Spanish - Eye Tracking - Reading - Information Structure

ORIGINAL RESEARCH
published: 06 April 2021

doi: 10.3389/fpsyg.2021.629724
Information Structure and Word

Order Canonicity in the
Comprehension of Spanish Texts: An
Eye-Tracking Study
Carolina A. Gattei 1,2,3*, Luis A. París 4 and Diego E. Shalom 1
1
Instituto de Física de Buenos Aires (CONICET), Universidad de Buenos Aires, Buenos Aires, Argentina, 2 Laboratorio de
Neurociencia, Universidad Torcuato Di Tella, Buenos Aires, Argentina, 3 Facultad de Filosofía y Letras, Pontificia Universidad
Católica Argentina, Buenos Aires, Argentina, 4 Grupo de Lingüística y Neurobiología del Lenguaje, Instituto de Ciencias
Humanas Sociales y Ambientales (CONICET), Mendoza, Argentina
Word order alternation has been described as one of the most productive information
structure markers and discourse organizers across languages. Psycholinguistic evidence
has shown that word order is a crucial cue for argument interpretation. Previous studies
about Spanish sentence comprehension have shown greater difficulty to parse sentences
that present a word order that does not respect the order of participants of the verb’s
lexico-semantic structure, irrespective to whether the sentences follow the canonical
word order of the language or not. This difficulty has been accounted as the cognitive cost
Edited by: related to the miscomputation of prominence status of the argument that precedes the
Valentina Cuccio,
University of Messina, Italy
verb. Nonetheless, the authors only analyzed the use of alternative word orders in isolated
Reviewed by:
sentences, leaving aside the pragmatic motivation of word order alternation. By means
Masatoshi Koizumi, of an eye-tracking task, the current study provides further evidence about the role of
Tohoku University, Japan information structure for the comprehension of sentences with alternative word order and
Tao Gong,
Educational Testing Service, verb type, and sheds light on the interaction between syntax, semantics and pragmatics.
United States We analyzed both “early” and “late” eye-movement measures as well as accuracy and
*Correspondence: response times to comprehension questions. Results showed an overall influence of
Carolina A. Gattei
[email protected]
information structure reflected in a modulation of late eye-movement measures as well
as offline measures like total reading time and questions response time. However, effects
Specialty section: related to the miscomputation of prominence status did not fade away when sentences
This article was submitted to
were preceded by a context that led to non-canonical word order of constituents,
Language Sciences,
a section of the journal showing that prominence computation is a core mechanism for argument interpretation,
Frontiers in Psychology even in sentences preceded by context.
Received: 15 November 2020
Keywords: information structure, word order, eye-tracking, text comprehension, prominence, psych verbs
Accepted: 22 February 2021
Published: 06 April 2021
Citation:
Gattei CA, París LA and Shalom DE
1. INTRODUCTION
(2021) Information Structure and
Word Order Canonicity in the
Word order alternation is a frequent feature in many languages across the world. Several works
Comprehension of Spanish Texts: An have tried to explain the psycholinguistic principles that govern comprehension of alternative
Eye-Tracking Study. word orders. Based on theoretical accounts of word order alternation or “scrambling,” many of
Front. Psychol. 12:629724. these studies assume the existence of a particular canonical word order for each language (e.g.,
doi: 10.3389/fpsyg.2021.629724 SVO for English, SOV for German, etc.), and alternative orders derived from it (Comrie, 1989).
Frontiers in Psychology | www.frontiersin.org 1 April 2021 | Volume 12 | Article 629724

Gattei et al. Information Structure and Word Order
Experimental evidence suggests that alternative word orders change from canonical SVO word order to a non-canonical one
are more difficult to process than canonical ones, as reflected (Zubizarreta, 1998) as it may be seen in (1):
by longer reading times, response times and lower accuracy
rates (Hyönä and Hujanen, 1997; Bader and Meng, 1999; (1) ¿Quién le gritó a Ana? [A Ana]
Kamide and Mitchell, 1999). Studies about the role of word Who CL-DAT yelled to Ana-DAT ? [To Ana-DAT]
order for incremental comprehension have also shown that le gritó María.
word order alternation is a relevant cue for lexico-semantic CL-DAT yelled María-NOM
argument interpretation and posterior realization of syntax-to-
semantics linking. In other words, incremental processing of “Who yelled at Ana? María yelled (at Ana=.”
word order features are useful to predict “who does what to
Typologically speaking, Spanish is considered to be a flexible
whom” in a given event (see Bader and Bayer, 2006; Bornkessel
language regarding both the possibility of alternating word order
and Schlesewsky, 2006, for two different reviews on this issue).
and the lack of constraints about the syntactic positions in which
For instance, a Spanish cloze task has shown that while the
focus can potentially be assigned (Van Valin, 1999; Belloro, 2012).
appearance of a nominative-marked argument in first position
Hence, the same question posited in (1) may present a response
leads readers to expect an activity verb, the appearance of a
in which new (focused) information takes place in first position,
dative-marked argument in first position leads them to expect
as shown in (2).
an Object Experiencer psychological verb (heareafter ObjExp
psych verb, Gattei et al., 2015b, Experiment 2). The violation (2) ¿Quién le gritó a Ana? María le
of these expectations generates higher error rates and response Who CL-DAT yelled to Ana-DAT ? María-NOM CL-DAT
times, longer reading times and amount of regressions to
gritó a Ana.
previous regions (Gattei et al., 2015a, 2017), and differential
yelled to Ana-DAT
neural correlates (Gattei et al., 2015b), even in the canonical
word order of the language. The interpretive function of word “Who yelled at Ana? María yelled at Ana.”
order has also been evaluated in a spectrum of languages with
different degrees of complexity regarding morphological case In this example, the response to the question may be
marking, such as German (Bornkessel et al., 2003, 2005), Italian interpreted as narrow focus, in the sense that it was María, and
(Dröge et al., 2014), and Chinese (Wang et al., 2012) with not Juan, for instance, who yelled at Ana. Thus, the appearance
very similar and robust results. Bornkessel and Schlesewsky of new information in first position may modify the way speakers
(2006) suggest that word order—as well as case marking, interpret the response. From a psycholinguistic point of view,
animacy, and definiteness—are key linguistic features for the not many studies have addressed the role of word order for
computation of argument prominence, which comprises the discourse on-going interpretation. In a study about processing of
hierarchical relation among arguments in a sentence (Lamers declarative sentences in Finnish with non-canonical information
and De Swart, 2012). The evidence suggests that the human structure Kaiser and Trueswell (2004) argue that readers may
sentence parser tends to interpret the first argument of a sentence need additional presuppositions in order to understand isolated
as the most “Actor-like” possible according to the prominence sentences with a non-canonical word order. Hence, showing the
status provided by those features. This proposal suggests that right discourse setting for this type of sentences should facilitate
speakers tend to compute arguments prominence status by comprehension. The authors showed that the presentation of
following a more-to-less prominent order, this is, following the a referent providing new information in first position entailed
stipulated order of arguments in the lexico-semantic structure longer reading times irrespective of word order (SVO vs. OVS),
of verbs. Hence, a Spanish animate, nominative-marked, definite and that overall, sentences with non-canonical word order (OVS)
argument in first position will most likely be the Actor of an were more difficult to understand that sentences that followed
activity event, and an animate, dative-marked, definite argument the canonical word order of the language. This means that the
in first position will most likely be the Experiencer of a presentation of a supportive discourse context partially alleviates
psych state. the usual difficulty associated to a non-canonical construction.
However, an aspect that has not been taken into account by In a series of Event-Related Potentials (ERPs) experiments
most studies that address word order alternation in sentence in German that manipulated position of the referents and
comprehension is that the appearance of non-canonical word givenness, Schumacher and Hung (2012) showed that new
order is not arbitrary but rather motivated by discursive factors inferred information in sentence-medial positions engender a
like, for instance, if a referent has been previously introduced or if Late Positivity when compared to new given information. This
it is part of a referent mentioned before (Givón, 1984; Lambrecht, difference does not take place when the constituents are in
1994; Birner and Ward, 1998). Along with prosody, word order is sentence-initial position. The authors claim that “information
considered one of the key information structure markers across presented in sentence-initial position is treated differently than
languages. The way in which given and new information is information in other positions during both language processing,”
conveyed can modulate pragmatic interpretation by stipulating and the construction of discourse representation structure.
the status of constituents as discourse topic and focus. For Burmester et al. (2014) also showed that topic-first order
instance, when unmarked, Spanish favors given information in eases OVS sentence processing in German-speaking adults, as
the left-most position of the sentence, even when that means to evidenced by an offline comprehensibility judgment task and a

