=Paper=
{{Paper
|id=Vol-2253/paper08
|storemode=property
|title=Grammatical Class Effects in Production of Italian Inflected Verbs
|pdfUrl=https://ceur-ws.org/Vol-2253/paper08.pdf
|volume=Vol-2253
|authors=Maria De Martino,Azzurra Mancuso,Alessandro Laudanna
|dblpUrl=https://dblp.org/rec/conf/clic-it/MartinoML18
}}
==Grammatical Class Effects in Production of Italian Inflected Verbs==
https://ceur-ws.org/Vol-2253/paper08.pdf
Grammatical class effects in production of Italian inflected verbs
Maria De Martino, Azzurra Mancuso, Alessandro Laudanna
LaPSUS, Laboratory of Experimental Psychology, University of Salerno
Via Giovanni Paolo II, 132 Fisciano, SA, 84084, Italy
mdemartino@unisa.it amancuso@unisa.it alaudanna@unisa.it
1. Introduction
Abstract
Models of lexical access share the assumption
English. We report a picture-word inter- that different kinds of linguistic information (se-
ference (PWI) experiment conducted in mantic, orthographic-phonological, syntactic-
Italian where target verbs were used to grammatical, and so on) have different levels of
name pictures in presence of semantically lexical representation (Caramazza, 1997; Levelt,
related and unrelated distracters. The Roelofs and Meyer, 1999; Dell, 1986). The pic-
congruency of grammatical class be- ture-word interference (PWI) paradigm has been
tween targets and distracters was manipu- widely exploited to test the dynamics of activa-
lated and nouns and verbs were used as tion of different properties of words during lexi-
distracters. Consistently with previous cal production. Such a task allows the observa-
studies, an expected semantic interfer- tion of specific lexical effects by manipulating
ence effect was observed but, interesting- the linguistic relation between words to be used
ly, such an effect does not equally apply in a picture naming task and written distracter-
to target-distracter pairs sharing or not words super-imposed to pictures. The basic as-
grammatical class information. This out- sumption is that linguistic information of a dis-
come seems to corroborate the hypothesis tractor influences the time needed to select the
of the intervention of grammatical con- appropriate word-form to name a picture. For
straints in word production as explored in instance, two well-known effects observed in
the PWI task. PWI, the semantic interference and the phono-
logical facilitation effects, are thought to reflect
Italiano. Questo lavoro descrive un respectively the competition at the lexical level
esperimento di interferenza figura-parola between the lexical representations of the target
sull’ italiano in cui le figure dovevano and the distracter and the co-activation of the
essere denominate usando verbi in pre- phonemes shared by the target and the distracter
senza di distrattori semanticamente col- during the phonetic encoding stage.
legati o non collegati alla figura. È stata Scholars have also tried to investigate the acti-
manipolata anche la congruenza di clas- vation of grammatical information in speech
se grammaticale tra target e distrattori; production through the PWI paradigm but con-
questi ultimi nella metà dei casi erano flicting evidence has been collected. For in-
nomi e nell’altra verbi. In linea con studi stance, Pechmann and Zerbst (2002), Pechmann
precedenti, abbiamo ottenuto un effetto and coll. (2004), Vigliocco and coll. (2005), Ro-
di interferenza semantica; il dato interes- driguez-Ferreiro and coll. (2014), De Simone and
sante è che quest’ultimo effetto interessa Collina (2016) obtained grammatical class ef-
in modo differente le coppie target- fects, while Mahon and coll. (2007), Iwasaki and
distrattore congruenti o non congruenti coll. (2008) and Janssen and coll. (2010) did not.
per classe grammaticale. Questo risulta- Arguably, the variability in the experimental evi-
to sembra corroborare l’ipotesi che nella dence can be ascribed to heterogeneous method-
di produzione di parole esplorata attra- ologies across studies: for instance, results ob-
verso il compito di interferenza figura- tained by Vigliocco and coll. (2005) could be
parola giochino un ruolo le proprietà biased by their methodological choice to admin-
grammaticali delle parole. ister noun-distracters with determiners, while in
the study of Rodriguez-Ferreiro and coll. (2014)
semantic categories (actions/objects/instruments) ten form frequency (CoLFIS; Bertinetto et al.,
partially overlapped grammatical classes and a 2005) length, semantic relatedness. Formal or-
confound due to an imageability bias (Exp. 3) thographic or phonological overlap between tar-
was present. gets and distracters was avoided. The mean val-
As a consequence, the intervention of grammati- ues and standard deviations for each of these var-
cal constraints during production processes, as iables are reported in Table 1.
