a well-learned symbol, which conveys a strong
spatial information that is reinforced every day,
for instance by means of road signs. On the con-
trary, gaze is likely to trigger an automatic attention
orienting because it may represent a special cue
characterized by a strong biological significance
given its relevance in everyday life (Ristic et al.,
2002), and, for such reasons, humans would have
developed a reflexive attention shift in response to
the view of an averted gaze that would also be sup-
ported by a dedicated neural circuit (Hietanen,
Nummenmaa, Nyman, Parkkola, & Hämäläinen,
2006; but see C. M. Tipper, Handy, Giesbrecht,
& Kingstone, 2008). Moreover, the presence of
gaze-mediated orienting of attention in both new-
borns as young as 2 days old (e.g., Farroni et al.,
2004) and several species of nonhuman primates
(e.g., Deaner & Platt, 2003; Tomasello, Call, &
Hare, 1998) suggests that this phenomenon can
be defined as innate, thus reinforcing the special
nature of gaze as compared to other cue types.
Previous studies comparing orienting of atten-
tion elicited by gaze and arrows have led to mixed
results. In the present experiments, a stronger test
of the resistance to suppression criterion was oper-
ationalized as compared to previous studies using
counterpredictive cues (e.g., Friesen et al., 2004;
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10) 1907
AUTOMATIC ORIENTING BY GAZE AND ARROWS
Downloaded by [Universita di Padova] at 07:23 02 October 2012
Tipples, 2008). Indeed, unlike counterpredictive
cues that have to be processed in order to infer
the most likely target location, in our experiments
neither gaze nor arrows were useful in order to
infer the target location. Participants were
informed with 100% certainty about the upcoming
target location by means of a direction word
(or instructions provided at the beginning of a
block of trials) and, therefore, were strongly motiv-
ated to voluntarily attend to that location.
Nonetheless, the results showed that the mere
view of gaze and arrows triggered an attention
shift in the corresponding direction.
The paradigm that we have adopted is more
similar to experiments using voluntary oculomotor
responses in which gaze and arrows are presented
as distractors (e.g., Kuhn & Benson, 2007).
However, in the present research, an important
change has been made: Indeed, in previous
studies, participants were typically instructed to
perform a saccade in a specific direction, depending
on the colour of a geometric shape (e.g., Kuhn &
Benson, 2007; Ricciardelli et al., 2002). In order
to perform this task, participants had to learn an
arbitrary association between a specific colour and
a specific spatial vector and retrieve it from
memory each time they saw the colour cue. In the
present research, we adopted a more self-evident
cue—namely, a direction word that does not
require any artificial mapping in order to be pro-
cessed. Moreover, arbitrary associations between a
symbol and spatial information take time to
develop and to become effective (Guzzon et al.,
2010), whereas direction words trigger attention
shifts even when uninformative as to target location
(Hommel et al., 2001). Thus, the fact that both
gaze and arrows interfered with attentional deploy-
ment even under conditions that were likely to
strongly encourage attending to a specific location
suggests that orienting of attention elicited by
these two signals can be defined as strongly
automatic.
It is worth noting, however, that the experimen-
tal procedure that was adopted in the present
research was much simpler than situations we face
in everyday life. Indeed, participants faced an
environment in which gaze and arrows were not
embedded in a complex context. We cannot rule
out the possibility that results would change by
adopting more ecological paradigms in which
gaze and arrows are surrounded by other meaning-
ful stimuli. Indeed, the salience of these two cues
and, consequently, their ability in pushing attention
might be influenced by the context. For instance, it
has recently been shown that when gaze and arrows
are presented in the same visual scene, gaze seems
to receive a stronger attentional priority than
arrows (Birmingham, Bischof, & Kingstone,
2009). In addition, even with a simple environment
such as that characterizing a spatial cueing para-
digm, evidence has been provided suggesting that
gaze-mediated orienting can be modulated by
social factors related to the identity of the cueing
face and the identity of the participants (e.g.,
Dalmaso, Pavan, Castelli, & Galfano, in press;
Pavan, Dalmaso, Galfano, & Castelli, 2011;
Shepherd, Deaner, & Platt, 2006), probably
because social variables deeply shape human
relations and the perception that we have of the
individuals conveying the gaze cues.