late positivity effect at the ERP mean voltage when comparing (6) A Ana le encanta María.
sentences preceded by a neutral context and those preceded by a To Ana CL-DAT loves María-NOM
topicalized context.
“Ana loves María”
The present study seeks to go a step forward and to evaluate
how the pragmatic use of word order alternation interacts with
its use as a cue for arguments prominence computation. In Altough sentence (6) carries a non-canonical word order, it
other words, if prominence is considered a hierarchy composed reflects the canonical order of arguments established by its lexico-
by other independent hierarchies (e.g., animacy features are semantic structure, as exemplified in (7), in which the verb’s left-
independent from case marking and word order), and in a most argument (“x”) is associated to an Experiencer of a state
particular sentence these hierarchies may conflict with each other predicate, and “y” is associated to the Theme that generates this
(e.g., the innanimate argument bares nominative case, Chow state (Van Valin, 2005, p. 45).
and Phillips, 2013), it is worth exploring when the language
word order (SVO) is incongruent with the canonical word order (7) encantar’(x,y)
stipulated by the lexico-semantic structure of the verb (SVO for
activity verbs and OVS for ObjExp psych verbs) and that of the This subclass of psych verbs has become relevant for
rhematic hierarchy (“given” referents precede “new” ones). understanding how the sentence processor uses both syntactic
The paper is organized as follows: We first present a brief and lexico-semantic information in order to predict the thematic
description of Spanish word order alternation, with special structure of a particular event. In a series of studies run
emphasis on Object Experiencer Psych Verbs (hereafter ObjExp in this language, when presented with sentences like (4)
psych verbs) and stipulate the hypotheses and predictions related and (6), readers found higher difficulty to integrate the verb
to the processing of these sentences when embedded in context. and the second argument of the sentence in trials with
We then present an eye-tracking study addressing these issues. activity verbs than in sentences with ObjExp psych verbs.
Finally, we discuss the results of the current experiment under The opposite pattern was found for subject-initial sentences,
the light of previous findings. showing longer reading times (Gattei et al., 2015a) and higher
amount of regressions (Gattei et al., 2017) to previous regions
1.1. Word Order Alternation in Spanish when the sentence included an ObjExp psych verb than an
Spanish is rather flexible in terms of word order, although it is activity verb.
argued to be a Subject-Verb-Object (SVO) language (Contreras, This pattern of results suggests that readers are not only
1976; Suñer, 1982; Ocampo, 1995). Take for instance sentences in guided by word order canonicity in order to interpret sentences,
(3) and (4) but that they use word order together with a semantic principle
that stipulates that the first argument will take the most
(3) María le responde a Ana. prominent status possible to form predictions about the type of
MaríaNOM CL-DAT responds to Ana-DAT thematic structure that the event will carry and assign a thematic
“María responds to Ana” role to the preverbal argument accordingly (Bornkessel et al.,
2005; Wolff et al., 2007; Haupt et al., 2008). The appearance of
a verb that required a correction of this assumption resulted
(4) A Ana le responde María. in longer reading times in the regions that comprised the
To Ana CL-DAT responds María-NOM second argument of the sentence. Furthermore, when asked
“who did/felt what for whom” after reading each sentence,
“María responds to Ana” accuracy rates were lower and response time longer when
the sentences arguments did not reflect the canonical order
The verb in these examples expresses the same type of event, of arguments of their lexico-semantic structure, showing that
in which an Actor (“María”) carries out an activity (“to respond”) the effects of not respecting the order established by the
that affects another participant (“Ana”). The main difference lexico-semantic structure of the verb are so robust that can
between both sentences is that, apart from showing the canonical persist even once all the processes of linguistic integration have
word order of the language, sentence (3) shows a canonical order been completed.
of its arguments, with an Actor preceding the affected participant
or “Undergoer” (Foley and Van Valin, 1984). Sentence (4), on the 1.2. Hypotheses and Predictions
contrary, exhibit both a non-canonical word order and a non- By means of an eye-tracking reading task we aim at weighing
canonical arguments order, with the Undergoer preceding the the relative processing load imposed by the violation of two
Actor. types of linguistic hierarchies related to word order alternation:
The same morphological case marking is applied to arguments the rhematic hierarchy—given referents precede new referents
in sentences with ObjExp psych verbs, as shown in (5) and (6) (Contreras, 1976)—and that related to arguments’ prominence—
“the Actor precedes the Undergoer” (Van Valin and LaPolla,
(5) María le encanta a Ana.
1997).
MaríaNOM CL-DAT loves to Ana-DAT
We propose to replicate the findings from the study of Kaiser
“Ana loves María” and Trueswell (2004), but using two different verb types (i.e.,

activity verbs and ObjExp psych verbs), as in Gattei et al. (2015a, sentences with Activity verbs) irrespective of whether they are
2017). preceded by a supportive or unsupportive context. Following
Following the results of Kaiser and Trueswell (2004) and Gattei et al. (2017), effects of prominence miscomputation
basing our hypotheses on the assumption that word order should take place at late eye-movement measures at later
alternation is motivated by discursive factors (Givón, 1984; regions of the sentence (i.e., verb region onward for reading
Lambrecht, 1994; Birner and Ward, 1998) we expect that measures, and at initial regions of the sentence and verb for
overall, the appearance of an adequate context facilitates sentence regression measures).
comprehension. In the current study, context adequacy is
provided by the pragmatic status of referents. This means that an
adequate context will lead to a sentence with a “given” referent 2. MATERIALS AND METHODS
in first position and a “new” referent as second argument, giving
rise to a canonical rhematic hierarchy. Conversely, an inadequate We designed a text reading task using the eye-tracking
context will give rise to a “new” referent in first position and a method in order to study the interaction between word
non-canonical information structure. order, verb type and information structure. This technique
We also expect that all effects related to the interaction allows to register with great temporal precision what eyes do
between syntactic, semantic and pragmatic factors are reflected during naturalistic reading, and what strategies readers use
in late eye-movement measures, since they are assumed to reflect in order to overcome cognitive difficulties that could arise
later parsing stages (see Clifton et al., 2007; Vasishth et al., 2013, from linguistic complexity (also see Just and Carpenter, 1980;
for a review on this discussion). Just et al., 1982, for a discussion on the advantages of this
In relation to the interaction between both prominence and paradigm).
rhematic hierarchies, we predict two possible outcomes:
1. Context adequacy causes possible effects of prominence 2.1. Participants
miscomputation fade away. The rationale of this prediction is Seventy-two native Spanish speakers (47 female, age range 18–
that thematic reanalysis effects found in previous studies could 54 years old; M = 22.6, SE = 0.74) participated in this study. All
have been the result of making additional presuppositions participants had normal or corrected-to-normal vision and had
related to the use of a non-canonical word order without any no history of prior neurological disease, drug or alcohol abuse,
previous context. This hypothesis predicts a main effect of psychiatric disorders, developmental speech/language disorders,
information structure, and a triple interaction between word or learning disabilities. All of them provided written consent
order, verb type and information structure, with sentences prior to the study. Sixty-nine of the participants entered the final
with non-canonical word order showing higher processing data analysis, the remaining three having been excluded on the
demand when an unsupportive context is used than when basis of equipment-related artifacts and/or insufficient accuracy
preceded with a supportive context. When a supportive in the comprehension task (an error rate higher than 40% in
context is used, prominence miscomputation effects should the critical conditions). All participants were compensated with
disappear. This interaction should take place once the verbs 150 Argentinian Pesos (approximately US$ 9 at that time) after
are read and in subsequent regions. finishing the experiment session.
2. Context adequacy plays a role at initial stages of sentence
processing but does not make the effect of prominence 2.2. Materials
miscomputation fade away. The rationale of this prediction A total of 384 texts were built following the studies of Gattei
is that the relation between syntax-to-semantics linking et al. (2017) and Kaiser and Trueswell (2004). The texts consisted
and word order involves a mechanism -semantic roles and of three sentences [hereafter S1 refers to the first sentence, S2
syntactic functions- that belongs to the grammatical nucleus refers to the second sentence and S3 , to the third sentence of
of any given language. Thus, the violation of the prominence the text, see example (8)]. S1 introduced the first referent (R1 :
hierarchy comprises the alteration of a core relationship Richard) and the situation in which s/he was. S2 introduced the
in a sentence. On the contrary, the relationship between a second referent (R2 : Mary/Ana) and stated that this person was
non-canonical rhematic structure and non-canonical word performing an action with a person whose name has not been
order involves the manipulation of a more flexible system mentioned (R3 : Ana/Mary). S3 comprised the target sentence,
(i.e., Pragmatics), designed to adapt linguistic form to the which described that R1 (Richard) saw or heard that one of
dynamics of context. In other words, this hypothesis predicts the two referents introduced in S2 did or felt something for
greater difficulty for sentences with unsupportive context the other person. Sentences were built in such way that R1
than for those with supportive one. This difficulty should be always had a different gender than R2 and R3 . This was done
reflected at the initial regions of sentences (i.e., where the in order to avoid possible ambiguity in the use of pronouns
new referent takes place). The hypothesis also predicts higher (“he” or “she”) in the text, so that it was always clear for
processing demands for those sentences that do not respect the reader that it referred to R1 . In other words, if R1 was
the prominence hierarchy than for those that respect it (i.e., feminine, then R2 and R3 were masculine proper names and
for SVO sentences with ObjExp psych verbs and for OVS vice versa.

(8) Ricardo estaba en la sala corrigiendo unos exámenes. texts included sentences with different syntactic complexity and
Richard was in the room grading some exams. length to the target texts, and referred to diverse semantic topics,
Él vio que María/Ana estaba en el pasillo so that participants could not realize what the main purpose of
He saw that Mary/Ana was in the corridor the study was.
charlando por teléfono. Poco después, él oyó que Finally, 123 questions were designed in order to test the
chatting by telephone. Little later, he heard that comprehension of each practice, critical and filler item. Questions
for the critical items were formulated in two ways: In order
María le respondía a Ana...
to respond 32 of the questions, participants had to retrieve the
Mary CL-DAT responded to Ana...
argument structure of the target sentence (S3 ) and participants
“Richard was in the room grading exams. He saw that had to respond whether one of the referents did / felt what for the
Mary/Ana was in the corridor chatting on the phone. Little other referent, while in the remaining 16 texts the question tested
later, he heard that Mary responded to Ana...” the comprehension of one of the two previous sentences (S1 and
S2 ). The rationale of doing this was to assure that participants
In S3 , 24 ObjExp psych verbs and 24 activity verbs with would read the context previous to the target sentence. Half
dative-marked objects were used. Both verb groups were matched of the questions were responded affirmatively and half of them
according to length (ObjExp: M = 6.8, SE = 0.31; Act: M = 6.3, were responded negatively. Half of the questions that referred
SE = 0.25) and log-transformed frequency (ObjExp: M = 4.32, to S3 asked about the subject constituent and half of them
SE = 0.17; Act: M = 4.47, SE = 0.11) according to the referred to the object constituent. Participants had to choose
LEXESP database (Davis and Perea, 2005). An independent the correct answer by clicking on it with the mouse. Position
samples t-test revealed that there were no significant differences of the correct answer was half of the times on the right side of
between groups [Length: t (46) = −1.35, p > 0.05; log Frequency: the screen and was randomly assigned between trials for each
t (46) = 0.71, p > 0.05]. participant. Table 1 shows an example of one of the 48 sets
Verbs from S3 were framed between a Noun Phrase (NP) and of 8 texts used in the current experiment. A complete list of
a Prepositional Phrase (PP) that consisted of 48 pairs of proper the experiment materials may be found at Appendix A of the
names matched in length and counterbalanced in gender (half Supplemental Material, available at https://osf.io/kp4dn/.
masculine and half feminine). Target sentences could also follow
the Subject-Verb-Object (SVO) order or the Object-Verb-Subject 2.3. Equipment
(OVS) word order. In this way, we tested the role of constituents Similarly to (Gattei et al., 2017), participants were seated in front
order for these types of sentences. Finally, information structure of a 19-inch screen (Samsung SyncMaster 997 MB, 1024 × 768
of S3 was also manipulated. In four of the target sentences, pixels resolution, 100 Hz refresh rate) at a viewing distance of
the referent that appeared in first position in S3 had already 65 cm. Head movements were prevented with a chinrest aligned
been mentioned in S2 , while in the other four, the referent with the center of the screen. Gaze locations of both eyes during
that appeared in first position in S3 had not been previously reading was recorded with an EyeLink 1000 eye-tracker (SR
mentioned by its proper name. This means that the configuration Research Ltd.) at a sampling rate of 1 kHz. As given by the
of half of the sentences’ information structure comprised a given manufacturer, nominal average accuracy was 0.5 ◦ and space
referent in first position while in the other half, a new referent resolution was 0.01◦ root mean square (RMS). A standard 13-
was provided at sentence-initial position. point grid for both eyes was used to calibrate participant’s gaze.
The 384 total sentences were divided into eight lists of 48 All recordings and calibration were binocular but only left eye
sentences each (six per condition) so that participants would see data were used for the analysis.
each verb twice, each time in a sentence with different word All eye movements were labeled as fixations, saccades
order and framed by two different pairs of proper names and a and blinks by the eye-tracker software using the default
different context. thresholds for Cognitive experiments (30◦ /s for velocity,
In order to avoid wrap-up effects (Just and Carpenter, 1980), 8,000◦ /s for acceleration, and 0.1◦ for motion, Cornelissen
additional phrases were added at the end of S3 so that regions et al., 2002). Stimuli presentation was developed using Matlab
of interest did not coincide with the last word of the text. These (http://www.mathworks.com/, Massachusetts, United States) and
phrases were semantically neutral so that they would not facilitate Psychophysics Toolbox Version 3.
the interpretation of S3 ’s argument structure. In order to facilitate
the posterior statistical analysis, both the syntactic structure and 2.4. Procedure
length of the first two sentences of the text were kept constant All texts were displayed on five lines, the critical sentence being
among the 48 sets, with a length range between 35 and 52 displayed on the fourth line. Neither the first nor the last word of
characters in S1 (M = 42.5 characters), and 29–40 characters each line displayed any of the main regions of interest from the
in S2 (M = 35.42 characters). Length of S3 would only vary critical sentence nor any of the referents from S2 .
according to the length of the additional phrase used in order to In sum, the design of the text was such that: (i) the critical
avoid “wrap-up” effects, with a length range of 75–88 characters sentence did not exceed one line of the text; (ii) the line of the
(M = 82.33 characters). critical sentence was always the same across trials (line four); and
In addition, a set of three practice items and 72 filler texts that (iii) the line of the critical sentence never started or ended with a
were unrelated to the purposes of the study were used. The latter critical word.

