Content uploaded by Oksana Tsaregorodtseva
Author content
All content in this area was uploaded by Oksana Tsaregorodtseva on Mar 05, 2021
Content may be subject to copyright.
SHORT REPORT
Remember down, look down, read up: Does a word modulate eye
trajectory away from remembered location?
Armina Janyan
1,3
•Ivan Vankov
2
•Oksana Tsaregorodtseva
3
•Alex Miklashevsky
3
Published online: 1 August 2015
ÓMarta Olivetti Belardinelli and Springer-Verlag Berlin Heidelberg 2015
Abstract Previous studies show that eye movement tra-
jectory curves away from a remembered visual location if a
saccade needs to be made in the same direction as the
location. Data suggest that part of the process of main-
taining the location in working memory is the mental
simulation of that location, so that the oculomotor system
treats the remembered location as a real one. Other
research suggests that word meaning may also behave like
a ‘real object’ in space. The current study aimed to com-
bine the two streams of research examining the effect of
word meaning on the memory of a dot location. The results
of two experiments showed that word meaning for ‘up’
(but not ‘down’) modulated both eye movement trajectory
and location recognition time. Thus, mental simulation of
task-irrelevant space-related word meaning affected both
earlier stages of memory processes (maintenance of the
location in the working memory) and later ones (location
recognition).
Keywords Visual working memory Mental simulation
Spatial location
Introduction
Previous research (Theeuwes et al. 2005) indicates that eye
trajectory deviates away from a remembered visual loca-
tion if a saccade needs to be made in the same direction as
that location. Data suggest that part of the process of
maintaining the location in working memory (WM) is the
mental simulation of that location so that the oculomotor
system treats the remembered location as a real one. Other
stream of research suggests that word meaning may also
behave like a ‘real object’ in space (Pecher et al. 2010),
when the ‘spatial’ meaning of a word is task relevant.
Participants usually respond faster to names of objects
associated with a typically higher position in vertical space
(e.g., sun) if the word appears in the upper part of the
screen compared to the lower part. The interaction between
spatial word meaning and spatial location on the screen is
interpreted in terms of sensory-motor simulation of a per-
ceived word. That is, reading the word sun would activate
sensory-motor information and would prompt the mental
simulation of a ‘looking up’ action that in turn would
influence other sensory-motor processes. For instance,
previous research suggests that participants make faster
saccades to the word sun, when located in the upper part of
the screen compared to the lower part (Dudschig et al.
2013). Moreover, it seems that even when irrelevant to the
task, word meaning can influence the time of saccade ini-
tiation: Saccades were faster when their direction (left/
right, above/below) from the centrally presented word (left/
right, above/below) were compatible than when they were
incompatible (Khalid and Ansorge 2013).
Thus, on one hand, data suggest that keeping a dot
location in the WM forces participants to treat the empty
space as an actual spatial obstacle and to ‘avoid’ it while
generating a saccade that deviates from the remembered
&Armina Janyan
ajanyan@cogs.nbu.bg
1
Research Center for Cognitive Science, New Bulgarian
University, 21 Montevideo Blvd., Room 401, 1618 Sofia,
Bulgaria
2
Department of Cognitive Science and Psychology, New
Bulgarian University, Sofia, Bulgaria
3
National Research Tomsk State University, Tomsk, Russia
123
Cogn Process (2015) 16 (Suppl 1):S259–S263
DOI 10.1007/s10339-015-0718-5
location. On the other hand, reaction time studies indicate a
meaning-location compatibility effect mostly for task-rel-
evant target words, even though eye tracking studies find a
compatibility effect with meaning-irrelevant tasks as well.
A question arises, then, whether the simulation of space-
related conceptual aspects of a task-irrelevant word is
powerful enough to modify eye deviation from a remem-
bered location.