In conclusion, the present set of experiments, in
which a stronger definition of automaticity was
operationalized as compared to previous studies,
demonstrated that eye gaze is not unique in elicit-
ing a strongly automatic orienting of attention
and that arrows are also able to induce a very
similar effect.
Original manuscript received 24 May 2011
Accepted revision received 30 January 2012
First published online 19 April 2012
REFERENCES
Akiyama, T., Kato, M., Muramatsu, T., Maeda, T.,
Hara, T., & Kashima, H. (2008). Gaze-triggered
orienting is reduced in chronic schizophrenia.
Psychiatry Research,158, 287–296.
Bayliss, A. P., di Pellegrino, G., & Tipper, S. P. (2005).
Sex differences in eye gaze and symbolic cueing of
attention. Quarterly Journal of Experimental
Psychology,58A, 631–650.
1908 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
Birmingham, E., Bischof, W. F., & Kingstone, A.
(2009). Get real! Resolving the debate about equival-
ent social stimuli. Visual Cognition,17, 904–924.
Christ, S. E., & Abrams, R. A. (2006). Abrupt onsets
cannot be ignored. Psychonomic Bulletin & Review,
13, 875–880.
Dalmaso, M., Pavan, G., Castelli, L., & Galfano, G. (in
press). Social status gates social attention in humans.
Biology Letters. doi:10.1098/rsbl.2011.0881
Deaner, R. O., & Platt, M. L. (2003). Reflexive social
attention in monkeys and humans. Current Biology,
13, 1609–1613.
Driver, J., Davis, G., Ricciardelli, P., Kidd, P., Maxwell,
E., & Baron-Cohen, S. (1999). Gaze perception trig-
gers reflexive visuospatial orienting. Visual Cognition,
6, 509–540.
Eimer, M. (1997). Uninformative symbolic cues may
bias visual-spatial attention: Behavioral and electro-
physiological evidence. Biological Psychology,46,
67–71.
Farroni, T., Massaccesi, S., Pividori, D., & Johnson, M.
H. (2004). Gaze following in newborns. Infancy,5,
39–60.
Friesen, C. K., & Kingstone, A. (1998). The eyes have it!
Reflexive orienting is triggered by nonpredictive gaze.
Psychonomic Bulletin & Review,5, 490–495.
Friesen, C. K., Ristic, J., & Kingstone, A. (2004).
Attentional effects of counterpredictive gaze and
arrow cues. Journal of Experimental Psychology:
Human Perception and Performance,30, 319–329.
Frischen, A., Bayliss, A. P., & Tipper, S. P. (2007). Gaze
cueing of attention: Visual attention, social cognition,
and individual differences. Psychological Bulletin,133,
694–724.
Frischen, A., & Tipper, S. P. (2004). Orienting attention
via observed gaze shift evokes longer-term inhibitory
effects: Implications for social interactions, attention,
and memory. Journal of Experimental Psychology:
General,133, 516–533.
Galfano, G., Mazza, V., Tamè, L., Umiltà, C., &
Turatto, M. (2008). Change detection evokes a
Simon-like effect. Acta Psychologica,127, 186–196.
Galfano, G., Rusconi, E., & Umiltà, C. (2006). Number
magnitude orients attention, but not against one’s
will. Psychonomic Bulletin & Review,13, 869–874.
Galfano, G., Sarlo, M., Sassi, F., Munafò, M., Fuentes,
L. J., & Umiltà, C. (2011). Reorienting of spatial
attention in gaze cuing is reflected in N2pc. Social
Neuroscience,6, 257–269.
Guzzon, D., Brignani, D., Miniussi, C., & Marzi, C. A.
(2010). Orienting of attention with eye and arrow
cues and the effect of overtraining. Acta Psychologica,
134, 353–362.
Hermens, F., & Walker, R. (2010). Gaze and arrow dis-
tractors influence saccade trajectories similarly.
Quarterly Journal of Experimental Psychology,63,
2120–2140.
Hietanen, J. K. (1999). Does your gaze direction and
head orientation shift my visual attention?
Neuroreport,10, 3443–3447.
Hietanen, J. K., Nummenmaa, L., Nyman, M. J.,
Parkkola, R., & Hämäläinen, H. (2006). Automatic
attention orienting by social and symbolic cues acti-
vates different neural networks: An fMRI study.
NeuroImage,33, 405–413.
Hommel, B., Pratt, J., Colzato, L., & Godijn, R. (2001).