TABLE 1 | Critical sentences used in the current eye-tracking study.
Condition Context Critical sentence Question
(a) ObjExp SVO G-N María le encantaba a Ana aunque no ¿Es María quien le
Ricardo estaba en la sala corrigiendo unos estuvieran de acuerdo. encantaba a alguien?
exámenes. Él vio que María estaba en el pasillo
Ana loves Mary although they wouldn’t agree.
charlando por teléfono. Poco después él oyó que...
(b) ObjExp OVS G-N A María le encantaba Ana aunque no Is it Mary who is loved by
estuvieran de acuerdo. someone?
Ana loved Mary although they wouldn’t agree.
(c) Act SVO G-N María le respondía a Ana aunque no estuvieran ¿Es María quien le
Richard was at the room grading exams. He saw de acuerdo. respondía a alguien?
María was at the corridor talking on the phone.
Mary responded to Ana although they wouldn’t
Later he heard that...
agree.
(d) Act OVS N-G A María le respondía Ana aunque no estuvieran Is it Mary who responded to
de acuerdo. someone?
Ana responded to Mary although they wouldn’t
agree
(d) ObjExp SVO N-G María le encantaba a Ana aunque no ¿Es María quien le
Ricardo estaba en la sala corrigiendo unos estuvieran de acuerdo. encantaba a alguien?
exámenes. Él vio que Ana estaba en el pasillo
Ana loves Mary although they wouldn’t agree.
charlando por teléfono. Poco después él oyó que...
(e) ObjExp OVS N-G A María le encantaba Ana aunque no Is it Mary who is loved by
estuvieran de acuerdo. someone?
Ana loved Mary although they wouldn’t agree.
(f) Act SVO N-G María le respondía a Ana aunque no estuvieran ¿Es María quien le
Richard was at the room grading exams. He saw de acuerdo. respondía a alguien?
Ana was at the corridor talking on the phone. Later
Mary responded to Ana although they wouldn’t
he heard that...
agree.
(g) Act OVS N-G A María le respondía Ana aunque no estuvieran Is it Mary who responded to
de acuerdo. someone?
Ana responded to Mary although they wouldn’t
agree
ObjExp, Object Experiencer Psych Verbs; Act, Activity Verbs; SVO, Subject-Verb-Object, OVS, Object-Verb-Subject; G-N, Given-New; N-G, New-Given.
Sentences were presented in Courier New Bold font. At a 2.5. Data Analysis
distance of 65 cm, each letter subtended 0.44◦ of visual angle Eye movement data from the 69 participants was screened
laterally. Subjects were instructed to read the texts at their own for blinks and track losses. Fixations shorter than 50 ms and
rate. No instructions were given to suppress eye blinks. longer than 1,000 ms were removed from the analysis. After
Before the eye-tracking experiment began, they had a practice this screening process, fixations were assigned to their respective
session of three texts. At the beginning of each trial, a dot word and line. Boundaries between words (x axis) were set
appeared at the top left edge of the screen and after participants by splitting the space between two words in half. Boundaries
fixated on this dot, the text appeared. The first letter of the between lines (y axis) were set by splitting the space between two
text was located at the position of the dot. Participants were lines in half. Upper and lower boundaries of the first and last
instructed to look at a second dot at the bottom right corner lines were calculated so that they were symmetrical with the lower
of the screen to indicate they had finished reading. The total and upper boundaries of these lines, respectively. Fixations that
reading time of each trial was measured starting from when fell outside the boundaries of the text were eliminated whenever
participants triggered the appearance of the text by fixating on participants continued reading after fixating outside the text area.
the left dot until they fixated on the bottom right dot and the text Visual inspection was carried out for each trial by providing
disappeared. Comprehension questions appeared after every text. a number to each fixation and a line that linked consecutive
Participants responded by mouse-clicking on one of two possible fixations. With this representation it could be easily established
answers (“Yes” or “No”) displayed horizontally. Response time whether participants were reading the whole text. Trials in
was measured starting from the appearance of the question which participants skipped sentences from the context or the
until participants clicked on one of the possible responses. A critical sentence were erased. Whenever there was a vertical
calibration procedure was performed at the beginning of the eye- misalignment between the fixations and the lines they belonged
tracking experiment. Experimental sessions lasted approximately to, manual correction of fixations was performed by taking into
45 min. account the trajectory of the reading path and realigning the

fixation to the correct line. Visual inspection and subsequent hand, was analyzed with a generalized mixed-effects model with
correction resulted in the removal of 5,293 fixations (0.79% of Poisson link function, which is appropriate for counts of events
the data) and realignment of 17537 fixations (2.62%). Besides, 44 in a fixed time window (Baayen, 2008, p. 322).
trials were removed due to track loss, the appearance of a random For the regression models, Verb Type, Word Order and
reading pattern, or incomplete text reading. This comprises 0.5% Information Structure were considered fixed effects and Subject,
of the total data. and Item were fit as random effects. Log Frequency and inverse
Eye-tracking measures were computed using em2 package for length of each word were included as control factors in every
R language for statistical computing (Logacev and Vasishth, 2013, region except for regions 1 and 5 (preposition “a”) and region
version 3.0.2). 3 (clitic). These two variables may explain a significant part of
For the purpose of analysis, we divided the sentences into ten the variability in reading times and amount of fixations on these
regions that consisted of the first ten words of each sentence, regions (Just and Carpenter, 1980; Rayner and Well, 1996; Kliegl
as shown in Table 2. Note that in order to facilitate statistical et al., 2004). As for collinearity between both factors, model
analysis and visual presentation of the results, we aligned the comparison among models that included one, the other or both
critical regions that comprised the proper names (regions 2 and were significantly different. AIC values indicated that models
6), the clitic (region 3) and the verb (region 4). The region of the where both factors were included were significantly better than
preposition has been labeled as (5) in subject-initial sentences, the other two. In consequence, the two of them were included.
and (1) in object-initial sentences. The regions “PP1,” “PP2,” A maximal random-effects structure was included in both
“PP3,” and “PP4” correspond to the first to the fourth word of LMMs and GLMMs whenever it was possible, as linear mixed-
the prepositional phrase following the second noun phrase. effects models that do not consider random intercepts and slopes
For each fixated word, we computed the following measures: involve the risk of Type I error inflation (Barr et al., 2013).
(1) First Fixation Duration (FFD; the duration of the first fixation When models either did not converge or the correlation between
on the word); (2) First Pass Reading Time (FPRT; the sum of variance components could not be estimated, the random effects
all fixation durations on the word before any other word was structure was simplified by removing the correlations. For large
fixated); (3) Regression Path Duration (RPD; also known as go- samples like the ones collected in this study, the t distribution
past time, it is the sum of all first-pass fixation durations on the approximates the normal distribution and an absolute value of t
word and all preceding words in the time period between the larger than 2 indicates a significant effect at α = 0.05. For all the
first fixation on the word up to the point where the reader leaves models presented in the study, covariates that involved reading
the critical region with a progressive saccade; (4) Right-Bounded time were scaled and centered.
Regression Count (RBRC; the number of regressions from the Finally, we used an orthogonal contrast coding to test the
word before any word further to the right has been fixated); (5) interactions among verb type, word order and information
Total Fixation Time (TFT; the sum of all fixations durations on structure at the pertinent regions. For the verb type contrast,
a word); and (6) Total Incoming Regressions (TIR; the number sentences with activity verbs were coded as −1 and sentences
of regressions to a specific word). Measures 1–2 are typically with ObjExp psych verbs were coded as 1. For the word order
considered early measures, whereas measures 3–6 are considered contrast, SVO sentences were coded as −1 and OVS sentences
late measures (Clifton et al., 2007; Vasishth et al., 2013). were coded as 1. Finally, for the information structure contrast,
Data analysis was conducted in the R programming sentences with a new referent in first position were coded as
environment (R Core Team, 2013). For measures comprising −1 and sentences with a given referent in first position were
reading or response time (i.e., Comprehension Task Response coded as 1.
Time, FFD, FPRT, RPD, and TFT) a linear mixed-effects model
was fit to the data using the package lme4 (Pinheiro and
Bates, 2000; Bates et al., 2014). For the accuracy measure,
3. RESULTS
the data was fit to a generalized linear mixed-effects model 3.1. Comprehension Task
with a binomial function, which is adequate for analyzing 3.1.1. Total Reading Time
data measured on a dichotomous scale, namely “Correct” and Figure 1A shows the average total reading time for the critical
“Incorrect” response. Count data (RBRC and TIR), on the other texts used in the current eye-tracking experiment. The statistical
TABLE 2 | Regions of interest used for the statistical analysis of the current eye-tracking experiment according to Word Order.
1 2 3 4 5 6 7 8 9 10
María le respondía|encantaba a Ana PP1 PP2 PP3 PP4

SVO
Mary clitic[DAT] responded|loved to Ana[DAT] PP1 PP2 PP3 PP4
A María le respondía|encantaba Ana PP1 PP2 PP3 PP4
OVS
To Mary[DAT] clitic[DAT] responded|loved Ana PP1 PP2 PP3 PP4
SVO, Subject-Verb-Object; OVS, Object-Verb-Subject; PP1, First Word of the Prepositional Phrase; PP2, Second Word of the Prepositional Phrase; PP3, Third Word of the Prepositional
Phrase; PP4, Fourth Word of the Prepositional Phrase.