Experiment 1: Eye tracking
The aim of the experiment was to examine the effect of
word meaning on the memory fora location during location
maintaining in the WM. We slightly modified Theeuwes
et al. (2005) study on memorized dot location for the
current experiment. Participants were asked to memorize
the location of a dot and then to execute a saccade
depending on the direction of a presented arrow (up/down).
Within the dot the word ‘up’ or ‘down’ was always present.
Method
Participants
Forty university students (nine males; mean age =18.6,
SD =1.6) participated voluntarily in the experiment. All
of them were native speakers of Russian, with normal or
corrected to normal vision.
Procedure and design
Participants were seated 70 cm from the computer screen
with the head positioned on a chin-rest. A 500-Hz SMI eye
tracking system controlled stimuli presentation and recor-
ded eye movements. The procedure was adapted from
Theeuwes et al. (2005). Participants had to fixate on a
centrally presented cross for 1 s (Fig. 1). Then, a dot
appeared for 500 ms filled with a word (up or down in
Russian). The dot, 1.55°in diameter, appeared in one of the
four lateral cells (left/right top/bottom) of the screen divi-
ded into 3 93 grid (grids were centered at x =±4.89°,
y=±3.18°from the central cross). Participants had to
remember the location of the dot. Then, a centrally pre-
sented cross appeared again and stayed on the screen until
the eye tracker system recorded a one-second duration gaze
onto the cross. After that an arrow pointing up or down was
presented for 300 ms. Participants had to execute a saccade
in line with the direction of the arrow. The saccade had to
be executed during 1 s to a cross-positioned ±9.55°ver-
tically from the central cross (cf. Figure 1). After the sac-
cade a recognition task followed. A dot was presented
either at the exactly same position as the previous one or at
a slightly different position. The word within the dot was
not changed. Participants had to press a corresponding
button (same/different) on a keyboard. A practice session
contained eight trials. Each participant was presented with
128 pseudo-randomized trials so that there were no more
than two consecutive trials of the same condition. The
experiment took approximately 15 min.
The design applied was dot position (up vs. down) 9-
word (up vs. down) 9arrow direction (up vs. down). Two
deviation measures (maximal angle deviation and average
angle deviation) served as dependent variables.
Results and discussion
Analyses were run on data of participants with more than
70 % response accuracy and more than 65 % fixation
capture. This resulted in overall analysis of data of 26
participants (six males, mean age =18.5, SD =1.6). Of
them, 15 % of data were excluded due to the failure in
fixation registration.
For each trial, maximal angle deviation and average
angle deviation were calculated (Van der Stigchel et al.
2006). These measures reflected how much the eye
movement trajectory deviated from the line connecting the
saccade starting and the saccade ending position. The
deviation was coded as a positive number if the eye
movements deviated away from the target location and it
was negative if the eye movements deviated toward the
target location.
Repeated measures analyses of variance (rANOVA)
were applied to subject means with word, dot position and
arrow direction as within-subject variables. Bonferroni post
hoc test was applied where needed.
Maximal angle deviation
The analyses revealed no main effect (all p
s
[0.07). A
significant two-way interaction between word and dot
position (Fig. 2) was obtained [F(1, 27) =7.78, p\0.01,
Fig. 1 Schematic
representation of a trial, not to
scale
S260 Cogn Process (2015) 16 (Suppl 1):S259–S263
123
g
p
2
=0.22]. The interaction suggested that the word mod-
ified eye trajectories only when the word was ‘up.’ When
the word ‘up’ coincided with the ‘up’ dot location, the eye
trajectory deviated away from the remembered location
compared to ‘down’ dot location. The word ‘down’ did not
influence eye deviation from the remembered location. A
position by arrow interaction was also significant [F(1,
27) =7.12, p\0.05, g
p
2
=0.21; Fig. 3]. It showed the
same pattern as the previous interaction with regard to
arrow direction: Saccades deviated away from the
remembered location only when the arrow required a sac-
cade to the top of the computer display. Thus, the
experiment replicated the remembered location-saccade
direction compatibility effect reported by Theeuwes et al.