explored in PWI tasks, is still debated. The experimental list was composed of 140 tri-
In this study on Italian we aimed at exploring the als where the 35 target-verbs were accompanied
problem by trying to avoid possible confounds by 70 verb-distracters (35 semantically related
existing in previous studies. and 35 unrelated) and by 70 noun-distracters (35
semantically related and 35 unrelated). Two ad-
ditional distracters were used as filler trials: for
2. Method each target a related and an unrelated word were
provided; these filler distracters differed from
Participants: Thirty-six undergraduate students
experimental distracters since they were word-
(28 females) from University of Salerno volun-
class ambiguous items. Instances of all experi-
tarily took part in the experiment. They were all
mental conditions are reported in Table 2 and an
native speakers of Italian and they all had normal
example of experimental item is reported in Fig-
or corrected-to-normal vision. Their age ranged
ure 1.
from 20 to 30 years (mean=22; sd=2.5). They
served for a session lasting about 45 minutes.
Semantically Semantically
related pairs unrelated pairs
Materials: Thirty-five black-and-white line
noun verb noun verb
drawings depicting actions were used as experi-
mental items. Participants were instructed to length 7.1 6.3 7.1 6.3
(1.6) (1.4) (1.6) (1.4)
name these pictures by using inflected verb
forms (either present indicative, or 3rd singular written 79.3 75.3 79.3 75.3
person). These verbs constituted the target items. form frequency (92.3) (97.7) (92.3) (97.7)
For each target-verb a semantically related dis-
3.5 3.7 3.5 3.7
tracter-verb and a semantically related distracter- imageability
(0.6) (0.6) (0.6) (0.6)
noun were selected, so that a list of 35 distracter-
verbs and a list of 35 distracter-nouns were built. shared letters 2 2 2 1.6
The selected nouns and verbs were not affected between targets (1.1) (1.1) (1.1) (1.0)
and distracters
by the semantic bias due to the object/action di-
chotomy. The semantic relatedness between tar-
gets and distracters was calculated on the basis of subjective
semantic 3.3 3.5 1.4 1.4
2 measures: corpus-based automatic semantic relatedness (0.9) (1.03) (0.4) (0.4)
metrics (WEISS, Word-embeddings Italian se- ratings
mantic spaces; Marelli, 2017) and subjective rat-
ings on a 5 point Likert scale1. WEISS metrics
0.7 0.6 0.9 0.9
(0.1) (0.2) (0.1) (0.1)
The same distracters were differently paired
with the target verbs so that two lists of unrelated Table 1: Mean values and standard deviations (in parenthesis) of
distracters’ characteristics
nominal (related-noun and unrelated-noun exper-
imental conditions) and verbal (related-verb and
unrelated-verb experimental conditions) distract-
ers were created. Distracters in the four experi-
mental conditions were matched for the main
psycholinguistics variables: imageability, writ-
1
The first measure provided objective values, based on
distributional estimates, for the semantic distance between
each target-word and its distracter. The second measure
allowed us to ascertain to what extent the specific word
sense evoked by the picture was related to the distracter-
word.
Figure 1. An example of a related distracter-picture pair
times from the appearance of the stimuli to the
onset of articulation were collected by a voice
Related noun: frittura (frying) key connected to the computer and participant
Related verb: frigge (he/she fries) responses were recorded. Upon a response, the
Distracters Unrelated noun: rumore (noise)
Unrelated verb: sente (he/she listens
picture and the distracter disappeared from the
to) screen. Both the presentation of the stimuli and
the recording of the responses were managed by
the E-Prime software 2.0. The responses of the
Target cuoce (he/she cooks) participants were checked for accuracy by an
experimenter.
Table 2. Distracter-target pairs
Each single trial consisted of the following
events: a fixation cross presented at the center of
In order to prevent any strategic bias due to
the screen for 300 ms; the stimulus until the re-
semantic and/or grammatical relationships
sponse or for a maximum of 2.5 seconds; a feed-
among targets and distracters, 15 additional pic-
back mask signaling the activation of the voice
tures were used as filler targets and were pre-
key of 500ms, a blank interval of 500 ms. The
sented with 6 different distracters. The whole list
SOA between pictures and distracter-words was
of both experimental and filler target-distracter
0 ms.
pairs was composed of 300 trials: 33% were se-
Words pronounced incorrectly, non-expected
mantically related trials and 67% were unrelated
picture names, hesitations in giving the respons-
trials.