FIGURE 1 | Mean total reading time for the critical texts (A), percentage of accurate answers (B), mean response times for the comprehension question (C), in the
current eye-tracking study according to verb type (ObjExp psych verb vs. Act), word order (SVO vs. OVS) and information structure (given-new vs. new-given). Error
bars correspond to Standard Error of the Mean. ObjExp psych verb, Object Experiencer Psych Verb; Act, Activity Verb; SVO, Subject-Verb-Object; OVS,
Object-Verb-Subject; GN, given-new; NG, new-given.
analysis revealed an interaction between verb type and word texts with sentences in SVO order (M = 84.51%, SE = 0.9) than
order; β = 0.022, SE = 0.007, t = 3.193, p < 0.01. Resolving this texts with sentences in OVS order (M = 70.64%, SE = 1.13;
interaction showed that participants spent significantly longer β = −0.626, SE = 0.056, z = −11.27, p < 0.001).
time reading sentences with SVO word order when they included
an ObjExp psych verb than when they included an activity verb; 3.1.3. Response Time
β = 0.071, SE = 0.020, z = 3.624, p < 0.01. Although this Figure 1C shows mean response time (RT) according to
difference was not significant among OVS conditions, sentences condition. Analysis of differences in RT between verb type,
with activity verbs were read slower when they followed the OVS word order and information structure revealed main effects of
word order than when they followed the SVO order; β = 0.077, the three factors. On average, response time was significantly
SE = 0.020, z = 3.955, p < 0.001. Information structure longer for questions about texts that included ObjExp psych verbs
also affected texts’ reading times significantly. Participants took (M = 4,629 ms; SE = 70) than for questions about texts with
longer time to read texts in which the Information Structure activity verbs (M = 3,889 ms, SE = 58; β = −0.118, SE = 0.013,
of the critical sentence included a new referent in first position t = −9.066, p < 0.001). Participants also took longer time
(M = 16,955 ms, SE = 210 ms) than when it included a given to respond to questions about texts that included sentences in
referent in first position (M = 16,015 ms, SE = 207 ms; β = 0.032, OVS order (M = 4,396 ms; SE = 67) than when they included
SE = 0.007, z = 4.552, p < 0.001). sentences in SVO order (M = 4,121 ms, SE = 61; β = 0.054,
SE = 0.011, t = 5.031, p < 0.001). Finally, questions about texts
that included critical sentences with non-canonical information
3.1.2. Question Accuracy structure were responded significantly slower (M = 4,383 ms;
Mean accuracy for all comprehension questions was 86.27% SE = 67) than questions about texts that included sentences
(SE = 0.38%). This indicates that participants were paying with a canonical rhematic hierarchy (M = 4135 ms, SE = 62;
attention to the content of the texts. Mean accuracy of β = 0.025, SE = 0.011, t = 2.270, p < 0.05). Interactions among
critical questions was 77.45% (SE = 0.74%). Figure 1B shows the three factors were not significant.
mean accuracy according to condition. Differences in accuracy
according to verb type and word order were analyzed with a 3.2. Eyetracking Measures
generalized linear mixed-effects model. The analysis revealed Figure 2 summarizes the contrast between sentences with activity
a significant interaction between verb type and word order; verbs and sentences with ObjExp psych verbs according to both
β = −0.198, SE = 0.055, z = −3.614, p < 0.001. Resolving word orders (SVO in red and OVS in blue) and information
this interaction revealed that accuracy was significantly higher structure (GN in dashed lines; NG in solid lines). Positive
for questions about sentences with Activity verbs and SVO word values mean that reading time is longer and regression counts
order than for the other three conditions (ActSVO - ActOVS; are higher for sentences with activity verbs than for sentences
β = 1.640, SE = 0.164, z = 9.981, p < 0.001; ActSVO - with ObjExp psych verbs. A positive blue line and a negative
ObjExpSVO; β = 0.532, SE = 0.172, z = 3.089, p = 0.011; red line correspond to an interaction between Verb Type
ActSVO - ObjExpOVS; β = 1.385, SE = 0.165, z = 8.397, and Word Order. Absolute values higher for solid lines than
p < 0.001). A significant effect of word order was also found. for dashed lines show an effect of Information Structure as
On average, participants responded more accurately after reading expected, with non-canonical information structure showing

FIGURE 2 | The Figure shows the difference (1) in mean fixation times (ms) and the amount of regressive saccades (counts) between conditions with Activity Verbs
and conditions with Psych verbs according to the sentence word order (SVO vs. OVS) and information structure (Given-New vs. New-Given) Error bars correspond to
Standard Error of the Mean. Eye-tracking measures: FFD, First Fixation Duration; FPRT, First Pass Reading Time; TFT, Total Fixation Time; RPD, Regression Path
Duration; RBRC, Right-Bounded Regression Count; TIR, Total Incoming Regressions. Word Order: SVO, Subject-Verb-Object; OVS, Object-Verb-Subject. The
asterisk shows that the interaction between Word Order and Verb Type was significant.
higher cognitive demand than canonical information structure. Fixation Time (TFT) showing longer fixation time for sentences
This representation makes the interaction and Information with activity verbs than for sentences with ObjExp psych verbs;
Structure effect visually clear. The asterisks show the regions β = 0.636, SE = 0.019, t = 3.311, p < 0.001.
where the interaction was significant.
We now provide the analysis of regions of interest for both the Region 2 (First proper name)
early and late measures mentioned in the section 2.5. For each Late eye-movement measures showed a significant interaction
region, we first present the analysis of the interactions among between Verb Type and Word Order for Total Fixation Time
factors since they comprise the contrasts of interest of the current (TFT) and Total Incoming Regressions (TIR); TFT: β = 0.086,
study. We then provide the relevant results of the multiple SE = 0.011, t = 7.923, p < 0.001; TIR: β = 0.192, SE = 0.018,
comparisons test whenever was needed. Finally, we report main z = 10.619, p < 0.001). Resolving these interactions revealed
effects of Verb Type, Word Order or Information Structure. that for subject-initial sentences, the probability of regressing
Appendix B shows the final converging models for each measure to this region and the total fixation time were significantly
at each region. A table with all statistical coefficients, standard longer when the sentence included an ObjExp psych verb
errors and t values may be found at Appendix C. than when it included an activity verb (TFT: β = 0.259,
SE = 0.031, z = 8.384, p < 0.001; TIR: β = 0.580, SE = 0.060,
Region 1 (Case marking preposition “a”) z = 9.590, p < 0.001). Conversely, for object-initial sentences, the
Analysis of late eye-movement measures revealed no interactions probability of regressing into this region and the total fixation
among factors. A main effect of Verb Type was found for Total time were significantly longer when the sentence included an