(2005). However, due to the differences in designs, we
observed the effect mostly pronounced in the ‘up’ direc-
tion, a detail that is hidden in the experiments that collapse
two ‘up’ and ‘down’ variables into one (compatibility).
All other interactions were not significant (all p
s
[0.1).
Average angle deviation
The analysis partially confirmed the results obtained on
maximal angle deviation. No main effects were obtained (all
p
s
[0.2). The position by arrow interaction [F(1,
27) =2.76, p=0.11, g
p
2
=0.09] as well the interaction
between word and arrow direction [F(1, 27) =4.02,
p=0.06, g
p
2
=0.13] did not reach the significance level.
Still, we obtained a significant interaction between word and
dot position [F(1, 27) =7.19, p\0.05, g
p
2
=0.21] reveal-
ing the same pattern as before (Fig. 4): Saccades deviated
from the remembered position only if the dot was in the top
display position with a written word ‘up’ in it. The three-way
interaction was highly insignificant (p[0.4).
The results indicated that the process of maintaining a dot
location can be influenced by task-irrelevant word meaning
that directs the participants’ attention to the specific part in
space and modifies saccade deviation away from the
remembered location. However, this is valid only for word
‘up’ but not for word ‘down’. It may be argued that word ‘up’
activates the upper space strengthening the effect of mental
simulation of dot location in the visual WM.
The next experiment tests the effect of word meaning on
dot location memory during location recognition.
down up
Word
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
Maximal angle deviation
**
Position
down
up
Fig. 2 Word by Dot position interaction, maximal angle devia-
tion. Vertical bars represent standard errors. **p\0.01
down up
Arrow
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
Maximal angle deviation
*
Position
down
up
Fig. 3 Arrow by Dot position interaction, maximal angle devia-
tion. Vertical bars represent standard errors. *p\0.05
down up
Word
-0.3
-0.2
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Average angle deviation
*
Position
down
up
Fig. 4 Word by Dot position interaction, average angle devia-
tion. Vertical bars represent standard errors. *p\0.05
Cogn Process (2015) 16 (Suppl 1):S259–S263 S261
123
Experiment 2: Reaction time
We constructed a reaction time study to fully replicate the
procedure of the eye tracking study except for the saccade
initiation task. The responses during location recognition
were gathered and analyzed.
Method
Participants
Twenty-nine participants (six males, mean age =19.2,
SD =2) participated voluntarily in the experiment. All of
them were native speakers of Russian, with normal or
corrected to normal vision.
Procedure and design
The procedure and design were exactly the same as in the
eye tracking experiment except that no saccade execution
was required after the arrow (cf. Figure 1). Participants had
to remember the dot location and then to recognize it by
pressing a corresponding button (same/different) on a
keyboard. Reaction time of dot position recognition served
as a dependent variable. Stimulus presentation as well as
recording of response time and accuracy of memorized
location recognition was controlled by E-prime program
(Schneider et al. 2002).
Results and discussion
Data of four participants were excluded from the analysis
due to low accuracy (\70 %). The analysis was based on
data of 25 participants (five males, mean age =18.9,
SD =1.7).
Prior to the RT analysis, erroneous responses (9.8 %)
and response times lying more than ±2 standard deviations
from the RT mean per condition were excluded (5 %).
rANOVA obtained a marginally significant main effect of
dot position [F(1, 24) =3.97, p\0.06, g
p
2
=0.14] that
showed faster RT in ‘up’ position than in ‘down’ position
(see Table 1for descriptive statistics). The arrow direction
main effect [F(1, 24) =4.52, p\0.05, g
p
2
=0.16] sug-
gested overall faster RT in the ‘up’ direction. The effect of
word was not significant (p[0.3). The word by dot
position interaction [F(1, 24) =20.43, p\0.001,
g
p
2
=0.46] is shown in Fig. 5. Overall, it confirmed the eye
tracking data in that the type of word influenced time of dot
location recognition only in ‘up’ word condition. Location
recognition was faster if the word meaning was compatible
with the dot position (‘up’) in comparison with the
incompatible condition (‘down’ position). Other interac-
tions were not significant (all p
s
[0.1).