es, word fragments, omissions, verbal dysfluen-
cies and responses given after the deadline were
Procedure: The participants were tested individ-
scored as errors. Invalid responses (e.g., trials in
ually; an experimental session consisted of three
which the voice key was triggered by external
parts: a familiarization, a practice and an experi-
noise) and responses shorter than 400 ms were
mental phase. The E-Prime software 2.0 (Psy-
considered as missing data.
chology Software Tools, Inc., Pittsburgh, PA)
At the end of the practice phase, the experiment
was used.
started and 6 experimental blocks of 50 trials (35
At the beginning of the experiment, each par-
experimental items and 15 filler items) were pre-
ticipant was familiarized with the whole set of
sented, for a total of 300 trials. An equal number
experimental and filler pictures in an untimed
of items from each experimental condition was
picture naming session. In this phase, the pic-
included in every block. Blocks were counterbal-
tures were presented on the computer screen with
anced across participants. In each block, stimuli
a superimposed row of Xs to simulate the dis-
underwent a randomization governed by the E-
tracter word. Participants learned to produce the
Prime software 2.0.
targets upon presentation of the corresponding
pictures. If participants named a picture with a
verb that differed from the one designed as the 3. Results
target by experimenters, a feedback was given:
the expected verb was provided to participants
and they were invited to use it in the experi- An analysis of variance (ANOVA) was per-
mental session. formed on naming latencies and accuracy rates
Following the familiarization phase, a practice by subjects (F1) and by items (F2) with the dis-
block was administered where participants were tractor type (four levels) as a variable. For the
asked to name each picture as inflected verb sake of conciseness only the statistically signifi-
forms (present indicative 3rd singular person, e.g. cant analyses will be reported and discussed.
beve, he/she drinks) and were instructed to re- A main effect of semantic relatedness has been
spond as quickly and accurately as possible, observed both in the ANOVA by participants
while ignoring the distracter word. The experi- (F1(1, 35) = 4.56, p< .05) and by items (F2(1,
menter was seated behind the participant and 30) = 4.46, p< .05) on response latencies. Re-
recorded errors and equipment failures. The sponses to target verbs were slower when they
stimuli presented in the training phase were part were accompanied by semantically related dis-
of the filler set. tracters (+17 ms).
The stimuli appeared on a video display unit Neither effects of grammatical class nor inter-
controlled by a personal computer. Reaction action between grammatical class and semantic
relation were found.
Two-tailed t tests comparing the semantic in- ly, the Italian inflected form “amavo”
terference effect within the grammatical class (indicative, imperfect, 1st singular per-
congruent and non-congruent target/distracter son, I used to love), is composed of a
pairs revealed that the semantic interference ef- stem, “am-”, which conveys the core
fect reaches the statistical significance with meaning of the verb, the vowel “-a-”,
noun-distracters (+24 ms, p = .02) but not with which specifies the inflectional pattern
verb-distracters (+9 ms, p = .43). The results are compatible with the verbal stem, the
graphically shown in Table 3. segment “–v-”, which encodes mood and
tense information, and the segment ”-o”
which encodes person and number in-
Noun Verb formation. None of these features, with
distracters distracters the exception of meaning and number
1020 ms 1011 ms features, can be part of the lexical repre-
Related sentation of noun-forms. This latter ma-
(125) (121)
996 ms 1002 ms nipulation has relevant consequences on
Unrelated the detection of grammatical class effect
(107) (111)
Table 3. Mean response latencies and standard deviations (in pa- in PWI, since it has been demonstrated
renthesis) for all conditions that, when finite verbs have to be pro-
duced, the naming context sets the re-
sponse-relevant criterion on the gram-
4. Conclusions matical class of verbs and then noun-
One of the aim of the present experiment was distracters tend to interfere significantly
to overcome some limitations of previous inves- more than verb-distracters (De Martino
tigations. The following constraints were adopt- & Laudanna, 2017)3.
ed: Consistently with previous PWI evidence, our
1. We contrasted the production of verbs experiment replicated a reliable semantic inter-
when presented with semantically related ference effect. This finding confirms that the se-
and unrelated distracters: the expected lection of an oral target response is slowed-down
semantic interference effect guaranteed by the activation of a semantically-related dis-
for the reliability of the paradigm. tracter because the lexical system has to manage
2. We selected experimental materials the level of activation of target lexical competi-
where the differences between grammat- tors, including the highly activated semantically
ical classes in terms of their semantic related distracter word. Interestingly, we ob-
domain (objects (nouns) vs. actions served that, at least when pictures have to be
(verbs)) was kept under control. named by using inflected verb forms, such an
3. Word-class ambiguous items were ex- effect does not equally affect all semantically
cluded by experimental materials. related target-distracter pairs: related pairs shar-
4. Inflected finite verbal-forms were used ing grammatical class information do not exhibit
both as targets and distracters: these ver- significant semantic interference but grammati-
bal forms allow to maximize the differ- cal-class incongruent pairs do.