activity verb than when it included an ObjExp psych verb regressed to this region significantly more when the sentence
(TFT: β = 0.087, SE = 0.031, z = 2.811, p = 0.025; TIR: contained an activity verb; TIR: β = 0.246, SE = 0.067, z = 3.748,
β = 0.181, SE = 0.056, z = 3.258, p = 0.005). Analysis of p < 0.001. This difference was not significant for TFT; β = 0.071,
Regression Path Duration (RPD) also showed a main interaction SE = 0.039, z = 1.817, p < 0.265. The analysis of these measures
between word order and information structure β =−0.027, also revealed main effects of Verb Type (TFT: β = −0.039,
SE = 0.0127, t = −2.158, p < 0.05. Tukey post-hoc test SE = 0.014; t = −2.791; TIR: β = −0.094, SE = 0.024 z = −3.833,
revealed that non-canonical word order (OVS) led to significantly p < 0.01), Word Order (TIR: β = −0.075, SE = 0.024 z = 3.095,
longer regression path duration before continuing reading when p < 0.01; RPD: β = 0.076 SE = 0.029, t = 2.639, p < 0.05),
the sentence presented a non-canonical information structure, and Information Structure (TFT: β = −0.042, SE = 0.016;
β =−0.107, SE = 0.035, z = −3.022, p < 0.05. This difference t = −2.597, p < 0.01). Participants showed higher processing
was not significant between sentences with canonical information load whenever the sentences included an ObjExp psych verb
structure. than when the included an activity verb. They also regressed
Analysis of early eye-movement measures revealed a main to this region significantly more when the sentence followed
effect of Information Structure for First Fixation Duration (FFD) the OVS order than when it followed the SVO word order, and
and First Pass Reading Time (FPRT), showing longer reading fixated for longer time on this region when the first NP belonged
time for sentences with non-canonical Information Structure to a new referent than when it belonged to an already given one.
than for sentences with canonical Information Structure (FFD:
β = −0.032, SE = 0.006, t = −5.018 , p < 0.001; FPRT: Region 4 (Disambiguating verb)
β = −0.052, SE = 0.008, t = −6.714,, p < 0.001.) A Analysis of early eye-movement measures revealed no
similar effect was found for Right-Bounded Regression Count interactions among factors. However, a main effect of
(RBRC), showing higher amount of regressions from this Information Structure was found for FPRT, showing significantly
region for sentences with non-canonical Information Structure: longer reading time for this region whenever the sentence
β = −0.016, SE = 0.039, t = −4.161, p < 0.01. presented a new referent in first position; β = −0.042,
Analysis of the probability of regressions into this region SE = 0.009, t = −4.706, p < 0.001. A significant interaction
(TIR), RBRC , RPD, and Total Fixation Time also revealed a main between Verb type and Word Order was found for all late
effect of Word Order. On average, participants regressed to this eye-movement measures (RPD: β = 0.053, SE = 0.012, t = 4.449,
word significantly more, fixated on this word for longer time p < 0.001; RBRC: β = −0.135, SE = 0.033, z = 4.081, p < 0.001;
and regressed for significantly longer time and higher amount TFT: β = 0.119, SE = 0.010, t = 11.852tcr, p < 0.001; TIR:
of times from this region in object-initial sentences than in β = 0.202, SE = 0.019, z = 10.531, p < 0.001) Tukey post-hoc
subject-initial sentences; (TIR: β = 0.155, SE = 0.033, z = 4.710, test showed that this interaction follows the same direction as
p < 0.001; TFT: β = 0.063, SE = 0.011, t = 5.744, p < 0.001; in the previous region (SVO: RBRC: β = 0.276, SE = 0.102,
RBRC: β = 0.010, SE = 0.040, t = 2.502, p < 0.05; RPD: z = 2.711, p < 0.05; TFT: β = 0.358, SE = 0.047, z = 7.701,
β = 0.026, SE = 0.013, t = 2.065, p < 0.05 . Finally, effects of Verb p < 0.001; TIR: β = 0.631, SE = 0.081, z = 7.751, p < 0.001;
Type and Information Structure were present for TFT and effects OVS: RPD: β = 0.122, SE = 0.038, z = 3.222, p = 0.007; TFT:
of Information Structure were found at RPD(TFT: Verb Type: β = 0.118, SE = 0.046, z = 2.544, p < 0.05; RBRC: β = 0.266,
β = −0.043, SE = 0.011, t = −3.963, p < 0.001; Information SE = 0.091, z = 2.930, p < 0.05). TIR also showed a significant
Structure:β = −0.110, SE = 0.013, z = −8.203, p < 0.001, RPD interaction among the three main factors, β = 0.041, SE = 0.019,
- Information Structure: beta = −0.083, SE = 0.013, z = −6.543, z = 2.112, p < 0.05. Tukey HSD multiple comparisons showed
p < 0.001. The sign of these effects reveal that participants fixated that this triple interaction depended on the interaction between
for longer time when the sentence included an ObjExp Psych Verb and Word Order: when new information was in both first
verb and fixated and regressed to previous regions for longer time and second position, and sentences included an ObjExp psych
when the noun corresponded to a new referent. verb, participants regressed significantly more to this region in
SVO conditions than in OVS sentences (New-Given:β = 0.033,
Region 3 (Clitic) SE = 0.073, z = 4.546, p < 0.001; Given-New: β = 0.040,
Analysis of early eye-movement measures revealed no SE = 0.071, z = 5.641, p < 0.001). The opposite pattern took
interactions among fixed factors nor main effects. Analysis place for sentences with activity verbs: participants regressed
of late eye-movement measures showed a significant interaction significantly more to this region when the sentence followed
between Verb Type and Word Order for total fixation time, the OVS word order than when it followed the SVO order
and for the probability of regressions into this region (TFT: (New-Given: β = 0.033, SE = 0.079, z = 4.160, p < 0.001;
β = 0.074, SE = 0.014; t = 5.331; TIR: β = −0.218, SE = 0.024 Given-New: β = 0.057, SE = 0.084, z = 6.762, p < 0.001). A
z = 8.977, p < 0.001). This interaction follows the same significant effect of Word Order was found for RPD, RBRC
direction as the interaction found on Region 2. For subject-initial and TFT in the same direction as in the previous region:
sentences, participants regressed and fixated on this region participants found higher processing cost at this region for
significantly more when the sentence contained an ObjExp psych OVS sentences than for SVO sentences (RPD: β = 0.044,
verb than when it included an activity verb (TFT: β = 0.226, SE = 0.012, t = 3.739, p < 0.001; RBRC: β = 0.110, SE = 0.033,
SE = 0.040, z = 5.685, p < 0.001; TIR: β = 0.624, SE = 0.072, z = 3.280; TFT: β = 0.419, SE = 0.010, t = 4.173, p < 0.001)
z = 8.630, p < 0.001). In object-initial sentences; participants Furthermore, participants fixated for significantly longer time

on this region and regressed significantly more to it whenever shows longer reading time for conditions with new information
it included an ObjExp psych verb than when it included an in second position than for conditions with new information in
activity verb (TFT: β = −0.060, SE = 0.021, t = −2.865, first position; β = −0.026; SE = 0.008; t = −3.101, p < 0.01.
p < 0.01; TIR: β = −0.115, SE = 0.035, z = −3.251, p < 0.01).
Finally, Information Structure modulated both RPD and Region 7 (First word of the Spill-over region)
TFT. Participants fixated for longer time on that region and Analysis of this region showed that the interaction between
previous regions before continuing reading and fixated for Verb Type and Word Order was significant for FFD and for
longer time on that word whenever the sentence included a three out of five late eye-movement measures (FFD: β = 0.020,
new referent in first position (RPD: β = −0.062, SE = 0.012, SE = 0.008, t = 2.649, p = 0.008; RPD: β = 0.066, SE= 0.016,
t = −5.175, p < 0.001; TFT: β = −0.073, SE = 0.011, t = −6.827, t = 4.206, p < 0.001; RBRC: β = 0.151; SE = 0.051, z=2.953,
p < 0.001). p < 0.05; TFT: β = 0.052; SE = 0.011, t = 4.579, p < 0.001).
The multiple comparisons test showed that in subject-initial
Region 5 (Case marking preposition “a”) sentences, participants fixated for longer time at this region
Analysis of this region showed a significant main effect of Verb when the sentence contained an ObjExp psych verb than when
Type for most late eye-movement measures, with participants it included an activity verb (TFT: β = 0.0332, SE = 0.032,
experiencing greater cognitive load and regressing significantly t = 10.477, p < 0.001). Participants also spent longer time reading
more to this region whenever the sentence included an ObjExp and regressing to previous regions and regressed significantly
psych verb (RPD: β = −0.126, SE = 0.041; t = −3.074, p < 0.01; more times from this region for sentences with ObjExp psych
RBRC: β = −0.28617, SE = 0.089; z = −3.225, p < 0.001;TFT: verbs than for sentences with activity verbs (RPD: β = 0.021,
β = −0.132, SE = 0.020; t = −6.544, p < 0.001; TIR: β = −0.348, SE = 0.045, z = 4.724, p < 0.001; RBRC: β = 0.665; SE = 0.146,
SE = 0.059; z = −5.921, p < 0.010). t = 4.551, p < 0.001).
Differences among object-initial sentences were marginally
Region 6 (Second proper name) significant for Total Fixation Time, with participant fixating
Analysis of this region showed an interaction between Verb Type for longer time on this region when the sentence included an
and Word Order for all late eye-movement measures (RPD: activity verb than when it included an ObjExp psych verb; TFT:
β = 0.116, SE = 0.015, t = 7.598, p < 0.001; RBRC: β = 0.177, β = 0.077, SE = 0.032, z = 2.436, p = 0.070. For the other four
SE = 0.028, z = 6.253, p < 0.01; TFT: β = 0.010; SE = 0.011; measures this difference was not significant.
t = 9.150, p < 0.001; TIR: β = 0.084; SE = 0.040; t = 2.125, Analysis of this region also showed main effects of word
p < 0.05). Resolving these interactions revealed that in SVO order and verb type for late measures RPD and RBRC.
sentences, participants fixated for longer time at this and previous Participants regressed significantly more from this region and
regions and regressed significantly more times from and to this spent significantly longer time on previous regions before
region when the sentence contained an ObjExp psych verb than continuing reading when the sentences included ObjExp psych
when it included an activity verb (RPD: β = 0.398, SE = 0.043, verbs than when they included activity verbs and when they
z = 9.235, p < 0.001; RBRC: β = 0.534, SE = 0.081, z = 6.573, followed the OVS word order than when they were subject-initial
p < 0.001; TFT: β = 0.332, SE = 0.033, z = 9.962, p < 0.001; TIR: sentences (Word Order: RPD: β = 0.049, SE = 0.016, t = 3.127,
β = 0.624, SE = 0.072, z = 8.693, p < 0.001). p < 0.01; RBRC: β = 0.021, SE = 0.048, z = 4.454, p < 0.001;
Difference among OVS conditions was only significant for Verb: RPD: β = −0.039, SE = 0.016, t = −2.517, p < 0.05; RBRC:
Total Incoming Regressions, with conditions with activity verbs β = −0.184, SE = 0.046, z = −3.983, p < 0.01).
showing a higher amount of regressions to this region than
conditions with ObjExp psych verbs (β = 0.246, SE = 0.065, Region 8 (Second word of the Spill-over Region)
z = 3.777, p < 0.001). A significant interaction between Verb Type and Word Order
A main effect of Word Order was found for FFD and was found for FFD and RPD (FFD: β = 0.023, SE = 0.010,
FPRT, with longer reading times for sentences with SVO word t = 2.555, p < 0.05; RPD: β = 0.042, SE = 0.019, t = 2.224,
order than for sentences with OVS order (FFD: β = −0.016; p < 0.05). The multiple comparisons Tukey HSD test revealed
SE = 0.007; t = −2.292, p = 0.002; FPRT: β = −0.034; significant differences among SVO conditions for RPD only,
SE = 0.008; t = −3.989, p < 0.01). with conditions with ObjExp psych verbs engendering longer
A significant effect of Verb Type was found for FPRT and regression path duration than sentences with activity verbs.
all late eye-movement measures except for TIR. The sign of Differences among OVS conditions were not significant for any
the effect shows longer reading time, regression duration and of the above-mentioned measures. Analysis of FFD and TFT
amount of regressions from this region when the sentences also showed a significant interaction between Word Order and
included an ObjExp psych verb than when they included an Information Structure (FFD: β = 0.021, SE = 0.010, t = 2.087,
activity verb (FPRT: β = −0.023; SE = 0.008; t = −2.724, p < 0.05; TFT: β = 0.028, SE = 0.013, t = 2.138). However,
p < 0.01; RPD: β = −0.083; SE = 0.015; t = −5.438, p < 0.001; the interaction was not confirmed by the multiple comparisons
RBRC: β = −0.090; SE = 0.028; z = −3.180, p < 0.01; TFT: tests from both measures. Finally, a main effect of Verb Type
β = −0.064; SE = 0.012; t = −5.172). Finally, a significant effect was found for RPD, with sentences with ObjExp psych verbs
of Information Structure was also for FPRT. Contrary to the effect engendering longer regression path duration than sentences with
of Information Structure found in previous regions, this region activity verbs; β = −0.046, SE = 0.019, t = −2.413, p < 0.05.