Symbolic control of visual attention. Psychological
Science,12, 360–365.
Jonides, J. (1981). Voluntary versus automatic control
over the mind’s eye movement. In J. B. Long &
A. D. Baddeley (Eds.), Attention and Performance
IX (pp. 187–203). Hillsdale, NJ: Lawrence Erlbaum
Associates.
Kuhn, G., & Benson, V. (2007). The influence of eye-
gaze and arrow pointing distractor cues on voluntary
eye movements. Perception & Psychophysics,69,
966–971.
Kuhn, G., & Kingstone, A. (2009). Look away! Eyes
and arrows engage oculomotor responses automati-
cally. Attention, Perception, & Psychophysics,71,
314–327.
Kuhn, G., Tewson, L., Morpurgo, L., Freebody, S. F.,
Musil, A. S., & Leekam, S. R. (2011).
Developmental changes in the control of saccadic
eye movements in response to directional eye gaze
and arrows. Quarterly Journal of Experimental
Psychology,64, 1919–1929.
Langton, S. R. H., & Bruce, V. (1999). Reflexive visual
orienting in response to the social attention of others.
Visual Cognition,6, 541–567.
Law, A. S., Langton, S. R. H., & Logie, R. H. (2011).
Assessing the impact of verbal and visuospatial
working memory load on eye-gaze cueing. Visual
Cognition,18, 1420–1438.
Pavan, G., Dalmaso, M., Galfano, G., & Castelli, L.
(2011). Racial group membership is associated to
gaze-mediated orienting in Italy. PLoS One,6,
e25608.
Posner, M. I., & Cohen, Y. (1984). Components of
visual orienting. In H. Bouma & D. Bouwhuis G.
(Eds.), Attention and Performance X (pp. 531–556).
Hillsdale, NJ: Lawrence Erlbaum Associates.
THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10) 1909
AUTOMATIC ORIENTING BY GAZE AND ARROWS
Downloaded by [Universita di Padova] at 07:23 02 October 2012
Posner, M. I., Cohen, Y., & Rafal, R. D. (1982). Neural
systems control of spatial orienting. Philosophical
Transactions of the Royal Society London B,298, 187–198.
Ricciardelli, P., Bricolo, E., Aglioti, S. M., & Chelazzi, L.
(2002). My eyes want to look where your eyes are
looking: Exploring the tendency to imitate another
individual’s gaze. NeuroReport,13, 2259–2264.
Ristic, J., Friesen, C. K., & Kingstone, A. (2002). Are
eyes special? It depends on how you look at it.
Psychonomic Bulletin & Review,9, 507–513.
Rizzolatti, G., Riggio, L., Dascola, I., & Umiltà, C.
(1987). Reorienting attention across the horizontal
and vertical meridians: Evidence in favor of a premotor
theory of attention. Neuropsychologia,25, 31–40.
Sanders, A. F. (1975). The foreperiod effect revisited.
Quarterly Journal of Experimental Psychology,27A,
591–598.
Sato, W., Okada, T., & Toichi, M. (2007). Attentional
shift by gaze is triggered without awareness.
Experimental Brain Research,183, 87–94.
Shepherd, S. V., Deaner, R. O., & Platt, M. L. (2006).
Social status gates social attention in monkeys.
Current Biology,16, R119–R120.
Tipper, C. M., Handy, T. C., Giesbrecht, B., &
Kingstone, A. (2008). Brain responses to biological
relevance. Journal of Cognitive Neuroscience,20,
879–891.
Tipper, S. P. (2010). From observation to action simu-
lation: The role of attention, eye-gaze, emotion, and
body state. Quarterly Journal of Experimental
Psychology,63, 2081–2105.
Tipples, J. (2002). Eye gaze is not unique: Automatic
orienting in response to uninformative arrows.
Psychonomic Bulletin & Review,9, 314–318.
Tipples, J. (2008). Orienting to counterpredictive gaze
and arrow cues. Perception & Psychophysics,70,
77–87.
Tomasello, M., Call, J., & Hare, B. (1998). Five primate
species follow the visual gaze of conspecifics. Animal
Behaviour,55, 1063–1069.
1910 THE QUARTERLY JOURNAL OF EXPERIMENTAL PSYCHOLOGY, 2012, 65 (10)
GALFANO ET AL.
Downloaded by [Universita di Padova] at 07:23 02 October 2012