Conclusion
The aim of the study was to examine the effect of word
meaning on the memory of a dot location during two
processing stages: location maintaining in the WM (‘Ex-
periment 1’ section) and location recognition (‘Experiment
2’ section). The results point to several important theoret-
ical and methodological considerations. Firstly, data sug-
gested that visual working memory for location, spatial
attention and language do not just go ‘hand in hand’ but
they interact on an oculomotor level. It suggested an
automatic impact of language on saccade programming and
execution; moreover, the impact was selective, observed
only for the ‘up’ words. The ‘up’ results are in line with
previous data reporting differential spatial bias, locus of
attention and type of processing in the upper and lower
space. Global processing, motion perception and color
discrimination advantages are observed in the lower space,
while local processing, object categorization/identification
and visual search prevail in upper space (cf. Thomas and
Elias 2011). Importantly, the current ‘up’ effects were
down up
Word
680
700
720
740
760
780
800
820
840
860
RT, ms
***
**
Position
down
up
Fig. 5 Word by Dot position interaction, RT. Vertical bars represent
standard errors. **p\0.01; ***p\0.001
Table 1 Means and standard deviations (in parentheses) per condi-
tion in ms, subject means, Experiment 2
Word Position up Position down
Arrow up Arrow down Arrow up Arrow down
Up 755 (154) 750 (134) 799 (146) 808 (155)
Down 756 (138) 798 (147) 751 (128) 765 (115)
S262 Cogn Process (2015) 16 (Suppl 1):S259–S263
123
stable across different measures (eye movements and
reaction time) and levels of processing (location main-
taining and location recognition). Secondly, our results
indicate that merely an activation of a part of space by a
word makes the memorized object in the space more tan-
gible, which is evident by eye curvatures. Thirdly, it
appears that meaning-location compatibility (e.g., Khalid
and Ansorge 2013) or the location-saccade direction
compatibility (Theeuwes et al. 2005) effects should not be
generalized over all spatial directions but rather tested
separately.
Overall, the results demonstrate that word meaning can
influence the process of mental simulation of remembered
location during location maintenance in visual WM as well
as during location recognition.
Acknowledgments We thank an anonymous reviewer for useful
comments and suggestions. This research was partially supported by
Tomsk State University Academic D.I. Mendeleev Fund Program,
Grant No. 8.1.37.2015.
References
Dudschig C, Souman J, Lachmair M, de la Vega I, Kaup B (2013)
Reading ‘‘sun’’ and looking up: the influence of language on
saccadic eye movements in the vertical dimension. PLoS One
8(2):e56872
Khalid S, Ansorge U (2013) The simon effect of spatial words in eye
movements: comparison of vertical and horizontal effects and of
eye and finger responses. Vis Res 86:6–14
Pecher D, Van Dantzig S, Boot I, Zanolie K, Huber DE (2010)
Congruency between word position and meaning is caused by
task-induced spatial attention. Front Psychol 1:30
Schneider W, Eschman A, Zuccolotto A (2002) E-Prime reference
guide. Psychology Software Tools Incorporated, Pittsburgh
Theeuwes J, Olivers CN, Chizk CL (2005) Remembering a location
makes the eyes curve away. Psychol Sci 16(3):196–199
Thomas NA, Elias LJ (2011) Upper and lower visual field differences
in perceptual asymmetries. Brain Res 1387:108–115
Van der Stigchel S, Meeter M, Theeuwes J (2006) Eye movement
trajectories and what they tell us. Neurosci Biobehav Rev
30(5):666–679
Cogn Process (2015) 16 (Suppl 1):S259–S263 S263
123