Region 9 (Third word of the Spill-over Region) predictions about the verb type that will take place in the
Analysis of early eye-movement measures revealed a significant sentence according to the prominence status of that argument.
interaction between Word Order and Information Structure for This proposal assumes that the language processing system
FPRT; β = 0.021, SE = 0.010, t = 2.037, p < 0.05. However, gives rise to predictions about arguments order by following
the multiple comparisons test showed no significant differences a prominence hierarchy that canonically stipulates that more
among conditions. prominent arguments precede less prominent ones (Bornkessel
Analysis of late eye-movement measures showed a main effect et al., 2005; Wolff et al., 2007; Haupt et al., 2008). Nonetheless,
of Verb Type for RPD and a main effect of Word Order for TIR it is relevant to ask whether the pragmatic use of constituents
(Verb: β = −0.055, SE = 0.026, t = −2.113, p < 0.05; Word order has any influence on the mentioned results. In the
Order: β = 0.056, SE = 0.026, z = 2.198, p < 0.05). The sign of the current study, we focused on the distinction between “given”
effects show that participants regressed significantly longer time and “new” referents in a sentence in relation with a previous
to previous regions when the sentence included an ObjExp psych context. According to Givón (1984), the use of a non-canonical
verb and that they regressed more times to this region when the word order is expected when mentioning a referent that has
sentence followed the OVS word order. already been introduced by the previous context so that the
rhematic hierarchy (i.e., given-new) is respected. Experimental
Region 10 (Fourth word of the Spill-over Region) evidence about the role of rhematic hierarchy during incremental
Analysis of this region showed a significant interaction among reading has shown that effects of word order non-canonicity
Word Order, Verb Type and Information Structure for FPRT are alleviated when an adequate context precedes the sentence
and RPD (FPRT: β = 0.024, SE = 0.010, t = 2.484, p < 0.05; (Kaiser and Trueswell, 2004; Burmester et al., 2014), suggesting
RPD: β = 0.056, SE = 0.021, t = 2.680, p < 0.01). Resolving that increasing reading times in isolated sentences with non-
these interactions revealed that for First Pass Reading Time, canonical word order could partly be due to higher cognitive
participants spent longer time reading this region in sentences demands related to making assumptions about possible contexts
with activity verbs and a new referent in first position when the in which a non-canonical word order could take place. However,
sentence followed the SVO order than when it followed the OVS the role of information structure in relation to alternative word
one; β = 0.122; SE = 0.039, t = 3.164, p < 0.05. Differences orders stipulated by different lexico-semantic configurations
among the other conditions did not reach significance for this had not been explored yet. Thus, the question that motivated
measure nor for RPD. A main effect of Word Order was the present study was whether the prominence effects found for
also sound for FPRT. Participants spent longer time on this Spanish sentence comprehension were caused by the lack of a
region when word order was SVO than when it was OVS, context that could motivate the election of a specific word order.
β =−0.0239, SE = 0.010, t = −2.486 , p < 0.05 Finally, a main We thus framed sentences used in Gattei et al. (2017) in texts
effect of Information Structure was found for RPD, β =−0.064, that would favor the appearance of a specific referent in first
SE = 0.021, t = −3.078, p < 0.01. The sign of the effect shows position of the sentence and compared them with sentences in
longer reading time for conditions with new information in which the first argument comprised a new referent. By means
second position than for conditions with new information in of a comprehension offline task, we also evaluated the cognitive
first position. cost of understanding “who did/felt what for whom” correctly.
Regarding the hypotheses and predictions outlined at the
4. DISCUSSION section 1, the current work shows that context adequacy plays
a role for processing of sentences with non-canonical word order
Evidence about the comprehension of isolated Spanish sentences but does not make effects of prominence miscomputation fade
with alternative word orders has shown that readers manifest away.
increasing difficulty to understand sentences with a word order On the one hand, results of the current study revealed that the
that does not respect the order of arguments at the lexico- use of an adequate context facilitated the comprehension of the
semantic structure of the verb, independently of whether the target sentences. Participants took significantly less time to read
sentence follows the canonical word order of the language (SVO) the texts when, in first position, the final sentence introduced
or not (Gattei et al., 2015a,b, 2017). In these studies, the authors a referent that had explicitly been presented before. They also
used sentences with activity verbs and object experiencer verbs took less time to respond the comprehension questions when
in order to compare events that required alternative linking the target sentence included a canonical information structure.
between syntax and semantics. When reading SVO sentences, In other words, these results replicate the findings that Kaiser
participants required significantly more time to read and figure and Trueswell (2004) showed for Finnish sentences with activity
out “who did / felt what for whom” when the sentence included verbs. Additionally, the current study revealed that a non-
an ObjExp psych verb. Conversely, when reading OVS sentences, canonical information structure is detrimental to comprehension
participants required more time to read when the sentence even in SVO sentences, as it is evidenced by sentences’ response
included an activity verb. This interaction was present when time of questions about sentences with ObjExp psych verbs.
using both self-paced reading and eye-tracking techniques. On the other hand, effects related to incorrect syntax-to-
These studies support the hypothesis that during incremental semantic linking were present during reading for late eye-
parsing, readers use the morphosyntactic and semantic movement measures as predicted, showing a disruption of
information provided by the first sentential argument to generate processes of higher-level text integration (Clifton et al., 2007).

When encountering a verb that did not match the predictions the dynamics of context -possibly through working memory
stipulated by the computation of prominence status of the capacity-, and prepared to deal with domain-general mechanisms
first argument, participants took longer time to read the word like attentional shift and reorienting. Although the current work
and following content words, and regressed more times and was not aimed at disentangle whether the effects produced by
for longer time to previous regions. These results replicate the the use of a non-canonical rhematic hierarchy were related
findings by Gattei et al. (2017) for isolated sentences, and yield to factors associated to referents’ accessibility, readers’ working
further evidence in favor of the hypothesis that one of the memory capacity or a failure of an expected structure and
central mechanisms for argument interpretation is prominence the subsequent need for attentional reorientation, the current
computation (Bornkessel and Schlesewsky, 2006), and that findings are informative with regards to the mechanisms
prominence computation follow a principle that assumes that the underlying incremental processing of new information.
first argument will be the most “Actor-like” possible. In the current study, eye-movement measures showed
The current study also provides interesting insight about information structure effects, evidenced by increasing reading
the cognitive cost and strategies used by readers to process times for the first and the second NP whenever readers found a
new information. Several proposals have tried to explain the new referent. However, it is generally argued that reading words
cognitive effects derived from the use of a non-canonical that are repeated throughout a text entails a decrease of reading
rhematic hierarchy. Although there is an agreement regarding time (Rayner et al., 1995; O’Brien et al., 1997; Kamienkowski
the type of effects caused by the unpredictable appearance et al., 2018). Hence, it is fair to ask whether the effects found at
of a new referent, there is not a unique view with respect new referents respond to the manipulation of rhematic hierarchy
to which mechanisms are involved in information structure or if they should be interpreted as lexical repetition effects
processing. For instance, it has been proposed that speakers tend related to word recognition processes. For instance, Lowder et al.
to choose the syntactic constructions that allow them to place (2013) ran en eye-tracking study in which participants had to
the most “accessible” (already mentioned) information earlier in read sentences with two NPs composed by two and one proper
the utterance (Ferreira, 2003), possibly because this allows them names, respectively. The authors manipulated proper names’
to postpone the difficult part of the utterance, which requires frequency and repetition (the second NP could mention one of
more resources to plan. The assumption behind this hypothesis the proper names from the first NP or not) and showed that when
is that when information has a strong representation in memory reading the second NP, repeated names were processed more
it is easier to retrieve and to process. Evidence in favor of this quickly than new names in both early and late eye-movement
view shows that speakers choose word order according to visual measures. Following Gordon and Hendrick (1998), the authors
attention (Gleitman et al., 2007). argue that “while basic word recognition goes on, the effort to
Kaiser (2012) argues that the pragmatic status of referent understand the meaning of a sentence or short discourse leads
emerges naturally from memory and attention. Theories about to the construction of a discourse model that represents patterns
memory distinguish between working memory, which stores of reference and co-reference and which captures the predicate-
information currently being used, and long-term memory, which argument relationships described in the text” (Lowder et al.,
stores the conceptual and procedural knowledge for posterior 2013). The results of the current study showed a similar pattern of
use. From this point of view, given referents can be defined results, with a modulation of both early and late eye-movement
as those accessed through working memory (and thus easier measures at the first NP and following two regions, when the
to retrieve) and new referents as those which have not been proper name comprised a new referent. Interestingly, when the
retrieved by long-term memory yet (Arnold et al., 2013). sentence followed a canonical information structure, effects of
From a neurobiological perspective of language and its a new referent were only present for First Pass Reading Time
processing, Bornkessel-Schlesewsky and Schumacher (2016) at the proper name region. We interpret this pattern of results
propose that, instead of postulating specific neural correlates for as a difference in the control and time course of oculomotor
information structure, a more promising approach is to consider processes for word recognition, with short-lived, early effects,
that information structure affects domain-general mechanisms and for information structure manipulation, which affected late
when hierarchically guiding predictive processing or when eye-movements and caused a longer comprehension disruption.
providing cues for attentional shift. The authors claim that As for offline measures collected in the current study, total
the status of discourse referents feeds the predictive processes reading time is informative of the time required by readers to
during discourse, as it is shown by how the preference for guarantee that they have understood the text. Although these
a continuity of the same referent or for certain types of were the instructions provided, this measure did not reflect
linearization (i.e., given referents precede new ones) facilitate comprehension success, as shown by accuracy rates. In particular,
language processing. Errors in the predictions at this level participants responded questions significantly better when the
elicit negative potentials (for instance N400 for unpredictable final sentence followed Spanish canonical word order (SVO),
information structure properties) and result in attentional independently of whether the initial constituent consisted of a
reorienting and mental model updating required in topic new referent or not. Although this was expected for sentences
shift scenarios. with activity verbs, a preference for SVO word order was not
A general idea that stems from these approaches is that expected for accuracy rates of sentences with ObjExp psych verbs,
the violation of rhematic structure involves the modification diverging from the results found in previous studies about this
of a more flexible system, designed to adapt linguistic form to issue with isolated sentences. In Gattei et al. (2017) participants