ence between nouns and verbs2. Actual- In conclusion, our data suggest that the PWI
task is sensitive to the manipulation of grammat-
ical class information. In other words, such a pat-
2
The distinction between finite and non-finite moods is tern of results is compatible with the intervention
motivated on morphological and syntactic grounds: finite- of grammatical constraints during production
forms are inflected for person and in syntactic context they
are used as verbal predicates. Conversely, non-finite forms processes, as explored in the PWI task.
lack for person inflection and are used in periphrastic con-
struction or in combination with auxiliary verbs to assemble References
the “composed tenses” of the paradigm. Under certain
circumstances, non-finite forms undergo syntactic trans-
categorization and behave as nouns or adjectives: “mi piace Bertinetto, P. M., Burani, C., Laudanna, A., Marconi,
ballare [infinitive]”, (I love dancing). “I partecipanti [pre- L., Ratti, D., Rolando, C., & Thornton, A. M.
sent participle], sono pronti” (participants are ready); “tre (2005). Corpus e Lessico di Frequenza
gare vinte [past participle, from “vincere”] e cinque perse
[past participle, from “perdere”], (three competitions won
and five lost). 3
This result was obtained regardless of semantic relation
between targets and distracters.
dell’Italiano Scritto (CoLFIS). digm. Language, Cognition and Neurosci-
http://linguistica.sns.it/CoLFIS/Home.htm ence, 29(1), 125-135.
Caramazza, A. (1997). How many levels of pro- Vigliocco, G., Vinson, D. P., & Siri, S. (2005). Se-
cessing are there in lexical access?.Cognitive Neu- mantic similarity and grammatical class in naming
ropsychology, 14(1), 177-208. actions. Cognition, 94(3), B91-B100.
Dell, G. S. (1986). A spreading-activation theory of
retrieval in sentence production. Psychological Re-
view, 93(3), 283.
De Martino, M., & Laudanna, A. (2017). The role of
grammatical properties in the word-picture inter-
ference paradigm: data from single verbs produc-
tion in Italian. In: Abstracts of the 20th Conference
of the European Society for Cognitive Psychology
(198-198).
De Simone, F., & Collina, S. (2016). The Picture–
Word Interference Paradigm: Grammatical Class
Effects in Lexical Production. Journal of Psycho-
linguistic Research, 45(5), 1003-1019.
Iwasaki, N., Vinson, D. P., Vigliocco, G., Watanabe,
M., & Arciuli, J. (2008). Naming action in Japa-
nese: Effects of semantic similarity and grammati-
cal class. Language and Cognitive Process-
es, 23(6), 889-930.
Janssen, N., Melinger, A., Mahon, B. Z., Finkbeiner,
M., & Caramazza, A. (2010). The word class effect
in the picture–word interference paradigm. The
Quarterly Journal of Experimental Psycholo-
gy, 63(6), 1233-1246.
Levelt, W. J., Roelofs, A., & Meyer, A. S. (1999). A
theory of lexical access in speech produc-
tion. Behavioral and Brain Sciences, 22(1), 1-38.
Mahon, B. Z., Costa, A., Peterson, R., Vargas, K. A.,
& Caramazza, A. (2007). Lexical selection is not
by competition: a reinterpretation of semantic in-
terference and facilitation effects in the picture-
word interference paradigm. Journal of Experi-
mental Psychology: Learning, Memory, and Cog-
nition, 33 (3), 503-535.
Marelli, M. (2017). Word-Embeddings Italian Seman-
tic Spaces: A semantic model for psycholinguistic
research. Psihologija, 50(4), 503-520.
Pechmann, T., & Zerbst, D. (2002). The activation of
word class information during speech produc-
tion. Journal of Experimental Psychology: Learn-
ing, Memory, and Cognition, 28 (1), 233–243.
Pechmann, T., Garrett, M., & Zerbst, D. (2004). The
time course of recovery for grammatical category
information during lexical processing for syntactic
construction. Journal of Experimental Psychology:
Learning, Memory, and Cognition, 30(3), 723.
Rodríguez-Ferreiro, J., Davies, R., & Cuetos, F.
(2014). Semantic domain and grammatical class ef-
fects in the picture–word interference para-