showed overall higher accuracy rates (around 90% accuracy for opposed to two as it occurs in the current version of the
critical questions), and higher accuracy rates for questions about experiment), enabling the possibility of introducing the confound
sentences with activity verbs than for those about ObjExp psych of structure repetition effect and other possible confounds due
verbs. While further investigation of this difference between to participants tiredness or boredom, or that the amount of
experiments is needed, it is possible that the use of additional subjects tested was doubled to approximately 150 to yield results
context and the requirement of keeping referents in working comparable to the current ones. Considering that the sentences
memory in order to reply the comprehension questions, had without context have been repeatedly tested (Gattei et al.,
a negative effect on the comprehension of Spanish overall less 2015a,b, 2017) we considered that the design of the current study
frequent word order. was a fair trade-off between running the ideal experiment and
With regards to the interaction between verb type and word getting reliable results. Still, adding further isolated conditions in
order found for accuracy rates, results replicate the findings of languages that have not been previously tested would be crucial
Gattei et al. (2017), with higher accuracy for questions about SVO for results’ replication.
sentences with activity verbs than for the other conditions. We A second aspect that needs further investigation is how
argue that this pattern is expected as in this type of sentences participants deal with referent’s activation when encountering
both semantic and syntactic canonical orders coincide, while the sentences with non-canonical information structure. In other
other conditions present an alteration of either semantic order words, can regressive saccades from regions comprising new
(as in SVO sentences with ObjExp psych verbs), constituents referents to previous sentences be informative of participants’
order (as in OVS sentences with ObjExp psych verbs) or both reading strategies related to referent updating? (Chafe, 1976,
(as in OVS sentences with Activity verbs). In other words, results 1994). While, this question was out of the scope of our
show that the alignment of both canonical linking and canonical work, we are currently addressing this issue with the data
word constituents order facilitates comprehension, whereas non currently collected.
canonical arrangement of either type of information makes it
more difficult.
A final aspect that needs to be taken into account is the 5. CONCLUSIONS
response time for comprehension questions, which show that
readers needed extra time to respond questions about sentences We presented an exploratory study that evaluated the interaction
with either OVS word order, non-canonical information between word order, lexico-semantic structure of the verb
structure or ObjExp psych verbs. We believe that the lack of and information structure in the comprehension of Spanish
information structure effects for overall accuracy shows that texts. Previous studies about this language have only evaluated
while the use of a non-canonical rhematic hierarchy require the role of the first two factors, leaving aside the pragmatic
longer reading time and response time for comprehension aspect involved in the election of constituents’ word order.
questions, the consequences of not following a canonical Understanding the role of information structure is crucial to
order for information structure are not as strong as to show explain sentence processing in this language, since previous
a modulation of comprehension success, as it occurs with evidence has shown that when sentences are presented in
OVS sentences. However, it is matter of future research to isolation, constituents order is a relevant cue for incremental
evaluate whether differences in the trade-off between response argument interpretation. It was pertinent to ask whether
time and accuracy for word order and information structure word order is still a relevant cue for argument interpretation
non-canonicity respond to task-related factors (as structural when the previous context justifies (or not) the appearance
complexity or types of questions used) or individual differences of a specific word order. By evaluating reading of texts that
(like working memory or attentional capacities). manipulated the relation between “given” and “new” referents
we showed that while information structure canonicity enhances
comprehension, the use of an adequate context for a specific
4.1. Possible Methodological Caveats and word order does not alleviate comprehension effects caused by
Future Directions argument misinterpretation. This type of evidence is crucial
Although results of the current study support previous results on for any model of language comprehension that attempts to
this issue, possible methodological caveats should be taken into explain sentence processing in languages that allow alternative
account for future research and replication in other languages. word orders.
Most importantly, while the study asks about the role of By conducting an eye-tracking experiment, we could also
prominence computation in sentences embedded in context, provide further information about the time course of on-going
the materials were designed in such way that they do not processing of new referents, which show a different gaze signature
directly compare comprehension of isolated sentences with to lexically-driven word retrieval.
comprehension of sentences embedded in texts within the same
group of subjects. The rationale of not doing so was that adding
no-context trials would have implied to double the amount of DATA AVAILABILITY STATEMENT
conditions to sixteen conditions. Considering the short amount
of ObjExp psych verbs available in Spanish, this would have The raw data supporting the conclusions of this article are
implied that participants either read each verb four times (as available at https://osf.io/kp4dn/.

ETHICS STATEMENT FUNDING

The studies involving human participants were reviewed and This work was part of the CG postdoctoral fellowship work
approved by Facultad de Filosofía y Letras - Universidad de funded by the National Council of Scientific and Technical
Buenos Aires. The participants provided their written informed Research (CONICET) for the project Prediction and integration
consent to participate in this study. during information structure parsing in Spanish and its
interaction with the syntax-semantics interface. This work
AUTHOR CONTRIBUTIONS was also funded by the James McDonnell Foundation JSMF
Grant #220020334, awarded to Laboratorio de Neurociencia at
CG designed the materials of the study, collected the data, ran the Universidad Torcuato Di Tella.
statistical analysis, and sketched the first draft of this manuscript.
DS programmed the Matlab script for data collection, data ACKNOWLEDGMENTS
preanalysis, and figures’ design. LP supervised the adequacy of
hypotheses and predictions to formal approaches of information We would like to thank the contribution of the attendees and
structure canonicity. All authors equally contributed with the organizers of the I Encuentro Latinoamericano de Interfaces
initial discussion of hypotheses, predictions and results of the Gramaticales (ELIG) for the thorough discussion of the current
current study, and writing the final version of this manuscript. results previous to this article’s submission.
REFERENCES Chafe, W. (1976). “Givenness, contrastiveness, definiteness, subjects, topics, and

point of view,” in Subject and Topic, ed C. Li (New York, NY: Academic Press).
Arnold, J. E., Kaiser, E., Kahn, J. M., and Kim, L. K. (2013). Information structure: Chafe, W. (1994). Discourse, Consciousness, and Time: The Flow and Displacement
linguistic, cognitive, and processing approaches. Wiley Interdiscipl. Rev. Cogn. of Conscious Experience in Speaking and Writing. University of Chicago Press.
Sci. 4, 403–413. doi: 10.1002/wcs.1234 Chow, W.-Y., and Phillips, C. (2013). No semantic illusions in the ’semantic P600’
Baayen, R. H. (2008). Analyzing Linguistic Data: A Practical Introduction phenomenon: ERP evidence from Mandarin Chinese. Brain Res. 1506, 76–93.
to Statistics Using R. Cambridge, MA: Cambridge University Press. doi: 10.1016/j.brainres.2013.02.016
doi: 10.1017/CBO9780511801686 Clifton, C., Staub, A., and Rayner, K. (2007). “Eye movements in reading words
Bader, M., and Bayer, J. (2006). Case and Linking in Language Comprehension: and sentences,” in Eye Movements: A Window on Mind and Brain, ed R. van
Evidence from German, Volume 34 of Studies in Theoretical Psycholinguistics. Gompel (Elsevier Science), 341–372. doi: 10.1016/B978-008044980-7/50017-3
Dordrecht: Springer Science & Business Media. Comrie, B. (1989). Language Universals and Linguistic Typology: Syntax and
Bader, M., and Meng, M. (1999). Subject-object ambiguities in German embedded Morphology. University of Chicago Press.
clauses: an across-the-board comparison. J. Psycholinguist. Res. 28, 121–143. Contreras, H. (1976). A Theory of Word Order with Special Reference to Spanish.
doi: 10.1023/A:1023206208142 Amsterdam: North-Holland.
Barr, D. J., Levy, R., Scheepers, C., and Tily, H. J. (2013). Random effects Cornelissen, F. W., Peters, E. M., and Palmer, J. (2002). The eyelink toolbox: eye
structure for confirmatory hypothesis testing: keep it maximal. J. Mem. Lang. tracking with MATLAB and the psychophysics toolbox. Behav. Res. Methods
68, 255–278. doi: 10.1016/j.jml.2012.11.001 Instrum. Comput. 34, 613–617. doi: 10.3758/BF03195489
Bates, D., Machler, M., Bolker, B., and Walker, S. (2014). Fitting linear Davis, C. J., and Perea, M. (2005). Buscapalabras: a program for deriving
mixed-effects models using lme4. arXiv preprint arXiv:1406.5823. orthographic and phonological neighborhood statistics and other
doi: 10.18637/jss.v067.i01 psycholinguistic indices in Spanish. Behav. Res. Methods 37, 665–671.
Belloro, V. (2012). “La estructura informativa,” in El funcionalismo en la teoría doi: 10.3758/BF03192738
lingüística: la Gramática del Papel y la Referencia; Introducción, avances y Dröge, A., Maffongelli, L., and Bornkessel-Schlesewsky, I. (2014). “Luigi piace
aplicaciones, eds R. Mairal, L. Guerrero, and C. González Vergara (Madrid: a Laura?” in Structuring the Argument: Multidisciplinary Research on Verb
Akal), 225–244. Argument Structure, eds A. Bachrach, I. Roy, and L. Stockall (Amsterdam: John
Birner, B. J., and Ward, G. L. (1998). Information Status and Noncanonical Benjamins Publishing Company), 83–118. doi: 10.1075/lfab.10.05dro
Word Order in English, Vol. 40. Amsterdam: John Benjamins Publishing. Ferreira, V. S. (2003). The persistence of optional complementizer production:
doi: 10.1075/slcs.40 why saying “that” is not saying “that” at all. J. Mem. Lang. 48, 379–398.
Bornkessel, I., and Schlesewsky, M. (2006). The extended argument dependency doi: 10.1016/S0749-596X(02)00523-5
model: a neurocognitive approach to sentence comprehension across Foley, W. A., and Van Valin, R. D. Jr. (1984). Functional Syntax and Universal
languages. Psychol. Rev. 113:787. doi: 10.1037/0033-295X.113.4.787 Grammar, Volume 38 of Cambridge Studies in Linguistics. London: Cambridge
Bornkessel, I., Schlesewsky, M., and Friederici, A. D. (2003). Eliciting thematic University Press.
reanalysis effects: the role of syntax-independent information during parsing. Gattei, C. A., Dickey, M. W., Wainselboim, A. J., and París, L. (2015a). The
Lang. Cogn. Process. 18, 269–298. doi: 10.1080/01690960244000018 thematic hierarchy in sentence comprehension: a study on the interaction
Bornkessel, I., Zysset, S., Friederici, A. D., von Cramon, D. Y., and Schlesewsky, between verb class and word order in Spanish. Q. J. Exp. Psychol. 68, 1981–2007.
M. (2005). Who did what to whom? The neural basis of argument doi: 10.1080/17470218.2014.1000345
hierarchies during language comprehension. Neuroimage 26, 221–233. Gattei, C. A., Sevilla, Y., Tabullo, A., Wainselboim, A., París, L., and Shalom,
doi: 10.1016/j.neuroimage.2005.01.032 D. (2017). Prominence in spanish sentence comprehension: an eye-tracking
Bornkessel-Schlesewsky, I., and Schumacher, P. B. (2016). “Towards a study. Lang. Cogn. Neurosci. 33, 587–607. doi: 10.1080/23273798.2017.
neurobiology of information structure,” in The Oxford Handbook of 1397278
Information Structure, eds C. Fery and S. Ishihara (Oxford: Oxford University Gattei, C. A., Tabullo, Á., París, L., and Wainselboim, A. J. (2015b). The role of
Press), 581–598. doi: 10.1093/oxfordhb/9780199642670.013.22 prominence in Spanish sentence comprehension: an ERP study. Brain Lang.
Burmester, J., Spalek, K., and Wartenburger, I. (2014). Context updating during 150, 22–35. doi: 10.1016/j.bandl.2015.08.001
sentence comprehension: the effect of aboutness topic. Brain Lang. 137, 62–76. Givón, T. (1984). Syntax: A Functional-Typological Approach. Amsterdam: J.
doi: 10.1016/j.bandl.2014.08.001 Benjamins. doi: 10.1075/z.17

Gleitman, L. R., January, D., Nappa, R., and Trueswell, J. C. (2007). On the give and Pinheiro, J. C., and Bates, D. M. (2000). “Linear mixed-
take between event apprehension and utterance formulation. J. Mem. Lang. 57, effects models: basic concepts and examples,” in Mixed-Effects
544–569. doi: 10.1016/j.jml.2007.01.007 Models in S and S-Plus, eds J. C. Pinheiro and D. M. Bates
Gordon, P. C., and Hendrick, R. (1998). The representation and (New York, NY: Springer), 3–56. doi: 10.1007/978-1-4419-0
processing of coreference in discourse. Cogn. Sci. 22, 389–424. 318-1_1
doi: 10.1207/s15516709cog2204_1 R Core Team (2013). R: A Language and Environment for Statistical Computing.
Haupt, F. S., Schlesewsky, M., Roehm, D., Friederici, A. D., and Bornkessel- Vienna: R Foundation for Statistical Computing.
Schlesewsky, I. (2008). The status of subject-object reanalyses in Rayner, K., Raney, G. E., and Pollatsek, A. (1995). “Eye movements and discourse
the language comprehension architecture. J. Mem. Lang. 59, 54–96. processing,” in Sources of Coherence in Reading, eds R. F. Lorch Jr., and E. J.
doi: 10.1016/j.jml.2008.02.003 O’Brien (Hillsdale), 9–35.
Hyönä, H., and Hujanen, J. (1997). Effects of case marking and word order on Rayner, K., and Well, A. D. (1996). Effects of contextual constraint on eye
sentence parsing in finnish: an eye fixation analysis. Q. J. Exp. Psychol. A 50, movements in reading: a further examination. Psychon. Bull. Rev. 3, 04–509.
841–858. doi: 10.1080/027249897391919 doi: 10.3758/BF03214555
Just, M. A., and Carpenter, P. A. (1980). A theory of reading: from eye fixations to Schumacher, P. B., and Hung, Y.-C. (2012). Positional influences on information
comprehension. Psychol. Rev. 87:329. doi: 10.1037/0033-295X.87.4.329 packaging: insights from topological fields in German. J. Mem. Lang. 67,
Just, M. A., Carpenter, P. A., and Woolley, J. D. (1982). Paradigms 295–310. doi: 10.1016/j.jml.2012.05.006
and processes in reading comprehension. J. Exp. Psychol. Gen. 111:228. Suñer, M. (1982). Syntax and Semantics of Spanish Presentational Sentence-Types.
doi: 10.1037/0096-3445.111.2.228 Georgetown University School of Language.
Kaiser, E. (2012). Taking action: a cross-modal investigation of discourse-level Van Valin, R. D., and LaPolla, R. J. (1997). Syntax: Structure, Meaning,
representations. Front. Psychol. 3:156. doi: 10.3389/fpsyg.2012.00156 and Function. Cambridge, MA: Cambridge University Press.
Kaiser, E., and Trueswell, J. C. (2004). The role of discourse context in doi: 10.1017/CBO9781139166799
the processing of a flexible word-order language. Cognition 94, 113–147. Van Valin, R. D. Jr. (1999). “A typology of the interaction of focus structure
doi: 10.1016/j.cognition.2004.01.002 and syntax,” in Typology and the Theory of Language: From Description to
Kamide, Y., and Mitchell, D. C. (1999). Incremental pre-head Explanation, eds E. Raxilina and J. Testelec (Moscow: Languages of Russian
attachment in Japanese parsing. Lang. Cogn. Process. 14, 631–662. Culture), 511–524.
doi: 10.1080/016909699386211 Van Valin, R. D. Jr. (2005). Exploring the Syntax-Semantics Interface. Cambridge,
Kamienkowski, J. E., Carbajal, M. J., Bianchi, B., Sigman, M., and Shalom, MA: Cambridge University Press. doi: 10.1017/CBO9780511610578.001
D. E. (2018). Cumulative repetition effects across multiple readings of Vasishth, S., von der Malsburg, T., and Engelmann, F. (2013). What eye movements
a word: evidence from eye movements. Discourse Process. 55, 256–271. can tell us about sentence comprehension. Wiley Interdiscipl. Rev. Cogn. Sci. 4,
doi: 10.1080/0163853X.2016.1234872 125–134. doi: 10.1002/wcs.1209
Kliegl, R., Grabner, E., Rolfs, M., and Engbert, R. (2004). Length, frequency, and Wang, L., Schlesewsky, M., Philipp, M., and Bornkessel-Schlesewsky, I. (2012).
predictability effects of words on eye movements in reading. Eur. J. Cogn. “The role of animacy in online argument interpretation in Mandarin Chinese,”
Psychol. 16, 262–284. doi: 10.1080/09541440340000213 in Case, Word Order and Prominence, eds M. Lamers and P. de Swart
Lambrecht, K. (1994). Information Structure and Sentence Form: Topic, Focus, and (Dordrecht: Springer), 91–119. doi: 10.1007/978-94-007-1463-2_5
the Mental Representations of Discourse Referents, Vol. 71. Cambridge, MA: Wolff, S., Schlesewsky, M., and Bornkessel-Schlesewsky, I. (2007). The interaction
Cambridge University Press. doi: 10.1017/CBO9780511620607 of universal and language-specific properties in the neurocognition of language
Lamers, M., and De Swart, P. (2012). Case, Word order and Prominence: Interacting comprehension: evidence from the processing of word order permutations in
Cues in Language Production and Comprehension, Vol. 40. Dordrecht: Springer Japanese. J. Cogn. Neurosci 288.
Science & Business Media. doi: 10.1007/978-94-007-1463-2 Zubizarreta, M. L. (1998). Prosody, Focus, and Word Order. Cambridge, MA: MIT
Logacev, P., and Vasishth, S. (2013). em2: A Package for Computing Reading Time Press.
Measures for Psycholinguistics. R package version 0.9. Available online at: http://
CRAN.R-project.org/package=em2 Conflict of Interest: The authors declare that the research was conducted in the
Lowder, M. W., Choi, W., and Gordon, P. C. (2013). Word recognition during absence of any commercial or financial relationships that could be construed as a
reading: the interaction between lexical repetition and frequency. Mem. Cogn. potential conflict of interest.
41, 738–751. doi: 10.3758/s13421-012-0288-z
O’Brien, E. J., Raney, G. E., Albrecht, J. E., and Rayner, K. (1997). Processes Copyright © 2021 Gattei, París and Shalom. This is an open-access article distributed
involved in the resolution of explicit anaphors. Discourse Process. 23, 1–24. under the terms of the Creative Commons Attribution License (CC BY). The use,
doi: 10.1080/01638539709544979 distribution or reproduction in other forums is permitted, provided the original
Ocampo, F. (1995). “The word order of two-constituent constructions in spoken author(s) and the copyright owner(s) are credited and that the original publication
Spanish,” in Word Order in Discourse, eds P. A. Downing and M. Noonan in this journal is cited, in accordance with accepted academic practice. No use,
(Amsterdam: John Benjamins), 425–447. doi: 10.1075/tsl.30.15oca distribution or reproduction is permitted which does not comply with these terms.

Spanish - Eye Tracking - Reading - Information Structure

Uploaded by

Copyright:

Available Formats

Spanish - Eye Tracking - Reading - Information Structure

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spanish - Eye Tracking - Reading - Information Structure

Uploaded by

Copyright:

Available Formats

ORIGINAL RESEARCH

published: 06 April 2021

Information Structure and Word

Frontiers in Psychology | www.frontiersin.org 1 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 2 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 3 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 4 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 5 April 2021 | Volume 12 | Article 629724

TABLE 1 | Critical sentences used in the current eye-tracking study.

Condition Context Critical sentence Question

Frontiers in Psychology | www.frontiersin.org 6 April 2021 | Volume 12 | Article 629724

María le respondía|encantaba a Ana PP1 PP2 PP3 PP4

Frontiers in Psychology | www.frontiersin.org 7 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 8 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 9 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 10 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 11 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 12 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 13 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 14 April 2021 | Volume 12 | Article 629724

ETHICS STATEMENT FUNDING

REFERENCES Chafe, W. (1976). “Givenness, contrastiveness, definiteness, subjects, topics, and

Frontiers in Psychology | www.frontiersin.org 15 April 2021 | Volume 12 | Article 629724

Frontiers in Psychology | www.frontiersin.org 16 April 2021 | Volume 12 | Article 629724

You might also like